SGKC/SSCC Annual Assembly 2019 28th-30th August 2019 Technopark, Zurich or Evidence Based Laboratory Medicine

A 2D HPLC-MS/MS Method for Several Antibiotics

Sophia Rehm¹, Katharina Rentsch¹

¹Laboratory Medicine, University Hospital Basel, Basel, Switzerland

An analytical method for the quantification of four beta-lactam antibiotics (amoxicillin, flucloxacillin, piperacillin, and benzylpenicillin), two beta-lactamase inhibitors (clavulanic acid and tazobactam) as well as the macrolide antibiotic clindamycin in different matrices is presented. The method uses 2D high performance liquid chromatography followed by tandem mass spectrometry. Sample preparation for all matrices was protein precipitation and dilution. As internal standards, deuterated forms of the beta-lactams were used. All analytes were measured in human plasma, while amoxicillin, clavulanic acid, flucloxacillin and clindamycin were also analyzed in human tissue samples. Additionally, the free fraction of flucloxacillin was measured after ultrafiltration. In plasma and plasma water, inaccuracy and imprecision for any analyte were always less than 15%. Measuring quality controls made in various tissues against a calibration in water, the accuracy and precision varied up to 16%, respectively 20% for clavulanic acid, while for all other analytes they were below 15%. Stability of analytes in whole blood and plasma is shown under different conditions and the importance of correct sample handling in beta-lactam analysis is emphasized. The method is in use for routine therapeutic drug monitoring and in several clinical studies and has proofed to be stable and efficient.

A Fast Method for Quantification of Venetoclax by LC-MS/MS

Regula Steiner¹, Daniel Müller¹

¹Institute of Clinical Chemistry, University Hospital Zurich, Zurich, Switzerland

Venetoclax, a BCL-2-inhibitor for the treatment of B-cell lymphoma associated with chromosome 17p deletion, is used more and more in the clinics. Because of severe toxicity, venetoclax has to be titrated at the beginning of the therapy. Further, it is metabolized over CYP3A4 and therefore prone to interactions, which makes therapeutic drug monitoring highly recommended. Therefore, a fast, simple and sensitive high-throughput LC-MS/MS method was developed, using 50 μL of heparin plasma. Plasma was mixed with internal standard (d7-venetoclax) and the precipitation solution and centrifuged afterwards. The supernatant was transferred into autosampler vials and 20 μL was injected into the LC system using reversed phase chromatography with isocratic conditions. Detection was done by electrospray ionization mass spectrometry in the positive ionization mode. The method has been fully validated. The method was shown to be linear over the entire calibration range (0.01-5.0 mg/L) with a lower limit of quantification at 0.01 mg/L. For within-day and between-day analysis, the CV’s were <4.8% and the accuracy was between 95-98%. The method was demonstrated to be free of matrix effects for serum, heparin, and EDTA-plasma. The presented LC-MS/MS method allows a fast, simple and reliable determination of venetoclax.

A New KI Rat Model of Creatine Transporter Deficiency

Lara Duran-Trio¹, Gabriella Fernandes-Pires¹, Marc Loup¹, Jérôme Grosse², Carmen Sandi², Cristina Cudalbu³, Olivier Braissant¹

¹Clinical Chemistry CHUV Lausanne, Lausanne, Switzerland; ²Brain Mind Institute EPFL Lausanne, Lausanne, Switzerland; ³CIBM EPFL Lausanne, Lausanne, Switzerland

Creatine is essential for recycling ATP, is synthesized by a 2-steps pathway (AGAT/GAMT) and transported by SLC6A8. Cerebral creatine deficiencies, due to AGAT, GAMT or SLC6A8 deficiencies and characterized by absence of brain creatine measured by 1H-MRS, cause severe neurodevelopmental delays, intellectual disability and behavioral problems. So far, the X-linked SLC6A8 deficiency cannot be treated and its pathological mechanisms are still unknown. We present a new rat model of SLC6A8 deficiency. Knock-in Sprague-Dawley rats were generated (CRISPR/Cas9; c.1166A>G; p.Tyr389Cys; human missense mutation abolishing SLC6A8 activity). Brain creatine was measured by 1H-MRS (horizontal 9.4T-MRI). Brain, kidney and liver were analyzed by immunohistochemistry and western blotting. Grooming and rearing were recorded in an arena (1m diameter) for 10 min and analyzed using TheObserver®XT11. KI males showed strong decreases in brain creatine (-80%), body weight gain and rearing unsupported, but significant more grooming time. Kidney AGAT was increased (suggesting up-regulation) while SLC6A8 was still expressed but intracellularly delocalized. KI males showed structural changes in cerebellum: disorganized astrocytic fibers, and thinner molecular layer of cerebellar cortex. Our results validate this model as promising tool to better understand and treat SLC6A8 deficiency.

Albumin-Adjusted Calcium: the Formula Matters!

