Conference Agenda

Overview and details of the sessions of this conference. Please select a date or location to show only sessions at that day or location. Please select a single session for detailed view (with abstracts and downloads if available).

Please note that all times are shown in the time zone of the conference. The current conference time is: 17th Sept 2021, 05:52:17am CEST

Session Overview
Date: Sunday, 19/Sept/2021
10:30am - 11:30amPre congress registration
Location: Room A - Angelicum Congress Centre
Room A - Angelicum Congress Centre 
10:30am - 11:30amPre congress registration
Location: Room B - Angelicum Congress Centre
Room B - Angelicum Congress Centre 
10:30am - 8:00pmCongress registration
Location: Sala delle Colonne - Angelicum Congress Centre
Sala delle Colonne - Angelicum Congress Centre 
11:00am - 3:45pmPre-Congress (CME): Pharmacometrics-enhanced TDM
Location: Room A - Angelicum Congress Centre

Session Chair: Sven van Dijkman
Session Chair: Brenda de Winter

11.00 am - 01.00 pm: SESSION 1: Precision Dosing

11.00 am: Talk 1 Population approach basics. Erwin Dreesen, Maurice Ahsman

11.30 am: Talk 2 Covariate modelling. Iris Minichmayr

12.00 pm: Talk 3 Hands-on 1: Developing a basic PK model. All present PMX committee members

01.00 pm: Lunch

01.45 pm - 03.45 pm: SESSION 2: Precision dosing and Clinical Trial Design

01.45 pm: Talk 4 Model-informed precision dosing. Sebastian Wicha, Brenda de Winter

02.15 pm: Talk 5: Clinical trial design. Sven van Dijkman

02.45 pm: Elective hands-on:

- Hands-on 2: 
Treatment individualisation using InsightRX and TDMx. Brenda de Winter, Sebastian Wicha
+ committee members

- Hands-on 3:
Clinical Trial Design. Sven van Dijkman + committee members

03.45 pm Pre-congress seminar closing

Room A - Angelicum Congress Centre 
11:00am - 3:45pmPre-Congress (CME): TDM in pediatric patients as well as in neonates
Location: Room B - Angelicum Congress Centre

Session Chair: Alberto Villani 
Session Chair: Bianca Maria Goffredo

11.00 am - 01.00 pm: SESSION 1

11.00 am: Welcome - Franco Locatelli 

11.10 am: TDM in children supported with ECMO. Matteo Di Nardo

11.40 am: Adverse drug reactions and pediatric intensive care. What impact? Morida Khalil Ramla

12.10 pm: Discussion/Questions

01.00 pm: Lunch

02.00 pm - 03.45 pm: SESSION 2

02.00 pm: Drug monitoring in human milk from women after labor analgesia. Andrea Dotta

02.30 pm: TDM in paediatric haematology. Pietro Merli

03.00 pm:Neonatal pharmacology: how to improve the dosing regimens for TDM compounds based on PK modeling. Anne Smits

03.30 pm: Discussion/Questions and closing considerations

Room B - Angelicum Congress Centre 
4:00pm - 4:30pmWelcome address
Location: Aula Magna - Angelicum Congress Centre

Welcome address : Mario Regazzi & Franco Locatelli

Presidential Welcome address: Yusuke Tanigawara

Aula Magna - Angelicum Congress Centre 
4:30pm - 5:00pmOpening Lecture: The role of PK and PD for treatment of childhood hematological disorders with innovative antibodies.
Location: Aula Magna - Angelicum Congress Centre

Opening lecture: Franco Locatelli

Aula Magna - Angelicum Congress Centre 
5:00pm - 5:45pmPlenary Lecture 1: Precision Dosing: A view from industry
Location: Aula Magna - Angelicum Congress Centre

Plenary Speaker: Richard W. Peck

Aula Magna - Angelicum Congress Centre 
5:45pm - 6:15pmIrving Sunshine Award
Location: Aula Magna - Angelicum Congress Centre
Aula Magna - Angelicum Congress Centre 
6:15pm - 9:00pmWelcome cocktail
Location: Angelicum Congress Centre
Angelicum Congress Centre 
Date: Monday, 20/Sept/2021
7:30am - 8:30amMorning roundtable session: Alternative sampling devices to collect dried blood microsamples: state-of-the-Art
Location: Hotel The Hive

Chair: Birgit Kock

Chair: Christophe Stove

Hotel The Hive 
7:30am - 8:30amMorning roundtable session: TDM of tirosyne kinase inhibitors
Location: Hotel The Hive

Chair: Dirk Moes

Hotel The Hive 
8:00am - 7:30pmCongress registration
Location: Sala delle Colonne - Angelicum Congress Centre
Sala delle Colonne - Angelicum Congress Centre 
8:45am - 9:00amOpening Remarks
Location: Aula Magna - Angelicum Congress Centre
Aula Magna - Angelicum Congress Centre 
9:00am - 9:45amPlenary Lecture 2 (CME): The New Psychoactive Substances: availability, effects and new diagnostic needs.
Location: Aula Magna - Angelicum Congress Centre

Plenary Speaker: Carlo Locatelli

Aula Magna - Angelicum Congress Centre 
9:45am - 10:15amPippenger Award
Location: Aula Magna - Angelicum Congress Centre
Aula Magna - Angelicum Congress Centre 
10:30am - 11:00amMorning coffee & tea, exhibition and poster viewing
Location: Angelicum Congress Centre
Angelicum Congress Centre 
11:00am - 12:30pmSymposium 1: New issues in Immunosuppression
Location: Aula Magna - Angelicum Congress Centre
Session Chair: Stein Bergan
Session Chair: Franco Locatelli

11.00 am: Talk 1: Biomarkers in solid organ transplantation. Stein Bergan

11.30 am: Talk 2: JAK inhibitors – a new mode of immunosuppression in rheumatic disease. Guro Lovick Goll

12.00 pm: Talk 3: Immunosup PK and physical activity. Massimo Baraldo

Aula Magna - Angelicum Congress Centre 
11:00am - 12:30pmSymposium 2: The “NPS (New Psychoactive Substances) phenomenon”
Location: Room A - Angelicum Congress Centre

Session Chair: Rossella Gottardo

11.00 am: Talk 4: Pharmaco-toxicological aspects of NPS. Matteo Marti

11.25 am: Talk 5: The European Early warning system on new psychoactive substances. Michael Evans-Brown

11.50 am: Talk 6: NPS testing in emergency departments: screenings and quantification. Antonella Valli

12.15 pm: Talk 7: Biological samples for the determination of recreational substances. Sabina Strano Rossi

Room A - Angelicum Congress Centre 
11:00am - 12:30pmOral Session: Pharmacogenetics
Location: Room B - Angelicum Congress Centre

Session Chair: Ron Van Schaik
Session Chair: Antonio D’Avolio

Room B - Angelicum Congress Centre 

Limited precision of CYP2D6 diplotype scores in predicting in vivo metabolizer phenotypes

Prof. Espen Molden

Diakonhjemmet Hospital, Center for Psychopharmacology, Norway

Introduction: Genetic differences in cytochrome P450 (CYP)-mediated metabolism have been known for decades. The clinically most important polymorphic CYP enzyme is CYP2D6, which plays a key role in the metabolism of about one fourth of all drugs. Guidelines have been published to predict CYP2D6 metabolizer phenotype and individual dose requirements of CYP2D6-metabolized drugs.

Methods: Metabolic phenotypes of the CYP2D6 substrates aripiprazole, risperidone and risperidone were compared between patients with different CYP2D6 genotypes, but similar CYP2D6 diplotype activity scores according to current consensus guidelines of the Clinical Pharmacogenetics Implementation Consortium (CPIC) and the Dutch Pharmacogenetics Working Group (DPWG).

Results: Large differences in CYP2D6 metabolizer phenotypes were found between patients with similar diplotype activity scores assigned from different genotypes, e.g. the metabolic ratios of all CYP2D6 indicator drugs were >5-fold lower for CYP2D6*41/Null carriers than for CYP2D6*10/*10 carriers, which both have a diplotype score of 0.5. The most obvious misclassification of current guidelines is the assignment of an activity score of 0.5 for CYP2D6*41. Its activity score should rather be defined as 0.1 and the CYP2D6*41/Null genotype considered to be merged into the poor metabolizer subgroup.

Conclusion: The CPIC/DPWG guidelines have limited precision in predicting CYP2D6 metabolizer phenotypes and individual dose requirements of CYP2D6 substrates. Use of the exact CYP2D6 genotype is probably more appropriate for estimating dosing to reach target concentrations than the activity score model. It is proposed that future research on personalized dosing of CYP2D6 substrates should focus on developing algorithms accounting for all variables determining clearance, and not only the partial intrinsic CYP2D6 clearance.

An easy, fast and inexpensive multiplex pharmacogenetics assay to simultaneously analyze clinically relevant genetic polymorphisms in CYP3A4, CYP3A5, CYP1A2, CYP2C9, CYP2C19, CYP2D6, ABCB1 and VKORC1 genes

Dr. Camille Tron1,2,3, Régis Bouvet4, Dr. Marie-Clémence Verdier1,2,3, Dr. Christèle Dubourg4, Prof. Marie-Dominique Galibert4, Prof. Eric Bellissant1,2,3

1Pharmacology Department, Rennes University Hospital, Rennes, France; 2CIC 1414 (Clinical Investigation Center), INSERM, Rennes, France; 3Experimental and Clinical Pharmacology Laboratory, University of Rennes, Rennes, France; 4Department of Molecular Genetics and Genomics, Rennes Hospital University, Rennes, France


In order to optimize genotyping of clinically relevant pharmacogenes, the trend is nowadays to analyze a panel of several genetic polymorphisms. This approach is valuable for patients but might be challenging to perform in some laboratories. Indeed, it usually needs to use next generation sequencing which requires expensive reagents and instruments and specific skills to interpret results. To overcome these hurdles, the aim of this work was to validate an easy, fast and inexpensive multiplex pharmacogenetics assay to simultaneously genotype a panel of 18 clinically actionable variants involved in drug pharmacokinetics/pharmacodynamics.


We designed primers to perform a multiplex PCR assay using a single mix. Primers were labeled by two different fluorescent dye markers (HEX/FAM). Those two labels were used to discriminate alleles, while the size of the PCR fragment allowed identifying a particular polymorphism location. The only instruments needed for the analysis were a thermocycler, a capillary sequencer (3500 Genetic Analyzer Applied Biosystems) and GeneMapperTM software. Polymorphisms of interest were CYP3A4*22, CYP3A5*3, CYP1A2*1F, CYP2C9*2 -*3, CYP2C19*2-*3-*17, VKORC11639G>A, ABCB1 rs1045642-rs1128503-rs2229109-rs2032582, CYP2D6*3-*4-*5-*6-*9.


The assay was repeatable (n=3 data sets), a minimum amount of 10 ng of DNA/ samples was needed. For CYP3A4, CYP3A5 and ABCB1 variants, the method was applied to a validation cohort of 60 samples and genotyping results were consistent with those previously obtained with reference methods. For other variants, analysis of a validation cohort is on-going. The assay is efficient since 96 samples can be genotyped in the same run. Results are available within one working-day. The average cost of reagents and consumable to analyze 10 patients sample is 10 euros.


This robust assay can be easily implemented in laboratories as an alternative to cumbersome simplex assays or expensive multiplex approaches. It could help to widespread access to pharmacogenetics in clinical practice.

Impact of CYP2B6 and CYP2C9 on serum concentration of sertraline in patients carrying a CYP2C19 *1/*1 genotype

Line Skute Bråten1,2, Prof. Espen Molden1,3, Dr. Marianne Kristiansen Kringen1,2

1Center for Psykofarmakologi, Diakonhjemmet Sykehus; 2Faculty of health sciences, Oslo Metropolitan University (OsloMet); 3School of Pharmacy, University of Oslo


Sertraline is a selective serotonin reuptake inhibitor (SSRI) metabolized mainly in the liver by the polymorphic enzyme CYP2C19. Previous studies have indicated that other enzymes such as CYP2D6, CYP2C9, CYP3A4 and CYP2B6 also contribute to the metabolism of sertraline but the effect of these genes has not been extensively studied yet. Therefore, the aim of this study was to investigate the effect of CYP2B6 and CYP2C9 genotype on serum concentrations of sertraline in a large population of psychiatric patients carrying a CYP2C19 *1/*1 genotype.


Patients who had performed both serum concentration measurements of sertraline and CYP2C19 genotyping were retrospectively retrieved from a therapeutic drug monitoring (TDM) database at Center for Psychopharmacology, Diakonhjemmet Hospital. Inclusion criteria were information about the prescribed sertraline dose and blood sampling for sertraline TDM 10-30 hours after the last dose intake. Patients genotyped as CYP2C19*1/*1 were further genotyped for the CYP2B6*6-allele (CYP2B6 c.516G>T (rs3745274) and CYP2B6 c.785A>G (rs2279343)) and CYPC2C9*2 and *3-alleles (rs1799853 and rs1057910, respectively). Dose-adjusted serum concentrations (C/D ratios) of sertraline were compared between the subgroups by linear mixed model analysis using CYP2B6 *1/*1 and CYP2C9*1/*1 as reference.


Overall, a total of 323 patients, representing 518 TDM measurements, were included. Compared with the reference group, the C/D ratio of sertraline was 1.9-fold increased in the CYP2B6*6/*6 subgroup (p < 0.001) and 1.2-fold increased in the CYP2B6*1/*6 subgroup (p = 0.011). A 1.72-fold increase in C/D ratio of sertraline was also detected for patients with CYP2C9*2/*3 or CYP2C9*3/*3 genotypes (p = 0.013), compared with the reference group.


This study shows that CYP2B6 and CYP2C9 have a significant impact on sertraline concentration in patients previously genotyped as CYP2C19*1/*1. Thus, multiple CYP genotypes may affect the clinical response of sertraline.

Escitalopram Metabolite Profiles According to CYP2C19 and CYP2D6 Genotypes

Kristine Hole1, Pari Faraj1,2, Astrid Hermansen1, Espen Molden1,2

1Centre for Psychopharmacology, Diakonhjemmet hospital, Norway; 2Department of Pharmaceutical Biosciences, School of Pharmacy, University of Oslo, Norway


Escitalopram (ESC) is a selective serotonin reuptake inhibitor used in the treatment of anxiety and depression disorders. ESC is metabolized to four main metabolites, and several CYP enzymes are involved. The aim of this study was to investigate the impact of CYP2C19 and CYP2D6 genotypes on the different ESC metabolites in vivo.


Serum samples were collected from a routine therapeutic drug monitoring service at Diakonhjemmet hospital in Oslo. Samples were included from patients who had undergone both CYP2C19 and CYP2D6 genotyping and serum concentration measurement of ESC. The serum samples were analyzed with high resolution Orbitrap MS to measure the concentrations of N-demethyl-ESC, N-didemethyl-ESC, ESC-N-oxide and ESC propionic acid. The metabolite/ESC ratios were compared between genotype groups using Kruskal-Wallis followed by Bonferroni-adjusted Dunn’s tests.


In total, 258 patients were included in the study, 166 women and 92 men. The patients had a median age of 41 years (range 15-99). The metabolic ratio most closely related to CYP2C19 genotype was ESC propionic acid/ESC, which was 46% lower in CYP2C19 IMs and 77% lower in PMs compared with NMs (p ≤ 0.005). CYP2C19 genotype was also associated with N-demethyl-ESC/ESC and N-didemethyl-ESC/ESC metabolic ratios. CYP2D6 genotype was most closely related to the ESC N-oxide/ESC metabolic ratio, which was 34% lower in CYP2D6 IMs and 82% lower in PMs compared with NMs (p < 0.001). CYP2D6 genotype was also associated with N-didemethyl-ESC/ESC metabolic ratio.


CYP2C19 genotype impacts serum levels of the ESC metabolites N-demethyl-ESC, ESC propionic acid and N-didemethyl-ESC in vivo, while CYP2D6 genotype impacts ESC-N-oxide and N-didemethyl-ESC. These metabolic ratios can be applied when investigating mechanisms underlying potential interaction between ESC and other drugs.

How strong is pharmacogenetics in psychiatry? Our experience in clinical practice

Dr. Cristina Montrasio1, Dr. Stefania Cheli1, Prof. Emilio Clementi1,2

1Unit of Clinical Pharmacology, Department of Laboratory Medicine, ASST Fatebenefratelli Sacco, Milan, Italy; 2Unit of Clinical Pharmacology ASST Fatebenefratelli Sacco, Department of Biomedical and Clinical Sciences L. Sacco, Università di Milano, 20157 Milan, Italy

Introduction: The treatment of psychiatric conditions is complicated by the large interindividual variability in exposure and response to psychotropic agents, in part due to genetic factors. There is growing evidence that pharmacogenetic (PGx) testing could help guide drug therapy in this field. At present, a crucial limitation to PGx implementation in clinical practice is the lack of consensus between laboratories on a minimum testing panel of clinical validity and on genotype to phenotype translation. We report our experience on the application of PGx testing in real life settings.

Methods: Genomic DNA was isolated using an automatic DNA extraction system. All genotypes were determined by Real-Time PCR. CYP2D6 genotyping was performed using the INFINITI® CYP450 2D6-BC Assay (AutoGenomics). In case of specific drug, we tested the genes encoding the main metabolizing enzymes. In case of unspecified drug, the analysis of CYP2C19, CYP2D6 and the analysis of CYP1A2, CYP3A4/5, CYP2D6 were performed for antidepressants and antipsychotics, respectively, in line with guidelines developed by the Clinical Pharmacogenetics Implementation Consortium (CPIC), the Dutch Pharmacogenetics Working Group (DPWG) and drug label information provided by the US Food and Drug Administration (FDA).

Results: From 2017 to date, we collected a total of 233 requests for pharmacogenetic analysis of antipsychotic or antidepressant drugs. Every year, antidepressant tests represented about 50% of total requests and SSRIs were the most frequently prescribed (68%). Among antipsychotics, clozapine, haloperidol and aripiprazole were the most requested (80.6%). Many requests were generic for antidepressants (35.2%) and antipsychotics (42.6%), without indication of a specific drug, suggesting in some patients a marked resistance or intolerance to conventional treatments.

Conclusions: Although far from widespread application, pharmacogenetics might represent an extremely useful tool to aim for an individualized therapy, especially in treatment‐refractory cases.

Pharmacogenetics of anti-cancer drugs in real life settings: a monocentric experience

Dr. Cristina Montrasio1, Dr. Stefania Cheli1, Prof. Emilio Clementi1,2

1Unit of Clinical Pharmacology, Department of Laboratory Medicine, ASST Fatebenefratelli Sacco, Milan, Italy; 2Unit of Clinical Pharmacology ASST Fatebenefratelli Sacco, Department of Biomedical and Clinical Sciences L. Sacco, Università di Milano, 20157 Milan, Italy

Personalized treatment in oncology is very important because the drugs used for chemotherapy often require dose adaptation to limit the toxic effects while maintaining optimal efficacy. Pharmacogenetics (PGx) may contribute significantly to optimization of anti-cancer therapies. We present the retrospective analysis of data collected during the last 4 years on the application of pharmacogenetic analysis in the routine clinical practice. In particular, we report the results relatively to dihydropyrimidine dehydrogenase (DPYD) genotyping in case of fluoropyrimidines administration and UDP-glucuronosylstransferase (UGT1A1) genotyping for irinotecan. Besides these two PGx tests, strongly recommended, we also report results from tests for which the level of diagnostic evidence is still under active debate: PGx testing for methotrexate (MTX) and taxanes as potential supporting clinical tools. Single nucleotide polymorphisms (SNPs) in MTHFR, SLCO1B1, ABCB1, ABCC2 genes and in CYP3A4/3A5, CYP2C8, ABCB1 genes are investigated for MTX and taxanes tests, respectively. We collected a total of 2637 requests for PGx analysis, of which 1713 DPYD genotyping, 741 UGT1A1 genotyping, 180 for taxanes and 3 for methotrexate. The requests for DPYD and UGT1A1 tests have doubled from 2019 to date, probably due to increasing evidence of their clinical usefulness. In addition to the four DPYD variants, included in the Clinical Pharmacogenetics Implementation Consortium (CPIC) guidelines, we analyse other two more frequent ones, c.496A>G and c.2194G>A, that especially in a homozygous or compound heterozygous status, seem to play a key role. Our data on taxanes show a possible influence of the CYP3A4*22 and CYP2C8*3/*4 on paclitaxel and nab-paclitaxel toxicity and of the ABCB1 c.3435C>T on taxanes effectiveness. We stress the use of methotrexate PGx tests as “a posteriori” strategy to assist the physician in the problematic treatment of patients with unusual and severe toxicities. PGx tests represent a useful and effective tools to prevent or limit anti-cancer drug-induced toxicity.

Exome sequencing in oncology: a way to obtain relevant pharmacogenetic information.

Simon Verdez1,2, Juliette Albuisson3,4, Yannis Duffourd1,2, Romain Boidot3,4,5, Manon Reda3,5,6, Christel Thauvin-Robinet2,4,7, Jean-David Fumet6, Sylvain Ladoire6, Sophie Nambot2,7, Maxime Luu8, Patrick Callier1,2, Laurence Faivre2,4,7, Francois Ghiringhelli3,4,5,6, Picard Nicolas9

1UF Innovation en diagnostic génomique des maladies rares, Dijon, France; 2UMR1231 GAD, Inserm - Université Bourgogne-Franche Comté, Dijon, France; 3Platform of Transfer in Cancer Biology, Georges François Leclerc Cancer Center - UNICANCER, Dijon, France; 4Genomic and Immunotherapy Medical Institute, Dijon, France; 5Department of Tumour Biology and Pathology, Georges François Leclerc Cancer Center - UNICANCER, Dijon, France; 6Department of Medical Oncology, Georges François Leclerc Cancer Center - UNICANCER, Dijon, France; 7Centre de Référence maladies rares « Anomalies du Développement et syndromes malformatifs », Centre de Génétique, FHU-TRANSLAD, CHU Dijon Bourgogne, Dijon, France; 8Centre d'Investigation Clinique, module Epidémiologie Clinique/Essais cliniques, CHU Dijon Bourgogne, Dijon, France; 9Inserm U1248, service de pharmacologie et toxicologie, université de Limoges, CHU de Limoges, F-87042, Limoges, France.

Introduction: Pangenomic sequencing plays an important role in cancer diagnosis and treatment. In addition, it has the potential to reveal several germline genetic variations with therapeutic impact including polymorphisms in pharmacogenes responsible for adverse drug reactions (ADRs) in relation with chemotherapy, antiemetic or pain treatments used in oncology.

Material and Methods: We evaluate the interest of such pharmacogenetic information in a cohort of 445 solid cancer patients with aggressive diseases or relapses, who previously benefited from exome sequencing (ES) for therapeutic issues. We applied a dedicated pipeline to identify relevant pharmacogenetic variants among a list of 67 variants in 8 genes. Based on patient’s clinical and treatments history, we retained 2 genes with variations of potential interest, namely DYPD (dihydropyrimidine dehydrogenase gene) and CYP2D6. For CYP2D6 we determined bioinformatically the metabolizers status for each patient. We retrospectively analyzed drug plasma concentrations and treatment outcomes in line of the variants identified.

Results: Four gene-drug pairs were analyzed: DPYD/5FU, CYP2D6/opioids, CYP2D6/Tamoxifen and CYP2D6/Ondansetron. Six patients treated with 5-fluorouracil and carrying one level 1A PharmGKB variant in DPYD showed a decreased drug mean clearance compared to non-carriers (p=0.01). All patients (n=5) with CYP2D6 poor or ultra-metabolizers status, experienced ADRs related to opioid therapy, a proportion significantly higher than in normal metabolizers (p=0.02 and p<0.01, respectively). Analysis of proportion between patients carrying one level 1A PharmGKB allele for CYP2D6 and non-carriers for the last two drugs (Tamoxifen and Ondansetron), did not reveal a statistical difference with the risk of ADRs.

Conclusion: Although the study was not designed to provide definitive evidence, it suggests that pangenomic germline sequencing in patients with solid tumor can provide pharmacogenetic information likely to be a useful tool to guide therapeutic decisions in particular for drugs interacting with the CYP2D6 and DPD enzyme.

11:00am - 12:30pmOral Session: TDM in real world
Location: Room C - Angelicum Congress Centre
Session Chair: Ugo de Grazia
Session Chair: Giuliana Cangemi
Room C - Angelicum Congress Centre 

Fosfomycin therapeutic drug monitoring in real-life: development and validation of a LC-MS/MS method on plasma samples

Dr. Sara Baldelli1, Dr. Matteo Cerea2, Dr. Davide Mangioni3, Dr. Alessandra Bandera3, Dr. Dario Cattaneo1

1ASST Fatebenefratelli Sacco, Italy; 2University of Milan; 3Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico

Background: More than 670.000 infections due to bacteria resistance to antibiotics occur each year in the EU, and approximately 33.000 people die as direct consequence of these infections. The paucity of new antibiotics active against drug resistant pathogens brings to a reassessment and reintroduction of already existing therapeutic options such as fosfomycin.

Aim: Develop and validate a rapid UHPLC-MS/MS method for the determination of fosfomycin and clinically apply to plasma patients undergoing fosfomycin therapy.

Methods: Sample preparation involved protein precipitation with [13C3]-fosfomycin benzylamine salt as internal standard. Quantification of fosfomycin was based on MRM (m/z 136.85 >78.87; for IS m/z 139.85 >62.83). A seven-point standard calibration curve was prepared over the concentration range of 2 to 800 mg/L. The present method was successfully applied to patients in therapy with fosfomycin from the IRCCS Ca’ Grande Ospedale Maggiore Policlinico of Milan.

Results: Within- and between-day precision and accuracy, sensitivity, selectivity, dilution integrity, recovery were investigated and the results met the acceptance criteria. Fosfomycin is stable at 4 °C, over a 24-hour and 72-hours storage period (mean concentration variation 2.3%). After three freeze–thaw cycles no significant loss for fosfomycin was observed (mean deviation from fresh QCs -3.6%).

Patients undergoing the routine TDM of fosfomycin were given fosfomycin in combination with other antibiotics. Daily fosfomycin doses ranged from 2 g to 24 g, depending on patient’s weight, renal function and MIC of the pathogen involved. Measured fosfomycin plasma concentrations in patient samples ranged from 7.4 mg/L and 644.6 mg/L (CV: 91.1%).

Conclusions: In critical care patient setting where the optimization of pharmacokinetic parameters has become a clinical necessity, TDM represents an important tool to identify the best fosfomycin dosing in single patients. The sensitive, simple, reproducible and robust method proposed for fosfomycin quantification in plasma can be applied in real life settings.

AUC-based therapeutic drug monitoring of tacrolimus after renal transplant recipients is challenged by real-life data

Dr. Marte T Gustavsen1,2, Dr. Karsten Midtvedt1, Dr. Ida Robertsen2, Prof. Jean-Baptiste Woillard3,4, Dr. Jean Debord3,4, Dr. Rolf A Klaasen1, Dr. Nils Tore Vethe1, Prof. Stein Bergan1,2, Prof. Anders Åsberg1,2

1Oslo University Hospital, Norway; 2University of Oslo, Norway; 3CHU Limoges, France; 4INSERM, University of Limoges, France

Background: With the use of capillary microsamples, limited sampling strategies (LSS) and population pharmacokinetic model derived Bayesian Estimators (popPK-BE), AUC-guided TDM of tacrolimus is clinically applicable. However, most popPK models are based on data from clinical trials, obtained under controlled conditions (i.e. fasting and without concomitant drugs), and may not reflect a real-life scenario. In addition, most data are obtained during the day, not taking tacrolimus circadian variation into account. The aim of the present study was to investigate the PK of twice-daily tacrolimus in a real-life setting.

Methods: Successive 12-hour tacrolimus PK investigations were performed after the morning- and the evening dose both in a fasting-state (±2 hours fasting rule) and under real-life conditions (administered as in the patients every-day life). A total of 31 renal transplant recipients were investigated in the early post-transplant phase (14 repeated the investigations after 1-2 weeks).

Results: Circadian variation was observed under fasting conditions, with 45% higher peak-concentration and 20% higher AUC following the morning dose (P<0.06). Under real-life administration circadian variation was not present: the PK-profiles were flat and comparable after the morning and evening dose. The absorption rate was slower and AUC lower when compared to the fasting morning administration. LSS with samples obtained at 0, 1, and 3 hours predicted AUC well after the fasting morning dose using a previously developed popPK model. Adapting this model to fit the slower absorption shown in real-life data, an LSS using sampling times at 1, 3 and 6 hours predicted AUC well for the fasting evening dose and in the real-life setting.

Conclusion: When administered in a real-life setting tacrolimus show flat PK-profiles and no circadian variation. In a fasting state circadian variation is present. LSS in combination with popPK-BE are able to predict AUC under both conditions.

Pharmacokinetic interaction between bupropion and escitalopram in patients with depression

Dr. Ragnhild Birkeland Waade, Dr. Tore Haslemo

Center for psychopharmacology, Diakonhjemmet Hospital, Oslo, Norway

Introduction: Escitalopram is metabolized primarily by CYP2C19, but also to some extent by CYP2D6 and CYP3A4. Bupropion is a known CYP2D6 inhibitor, and it is speculated whether this antidepressant inhibits other CYP enzymes as well. The aim of the present study was to investigate the impact of bupropion comedication on escitalopram serum concentrations in patients with depression.

Methods: Serum concentration measurements from CYP-genotyped patients treated with the antidepressants escitalopram (n=4025, 7310 samples) and bupropion (n=188, 261 samples) were collected retrospectively from a therapeutic drug monitoring database during an 11-year period. The impact of bupropion comedication on dose-adjusted serum concentrations [i.e., nmol/L/mg/day] of escitalopram was evaluated by multivariable mixed model analysis. We also investigated to what extent genetic variation in CYP2D6 affects the dose-adjusted serum concentrations of escitalopram.

Results: In CYP2C19 and CYP2D6 normal metabolizers (NMs), the mean dose-adjusted serum concentration of escitalopram was 1.7-fold higher in patients comedicated with bupropion (n=22, 33 samples) compared with controls (n=592, 1044 samples) (p<0.001). In comparison, in CYP2C19 NMs not comedicated with bupropion, the impact of being a CYP2D6 poor metabolizer (PM) on dose-adjusted escitalopram serum concentrations was less pronounced and not statistically significant (1.2-fold difference between CYP2D6 PMs (n=96, 188 samples) and NMs (n=592, 1044 samples), p=0,230).

Conclusions: This study indicates that bupropion inhibits the metabolism of escitalopram, and suggests that bupropion also inhibits other enzymes than CYP2D6, possibly CYP2C19.

Effect of Therapeutic Plasma Exchange on the Pharmacokinetics of Itraconazole

Dr. Vincent Seah1, Dr. Sophie Stocker2, Dr. Danijela Kocic1, Dr. Stephanie Reuter3, Prof. Deborah Marriott4,5

1Department of Clinical Pharmacology & Toxicology, St Vincent's Hospital Sydney, Australia.; 2School of Pharmacy, Faculty of Medicine and Health, The University of Sydney, Australia.; 3School of Pharmacy and Medical Sciences, University of South Australia, Adelaide, SA, Australia.; 4Department of Infectious Diseases and Clinical Microbiology, St Vincent's Hospital, Sydney, NSW, Australia; 5School of Medicine, University of New South Wales, Sydney, NSW, Australia.

Introduction. Therapeutic plasma exchange (TPE) is the process of extracorporeal plasma removal to reduce autoantibody levels in specific clinical settings. Patients are frequently immunosuppressed and require antimicrobial treatment or prophylaxis. TPE prescription changes may require drug dose adjustments. Understanding the influence of TPE on the pharmacokinetics of antimicrobials is essential to optimize their effectiveness and achieve therapeutic targets.

Itraconazole is a commonly used antifungal lacking published data regarding clearance during TPE.

Theoretically, drugs with a low Vd (<0.2 L/kg) and highly protein-bound are more susceptible to clearance via TPE. Itraconazole is highly protein-bound (>99%); however it possesses a large Vd. Therefore, the effect of TPE on the pharmacokinetics of itraconazole remains unclear. Our report examines the pharmacokinetics of itraconazole in a patient on regular TPE.

Case description. A 34-year-old Malaysian male presented with anti-MDA5 amyotrophic dermatomyositis. Management included twice-weekly TPE and immunosuppression with rituximab and high dose corticosteroids. Oral itraconazole (200mg BD) was commenced for invasive Aspergillus lentulus infection after voriconazole failed to reach therapeutic concentrations despite dose escalation. Multiple measurements of itraconazole were performed to guide dose adjustment. One illustrative episode is described.

TPE was 142 minutes, 1.1 total body volume exchanged with albumin. Serial venous blood at TPE start, mid-point and completion and multiple points the same day were obtained to determine itraconazole concentrations. Pre-and post-TPE serum itraconazole concentrations were 3956 ug/L and 1850 ug/L respectively - a 53.2% reduction. Waste plasma volume was 3740 mL with itraconazole concentration of 2451 ug/L, giving a calculated clearance of 1.36 L/h via TPE.

An increase in drug levels was observed during the inter-TPE period, probably reflecting redistribution from extravascular compartments.

Conclusion. Despite the large volume of distribution, a significant proportion of itraconazole is removed by plasma exchange. TDM is required to guide dose adjustments during a TPE treatment course.