Ursula Amstutz¹, Anca Liliana Cismaru^1,2, Livia Grimm¹, Christof Schild¹, Michael Nagler¹

¹University Institute of Clinical Chemistry, Inselspital, Bern University Hospital, University of Bern, Bern, Switzerland; ²Graduate School for Cellular and Biomedical Sciences, University of Bern, Bern, Switzerland

Plasma concentrations of total calcium and albumin are correlated. For calculation of albumin-adjusted calcium concentrations, Payne’s formula is widely used. However, this formula was developed based on albumin measurement using bromcresol green (BCG) while our current method uses bromcresol purple (BCP). Here, we assessed different formulas for albumin adjustment of calcium concentrations. Anonymized data from laboratory tests performed at the Inselspital Bern University Hospital were obtained. Payne’s formula and a formula developed using BCP-based albumin measurement (James formula) were compared by assessing the correlation of albumin-adjusted calcium with ionized calcium. Additionally, a formula based on the local patient population was derived using the correlation between total calcium and albumin. Albumin-adjusted calcium based on Payne’s formula showed poorer correlation (Spearman’s r=0.56) with ionized calcium compared to uncorrected total calcium, the James formula or the in-house formula (all Spearman’s r≥0.73). The area under the receiver operating characteristics curve for the identification of hypoalbuminemic patients (albumin <33 g/L) with hypocalcemia (ionized calcium <1.1 mmol/L) was highest for the locally derived and the James formula compared to unadjusted calcium or adjusted calcium using Payne’s formula, indicating that the latter is not suitable for calculation of albumin-adjustment based on measurements using BCP.

Anti-Inflammatory Effect of Omega-3 Fatty Acids on Activated CD4+T Cells

Paulina Wawrzyniak¹, Pakeerathan Srikanthan¹, Claudio Gemperle¹, Déborah Mathis¹, Martin Hersberger¹

¹Division of Clinical Chemistry and Biochemistry, University Children’s Hospital Zurich, Zurich, Switzerland

Triglyceride (TG)-based lipid emulsions form an integral part of life-saving total parenteral nutrition (TPN) and are provided to patients who are unable to orally ingest or digest/absorb the necessary daily amount of food. Currently available therapies often cause considerable side-effects, with evidence that lipid emulsions containing high amounts of omega-3 polyunsaturated fatty acids, have less side-effects. We compared the effect of a commercial fish oil lipid emulsion (Omegaven) to an omega-3 poor soybean oil lipid emulsion (Intralipid) on the activation of primary human T cells. For this, CD4+T cells were isolated by magnetic cell sorting from primary human PBMCs and were stimulated with anti-CD2, anti-CD3 and anti-CD28 antibodies for 48 hours in the presence of lipid emulsions. Intracellular production of cytokines was analyzed using flow cytometry following 6 hours stimulation with PMA/Ionomycin and brefeldin. Treating human primary CD4+T cells with Omegaven led to a dose-dependent increase in cellular omega-3 fatty acids and to a reduced cytokine production of TNF-α, IFN-γ and IL-4 when compared to CD4+T cells treated with Intralipid. Our results demonstrate that CD4+T cells treated with Omegaven incorporate the omega-3 polyunsaturated fatty acids, which reduces the production of pro-inflammatory cytokines and the activation of CD4+T cells.

Assessment of the Phenotypic Impact of DPYD Polymorphisms in Healthy Blood Donors

Seid Hamzic^1,2, Dominic Schärer^1,2, Alban Ramette³, Stefano Fontana⁴, Ursula Amstutz¹, Carlo R. Largiadèr¹

¹Institute of Clinical Chemistry, Inselspital, Bern University Hospital, University of Bern, Bern, Switzerland; ²Graduate School for Cellular and Biomedical Sciences, University of Bern, Bern, Switzerland; ³Institute forInfectious Diseases, University of Bern, Bern, Switzerland; ⁴Interregional Blood Transfusion SRC Bern and University of Lausanne, Bern, Switzerland

Dihydropyrimidine dehydrogenase (DPD) is the rate limiting catabolizing enzyme of the antimetabolite 5-Fluorouracil (5-FU). Currently, four DPD-activity impairing mutations in DPYD are used as 5-FU-toxicity markers. Several other variants, which are suspected to moderately affect enzyme activity, remain to be validated for their phenotypic effects. Here, we assessed the impact of c.85T>C, c.496A>G, and c.1601G>A in DPYD, together with the four clinically relevant DPYD-variants on Dihydrouracil/Uracil (UH2/U) ratios in plasma. Ratios were assessed by LC-MS/MS in N=628 healthy volunteers. Genotyping was performed with TaqMan or KASP-assays. Associations were tested with multivariate linear regression. All clinically relevant DPYD-variants and c.496G were individually associated with lower UH2/U-ratios (p<0.05), indicating reduced DPD-activity. Carriers of c.85CC had significantly higher UH2/U-ratios (p<0.05). C.85T>C and c.496A>G were in significant linkage (D’=0.69). When both variants were jointly analyzed, the effect increased 2.3 fold for c.85C and 1.6 fold for c.496G carriers. To investigate the effects of the cis- and trans-configurations in compound-heterozygous carriers (N=51), we performed long-read nanopore-sequencing of DPD-cDNA for phased-genotyping. Our results suggest that c.496G carriers have impaired DPD-activity. Furthermore, we observed increased DPD-activity in c.85CC carriers with a potential interaction with c.496A>G.