Hydroxychloroquine blood concentrations in COVID-19 patients decline slowly after treatment cessation

Dr. Simona De Gregori1, Dr. Alessia Ballesio2, Dr. Alessandra Fusco2, Carlo Soffiantini2, Dr. Alessandro Vicentini3, Dr. Annalisa De Silvestri4, Dr. Francesco Falaschi5

1Clinical and Experimental Pharmacokinetics Unit, Fondazione IRCCS Policlinico San Matteo, Italy; 2Department of Internal Medicine, Fondazione IRCCS Policlinico San Matteo, University of Pavia, Pavia, Italy; 3Cardiac Intensive Care Unit, Arrhythmia and Electrophysiology and Laboratory of Clinical and Experimental Cardiology, Fondazione IRCCS Policlinico San Matteo, Pavia, Italy; 4Clinical Epidemiology and Biometry Unit, Fondazione IRCCS Policlinico San Matteo, Pavia, Italy; 5Internal Medicine 2, Fondazione I.R.C.C.S. Policlinico San Matteo, University of Pavia, Pavia, Italy

INTRODUCTION: Hydroxychloroquine (HCQ) has been widely used during the SARS-COV2 pandemic as an antiviral drug. Most previous pharmacokinetic/pharmacodynamic studies on HCQ were conducted in healthy volunteers or patients in chronic therapy. There are no studies on the elimination of HCQ after short treatments. HCQ is known to have a pro-arrhythmic effect through QT interval prolongation, but data in this setting are not conclusive. Our aims are to estimate the time needed for HCQ whole blood concentrations (CHCQ) to drop to a safety level (500 ng/mL) after cessation of a short-term therapeutic cycle and to correlate corrected QT interval (QTc) with CHCQ in this particular setting.

METHODS: We collected blood samples and electrocardiograms of patients who underwent a short-term therapy with HCQ during drug intake and after discontinuation. CHCQ were measured by a HPLC-MS/MS assay and analysed with a linear regression model to estimate the elimination time of the drug after its discontinuation correcting for total dose taken. We analysed through a multivariate analysis QTc correlation with CHCQ.

RESULTS: The Figure shows the correlation between normalized (for drug dose) HCQ concentrations (nCHCQ) and time from treatment cessation. The linear regression model of nCHCQ versus days since the last HCQ dose taken has an upper 95% CI that crosses the concentration of 500 ng/mL 16 days after treatment cessation. QTc prolongation significantly correlates with CHCQi in a model that includes also age, sex and concomitant treatment with QT prolonging drugs.

CONCLUSIONS: The data confirm the long half-life of HCQ and its effect on QT interval even after short treatment courses. We suggest that 16 days after the last administered HCQ dose, physicians could safely introduce (or reintroduce) treatments with a QT prolonging potential.

Therapeutic Drug Monitoring Of Quinidine In Patients With Kcnt1 Genetic Variants

Dr. Sara Cairoli1, Dr. Alessandro Ferretti2, Dr. Nicola Pietrafusa2, Dr. Alessia Vitale1, Dr. Nicola Specchio2, Dr. Carlo Dionisi Vici1, Dr. Bianca Maria Goffredo1

1Department of Pediatric Subspecialties, Division of Metabolism, Bambino Gesù Children's Hospital, IRCCS, Rome, Italy; 2Department of Neuroscience, Rare and Complex Epilepsy Unit, Bambino Gesù Children's Hospital, IRCCS, Rome, Italy


Quinidine is a repurposing drug for patients with Developmental and Epileptic Encephalopathies (DEE) due to KCNT1 genetic variants, with promising good effects. Cardiological toxic effects should be closely monitored. It is therefore essential to determine optimal plasma quinidine concentration for safety and efficacy purposes .


We determined quinidine plasma levels in 3 pediatric patients, all of them carrying KCNT1 genetic variants, using a validated method Liquid Chromatography coupled to Mass Spectrometry (LC-MS/MS). Samples were collected at 12 hours after last administration.


Patient 1 4 years and 6 months (c.337G>A). Epilepsy onset was at 9 months, very soon drug resistant. The dose of quinidine was increased gradually and his seizure frequency decreased by approximately 90% upon reaching the quinidine dose of 25 mg/kg/day. Plasma quinidine levels were between 0.072 and 1.12 μg/mL.

Patient 2 1 year old (c.1283G>A). Epilepsy onset at the age of 2 months. Quinidine was started at the age of 7 months. His seizure frequency decreased by approximately by 50% at a dose of 60 mg/kg/day. Plasma quinidine levels were between 0.14 and 0.26 μg/mL.

Patient 3 1 year and 6 months old (c.862G>A). Epilepsy started at the age of 40 days, resistant to conventional ASMs and ketogenic diet. At the age of 11 months, he started quinidine and seizure frequency decreased by approximately 75% upon reaching the quinidine dose of 40 mg/kg/day. Plasma quinidine levels were 0.007 and 0.02 μg/mL.


Quinidine should be considered as a promising drug for patients with DEE due to KCNT1 genetic variants. Quinidine blood levels were significantly lower if compared with the supposed therapeutic range (1-5 μg/mL), even if we observed an improvement of seizure burden. Plasma quinidine concentrations should be closely monitored and more cases should be analyzed in order to define an appropriate therapeutic range.

Increased accumulation of erythrocyte methotrexate polyglutamates during early phase subcutaneous versus oral methotrexate treatment of rheumatoid arthritis patients

Renske Hebing1,2, Ittai Muller3, Sohaila Mahmoud1, Marry Lin3, Dr. Sandra Heil4, Prof. Willem Lems1,2, Prof. Michael Nurmohamed1,2, Prof. Robert de Jonge3, Dr. Gerrit Jansen2

1Reade, Rheumatology, Amsterdam, The Netherlands; 2Amsterdam UMC, location VUmc, Rheumatology, Amsterdam, Netherlands; 3Amsterdam UMC, location VUmc, Clinical Chemistry, Amsterdam, Netherlands; 4Erasmus MC, Clinical Chemistry, Rotterdam, Netherlands


Assessment of erythrocyte methotrexate (eMTX)-polyglutamates (PGs) levels has been employed as a tool to monitor clinical response of rheumatoid arthritis (RA) patients in the first 3-12 months of treatment. Some MTX-PGs were associated with a lower DAS28 over 9 months. Data from earlier timepoints, MTX-PG6 and per route of administration are unavailable.


In this clinical prospective cohort study, we investigated the pharmacokinetics and -dynamics of eMTX-PG accumulation in RA patients starting MTX. Patients were administered oral (n=24) or subcutaneous (n=22) MTX, mostly according to the COBRA-light schedule (start 10 mg, increased to 25 mg in 8 weeks). At 1, 2, and 3 months after start of therapy, blood was collected and individual MTX-PGs (MTX-PG1 – MTX-PG6) were analysed in erythrocytes at a minimal detection limit of 1 nmol/L, using a validated UHPLC-MS/MS method with labelled internal standards.


46 consecutive patients were included in this study; 76% female, mean age: 57.8 years, BMI: 25.8, 20% smokers, mean baseline DAS28-ESR: 3.5. Notwithstanding marked interpatient variability, patients starting subcutaneous MTX accumulated significantly higher (approximately 2-fold) long chain MTX-PGs (MTX-PG4-6) compared to the oral MTX group at 1 and 2 months, see figure. Similarly, MTX-PG1-6 and MTX-PG3 accumulation was higher in subcutaneous MTX-users at month 1 (p=0.022 and p=0.011) compared to the oral group (median 68.6 nmol/L (IQR:40.5) vs 51.9 (55.6) and 17.4 (11.1) vs 11.2 (15.6).

With nor without corrections for age, baseline DAS28, eGFR, MTX dose (1 month before sampling), smoking and BMI, no significant relation between MTX-PG concentrations and DAS28 was observed.


This study shows the feasibility of measuring eMTX-PGs early in the treatment of RA patients starting MTX and demonstrated significantly higher accumulation of MTX-PGs following subcutaneous versus oral MTX administration. These data will be applied to ameliorate pharmacokinetic models developed with NONMEM.

11:00am - 12:30pmIntersociety Seminar (CME): Pharmacometrics Practical approaches to dose optimization
Location: Room D - Angelicum Congress Centre
Session Chair: Antonello Di Paolo

11.00 am: Opening remarks

11.15 am: Effects of interindividual variability in pharmacokinetics in children. Oscar Della Pasqua

11.25 am: Personalized medicine in children. Anne Smits

11.45 am: Dose optimization of biologics. Erwin Dreesen

12.05 pm: Discussion

12.05 pm: Final considerations

Room D - Angelicum Congress Centre 
12:30pm - 1:00pmThermoFisher workshop: LC-MS/MS Platforms for Accurate, Confident Results and Regulatory-Compliant Workflows
Location: Aula Magna - Angelicum Congress Centre

Moderator: Edward Goucher -  Global Senior Marketing Manager at Thermo Fisher Scientific

- Speaker 1: Dr. Jonas Brand - Head of Department LCMS/HPLC/Metabolism at MVZ Labor Dr.Limbach & Kollegen GbR 

   Presentation title: Implementation of a CE/IVD Clinical LC-MS/MS Analyzer in Routine Diagnostics

- Speaker 2: Dr. Maurizio Suppo - Co-owner & Principal Consultant at Qarad BV European Regulatory Services

   Presentation title: Navigating the IVDR Landscape

Aula Magna - Angelicum Congress Centre 
12:30pm - 1:15pmSaladax workshop: Individualize and optimize busulfan dosing in-house with a rapid end-to-end solution.
Location: Room A - Angelicum Congress Centre

Speakers: Salvatore J. Salamone Sirj Goswami 

Room A - Angelicum Congress Centre 
12:30pm - 2:00pmBusiness meeting - TDM of Biologics
Location: Room B - Angelicum Congress Centre

Coordinators: Annick de Vries, Dirk Jan Moes

Room B - Angelicum Congress Centre 
12:30pm - 2:00pmBusiness meeting - Immunosuppressive Drugs Committee
Location: Room C - Angelicum Congress Centre

Coordinators: Stein Bergan, Satohiro Masuda

Room C - Angelicum Congress Centre 
12:30pm - 2:00pmBusiness meeting - Pharmacogenetics Committee
Location: Room D - Angelicum Congress Centre

Coordinators: Jesse Swen, Nicolas Picard

Room D - Angelicum Congress Centre 
12:30pm - 2:00pmLunch with Exhibitors
Location: Angelicum Congress Centre
Angelicum Congress Centre 
2:00pm - 3:30pmSymposium 3: Update in oncology
Location: Aula Magna - Angelicum Congress Centre
Session Chair: Romano Danesi

02.00 pm: Talk 8: TDM: small molecules vs biologics. Antonello Di Paolo

02.30 pm: Talk 9: Next generation therapeutic antibodies. Josee Golay

03.00 pm: Talk 10: Nano-drug delivery; dream or reality. Fabio Corsi

Aula Magna - Angelicum Congress Centre 
2:00pm - 3:30pmSymposium 4: Clinical toxicology laboratory: broad-spectrum drug screening
Location: Room A - Angelicum Congress Centre
Session Chair: Alain Gaston Verstraete

02.00 pm: Talk 12: High Resolution Mass Spectrometry drug screening in plasma and urine. Alain Verstraete

02.30 pm: Talk 11: Point of Care urine drug screen in supporting ambulatory substance use clinic. Lei Fu

03.00 pm: Talk 13: The LC-Orbitrap HR/MS in the clinical toxicology laboratory: a quality management perspective to broad-spectrum drug screening. Cristiana Stefan

Room A - Angelicum Congress Centre 
2:00pm - 3:30pmOral Session: Miscellaneous 1
Location: Room B - Angelicum Congress Centre
Session Chair: Teun van Gelder

Session Chair: Rossella Gottardo

Room B - Angelicum Congress Centre 

Multiple and sequential quantitative evaluations of SARS-Cov-2 IgG after vaccination: A perspective study in healthcare professionals.

Dr. Elisa Roda1, Dr. Lucia Bernasconi1,2, Dr. Cristina Grazioli1,2, Dr. Benedetta Brolli1,2, Dr. Adam Osuchowski1,2, Dr. Silvia Biserni1,2, Dr. Valentina Negrini1,2, Dr. Paola Baiardi3, Prof. Fabio Blandini4, Dr. Anna Losurdo5, Dr. Luisa Gervasio6, Dr. Francesco Saverio Robustelli Della Cuna6, Dr. Teresa Coccini1, Dr. Davide Lonati1, Dr. Azzurra Schicchi1, Dr. Valeria M Petrolini1, Prof. Carlo Locatelli1

1Toxicology Unit, Pavia Poison Centre-National Toxicology Information Centre, Clinical and Experimental Laboratory, IRCCS Pavia Hospital, ICS Maugeri SpA-SB, Pavia, Italy; 2School of Specialization in Pharmacology and Clinical Toxicology, University of Pavia, Pavia, Italy; 3Scientific Directorate, IRCCS Pavia Hospital, ICS Maugeri SpA-SB, Pavia, Italy; 4Scientific Directorate, Mondino Foundation IRCCS, Pavia, Italy; 5Hospital Pharmacy, IRCCS Pavia Hospital, ICS Maugeri SpA-SB, Pavia, Italy; 6Hospital Pharmacy, Mondino Foundation IRCCS, Pavia, Italy

A perspective study evaluating SARS-Cov-2 IgG in healthcare professionals enrolled from two research hospitals in Pavia would provide the monitoring of effects and adverse reactions after vaccination (carried out with Pfizer Comirnaty). The qualitative and quantitative evaluation of the neutralizing S1-RBD IgG produced in response to vaccination was achieved using three different methods i.e. ELISA, immunoassay and biochip array technology. Specifically, the following diagnostic assays have been employed: Anti-SARS-CoV-2 QuantiVac ELISA (IgG)® Euroimmun, Siemens-Healthineers ADVIA Centaur® SARS-CoV-2 IgG, and Randox SARS-CoV-2 IgG (RBD & NP) array, this latter able to identify both S1-RBD and NP antibodies. More than 1200 subjects were included in the study with a first sampling before vaccination (T0), followed by subsequent evaluations at different time-points: T1, T2, T3 and T4 (30, 90 180 and 360 days, respectively). The early available data concern T0 and T1 blood specimens and were obtained using ELISA and immunoassay methods. Interestingly, these preliminary determinations allowed to identify about a 0.5 percent of non-responders, 20% of low-responders, and 35% of high-responders, also performing a close tracking of adverse effect, which revealed very rare severe events (accounting for not more than 3% of the total ADR). Based on these preliminary results, evaluated in light of the current worldwide Covid-19 pandemic, the neutralizing antibodies monitoring is crucial to identify the duration of the immune antibody response, answering the key question if a concrete relation exists between IgG amount and disease protection/prevention. The monitoring of these biomarkers of effect could be proposed as a fundamental evaluation to be adopted in the existing TDM units, with the main goal to make available worldwide the unique standardized and easy-to-apply measurement able to evaluate the efficacy of the SARS-Cov-2 vaccination.

Evaluation of pre-analytical practices and intraindividual variability of uracilemia for DPD phenotyping: results of a study from the French Group of Clinical Oncological Pharmacology.

Maud Maillard1, Dr. Elodie Gautier-Veyret2, Prof. Jérôme Guitton3, Dr. Bernard Royer4, Dr. Manon Launay5, Dr. Sophie Broutin6, Dr. Céline Narjoz7, Dr. Camille Tron8, Prof. Joseph Ciccolini9, Dr. Damien Richard10, Hugo Alarcan11, Prof. Vincent Haufroid12, Naïma Tafzi13, Dr. Antonin Schmitt14, Dr. Marie-Christine Etienne-Grimaldi15, Dr. Fabienne Thomas1

1Laboratoire de Pharmacologie, Institut Claudius Regaud, IUCT-Oncopole et Centre de Recherches en Cancérologie de Toulouse, Inserm UMR1037, Université Paul Sabatier, Toulouse, France; 2Laboratoire de Pharmacologie, Pharmacogénétique et Toxicologie, CHU Grenoble-Alpes et Université Grenoble-Alpes, laboratoire HP2, INSERM U1042, Grenoble, France; 3Laboratoire de Pharmacologie Toxicologie, CHU de Lyon, Lyon, France; 4Laboratoire de Pharmacologie Clinique et Toxicologie, CHU de Besançon, Besançon, France; 5Laboratoire de Pharmacologie et Toxicologie, CHU de Saint-Etienne, Saint-Etienne, France; 6Département de Biologie et Pathologie Médicale, Service de Pharmacologie, Gustave Roussy, Villejuif, France; 7Assistance Publique des Hôpitaux de Paris, Hôpital européen Georges-Pompidou, Service de biochimie, Paris; 8Laboratoire de Pharmacologie, CHU de Rennes, Rennes, France; 9SMARTc Unit, CRCM Inserm U1068 et Laboratoire de Pharmacocinétique, CHU La Timone, Marseille, France; 10Laboratoire de Pharmacologie et Toxicologie, CHU de Clermont-Ferrand, Clermont-Ferrand, France; 11Service de Biochimie et Biologie Moléculaire, CHRU de Tours, Tours, France; 12Louvain centre for Toxicology and Applied Pharmacology (LTAP), Clinical and Experimental Research Institute (IREC), Université catholique de Louvain, Brussels, Belgium; and Clinical Chemistry Department, Cliniques Universitaires Saint-Luc, Brussels, Belgium; 13INSERM, Université de Limoges, Service de Pharmacologie et Toxicologie, CHU de Limoges, U1248 IPPRITT, Limoges, France; 14Service Pharmacie, Centre Georges-François Leclerc et INSERM U1231, Université de Bourgogne, Dijon, France; 15Laboratoire d’Oncopharmacologie, UPRC EA 7497, Centre Antoine Lacassagne, Nice, France

Background: Determination of dihydropyrimidine dehydrogenase (DPD) activity before fluoropyrimidines (FP) administration is currently mandatory in France and recommended by EMA. In this on-going study, we aimed to refine the pre-analytical recommendations for uracil (U) and dihydrouracil (UH2) plasma determination, which are essential for the reliability of DPD deficiency testing.

Methods: Plasma concentrations of U and UH2 are being collected from 11 francophone laboratories. Different parameters were evaluated: nature of sampling tube, whole blood and plasma stability before and after centrifugation at 4°C and at room temperature (RT), and long-term plasma storage at -80°C. Bias for measured concentrations were calculated from a reference value and acceptability threshold for stability variability was ± 20%. Intra-individual variability (IIV) was assessed in 527 patients sampled on two occasions for DPD-deficiency testing before FP administration.

Results: Preliminary analyses showed that U in whole blood was stable until 4 hours when stored at 4°C (+10% variation on average) but not at RT (+18% after 1h30, +24% after 2h). A similar trend was observed for U in plasma stored at RT (+23% after 1h30) or at 4°C (+12% after 6 hours) after centrifugation. Long-term freezing (24 months) did not affect U (+0.8%) and UH2 (+3.2%). For patients having two samples compliant with pre-analytical recommendations, mean IIV coefficients of variation (CV) reached 23.7% for U and 20.7% for UH2. Among them, 20.3% had a contradictory DPD status (based on U > or < 16 ng/ml). For those having one non-compliant sample, this misclassification rate increases up to 32% and CVs up to 31.6% and 26.3%, respectively.

Conclusions: These preliminary results consolidate the pre-existing recommendations about sample handling for DPD phenotyping. More importantly, these unpublished IIV data may help biologists interpreting discrepancies between U concentrations obtained at two occasions for a given patient.

Short- and long-term effect of gastric bypass and calorie restriction on CYP3A activity in patients with severe obesity

Kine Eide Kvitne

University of Oslo, Norway


It remains uncertain whether potential pharmacokinetic changes following Roux-en-Y gastric bypass (RYGB) are attributed to the surgery-induced gastrointestinal alterations per se and/or the subsequent weight loss. The objective was to disentangle these effects by comparing short- and long-term effects of RYGB and non-surgical calorie-restriction on CYP3A-activity (midazolam as probe).


This non-randomized, single-center trial included patients with severe obesity preparing for RYGB or diet-induced weight loss. Both groups underwent an initial 3-week low-energy-diet (<1200 kcal/day) followed by a 6-week very-low-energy-diet (<800 kcal/day) or RYGB (<800 kcal/day). Patients were followed for 2 years, with four 24-hour midazolam pharmacokinetic investigations (week 0, 3, 9 and year 2) using semi-simultaneous oral and intravenous dosing. The primary outcomes were changes in absolute bioavailability and clearance of midazolam, a standard CYP3A probe drug, within and between groups.


In total, 82 patients (28 males) were included, with a mean±SD body weight of 132±24 kg and 124±23 in the RYGB- (n=41) and diet group (n=41), respectively. The RYGB- and diet group showed similar weight loss at week 9 (13±2.2% vs. 11±3.6%), but differed substantially after 2 years (-30±6.7% vs. -3.1±6.3%). Midazolam absolute bioavailability and clearance were similar in the two groups at baseline. Absolute bioavailability was unaltered after 9 weeks both within and between the groups, but decreased after 2 years (RYGB: 10% [95% CI: 4.0, 16], diet: 7.0% [95% CI: 0.97, 13], mean group difference 0.22% [95% CI: -5.4, 5.8]). No difference in clearance was observed over time within, or between, the groups.


Neither RYGB per se nor the subsequent weight loss impacted absolute bioavailability or clearance of midazolam short-term, suggesting that dose adjustments of CYP3A-substrates may not be needed in the early period after RYGB. The decrease in absolute bioavailability long-term supports the hypothesis that CYP3A-activity recovers following a substantial weight loss.

Influence of green tea flavonoids on the bioavailability of nintedanib: a randomized, cross-over study in patients with pulmonary fibrosis

Dr. GD Marijn Veerman1, Dr. Marlies S Wijsenbeek2, Sanne C van der Werff1, Dr. Jelle R Miedema2, Dr. Esther Oomen-de Hoop1, Dr. Sophie C van der Mark2, Dr. Prewesh P Chandoesing2, Dr. Stijn LW Koolen1,3, Prof. Ron HJ Mathijssen1

1Dept. of Medical Oncology, Erasmus MC Cancer Institute, Rotterdam, the Netherlands; 2Dept. of Pulmonology, Erasmus MC, Rotterdam, the Netherlands; 3Dept. of Hospital Pharmacy, Erasmus MC, Rotterdam, the Netherlands


Nintedanib is an oral tyrosine kinase inhibitor used in fibrotic interstitial lung disease (ILD) and metastatic lung cancer. Drug absorption in the gastrointestinal tract is -with <5% bioavailability- limited. Nintedanib is a substrate of efflux transporter ABCB1 (P-gp) and to a minor extend of CYP3A4. It is known that potent inhibitors of ABCB1 significantly increase nintedanib’s bioavailability. Flavonoids have the potency to influence ABCB1 in vitro and in vivo. One of the most potent flavonoids is epigallocatechin gallate (EGCG), that is highly present in green tea. Hence, we hypothesized that administration of green tea could increase nintedanib uptake. We performed a two-period randomized cross-over study to investigate the influence of green tea capsules on the exposure of nintedanib in patients with fibrotic ILD.


Patients were randomized between phases A>B and B>A. In both phases nintedanib was administered twice daily after a meal and each phase lasted 7 days. In phase B patients also received green tea capsules (GTC; 500mg BD, >60% EGCG) which were taken with water, 30 min prior to a meal. Pharmacokinetic sampling (12h) was performed at day 7 of each phase. Primary endpoint was change in geometric mean for the area under the curve (AUC0-12h). A linear mixed model was used to analyze AUCs and maximal concentration (Cmax).


In the first 16 patients, the nintedanib AUC0-12h was 23% lower (95%CI: -33 to -10%; P=0.002) in phase B (with GTC) compared to phase A. Cmax did not differ significantly between phases; -20% (95%CI: -40 to +8%; P=0.131). No differences in toxicities were observed.


Exposure to nintedanib significantly decreased when administered 60 min after GTC (500mg BD, >60% EGCG) for 7 days. This is a clinically relevant interaction according to the FDA, which could impair treatment efficacy and therefore should be avoided.

Assessing the binding interaction of polystyrene sulfonate with amitriptyline in healthy volunteers, a cross-over design: The BIND study

Ilona Prins-Can1, Inge Regina Francina van Berlo-van de Laar1,2, Marieke Zeeman3, Cornelis Gerard Vermeij4, Esther van 't Riet5, Katja Taxis2, Franciscus Gerhardus Antonius Jansman1,2

1Deventer Teaching Hospital, The Netherlands, Department of Clinical Pharmacy; 2University of Groningen, The Netherlands, Groningen Research Institute of Pharmacy (GRIP),Unit of Pharmacotherapy, -Epidemiology &-Economics; 3Deventer Teaching Hospital, The Netherlands, Department of Geriatrics; 4Deventer Teaching Hospital, The Netherlands, Department of Nephrology; 5Deventer Teaching Hospital, The Netherlands, Research Department

Introduction: Polystyrene sulfonate is used for binding potassium in patients with chronic kidney disease (CKD). Because of its binding properties, it can potentially bind other medications and thereby decrease their bioavailability and effectiveness. Amitriptyline, often used by CKD patients for neuropathic pain, shows significant binding to polystyrene sulfonate in vitro. To assess the clinical relevance of this interaction, in vivo information is needed.

Aim: To determine the effect of polystyrene sulfonate on the exposure of amitriptyline when taken simultaneously, in healthy volunteers.

Methods: We performed a prospective cross-over study in nine healthy volunteers. Participants were 18 years of age or older, did not use any medication and had no known allergy to amitriptyline or polystyrene sulfonate. Participants visited the hospital on two separate days. Once they received a single dose of amitriptyline 50mg and once a single dose of both polystyrene sulfonate 15g and amitriptyline 50mg. After drug intake six blood samples were collected; at 2, 3, 4, 5, 6 and 8 hours. Cmax and AUC0-8h of amitriptyline were determined on both days. A Wilcoxon signed-rank test was performed to determine whether Cmax and AUC0-8h values were different between days. A two sided p-value <0.05 was considered significant.

Results: Of the nine participants included, eight participants completed both visits to the hospital. Mean Cmax of amitriptyline was 35.61µg/l (±9.23µg/l) when taken alone, compared to 9.25µg/l (±3.19µg/l) when taken with polystyrene sulfonate (p=0.012). Mean AUC0-8h of amitriptyline was 168.20h*µg/l (±33.79h*µg/l) when taken alone and 45.78h*µg/l (±18.64h*µg/l) when taken with polystyrene sulfonate (p=0.012).

Conclusion: These results show a 75% decrease in exposure to amitriptyline when taken concomitantly with polystyrene sulfonate, compared to amitriptyline alone. Therapy efficacy of amitriptyline is likely to be decreased when taken concomitantly with polystyrene sulfonate. Patients using both amitriptyline and polystyrene sulfonate should apply a staggered administration regimen.

Multi-Site Evaluation of Immunoassays for Antipsychotic Drug Measurement in Clinical Samples

Prof. William A Clarke1, Bruce Slayer1, Casey Hussey1, Dr. JoAnn Gardiner2, Prof. Kamisha Johnson-Davis3,4, Dr. Michael C Milone2, Dr. Salvatore J Salamone5

1Johns Hopkins University School of Medicine, United State of America; 2Perelman School of Medicine of the University of Pennsylvania, United State of America; 3University of Utah Health Sciences Center, Department of Pathology, United States of America; 4ARUP Institute for Clinical and Experimental Pathology, United State of America; 5Saladax Biomedical, Inc., United States of America

Background: Atypical antipsychotic drugs are frequently used in the treatment of serious mental illness (SMI), specifically in schizophrenia and bipolar disorder. Adherence to these prescribed drug regimens is a challenge to successful treatment. For some of the more common drugs in this class, novel turbidimetric immunoassays have been developed for TDM to aid in the management of patients prescribed these drugs.

Methods: Immunoassays for aripiprazole, clozapine, olanzapine, paliperidone, quetiapine, and risperidone were set up at two centers: Johns Hopkins Hospital (JHH) on the Roche Cobas® c501, and the Hospital of the University of Pennsylvania (HUP) on the Beckman AU480. Assay precision, limit of quantification (LOQ), functional sensitivity, linearity, and recovery were assessed. Remnant clinical samples were obtained from a reference laboratory (ARUP), and immunoassay results were compared with those obtained by LC-MS/MS.

Results: Imprecision at both sites for all analytes and concentrations tested was <8%. The manufacturer’s LOQ was confirmed for each assay, and the functional sensitivity for each assay was found to be lower than the LOQ. All assays were found to be linear over the measuring range, with recoveries ranging from 91-123%. For method comparison, Deming regression slopes were found to be between 0.84 and 1.28.

Conclusion: The immunoassays evaluated here are suitable for quantifying drug concentrations to be used in TDM for all 6 drugs. Commercialization of these assays will enable increased access for TDM in psychiatric patient management.

Validation of Point of Care Tests (POCT) of Antipsychotic Drugs Aripiprazole and Risperidone

DeWayne M. Davenport, Mary Rose Hilaire, Nathan A. Siegfried, Jodi Courtney, Irina Baburina, Regina V. Gill, Megan S. Konrath, Tasfia Habiba, Barrie A. Rogers, Adrienne Drisgill, Salvatore J. Salamone

Saladax Biomedical, Inc. United States of America

Introduction: Treatment guidelines, clinical support tools, and expert consensus guidelines recommend measurement of antipsychotic blood levels, as adherence to antipsychotic therapy is critical to positive treatment outcomes. POCT provides immediate, objective drug level results during the patient visit so clinicians can provide personalized treatment plans to combat problems such as nonadherence, pharmacokinetic interactions (drug-drug, changes in lifestyle), poor response, or unacceptable toxicity. The aims of the study were to validate whole blood (WB) POCT assays for total aripiprazole (aripiprazole + major metabolite dehydroaripiprazole; ARI), and total risperidone (risperidone + major metabolite 9-hydroxyrisperidone; RSP).

Methods: Automated homogenous immunoassays (MyCare Psychiatry Assay Kits for Total Aripiprazole (CE Mark) and Total Risperidone (CE Mark), were modified to test analyte levels in WB with the MyCare Insite, a small, portable analyser used for rapid POCT. Analytical performance of MyCare Insite ARI Test (ARI-POCT) and MyCare Insite RSP Test (RSP-POCT) was evaluated according to CLSI guidelines with 3 reagent lots. Testing was performed using buffered controls and spiked WB samples.

Results: The ARI-POCT was linear from 56 – 1,000 ng/mL. Within-laboratory precision over the measuring range was 7 – 15%. The limit of quantification (LoQ) was 120 ng/mL. No bias from hematocrit (35 – 50%) was observed for either ARI or RSP. The recovery over the measuring range was 90 – 110%. ARI-POCT Insite results from 41 patient samples were compared to a validated LC-MS/MS; Passing-Bablok regression statistics: 0.1.08 (slope), -23.44 (intercept), and 0.96 (R). The RSP-POCT was linear from 10 - 120 ng/mL. LoQ was 16 ng/mL. Within-laboratory precision over the measuring range was 9 – 18%. Recovery from 15 to 120 ng/mL was 92–115%.

Conclusions: Precise, accurate blood levels of the antipsychotics aripiprazole and risperidone can be reported in less than 8 minutes using capillary human whole blood samples at point of care.

2:00pm - 3:30pmOral Session: Clinical Toxicology - 1
Location: Room C - Angelicum Congress Centre

Session Chair: Roberta Pacifici
Session Chair: Caterina Grassi

Room C - Angelicum Congress Centre 

National early warning system on New Psychoactive Substances

Dr. Simona Pichini

Analytical Pharmacotoxicology Unit Head, National Centre on Addiction and Doping, Istituto Superiore di Sanità

The globalization, the evolution of e-commerce and the growing popularity of New Psychoactive Substances (NPS), facilitated the development of a wide illegal market in constant expansion. The dynamic nature of this phenomenon has led to an evolution in the prevention and monitoring of NPS trafficking within the European Union (EU). The European legislative system has been amended with the aim of creating a faster and more effective regulatory system to tackle NPS diffusion and ban their sale and circulation. At the end of 2008, in compliance with the European Council Decision 2005/387/JHA, the Anti-Drug Policies Department of the Presidency of the Council of Ministers activated in the Italian National Early Warning System (NEWS) to promote a rapid exchange of information on NPS between Italy and the EU.

In June 2016, the National Center for Addiction and Doping (CNDD) of the Istituto Superiore di Sanità (ISS) was designated to organize and manage NEWS, supported by the Poison Center of the Maugeri Scientific Clinical Institutes of Pavia to manage NPS clinical and toxicological aspects, the Research Unit of Forensic Toxicology of the Sapienza University of Rome to manage biotoxicological and analytical/technical aspects, and the Central Antidrug Service to ensure law enforcement and disclose seizures and risky situations in National territory. More than 200 Collaborative Centres across the country joined the system, such as law enforcement agencies and their laboratory networks, analytical toxicology laboratories, forensic toxicology laboratories, poison centres, health and care systems, universities and research institutes, medicines regulatory authorities.

The activities, the challenges, the help to collaborative centres in term of pure standards of NPS, methodologies, quality control exercise and etc. will be described to explain how to detect and assess any potential threats, and identify and implement any response measures that might be required for NPS health and social threats.

Evaluation of a new immunoassay method for the evaluation of traditional and designer benzodiazepines in urine

Prof. Luca Morini, Brian Rossi, Francesca Freni, Giancarlo Collo, Dr. Cristiana Stramesi, Claudia Vignali

Department of Public Health, Experimental and Forensic Medicine, University of Pavia, Italy

Introduction: The aim of the study was: a) to evaluate a new assay, not yet commercially available (ARK™ HS Benzodiazepine II), on a Siemens Advia 1800 system in comparison to the currently used one (Siemens EMIT II PLUS); to correlate the results obtained from the two immunoenzymatic methods with the ones achieved through a liquid chromatographic tandem mass spectrometric (LC-MS/MS) confirmation test. Methods: The ARK™ HS Benzodiazepine II assay was tested and validated using etizolam as reference substance (calibration curve set in the range 100-3000 ng/mL), while the EMIT II PLUS curves (100-1000 ng/mL) were performed by means of kits prepared with lormetazepam. A simple liquid/liquid extraction was applied to 100 µL urine samples before injection into LC-MS/MS system. The confirmation test was qualitatively validated for a total of 41 benzodiazepines. Quantitative LC-MS/MS determination (10 calibration points within the range 10-7500 ng/mL), was performed for 22 molecules, dependently on the reference standards available in the lab. Results: All the validation tests performed on the new assay satisfied the acceptance criteria. The LC-MS/MS was fully validated, in accordance to the international guidelines. The three methods were applied to real urine samples, before and after hydrolysis, collected from healthy subjects, patients under treatment, and postmortem cases. Preliminary results confirmed a good qualitative correlation between the two immunoassays; the ARK™ HS Benzodiazepine II provided a higher sensitivity in detecting some benzodiazepines such as lorazepam and 7-aminoclonazapem, while a lower sensitivity for alprazolam was observed. In particular, the new method appeared to yield to better cross-reactivity for glucuronide and 7-amino metabolites. Conclusion: The new immunoassay proved to be reliable and highly sensitive in detecting benzodiazepines and metabolites in urine, independently on the molecule, thus suggesting a good cross-reactivity for the most widespread BZs. The preliminary results should now be confirmed on a larger population.

Identification of the metabolites of four new designer benzodiazepines with the help of molecular networks.

Michiel Notelaers1, Prof. Alain Gaston Verstraete1,2

1Department of diagnostic sciences, Ghent university, Belgium; 2Department of laboratory medicine, Ghent university hospital, Belgium

Introduction: More than a thousand new psychoactive substances (NPS) were reported by the UNODC in December 2020. In the same year, etizolam, flualprazolam and flubromazolam, 3 designer benzodiazepines, were reported as the top 3 NPS in toxicologic cases by the UNODC. Difficulties with legislation, identification and detection are a worldwide problem. The objective of our research was to determine whether or not Feature Based Molecular Networking (FBMN) gives additional value in NPS metabolite detection, exploration and identification when compared to literature and in silico calculation.

Method: Urine samples that contained one of the target designer benzodiazepines etizolam (3 urine samples), flubromazolam (1), bromazolam (1) or flualprazolam (7) were used and prepared with and without ß-glucuronidase. Positive electrospray ionization was used in full-scan (70-700 m/z) on a Thermo Q Exactive with data-dependent acquisition based on an inclusion list with metabolites compiled primarily from literature and the BioTransformer website. The raw data were converted using MassIVE and imported into MzMine for pre-processing using mass detection, chromatogram building and deconvolution, isotopic peak grouping, alignment, and filtering. The data were exported to the Global Natural Products Social Molecular Networking (GNPS) platform for FBMN. The output was visualised using Cytoscape.

Results: A total of 7 bromazolam metabolites were found compared to 8 found in the literature. For flubromazolam 3 metabolites were found (6 in literature). Seven flualprazolam metabolites were identified and 1 tentatively (7 in literature). For etizolam 9 metabolites were suspected, and 1 metabolite was identified, while only 2 have been reported in the literature. In samples containing more than 1 benzodiazepine, the networks overlapped, e.g. when they differed by one substituent.

Conclusion: GNPS and FMBN might prove useful for the detection and differentiation of potential metabolites of newly discovered NPS which can be further isolated and identified.

In vitro toxicokinetics of three novel stimulants of the (2-aminopropyl)benzothiophene type including cellular uptake, metabolic fate, isozyme mapping, and monoamine oxidase inhibition

Dr. Lea Wagmann1, Fabian Frankenfeld1, Dr. Simon D Brandt2, Ole Jensen3, Prof. Jürgen Brockmöller3, Prof. Markus R Meyer1

1Saarland University, Homburg, Germany; 2Liverpool John Moores University, Liverpool, UK; 3University Medical Center Göttingen, Göttingen, Germany

Introduction. The availability of amphetamine-based new psychoactive substances has attracted the attention of clinical and forensic toxicologists. The psychostimulant character extends to sulfur-based compounds such as methiopropamine with a thiophene ring. Three novel (2-aminopropyl)benzothiophene (APBT)-based stimulants, by name 3-APBT, 5-APBT, and 6-APBT, have been included in the present study to investigate their in vitro toxicokinetics.

Methods. Cellular uptake was studied using HEK293 cells overexpressing the organic cation transporter OCT1 and intracellular concentrations were compared to those in an empty vector control cell line. In vitro metabolites were identified in incubations with pooled human liver S9 fraction containing cosubstrates for most phase I and II reactions, while individual incubations with 11 human phase I isozymes were used for isozyme mapping. Monoamine oxidase (MAO) inhibition was tested in incubations with human MAO-A and MAO-B. All analyses were performed by liquid chromatography-high resolution tandem mass spectrometry.

Results. The three APBT isomers were identified as OCT1 substrates with preliminary Km values of 2.0 (6-APBT), 2.2 (5-APBT), and 2.7 µM (3-APBT). Hydroxylations were identified as main metabolic phase I steps and were found to be mainly catalyzed by CYP2D6, CYP3A4, and CYP3A5. Glucuronidation and sulfation could be detected as subsequent phase II metabolic transformations. The APBT isomers inhibited MAO-A with IC50 values of 0.4 (5-APBT), 0.6 (6-APBT), and 4 µM (3-APBT), but also MAO-B with IC50 values between 23 (5-APBT) and 49 µM (6-APBT).

Conclusions. The toxicokinetics of three novel APBT-based stimulants was extensively studied in vitro. OCT1 is expected to play an important role in their hepatic uptake. 5-APBT and 6-APBT were identified as potent MAO-A inhibitors and a clinical relevance after intake cannot be excluded. The presented results may help toxicologists in the interpretation of future cases.

A LC-MS/MS method for the detection of two phosphatidylethanol homologues as long-term biomarkers of alcohol exposure

Prof. Kamisha L. Johnson-Davis1,2, Dr. Nkemakonam C. Okoye1, Dr. Chad D. Moore3

1University of Utah / ARUP Laboratories, United States of America; 2ARUP Institute for Clinical and Experimental Pathology, United States of America; 3Sports Medicine and Research Testing Laboratory, United States of America

Introduction: Alcohol use disorders are a global healthcare problem. Long-term alcohol biomarkers are needed to monitor alcohol abstinence for transplant pre-qualification, drug rehabilitation programs and to assess in utero alcohol exposure. Phosphatidylethanol (PEth) is a group of ~48 glycerophospholipid homologues that are formed from ethanol and phospatidylcholine through a transphosphotidylation reaction catalyzed by phospholipase D. PEth is incorporated into RBCs with a half-life of 4-10 days following ethanol consumption. The purpose of this project was to develop and validate a LC/MS-MS method for quantifying the predominant homologues PEth 16:0/18:1 (POPEth) and PEth 16:0/18:2 (PLPEth) in blood.

Methods: Fifty µL of internal standards (POPEth-d5 and PLPEth-d5) was added into 100 µL aliquot of whole blood samples and vortexed. Protein precipitation was achieved using a 10% (v/v) isopropanol (IPA) / acetonitrile (ACN) solution. Upon centrifugation, the supernatant was extracted and buffered with 15 mM ammonium acetate. Sample extracts were analyzed by LC-MS/MS using an Agilent 6470 triple quadrupole mass spectrometer coupled to an Agilent LC system equipped with two 1260 Infinity II binary HPLC pumps, a 1260 Infinity II autosampler, and a 1260 Infinity II thermostat column compartment. The mass spectrometry method was performed using negative mode ionization with MRM acquisition. Analytical validation of the LC-MS/MS method was performed for linearity, accuracy, imprecision, sensitivity, specificity and ion suppression/matrix effect, in accordance to CLSI guidelines.

Results: The analytical measurement range, 10 – 2000 ng/mL, was linear with R2 of 0.999 for both PLPEth and POPEth. The within run and total imprecision was < 5% CV for the low (20 ng/mL), medium (200 ng/mL) and high QC (1000 ng/mL). Results for accuracy and method comparison experiments met the bias criteria of ±20%.

Conclusion: The LC-MS/MS method for the detection of POPEth and PLPEth in blood demonstrated high sensitivity and specificity for detecting alcohol exposure.

Severe intoxications associated with analytically confirmed use of NBOMe and 2C phenethylamines

Dr. Azzurra Schicchi1, Dr. Davide Lonati1, Dr. Valeria M Petrolini1, Dr. Eleonora Buscaglia1, Dr. Antonella Valli2, Dr. Giulia Scaravaggi1, Dr. Pietro Papa2, Prof. Carlo Locatelli1

1Toxicology Unit, Pavia Poison Center-National Toxicology Information Centre, Clinical and Experimental Laboratory, IRCCS Pavia Hospital, Istituti Clinici Scientifici Maugeri, Pavia, Italy; 2Laboratory of Analytical Toxicology, Clinical Chemistry Service, IRCCS Policlinico San Matteo Foundation, Pavia (Italy)

NBOMe are potent hallucinogens, active at sub-milligram dose, and are N-2-methoxybenzyl analogues of the respective 2C-X substituted phenethylamines. We report the prevalence and clinical features of analytically confirmed intoxications by NBOMe and 2C phenethylamines in a six-months-period among the cases referred to our Poison Centre (the reference Unit for suspected/confirmed NPS poisoning in Italy). Six cases of phenethylamines intoxication were evaluated (age ranging from 16 to 27 years-old; 4 males). Four patients declared assumption of LSD or another hallucinogenic substance, and two an unknown substance of abuse: two patients attended a rave-party. Urine quantitative analyses were carried out by LC-MS/MS, and all patients resulted positive for 25I-NBOMe and 2C-I at concentrations of 16 and 6 ng/ml, 2.8 and 2.5 ng/ml, 3.2 and 2.5 ng/ml, 43.4 and 11.1 ng/ml, 1.8 and 3.5 ng/ml, 4.25 and 1.9 ng/ml, respectively. Their urine samples tested positive also for THC (5 cases), amphetamines (2), MDMA (2) and ketamine (1). The most represented clinical manifestations were severe psychomotor agitation (83%), hallucinations, seizures, and rhabdomyolysis (50% each one), tachycardia, coma and hyperthermia (33% each one), confusion and mydriasis (17% each one); no lethal cases were registered. The treatment consisted in sedation with benzodiazepines (3 cases), intubation and respiratory support (3 cases). All our cases tested positive for 25I-NBOMe and 2C-I: this may be due to the metabolism of NBOMe to 2C analogues, or to the simultaneous abuse of 25I-NBOMe and 2C-I. All our cases have been reported to the Italian Early Warning System. Clinicians should be aware of the toxicity due to these NPS: NBOMe assumption can be suspect and analytically confirmed in patients reporting a recent use of LSD or other hallucinogens.

Acknowledgements. Study carried out with the support of Antidrug Policy Department, Italian Presidency of the Council of Ministers.

Prognostic value of liver and kidney function parameters and their correlation with the ratio of urine-to-plasma paraquat in patients with paraquat poisoning

Prof. Hualin Cai, Shuangyang Zhang, Prof. Bikui Zhang

The Second Xiangya Hospital and Institute of Clinical Pharmacy, Central South University, China

Introduction: Acute paraquat poisoning resulting from multiple organ failure usually has a high mortality rate. Liver and kidney, serve as the key oranges of paraquat detoxification and elimination. We intended to explore the prognostic value of liver and kidney function parameters, and further evaluate their correlation with urine-to-plasma paraquat ratio.

Methods: We retrospectively analyzed the records of patients with paraquat poisoning admitted to four centers in China from January 2018 to December 2019. All patients received the same drug protocols including immune-suppressing drugs, glucocorticoid, vitamin C, stomach- and liver-protective drugs and some antimicrobials. The initial plasma samples for determining paraquat concentrations were collected before taking other drugs and surgical treatments.

Results: Thirty-three patients with full medical records were included and 16 of them died from paraquat poisoning. Paraquat plasma concentrations (Fig. a) were significantly associated with urine paraquat concentrations (R2 = 0.19, P < 0.01) (Fig. b). The ratio of urine-to-plasma paraquat is negatively correlated with ALT (r = −0.94, P = 0.02) and BUN (r = −0.82, P = 0.03), which represent liver/kidney function, respectively. For receiver operating characteristic curve (Fig. c), new findings were as follows: ALT had an area of 0.85 (95% CI, 0.76-0.94) and the optimal cutoff value was 25.60 mmol/L (sensitivity, 56.67%; specificity, 89.57%). The BUN had an area of 0.99 (95% CI, 0.98-1.00) and the optimal cut-off value was 26.60 mmol/L (sensitivity, 96.67%; specificity, 100.00%). The ratio of urine-to-plasma paraquat had an area of 0.94 (95% CI, 0.81-1.00), and the optimal cut-off value was 20.97 (sensitivity, 88.89%; specificity, 100.00%).

Conclusions: The ratio of urine-to-plasma paraquat dynamically reflecting the excretion speed of paraquat, and ALT, BUN (Fig. d) which both reflect organ injuries and the capability of paraquat excretion of an individual patient, should be considered simultaneously with urine-to-plasma paraquat ratio for prediction of survival.

3:30pm - 4:20pmAfternoon tea&Coffee, exhibition and poster viewing
Location: Angelicum Congress Centre
Angelicum Congress Centre 
4:30pm - 6:00pmSymposium 5: TDM of biologics in inflammatory bowel diseases
Location: Aula Magna - Angelicum Congress Centre
Session Chair: Dario Cattaneo

04.30 pm: Talk 14: Role of proactive versus reactive TDM of anti-TNF drugs or standard of care in inflammatory bowel diseases. Dario Cattaneo

05.00 pm: Talk 15: Value of TDM of ustekinumab and vedolizumab in inflammatory bowel diseases. Gionata Fiorino

05.30 pm: Talk 16: How do we implement TDM to maximise treatment success? What target concentrations should we aim for? Anne Strik

Aula Magna - Angelicum Congress Centre 
4:30pm - 6:00pmSymposium 6: Role of metabolite measurement in risk assessment of xenobiotic compounds: analytical and toxicokinetic aspects
Location: Room A - Angelicum Congress Centre

Session Chair: Nicolas Venisse

04.30 pm: Talk 17: The complexity of human metabolism and how it can be explored through in vitro and in vivo approach. Markus R Meyer

05.00 pm: Talk 18: Bisphenol: weakness of deconjugation methods and its impact on biomonitoring. Roy Gerona

05.30 pm: Talk 19: The role of metabolite determination in risk assessment. Antoine Dupuis

Room A - Angelicum Congress Centre 
4:30pm - 6:00pmOral Session: Anti-infectives: antibiotics
Location: Room B - Angelicum Congress Centre
Session Chair: Antonello Di Paolo
Session Chair: Bianca Maria Goffredo
Room B - Angelicum Congress Centre 

A regression model to predict augmented renal clearance in critically ill obstetric patients and effects on vancomycin treatment

Prof. Lian Tang

The Affiliated Suzhou Hospital of Nanjing Medical University, China, People's Republic of

Background: Augmented renal clearence (ARC) risk factors and effects on vancomycin (VCM) of obstetric patients were possibly different from other populations based on pathophysiological characteristics. Our study was to establish a regression model for prediction of ARC and the effects on VCM treatment in critically ill obstetric patients.

Methods: We retrospectively included 427 patients, grouped into ARC and non-ARC patients. Logistic regression analysis was used to analyze the factors related to ARC. Patients who received VCM therapy were collected. Vancomycin population pharmacokinetic (PPK) model was used to calculate pharmacokinetic parameters.

Results: Of the 427 patients, ARC was present in 214 patients (50.1%). According to the results of Logistic regression analysis, we established the above nine-variable prediction regression model and calculated the predicted probability. ROC curve showed that the predicted probability of combined weight, albumin, height and gestational age had better sensitivity (93.8%) and specificity (54.2%) as well as the maximal AUC (AUC = 0.866). 41 cases received VCM; 17 cases (41.5%) had ARC. The initial trough concentration in ARC patients was lower than in non-ARC patients (p = 0.032). Comparing the predicted trough concentration of two VCM PPK models with the measured trough concentration, correlation coefficients (r) were all more than 0.8 in the ARC group and non-ARC group. AUC was significantly decreased in the ARC group (p = 0.026; p = 0.043), and CL increased in the ARC group (p<0.001; p<0.001) when compared with the non-ARC group.

Conclusions: ARC is a common state in critically ill obstetric patients. The regression model of nine variables had high predictive value for predicting ARC. Vancomycin PPK model had good predictive performance for predicting trough concentrations of obstetric patients. Pharmacokinetic parameters of VCM are different in ARC patients, which results in enhanced VCM clearance and decreased trough concentration.

Therapeutic drug monitoring of vancomycin in Coagulase Negative Staphylococci (CoNS) bacteremia: a real-life experience in a tertiary care Pediatric hospital.

Dr. Carolina Saffioti1, Dr. Alessio Mesini1, Dr. Marcello Mariani1,2, Dr. Luca Grasselli1, Dr. Claudia Sette1, Dr. Sebastiano Barco1, Angelo Maffia1, Laura Barbagallo1, Dr. Elio Castagnola1, Dr. Giuliana Cangemi1

1IRCCS Istituto Giannina Gaslini, Italy; 2Università degli Studi di Genova, Italy

Introduction: Methicillin-resistant CoNS bloodstream infections are an important cause of morbidity and mortality in paediatics. Treatment could be challenging especially for difficulties occurring during vancomycin therapy. It is necessary to monitor drug plasma levels to assess efficacy and toxicity. While large amount of data are available for vancomycin PK/PD in MRSA infections, data of effective AUC/MIC ratio in CoNS bacteremia are lacking.

Methods: data on vancomycin plasma levels during CoNS bacteremia occurring in a 1 year period at Istituto Giannina Gaslini Pediatric Hospital were collected. Episodes were considered significant after at least 2 consecutive positive blood cultures, for all patients aged >28 days, while a single culture was considered sufficient for neonates. AUC/MIC ratio, based on average AUC, was calculated for each episode. Vancomycin was usually administered at the dose of 40mg/kg/day in continued infusion, after a loading dose of 10-15mg/kg. In patients aged <28 days dose was adjusted on pre-menstrual age and weight. Vancomycin was measured by commercial immunoassay with automatic analyzer (Roche C501; Roche, Milan, Italy).

Results: A total of 46 episodes were collected from January to December 2020. Median AUC/MIC was 337 (IQR 178-462). AUC/MIC was >300 in 24 (52%) episodes, and > 400 in 18 (39%). Average AUC was > 600 in 3 episodes (6.5%) and > 400 in 18 episodes (39%). Median albumin was 3432 mg/dl (IQR 3332-3678).

Conclusions: the standard dose of vancomycin resulted inadequate to guarantee the more effective AUC/MIC ratio in the majority of our patients, in the presence of normal serum albumin levels. This result could be at least partially influenced by the limited number of episodes and the inhomogeneity of the sample deserving further studies.

Optimisation of antibiotic dosing regimens in neonatal and paediatric sepsis: Impact of a model-based approach to antimicrobial stewardship

Dr. Raffaele Simeoli1, Dr. Bianca Maria Goffredo1, Dr. Salvatore D'Agate2

1Bambino Gesù Children's Hospital IRCCS, Italy; 2UCL - LONDON'S GLOBAL UNIVERSITY

Introduction and objectives: Sepsis in neonates and children is still associated with high mortality and morbidity. Nevertheless, first-line treatment includes high-dose empirical antibiotic therapy, ignoring the implications of interindividual differences in pharmacokinetics and bacterial susceptibility. There is an urgent need to ensure personalised interventions and improve antibiotic stewardship in paediatric intensive care units , enhancing the probability of clinical cure. The aim of the current investigation was to evaluate treatment performance, as assessed by PKPD indices (i.e., time above the minimum inhibitory concentration (T>MIC) and probability of target attainment (PTA)) for currently used doses and dosing regimens of amoxicillin, ampicillin, ceftazidime, ceftriaxone in patients with suspected sepsis.

Methods: A total of 59 blood samples from 29 patients enrolled into an observational, prospective, single-centre study in patients with at “Bambino Gesù Paediatric Hospital IRCCS, Rome were used for the purpose of this analysis. Using nonlinear mixed effects modelling in conjunction with priors for pharmacokinetic parameter distributions obtained from previously published population pharmacokinetic models, a post-hoc analysis was performed to describe the individual concentration vs. time profiles and secondary parameters, including the area under the concentration vs. time curve (AUC). The probability of target attainment was then calculated to evaluate the suitability of the antibiotic doses and dosing regimens.

Results: Our results show that the currently recommended doses of ampicillin, ceftazidime and ceftriaxone resulted in exposure levels that ensure efficacy targets of 50%fT>MIC for MICs up to 32 mg/L in most patients. By contrast, the standard dosing regimen of amoxicillin failed to achieved the desired PTA.

Conclusions: This investigation shows how a model-based approach can be used to enhance data integration and support antimicrobial stewardship. In addition, simulations provide insight into dose(s) and dosing regimens that are required to improve clinical response and prevent antibiotic overuse in this vulnerable population.

The possible impact of CYP3A genotypes in linezolid underexposure

Dr. Sara Baldelli, Dr. Stefania Cheli, Dr. Dario Cattaneo, Dr. Cristina Montrasio

ASST Fatebenefratelli Sacco, Italy

Background: In adults, linezolid is licensed at the standard dose of 600 mg twice daily. However, the exposure to linezolid is characterized by a large inter-individual variability. Age, renal function and body weight explain this variability only to a limited extent and a considerable portion of it remains unexplained. Linezolid is predominantly metabolized through oxidation and cytochrome P450 enzymes seem not to play a significant role. However, some drug-to-drug interactions potentially related to the polymorphic CYP3A and ABCB1 genes have been reported in patients treated with linezolid, suggesting an involvement of the genetic background on linezolid disposition.

Aim: Investigate the role of individual genetic background in linezolid underexposure.

Methods: We conducted a retrospective analysis of routine requests for therapeutic drug monitoring of linezolid and pharmacogenetic analysis carried out by our Unit.

Eligible population included patients who received linezolid therapy for ≥72 h, with a blood sample collected 12 h after the last drug intake.

Linezolid plasma concentrations were classified as “below”, “within” or “above” reference range (2-8 mg/L). Genetic polymorphisms for ABCB1, CYP3A4 and CYP3A5, and POR were investigated

Results: 72 out of 190 patients (37.9%) had linezolid trough concentrations within the therapeutic range; 52.4% and 9.4% had, respectively, linezolid trough concentrations above and below therapeutic range. Age significantly correlated with drug exposure. No association was found between ABCB1 polymorphisms and linezolid exposure. A significant relationship was instead found in the frequency distribution of linezolid trough concentrations when looking at the rs776746 polymorphism of the CYP3A5 gene. The multinomial logistic regression analysis confirmed that CYP3A5 expressers (*1/*1 and *1/*3) were found to be more at risk to be underdosed respect to CYP3A5 non expressers (*3/*3).

Conclusion: Our data suggest that CYP3A5, but not ABCB1, polymorphisms significantly affect linezolid disposition, putting patients carrying allelic variants at higher risk to linezolid underexposure.

Daptomycin and methadone: a possible drug – drug interaction

Dr. Simona De Gregori1, Dr. Annalisa De Silvestri2, Dr. Antonella Bartoli1, Dr. Riccardo Albertini3, Prof. Raffaele Bruno4, Dr. Elena Seminari4

1Fondazione IRCCS Policlinico San Matteo - Pavia, Italy - Clinical and Experimental Pharmacokinetics Unit; 2Fondazione IRCCS Policlinico San Matteo - Pavia, Italy - Clinical Epidemiology and Biometry Unit; 3Fondazione IRCCS Policlinico San Matteo - Pavia, Italy - Clinical Chemistry Laboratory; 4Fondazione IRCCS Policlinico San Matteo - Pavia, Italy - Infectious Diseases I, Department of Medical Sciences and Infectious Diseases

INTRODUCTION: Staphylococcus aureus is the predominant pathogen causing infective endocarditis (IE) and injection drug use due to opioid abuse is one of its leading causes. This infection is severe and still associated with a significantly high in-hospital mortality. Daptomycin administration in combination with β-lactams is strongly recommended, but its interaction with methadone, a drug widely used in the maintenance of opioid use disorder, is not yet well understood.

METHODS: Five intravenous drug users (IDUs) in treatment with methadone under medical supervision and hospitalized for IE were administered with a Daptomycin daily dose of 12 mg/kg. All patients were examined by echocardiography for confirmation of IE. After five days of treatment, their exposure (AUC: µgxh/mL) to Daptomycin was calculated. The target AUC0-24/MIC was 666 whereas a safety threshold concentration (Cmin) was chosen <24.3 µg/mL. Pharmacokitetic results were compared with those described in literature for healthy volunteer treated with the same daily dose for 14 days.

RESULTS: The calculated average CMAX and AUC0-24 were lower than expected for healthy volunteers and statistically different (p<0.001 and p=0.001, respectively). Cmin was always lower than 24.3µg/mL ( : 8.7µg/mL). Conversely, the mean Daptomycin plasma clearance normalized for the body weight (CLwp) and the volume of distribution at the steady state (Vd,ss) were both higher than expected (p=0.001 and p<0.001, respectively). Median AUC0-24/MIC was 1337.

DISCUSSION: Daptomycin pharmacokinetics in 5 IDUs treated in combination with methadone were significantly different from those reported in healthy volunteers at the same antibiotic daily dosage. Low calculated CMAX AUC0-24 could predict therapeutic failures, whereas changes in Vd,ss and CLwp may modify the antibiotic concentration at the target site, affecting the drug’s efficacy. Median AUC0-24/MIC =1337 was due to a very low MIC. We suggest Daptomycin daily dose higher than 12 mg/kg for IDUs on methadone treatment and a rigorous TDM.

Comparison of dalbavancin pharmacokinetics in peripheral blood mononuclear cells (PBMCs) and plasma in a clinical real-life context.

Dr. Jacopo Mula1, Dr. Amedeo De Nicolò1, Dr. Miriam Antonucci1, Dr. Elisa De Vivo1, Dr. Jessica Cusato1, Dr. Alessandra Manca1, Dr. Alice Palermiti1, Dr. Alice Ianniello1, Dr. Valeria Avataneo1, Dr. Giacomo Stroffolini2, Dr. Giuseppe Cariti2, Prof. Giovanni Di Perri2, Prof. Antonio D'Avolio1

1University of Turin, Department of Medical Sciences, Laboratory of Clinical Pharmacology and Pharmacogenetics, Amedeo di Savoia Hospital, Turin (Italy).; 2University of Turin, Department of Medical Sciences, Unit of Infectious Diseases, Amedeo di Savoia Hospital, Turin (Italy)

Dalbavancin (DBV) is a lipoglycopeptide antibiotic with high efficacy and tolerability against Gram+ infections: its effectiveness is partially due to its long half-life. Nevertheless, little is known about DBV intracellular pharmacokinetics (PK) in the real-life.

For this reason 7 patients were enrolled, treated with one (n=3) or two (n=4, 2 weeks delay) high 1500 mg doses. Peripheral blood mononuclear cells (PBMCs) and plasma concentrations were evaluated through a modified version of UHPLC-MS/MS CoQua Lab Kit-System Antibiotics® before infusion, end-of-infusion, 1h post-infusion and then weekly. PBMCs were isolated through Cell Preparation Tubes, while cell counts and mean cell volumes were determined through an automated counter. Intracellular concentrations were compared with plasma concentrations: AUCs (0-2w and 0-∞) and half-lives (t1/2) for both doses were calculated through Phoenix WinNonLin ® (Certara) software on mean concentration data.

DBV mean terminal t1/2 resulted longer in PBMCs than in plasma both for the first and second dose (696h and 611h in PBMC vs. 473h and 336h in plasma), in accordance with mean intra-PBMCs AUC0-∞: 73123 h*mg/L and 78524 h*mg/L for PBMCs vs. 50313,8 h*mg/L and 55077 h*mg/L for plasma, for the first and second dose, respectively. As expected, AUC0-2w in plasma and PBMCs were both higher after the second dose: 33133 h*mg/L vs 39472,8 h*mg/L in PBMCs and 31307,5 h*mg/L vs 35733,9 h*mg/L in plasma. Mean PBMCs/plasma AUC0-2w ratios were 1.05 and 1.10 for the first and second dose, respectively, indicating a progressive accumulation within cells.

Mean Cmax in PBMCs and plasma were 371.1 and 378.8 mg/L for the first dose and 320.5 and 385.4 mg/L for the second dose, respectively. These data suggest a slightly delayed Cmax in PBMCs and a longer permanence time of DBV within cells, confirming the proposed mechanism of tissue redistribution at the basis of its extremely long terminal halflife.

Dalbavancin pharmacokinetics with single or dual infusions in patients with soft-tissues or osteoarticular infections.

Dr. Amedeo De Nicolò1, Dr. Giacomo Stroffolini2, Dr. Miriam Antonucci1, Dr. Jacopo Mula1, Dr. Jessica Cusato1, Dr. Alice Palermiti1, Dr. Alessandra Manca1, Dr. Giuseppe Cariti2, Prof. Stefano Bonora2, Prof. Giovanni Di Perri2, Prof. Antonio D'Avolio1

1Laboratory of Clinical Pharmacology and Pharmacogenetics, Department of Medical Sciences, University of Turin, Italy; 2Unit of Infectious Diseases, Department of Medical Sciences, University of Turin, Italy

Dalbavancin (DBV) is a potent long-acting lipoglycopeptide active against Gram+ bacteria with high efficacy and excellent tolerability. Nowadays, little is known about its pharmacokinetic/pharmacodynamic (PK/PD) profile, particularly in the osteoarticular setting and with two high 1500 mg infusions.

Two groups of patients treated with DBV were enrolled, receiving one or two high doses (1500 mg vs 2 x 1500 mg weekly) respectively. Plasma DBV concentrations were measured through a UHPLC-MS/MS analytical Kit (CoQuaLab) at the end of infusion, after one hour and then weekly. AUCs and half-lives (t1/2) were calculated through Phoenix WinNonLin ® (Certara) software. Then, AUC0-2w/MIC and the estimated T>MIC were calculated according to the breakpoint-MIC for susceptible microorganisms (0.125 mg/L) and adjusted by 93% protein binding.

Ten patients were enrolled: 3 with soft-tissues infections (ABSSSI) and 7 with osteoarticular infections. Six and 4 patients received 1 and 2 infusions, respectively. Mean Cmax after the first and second dose were 355.7 mg/L and 376.1mg/L, respectively. DBV concentrations showed multiphasic PK, with an extremely long terminal t1/2 (554.3 h). Mean observed AUC0-2w for the single dose and double dose groups were 20590 and 56010 mg/L*h, respectively.

Mean protein-binding-adjusted AUC0-2w/MIC were 20590 and 31366 in the single and dual dose groups, respectively; mean estimated T>MIC were 17.3 w and 20.5 w, respectively. There were no adverse events; Two patients were not cured, both with single dose and osteoarticular infections and slightly lower mean AUC0-2w/MIC (21406 vs 27863, P=0.057)

By a theoretical point of view, our data highlight very high AUC0-2w/MIC both with single and dual infusions, abundantly higher than the previously suggested cut-off value (nearly 1000 for ABSSSI), even considering protein binding. This suggests the need for dedicated PK/PD targets in the osteoarticular setting. The evidence of high T>MIC is reassuring in terms of prophylaxis from re-infections or systemic spreading.

Belgian experience with a flucloxacillin absorption test to guide antibiotic treatment.

Nick Verougstraete1, Franky Buyle2, Pieter De Cock2, Jerina Boelens1, Sanne De Smet3, Diana Huis in 't Veld3, Alain Verstraete1

1Department of Laboratory Medicine, Ghent University Hospital, Ghent, Belgium; 2Department of Pharmacy, Ghent University Hospital, Ghent, Belgium; 3Department of Internal Medicine and Infectious Diseases, Ghent University Hospital, Ghent, Belgium

Background: Flucloxacillin is the first-choice antibiotic to treat severe methicillin-sensitive Staphylococcus aureus infections. Intravenous (IV) therapy is typically followed by a course of oral flucloxacillin enabling discharge of the patient from the hospital. Given the high inter-individual variability in absorption after oral administration, a switch from IV to oral flucloxacillin therapy can only be justified after documenting adequate absorption by performing an oral absorption test (OAT). In this study we evaluated the use of an OAT in a routine clinical setting.

Methods: An oral test dose of 1g flucloxacillin was given after a two-hour fasting period and blood samples were taken at baseline and 45 and 90min after administration, to estimate peak levels. Continuous infusion was not interrupted for the test. If flucloxacillin was administered via an intermittent infusion, one IV administration was replaced by the oral test dose. Absorption was considered sufficient if the increase after 45 or 90min was ≥10 mg/L from baseline. Flucloxacillin concentrations in plasma samples were determined by an UHPLC-HRMS (Q-Exactive, ThermoFisher) method after protein precipitation.

Results: OAT results were obtained from 90 adult patients (40 females/50 males) with a median age of 60.5 years (IQR 49.3-69.0). In 66% of the patients the maximal increase in concentration reached the target of 10 mg/L. For 73% of the patients the maximal increase of flucloxacillin concentration was obtained after 45min (median 19.8 mg/L, IQR 13.4-30.2) and for the other 27% the maximal increase was obtained after 90min (median 20.5 mg/L, IQR 15.5-40.1). The patients with adequate absorption were switched to oral flucloxacillin administration, for the poor absorbers IV therapy was prolonged or switched to another oral antibiotic.

Conclusions: An adopted OAT (protocol based on Dijkmans et al.) has been successfully implemented and applied in a Belgian tertiary hospital to guide antibiotic treatment.