Associations between Classical HLA Genes and Metamizole-Induced Neutropenia

Livia Grimm¹, Anca Liliana Cismaru^1,2, Deborah Rudin^3,4, Manuel Haschke^5,6, Carlo R. Largiadèr¹, Ursula Amstutz¹

¹University Institute of Clinical Chemistry, Inselspital, Bern University Hospital, University of Bern, Bern, Switzerland; ²Graduate School for Cellular and Biomedical Sciences, University of Bern, Bern, Switzerland; ³Division of Clinical Pharmacology and Toxicology, University Hospital Basel, Basel, Switzerland; ⁴Department of Biomedicine, University of Basel, Basel, Switzerland; ⁵Clinical Pharmacology and Toxicology, Department of General Internal Medicine, Inselspital, Bern University Hospital, Bern, Switzerland; ⁶Institute of Pharmacology, University of Bern, Bern, Switzerland

Given the increase in prescriptions of the analgesic metamizole in Switzerland and other European countries, a better understanding of possible genetic susceptibility factors for metamizole-induced neutropenia or agranulocytosis is needed. Here, we investigated potential associations of classical HLA alleles with this adverse drug reaction. Six-digit resolution HLA typing results were obtained by targeted full-length resequencing of HLA-A, -B, -C, -DRB1, -DQA1, -DQB1, -DPA1, and -DPB1 using long-range PCR amplification and high-throughput sequencing in a discovery cohort of 32 cases of metamizole-induced neutropenia and 39 metamizole-tolerant controls. Nine alleles in six different HLA genes were identified as candidates with evidence for an association: HLA-A*01:01:01 [odds ratio (OR) 3.03, 95% confidence interval (CI) (1.00-10.41), p=0.045], -B*35:01:01 [OR=0.11, CI (0.00-0.81), p=0.013], -C*04:01:01 [OR=0.09, CI (0.00-0.63), p=0.006], -C*07:04:01 [OR->∞, CI (0.82-1), p=0.039], -DQA1*01:02:01 [OR=3.96, CI (1.35-13.27), p=0.005], -DQB1*05:01:01 [OR=0.28, CI (0.06-0.95), p=0.027], -DQB1*06:04:01 [OR=7.87, CI (0.92-370.35), p=0.046], -DRB1*04:01:01 [OR=6.15, CI (1.20-60.68), p=0.023], and -DRB1*13:02:01 [OR=9.33, CI (1.15-430.65), p=0.023]. Exhaustive multivariable logistic regression modelling suggests that, despite extensive linkage in this region, these associations represent at least two independent potential effects, which require confirmation in a larger patient cohort. The causal variation may, however, be located outside the coding region of classical HLA genes.

Genetic Case-Control Association Study of Metamizole-Induced Agranulocytosis

Anca Liliana Cismaru¹, Deborah Rudin², Luisa Ibañez³, Evangelina Liakoni⁴, Rheinhold Kreutz⁵, Alfonso Carvajal⁶, M.Isabel Lucena⁷, Javier Martin⁸, Esther Sancho Ponce⁹, Niclas Eriksson¹⁰, Carlo Largiadèr¹, Manuel Haschke^4,11, Pär Hallberg¹², Mia Wadelius¹², Ursula Amstutz¹

¹University Institute of Clinical Chemistry, Inselspital Bern University Hospital, University of Bern; Bern, Switzerland; ²Department of Clinical Pharmacology & Toxicology, University Hospital Basel, University of Basel; Basel, Switzerland; ³Hospital Universitari Vall d’Hebron, Universitat Autònoma de Barcelona, Fundació Institut Català de Farmacologia, Barcelona, Spain; ⁴Department of Clinical Pharmacology & Toxicology, Inselspital Bern University Hospital, University of Bern, Bern, Switzerland; ⁵Charité – Universitätsmedizin Berlin, Institut für Klinische Pharmakologie und Toxikologie, Berlin, Germany; ⁶Centro de Estudios sobre la Seguridad de los Medicamentos, Universidad de Valladolid, Valladolid, Spain; ⁷S Farmacologia Clinica, IBIMA, H Universitario Virgen de la Victoria, Universidad de Málaga, CIBERehd, Málaga, Spain; ⁸Instituto de Parasitologia y Biomedicina Lopez-Neyra, CSIC, Granada, Spain; ⁹Servei d’Hematologia i Banc de Sang, Hospital General de Catalunya, Sant Cugat del Vallès, Spain; ¹⁰Uppsala Clinical Research Center and Department of Medical Sciences, Uppsala University, Uppsala, Sweden; ¹¹Institute of Pharmacology, University of Bern, Switzerland; ¹²Department of Medical Sciences, Clinical Pharmacology and Science for Life Laboratory, Uppsala University, Uppsala, Sweden

Metamizole-induced agranulocytosis (MIA) is a rare and potentially fatal adverse drug reaction characterized by a loss of granulocytes. Previous studies have uncovered genetic associations for agranulocytosis caused by other drugs but none have investigated the genetic basis of MIA using. Here, we performed a case control association study to uncover potential susceptibility factors associated with MIA in a European population. Illumina Infinium BeadArray data of 254 individuals (53 patients, 39 tolerant controls and 162 unexposed controls) was assessed and tested for association with MIA using logistic regression. Among the tested SNPs of a candidate gene analysis focusing on genes previously associated with agranulocytosis caused by other drugs, we found two significant associations of rs10041917 and rs7726099 in the Serine incorporator 5 gene with MIA (p=2.3x10⁻⁴, OR=0.36, 95% CI=0.21-0.62; p=3.3x10⁻⁴, OR=0.37, 95% CI=0.21-0.64 respectively). On a genome-wide level, a candidate marker on chromosome 6 (rs191786) was identified but the association did not reach genome-wide significance. A follow-up imputation and meta-analysis including genome-wide genotype data from two European patient cohorts is ongoing and will further investigate the GWAS-implicated lead variants in all three cohorts as well as test over 38 million additional imputed SNPs for association. The identification of susceptibility markers for MIA can provide insights into the genetic background as well as the mechanisms underlying MIA and ultimately enable the identification of at-risk patients.

Group IIA Secretory Phospholipase A2 PredictsKidney Graft Failure and Mortality

Wijtske Wallimann¹, Jan Freark de Boer², Arne Dikkers², Markus van der Giet³, Stephan J.L. Bakker², Uwe J.F. Tietge²

¹Institute of Clinical Chemistry, University Hospital Zurich, Zurich, Switzerland; ²Department of Pediatrics, Center for Liver, Digestive, and Metabolic Diseases, University of Groningen, University Medical Center Groningen, Griningen, The Netherlands, ³Medizinische Klinik IV—Nephrology, Charite,Berlin, Germany

The acute phase protein group IIA secretory phospholipase A2 has potent proinflammatory and proatherosclerotic properties. The present study prospectively investigated whether plasma group IIA secretory phospholipase A2 associates with graft failure, cardiovascular and all-cause mortality in long-term renal transplant recipients (stable renal function at least one year after transplantation). In 511 renal transplant recipients (median follow-up 7.0 years) baseline group IIA secretory phospholipase A2 was significantly higher in renal transplant recipients than in healthy controls, but lower than in end-stage renal disease patients. Kaplan-Meier analysis demonstrated an increased risk for graft failure, cardiovascular and all-cause mortality with increasing gender-stratified quartiles of group IIA secretory phospholipase A2. Cox regression analyses revealed increased hazard ratios of graft failure, cardiovascular and all-cause mortality in patients with higher group IIA secretory phospholipase A2 levels. This association was largely explained by parameters of kidney function. Additional analyses demonstrated that renal function during follow-up declined faster in renal transplant recipients with higher baseline group IIA secretory phospholipase A2 levels. In summary, this prospective study demonstrates that group IIA secretory phospholipase A2 is a significant predictive biomarker for the occurrence of chronic graft failure, overall and cardiovascular disease mortality in renal transplant recipients dependent on kidney function.

Innovation in Biobanking: Reducing Errors through Automation

Tanja K. Froehlich¹, Martin G. Fiedler¹, Carlo R. Largiadèr¹

¹Liquid Biobank Bern, University Institute of Clinical Chemistry, Bern University Hospital, University of Bern: Inselspital, Bern, Switzerland

Despite advances in sample processing automation, generating sampling-kits for multi-centric studies is still done manually, as no system so far could generate complex kits. Although aliquoting robots exist, they require manual selection of protocols. Manual steps are time-consuming, and represent a considerable source of error, reducing the quality of biospecimen collections. Regarding the reproducibility of research results, a high level of accuracy during these steps of biobanking is key.

We present two robotic systems improving those limitations for the healthcare-integrated biobanking process of the Liquid Biobank Bern, Switzerland. For sample-kit generation, a newly designed SampliKit robot - in collaboration with Samplision – performs the picking of labware, label printing, labeling, and packing into sealed bags in one device. By replacing monotonous error-prone manual packing, errors in kit manufacturing are avoided and hands-on-time is highly reduced. For optimization of aliquoting, the processing of a Hamilton Star system was redesigned to enable a sample-by-sample processing, requesting for each loaded sample its own specific profile from central LIMS. Thus, we can process an infinite number of protocols on the same aliquoting system without a user having to select a protocol. Aliquoting errors– as they occur with manual selection - are avoided with this system.

LC-MS/MS Method for Quantification of Folate Metabolites in Human Plasma

Dominic Schärer^1,2, Jean-Christophe Prost1, Ursula Amstutz¹, Martin G. Fiedler¹, Cédric Bovet¹, Carlo R. Largiadèr¹

¹Institute of Clinical Chemistry Inselspital Bern University Hospital and University of Bern, Bern, Switzerland; ²Graduate School for Cellular and Biomedical Sciences University of Bern, Bern, Switzerland

The folate cycle involves a variety of inter-dependent enzymes, for many of which common polymorphisms in the encoding genes are known, but studies assessing their impact on single metabolite level are lacking. In the context of fluoropyrimidine (FP)-based chemotherapy, inter-individual variability in the folate cycle may strongly affect the response to FP-therapy, since the primary cytotoxic effect of FPs is enhanced by the folate-cycle derived co-factor 5,10-methylene tetrahydrofolate. In order to assess individual folate cycle phenotypes, we developed a LC-MS/MS assay for the quantification of five folate metabolites and one oxidation product of 5-methyl tetrahydrofolate in human plasma. An amount 275μL of plasma was prepared by protein precipitation followed by a solid-phase extraction with phenyl-sorbent cartridges. The extracted samples were injected on a Waters Acquity UPLC HSS T3 column using 0.1% formic acid in water and 0.1% formic acid in methanol as mobile phases. The analytes were separated with a run time of 10.8 min and detected with a triple quadrupole mass spectrometer (Xevo TQ-S, Waters) in the selected reaction monitoring mode (SRM) at positive polarity. This simultaneous quantification of folate metabolites enables us to determine individual folate metabolism phenotypes and to assess their correlation with genetic variability.