4:30pm - 6:00pmOral Session: Immunosuppressants
Location: Room C - Angelicum Congress Centre
Session Chair: Florian Lemaitre
Session Chair: Massimo Baraldo

Session Chair: Åsberg Anders

Room C - Angelicum Congress Centre 

Contribution of the Bayesian estimate of the area under the curve in the optimization of calcineurin inhibitors's therapeutic monitoring

Dr. Habiba Fetati1, Dr. Fatma Boudia1, Dr. Abelghani Chaib1, Dr. Nadjet Mekaouche1, Dr. Asmaa Memou1, Dr. Imene Ouden1, Dr. Kada Djoudad2, Dr. Brahim Bendifallah2, Prof. Faiza Zerdoumi2, Prof. Houari Toumi1

1Faculty of Medecine, University Oran1 Ahmed Ben Bella, Oran, Pharmacovigilance Department of the Hospital and university Establishment of Oran, Algeria; 2Nephrology Departement of the Hospital and university Establishment HUE of Oran, Algeria

Background: Calcineurin inhibitors (CNIs) are immunosuppressive drugs that have been shown to significantly improve the survival rate of grafts in organ transplants. The residual concentration(C0) is the exposure index most commonly used in therapeutic monitoring of these drugs. However, the Bayesian estimator approach to assess the area under the curve (AUC) appears to have better reliability for therapeutic optimization.

The adjective is to assess the contribution of the calculation of the AUC of ICNs by the Bayesian method in the TDM of kidney transplant recipients at the Hospital and University Etablishment of Oran in Algeria.

Materials and methods: A one-year study of 35 renal transplant patients treated with cyclosporine (CSA) and tacrolimus (Tac) who benefited from a TDM based on the evaluation of Co and AUC. The AUC estimate was made after measuring three concentrations (Co, C1 and C 3) on the CHU platform "ABIS et sites pour l'individualisation posologique en transplantation"of CHU Limoge, France.

Results and discussion : A total of 153 Co evaluated including 72 Co of Ciclosporin and 81 Co of Tacrolimus. 51 AUC0-12h estimated by Bayesian method. The results showed a good C0 / AUC correlation: CSA r² = 0.78 and r² = 0.87 for the Tac. 53% of AUC0-12h corresponding to normal CSA's Co were outside the therapeutic range while only 20% in the case of TAC. 32% of dosage requests were in the context of dosing adjustment, 14% risk of rejection and 5 % in the case of CNIs switch.

Conclusion: The Bayesian estimate of the AUC of ICNs should be considered a posteriori as a tool that can help solve clinical problems and improve the management of patients treated with these drugs.

Keywords: Ciclosporin, Tacrolimus, kidney transplantation, STP, C0, AUC, Bayesian method.

Avoiding tacrolimus under- and overexposure with a dosing algorithm for renal transplant recipients: a single arm prospective intervention trial

Marith I. Francke1,2,3, Louise M. Andrews4,5, Hoang Lan Le4, Jaqueline van de Wetering1,2, Marian C. Clahsen-van Groningen2,6, Teun van Gelder7, Ron H.N. van Schaik8, Bronno van der Holt9, Brenda C.M. de Winter2,4, Dennis A. Hesselink1,2

1Department of Internal Medicine, Division of Nephrology and Transplantation, Erasmus MC, University Medical Center Rotterdam, the Netherlands; 2Rotterdam Transplant Group; 3Netherlands Institute for Health Sciences; 4Department of Hospital Pharmacy, Erasmus MC, University Medical Center Rotterdam, Rotterdam, The Netherlands; 5Department of Hospital Pharmacy, Meander Medical Center, Amersfoort, The Netherlands; 6Department of Pathology, Erasmus MC, University Medical Center Rotterdam, Rotterdam, The Netherlands; 7Department of Clinical Pharmacy and Toxicology, Leiden University Medical Center, Leiden, the Netherlands; 8Department of Clinical Chemistry, Erasmus MC, University Medical Center, Rotterdam, the Netherlands; 9Department of Hematology, Erasmus MC Cancer Institute, Rotterdam, the Netherlands

Introduction: Bodyweight-based tacrolimus dosing followed by therapeutic drug monitoring is standard clinical care after renal transplantation. However, after transplantation, a meager 38% of patients are on target at first steady state and it can take up to 3 weeks to reach the target tacrolimus pre-dose concentration (C0). Tacrolimus underexposure and overexposure is associated with an increased risk of rejection and drug-related toxicity, respectively. To minimize subtherapeutic and supratherapeutic tacrolimus exposure in the immediate post-transplant phase, a previously developed dosing algorithm to predict an individual’s tacrolimus starting dose was tested prospectively.

Methods: In this single-arm, prospective, therapeutic intervention trial, 60 de novo kidney transplant recipients received a tacrolimus starting dose based on a dosing algorithm instead of a standard, bodyweight-based dose. The algorithm included cytochrome P450 (CYP) 3A4 and CYP3A5 genotype, body surface area and age as covariates. The target tacrolimus C0, measured for the first time at day 3, was 7.5-12.5 ng/mL.

Results: Between February 23, 2019 and July 7, 2020, 60 patients were included. One patient was excluded because of a protocol violation. On day 3 post-transplantation, 34 of 59 patients (58%, 90% CI 47%-68%) had a tacrolimus C0 within the therapeutic range. Markedly subtherapeutic (<5.0 ng/mL) and supratherapeutic (>20 ng/mL) tacrolimus concentrations were observed in 7% and 3% of the patients, respectively. Biopsy-proven acute rejection occurred in three patients (5%).

Conclusion: Algorithm-based tacrolimus dosing leads to the achievement of the tacrolimus target C0 in as many as 58% of the patients on day 3 after kidney transplantation.

Drug transporters expression on PBMC membrane is associated with tacrolimus intracellular concentration

Gwendal Coste, Dr. Jonathan Chemouny, Dr. Pauline Houssel-Debry, Prof. Mikael Roussel, Dr. Caroline Jézéquel, Dr. Camille Tron, Dr. Michel Rayar, Dr. Fabien Robin, Prof. Cécile Vigneau, Dr. Marie-Clémence Verdier, Prof. Eric Bellissant, Prof. Karim Boudjema, Prof. Bruno Laviolle, Dr. Florian Lemaitre

Rennes University Hospital, France

Measuring tacrolimus (TAC) its whole blood concentrations (Cmin) is mandatory in liver and kidney transplant recipients (LTR and KTR). However, while having adequate Cmin, some patients still experience rejection/adverse events, highlighting the limitation of the actual monitoring. As TAC exerts its effect inside T cells, monitoring intracellular instead of whole blood concentrations is an appealing strategy. Some drug transporters expressed at the cells membrane could be major factors of TAC diffusion into peripheral blood mononuclear cells (PBMC).

The aim of the study was to explore the relationship between TAC diffusion into PBMC and cell membrane expression of four drug transporters (P-glycoprotein (P-gp), Multidrug resistance-associated protein 2 (MRP-2), Concentrative nucleoside transporter 3 (CNT-3) and Equilibrative nucleoside transporter 1(ENT-1)).

Sixty stable LTR and KTR were included in the study. TAC Blood and PBMC concentrations were determined by liquid chromatography coupled with tandem mass spectrometry. P-gp, MRP-2, CNT-3 and ENT-1 expression at the PBMC surface was evaluated using flux cytometry. Correlation between TAC PBMC/blood ratios and transporters expressions were explored using correlation tests.

TAC PBMC/blood ratios were not different between LTR and KTR and the results were then analyzed altogether. P-gp, CNT-3 and ENT-1 expressions show positive correlation with TAC PBMC/blood ratio. Correlation was stronger for ENT-1 (R2=0.31,p<0.0001) than for P-gp and CNT-3 (R2=0.17,p=0.001 and R2=0.14,p=0.003, respectively). MRP-2 expression was not correlated with PBMC/blood ratio(figure 1).

In this study, TAC diffusion into PBMC increased as a function of membrane expression of ENT-1, CNT-3 and P-gp. ENT-1 expression seems to be the strongest driver of TAC diffusion, which is consistent with its role as an influx transporter. Findings for P-gp might be explained by an induction of its expression when intracellular TAC concentration increases. ENT-1 and CNT-3 influx effects on TAC diffusion would then outweigh the P-gp efflux effect.

Intracellular tacrolimus monitoring: intra-patient medium-term variability in a cohort of pediatric patients

Dr. Alice Palermiti1, Dr. Amedeo De Nicolò1, Dr. Michele Pinon2, Dr. Jessica Cusato1, Dr. Alessandra Manca1, Dr. Jacopo Mula1, Dr. Miriam Antonucci1, Dr. Valeria Avataneo1, Dr. Elisa Delia De Vivo1, Dr. Alice Ianniello1, Dr. Pierluigi Calvo2, Prof. Antonio D'Avolio1

1University of Turin, Amedeo di Savoia Hospital, Italy; 2Regina Margherita Children's Hospital, Turin, Italy

Tacrolimus (TAC) is a calcineurin-inhibitor representing one of the most used immunosuppressants in organ transplantation. Immunosuppressants require careful monitoring, as they are characterized by drug-drug interactions, significant inter- and intra-patient variability and a narrow therapeutic index: particularly, high TAC concentrations are associated with adverse events, whereas low levels increase organ rejection risk. Therapeutic Drug Monitoring (TDM) has proven to reduce the incidence of these negative events.

TAC is currently quantified in whole blood, despite it acts within lymphocytes: therefore, studying TAC penetration in peripheral blood mononuclear cells (PBMCs) may represent a substantial improvement for future TDM. For these reasons, the aim of this study was to assess the intracellular TAC concentrations in PBMC and their medium-term intra-patients variability in pediatric patients who underwent liver transplantation assuming constant absolute TAC doses.

Thirty-nine patients were enrolled, 12 patients excluded due to changes in TAC posology, 27 were finally included. TAC intra-PBMCs quantification was performed through a validated method in UHPLC–MS/MS method coupled with automated online solid-phase extraction and the resulting concentrations were tested for their variability over a period of 2-8 months. TAC intracellular concentrations by weight-adjusted dose varied in a not-significant manner (p=0.683), with a median deviation of 1.7% (IQR -14.0% – +25.1%): the median deviation was -7.8% (IQR -17.5% – +25.2%) if the weight-adjusted dose was not considered (p=0.517). Significant correlations were observed between TAC intracellular levels and the weight-adjusted doses (r=0.660; p<0.001) and between the first and second measurement (r=0.780; p<0.001).

Concluding, these data highlight a slight, yet not negligible, intra-individual variability in the TAC intracellular exposure; the slightly lower differences observed with weight-based adjustment suggest patients growth as a contributor to this variability. Further studies are needed to translate this evidence to the clinical practice and to confirm the clinical usefulness of intra-PBMC TAC concentration monitoring.

Clinically significant drug‐drug interaction between tacrolimus and fluconazole in stable renal transplant recipient

Dr. Jiake He1,2,3, Chenglong Yin2, Hailang Liu2, Hua Zou2, Jingsheng Ma2, Wentao Yang2, Dr. Qigen Li2, Lin Zhong2, Prof. Xijing Chen3

1Department of Pharmacy, The Second Affiliated Hospital of Nanchang University, Nanchang, People's Republic of China; 2Department of Organ Transplantation, The Second Affiliated Hospital of Nanchang University, Nanchang, People's Republic of China; 3Clinical Pharmacokinetics Laboratory, School of Basic Medicine and Clinical Pharmacy, China Pharmaceutical University, Nanjing, People's Republic of China

Introduction: Clinical use of fluconazole against fungal infections in renal transplant patients is complicated by the profound and unpredictable nature of drug-drug interactions (DDIs).

Methods: We reported a case of tacrolimus-fluconazole DDI in stable renal transplant recipient and analyzed the onset, mechanism, magnitude, duration and clinical characteristics of this DDI. An individualized regimen was adapted by therapeutic drug monitoring of tacrolimus trough concentration and genetic screening of CYP3A5 6986 A>G and ABCB1 3435 C>T.

Results: A 38-year-old woman experienced a 9.13-fold increase in dose-normalized tacrolimus trough level (trough concentration/weight-normalized daily dose) and an 87 % decrease in weight-normalized daily dose (daily dose/body weight) in the treatment of documented Candida albicans esophagitis by fluconazole. After discontinuation of fluconazole for 161-day, a 26 % reduction in weight-normalized daily dose was required to maintain therapeutic exposure. This patient was homozygous for the CYP3A5 6986 G allele (designated as CYP3A5*3/*3) and ABCB1 3435 T allele and therefore was expected to lack CYP3A5 and P-gp activity. Thus, she experienced an enhanced magnitude and prolonged duration towards the inhibitory effects of fluconazole.

Conclusion: Oral fluconazole has a more significant impact on its drug interactions with tacrolimus than intravenous fluconazole. Proactively gene screening of CYP3A5 6986 A>G and ABCB1 3435 C>T in organ transplant recipients could help in determining the potency of DDI and facilitating tacrolimus dose adjustment in real clinical setting.

Does inflammation impact tacrolimus metabolism in kidney transplant recipients?

Dr. Florian Lemaitre, Gwendal Coste, Dr. Camille Tron, Dr. Jonathan Chemouny

Rennes University Hospital, France


There is more and more evidence of an impact of inflammatory state on drugs metabolism. Thus, pro-inflammatory cytokines has been shown to downregulate cytochrome activity leading to a decrease in drugs biotransformation. Tacrolimus is altogether the major immunosuppressive drugs, a drug with a narrow therapeutic range and a remarkably good substrate for cytochrome P450 enzymes. Its metabolism could then be influenced by inflammatory state.

The aim of this study was to explore the impact of inflammation on tacrolimus pharmacokinetics.


Every consecutive kidney transplant recipients with at least one measurement of C-reactive protein (CRP) hospitalized in our institution was included in this retrospective monocenter study. To evaluate the impact of inflammation on metabolism and to account for patients difference in basal metabolism, we calculate the ratio of the maximum tacrolimus concentration over dose ratio (MaxC/D) over the minimun concentration over dose ratio (MinC/D) during the patient's hospital stay. We finally explore the relationship between the maximum CRP expressed as quartiles and the MaxC/D over MinC/D ratio using a Kruskal-Wallis test.


Two-hundred and ninety-seven hospital stays in 195 patients were collected. Patients in the first CRP quartile (CRP below 16.6 mg/L) have lowest Max over Min C/D ratio than patients in the third (CRP between 58.8 and 141.3 mg/L) and in the fourth (CRP above 141.3 mg/L) quartile. Mean Max over Min C/D ratio were respectively 1.48 ± 0.51; 1.87 ± 2.04; 1.99 ± 0.99 and 3.92 ± 6.13 (p<0.0001).


Inflammation might induce a decrease in tacrolimus metabolism leading to an increase in the risk of patients overexposure. Inflammation parameters could be used as a complementary tool to therapeutic drug monitoring of tacrolimus in kidney transplantation.

Utility of Mycophenolic Acid Therapeutic Drug Monitoring in pediatric hematopoietic stem cell transplantation patients in the first month after transplant, using a Limited Sampling Strategy developed for renal pediatric patients

Dr. Francesco Lo Re1, Dr. Antonella Zucchetto2, Dr. Natalia Maximova3, Prof. Massimo Baraldo1,4

1Clinical Pharmacology and Toxicology Institute, “S. Maria della Misericordia” University Hospital Friuli Centrale, Udine, (Italy); 2Scientific Directorate, Centro di Riferimento Oncologico di Aviano (CRO), IRCCS, (Italy); 3Institute for Maternal and Child Health-IRCCS Burlo Garofolo, Trieste, Italy; 4Chair of Pharmacology, Department of Medical Area (DAME), UNIUD (Italy)


Mycophenolate mofetil (MMF), an ester pro-drug of Mycophenolic Acid (MPA), is an immunosuppressive drug used as a prophylactic regimen in hematopoietic stem cell transplantation (HSCT) to prevent the developing of graft-versus-host disease (GVHD). Analogously to solid organ transplantation, MPA exposure, in terms of its Area Under the 12h time concentration Curve (AUC0-12h) correlates with treatment outcome. The therapeutic range has been established (30-60 mg×h/L), but regarding pediatric HSCT there is a lack of exposure–response data, and only few Limited Sampling Strategies (LSSs) has been developed.

The aim of this study is to demonstrate the capability of a LSS developed for renal pediatric patients for the execution of Therapeutic Drug Monitoring (TDM) in pediatric HSCT patients.


In this retrospective study 30 HSCT pediatric patients were enrolled. The patients underwent TDM, executed with the following LSS: MPA AUC0-12h= 18.6+4.3×C0+0.54×C0.5+2.15×C2. MPA abbreviated AUC0-12h values were checked at baseline and at 15.9 (±4.3) days after the first TDM. The difference between MPA AUC0-12h values from baseline to after the TDM, was tested by Mann Whitney test, considering as statistically significant p values of a 2-sided Test < 0.05.


At baseline 13 patients were under the therapeutic range, 16 patients were in the correct range, and 1 patient was over the range. After the TDM only 3 patients were under the therapeutic range and 27 patients were in the correct range. The median value of AUC0-12h (mg×h/L) from baseline vs after TDM was: 30.70; Interquartile range (IQR): 26.50 to 40.47 vs. 39.41; IQR: 35.80 to 42.90 (p = 0.004).


These results suggest the capability of a renal transplant LSS to reach an adequate MPA exposure in a group of HSCT patients, evidencing the pivotal role of TDM in the first period after HSCT. Future studies are needed to confirm these results.

Mycophenolate dose adjustment in paediatric transplantation patients: Impact of a model-based dosing algorithm

Dr. Bianca Maria Goffredo1, Sara Cairoli1, Raffaele Simeoli1, Marco Dionisi2, Oscar della Pasqua3

1Bambino Gesù Children's Hospital IRCCS, Italy; 2Istituto per le Applicazioni del Calcolo, National Research Council, Rome, Italy; 3UCL - LONDON'S GLOBAL UNIVERSITY

Introduction and objectives: The immunosuppressant effect of Mycophenolate mofetil and mycophenolate sodium is caused by mycophenolic acid (MPA), which is the active component released in plasma. Even though MPA prodrugs are recognised to be well tolerated, there is large inter- and intraindividual variability in the dose-exposure relationship. In fact, the incidence rate of acute rejection in paediatric renal transplant patients is relatively high ( 15 -35%). Using a model-based approach the present study aimed to assess the influence of covariate factors on the pharmacokinetic variability of MPA and develop an algorithm for optimisation of the dosing regimen in paediatric patients receiving solid organ transplantation.

Methods: This was a retrospective, observational study in 103 paediatric solid organ transplant patients. Using sparse therapeutic monitoring (TDM) data, a nonlinear mixed effects modelling approach was implemented in conjunction with priors parameter distributions from a population pharmacokinetic model previously developed for MPA. Evaluation of covariate factors affecting disposition parameters was performed using stepwise forward inclusion/backward elimination procedures. Following the assessment of the predictive performance of the final model, simulations were performed to identify an algorithm for optimised dose adjustment.

Results: Systemic exposure to mycophenolate/MPA can be described by a two-compartment model. Inclusion of body weight and organ function appear to be significant covariate factors on clearance, despite considerable inter-occasion variability in absorption. Dosing adjustment using longitudinal data and baseline covariates yields an increased proportion of patients reaching the desired target therapeutic range.

Conclusion: In contrast to empirical dose adjustment, the use of a model-based dosing algorithm allows the integration of baseline covariate effects into estimation of individual post-hoc estimates of the parameter(s) of interest. Moreover, the use of longitudinal data from previous monitoring can further contribute to enhanced performance of the model, disentangling the contribution of intraindividual variability in clearance from other sources of variation.

Nuclear factor of activated T cells as potential pharmacodynamic biomarker for the risk of acute and subclinical rejection in de novo liver recipients

Dr. Olga Millán1,2, Dr. Pablo Ruiz3, Virginia Fortuna1, Dr. Miguel Navasa1,3, Dr. Constantino Fondevila1,4, Dr. Mercè Brunet1,2

1Pharmacology and Toxicology, Biochemistry and Molecular Genetics, Biomedical Diagnostic Centre (CDB), IDIBAPS, Hospital Clinic of Barcelona, University of Barcelona, Spain.; 2Biomedical Research Centre in Hepatic and Digestive Diseases (CIBERehd), Instituto de Salud Carlos III, Spain.; 3Liver Transplant Unit, IDIBAPS, Hospital Clinic of Barcelona, University of Barcelona, Spain; 4Hepatopancreatobiliary Surgery and Transplantation, General and Digestive Surgery Service, Hospital Clínic, IDIBAPS, University of Barcelona, Barcelona, Spain.

Background:NFAT-regulated gene expression (NFAT-RGE) has been proposed as a pharmacodynamic biomarker for Tac and CsA. Our aim was to evaluate the role of NFAT-RGE in modulating intralymphocytary IL-2 and IFN- expression and its clinical utility as an early predictive biomarker for the risk of acute rejection (AR) and infection in de novo adult liver transplant (LT) recipients.

Methods:Fifty-six LT recipients treated with Tac or CsA [with and without mycophenolate mofetil (MMF)] were included: 30 free of rejection or infection, 11 rejectors (TCMAR), 5 with subclinical rejection (SCR) and 10 with cytomegalovirus (CMV) infection. Within the first 3 months after transplantation, NFAT-RGE and intralymphocytary synthesis of IL-2 and IFN- were evaluated by real-time PCR and flow cytometry, respectively. The trough and 1.5h concentrations after CNi administration were analysed within the 1st week, on the 15th day, and at 1,2 and 3 months post-transplantation.

Results:A significant increase in NFAT-RGE was observed in patients who experienced TCMAR [75% (42%-100%)] or SCR [41% (18%-78%)] compared with patients without rejection or infection [14% (2%-23%)]. Positive correlations between the %NFAT-RGE-IFN and both the %CD8CD69IFN- and %CD4CD69IFN-and between the %NFAT-RGE-IL2 and the %CD8CD69IL2 were observed. NFAT-RGE was significantly lower in CMV+ patients than in non-infected patients. No statistically significant differences between patients without rejection, patients with rejection (TCMAR or SCR) and patients who had CMV infections were observed with regard to drug exposure.

Conclusions:The results suggest that NFAT-RGE is a potential biomarker for the risk of TCMAR and SCR and may provide guidance for CNi therapy. The combination of sequential pre- and post-transplantation monitoring of intralymphocytary IL-2 and IFN- levels with NFAT-RGE may be considered to improve patient risk stratification, mainly pre-transplantation and at the 1st and 3rd months post-transplantation and provide physicians with broad information for early personalized treatment adjustment and better prevention of rejection.

6:00pm - 7:30pmCocktail reception and poster viewing
Location: Angelicum Congress Centre

Dinner at Leisure

Angelicum Congress Centre 
8:30pm - 11:00pmYoung Scientists dinner
Location: Stadium of Domitian
Stadium of Domitian 
Date: Tuesday, 21/Sept/2021
7:30am - 8:30amMorning roundtable session: TDM of taxanes
Location: Hotel The Hive

Chair: Salvatore J. Salamone

Hotel The Hive 
7:30am - 8:30amMorning roundtable session: New drug-drug interactions in organ transplant recipients
Location: Hotel The Hive

Chair: Florian Lemaitre

Hotel The Hive 
7:30am - 8:30amMorning roundtable session: TDM of long-acting formulations of antipsychotics
Location: Hotel The Hive

Chair: Sara Baldelli

Hotel The Hive 
8:00am - 7:30pmCongress registration
Location: Sala delle Colonne - Angelicum Congress Centre
Sala delle Colonne - Angelicum Congress Centre 
8:45am - 9:30amPlenary lecture 3 (CME): Precision medicine and information technology in ICUs.
Location: Aula Magna - Angelicum Congress Centre

Plenary Speaker: Stefano Finazzi

Aula Magna - Angelicum Congress Centre 
9:30am - 10:00amVictor Armstrong Young Investigator Award
Location: Aula Magna - Angelicum Congress Centre
Aula Magna - Angelicum Congress Centre 
10:00am - 10:15amPatsalos Prize
Location: Aula Magna - Angelicum Congress Centre
Aula Magna - Angelicum Congress Centre 
10:15am - 10:50amMorning coffee & tea, exhibition and poster viewing
Location: Angelicum Congress Centre
Angelicum Congress Centre 
11:00am - 12:30pmSymposium 7: The issues in pediatric patients
Location: Aula Magna - Angelicum Congress Centre
Session Chair: Bianca Maria Goffredo

11.00 am: Talk 20: TDM of drugs used for analgesia/sedation. Bianca Maria Goffredo

11.30 am: Talk 21: Management of IFI in immunocompromised children. Roger Bruggemann

12.00 pm: Talk 22: Antipsychotics monitoring in pediatric patients. Emilio Clementi

Aula Magna - Angelicum Congress Centre 
11:00am - 12:30pmSymposium 8: Microsampling devices: assay developments and applications
Location: Room A - Angelicum Congress Centre
Session Chair: Antonio D'Avolio

11.00 am: Talk 23: Dried blood microsampling: new devices, automation, and toxicological and clinical implementation. Sigrid Deprez

11.30 pm: Talk 24: DBS and the difficulties one encounters when implementing such a strategy in practice. Marith Francke

12.00 pm: Talk 25: Direct mass spectrometric analysis from VAMS. Karin Kipper

Room A - Angelicum Congress Centre 
11:00am - 12:30pmSymposium 9: Diagnosing drug-induced liver injury (DILI)
Location: Room B - Angelicum Congress Centre

Session Chair: Yusuke Tanigawara

11.00 am: Talk 26: Adverse drug reactions caused by genetic variation of drug transporters. Yusuke Tanigawara

11.30 am: Talk 27: Predictings hypersensitivity drug reactions mitochondrial toxicity and cytokine storm. Manuela Neuman

12.00 pm:Talk 28: Liver injury produced by amatoxins and phallotoxins. Eberhard Wieland

Room B - Angelicum Congress Centre 
11:00am - 12:30pmOral Session: Pharmacometrics
Location: Room C - Angelicum Congress Centre

Session Chair: Sebastian Wicha
Session Chair: Oscar Della Pasqua

Room C - Angelicum Congress Centre 

Boosting Model-Informed Precision Dosing with Machine Learning: Improving predictions in atypical patients

Dr. Jasmine Hannah Hughes, Dr. Ron Keizer

InsightRX, United States of America

Introduction: Bayesian estimation of individual pharmacokinetic (PK) parameters combines knowledge about the patient, in the form of therapeutic drug monitoring (TDM) samples or pharmacodynamic measurements, with prior knowledge about the drug PK, in the form of a population PK model. Although maximum a posteriori (MAP) Bayesian estimation underpins most recommendations in model-informed precision dosing (MIPD) software, there is little clinical guidance for how to handle patients that are poorly reflected by the model priors. One previously proposed approach is “flattened priors” (FP), in which the weight of the model prior is decreased relative to observed TDM data during estimation of individual PK parameters. However, little is known about the performance of this approach, nor when it should be applied.

Methods: FP was evaluated retrospectively in data collected over the course of routine clinical care from 4679 adult patients treated with vancomycin. Machine learning (ML) models were developed to recommend when FP should be used in place of MAP during parameter estimation.

Results: We find that FP improves PK model predictive performance in 42-55% of PK parameter estimations. The ML models correctly recommended the use of FP over MAP in 81-85% of decisions, reducing root mean squared error (RMSE) of PK model predictions by 12-22% (0.5-1.2 mg/L). Two features were consistently the most indicative of using FP: past prediction error magnitude and consistent bias in prediction error. A minimal model developed using only these two features reduced RMSE by 5-18% (0.20-0.93 mg/L) relative to MAP.

Conclusions: This work validates the use of FP for predicting PK in patients poorly described by model priors. The proposed models leverage machine learning to improve model predictive accuracy in precision dosing, while maintaining the mechanistic insight and explainability of pharmacokinetic models.

A Reinforcement Learning Tool for Population Modeling

Julian Otalvaro1,2,3, Michael Neely1, Walter Yamada1, Alher Hernandez3, Andres Zuluaga2

1LAPKB, United States of America; 2LIME, University of Antioquia; 3GIBIC, University of Antioquia

Population pharmacokinetic modeling is a proven tool to inform decision making across all phases of the drug development process and subsequent determination of dosing protocols. This process can take a considerable amount of time, influenced by the complexity of the model, the experience of the modeler, and the set of tools selected [1]. Reinforcement Learning (RL) is a type of Machine Learning, with the goal of learning the sequence of decisions that maximize a total reward.

We designed, developed, and implemented an RL tool to assist the pharmacometrician with a workflow when no prior information about the drug’s pharmacokinetics is known by testing a pool of compartmental models, parameter ranges and algorithms in an unbiased manner. The first version of this tool supports one and two compartment models with all their parameters. We used the Nonparametric Adaptive Grid (NPAG) algorithm [2] and a newer Nonparametric Optimal Design (NPOD) algorithm to drive the optimization. Both NPAG and NPOD are maximum likelihood population algorithms, but NPAG searches parameter space randomly, while NPOD searches with a gradient method and is generally faster than NPAG.

To test the RL tool, we used a previously published simulated dataset for a one-compartment model of an intravenous infusion with unimodal distribution for volume and bimodal for elimination, plus one outlier [3]. We purposely initialized the RL process far from the previously published results. The RL implementation obtained a final log-likelihood of 134.61 and 131.67, when using NPAG or NPOD as the driver compared to 130.67 obtained manually, respectively. In both cases the different distributions of the parameters and the values for the outlier were detected accurately.

Neonatal PBPK model of buprenorphine: Identification of factors driving PK variability in newborns with neonatal opioid withdrawal syndrome

Matthijs W. van Hoogdalem1,2, Dr. Trevor N. Johnson3, Dr. Brooks T. McPhail1,4, Dr. Suyog Kamatkar5,6, Dr. Scott L. Wexelblatt5,7,8, Dr. Laura P. Ward5,7, Prof. Uwe Christians9, Prof. Henry T. Akinbi5,7, Prof. Alexander A. Vinks1,7,8, Dr. Tomoyuki Mizuno1,7,8

1Division of Clinical Pharmacology, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, USA; 2James L. Winkle College of Pharmacy, University of Cincinnati, Cincinnati, OH, USA; 3Certara UK Limited, Sheffield, UK; 4School of Medicine Greenville, University of South Carolina, Greenville, SC, USA; 5Perinatal Institute, Division of Neonatology, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, USA; 6Community Hospital East, Indianapolis, IN, USA; 7Department of Pediatrics, College of Medicine, University of Cincinnati, Cincinnati, OH, USA; 8Center for Addiction Research, College of Medicine, University of Cincinnati, Cincinnati, OH, USA; 9iC42 Clinical Research and Development, University of Colorado, Aurora, CO, USA

Introduction: Prenatally opioid-exposed infants are at risk for neonatal opioid withdrawal syndrome (NOWS). NOWS is a major public health issue resulting from the expanding opioid epidemic. Sublingual buprenorphine is an emerging treatment for NOWS, but optimal dosing strategies are an unmet clinical need. In this study, we developed a physiologically-based pharmacokinetic (PBPK) model to identify factors predictive of buprenorphine PK variability to accelerate precision dosing in this vulnerable patient population.

Methods: Sequential adult and neonatal PBPK models were developed using Simcyp, with inclusion of ontogeny profiles of CYP3A4, CYP2C8, UGT1A1, UGT1A3, UGT2B7, and UGT2B17. Biliary clearance ontogeny was estimated through PBPK model-based simulations using buprenorphine blood concentration data obtained from neonates through sparse sampling. Following adult and neonatal model validation, sensitivity analysis was performed to investigate the effect of patient-specific factors on clearance (CL/F) in virtual neonates aged 1 day, 1 week, and 1 month postnatal age (PNA).

Results: Following sublingual administration in 127 healthy adult volunteers, the geometric mean (95% confidence interval [CI]) of PBPK model-based predicted/observed ratios (P/O ratios) of CL/F, peak concentration (Cmax), and time to reach Cmax (Tmax) were 1.00 (0.75–1.32), 1.04 (0.84–1.29), and 0.95 (0.72–1.26), respectively. Following sublingual administration in 19 neonates with NOWS, inclusion of a newly outlined biliary clearance developmental profile yielded a geometric mean (95% CI) P/O ratio of buprenorphine concentrations equaling 0.75 (0.64–0.87). The PBPK model was sensitive to changes in sublingual absorption, biliary clearance, and CYP3A4 abundance, especially in neonates aged 1 day PNA. Other enzyme abundances did not affect CL/F.

Conclusions: The PBPK model demonstrated that variability in sublingual absorption, biliary clearance, and CYP3A4 abundance likely drive buprenorphine PK variability, especially in neonates early after birth when NOWS treatment is initiated. These findings may enhance refinement of buprenorphine dosing protocols for the treatment of NOWS.

Population pharmacokinetics of oxycodone and active metabolites in patients with cancer-related pain

Bram Agema1,2, Astrid Oosten1, Sebastiaan Sassen2, Wim Rietdijk2, Carin Van der Rijt1, Birgit Koch2, Ron Mathijssen1, Stijn Koolen1,2

1Dept. of Medical Oncology, Erasmus MC Cancer Institute, Erasmus University Medical Center, Rotterdam, The Netherlands; 2Clinical Pharmacy, Erasmus University Medical Center, Rotterdam, The Netherlands

Background Oxycodone is frequently used for the treatment of cancer-related pain, while not much is known about the factors that influence treatment outcomes in these patients. We aim to unravel these factors by developing a population-pharmacokinetic model to assess the pharmacokinetics (PK) of oxycodone and its metabolites in cancer patients. Further, we associate oxycodone (metabolite) exposure with pain scores, and the occurrence and severity of adverse events.

Methods Hospitalized patients with cancer-related pain, who were titrated with oral oxycodone could participate. PK samples of oxycodone, oxymorphone, nor-oxycodone, and nor-oxymorphone were taken every 12h and occasionally at 5, 15, 30, and 60 min after oxycodone administration. Patient-reported pain scores and occurrence and severity of nausea, vomiting, dry mouth, drowsiness, obstipation, myoclonia, confusion, hallucinations and transpiration were collected on a 4-point Likert scale every 12h. PK data was analyzed using NONMEM and PK/pharmacodynamic (PK/PD) associations were tested using mixed-effects linear regression.