LDL versus CRP Association to Oxidised LDL

Noureddine Brakch¹, Selena Hamadet¹, Youssef Daaly²

¹Laboratoire ANAMED, Lausanne, Switzerland; ²Université de Genève Centre Médical Universitaire (CMU), Genève, Switzerland

Several studies have demonstrated an atherogenic role of oxidized low density lipoprotein (Ox-LDL). The present study aimed to assess the possible association of Ox-LDL to LDL, HDL, total cholesterol (CT) and. Triglycerides (TG).

We analyzed serum addressed to our laboratory for diagnosis (n=187 individual, 107 females and 80 males aged from 20 to 90) for LDL, CT, HDL, TG, Ox-LDL and C-reactive protein (CRP).

Data showed that females have significantly higher level of HDL than males 1.32±0.06 vs 1.16±0.04 p<0.01 and males have significantly higher TG than females 1.93±0.23 vs 1.6±0.2 p= 0.05. Moreover, LDL and CT were highly correlated (R2=0.918; p<0.01), HDL-CT (R2= 0.383; p= <0.01), HDL-LDL (R2= 0.32; p=<0.01). Surprisingly Ox-LDL showed no correlation to any of these lipids, otherwise it has showed a correlation to C-reactive protein CRP (R2=0.35, p<0.05).

When inflammatory state was combined to the levels of all lipids CT.CRP, LDL.CRP, HDL.CRP parameters showed a correlation R2= 0.36, p<0.05; R2=036 p<0.05 and R2=0.33, p<0.05 respectively.

This study indicated that inflammatory state expressed through CRP levels plays a role in Ox-LDL generation. Thus, association was slightly enhanced when combined to LDL, HDL and CT levels.

Performance of the RA Anylazer 10 for the Serological Screening of Lyme Borreliosis

C. Schilt¹, M.L. Tritten¹, A.Bizzini¹, R. Lienhard¹

¹ADMed Microbiologie, La Chaux-De-Fonds, Switzerland

Purpose:

Qualification of a new Assay on the RA Analyzer 10 instrument (Euroimmun).

Material and Method.

Qualification Panel included sera from blood donors and sera from blood donors declaring tick bite or treated Lyme borreliosis. Sera from Lyme patients were confirmed cases of erythema migrans (EM), neuroborreliosis, acrodermatitis, arthritis. Comparison was made with the Liaison XL (Diasorin).

Results and Discussion:

Specificity for RA versus XL was 85.2% (340/405) versus 87.2% (353/405) for IgG and 82.7% (40/405) versus 91.3% (380/405) for IgM. This low specificity was this is due to the Lyme borreliosis endemic situation in Switzerland. The concern is about the low specificity for IgM compared to XL.

Sensitivity for RA versus XL on EM is 58% (40-69) versus 76% (54/71) for IgG and 58.8% (41/70) versus 40.8% (29/71) for IgM. At early disseminated stadium the sensitivity for RA versus XL was 97.9% versus 90% in IgG and 65.3% (32/49) versus 34% (17/50) for IgM. All assays showed 100% sensitivity for late disseminated stadium in IgG.

Conclusion:

This new instrument is qualified for Lyme borreliosis serology. An improvement in IgM positivity limit is needed to avoid low specificity.

Rapid Centrifugation in the Routine Hemostasis Laboratory

Nathan Wolfensberger¹, Georgios Georgiou², Evelyne Giabbani³, Marianne Reusser³, Linet M. Njue³, Martin Fiedler⁴, Alexander B. Leichtle⁴, Michael Nagler⁴

¹Department of Medical Oncology and Hematology, University and University Hospital Zurich, Zurich, Switzerland; ²Unilabs Lausanne, Lausanne, Switzerland; ³Department of Hematology and Central HematologyLaboratory, Inselspital, Bern University Hospital, Bern, Switzerland; ⁴University Institute of ClinicalChemistry, Inselspital, Bern University Hospital, University of Bern, Bern, Switzerland

It is unclear whether short and rapid centrifugation schemes are applicable to the hemostasis laboratory. We assessed the accuracy of measurements obtained with a rapid, high-speed centrifugation scheme in a large set of hemostasis tests, covering the full spectrum of values obtained in clinical practice, and using meaningful statistical measures. Two citrated plasma samples were obtained from consecutive patients of a tertiary hospital with suspected abnormal hemostasis tests and processed with two centrifugation schemes in parallel: 1500 g for 10 minutes and 3137 g for 7 minutes. The following tests were conducted: prothrombin time (n=125), INR (n=146), activated partial thromboplastin time (n=119), thrombin time (n=105), fibrinogen (n=125), factor (f)II (n=69), fV (n=64), fVII (n=64), fX (n=67), fVIII (n=55), fIX (n=37), fXI (n=35), and fXIII (n=20), D-dimer (n=34), antithrombin (n=31), anti-Xa activity (n=30), von Willebrand antigen (n=25), and von Willebrand activity (VWF:GPIbM; n=27). Spearman’s rank correlation coefficient was at least 0.95 for all tests except fV, fIX, and fXI. The coverage probability at a given deviation index of 15% was above 0.9 for all tests except fV, fVII, fX, fVIII, fIX, fXI, and VWF:GPIbM. Our results suggest that high-speed centrifugation is applicable to the majority of routine hemostasis parameters.

Resolvin D1 Reduces Adipose Tissue Inflammation by Triggering Human Macrophages

Claudio Gemperle¹, Syndi Tran¹, Jacqueline Marti-Jaun¹, Paulina Wawrzyniak¹, Martin Hersberger^1,2

¹Division of Clinical Chemistry and Biochemistry, Children’s Research Center, University Children’s Hospital Zurich, Zurich, Switzerland; ²Zurich Center for Integrative Human Physiology, Zurich, Switzerland

Obesity leads to chronic inflammation of the adipose tissue which is tightly associated with the metabolic syndrome, type 2 diabetes and cardiovascular disease. Inflammation of the adipose tissue is mainly characterized by the presence of crown-like structures composed of inflammatory macrophages in the neighborhood of adipocytes. Resolvin D1 (RvD1), a potent anti-inflammatory and pro-resolving lipid mediator derived from the omega-3 fatty acid docosahexaenoic acid, has been shown to reduce the inflammatory tone of adipose tissue in animal models but the underlying mechanism is not clear. We investigated the effect of RvD1 on the inflammatory state of a human co-culture system of adipocytes and macrophages. For this, human mesenchymal stem cells were differentiated into mature adipocytes and overlayed with human primary macrophages. In this co-culture, 10-500 nM RvD1 reduced the secretion of the pro-inflammatory cytokine IL-6, of the chemokine MCP-1, and of the adipokine leptin. Similarly, we observed a reduction in secretion of IL-6 and TNF-α when macrophages were treated alone with RvD1, while no change of cytokine secretion was observed when adipocytes were treated with RvD1. We conclude that RvD1 polarizes macrophages to an anti-inflammatory phenotype, which in turn modulates the inflammatory tone of adipocytes.

Spectrum of Acute Drug Toxicity During the Street Parade

Daniel Müller¹, Ksenija Slankamenac², Aline Herzog², Hugo Kupferschmidt³, Arnold von Eckardstein¹, Dagmar I. Keller²

¹Institute of Clinical Chemistry, University Hospital Zurich, Zurich, Switzerland; ²Emergency Department, University Hospital Zurich, Zurich, Switzerland; ³Tox Info Suisse, National Poisons Information Centre, Associated Institute of the University of Zurich, Zurich, Switzerland

Since 1991, the Street Parade, world’s most popular house and techno parade in Zurich, is a mecca for ravers. Extensive drug abuse has been commonly seen. The prevalence of acute drug toxicity during the Street Parade is unknown. Our aim was to investigate the drug spectrum of acute intoxicated patients from the Street Parade presenting in the Emergency Department.

Samples were analyzed by liquid chromatography coupled to high-resolution mass spectrometry. We investigated urine samples of acute intoxicated patients who participated at the Street Parade and presented in the Emergency Department in 2017 and 2018. The endpoints were the analysis of the drug spectrum and assessment of the prevalence of acute drug toxicity.

In total, we analyzed 47 urine samples. The three leading drugs were cocaine (21.3%), 3,4-methylenedioxymethamphetamine (MDMA) (19.1%) and tetrahydrocannabinol (THC) (17.0%) followed by methamphetamine (8.5%), methylphenidate (6.4%) and 2.1% for each lysergic acid diethylamide and amphetamine. Furthermore, one patient (2.1%) showed an abuse of a new psychoactive substance (methylon) in combination with alcohol, cocaine and MDMA. Methamphetamine intoxications seem to increase from 2017 to 2018. Thus, future preventive strategies need to sensitize the rave scene about the drug spectrum and possible health consequences.

Structure–function Relationships of HDL in Diabetes and Coronary Heart Disease

Mustafa Yalcinkaya¹, Mathias Cardner², Sandra Goetze³, Edlira Luca⁴, Miroslav Baláz³, Monika Hunjadi⁵, Johannes Hartung⁶, Nicolle Kränkel⁶, Andrej Shemet⁷, Silvija Radosavljevic¹, Michaela Keel¹, Alaa Othman¹, Thorsten Hornemann¹, Manfred Claassen⁹, Gerhard Liebisch¹⁰, Erick Carreira⁷, Andreas Ritsch⁵, Ulf Landmesser⁵, Jan Krützfeldt⁴, Christian Wolfrum³, Bernd Wollscheid³, Niko Beerenwinkel², Lucia Rohrer¹, Arnold von Eckardstein¹

¹Institute of Clinical Chemistry, University and University Hospital of Zurich, Zurich, Switzerland; ²Department of Biosystems Science and Engineering, ETH Zurich, Zurich, Switzerland; ³Department of Health Sciences and Technology, ETH Zurich, Zurich, Switzerland; ⁴Department of Diabetology and Endocrinology, University and University Hospital of Zurich, Zurich, Switzerland; ⁵Department of Internal Medicine, University Medicine Innsbruck, Austria, ⁶Department of Cardiology, University Medicine Charité Berlin, Germany, ⁷Department of Chemistry and Applied Biosciences, ETH Zurich, Zurich, Switzerland; ⁸Institute of Clinical Chemistry, University and University Hospital of Zurich, Zurich, Switzerland; ⁹Institute of Molecular Systems Biology, ETH Zurich, Zurich, Switzerland; ¹⁰Institute of Clinical Chemistry, University of Regensburg, Germany