Results In 28 patients, 309 samples for oxycodone and its metabolites were collected. A five-compartment model best described oxycodone, nor-oxycodone, and nor-oxymorphone pharmacokinetics. None of the tested covariates (age, gender, albumin eGFR, weight, BMI, and CYP3A4, CYP2D6, and UGT2B7 genotype) significantly improved the PK model, mainly due to lack of power. Furthermore, oxycodone exposure, quantified as area under the curve, was not associated with average and maximal pain scores and oxycodone, nor-oxycodone, and nor-oxymorphone exposure were not associated with adverse events (all p>0.05).

Conclusion This is the first model to describe the pharmacokinetics of oxycodone including the metabolites nor-oxycodone and nor-oxymorphone in hospitalized patients with cancer pain. Additional research, including more patients and more timely collection of PD data, is needed to further elucidate oxycodone (metabolite) PK/PD relationships. This model is a good starting point for further studies to optimize oxycodone dosing regiments in patients with cancer-related pain.

Neural network modelling does not yield better results than Xgboost for the prediction of MPA exposure in large kidney transplantation datasets

Dr. Jean-Baptiste Woillard1,2, Dr. Marc Labriffe1,2, Dr. Jean Debord1,2, Prof. Pierre Marquet1,2

1INSERM U1248, University of Limoges, France; 2Department of Pharmacology and Toxicology, Limoges University Hospital Center, France

Introduction: Therapeutic drug monitoring of mycophenolic acid (MPA) based on the interdose area under the curve (AUC) is well established and we previously used machine learning (Xgboost) to accurately estimate MPA AUC0-12h based on a limited sampling strategy. The aim of this work was to estimate MPA AUC0-12h in organ transplant patients using neural network (NN) modelling, the state of the art technique for analyzing very large and complex datasets.

Methods: We used a total of 12,877 MPA AUC0-12h estimated through our ISBA expert system for 6884 different patients using at least 3 plasma concentrations (at 20min, 1 and 3h after dosing), to develop 2 independent NN models based on 3 or 2 concentrations. Data were split into a training set (75%) and a test set (25%) and the NN models in the training set with the lowest root mean square error (RMSE) in a five-fold cross-validation experiment were evaluated in the test set and in one independent full-pk dataset from kidney transplant patients. The analyses were performed using the tidymodel framework and keras/tensorflow in R.

Results: NN modelling yielded good performance in the training set after cross validation, with median[min-max] RMSE = 6.72[6.43-6.83] and 5.55[5.12-5.93] h*mg/L for the 2- and 3-sample NN models, respectively. Performance in the test set was similar, with relative Mean Prediction Error (MPE)/RMSE of 0.5/16.6% and -0.8/13.8% for 2 and 3 samples, respectively. AUC estimation in the external full-PK AUC with NN was less accurate, but still equivalent to the performance of Maximum A Posteriori-Bayesian estimation (MAP-BE), with relative MPE/RMSE/’number of profiles out-of the ±20% interval’ as follows: NN 2 samples: 0.15/24.7/36%; NN 3 samples: -3.8/19.8/31%; MAP-BE: 0.9/19.9/29%).

Conclusions: Neural network modelling does not improve MPA AUC estimation in renal transplant patients over Xgboost or MAP-BE.

An open-source R package for maximum a posteriori Bayesian estimation of pharmacokinetic parameters : mapbayr.

Dr. Félicien Le Louedec1,2,3, Dr. Florent Puisset2,3,4, Dr. Fabienne Thomas1,2,3, Dr. Mélanie White-Koning2,3, Prof. Etienne Chatelut1,2,3

1Pharmacology Department, Institut Universitaire du Cancer de Toulouse - Oncopole, Toulouse, France; 2Team 14 (Dose Individualization of Anticancer Drugs), Centre de Recherches en Cancerologie de Toulouse, Inserm UMR1037, Toulouse, France; 3Faculté de Pharmacie, Université Paul Sabatier, Toulouse, France; 4Pharmacy Department, Institut Universitaire du Cancer de Toulouse - Oncopole, Toulouse, France

Introduction. Maximum a posteriori Bayesian estimation (MAP-BE) of pharmacokinetic (PK) parameters is the cornerstone of model-based therapeutic drug monitoring (mbTDM). Because current mbTDM tools present several drawbacks (limited drug library, restricted settings, not free of charge), we developed the free and open-source R package mapbayr to perform MAP-BE easily for any kind of PK model, independently of external modeling software. Our objective was to validate its performance versus NONMEM.

Methods. mapbayr relies on mrgsolve for model coding and differential equation solving, and on the L-BFGS-B algorithm to minimize objective function value (OFV). Thirty-four PK models were coded with the possible features: oral or intravenous administration ; first- or zero-order absorption ; lag time ; time-varying covariates ; Michaelis-Menten elimination ; additive, proportional, mixed and exponential residual error ; parent drug and metabolite ; limited or large inter-individual variability. For each model, 4,000 PK profiles were simulated, with combinations of single/multiple dosing and rich/sparse sampling. MAP-BE of PK parameters was performed in both mapbayr and NONMEM. For every patient, the estimation was qualified as excellent if the maximum difference in PK parameter values (∆η) was lower than 0.1%, discordant if higher than 10%, and acceptable in between.

Results. Of the 136,000 estimations, median ∆η was 0.0003%, 98.8% had an excellent estimation of PK parameters, 0.3% acceptable and 0.9% discordant. The highest percentages of discrepancy (5.7% discordant) were seen with models with lag time or zero-order parameters to estimate in rich sampling settings. However, in those cases, OFV was often lower with mapbayr than NONMEM and discrepancies affected absorption parameters and rarely elimination such as clearance.

Discussion. Thanks to mapbayr, any model can be coded in R and be used for MAP-BE with reliable estimates. It is intended to be an universal tool for in-house Shiny apps to perform mbTDM. mapbayr will soon be available on CRAN and Github.

Population pharmacokinetic and target attainment of fosfomycin disodium in healthy male volunteers

Angela Elma Edwina1, Birgit Koch1, Valentin al Jalali2, Peter Matzneller2, Markus Zeitlinger2, Sebastiaan Sassen1

1Department of Hospital Pharmacy, Erasmus University Medical Center, Rotterdam; 2Department of Clinical Pharmacology, Medical University of Vienna, Vienna, Austria

Introduction: Fosfomycin disodium (FOD) is a wide-range antibiotic used against resistant bacteria. We developed a plasma and urine population PK (PopPK) model to evaluate: (1) FOD disposition and subject characteristics impact on PK parameters, and (2) the adequacy of standard dose regimens to attain efficacy targets.

Methods: In this two-way crossover study, eight healthy men were randomised to receive both intermittent 8g tid and continuous infusion 1g/h over 18h with a loading dose of 8g separated by a wash-out period of at least 48h. Per subject 22 blood and 9-11 urine samples were collected. Concentrations were determined using LC-MS/MS. PopPK modelling was performed using Non-Linear Mixed Effects Modelling (NONMEM, FOCE-I). Monte Carlo simulations using the final model were conducted to evaluate Probability of Target Attainment (PTA) of efficacy targets (AUC24h/MIC of 25 and 70%T>MIC).

Results: The data was best described by a two-compartment model with a urine compartment. All parameters were scaled by BMI. Final model included glomerular filtration rate (GFR) calculated with CKD-EPI formula which showed a significant positive correlation with the renal clearance. For MIC ≤32mg/L, 4g every 8h reached 100% and 99.9% of attainment for AUC/MIC and 70%T>MIC targets, respectively. For more resistant pathogens with a MIC 128mg/L, 100% of subjects who received 1g/h continuous infusion attained target of 70%T>MIC, while none of those received intermittent infusion achieved 70%T>MIC. Subjects with reduced kidney function and low BMI resulted in highest PTA.

Conclusion: Intermittent infusion dose of 4g every 8h might be sufficient and the most optimal dosing to treat susceptible bacteria with MIC ≤32mg/L. Continuous infusion is preferred for treatment which requires time-dependent bactericidal activity. The model suggests that FOD dosing should be based on GFR and BMI.

11:00am - 12:30pmSeminario Intersocietario (ECM): Il paziente fragile
Location: Room D - Angelicum Congress Centre
Session Chair: Antonello Di Paolo

11.00 am: Saluti ed apertura dei lavori

11.05 am: Terapie croniche e comorbidità nel paziente anziano. Fabio Monzani

11.25 am: Terapie concomitanti nel paziente trapiantato. Paolo De Simone

11.45 am: TDM nel paziente fragile. Mariela Marinova

12.05 pm: Discussione

12.25 pm: Considerazioni conclusive

Room D - Angelicum Congress Centre 
12:30pm - 1:00pmThermoFisher workshop: TDM by LC-MS/MS in pediatrics: from bench to bedside
Location: Aula Magna - Angelicum Congress Centre

Moderator: Edward Goucher - Thermo Fisher Scientific

 Speaker: Giuliana Cangemi - Gianina Gaslini Institute

Aula Magna - Angelicum Congress Centre 
12:30pm - 1:00pmPromise Proteomics workshop: Pharmacometrics of biologics in oncology: any room for therapeutic drug monitoring?
Location: Room A - Angelicum Congress Centre

Speaker: Dr Joseph Ciccolini - Professor of Pharmacokinetics, Clinical Pharmacologist (PU-PH)

Room A - Angelicum Congress Centre 
12:30pm - 2:00pmBusiness meeting - Pharmacometrics Committee
Location: Room B - Angelicum Congress Centre

Coordinators: Brenda De Winter, Jean-baptiste Woillard   

Room B - Angelicum Congress Centre 
12:30pm - 2:00pmBusiness meeting - Young Scientists Committee
Location: Room C - Angelicum Congress Centre

Coordinators: Laura Elens, Brenda DeWinter

Room C - Angelicum Congress Centre 
12:30pm - 2:00pmBusiness meeting - Clinical Toxicology/Drugs of Misuse Committee
Location: Room D - Angelicum Congress Centre

Coordinators: Manuela Neuman, Eric Franssen

Room D - Angelicum Congress Centre 
12:30pm - 2:00pmLunch with Exhibitors
Location: Angelicum Congress Centre
Angelicum Congress Centre 
1:00pm - 1:30pmThermoFisher workshop: New workflows for the detection of TDM with LC HRMS Orbitrap
Location: Aula Magna - Angelicum Congress Centre

Moderator: Edward Goucher - Thermo Fisher Scientific

Speaker: Petra Gerhards - Thermo Fisher Scientific

Aula Magna - Angelicum Congress Centre 
1:40pm - 1:55pmGroup Picture
Location: Aula Magna - Angelicum Congress Centre
Aula Magna - Angelicum Congress Centre 
2:00pm - 3:30pmSymposium 10: Therapeutic drug monitoring of biologics in immune-mediated inflammatory diseases other than IBD
Location: Aula Magna - Angelicum Congress Centre

Session Chair: Annick de Vries

02.00 pm: Talk 29: Exposure-Response Relationship of biologics in Patients With Rheumatoid Arthritis: individualised dose reduction based on drug concentration. Gertjan Wolbink

02.30 pm: Talk 30: Clinical utility of therapeutic drug monitoring for biologics in patients with psoriasis. Teresa Tsakok

03.00 pm: Talk 31: Personalized dosing of natalizumab in multiple sclerosis. Floris Loeff

Aula Magna - Angelicum Congress Centre 
2:00pm - 3:30pmSymposium 11: Laboratory monitoring of COVID-19 patients
Location: Room A - Angelicum Congress Centre
Session Chair: Birgit Koch

02.00 pm: Talk 32: Variants, epidemiology: the impact on vaccinations. Fausto Baldanti

02.30 pm: Talk 33: Performance of the different serological tests for anti-SARS-CoV-2 antibodies. Giulia Matusali

03.00 pm: Talk 34: Molecular diagnostic assays: considerations for SARS-COV-2 infection. Valeria Micheli

Room A - Angelicum Congress Centre 
2:00pm - 3:30pmOral Session: Anti-infectives: antiviral drugs
Location: Room B - Angelicum Congress Centre
Session Chair: Antonio D'Avolio
Session Chair: Nunzia Decembrino
Room B - Angelicum Congress Centre 

Intracellular evaluation of the pharmacokinetic interaction between rifampin and darunavir/ritonavir with dolutegravir

Dr. Elisa Delia De Vivo1, Prof. Andrea Calcagno2, Dr. Amedeo De Nicolò1, Dr. Ilaria Motta2, Prof. Antonio D'Avolio1, Prof. Giovanni Di Perri2, Prof. Lubbe Wiesner3, Dr. Isma-eel Ebrahim3, Prof. Gary Maartens3, Prof. Catherine Orrell4, Prof. Helen McIlleron3

1University of Turin, Department of Medical Sciences, Laboratory of Clinical Pharmacology and Pharmacogenetics, Amedeo di Savoia Hospital, Turin (Italy); 2Unit of Infectious Diseases, Department of Medical Sciences, University of Turin, Italy; 3Division of Clinical Pharmacology, Department of Medicine, University of Cape Town, South Africa; 4Desmond Tutu HIV Centre, Institute of Infectious Diseases and Molecular Medicine (IDM) and Department of Medicine, University of Cape Town, South Africa

Darunavir and ritonavir (DRV/r) are victims of drug-drug interactions (DDI) with strong cytochrome inducers, such as rifampin (RIF). Recent data showed that doubling DRV/r dose did not compensate RIF-induced decrease in both drugs plasma exposure. The impact of this DDI within Peripheral Blood Mononuclear Cells (PBMC) is still unknown. Therefore we investigated the intra-PBMC pharmacokinetics (PK) of DRV/r, dolutegravir (DTG) and RIF.

People living with HIV were enrolled in a prospective dose-escalation cross-over study. Patients started with a DRV/r 800/100mg QD regimen, then RIF (600-750mg) and DTG 50mg BD were added, RTV dose was escalated to 200mg, then they received either DRV/r 800/100 BD and then 1600/200 QD or vice versa. RIF was withdrawn. Plasma and intra-PBMC concentrations of DRV/r, DTG and RIF were measured through validated LC-MS/MS methods at the Ctrough at each treatment switch, and at 2-6h at the start and after DRV/r dose escalations.

17 patients were enrolled but due to high incidence of liver toxicity only 4 patients completed the protocol.

In this patients, after the addition of RIF, intra-PBMC DRV Ctrough dropped significantly (P=0.039) to 119ng/mL (IQR 13–694) and 68ng/mL (16–164) for 800/100 BD and 1600/200 QD dosages, respectively. Differences were less pronounced at 2-6h (P=0.114). RIF addition and dose escalation was associated with a significant increase in the intra-PBMC/plasma ratio for DRV, from a median starting value of 0.17 (0.09–0.26) to 0.23 (0.20–0.26) and 0.28 (0.21–0.41) for 800/100 BD and 1600/200 QD regimens, respectively. This indicates that the reduction within PBMC is less pronounced than in plasma. DTG and RIF intra-PBMC concentrations were comparable with data in literature.

The observed different impact of this DDI in plasma and in PBMC suggests that the intra-PBMC quantification may be useful to fully understand the relevance of pharmacokinetic DDIs with strong inducers.

Antiretroviral drug plasma exposure according to seasonality: a TDM register of 10 years.

Dr. Jessica Cusato, Prof. Andrea Calcagno, Dr. Jacopo Mula, Dr. Valeria Avataneo, Dr. Miriam Antonucci, Dr. Alice Palermiti, Dr. Amedeo De Nicolo', Dr. Elisa Delia De Vivo, Dr. Alice Ianniello, Dr. Alessandra Manca, Prof. Antonio D'Avolio

University of Turin

Therapeutic drug monitoring is a clinical practice which is able to quantify drugs in a biological matrix at specific timings to maintain a constant concentration in patient bloodstream, thereby optimizing individual dosage regimens in various clinical situations. In literature, studies showed that some drug concentrations, as the immunosuppressant drugs tacrolimus and sirolimus, may differ according to seasonal variation; authors justify this variability consistently with changes in vitamin D level, which is able to influence the expression of cytochrome P-450 3A4, main responsible of these drugs metabolism.

No data are available in literature concerning antiviral drug exposure variation according to seasonality.

For these reasons, plasma antiviral drug concentrations were evaluated and their level variation was investigated during the year.

Plasma concentrations of raltegravir, ritonavir, darunavir, elvitegravir, dolutegravir, etravirine, atazanavir, tenofovir, emtricitabine, lamivudine, efavirenz, nevirapine, maraviroc, abacavir, rilpivirine, lopinavir and enfuvirtide were measured through liquid chromatography coupled with mass spectrometry or UV detection.

A number of 4148 samples with blood withdrawn date were available from 2010 to 2020: maraviroc and etravirine levels were lower in winter compared to summer, whereas lopinavir concentrations had an opposite trend, with an increased exposure in summer than in winter.

Furthermore, an age higher than 50 years and summer resulted predictive factors of etravirine 300 ng/mL, the cut-off value associated with therapeutic efficacy.

This study reports for the first time the variability of antiviral drug concentrations during the year and seasons and it highlights the needed of therapy personalization. However, further studies in different cohorts are required to confirm this data and to clarify the role of vitamin D (certainly evaluating its levels) in drug pharmacokinetics.

Therapeutic drug monitoring and virological response in a cohort of HIV patients switching to dolutegravir/rilpivirine dual maintenance therapy (ANRS-BIRIDER Study)

Dr. Florian Lemaitre1, Dr. Jennifer Lagoutte-Renosi2, Dr. Marie-Claude Gagnieu3, Dr. Nicolas Venisse4, Dr. Matthieu Grégoire5, Dr. Stéphane Bouchet6, Dr. Rodolphe Garraffo7, Dr. Minh Lê8, Dr. Patrice Muret2, Dr. Emmanuelle Comets9, Dr. Caroline Solas10, Dr. Gilles Peytavin11, Anrs; Birider study group12, Anrs; AC-43. Pharmacology subgroup12

1Univ Rennes, CHU Rennes, Inserm, EHESP, Irset (Institut de recherche en santé, environnement et travail)-UMR_S 1085, Rennes, France; 2Clinical Pharmacology Department - Minjoz University Hospital - Besancon – France; 3Hospices Civils de Lyon - Groupement Hospitalier Sud - Service de Biochimie et Biologie Moléculaire - UM Pharmacologie-Toxicologie - 69495 Pierre-Benite; 4CHU Poitiers, Laboratoire de Toxicologie et Pharmacocinétique, CIC Inserm 1402, Poitiers, France; 5Clinical Pharmacology Department, Nantes University Hospital, Nantes, France; 6Laboratoire de Pharmacologie et Toxicologie, Service de Pharmacologie Médicale, CHU Pellegrin, INSERM U1219, Bordeaux, France; 7Pharmacologie Médicale, Faculté de Médecine et CHU de Nice, Hôpital Pasteur, Nice; 8AP-HP, Bichat Claude Bernard Hospital, Pharmacology-Toxicology Department, 75018 Paris, France; INSERM, UMRS-1144, Université de Paris, 75006 Paris, France; 9INSERM, CIC 1414, 35700 Rennes, France; Univ Rennes-1, 35700 Rennes, France; Université de Paris, INSERM, IAME, F-75006 Paris, France; 10APHM, Hôpital La Timone, Laboratoire de Pharmacocinétique et Toxicologie, Unité des Virus Émergents (UVE: Aix-Marseille Univ-IRD 190-Inserm 1207), Marseille, France; 11AP-HP, Bichat Claude Bernard Hospital, Pharmacology-Toxicology Department, 75018 Paris, France, Université de Paris, INSERM, IAME, F-75006 Paris, France; 12ANRS-Maladies Infectieuses Emergentes

Human Immunodeficiency Virus (HIV) treatment with dolutegravir(DTG)-rilpivirine(RPV) dual therapy is now increasingly recommended as a switch option in patients virologically suppressed. The aim of the study is to explore the relationship between DTG and RPV plasma trough concentrations (Cmin) and treatment response during maintenance treatment.

Patients switched to a dual DTG-RPV (50/25mg qd) therapy and with a plasma HIV-RNA (VL)<50 cp/mL were included in this retrospective, multicenter (n=9) study. Variables collected at W4, W12, W24 and W48 included demographics, VL, DTG and RPV Cmin, CD4, creatinine, eGFR. DTG and RPV Cmin were estimated using existing popPK models. The relationship between virological failure (defined as VL>50 cp/mL) and Cmin was explored at each visit using t-tests. Median [IQR25-75%] were presented.

Two hundred and two patients were enrolled in the study. Baseline characteristics were: age 56.0 years[50.2-63.0], 26% female, zenith VL 90,209 cp/mL[26,893-292,250] and nadir CD4 202/mm3[97-336]. Before switching, patients were mainly treated with triple therapy (n=149; 44 with DTG and 34 with RPV). DTG and RPV Cmin were 2175 ng/mL[1,327-3,150; n=309] and 112 ng/mL[77-176; n=310), respectively. Besides, 4.2 and 8.0% of patients had concentrations below clinical threshold (600 and 50 ng/mL) for DTG and RPV, respectively.During the 48-week follow-up, 16 patients (7.5%) exhibited a VL>50 cp/mL. No relationship between virological response and Cmin was evidenced at any time-point. At W48, 20 patients (9.4%) have discontinued their dual therapy: 5 for virological failure (2.4%), 7 for adverse events (3.3%) and 8 for other causes (3.8%).

A majority of patients receiving DTG/RPV maintenance dual therapy demonstrated virological success at W48 suggesting treatment efficacy of the one pill and once daily regimen. Drug exposure was not associated with VL>50 cp/mL during the 48-week study period possibly due to a low power/low number of events and a relatively drug exposure for both drugs.

Evaluation of interactions between antiretroviral drugs and food using chemometric methods

Agnieszka Wiesner, Prof. Paweł Zagrodzki, Marek Jamrozik, Dr. Monika Marcinkowska, Prof. Paweł Paśko

Jagiellonian University Medical College, Faculty of Pharmacy, Poland

Introduction: The occurrence of interactions may significantly alter the effectiveness of highly active antiretroviral therapy (HAART). Whereas awareness of drug-drug interactions exists, contrastingly, interactions between drugs and food (including dietary supplements) are often underestimated by patients, physicians, and pharmacists. In our study we used chemometric methods to analyze simultaneously many different features of antiretroviral drugs and food, that may influence their mutual interactions. Methods: 1) A comprehensive literature review of clinical studies investigating interactions with food, beverages, dietary supplements, and alcohol was performed for 33 antiretroviral drugs. 2) Data concerning different factors related to the drug and food was extracted and aggregated. Among evaluated drug features were e.g.: physicochemical properties, formulation, dose, pre- and postprandial values of pharmacokinetic and pharmacodynamic parameters. Investigated food features were e.g.: caloric density, quantitative and qualitative meal composition. 3) pKa values for the main chemical groups of each drug were calculated in silico, as well as the set of molecular descriptors. 4) Both extracted and calculated data were analyzed using chemometric methods based on data mining, e.g.: cluster analysis, principal component analysis, partial least-squares regression, and correspondence analysis. Results: With the review, evidence for either positive or negative interactions between HAART and food, beverages, dietary supplements, and alcohol was found. The result of interaction depended on the studied drug features and meal composition. Correlation structure between parameters investigated was evaluated and the strength of bivariate associations, for the pairs of correlated parameters was calculated. The apparent similarities and dissimilarities between studied drugs were revealed. Conclusions: The results of our study will contribute to a better understanding of HAART-food interactions and emphasizing their importance. The outcomes may also help to predict possible interactions based on the chemical structure and physicochemical properties of specific drugs and improve the safety and effectiveness of HAART.

Bioequivalence study of two Favipiravir formulations performed during COVID-19 pandemic

Katerina Kovalyova, Gregory Tsapko, Sergey Mashchenko, Vera Volkova, Prof. Igor Kuznetsov

Pharmbiotest LLC, Ukraine

One year after the COVID-19 pandemic announcement, the disease still expands and spreads that increases the necessity to develop specific antiviral drugs. Viral RNA Polymerase inhibitors is a pharmacologic group that includes Remdesivir approved by FDA for COVID-19 treatment. Favipiravir is another drug from this group attracting the interest of society and generic manufacturers.

The aim of this study was to assess the bioequivalence of a new generic 200 mg Favipiravir enteric-coated tablet “Favipiravir-Microchim” (Microchim, Ukraine) to the reference medicine “Avigan” (FUJIFILM Toyama Chemical Co., Japan).

This randomized, two-period, two-sequence, crossover bioequivalence study was conducted in 24 healthy volunteers of both sexes under fasting conditions. To reduce the risk of infection, a set of special measures was carried out, including multiple PCR-testing. Blood samples were collected just before and after dosing during a 24-h time period. Blood concentrations of favipiravir were measured by HPLC-MS/MS. Cmax, Tmax, t1/2, AUC0-t, AUC0-∞ AUCres, and Kel were determined for the test and reference drugs. Bioequivalence was to be concluded if the 90%-confidence intervals were within the range of 80% to 125% for Cmax and AUC0-t.

24 healthy volunteers were enrolled: 14 men and 10 women; mean age 32.8±6.9 31.9±10.8 years; mean weight 79.1±9.4 and 61.4±6.8 respectively. The 90%-confidence intervals for Favipiravir-Microchim/Avigan mean ratios of the log-transformed pharmacokinetic variables were within the conventional bioequivalence range: 91.14 – 114.04% for Cmax and 97.50 – 111.66% for AUC0-t (point estimate: 101.95 % and 104.34 % respectively). None of the volunteers reported any adverse events during the study.

Conclusions: The two favipiravir enteric-coated tablet formulations demonstrated similar pharmacokinetic profiles and were found to be bioequivalent, and therefore, interchangeable. Considering the limited number of favipiravir pharmacokinetics studies in COVID-19 patients, Pharmbiotest has initiated the project of therapeutic drug monitoring with home blood sampling using VAMS Mirta® (Neoteryx LLC, USA).

Pharmacokinetics of zanamivir in critically ill patients undergoing CVVH

Andre Wieringa1, Dr. Jasper Haringman3, Dr. Peter ter Horst1, Dr. Gert Jan Wagenvoort2, Dr. Birgit Koch4

1Isala Clinics, dept. of Clinical Pharmacy, The Netherlands; 2Isala Clinics, Laboratory of Clinical Microbiology and Infectious Diseases, The Netherlands; 3Isala Clinics, dept. of Intensive Care, The Netherlands; 4Erasmus MC, dept. of Clinical Pharmacy and Pharmacology, The Netherlands



In the critically ill zanamivir is indicated for the treatment of complicated and potentially life-threatening influenza A or B virus infection in adult and paediatric patients (aged ≥6 months). Little is known about the pharmacokinetics of zanamivir in patients undergoing continuous venovenous hemofiltration (CVVH) and the dosage to be used for optimal treatment. Our objective is to report the sieving coefficient (Sv) for zanamivir in patients receiving CVVH.


Patients of ≥18 years admitted to the intensive care unit (ICU) with a life-threatening Influenza A or B infection, treated with zanamivir i.v. and undergoing CVVH were included in this prospective observational study. Patients received a zanamivir loading dose of 600mg i.v., 12 hours later followed by maintenance dosages two times daily according to the treating physician. Zanamivir plasma and ultrafiltrate samples were drawn at t= 0, 0.5, 1, 2, 3, 4, 6, 8, 12h on day 3 of treatment and analysed with a validated HPLC-MS/MS method. The AUC, V, T1/2 were estimated with non-compartmental analysis using PK Solver software, a validated platform for non-compartimental analysis. The CLtotal was calculated as the ratio of zanamivir maintenance dose/plasma AUC0-12HR. The CLCVVH Sv was calculated as the Sv × ultrafiltration flow rate. The CLCVVH AUC was calculated as the ratio (dialysate AUC0-12HR/plasma AUC0-12HR)*CLtotal.


Four patients were included in the study. A Sv of approximately 1.0 was identified, with unrestricted transport of zanamivir to the ultrafiltrate. The calculated average CLCVVH AUC, CLCVVH Sv and CLtotal were comparable with 48 ± 10ml/min, 48 ± 11ml/min resp. 46 ± 10ml/min. The half-life was prolonged with 5.6–9.9 hours compared to the 2-3 hours in patients with normal renal function.


Zanamivir is well cleared by CVVH. Drug exposure is potentially predictable based on ultrafiltration flow rates and may help guide dosing of zanamivir.

2:00pm - 3:30pmOral Session: Clinical Toxicology - 2
Location: Room C - Angelicum Congress Centre

Session Chair: Elisa Roda
Session Chair: Guido Mannaioni

Room C - Angelicum Congress Centre 

An innovative and efficient tool for clinical toxicology: a 60 patients retrospective study on drug annotation combining a Molecular Networking Approach with MetWork web application

Romain Magny1,2, Yann Beauxis1, Dr. Gregory Genta-Jouve1,3, Dr. Emmanuel Bourgogne4,5

1Université de Paris, faculté de Pharmacie, UMR 8038, Paris, France; 2Sorbonne Université UM80, INSERM UMR 968, CNRS UMR 7210, Institut de la Vision, IHU ForeSight, Paris, France; 3USR 3456 CNRS LEEISA, Guyane, France; 4Université de Paris, faculté de Pharmacie, laboratoire de toxicologie, Paris, France; 5APHP, Hopital St Antoine, France

Background: In Toxicology, analysts are confronted with complex problems, resulting in important clinical consequences. Untargeted screening remains a challenge, given the high number of molecules to be detected. LC-HRMS, the reference method for screening, generates a large volume of high quality spectral data, without tools for visualizing and organizing MS data. We applied molecular networking (MN) for screening interpretation. Objectives were to build a MS library and apply this database (DB) for drug’s identification; compare theoretical MS libraries obtained in-silico with our DB; use MetWork webserver for metabolites identification.

Methods: Patient’s plasma was crushed with acetonitrile. After centrifugation, supernatant was injected onto an LC-Orbitrap QExactive LC-HRMS system. For MS, positive MS and two data-dependent MS/MS scans were recorded using CID and HCD activation types. For MN, data were analyzed and visualized using Metgem. Finally, MetWork annotations were filtered and this file was used to annotate the previously obtained MN.

Results: For the DB, 155 compounds were recorded. Using this DB, intake of various commercialized drugs as well as drugs of abuse were confirmed in 60 patients. The MN approaches confirmed results obtained by references methods (80% of the compounds were in common). Comparison was made between our experimental DB and in-silico MS/MS spectra using CFM-ID to annotate drugs metabolites. CFM-ID gave comparable results in 50% of cases whatever the activation types used. Finally, using MetWork, additional metabolite annotations were possible, such as mephedrone major metabolites; phase 2 metabolites like oxazepam glucuronides; isobaric drugs like tramadol, amitriptyline, venlafaxine and their metabolites were identified in patients without ambiguity.

Conclusions: For untargeted screening, our results show that MN opens perspectives in toxicology using LC-HRMS. Combined with CFM-ID and MetWork to extend the annotation of new potential drugs or old drug’s metabolites, even without reference standard, it could help clinicians to explain potential toxicity.

Analysis of Synthetic Cannabinoids Paraphernalia in two Patients admitted for drug Rehabilitation in UAE

Dr. Abuelgasim Elrasheed Alhassan1, Dr. Hamad Al Ghaferi2, Dr. Muez Siddig3, Dr. Amir Haju4

1National Rehabilitation Center, United Arab Emirates; 2National Rehabilitation Center, United Arab Emirates; 3National Rehabilitation Center, United Arab Emirates; 4National Rehabilitation Center, United Arab Emirates

The use of synthetic cannabinoids has been become wide spread around the world and especially within the last 10 years. Compared to some other regions (e.g. Europe and North America), such use has been more recent in the Gulf region. Potential drug material from two clinical patients was analysed in association with their admission and treatment programmes at the National Rehabilitation Center (NRC) in Abu Dhabi. Patient 1 had a history of drug use (in particular, heroin, methamphetamine, pregabalin, tramadol, benzodiazepines, cannabis and eye drops). As part of the out-patient programme after in patient care, urine drugs of abuse monitoring detected synthetic cannabinoids on one occasion but the patient denied use. The family were concerned of sporadic episodes of patient confusion, self-talking, restlessness, inappropriate laughter and nonsensical talking but the patient denied any drug use. Small pieces of paper were found next to the patient’s smoking kit at home and were submitted to the NRC toxicology laboratory. The synthetic cannabinoids 5F-MDMB-PICA and 5F-MDMB-PINACA were detected with use of such drugs potentially accounting for the patient symptoms exhibited. Patient 2 had a history of pregabalin and methamphetamine drug use. They had started to use synthetic cannabinoids within the last 2 years prior to admission for rehabilitation and had experienced a rapid health deterioration, including uncontrolled blood sugar (alternating hyperglycemia and hypoglycemia), repeated nausea and vomiting, repeated infections and bouts of irritability and anger. As part of the patient programme a small piece of paper was submitted to the NRC toxicology laboratory with 5F-MDMB-PICA and 5F-MDMB-PINACA being detected. Whilst synthetic cannabinoids are more commonly encountered as substances on plant material for smoking, in some situations (typically prisons), they may be present in paper items. These cases represent the less common form but provide useful confirmation of synthetic cannabinoids in the UAE.

Probe Electrospray Ionization Mass Spectrometry (PESI) : an application for testing driving under the influence of drugs (DUID) in saliva

Dr. Pauline Griffeuille, Dr. Elies Zarrouk, Dr. Souleiman El Balkhi, Prof. Franck Saint-Marcoux

Limoges University Hospital, France


Probe ElectroSpray Ionization (PESI) is an ambiant ionization method where a disposable solid needle is used as a sample probe and an Electrospray Ionization (ESI) is used as an emitter. Coupled with a mass spectrometer, it allows very fast analyses with a minimal sample preparation.

Our objectives were to develop a method for the measurement of illicit drugs in saliva using a PESI-MS/MS approach.