HDL contain hundreds of lipid species and proteins and exert many vasoprotective and anti-diabetogenic activities on cells. Structure-function-disease relationships are little understood. HDL subclasses in plasma from 51 healthy subjects and 98 patients with T2DM, CHD, or both were analysed by NMR spectroscopy. HDL was characterized for protein and lipid composition as well as the following functions: induction of cholesterol efflux, inhibition of apoptosis of HAECs and beta cells, regulation of mitochondrial potential and respiration in myotubes and brown adipocytes, respectively. Structure-function-disease relationships were inferred by high-dimensional probabilistic graphical models. Predicted structure-function-relationships were validated by the use of rHDL reconstituted with or without the candidate protein or lipid. The bioinformatic integration of 29 clinical characteristics, 33 NMR features, 182 proteins, 227 lipid species, and twelve functional readouts gave the following main findings: First, CHD and T2DM are associated with different changes of HDL in size distribution, protein and lipid composition as well as function. Second, the different functions of HDL are weakly correlated with each other and determined by different structural components. Third, three novel determinants of HDL function were identified, namely the SM 42:3 and GPLD1 for the ability of HDL to inhibit starvation induced apoptosis of HAECs and Apo-F for the ability of HDL to promote maximal respiration.

Swiss BioRef: Personalized Reference Values for Precision Medicine

Alexander Leichtle¹

¹Inselspital for the Swiss BioRef Consortium, Bern, Switzerland

Reference values for laboratory tests are crucial to define pathological or disease states as well as treatment thresholds. Since they depend on many factors, including patient-related variables (e.g. sex, genetic background) and sample-related variables (e.g. mode of collection, storage), definitions of “medical norms” may vary across disciplines and contexts. With the rise of personalized medicine, the concept of “classical” reference ranges has become outdated, since comparing a patient to a more or less adequate “healthy” control population cannot take into account intra-individual thresholds of what for a specific patient might be a “normal” or “healthy” state. Especially in a population increasingly suffering from a wide range of comorbidities, normal testing results might be an exception and not even be desirable. To properly assess a patient’s measurements, it is important to take the whole spectrum of potential sources of bias, comorbidities and confounders into account and to generate personalized “target values” we would expect from a healthy “digital twin” of the specific patient. With the concept of the Swiss BioRef SPHN infrastructure development project, we address these challenges by building a nation-wide computational resource for personalized reference values.

Targeted Analysis of 44 Lipid Mediators in Plasma

Ivan Hartling¹, Alessio Cremonesi¹, Martin Hersberger^1,2

¹Division of Clinical Chemistry and Biochemistry, Children’s Research Center, University Children’s Hospital Zurich, Zurich, Switzerland; ²Zurich Center for Integrative Human Physiology, Zurich, Switzerland

Lipid mediators are bioactive lipids produced from omega-3 and omega-6 polyunsaturated fatty acids (PUFAs) and are vital mediators of inflammation. Metabolites of the omega-6 PUFA, arachidonic acid, are essential for initiating inflammation and recruiting neutrophils, whereas metabolites of the omega-3 PUFAs eisosapentaenoic acid and docosahexaenoic acid are responsible for resolving inflammation and returning tissue to homeostasis. Quantifying lipid mediators in biological samples including plasma, tissue, and cell culture remains a significant hurdle to further understanding the relationship between lipid mediator concentrations, inflammation, disease, and treatments. To address this issue, we developed an UHPLC-MS/MS method to quantify 44 lipid mediators representing pro-inflammatory prostaglandins and leukotrienes; pro-resolving lipoxins, resolvins, protectins, and maresins; as well as epoxy eicosatrienoic acids and endocannabinoids. Quantification was performed using 13 deuterated internal standards for external calibration. Mass spectrometry parameters were optimized for each lipid and all lipids were either fully separated by liquid chromatography or had a unique mass transition detectable by mass spectrometry. Extraction recovery ranged from 55-96% and accuracy was 68-124% with the exception of 9,10-EpOME (44%). The developed method will allow us to better understand the role of lipid mediators in inflammation and disease as well as to monitor their response to treatment.

Towards LC-MS/MS Steroid Panel Measurements

Valentin Braun^1,2, Christian Timm¹, Martin Risch^1,3, Christoph Seger^1,2

¹lmz Dr Risch Ostschweiz AG, Buchs SG, Switzerland; ²Leopold-Franzens-Universität, Innsbruck, Austria, ³Kantonsspital, Chur GR, Switzerland

A main obstacle in modern day endocrinology is that metabolite testing must rely on immunological methods based on different platform chemistries. Parallel analyte assessment is impossible and assay selectivity is often limited. Consequently, overcoming these hurdles is a major task in contemporary laboratory medicine. An LC-ESI-MS/MS assay is presented allowing the parallel quantification of testosterone, androstendione, 17-OH-progesterone, progesterone, cortisol, cortisone, 11- and 21-deoxycortisol, DHEA, and DHEAS. The 6.5 min gradient elution assay relies on biphenyl stationary phase chemistry and the mobile phase carries ammonium fluoride as modifier. A protein crush with methanolic zinc sulfate solution carrying the stable isotope labelled internal standards allowed swift sample preparation. Inter-day IQC CVs better than 4-8% were found for all analytes. Quantification limits were optimized such that value assignment in the reference range is possible as patient sample (n=14) measurements did prove. EQC data did prove the trueness of the assay results. Pending further robustness experiments the assay is deemed fit for routine application.