The following conditions were defined : 50 µL of saliva are mixed in a methanol/ethanol/ammonium formate buffer. After adding deuterated internal standards, 10 µL are set in the PESI. A scheduled Multiple Reaction Monitoring (MRM) approach with 2 transitions per compound was developed. Total run time was about 1,5 minute. According to the ISO 15189 standards, the validation steps included a determination of the limit of detection (LDD), the limit of quantification (LDQ), repetability and reproducibility analysis.


Satisfactory results were obtained for 8 opiates, amphetamines and cocaine derivatives. Especially, LDD were below the limits defined by the law (10 ng/mL). In a sample of 52 real cases, the new method yielded results comparable to those of a LC-MS/MS method routinely used in the lab for confirmation of a positive test to Driving Under the Influence of Drugs (DUID).


It is feasible to use PESI-MS/MS for a very fast measurement of illicit drugs in saliva, especially in authorized labs using LC-MS/MS methods for DUID testing.

Confirmation of Methylphenidate and Ritalinic Acid in Urine Using Liquid Chromatography Tandem Mass Spectrometry (LC-MS/MS)

Prof. Loralie J Langman, Darlington Danso, Dr. Paul J Jannetto

Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, MN, USA

Introduction: Methylphenidate is a phenethylamine derivative with psychostimulant properties similar to amphetamine and commonly used to treat attention-deficit hyperactivity disorder and narcolepsy. Methylphenidate is metabolized to ritalinic acid (RA). The aim of this study was to validate an LC-MS/MS method for the quantification of methylphenidate and RA in urine.

Methods: A “dilute-and-shoot” method for urine was developed in which 10 µL of standards, controls, blanks, and patients were mixed with 50 µL of the deuterium labeled internal standards, and 1700 µL reagent grade water. Five µL was then injected on the ABSciex 6500 LC-MS/MS

Results: The analytical measuring range was linear from 10-2000ng/mL for methylphenidate and 50-10,000ng/mL for RA. The intra- and inter-assay (n=20) imprecision CVs were <5% throughout the analytical range. Analytical accuracy was determined with a minimum of 40 patient samples, and showed good agreement (slope=1.088, r2=0.9902, and slope=0.9508, r2=0.9857 for methylphenidate and RA respectively). Studies showed no significant (<20% of the LOQ) carryover up to 50,000ng/mL. No interferences were seen with the top 25 prescribed and over the counter, and, commonly co-prescribed drugs in the USA, potential known interferences, and common drugs of abuse. Additionally, no significant ion suppression or enhancement was identified.

Conclusions: A highly specific and sensitive LC-MS/MS assay was validated and shown to be free of interferences. The timely and accurate quantification of methylphenidate and RA in urine is important for both clinical and forensic applications for the monitoring of therapeutic compliance and potential misuse. A quantitative assay can provide insight into atypical metabolic ratios suggestive of alternative metabolic pathway or altered pharmacokinetics. Additionally, it can be used to identify if an individual may have added (spiked) the drug directly to the urine sample in attempt to simulate compliance with their therapeutic regimen.

Analytical identification of poisoning by the new synthetic opioid 2-methyl–AP-237

Dr. Azzurra Schicchi1, Dr. Antonella Valli2, Dr. Pietro Papa2, Dr. Maria Teresa Novelli3, Dr. Teresa Coccini1, Dr. Elisa Roda1, Dr. Davide Lonati1, Dr. Valeria M Petrolini1, Prof. Carlo Locatelli1

1Toxicology Unit, Pavia Poison Centre-National Toxicology Information Centre, Clinical and Experimental Laboratory, IRCCS Pavia Hospital, ICS Maugeri SpA-SB, Pavia, Italy; 2Laboratory of Analytical Toxicology, Clinical Chemistry Service, IRCCS Policlinico San Matteo Foundation, Pavia, Italy; 3Intensive Care Unit, Cardinal Massaia Hospital, Asti, Italy

2-methyl-AP-237 is novel psychoactive substance (NPS) classified as non-fentanyl opioid: no toxicological effects in humans are yet available in the scientific literature. We describe a case of a 20 years-old man that presented to emergency department (ED) with worsening dyspnoea that appeared over 48-hours after sniffing this NPS. He was conscious (GCS 15) and hearth rate, blood pressure and body temperature were normal. Arterial blood gas showed hypoxemic normocapnic type I respiratory failure. Biochemistry was normal except for WBC count 20.006 cells/µL and PCR 64 mg/L. Chest X-ray and CT-scan showed respectively a mediastinal widening and bilateral parenchymal ground glass opacity with mostly bilateral near-hilum distribution. The patient was treated with humidified oxygen via high-flow nasal cannula and fully recovered 10-days after hospitalization. Blood sample and the consumed substance (collected with patient’s consent) were both analyzed. The product, a powder labelled as “Research Chemical Powder 2-methyl-AP-237”, was analysed by gas-chromatography mass-spectrometry (GC-MS) full scan mode and identified as methyl-AP-237 by comparison with electronic libraries (unavailability of certified standards). Blood was screened for conventional drugs of abuse (opiates, cocaine, amphetamines, cannabinoids, methadone, buprenorphine, oxycodone, and tramadol) by LC-MS-MS, Multiple Reaction Monitoring (MRM) mode, and for NPS by LC-MS-MS, MRM mode and GC-MS, full scan mode. Blood concentration resulted 25 ng/mL for methyl-AP-237 (LC-MS/MS, using the patient’s product as analytical standard). No other substances were detected in the product and blood. The availability of new synthetic opioids on the drug market is increasing, and the related clinical pictures can be different from the typical opioid toxidrome. This case underlines the importance of specialized laboratory analysis to identify NPS which can explain unusual clinical presentation. Electronic libraries circulate by the Early Warning Systems (such as the European, by EMCDDA) are essential for the identifications of NPS until the availability of certified standards.

Haloperidol serum level after accidental intravenous administration of the decanoate salt

Dr. Valeria M Petrolini1, Dr. Azzurra Schicchi1, Dr. Davide Lonati1, Dr. Eleonora Buscaglia1, Dr. Giulia Scaravaggi1, Dr. Antonella Valli2, Dr. Pietro Papa2, Prof. Carlo Locatelli1

1IRCCS Pavia Hospital, Istituti Clinici Scientifici Maugeri, Pavia, Italy, Italy; 2Laboratory of Analytical Toxicology, Clinical Chemistry Service, IRCCS Policlinico San Matteo Foundation, Pavia, Italy

Haloperidol-decanoate (HD) is synthesized by esterification of the active drug to a long chain fatty acid (decanoic acid), and then dissolved in a vegetable oil. After intramuscular absorption, the hydrolysis to haloperidol occurs in lymphatic system, and the plasma concentration gradually rises (peak: 6 days; apparent half-life: 3 weeks). We report haloperidol serum analysis in two cases of erroneous intravenous (IV) administrations of HD (1 vial: 50 mg) instead of intramuscular route.

Clinical cases. A 55-year-old men received 1 vial in a one-hour intravenous infusion. Serum haloperidol resulted undetectable 8 hours later. The patient was admitted into ICU to be monitored, remained asymptomatic and returned to the psychiatry ward the day after. The second case refers to 89-year-old women that received 1 vial IV. Serum level resulted 1.1 ng/ml one hour after administration and undetectable 24, 48 and 72 hours later. She was monitored for two days: no neurological impairment or rhythm disturbances were observed.

Intravenous haloperidol administration is currently considered contraindicated because of its relationship with cardiotoxicity. However, the effects of the erroneous intravenous administration of decanoate form are unknown. A previous report described tachycardia in an agitated patient who received intravenous haloperidol decanoate (1). In our cases the accidental administration of intravenous haloperidol decanoate was not associated with adverse effects, neither pulmonary embolism due to oily vehicle. Serum level were undetectable (HPLC method, LOD 0.5 ng/mL) or very low (1.1 ng/ml while therapeutic range of 5.6-16.9 ng/ml and a recommended target concentration of 10 ng/ml) (2). It is possible to speculate that hydrolysis of the decanoate formulation occurs slowly also in the blood.

References: 1. Heinz P. Versehentliche intravenose Applikation von Haloperidol-Decanoat. Psichiatr Prax, 1996;23(1):42;2. Ulrich S et al. The relationship between serum concentration and therapeutic effect of haloperidol in patients with acute schizophrenia. Clin Pharmacokinet 1998;34(3):227-263

3:30pm - 4:20pmAfternoon tea&Coffee, exhibition and poster viewing
Location: Angelicum Congress Centre
Angelicum Congress Centre 
4:30pm - 6:00pmSymposium 12: Optimal sampling strategies for TDM exemplified by vancomycin
Location: Aula Magna - Angelicum Congress Centre
Session Chair: Oscar Della Pasqua

04.30 pm: Talk 35: Common PK/PD targets for TDM and model-informed precision dosing and what the trough sample does (not) tell. Sebastian Wicha

05.00 pm: Talk 36: Optimal sampling is not just for trial design: application to individual patients. Michael Neely

05.30 pm:Talk 37: An interactive introduction to AUC-based dosing of vancomycin. Iris Minichmayr

Aula Magna - Angelicum Congress Centre 
4:30pm - 6:00pmSymposium 13: New frontiers of clinical toxicology labs
Location: Room A - Angelicum Congress Centre
Session Chair: Eberhard Wieland

04.30 pm: Talk 38: Methadone, oxycodone and buprenorphine: are polymorphisms and metabolic evaluation of clinical relevance? Donata Favretto

05.00 pm: Talk 39: Amanitins testing in mushrooms poisoning. Elisa Roda

05.30 pm: Talk 40: Biosensor-enabled on-site TDM. Ceren Ates

Room A - Angelicum Congress Centre 
4:30pm - 6:00pmSymposium 14: Clinical implementation of Pharmacogenomics
Location: Room B - Angelicum Congress Centre

Session Chair:Jesse J Swen

04.30 pm: Talk 41: Development of pharmacogenetics guidelines: from reactive to pre-emptive. Jesse J Swen

05.00 pm: Talk 42: Implementation of pharmacogenetics in the oncological clinical practice: the experience of an Italian IRCCS- Cancer Research Center. Erika Cecchin

05.30 pm: Talk 43: Pharmacogenetics for CYP3A4: ready for clinical implementation? Ron van Schaik

Room B - Angelicum Congress Centre 
4:30pm - 6:00pmOral Session: TDM in oncology
Location: Room C - Angelicum Congress Centre
Session Chair: Antonello Di Paolo
Session Chair: Romano Danesi
Room C - Angelicum Congress Centre 

Therapeutic drug monitoring of docetaxel in patients with advanced non-small cell lung cancer (NSCLC): An open label randomized study.

Dr. Salvatore J Salamone1, Dr. Yungpeng Yang2, Dr. Yuniang Ma2, Dr. Yuehao Lin2, Dr. Weiting Liang2, Dr. Hungyun Zhao2, Dr. Li Zhang2

1Saladax Biomedical, Inc., United States of America; 2Sun Yat-Sen University Cancer Center, Guangzhou, China


This study (NCT01891123) aimed to compare the safety and efficacy of dosing by therapeutic drug monitoring (TDM) versus body surface area (BSA) of docetaxel in patients with non-small cell lung cancer (NSCLC).


Patients with advanced NSCLC were randomly assigned (1:1) to receive 3-weekly docetaxel (≤6 cycles) either with TDM (arm-A) or standard 75mg/m2 BSA dosing (arm-B). In arm-A, subsequent doses after cycle 1 were adjusted to an area under time-concentration curve (AUC) of 2.5-3.7μg/mL·h by previous-cycle AUC determined from two blood samples. Patients in arm-B were dosed based on BSA and dose adjustments were made according to the drug label. Primary endpoint was severe neutropenia and secondary endpoints were neuropathy, response rate and progression free survival. The plasma concentration of docetaxel was determined utilizing the MyDocetaxel™ immunoassay and AUC was calculated using nonlinear mixed-effect modelling program.


99 patients were randomized (arm-A 50, arm-B 49) and treated. TDM significantly reduced doses (75 to 61mg/m2, p<0.001), AUC overexposure (53% to 32%, p=0.012) and variation (CV 63% to 42%). Grade 3-4 neutropenia rate was significantly (p<.001) reduced in arm-A versus arm B at cycle 2 (35% vs. 83%, respectively. A significant lower cumulative hazard (0.41, 95%CI 0.17-0.98, p=0.037) of grade 2-4 neuropathy was observed in arm-A versus arm B. The overall response rate ORR and progression free survival between the arms was not significant.


TDM reduces the risk of docetaxel related neutropenia and neuropath and improves benefit-risk profile in patients with NSCLC.

Therapeutic drug monitoring based dosing of oral targeted anticancer therapies: a prospective multicenter study

Ruben Alexander Georg van Eerden1, Dr. Stefanie L. Groenland2, Kim Westerdijk3, Marinda Meertens2, Dr. Stijn L.W. Koolen1,4, Dr. Dirk Jan A.R. Moes5, Niels de Vries6, Dr. Hilde Rosing6, Dr. Hans-Martin Otten7, Dr. Annelie J.E. Vulink8, Dr. Ingrid M.E. Desar3, Dr. Alexander L.T. Imholz9, Prof. Hans Gelderblom10, Dr. Nielka P. van Erp11, Prof. Jos H. Beijnen6,12, Prof. Alwin D.R. Huitema6,13, Prof. Ron H.J. Mathijssen1, Dr. Neeltje Steeghs2

1Department of Medical Oncology, Erasmus MC Cancer Institute, Rotterdam, The Netherlands; 2Department of Clinical Pharmacology, Division of Medical Oncology, The Netherlands Cancer Institute, Amsterdam, The Netherlands; 3Department of Medical Oncology, Radboud University Medical Center, Nijmegen, The Netherlands; 4Department of Pharmacy, Erasmus Medical Center, Rotterdam, The Netherlands; 5Department of Clinical Pharmacy & Toxicology, Leiden University Medical Center, Leiden, The Netherlands; 6Department of Pharmacy & Pharmacology, The Netherlands Cancer Institute, Amsterdam, The Netherlands; 7Department of Medical Oncology, Meander Medical Center, Amersfoort, The Netherlands; 8Department of Medical Oncology, Reinier de Graaf Hospital, Delft, The Netherlands; 9Department of Medical Oncology, Deventer Hospital, Deventer, The Netherlands; 10Department of Medical Oncology, Leiden University Medical Center, Leiden, The Netherlands; 11Department of Pharmacology, Radboud University Medical Center, Nijmegen, The Netherlands; 12Department of Pharmaceutical Sciences, Utrecht University, Utrecht, The Netherlands; 13Department of Clinical Pharmacy, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands


Oral targeted anticancer therapies are still prescribed at a fixed dose despite their high interpatient variability. Consequently, the risk for overdosing resulting in toxicity or underdosing leading to ineffectiveness is high. For several of these drugs exposure-response relationships have been described which makes them highly suitable for Therapeutic Drug Monitoring (TDM), which is dose personalization based on drug levels.


We performed a prospective nationwide multicenter study (;NL6695) in which TDM was implemented if a patient started with one of the 24 drugs included in this protocol. Trough levels were measured after 4, 8, and 12 weeks and every 12 weeks hereafter. If a trough level was below the predefined target and toxicity profile was manageable, a PK-guided intervention (i.e. checking compliance and drug-drug interactions, concomitant intake with food, splitting the intake or dose increments) was performed. The primary outcome was to halve the proportion of patients with a exposure below the target at the third Ctrough measurement compared with historical data.


Up to now, 600 patients were included of whom 386 patients are evaluable for the primary outcome. In 543 patients ≥1 PK sample(s) were taken and are therefore evaluable for the overall analyses. At the third PK measurement, PK-guided dosing reduced the proportion of underexposed patients by 44% (95% confidence interval 32-54%) compared with historical data. At a certain time point during treatment 261 patients (48.1%) had ≥1 PK samples below the target. PK-guided interventions were performed in 139 of these patients (53.3%), which were successful in 95 out of 124 evaluable patients (76.6%). PK-guided interventions could not be performed in 122 patient, mainly due to toxicity (n=78), physician decision (n=21), or treatment discontinuation (n=16).


TDM based dosing of oral targeted anticancer therapies reduced the proportion of underexposed patients and was feasible in clinical practice.

Therapeutic drug monitoring of tamoxifen to improve adjuvant treatment of hormone sensitive breast cancer

Louwrens Braal1, Agnes Jager1, Esther Oomen-de Hoop1, Justin Westenberg1, Koen Lommen1, Peter de Bruijn1, Mijntje Vastbinder2, Quirine van Rossum-Schornagel3, Martine Thijs-Visser4, Robbert van Alphen5, Liesbeth Struik4, Hanneke Zuetenhorst3, Ron Mathijssen1, Stijn Koolen1

1Erasmus Medical Center, Rotterdam, The Netherlands; 2IJsselland Hospital, Capelle aan den IJssel, The Netherlands; 3Franciscus Gasthuis & Vlietland, Schiedam, The Netherlands; 4Ikazia Hospital, Rotterdam, The Netherlands; 5Elisabeth Tweesteden Hospital, Tilburg, The Netherlands


Endoxifen is the most important active metabolite of tamoxifen. Several (retrospective) studies suggested a minimal, or threshold, endoxifen systemic concentration of 14–16 nM for a lower recurrence rate. The aim of this study was to investigate the feasibility of reaching a predefined endoxifen level (≥16 nM; 5.97 ng/mL) --over time-- using therapeutic drug monitoring (TDM).


In this prospective, open label, intervention study, patients who started treatment with a standard dose of tamoxifen -- 20 mg once daily -- for early breast cancer were enrolled. An outpatient visit was combined with a TDM sample at 3, 4.5 and 6 months after initiation of the tamoxifen treatment. The tamoxifen dose was escalated to a maximum of 40 mg if patients had an endoxifen concentration below 16 nM. Primary endpoint of the study was the percentage of patients with an endoxifen level ≥16 nM at 6 months after the start of therapy compared with historical data.


In total, 145 patients were included. After 6 months, 89% of the patients had endoxifen levels ≥16 nM, compared with a literature based 80% of patients with endoxifen levels ≥16 nM at baseline (95% CI, 82 to 94%; P = .007). In patients with an affected CYP2D6 allele, it was not always feasible to reach the predefined endoxifen level ≥16 nM. No increase in tamoxifen related adverse events was reported after dose escalation.


This study demonstrates that it is feasible to increase the percentage of patients with endoxifen levels ≥16 nM by means of TDM. TDM is a safe strategy and offers a possibility to nearly halve the number of patients with endoxifen levels <16 nM.

Determination of blood-plasma ratios of tyrosine kinase inhibitors to allow implementation of (dried) blood microsampling methods.

Nick Verougstraete1,2, Veronique Stove2, Alain Verstraete2, Christope Stove1

1Laboratory of Toxicology, Department of Bioanalysis, Faculty of Pharmaceutical Sciences, Ghent University, Ghent, Belgium; 2Department of Laboratory Medicine, Ghent University Hospital, Ghent, Belgium

Background: Therapeutic drug monitoring (TDM) of tyrosine kinase inhibitors (TKIs) in cancer therapy offers the potential to improve treatment efficacy while minimizing toxicity. Compared to traditional venous blood sampling, (dried blood) microsampling is minimally invasive and easy to perform, enabling sampling at home by the patient. As plasma is still the standard matrix for performing TDM of TKIs, blood-plasma ratios should be known when aiming at implementation of a TDM method via blood microsampling. In this study we determined blood-plasma ratios of eight TKIs used in the treatment of hematologic malignancies.

Material & methods: A high-throughput sensitive LC-MS/MS method, involving simple protein precipitation on merely 50 µL plasma or whole blood, for the simultaneous quantification of bosutinib, dasatinib, gilteritinib, ibrutinib, imatinib, midostaurin, nilotinib and ponatinib was developed and validated. Calibration curve ranges and LLOQs were designed to cover the expected trough therapeutic concentrations. Blood-plasma ratios, and the variation thereof, were determined by analyzing authentic patient samples.

Results: The method was successfully validated according to EMA guidelines and applied on a total of 132 coupled blood and EDTA-plasma samples from 65 patients on TKI treatment. The mean obtained blood-plasma ratios (CV%) were 1.12 (16%) for bosutinib, 1.40 (24%) for dasatinib, 0.97 (17%) for gilteritinib, 1.10 (22%) for ibrutinib, 0.73 (8%) for imatinib, 0.71 (11%) for midostaurin, 0.70 (13%) for nilotinib and 0.92 (13%) for ponatinib. For dasatinib and ibrutinib wide interpatient variabilities were observed; to allow adequate interpretation of the concentrations of these compounds in blood, blood-specific reference values should be established. For bosutinib and gilteritinib larger datasets are required.

Conclusions: The derived blood-plasma ratios can be used for calculating plasma concentrations from dried blood microsamples. Method validation for the quantification of the same panel of TKIs in dried blood microsamples, obtained via volumetric absorptive microsampling (VAMS), is currently in progress.

A Population Pharmacokinetic Approach to Evaluate the Performance of 5-Fluorouracil Therapeutic Drug Monitoring Strategies

Dr. Hayley B Schultz, Dr. Stephanie E Reuter

Quality Use of Medicines and Pharmacy Research Centre, University of South Australia, Australia

Introduction: 5-fluorouracil (5FU) is widely used in the treatment of various cancers. 5FU pharmacokinetics are characterised by high variability, an established concentration-effect relationship and narrow therapeutic index, which support the use of therapeutic drug monitoring (TDM) approaches to improve 5FU safety and efficacy. Nonetheless, dose adjustment of 5FU is complex due to saturable pharmacokinetics and multiple TDM dose adjustment algorithms have been proposed.

Methods: A population pharmacokinetic approach was used to evaluate the probability of target attainment (AUC 18-30 of established TDM algorithms over four treatment cycles. Monte Carlo simulations were conducted for 1000 individuals with characteristics representative of patients receiving 5FU therapy. The results informed the development of a new and improved 5FU TDM algorithm.

Results: Standard BSA-based dosing of 5FU was predicted to result in supra-therapeutic exposure in 80% of individuals. After three subsequent TDM-adjusted cycles, previously proposed dose adjustment algorithms were predicted to reduce the risk of toxicity, with 40-50% of patients exhibiting drug exposure above the upper limit of the therapeutic range. Significant intra-individual variability in drug concentrations was also identified. Based on these findings, an alternate TDM algorithm was developed, which is predicted to result in supra-therapeutic drug exposure in only 5% of patients after one TDM-adjusted cycle.

Conclusion: Due to wide inter-individual variability in 5FU exposure after BSA-based dosing, existing TDM dosing algorithms require numerous dose adjustment cycles to achieve therapeutic targets in the majority of patients. Furthermore, high variability in quantification of drug concentrations limits the ability of current TDM practice to achieve optimal 5FU exposure. An alternate TDM-based dose adjustment approach has the ability to decrease the risk of 5FU-related toxicity, whilst maintaining treatment efficacy.

A combination of pre-screening for DPD deficiency by geno/phenotypinh and PK-guided dosing of 5-FU to prevent severe toxicity in GI-cancer patients (CombiGapp): interim results

Pawida Veluwenkamp-Worawutputtapong1, Daphne Den Besten-Bertholee1, Dr. Helle-Brit Fiebrich-Westra2, Dr. Richard Brohet3, Dr. Jan Gerard Maring1

1Dept. of Clinical Pharmacy, Isala Hospital, The Netherlands; 2Dept. of Oncology, Isala Hospital , The Netherlands; 3Dept. of Sciences & Innovation, Isala Hospital , The Netherlands

Introduction: Dihydropyrimidine dehydrogenase (DPD) is a major enzyme in the 5-FU metabolism pathway. DPD deficient pts have a strong reduced capacity to degrade 5-FU which may result in life-threatening toxicity. For 4 common DPYD variants, the Dutch Pharmacogenetics Working Group recommends a 50% reduced fluoropyrimidine starting dose in heterozygote pts. We studied the pharmacokinetics of 5-FU for these 4 variants, in order to define a safe starting dose for 5-FU in DPD deficient pts. We present the interim results for 2 common DPYD variants.

Methods: All patients are screened for DPD deficiency by DPYD genotyping and separated into five groups: 4 DPYD variants and a DPYD wild type control group (target 8-12 pts per group). TDM is performed to follow-up the 5-FU plasma concentration during chemotherapy. Pts are monitored until a steady state AUC46hr of 20-30 mg.h/L is reached.

Results: Until March 01.2021 we included 9 pts with DPYD variants (E412E(n=6) and D949V(n=3)) and 7 DPYD wt.control pts. The AUC0-48h of 5 out of 7 pts in the control group (100% dose) reached the target AUC (20-30 mg.h/L) within the first cycle whereas of none of the variant pts reached the target AUC on a 50% 5FU dose (range 8.74 to 14.26 mg.h/L). Three E412E variant pts needed an increase to 100% of the standard dose in order to reach the target AUC. Two DPYD D949V variant pts needed 75% and 100% respectively of the standard dose to reach the target AUC.

Conclusion: The current recommendation to treat heterozygotes for the common E412E and D949V DPYD variants with a 50% reduced 5-FU dose resulted in underdosing in all variant pts in our study. Based on these interim results, we urgently advice to treat DPYD variant carriers with a 5-FU infusion based chemotherapy protocol combined with 5-FU TDM for adequate dose-guiding.

Measurement of UH2/U ratio as prognostic screening of Dihydropyrimidine Dehydrogenase activity in 5-Fluorouracil therapy management

Dr. Bruno Charlier1,2, Dr. Albino Coglianese1, Dr. Emanuela De Bellis1, Dr. Federica De Rosa1,2, Dr. Francesca Mensitieri1, Dr. Valentina Manzo2, Dr. Berenice Stefanelli2, Prof. Valeria Conti1,2, Prof. Viviana Izzo1,2, Prof. Fabrizio Dal Piaz1,2, Prof. Amelia Filippelli1,2

1Department of Medicine, Surgery and Dentistry "Scuola Medica Salernitana", University of Salerno, Baronissi.; 2University Hospital "San Giovanni di Dio e Ruggi D'Aragona",

5-fluorouracil (5-FU) is a first-line chemotherapeutic drug, widely used for the treatment of solid tumors. It has been estimated that 5-FU administration can cause severe adverse drug reactions (ADRs) in 10-30% of treated patients. The dihydropyrimidine dehydrogenase (DPD) enzyme, encoded by the DPYD gene, is responsible for the conversion of fluorouracil to inactive form dihydrofluorouracil (5-FUH2). Several single nucleotide polymorphisms (SNPs) in DPYD sequence, causing a partial or complete DPD deficiency, led to active drug accumulation and consequent increased toxicity.The pharmacogenetic analysis of the DPYD gene sequence represents the method of choice for evaluating enzyme activity. However, genotyping does not consider all DPD deficiencies, such as unknown DPYD variants, or non-genetic factors. Other prognostic phenotype-based approaches have been developed to determine the activity of the DPD enzyme. Among these, the measurement of the endogenous dihydrouracil and uracil (UH2/U) ratio has proved reliable in some clinical studies, since uracil is physiologically metabolized to dihydrouracil by DPD. Moreover, European Medicines Agency (EMA) has recognized in recent guidelines the importance of measuring the UH2/U ratio as an additional predictive method for DPD activity evaluation. Main objective of our work was to evaluate DPD activity through plasma quantification of its natural substrate (U) and product (UH2) ratio. Patients plasma samples were analyzed by high performance liquid chromatography coupled with UV detector (HPLC-UV) after protein precipitation and subsequent liquid-liquid extraction. A perspective study was conducted on more than 700 patients, classifying them as extensive or poor metabolizers, depending on their UH2/U ratio. The results obtained from this study highlighted the prognostic importance of the UH2/U ratio. This evaluation, associated with the genetic DPYD test, added important information about DPD activity reduction, thus helping to significantly improve 5-FU safety therapy, thus avoiding patient serious or fatal adverse events due to drug accumulation.

Pilot Therapeutic Drug Monitoring (TDM) Study of Pembrolizumab and Nivolumab Trough Concentrations in Melanoma Patients

Dr. Paul J. Jannetto1, Dr. Maria A. Willrich1, Paula M. Ladwig1, Lisa A. Kottschade2, Jill M. Schimke2, Dr. Svetomir N. Markovic2

1Mayo Clinic, Department of Laboratory Medicine & Pathology, Rochester, MN, United States of America; 2Mayo Clinic, Department of Oncology, Rochester, MN, United States of America

Introduction: Pembrolizumab is a humanized monoclonal IgG4 antibody directed against the programmed death receptor-1 (PD-1). It blocks the interaction between PD-1 and its ligands which enhances the activation of T-cell mediated immune responses against tumor cells. Nivolumab, another human monoclonal antibody (IgG4), also targets PD-1 resulting in enhanced T-cell proliferation and interferon-gamma release. Both are FDA-approved for malignant melanoma monotherapy as intravenous injections (240 mg or 480 mg every 2 or 4 weeks for nivolumab and 200 mg every 3 weeks for pembrolizumab). The purpose of the study was to determine if peak or trough serum concentrations correlated to clinical outcomes or toxicity.

Methods: Twelve melanoma patients (5 prescribed nivolumab, 7 pembrolizumab) at steady state were consented into this IRB approved study. Trough and peak blood samples were collected immediately before and after infusions, processed, and frozen until analysis using monoclonal immunoglobulin Rapid Accurate Mass Measurement (miRAMM). Briefly, IgG4 immunoglobulins and the therapeutics were extracted from serum using a Human IgG4 affinity matrix. After wash steps and elution, a reduction step with dithiothreitol was used to release the light chains for quantification following chromatographic separation and a full scan mass spectrum detection on a Q-TOF mass spectrometer. All patients were also evaluated for adverse events prior to every treatment cycle and radiographically for disease progression.

Results: Trough concentrations of nivolumab and pembrolizumab ranged from 45-198 and 11- 74 mcg/mL, with peak concentrations ranging from 122-416 and 37-163 mcg/mL, respectively. Using a Kruskal-Wallis test, the pembrolizumab trough concentrations were higher in patients who achieved optimal clinical response (complete remission vs. partial response, p=0.0237) and toxicity (e.g. pancreatitis, thyroiditis) compared to no side effects (p=0.0019).

Conclusions: Preliminary data indicated that trough pembrolizumab concentrations may correlate with efficacy and toxicity, but larger studies will be needed to demonstrate the full benefit of TDM.

TDM-guided empirical treatment with beta-lactams in CAR-T patients with febrile neutropenia: proof of concept

Chun Liu1,3, Dr. Pier Giorgio Cojutti1,2, Dr. Milo Gatti1,2, Carla Troisi1,3, Prof. Federico Pea1,2,3

1Department of Medical and Surgical Sciences, Alma Mater Studiorum, University of Bologna - Bologna (Italy); 2Clinical Pharmacology Unit, IRCCS Policlinico Sant'Orsola - Bologna (Italy); 3This project has received funding from the European Union’s Horizon 2020 research and innovation program under the Marie Skłodowska-Curie grant agreement No 861323.


Chimeric Antigen Receptor T-cell (CAR-T) immunotherapy is a novel targeted therapy for cancer. Treatment–related cytokine storm may significantly alter the pharmacokinetics (PK) of beta-lactams that are mainstay of treatment in febrile neutropenic CAR-T patients. Therapeutic drug monitoring (TDM) may be helpful in guiding therapy. Our aim was to explore the PK changes of beta-lactams in relation to cytokine burden trend in CAR-T patients.


TDM-guided individual dose adjustment of piperacillin/tazobactam or meropenem steady-state concentrations (Css) administered by 24h-continuous infusion (CI) were provided and drug clearances (CL) were calculated. Relationships between drug exposure and trend over time of interleukin (IL)-8/10, estimated glomerular filtration rate (eGFR), C-reactive protein (C-RP) and procalcitonin (PCT) were also assessed.


Nine patients were included (5 piperacillin/tazobactam and 4 meropenem). Median(IQR) Css of piperacillin/tazobactam and meropenem were 69.9(51.0-83.2) and 4.85(3.9-5.7) mg/L, respectively. Median(IQR) CL were 10.73(9.01-14.71) L/h for piperacillin/tazobactam and 34.99(29.45-44.65) L/h for meropenem. Median(IQR) eGFR in patients receiving piperacillin and meropenem were 89.8(79.5-96.6) mL/min/1.73m2 and 100.9(96.08-111.10) mL/min/1.73m2, respectively.

Median reduction in interleukin levels after antibiotic treatment was 64.6% for IL8 and 58.3% for IL10 in the piperacillin/tazobactam group. A reduction of 63.9% for IL8 and of 73.3% for IL10 was observed in the meropenem group. Median C-RP and PCT reductions were 87.4% and 65.1%, respectively, for the piperacillin/tazobactam, and 93.5% and 85.7%, respectively, in the meropenem group.


CAR-T patients display higher CL both for piperacillin/tazobactam and meropenem compared to other populations, in which CL values are reported to range 5.6-17.2L/h for piperacillin/tazobactam and 9.4-13.6L/h for meropenem. This may be partly explained the cytokine storm that may play an important role in increasing the pharmacokinetic variability of beta-lactams administered in this population. The findings suggest that in CAR-T patients, beta-lactam dosing should rely on measured rather than estimated creatinine clearance and on real-time TDM-guided adjustments.

6:00pm - 7:30pmCocktail reception and poster viewing
Location: Angelicum Congress Centre

Dinner at Leisure

Angelicum Congress Centre 
Location: Palazzo Brancaccio
Palazzo Brancaccio 
Date: Wednesday, 22/Sept/2021
7:30am - 8:30amMorning roundtable session: Interpretation of metal concentrations in blood and urine
Location: Hotel The Hive

Chair: Souleiman El Balkhi

Co-Chair: David Kinniburgh

Antibiotic stability in elastomeric pumps used for outpatient parenteral antimicrobial therapy Alternative sampling devices to collect dried blood microsamples: state-of-the-Art TDM in an immune-oncology word

Hotel The Hive 
7:30am - 8:30amMorning roundtable session: Pharmacogenetics in psychiatry: is the glass half empty or half full?
Location: Hotel The Hive

Chair: Ron Van Schaik

Antibiotic stability in elastomeric pumps used for outpatient parenteral antimicrobial therapy Alternative sampling devices to collect dried blood microsamples: state-of-the-Art TDM in an immune-oncology word

Hotel The Hive 
7:30am - 8:30amMorning roundtable session: The use of machine learning approaches in clinical pharmacology
Location: Hotel The Hive

Chair: Jean-Baptiste Woillard 

Antibiotic stability in elastomeric pumps used for outpatient parenteral antimicrobial therapy Alternative sampling devices to collect dried blood microsamples: state-of-the-Art TDM in an immune-oncology word

Hotel The Hive 
8:00am - 5:30pmCongress registration
Location: Sala delle Colonne - Angelicum Congress Centre
Sala delle Colonne - Angelicum Congress Centre 
8:30am - 10:00amSymposium 15: Infection site measurement of anti-microbial agents: unnecessary, a luxury or essential?
Location: Aula Magna - Angelicum Congress Centre

Session Chair: Federico Pea

08.30 am: Talk 44: Bone and joint infections. Birgit Koch

09.00 am: Talk 45: Lungs and pleural space. Federico Pea

09.30 am: Talk 46: The role of microdialysis. Markus Zeitlinger

Aula Magna - Angelicum Congress Centre 
8:30am - 10:00amSymposium 16: Herbal remedies and dietary supplements
Location: Room A - Angelicum Congress Centre
Session Chair: Simona Pichini

08.30 am: Talk 47: Amanita Phalloides intoxications: clinical outcomes and efficacy of antidotes – Review of 40 years of case reports. Bart G.J. Dekkers

08.55 am: Talk 48: Herbal remedies: toxicity and effect on therapeutic drug monitoring. Vanessa Steenkamp

09.20 am: Talk 49: Case study of curcumin food supplements: an example of public health issue. Marco Silano

09.45 am: Talk 50: Herbal and dietary supplements use leading to toxicities. Amitava Dasgupta 

Room A - Angelicum Congress Centre 
8:30am - 10:00amOral Session: Alternative sampling strategies
Location: Room B - Angelicum Congress Centre
Session Chair: Ugo de Grazia
Session Chair: Giuliana Cangemi
Room B - Angelicum Congress Centre 

Alternative microsampling approaches and HRMS for the monitoring of neurodegenerative disorders

Dr. Michele Protti1, Dr. James Rudge2, Prof. Mario Amore3, Prof. Alessandro Serretti4, Prof. Laura Mercolini1

1Research group of Pharmaco-Toxicological Analysis (PTA Lab), Department of Pharmacy and Biotechnology (FaBiT), Alma Mater Studiorum - University of Bologna, Bologna, Italy; 2Neoteryx LLC, Torrance, CA, USA; 3Department of Neuroscience, Ophthalmology, Genetics and Infant-Maternal Science, University of Genoa, Genoa, Italy; 4Department of Biomedical and Neuromotor Science, Alma Mater Studiorum - University of Bologna, Bologna, Italy


Neurodegenerative disorders refer to neuronal degradation and dysfunctions: Alzheimer's, Parkinson's diseases and amyotrophic lateral sclerosis are regarded as the major ones. Currently, effective treatments for CNS depressive symptoms associated with neurodegenerative diseases involve the use of “new-generation” antidepressant agents (ADA). In order to correctly assess therapeutic options, alternative and patient-friendly microsampling strategies are particularly attractive to promote accurate TDM practices.


Microsampling approaches were developed, based on dried matrix spot (DMS) and volumetric absorptive microsampling (VAMS). In comparison with routine protocols, miniaturised haematic samples are minimally invasive, using few drops of capillary blood from a fingerprick, with promising future applications for home-sampling. Moreover, when using dried samples, cryopreservation for transport and storage is not required while usually granting high analyte stability. Whole blood microsampling was studied for several ADA together with their active metabolites, exploiting high resolution mass spectrometry (HRMS). In addition, oral fluid microsampling was investigated as possible alternative matrix in TDM. All the original procedures were optimised and validated, then compared to classic fluid samples.


Method validation results were satisfactory in terms of sensitivity, extraction yield (85%-95% range) and precision (RSD always <8.5%). Moreover, enhanced stability was observed for dried microsamples stored at room temperature (>88% after 3 months), when compared to fluid counterparts stored at controlled temperatures. Finally, good agreement from patient TDM data was demonstrated between microsampling and reference in-tube sample analysis.


Dried microsampling approaches provided sound and reliable results allowing a feasible analysis of ADA in subjects suffering from neurodegenerative diseases, with high potential within an increasingly patient-centric scenario. In fact, more frequent TDM could enhance precision medicine practices and provide novel diagnostic tools for remote sampling, when coupled to high-throughput, cost-effective analysis workflows. This collaborative research work was made possible by the Research Projects of National Relevance (PRIN) 2017 funds.

Fully Automated Dried Blood Spot-Based Therapeutic Drug Monitoring of Immunosuppressants

Sigrid Deprez, Prof. Christophe Stove

Laboratory of Toxicology, Department of Bioanalysis, Faculty of Pharmaceutical Sciences, Ghent University, Ottergemsesteenweg 460, 9000 Ghent, Belgium

Introduction: In therapeutic drug monitoring (TDM), there is an increased interest in dried blood microsampling. Sampling of dried blood spots (DBS) on filter paper is the best known strategy, as a less invasive alternative to venous blood sampling. Key advantages are the possibility to perform home sampling and the amenability for automation. The aim of this study was to develop, validate and compare LC-MS/MS methods for the quantification of four immunosuppressants (tacrolimus, everolimus, sirolimus and cyclosporin A) in blood and in DBS, for the latter using a DBS-MS-500 autosampler.

Methods: Method development included optimization of the fully automated extraction procedure. This optimized procedure was applied on venous left-over samples from patients taking immunosuppressants at Ghent University Hospital and was compared with an optimized procedure in whole blood.

Results: The LC-MS/MS methods were successfully validated based on international guidelines, for the DBS method also taking into account DBS-specific parameters. Reproducible (CV<15%) IS-compensated relative recovery values were obtained. However, a hematocrit-dependent relative recovery was observed for the DBS, with lower hematocrit values yielding higher relative recoveries (and vice versa). Relative to the reference hematocrit of 0.37, this difference exceeded 15% at hematocrit extremes (0.18 and 0.60), implying that ideally, for real samples outside the validated recovery range, a hematocrit correction should be applied.

Comparative analysis of blood and DBS for all analytes demonstrated applicability of fully automated DBS analysis. Using a validation set, a hematocrit-based correction algorithm was set up for the automated DBS procedure, which, upon application on an independent set, led to an acceptable agreement of DBS results with whole blood.

Conclusion: We demonstrated the potential of fully automated DBS-based analysis of immunosuppressants in the context of TDM (from card to result, with no hands-on). We are currently evaluating the clinical applicability of the DBS method in a real-life setting.

Development and validation of an analytical procedure suitable for adherence monitoring of antihypertensive drugs using volumetric absorptive microsampling of finger prick blood

Cathy M Jacobs, Dr. Lea Wagmann, Prof. Markus R Meyer

Department of Experimental and Clinical Toxicology, Institute of Experimental and Clinical Pharmacology and Toxicology, Center for Molecular Signaling (PZMS), Saarland University, Homburg, Germany

Introduction: Volumetric absorptive microsampling (VAMS), an emerging microsampling technique, claims to maintain the advantages and overcome the limitations of dried blood spots such as volume inaccuracy and hematocrit (HT) effects. Thus, the current study aimed to develop and validate a VAMS-based analytical procedure suitable for adherence monitoring of 10 frequently prescribed antihypertensive drugs.

Methods: Amlodipine, bisoprolol, candesartan, carvedilol, lercanidipine, losartan (assessed via the carboxylic acid metabolite), metoprolol, nebivolol, telmisartan, and valsartan were included. Sample preparation was performed by hydration of the VAMS-tip and subsequent precipitation with methanol:acetonitrile (30:70, v/v). The supernatant was injected onto a Dionex LC coupled to a ThermoFisher (TF) Q-Exactive. Full chromatographic separation was performed on a TF Accucore Phenyl-Hexyl column (100 mm x 2.1 mm, 2.6 µm particle size) within a total runtime of 10 min. The mass spectrometer was operated in positive parallel reaction monitoring mode. The analytical procedure was validated based on the EMA guideline on bioanalytical method validation and the IATDMCT guideline for development and validation of dried blood spot-based methods for therapeutic drug monitoring.

Results: VAMS enabled a fast and easy sample preparation. Carry over, selectivity, within/between day accuracy and precision met recommended criteria for most analytes, except of e.g. amlodipine. Matrix effects and recovery, determined at HT20%, HT40%, and HT60%, were sufficient even at low or high HT values in most cases. Dilution integrity achieved good results in case blank matrix was used for dilution of the extract, but not in case the precipitation agent was used.

Conclusions: A quantitative method for adherence monitoring of 10 antihypertensive drugs in finger prick blood was successfully developed and validated. The next step will be the application of the current method to assess the adherence of hypertensive patients to their medication by calculating individual dose related references ranges.

Interest of micro-sampling on Mitra® system for baclofen pharmacokinetics study in rats

Dr. Marie Allard1,2, Dr. Lucie Chevillard2, Nefertari Garcia-Moro1, Dr. Minh Le2,4, Dr. Bruno Megarbane2,3, Dr. Pascal Houze1,2, Dr. Laurence Labat1,2

1Laboratoire de Toxicologie, Hopital Lariboisière, APHP, Paris, France; 2Inserm UMR-S 1144, Université de Paris, France; 3Réanimation médicale et toxicologique, Hopital Lariboisière, APHP, Paris, France; 4Laboratoire de Biochimie métabolique et nutrition, Hopital Bichat, APHP, Paris, France

Small animal’s blood volume limits the amount of samples that pre-clinical pharmacokinetic studies require. VAMS (Volumetric Absorptive MicroSampling) systems allow reproducible sampling of few microliters of blood. To validate their use to describe baclofen pharmacokinetics in rats, we developed and validated a new method with the Mitra® system.

Previous validated LCMSMS method for plasma baclofen determination was adapted. 20µL-Mitra® were impregnated with whole blood and dried. Baclofen extraction was performed with water (200µL) and 1% (v/v) formic acid and baclofen-d4 (300ng) was added. Supernatant was evaporated and dry residue was taken up with 50µL of 80/20/0.1 water/acetonitrile/formic acid solution (%, v/v). Ten microliters were injected into Quantum UltraTM LCMSMS system coupled to turbulent-flow chromatography (ThermoFisher). To validate sampling protocol, two baclofen dosages (5mg/kg; 116mg/kg, i.e. 80% of the LD50) were tested on rats. Samples were drawn between 5min and 24h, simultaneously from a catheter and from tail. Plasma determinations in classical conditions were also performed.

This method was linear from 5 to 2000 ng/mL (R2mean=0.992; n=6) with Loq at 5ng/mL. Repeatability and intermediate precision were below 9.74% on three different levels (15, 750 and 1500ng/mL; n=6). Extraction efficiencies were between 98.2 and 115.8% (n=6). Excellent correlations were described between baclofen concentrations on Mitra® system on the two sampling sites, catheter and tail (p<0.0001) and between plasma values. First results allowed description of elimination half-life estimated at 5.5h.

A new baclofen quantification on VAMS system was validated. Sampling with Mitra® on catheter appeared to be the most practical and will allow us to precisely describe baclofen’s kinetic phases in rats, particularly the short absorption phase, which requires multiple samples. Elimination half-life estimation obtained was similar to previously reported data on plasma samples. Mitra® system therefore appears to be promising for kinetic studies in pharmacology or toxicology in small animals.

Near-infrared-based hematocrit prediction of dried blood spots: an in-depth evaluation

Liesl Heughebaert1, Lisa Delahaye1, Dr. Christoph Lühr2, Stijn Lambrecht3, Prof. Christophe Stove1

1Ghent University, Belgium; 2BÜCHI Labortechnik, Germany; 3Ghent University Hospital, Belgium

Background: Dried blood spot (DBS) microsampling has gained interest in different clinical fields, including pediatrics, owing to its many advantages compared to conventional blood sampling. However, whilst being applied for decades for screening purposes, some challenges, such as the hematocrit (Hct) effect, hinder further widespread use of DBS for quantitative purposes in clinical practice. Amongst the approaches that were developed to cope with this issue, is the Hct prediction of DBS using near-infrared (NIR) spectroscopy.

Objectives: The aim of this study was to extensively evaluate a commercially available NIR set-up for the prediction of the Hct from DBS.

Methods: Using left-over venous EDTA-anticoagulated blood from patients, the accuracy and precision, stability, and robustness were assessed. Furthermore, applicability of the method on capillary DBS was evaluated via finger prick samples.

Results: Following actualization of an in-built calibration model, which was needed as an unacceptable negative bias was observed, the method validation resulted in a maximal bias, respectively CV, of 0.013 L/L and 4.5%. The method was robust towards several aspects, including storage (except for storage at 60°C), measurement location, type of filter paper (Whatman 903 vs Ahlström 226) and spotted volume (except for 10 µL spots). Furthermore, the method allowed to discern an altered blood spreading in DBS that had been pressed following collection. In contrast, holding the filter paper at an angle of approximately 45° while collecting the DBS did not relevantly affect the Hct predictions. Finally, the potential to predict the Hct of capillary DBS was demonstrated.

Conclusion: A commercially available NIR set-up was extensively and successfully validated, allowing non-contact Hct prediction of DBS with excellent accuracy and precision. This allows to correct for the Hct-based bias observed in partial-punch DBS analysis and the set-up of blood-plasma conversion factors, increasing the application potential of patient-centric sampling.

Residual blood from blood gas tests can be used for therapeutic drug monitoring of vancomycin: a method comparison

Tim JL Smeets1, Anouk Rietveld1, Ruud Huisman1, Dr. Henrik Endeman2, Dr. Birgit CP Koch1, Dr. Nicole GM Hunfeld1,2

1Erasmus MC, University Medical Center Rotterdam, Department of Hospital Pharmacy, Doctor Molewaterplein 40, 3015 GD Rotterdam, the Netherlands; 2Erasmus MC, University Medical Center Rotterdam, Department of Intensive Care, Doctor Molewaterplein 40, 3015 GD Rotterdam, the Netherlands


Therapeutic drug monitoring (TDM) of vancomycin is standard care during treatment with vancomycin. Regarding sustainability, there is increasing interest in combining different analyzes in the same tube and the use of residual material. In the intensive care unit (ICU) on daily basis up to 12 blood gas tests per patient are analyzed. After analysis, with each syringe a variable volume of blood is discarded. Presumably, this residual material is suitable for TDM of vancomycin. However, the specifics of the heparin-coated syringes for blood gas analysis differ from the normally used ethylenediaminetetraacetic acid (EDTA tubes). Therefore, the aim of our study was to compare both different tubes and determine whether residual blood from blood gas sampling is suitable for TDM of vancomycin.


Samples from both EDTA tube and residual material from the blood gas test were simultaneously collected at the same time. Vancomycin was determined in both matrices with a particle enhanced turbid metric inhibition immunoassay method for EDTA tubes. Based on the ‘guidelines of Clinical & Laboratory Standards Institute’ a method comparison between EDTA tubes and heparin-coated syringes was performed with the Bland-Altman plot analysis.


Prospectively, a total of 24 samples from 18 patients in both EDTA tubes and blood gas syringe were obtained. The Bland-Altman plot revealed no clinically relevant difference between vancomycin plasma concentrations (Figure 1). The mean difference is 2.0% and all samples are within the agreement limit of ± 15%.


Residual blood from blood gas samples can be used for vancomycin analysis for TDM. No relevant differences were found in vancomycin concentration between the two analyzes. The use of residual material may prevent additional blood sampling and reduce the number of nursing procedures. These results may serve as an example for monitoring of other drugs to create a more sustainable TDM process

TDM In Haematic Miniaturised Samples Of Patients Under Polypharmacy

Camilla Marasca1,2, Prof. Marco Menchetti3, Dr. Carmine Petio4, Dr. Roberto Muratori4, Prof. Andrea Cavalli1,2, Prof. Laura Mercolini1, Dr. Michele Protti1

1Department of Pharmacy and Biotechnology (FaBiT), Alma Mater Studiorum – University of Bologna, 40126 Bologna, Italy; 2Italian Institute of Technology (IIT), 16163 Genoa, Italy; 3Department of Biomedical and Neuromotor Sciences, 40123 Bologna, Italy; 4Psychiatric Diagnosis and Care Service, S. Orsola-Malpighi University Hospital, 40138 Bologna, Italy

Introduction: Personalisation of therapeutic drug regimens is crucial for patients under polypharmacy, especially for central nervous system (CNS) drug treatments, in order to increase the therapeutic efficacy and minimise the side effects. For this reason, frequent and patient-friendly sampling and analytical monitoring protocols are advisable especially for polypharmacy regimens involving antipsychotics and antiepileptics.

Methods: With the aim of carrying out frequent and accurate blood sampling in a minimally invasive way and obtaining reliable quantitation of drug and metabolite levels, an alternative haematic microsampling strategy based on an innovative microfluidic technology was designed and developed. This miniaturised approach exploits a microfluidic chip with microchannels for the accurate collection of 10 µL each of whole blood directly from a fingerprick. The device generates four fixed-volume dried blood spots (DBS) independently from the haematocrit value. This novel technology can therefore overcome the well-known volumetric issues associated with classic DBS sampling, while maintaining all the advantages relating to storage and transport at room temperature, thus avoiding cryopreservation otherwise mandatory for classic fluid samples. In addition, a miniaturised pretreatment and a fast HPLC analysis were developed.

Results: The optimised method was validated, showing satisfactory results (sensitivity >10 ng/mL; extraction yield >85%; precision RSD <8.0%) and then successfully applied for the monitoring of patients under polypharmacy, demonstrating to be a patient-friendly alternative to traditional whole blood withdrawal. The comparison with classical reference and routine methodologies have shown how the microsampling proposed herein represents a readily available tool for achieving TDM sound and reliable results.

Conclusion: An accurate, frequent but feasible TDM is necessary in particular for those patients under complex polypharmacy regimens to decrease side effects and hospitalisations, by means of therapy personalisation and precision medicine approaches.

8:30am - 10:00amOral Session: Miscellaneous 2
Location: Room C - Angelicum Congress Centre

Session Chair: Pierluigi Navarra
Session Chair: Dario Cattaneo

Room C - Angelicum Congress Centre 

Tisagenlecleucel in Children and Young Adults: Reverse Translational Research by Using Real-World Safety Data

Dr. Concetta Rafaniello, Dr. Carmen Ferrajolo, Dr. Mario Gaio, Dr. Alessia Zinzi, Prof. Cristina Scavone, Dr. Maria Giuseppa Sullo, Prof. Francesco Rossi, Prof. Liberato Berrino, Prof. Annalisa Capuano

University of Campania "L. Vanvitelli", Italy

Introduction: Tisagenlecleucel has revolutionized the pharmacological approach of relapsed or refractory B-cell acute lymphoblastic leukaemialeukaemia in paediatrics. The safety profile of tisagenlecleucel still needs to be better defined. The aim of this study was a post-marketing evaluation of the safety of tisagenlecleucel through the analysis of the Eudravigilance database with focus on the paediatric population. Method: From 2017 to 2020, one third of Individual Case Safety Reports referring to tisagenlecleucel (117/364) have been collected in paediatrics, on average nine year-old boys. Results: Overall, 92% of the638 adverse events were serious and caused or prolonged hospitalisation. A total of 55 adverse events presented a fatal outcome, mainly due to progression of malignant neoplasm (N = 10; 18.2%), recurrence of acute lymphocytic leukaemia (N = 6; 10.9%) or occurrence of acute lymphocytic leukaemia (N = 5; 9.1%). Cytokine release syndrome was commonly reported after tisagenlecleucel infusion (54/638), followed by pyrexia (45/638) and hypotension (27/638). Only 18/638 events referred to neurotoxicity, none of them resulted in death. More than one third of cases (41/117) were suggestive of therapeutic failure. Conclusions:This first post-marketing analysis confirms pre-approval evidence of the safety profile of tisagenlecleucel in paediatrics. Since only a few years of marketing is available, further followed-up studies need to be performed to investigate longer-term safety of tisagenlecleucel.

Hair analysis, the ideal tool for monitoring adherence to therapy in chronic pharmacological treatments?

Prof. Cristiano Fava, Dr. Francesco Taus, Dr. Marco Ballotari, Prof. Rossella Gottardo, Prof. Pietro Minuz, Prof. Franco Tagliaro

Università di Verona, Verona, Italy


A poor adherence to therapy in chronic treatments has recently been reported in several pathological conditions including hypertension, dyslipidemic conditions and diabetes. This impacts severely public health and pharmaco-economy (see:

On the grounds of the poor effectiveness of the current approaches to evaluate the patients’ adherence to long term pharmacological treatments, the present work aims at testing the potential usefulness of hair analysis based on modern chromatographic and electrophoretic techniques coupled to mass spectrometry. Because of its wide retrospective detection window and its limited invasivity, as already assessed in the field of forensic toxicology, hair analysis looks suitable to provide objective information on the intake of drugs in the recent months before hair collection. However, the limited amount of hair which can be collected and the low concentrations of drugs and metabolites present in the hair matrix require highly sensitive analytical methods, which only recently have become available in pharmaco-toxicology laboratories. On the other hand, the complex composition and still largely unknown interaction of drugs with the hair matrix require high analytical selectivity.


The adopted approach based on liquid chromatography and capillary electrophoresis hyphenated with mass spectrometry proved situatable for a quantitative determination of drugs and metabolites in hair of subjects undergoing chronic treatments with statins and antihypertensive drugs (beta-blockers, calcium antagonists, ACE inhibitors). The methods have been validated according to the guidelines of forensic toxicology.

Results and conclusions.

The preliminary results show significant correlations between parent drug and/or metabolite concentrations in the hair matrix and the medium term “exposure” to therapy. These data strongly support the potential use of hair analysis for an objective monitoring of adherence to long term therapy.

Binding interactions with sevelamer and polystyrene sulfonate in vitro

Inge Regina Francina van Berlo-van de Laar1, Ilona Prins-Can1, Aliesa Anne de Lange2, Prof. Katja Taxis2, Dr. Franciscus Gerhardus Antonius Jansman1

1Deventer Teaching Hospital, The Netherlands; 2University of Groningen, The Netherlands

Introduction: Sevelamer and polystyrene sulfonate can bind concomitantly administered drugs in the gastrointestinal tract, decreasing their bioavailability and clinical effectiveness. The aim of this study was to identify potential new binding interactions by assessing the relative binding of different drugs to sevelamer and to polystyrene sulfonate in vitro.

Methods: Twenty-eight drugs were selected from a previous in silico study on co-dispensed drugs with sevelamer and polystyrene sulfonate and an assessment of pKa- and log P values of these drugs. The relative binding was determined by dissolving these drugs alone (= control), with 800mg sevelamer and with 15g polystyrene sulfonate at different pH levels (1.5, 5.5 and 7.4), respectively. After incubation at 37°C and shaking for 60 minutes, the solutions were diluted, centrifuged and drug concentrations were quantified with validated analytical assays. The binding assays were performed in three-fold. The mean relative binding (MRB) at each pH level was calculated. A MRB > 20% for at least one pH level was considered as relevant binding.

Results: Drugs with relevant binding to sevelamer (highest MRB (%±standard deviation) of the different pH levels) were salicylic acid (85±2%), flucloxacillin (74±6%), amiodarone (58±20%), sulfamethoxazole (54±3%), trimethoprim (53±4%), sertraline (45±22%), amitriptyline (44±14%), imipramine (38±7%), mirtazapine (38±1%), clomipramine (31±13%), duloxetine (29±12%), haloperidol (24±6%), fluvoxamine (22±6%) and phenytoin (21±10%). For polystyrene sulfonate, these concerned duloxetine (100±0%), sertraline (100±0%), amitriptyline (99±0%), aripiprazole (99±0%), citalopram (99±0%), clomipramine (99±0%), clozapine (99±0%), imipramine (99±0%), nortriptyline (99±0%), risperidone (99±0%), venlafaxine (99±0%), fluoxetine (98±0%), fluvoxamine (98±0%), haloperidol (98±0%), mirtazapine (98±0%), pipamperone (98±0%), lamotrigine (97±0%), clonazepam (96±0%), metformin (96±0%), paroxetine (95±2%), trimethoprim (94±0%), amiodarone (87±4%) and sulfamethoxazole (86±4%).

Conclusions: This in vitro study identified 14 and 23 potentially new binding interactions with sevelamer and with polystyrene sulfonate respectively, which have to be confirmed in vivo to assess their clinical relevance.

A Systematic Review And Meta-Analysis Of Enzyme-Linked Immunosorbent Spot (Elispot) Assay For Bk Polyomavirus Immune Response Monitoring After Kidney Transplantation

Suwasin Udomkarnjananun1,2, Stephen J. Kerr2, Marith I Francke1, Yingyos Avihingsanon2, Nicole M. van Besouw1, Carla C. Baan1, Dennis A Hesselink1

1Erasmus MC, Netherlands, The; 2Chulalongkorn University, Thailand

Background: BK virus (BKV) infection after kidney transplantation can cause BKV nephropathy (BKVAN) resulting in graft dysfunction and allograft loss. The treatment for BKVAN is reduction of the immunosuppressive load which increases the risk of kidney transplant rejection. There is no biomarker to monitor BKV activity besides BK viral load. The value of the Enzyme-Linked Immunosorbent Spot (ELISPOT) assay as a tool to monitor the recipient's anti-BKV immune response after transplantation was investigated systematically.

Methods: Electronic databases, including MEDLINE, Scopus, and the Cochrane Central Register of Controlled Trials were searched for studies of ELISPOT evaluating the immune response against BKV. BKV status was categorized as "active BKV replication" and as "inactive BKV replication". Random-effects model meta-analysis was performed to determine the diagnostic performance of the ELISPOT assay, which stratified patients into non-responders and responders.

Results: One-hundred twenty-seven articles were identified of which nine were included. Non-responders had an increased risk of having active BKV replication (odds ratio of 71.9 (95%-CI 31.0-167.1). Pooled sensitivity was 0.95 (95%-CI 0.89-0.98) and specificity was 0.88 (95%-CI 0.78-0.94). The standardized mean difference of the number of IFN-γ producing cells between patients with active BKV replication compared with patients who had inactive BKV was -2.09 (95%-CI -2.50, -1.68).

Conclusions: IFN-γ ELISPOT non-responders have a 71.9 higher risk of having active BKV replication. The ELISPOT assay is a useful tool for BKV risk assessment and in combination with BKV load may support clinicians in guiding immunosuppressive therapy in patients with BKV replication.

The Well Described, but Often Overlooked Mycotoxin 3-Nitroproprionic Acid (3-NPA)

Dr. Philip David Walson1,2

1University Medical Center Goettingen, Germany,; 2Department of Clinical Pharmacology

INTRODUCTION: Poisoning by mycotoxin contaminated food is a common global problem, especially in developing countries, that is often overlooked by toxicologists from developed countries. A result of this was demonstrated by a recent case report describing the delayed diagnosis of lethal 3-NPA poisoning after ingestion of ‘spoiled’ coconut milk (Birkelund et al, Emerging Infectious Diseases. 2021;27:278-280).

METHODS: This delayed diagnosis and my toxicology experience prompted this brief review of 3-NPA toxicity.

RESULTS: Ingestion of edible plants contaminated by the mycotoxin 3-NPA produced by a fungus (Arthrinium) can cause irreversible damage to the basal ganglia(Ludolph et al, Can J Neurol Sci.1991;18:492–8); similar to that caused by carbon monoxide, cyanide, hydrogen sulfide, manganese, or methanol. I was unaware of this until presented with a 5 year old, Chinese girl at the Beijing University’s First Hospital who had bilateral lenticular lesions on CT and severe Parkinson’s symptoms secondary to “Moldy Sugar-Cane Poisoning.” This syndrome had been well described, but mostly in Chinese medical journals (Liu et al, Biomed Environ Sci.1992;5:161–77). This case was published as a case report in an English language toxicology journal(Li Ming, J Toxicol Clin Toxicol.1995;33:363-7), but recognition of this toxin remains underappreciated by many western toxicologists.

CONCLUSIONS: These two 3-NPA poisoning cases, and my experience as an Editor-in-Chief of three English language medical journals, have convinced me that western toxicologists, toxicology journals, and medical societies (e.g., IATDMCT) would benefit if we could overcome the publication barriers and lack of recognition faced by practitioners in resource poor, non-English speaking environments.

8:30am - 10:00amSeminario Intersocietario (ECM): Anticorpi Monoclonali
Location: Room D - Angelicum Congress Centre
Session Chair: Antonello Di Paolo

8.30 am: Saluti ed apertura dei lavori

8.35 am: Talk A: Anticorpi monoclonali nel crohn pediatrico. Monica Malamisura 

9.05 am: Talk B: Impiego dei monoclonali nelle malattie infiammatorie intestinali dell'adulto. Stefano Festa 

9.20 am: Talk C: Impiego dei monoclonali nel paziente adulto reumatologico. Fabrizio Conti 

9.35 am: Talk D: TDM nei monoclonali: prospettive attuali e future. Antonio D'Avolio 

9.50 am: Discussione 

9.55 am: Considerazioni conclusive

Room D - Angelicum Congress Centre 
10:10am - 10:30amMorning coffee & tea, exhibition and poster viewing
Location: Angelicum Congress Centre
Angelicum Congress Centre 
10:30am - 12:00pmSymposium 17: Young Scientist Committee Symposium 2021
Location: Aula Magna - Angelicum Congress Centre

Session Chair: Laure Elens

10.30 am: Talk 51: Social media, a tool for phamacology education. Picard Nicolas

11.00 am: Talk 52: SimPHARM: a state-of-the-art simulation software to empower students. Stephen Duffull

11.30 am: Talk 53: Bioassays for activity-based screening and functional characterization of New Psychoactive Substances. Eline Pottie


Aula Magna - Angelicum Congress Centre 
10:30am - 12:00pmSymposium 18: Special Analytical evaluations in child/sexual abuse and in hospitalized patients
Location: Room A - Angelicum Congress Centre
Session Chair: Donata Favretto

10.30 am: Talk 54: Delirium in hospital-admitted patients: is pharmaco-toxicological testing appropriate? Philippe Hantson

11.00 am: Talk 55: Child abuse: relevance of the tox lab. Alberto Salomone

11.30 am: Talk 56: DFSA-Sexual abuse. Sabina Strano Rossi

Room A - Angelicum Congress Centre 
10:30am - 12:00pmOral Session: TDM of biologic drugs
Location: Room B - Angelicum Congress Centre
Session Chair: Mario Regazzi
Session Chair: Antonello Di Paolo
Room B - Angelicum Congress Centre 

The utility of Therapeutic Drug Monitoring to identify predictive factors associated to low drug levels and immunogenicity risk during therapy with Anti-Tnf Biological Agents

Dr. Raffaele Simeoli1, Dr. Erminia F. Romeo2, Dr. Valeria Ventura1, Dr. Paola De Angelis2, Dr. Fiammetta Bracci3, Dr. Daniela Knafelz3, Dr. Carlo Dionisi Vici1, Dr. Bianca Maria Goffredo1

1Department of Pediatric subspecialties, Division of Metabolism, Bambino Gesù Children's Hospital-IRCCS, Rome, Italy.; 2Department of Surgery subspecialties, Digestive Endoscopy and Surgery Unit, Bambino Gesù Children's Hospital-IRCCS, Rome, Italy.; 3Department of Pediatric subspecialties, Hepatology Gastroenterology and Nutrition Unit, Bambino Gesù Children's Hospital-IRCCS, Rome, Italy.

Background & Aims: The use of anti-Tumor Necrosis Factor (anti-TNF) biological agents for treatment of moderate to severe inflammatory bowel disease (IBD) has revolutionised the clinical practice. However, many patients with Crohn’s disease (CD) and ulcerative colitis (UC) are often unresponsive to anti-TNF therapy due to an increased drug elimination [1]. Several factors, including serum albumin, C-reactive protein (CRP) and presence of anti-drug antibodies (ADA), increase clearance of biologics [2]. Therapeutic drug monitoring (TDM) of serum drug and ADA levels is helpful to optimize therapy with biologics in IBD patients [3]. Therefore, the aim of this study is to identify factors influencing drug levels combining TDM practice and laboratory parameters. Methods: 33 IBD-pediatric patients, in therapy with Infliximab (IFX) or Adalimumab (ADL), were divided in positive (ADA POS) and negative (ADA NEG) according to their ADA levels. Immunogenicity was defined as presence of ADA titer ≥ 10 AU/mL following at least two consecutive measurements. Both serum ADA and drug levels were immunoenzymatically measured using an ELISA kit (Immunodiagnostik AG). Results: In both ADA positive and negative patients there was a significant inverse correlation between drug and ADA levels (P<0.05 and P<0.01, respectively). Meanwhile, high serum CRP was significantly and inversely correlated with drug concentration only in ADA positive patients (P<0.05). Similarly, significant correlation (P<0.05) between drug and both ADA and CRP levels was confirmed exclusively in ADA positive patients without concomitant immunosuppressant (IMM). Conclusions: In our pediatric population, high ADA and CRP levels could be predictive of low drug amount and therefore, considered as potential risk factors for loss of response. In absence of contemporary IMM therapy or before switching to a second anti-TNF agent, early proactive rather than reactive TDM approach together with human leukocyte antigen (HLA) genotyping, could be useful to avoid therapeutic failures or adverse drug reactions.

One-year experience of therapeutic drug monitoring and pharmacogenetics in anti-TNF therapy

Dr. Stefania Cheli1, Dr. Valeria Cozzi2, Dr. Cristina Montrasio1, Prof. Emilio Clementi1,2, Dr. Dario Cattaneo1

1Unit of Clinical Pharmacology, ASST Fatebenefratelli Sacco University Hospital, Milan, Italy.; 2Clinical Pharmacology Unit, Consiglio Nazionale delle Ricerche Institute of Neuroscience, Department of Biomedical and Clinical Sciences L, Sacco University Hospital, Università degli Studi di Milano, Milan, Italy.

Introduction: The therapeutic use of tumor necrosis factor inhibitors (anti-TNF) has improved the management of chronic inflammatory diseases. However, one third of patients does not respond adequately to the treatment. Lack of response is related mainly to sub-therapeutic drug concentrations due to increased non immune drug clearance and/or immunogenicity. In both cases, the use of anti-TNF agents can be optimized by therapeutic drug monitoring (TDM) coupled with antidrug antibodies (ADAs) monitoring. Here, we reported our one year experience the TDM of TNF-alpha inhibitors. As exploratory analysis, we also looked at the role of genetic variations in the HLA-DQA1 locus and TNF-alpha gene on the response to anti TNF treatment. Methods: TDM was performed using lateral flow rapid tests and ADAs were quantified using ELISA-based kits. All patients were typed for the HLA-DQA1*05 allele by polymerase chain reaction sequence specific oligonucleotide probes and screened for HLA-DQA and TNF-alpha gene polymorphisms by Real-Time PCR. Results: 75% and 25% of patients were treated with infliximab and adalimumab, respectively. The target concentration was reached in about 70% of patients, whereas 19% and 11% had sub-therapeutic and undetectable drug concentrations, respectively. All patients with undetectable concentration developed ADAs and carried the HLA-DQA1*05 variant allele. Of note, patients with TNF-alpha -308AA genotype were found to respond poorly to adalimumab compared with TNF-alpha -308AG and GG. Conclusions: TDM of anti-TNF agents is helpful to identify patients with primary non response and secondary loss of response mainly due to the formation of ADAs. Despite pharmacogenetics is still of limited use in the clinical practice, we have provided preliminary evidence that an association may exists between HLA-DQA1*05 and the development of antibodies against anti-TNFalpha. Moreover individuals homozygous for TNF-alpha -308A seemed to have a poorer response to adalimumab therapy than those with at least one G allele.

Evaluation of six population pharmacokinetic models of adalimumab in patients with inflammatory bowel disease in clinical practice

Silvia Márquez-Megías1, Dr. Amelia Ramón-López1,2, Dr. Patricio Más-Serrano1,2,3, Marcos Díaz-González3, Dr. Ricardo Nalda-Molina1,2

1Miguel Hernández University of Elche, School of Pharmacy, Alicante, Spain; 2Alicante Institute for Health and Biomedical Research (ISABIAL), Alicante, Spain; 3General University Hospital of Alicante, Pharmacy Department, Clinical Pharmacokinetics Unit, Alicante, Spain.

Introduction: Adalimumab is an anti-TNF-α recombinant drug used to treat inflammatory bowel disease (IBD). The dosing of adalimumab can be benefited by model-based therapeutic drug monitoring (TDM). However, there are several Population Pharmacokinetic models (PopPKmod) in the literature. A proper evaluation of these models in the target population should be performed before implanted in a clinical routine.

Objective: To evaluate the model adequacy and the predictive performance of six PopPKmod of adalimumab in adult patients diagnosed with inflammatory bowel disease.

Methods: A retrospective observational study was performed in the General University Hospital of Alicante. Adult patients with IBD treated with adalimumab, with at least two trough plasma concentrations (PCT) between 2014 and 2020 were included.

Six different PopPKmod were evaluated and implemented in NONMEM® v7.4: FDA, 2008 (Mod-1), Ternant et al, 2015 (Mod-2), Sharma et al, 2015 (Mod-3), Berends et al, 2018 (Mod-4), Vande-Casteele et al, 2019 (Mod-5) and Sánchez-Hernandez et al, 2020 (Mod-6).

The individual and population predictions of adalimumab concentrations were estimated from the six PopPKmod, by calculating the EBEs of the individual pharmacokinetic parameters.

To validate the six PopPKmod, the model adequacy (distribution of the EBEs and the NPDE evaluation) and the predictive performance (bias and precision) were performed.

Results: The dataset comprised 134 patients and 398 PCT. The mean values (95% CI) of weight, basal albumin and TC were: 66.7 kg (37 - 102), 3.80 g/dL (2.10- 4.90) and 6.72 mg/L (0 – 30.42), respectively. 8% of patients developed antibodies anti-adalimumab. The model adequacy and the predictive performance are significantly better in Mod-2 and Mod-4

Conclusions: The evaluation of the predictive performance of the six popPKmod found in the literature showed that Mod-2 and Mod-4 performed better in terms of model adequacy and predictive performance.

The impact of anti-infliximab antibodies, infliximab, circulating IL-6 and soluble TNFα in patients with rheumatoid arthritis: a repeated measures study

Miriam Casellas-Gibert1, Helena Colom-Codina2,3, Xavier Juanola-Roura4, Ariadna Padullés-Zamora1,2, F Javier Narvaez-Garcia4, Jordi Bas-Minguet5, Francisco Morandeira-Rego5, Núria Padullés-Zamora1,2,3

1Department of Pharmacy, Hospital Universitari de Bellvitge-HUB; 2Pharmacotherapy, Pharmacogenetics and Pharmaceutical Technology Program, Institut d’Investigació Biomèdica de Bellvitge-IDIBELL; 3School of pharmacy, Universitat de Barcelona-UB; 4Department of Rheumatology, Hospital Universitari de Bellvitge-HUB; 5Department of Immunology, Hospital Universitari de Bellvitge-HUB

Background: Therapeutic drug monitoring of infliximab (IFX) in combination with inflammation biomarkers and clinical response might be relevant to the optimization of IFX in rheumatoid arthritis (RA) patients. Main objective: to identify patient, disease and treatment characteristics that affect Cmin IFX and clinical response. Secondary: to evaluate the effect of patient characteristics, concomitant immunosuppressants, ATI and Cmin IFX in TNFα and IL-6.

Methods: prospective, observational study of repeated measures. 120 blood samples were obtained from a cohort of 22 RA adult patients treated with IFX (July 2014 to August 2017). Cmin IFX, ATI, TNFα, IL-6, erythrocyte sedimentation rate (ESR) and C reactive protein (CRP) were measured. The clinical response was evaluated with the disease activity score in 28 joints using ESR (DAS28-ESR). Clinical, biochemical and demographic data were recorded. We performed a mixed effects analysis to determine the predictive factors of Cmin IFX and clinical response. Statistical analysis was conducted with R 3.4.4.

Results: 22 patients (16 women (72.7%)) were included. The median age and body mass index (BMI) were 62.0 (49.9-70.2) years and 27.6 (23.5-30.5) kg/m2, respectively. 90.9% of the patients received concomitant immunomodulatory (IMM) treatment. Median DAS28-ESR at inclusion was 2.54 (1.95-3.57) (63.6% were in remission or moderate activity). During the study period, median Cmin IFX, IL-6 and TNFα were 0.75 mg/L, 5.90 pg/mL and 19.4 pg/mL, respectively. The intra-patient variability (IPV) for Cmin IFX was 57.3% (41.3-99.5). Four patients developed ATI. Cmin IFX was significantly associated with the presence of ATI in all patients, and with IL-6 concentration in patients with undetectable ATI. Male sex and IL-6 were significantly associated with DAS28-ESR value.

Conclusions: Cmin IFX was significantly associated with the presence of ATI and with IL-6 in ATI negative patients. Predictive variables of clinical response were sex, CRP, ESR and IL-6.

Simultaneous quantification of infliximab and adalimumab in human plasma by liquid chromatography-tandem mass spectrometry using ready-to-use kit and comparison with two ELISA methods

Dr. Camille Tron1,2, Dr. Florian Lemaitre1,2, Dr. Pauline Bros3, Dr. Claire Goulvestre4, Dr. Benedicte Franck1,2, Dr. Sandrine Bagnos3, Dr. Romain Coriat5, Dr. Nihel Khoudour6, Dr. Dorothée Lebert3, Dr. Benoit Blanchet6,7

1University of Rennes, CHU Rennes, Inserm, EHESP, Irset (Institut de recherche en santé, environnement et travail) – UMR_S 1085, F-35000 Rennes, France; 2Inserm, CIC-P 1414 Clinical Investigation Center, Rennes, France; 3Promise Proteomics, 7 Parvis Louis Néel, F-38040 Grenoble, France; 4Department of Immunology, Cochin Hospital, AP-HP Centre, 75014 Paris, France; 5Department of Gastroenterology, Cochin Hospital, AP-HP Centre, Université de Paris, 75014 Paris, France; 6Department of Pharmacokinetics and Pharmacochemistry, Cochin Hospital, AP-HP Centre, 75014 Paris, France; 7UMR8038 CNRS, U1268 INSERM, Faculty of Pharmacy, Université de Paris, PRES Sorbonne Paris Cité, CARPEM, 75006 Paris, France


Most of bioanalytical methods used for therapeutic drug monitoring (TDM) of infliximab (IFX) and adalimumab (ADA) are immunoassays such as enzyme-linked immunosorbent assay (ELISA). Alternative methods like liquid chromatography tandem mass spectrometry (LC-MS/MS) are of increasing interest and offer valuable analytical benefit. This work aimed at validating a LC-MS/MS method using a ready-to-use kit for the simultaneous quantification of IFX and ADA in human plasma, and at comparing its performance with two ELISA methods.


IFX and ADA were extracted from plasma using mAbXmise kit. Full-length stable-isotope-labeled antibodies were used as internal standards. Chromatography was performed on a column BioZen™ 2.6µm Peptide XB-C18 LC 100x2.1 mm (Phenomenex). Mobile phase was composed of a gradient of water/acetonitrile + formic acid 0.1%. Analysis was performed using a 6500+ QTRAP mass spectrometer (Sciex, Framingham, USA).


The method was linear from 2 to 100 µg/mL for both mAbs. Intra-assay and inter-assay accuracy and precision fulfilled acceptance criterion for quantification of IFX and ADA for innovator drugs as well as two biosimilars of IFX. No carry-over was observed. No concerning relative matrix effect was observed while there was an absolute matrix effect on analytes and internal standards. For methods comparison, plasma samples from 105 patients treated for inflammatory bowel diseases (n=70 for IFX, 35 for ADA) were analyzed with the LC-MS/MS method. For IFX, results were in good agreement with those obtained from an in-house ELISA assay (mean bias of -1.8 μg/mL) but there were a significant bias with those obtained from the ELISA commercial kit Lisa-Tracker® (mean bias of -6.1 μg/mL). Regarding ADA, the LC-MS/MS assay was compared to the in-house ELISA assay and results were consistent (mean bias of -1.2 μg/mL).


The LC-MS/MS method appears suitable to be implemented in clinical practice for TDM.

Is trough concentration the best pharmacokinetic parameter for real-world infliximab exposure evaluation?

Dr. Bénédicte Franck1, Dr. Florian Lemaitre1, Xavier Duval1, Dr. Sébastien Lalanne1, Dr. Marie-Clémence Verdier1, Prof. Eric Bellissant1, Prof. Guillaume Bouguen2, Dr. Camille Tron1

1Department of Pharmacology, University Hospital of Rennes, France; 2Department of Gastroenterology, University Hospital of Rennes, France


Infliximab is indicated for the treatment of crohn disease (CD) and ulcerative colitis (UC). The recommended dose is 5 or 10mg/kg at week 0, 2 and 6 then every 8 weeks. The therapeutic drug monitoring of infliximab is commonly performed based on trough concentration (C0) with a target of 3-7µg/mL. However, doses and dosing interval may be adapted to patients' outcome and this C0 target could correspond to a large range of exposure in terms of exposure (area under the curve of the concentrations over the time (AUC)). The aim of the study was to assess the relationship between C0 and AUC0-tau/AUC0-8wk through different dosing regimen.

Methods: We conducted a single center retrospective pharmacokinetic study in adults who received infliximab to treat CD or UC and had available infliximab concentrations between January 2014 and January 2021, per standard of care. A 2-compartment model with first-order elimination and including weight on volumes and weight, antibodies to infliximab, albuminemia and immumodulators drugs on clearance as significant covariates previously published was used to estimate the AUC0-tau/AUC0-8wk (Fasanmade et al., 2011). The correlations between C0 and AUC0-tau/AUC0-8wk were also assessed.

Results: A total of 247 patients contributed 5535 infliximab concentrations. The median [range] age and weight were 33years [16-94] and 65kg [39-150], respectively. The median [range] doses and maintenance dosing interval were 5mg/kg [4.2-10] and 8weeks [3-17], respectively. The median infliximab C0, AUC0-tau and AUC0-8wk were 7.2µg/mL [0.1-88], 26,637µg.h/mL [14,242-69,864] and 28,390µg.h/mL [10,698-170,454], respectively. The correlation between C0 and AUC0-tau and AUC0-8wk showed a correlation coefficient of 0.05 and 0.24, respectively.

Conclusion: Wide ranges of C0, AUC0-tau and AUC0-8wk were found with poor correlation between C0 and AUC. Exposure surrogate using C0 appears to be inappropriate in a clinical setting with various doses and dosing interval. AUC-guided drug monitoring should be preferred.

Validation of a population pharmacokinetic model of vedolizumab: application in clinical practice.

Marcos Díaz-González1,3, Dr. Ricardo Nalda-Molina2,3, Dr. Amelia Ramón-López2,3, Seira Climent1, Dr. Ana Gutierrez-Casbas3, Dr. Juan Selva1,2, Josefa Boada1, Silvia Márquez-Megías2, Mari Luz Boquera1,2, Dr. Patricio Más-Serrano1,2,3

1General University Hospital of Alicante, Pharmacy Department, Clinical Pharmacokinetics Unit, Alicante, Spain; 2Miguel Hernández University of Elche, School of Pharmacy,Elche, Spain; 3Alicante Institute for Health and Biomedical Research (ISABIAL), Alicante, Spain


Vedolizumab is a monoclonal antibody used in the treatment of inflammatory diseases (IBD). The objective was to perform an external validation of a published population pharmacokinetic model (PopPK) of vedolizumab in adult patients with inflammatory bowel disease with the aim of using it for model-based Therapeutic Drug Monitoring-

Material and methods:

Retrospective observational study. All adult patients with IBD treated with vedolizumab with at least one trough plasma concentration (ConcTrough) were included. A population pharmacokinetic model (Rosario et al. 2015) was implemented in NONMEM (v7.3) (bicompartmental model with linear and nonlinear parallel clearance, where albumin and weight are covariates of clearance).

Model evaluation was determined by accuracy (mean relative prediction error (MRPE)) and precision (square root of the mean squares of the relative prediction errors (RMSRPE)). Determination of plasma concentrations was performed by ELISA (TheraDiag (R)) and statistical analysis using the RStudio(R) computer package.


A total of 158 Cpvalle from 32 patients (50% female) were included. The mean Cpvalle was 14.95 mg/L (95%CI: 0.942-35.9 mg/L). The mean population weight was 72.2 kg (50.4 -101.2 kg), with mean albumin levels of 3.62 g/L (2.95 - 4.50). Fifty-six percent of the patients had a diagnosis of ulcerative colitis, while the rest had Crohn's disease. The accuracy (95%CI) was -16.9 (-19.0;-14.8). The precision (95%CI) was 21.5% (19.6;23.4).


The PopPK model of vedolizumab for adult population with IBD evaluated presents an overestimation of ConcTrough, as demonstrated by a negative accuracy and a confidence interval that does not include zero. Before implement a PopPK to perform model-based TDM in clinical practice it must be validate in the population to study.

10:30am - 12:00pmOral Session: Clinical Toxicology - 3
Location: Room C - Angelicum Congress Centre

Session Chair: Carlo Locatelli
Session Chair: Giampiero Frison

Room C - Angelicum Congress Centre 

Accidental intrathecal administration of vincristine: analytical data in a lethal case

Dr. Davide Lonati1, Dr. Azzurra Schicchi1, Dr. Valeria M Petrolini1, Dr. Sara Negri2, Dr. Elisa Roda1, Dr. Teresa Coccini1, Prof. Carlo Locatelli1

1Toxicology Unit, Pavia Poison Center-National Toxicology Information Centre, Clinical and Experimental Laboratory, IRCCS Pavia Hospital, Istituti Clinici Scientifici Maugeri, Pavia, Italy; 2Environmental Research Center, IRCCS Pavia Hospital, Istituti Clinici Scientifici Maugeri, Pavia, Italy

Inadvertent intrathecal administration of vincristine produces severe neurotoxicity with ascending paralysis that can progress to respiratory failure and coma. This therapeutic error, usually lethal, occurs due to drugs confusion (vincristine instead of methotrexate). Immediate cerebrospinal fluid drainage may be useful to prolong survival (Pongudom et al. J Med Assoc Thai, 2011;94:S258-631). We report a lethal case of accidental intrathecal vincristine (1,3 mg) administration instead of methotrexate to a 56-year-old-male with history of leukemia, HCV, and HIV. Despite the patient was initially asymptomatic, an early (4 hours later) external-ventricular-drainage (EVD) was placed and continued for 5 days considering the potential severe toxicity. Pirydossine, folinic acid and cyanocobalamin were also administered. Clinical conditions worsened progressively, and the patient developed paralysis (T10 level) at day 8, respiratory failure and ARDS requiring respiratory support. The patient died on day 31. To monitor the invasive treatment (EVD), a quantitative method in liquor for the vincristine determination was developed (UPLC-MS/MS; Acquity-TQD, Waters) showing the following results in liquor: 4,9 mg/L (before EVD), 0,01 mg/L (11 hours after EVD start), undetectable (starting from day-2). Vincristine resulted all the time negative (LOD=0.004 mg/L) in serum and urine. In conclusion, intrathecal accidental administration of vincristine represents a clinical emergency with possibly dramatic outcome. Continuous EVD (for at least 48 hours) decrease the concentration of vincristine in liquor. Moreover, in our case this was not sufficient to reduce the neurotoxicity, that may appear delayed (1 week). Clinical condition in our case worsened probably also in relation to the severe pre-existing diseases. UPLC-MS/MS-method is suitable for the rapid detection of vincristine in the liquor and can be useful to guide the duration of continuous EVD.

Identification of 3-methylmethcathinone and its metabolites with artificial intelligence and molecular networks.

Florian Stijven1, Prof. Alain Gaston Verstraete1,2

1Department of diagnostic sciences, Ghent University; 2Department of laboratory medicine, Ghent university hospital, Belgium

Introduction: 3-methylmethcathinone (MMC) is a synthetic ring-substituted cathinone. It is a positional isomer of mephedrone, differing only in the position of the methyl group on the phenyl ring. In March 2021, the EMCDDA added it to the list of new psychoactive substances (NPS) under intensive monitoring. The purpose of our research was to differentiate 3-MMC from mephedrone and to explore the metabolism of 3-MMC using Feature Based Molecular Networking (FBMN) Global Natural Products Social Molecular Networking (GNPS) platform. FBMN is a bioinformatic tool that organizes the LC-HRMS data in molecular networks based on similarity between MS/MS-spectra.

Methods: Positive electrospray ionization was used in full-scan (70-700 m/z) on a Thermo Q Exactive with data-dependent acquisition based preferentially, but not exclusively, on an inclusion list of known mephedrone metabolites. The raw data were converted to an open MS format (.mzML) using Proteowizard, and imported into MZMine 2.53 for mass detection, chromatogram building and deconvolution, isotopic peak grouping, alignment and filtering. Features present in the blank urine sample were discarded and only features present in two or more urine samples were retained. The data were exported to GNPS for FBMN. 3-MMC and 4-MMC were differentiated based upon the retention time and a principal components analysis linked to a linear discriminant analysis (PCA-LDA) of the MS/MS-spectra.

Results: The retention times of 3- and 4-MMC differed by only 0.05 minutes, but PCA-LDA confirmed that all samples contained 3-MMC. Currently, 3-carboxy dihydroMMC, 3-carboxy-dihydronorMMC, 3-carboxy-MMC, 3-carboxyMMC glucuronide 3-carboxy-norMMC, 3-hydroxy-dihydroMMC, 3-hydroxytolyl N-acetyl norMMC, 3-norMMC, DihydroMMC, DihydronorMMC, hydroxytolyl-3-MMC, N-acetyl-3-MMC and N-succinyl-3- MMC have been (tentatively) identified using FBMN; only 2 human metabolites of 3-MMC had previously been reported in literature.

Conclusions: FBMN has potential for the identification of NPS metabolites in samples of users, in addition to existing techniques like liver microsome studies.

Albumin isoforms and Serum enhanced binding (SEB) test for early Drugs Induced Liver Injuries (DILI) detection

Dr. Mohamed Ali Rahali1,2, Roy Lakis2, François Ludovic Sauvage2, Prof. Franck Saint-Marcoux1,2, Dr. Souleiman El Balkhi1,2

1CHU Limoges, France; 2University of Limoges, UMR1248, INSERM, France

Background: As Human Serum albumin (HAS) matures inside the hepatocytes, we hypothesized that HSA modifications occur at early stages of cellular hepatic injuries, long before liver parameters. Such modifications produce albumin isoforms and alter the conformation and binding properties of HSA. Therefore, we developed the Serum Enhanced Binding (SEB) Test as a new biomarker for the detection of DILI.

Methods: Hepatotoxicity was induced in 84 Wistar rats. Rats were intoxicated with ethanol, CCl4 or paracetamol for several days, or subjected to bile duct ligation (BDL) (n=6/group/delay). Blood was collected for HSA isoforms determination, the SEB test and liver parameters. Histological injuries were determined on liver tissues. For isoforms determination: serums were diluted and injected on a µLC-tripleTOF 5600 Plus system. HSA spectra were deconvoluted by using PeakView 2.1 for the identification of HSA isoforms. For the SEB test: Cu, Au, Cd, L-thyroxine (LT) and Dansylsarcosine (DS) were incubated with rats serums for 30 min and ultrafiltrated using Amicon® 30 KDa filters. The presence of ligands in the ultrafiltrates signed a modification of albumin. All ligands were measured by ICP-MS.

Results: Several isoforms were detected: acetylated, cysteinylated, glycosylated and glutathinoylated albumin. SEB ligands and the isoforms increases were significantly higher than the controls and had different profiles upon the degree of the liver injuries. The apparition of isoforms and the positivity of the SEB test were observed early, starting day3 in all groups. The classical tests (ALT, AST, PAL,...) significantly increased starting only D7. Cu, DS and Au increases were rapid, starting D1 for both ethanol and paracetamol. Cd increased only at D7 for paracetamol and D14 for ethanol.

Conclusions: The increase of HSA isoforms could be used as an early biomarker for DILI. These liver injuries could be revealed by a very rapid and simple test, the SEB test.

Quantitative plasma concentrations in metformin-related metabolic acidosis

Dr. Azzurra Schicchi1, Dr. Elisa Roda1, Dr. Antonella Valli2, Dr. Pietro Papa2, Dr. Teresa Coccini1, Dr. Davide Lonati1, Dr. Valeria M Petrolini1, Prof. Carlo Locatelli1

1Toxicology Unit, Pavia Poison Center-National Toxicology Information Centre, Clinical and Experimental Laboratory, IRCCS Pavia Hospital, ICS Maugeri SpA-SB, Pavia, Italy; 2Laboratory of Analytical Toxicology, Clinical Chemistry Service, IRCCS Policlinico San Matteo Foundation, Pavia, Italy

Metformin is widely used in the treatment of diabetes mellitus. Due to the important renal excretion, it can accumulate with development of severe metabolic acidosis in patients with dehydration and/or renal failure. Metformin therapeutic drug monitoring can be crucial to prevent accumulation in patients with dehydration risk factors, and to confirm the cause of the acute metabolic acidosis in patients suffering overdoses related to the drug accumulation during chronic treatments. We evaluated the diagnostic value of the metformin plasma determination (MPD) in chronically treated patients referred to our Poison Centre (PPC) between 01/2020-11/2020. Thirty-seven patients were included [27 males, median age 72 (range: 61-80)]. PCC consultation was requested by emergency departments (19/37), ICUs (10/37), and Nephrology units (8/37). In 6 cases the MPD was performed (HPLC) in cases of renal failure without metabolic acidosis. All the remaining 31 patients presented severe lactic acidosis and renal failure at admission. Their dehydration predisposing factors were: gastroenteritis (12/31), infectious disease (10/31), malnutrition (7/31), COVID-19 (2/31). Patients were treated with dialysis or continuous-hemofiltration (34/37) and with volume replacement in 3/37. MPD resulted into the therapeutic range (0-1.99 mg/L) in 13 cases, and above the therapeutic range [median concentrations 15 mg/L (range 7-22)] in 24 cases, which outcome was lethal in 7. In conclusion, metformin plasma determination is important to make the correct diagnosis and to avoid severe complications in patients with dehydration predisposing factor (e.g. gastroenteritis, infectious diseases). Metformin poisoning from accumulation is a severe complication that must be suspected in diabetic patients with lactic acidosis and renal failure: testing this drug is not readily available in health systems, and from the toxicological point of view, MDP should be included in TDM.

First report on in vivo toxicokinetics of Dichlorobisphenol A

Noémie Plattard1,2, Dr. Sami Haddad1, Dr. Antoine Dupuis2,3, Dr. Nicolas Venisse2,3

1Department of Environmental and Occupational Health, School of Public Health, CresP, Université de Montréal, Montreal, Quebec, Canada ; 2INSERM, University Hospital of Poitiers, CHU, University of Poitiers, CIC1402, HEDEX Research Group, 86021, Poitiers Cedex, France ; 3Biology-Pharmacy-Public Health Department, University Hospital of Poitiers, 2 rue de la Milétrie, 86201, Poitiers Cedex, France 

Context: During the water disinfection process, bisphenol A (BPA) present in water can react with chlorine to form chlorinated derivatives of BPA (ClxBPA), including 3,3’-dichlorobisphenol A (Cl2BPA). This emerging pollutant was found in tap water as well as in human biological matrices such as urine, adipose tissue, placental tissue and maternal milk. As an endocrine disruptor, Cl2BPA constitutes a potential risk to human health. Currently, no toxicokinetic data are available to assess exposure to Cl2BPA, which is essential for proper risk assessment.

Objective: The aim of this study is to determine in vivo toxicokinetics of Cl2BPA after administration in rats.

Materials and Methods: 20 males Sprague-Dawley rats (280g) were exposed intravenously to Cl2BPA at two doses: 2.9 mg/kg and 45.1 mg/kg (10 rats per dose). Blood samples were taken from the lateral vein of the tail at 0.25, 0.5, 1, 2, 4, 8 and 24h. After euthanasia at 4 and 24h, liver, brain, adipose tissue and muscles were collected, and urine and feces were harvested. Cl2BPA concentrations in plasma and tissues were determined by using a validated LC-MS/MS assay.

Results and discussion: For both doses, plasma kinetic curves show a rebound in concentration. Quite marked differences are present between the doses in terms of elimination and distribution of Cl2BPA, suggesting dose-dependent kinetics. Indeed, there was no proportional increase of AUC with dose. Lower tissue: plasma concentration ratios were found in adipose tissue in comparison to liver, brain and muscles. At 24h, excretion in feces was much greater compared to urines.

Conclusion: This study reports for the first time in vivo toxicokinetics of Cl2BPA but further investigations are required to understand the mechanisms involved in its disposition and to develop a PBPK model for exposure assessment. Moreover, the toxicokinetic of other ClxBPA are also currently under investigation.

Human biomonitoring: a proposal for a Global Biomonitoring Institute (GBI)

Prof. John Ibhagbemien Anetor, David W. Kinniburgh, Amy M. MacDonald

Alberta Center For Toxicology, Cumming School of Medicine, University of Calgary, Canada, Canada

Introduction: The persistent exposure of the population, including pregnant women and their fetuses in utero to environmental chemicals from varied sources, has given rise to the expanding field of biomonitoring. Biomonitoring is, however, currently disproportionately conducted mainly in advanced countries, calling for globally inclusive approach.

Comment: Human biomonitoring (HBM) is generally regarded as the method of assessing human exposure to chemicals and their effects by measuring specific chemicals, their metabolites or reaction products in human biospecimens. Biomonitoring contributes very useful information to the understanding of the effects of chemical exposures on human health. It is an integrated method of evaluating health risks, their early recognition, intervention, mitigation, and prevention. It may also provide clues to pressing global concerns, shaping public policies. Numerous chemicals and related metabolites may be measured relatively easily with progress in science and technology, but are largely advanced and expensive. Owing to the prohibitive cost involved in biomonitoring, many Low and Medium Income Countries (LMICs) do not conduct formal biomonitoring, rather they are involved in indirect or ‘surrogate biomonitoring ’program (SBP). As HBM is an important tool for supporting environmental health policy formulation, extending beyond regional to global levels, it appears imprudent to leave LIMCs to their own limitations, which may have unintended global backlash, including global insecurity/ terrorism from chemical induced-neurological dysfunction.

Conclusion: A Global Biomonitoring Institute (GBI) is proposed, which will globally evaluate environmental chemicals exposure in maternal population and attendant offspring, where the more economically and technically endowed countries lift up LIMCs to avoid late lessons from early warning or neglect of chemical-induced individual and societal disorders. One lesson the current pandemic has thought the scientific community is to act globally and proactively. Environmental pollutants know no national boundaries, as the SARS- CoV-2 virus has demonstrated very painfully and at a great cost.

Measurement of pesticide residues in bee products: a French study

Dr. Elies Zarrouk, Arnaud Gardère, Dr. Souleiman El Balkhi, Prof. Franck Saint-Marcoux

Limoges University Hospital, France


A decrease in the population of bees has been observed in the last decades. Among the supposed factors, the increase of pesticides use and the nature of the new pesticides are the most suspected culprits. . In the present study, we analyzed different bee products, collected in 2 areas where bees population declined significantly (Lot and south Corrèze: about 2500 km² in the Centre of France): honey, pollen, bee-food, bee-bread, honey-bees and beeswax.


All along 2019, we collected 80 honey samples from 40 different beekeepers, and 27 pollen, bee-food, bee-bread, honey-bee and beeswax samples. Each sample was analysed by taking into account to its location and the human activities around the hive. Different LC-MS/MS and GC-MS/MS methods (accredited according to the International Standards Organization (ISO) 17025 standard) were applied to determine more than 320 different pesticides in each sample.


At least one pesticide residue was detected in 61% of honey samples and 32% of them contained more than 2 molecules. About a half of pollens and more than 80% of the bee-bread and beeswax samples contained at least one pesticide. Interestingly, we detected pesticides in more than one half of the bees. The molecules detected were herbicides (ex: clonidafop, metamitron), insecticides (ex: cyfluthrin, ethofenprox) and antifungals (thiabendazol, propiconazol). No concentration above the Maximum Levels of Residue (MLR) was measured in honey samples and no significant difference was observed when considering the human activities around a hive.


The risk associated with the presence of pesticides in bee products is hardly interpretable but it is likely to be partially responsible for the decline in the population of bees that is observed particularly in the area considered herein.

12:30pm - 1:00pmThermoFisher workshop: Trends & perspectives of anti-epileptic drug therapies and therapeutic drug monitoring, from the clinical lab to the physician.
Location: Aula Magna - Angelicum Congress Centre

Moderator: Dario Cattaneo - Professor at L. Sacco University Hospital

- Speaker 1: Dr Karin Kipper - Head of Therapeutic Drug Monitoring Unit at Epilepsy Society

- Speaker 2: Dr Sara Zagaglia - Honorary Clinical Assistant and Clinical Research Fellow at University College London Hospitals NHS Foundation Trust

Aula Magna - Angelicum Congress Centre 
12:30pm - 1:00pmWaters workshop: The importance of "fast pharmacology" in infectious diseases: the case of antibiotics
Location: Aula Magna - Angelicum Congress Centre

Moderator: Sara Baldelli -  Sara Baldelli, Unit of Clinical Pharmacology, Luigi Sacco University Hospital, ASST Fatebenefratelli Sacco, Milan (Italy)


Aula Magna - Angelicum Congress Centre 
12:30pm - 2:00pmBusiness meeting - TDM in Oncology Committee
Location: Room B - Angelicum Congress Centre

Coordinators: Gareth Veal

Room B - Angelicum Congress Centre 
12:30pm - 2:00pmBusiness meeting - Toxicology and Environmental Health Committee
Location: Room C - Angelicum Congress Centre

Coordinators: Nicolas Venisse, David Kinniburgh

Room C - Angelicum Congress Centre 
12:30pm - 2:00pmBusiness meeting - Anti-Infective Drugs Committee
Location: Room D - Angelicum Congress Centre

Coordinators: Debbie Marriott, Dario Cattaneo

Room D - Angelicum Congress Centre 
2:45pm - 3:30pmPlenary lecture 4: Clinical importance and pharmacological utility of using biomarkers in diagnosis of dementia.
Location: Aula Magna - Angelicum Congress Centre

Plenary Speaker: Manuela Neuman


Aula Magna - Angelicum Congress Centre 
Location: Aula Magna - Angelicum Congress Centre
Aula Magna - Angelicum Congress Centre 
4:30pm - 4:45pmClosing remarks
Location: Aula Magna - Angelicum Congress Centre

Announcement of:

- Best oral presentation Award

- Best Poster Award

- Best Oral on Precision Dosing (in memory of Maria Delfina Molinaro)

Aula Magna - Angelicum Congress Centre 

Contact and Legal Notice · Contact Address:
Privacy Statement · Conference: IATDMCT 2021
Conference Software - ConfTool Pro 2.6.141+TC
© 2001 - 2021 by Dr. H. Weinreich, Hamburg, Germany