Validation of a Method for Volatile Amines in Urine and Plasma by HS-GC-MS

A. Fuchs¹, Peter J. Neyer², Luca Bernasconi², Martina D. Allenspach¹, Christian Steuer¹

¹ETHZ, Zurich, Switzerland; ²KSA, Aarau, Switzerland

We present a fast analytical procedure for four short chain volatile amines (DMA, TMA, DEA, TEA) in different human biosamples. The method uses head-space gas chromatography coupled to mass spectrometry for the simultaneous quantitation. Separation was performed on a volatile amine column within 7.5 min followed by detection on a Thermo DSQ II mass spectrometer (single quadrupole) scheduled in single ion monitoring mode. Four hundred microliters of Li-heparin plasma and urine were analyzed. Thermal liberation of volatile amines under in an alkaline and saline aqueous environment was automated in a gradient program with the autosampler. It was validated for accuracy, precision, recovery and stability. Accuracy and precision criteria were below 12% for all target analytes at two different concentration levels. The selected extraction procedure provided recoveries>92% from both matrices for TMA, DEA and TEA. DMA could be recovered from Li-heparin plasma only>75%. Futhermore, the method was successfully applied to plasma and urine samples from eleven healthy volunteers. Detected concentrations of the endogenous metabolites DMA and TMA are in accordance with the known reference ranges from the literature.

Validation of Ultrafiltration Techniques for TDM

Sophia Rehm¹, Katharina Rentsch¹

¹Laboratory Medicine, University Hospital Basel, Basel, Switzerland

Ultrafiltration is a widely used technique to measure free fractions of drugs, especially in the clinical laboratory, when other techniques, such as membrane filtration, may be too costly and elaborate. Conditions during filtration can have a huge impact on the measured free fraction, resulting in clinically relevant variations of the results. Some factors influencing protein binding of drugs and therefore free fractions can be influenced by laboratory conditions. These are temperature, pH, and the used filtration device and filtration settings. We compared the free fractions of the antiepileptic drugs valproic acid and phenytoin as well as antibiotics from different classes like beta-lactams, cephalosporins and carbapenems under different conditions and using different filter devices. The results show a big impact of temperature on some of the drugs, while others are only marginally affected. Some filtration devices strongly absorb drugs, leading to seemingly zero percent free fraction. Only one of the four tested devices gave satisfactory results. Filtration devices are mostly produced for DNA concentration and not protein filtration. Therefore, validation of the ultrafiltration process is the key in obtaining reliable free fractions.

Volatile Compounds (VOC) in Pathogens Related to Cystic Fibrosis by SESI-HRMS

N. Perkins¹, T. Bruderer^2,3, S. Müller³, S. Micic², R. Weber², R. Zenobi³, C. Berger⁴, A. Möller⁵

¹Division of Clinical Chemistry and Biochemistry, University Children’s Hospital Zurich, Zurich, Switzerland; ²Division of Respiratory Medicine, University Children’s Hospital Zurich, Zurich, Switzerland; ³ETH Zurich, Zurich, Switzerland; Department of Chemistry and Applied Bioscience, Zurich, Switzerland; ⁴Division of Infectious Diseases, University Children’s Hospital Zurich, Zurich, Switzerland; ⁵Division of Respiratory Medicine and Childhood Research Center, University Children’s Hospital Zurich, Zurich, Switzerland

Cystic fibrosis (CF) is the most common metabolic disorder in Western Europe. This autosomal recessive disease results in severely reduced mucus clearance in the lungs.

VOC’s in breath can reflect pathogen metabolism and could offer a non-invasive procedure for early detection of pulmonary infections.

We propose to differentiate between 6 pathogens detected in the airways of CF patients using headspace coupled secondary electrospray ionisation high-resolution mass spectrometry and to identify pathogen-specific markers for later comparison with the breath fingerprint of CF patients.

Cultures of the 6 pathogens were measured using an in-house headspace set-up (n=30). Principal component analysis was successful in discriminating between the 6 pathogens. In a supervised approach, linear discriminant analysis showed an average accuracy of 99.9%. Utilizing an extensive list of published pathogen biomarkers in conjunction with a correlation matrix to assign common adducts, we were able to put together a list of markers for each pathogen strain for further compound identification.

Employing a robust methodology, we were able to successfully differentiate between all 6 pathogens. Comparing these pathogen-specific markers with the breath fingerprint of CF patients will allow us to evaluate the applicability of secondary electrospray ionisation high-resolution mass spectrometry for clinical use in early CF diagnosis.

Xenobiotics Affect Human Semen Quality and Endocannabinoid Levels

Fanny Zufferey¹, Widmer Sabina¹, Nicolas Donzé¹, Rita Rhaban², Alfred Senn², Eric Stettler², Serge Rudaz³, Serge Nef², Michael F. Rossier^1,4

¹Service of Clinical Chemistry & Toxicology, ICH, Hôpital du Valais, Switzerland; ²Department of Genetic Medicine & Development, University of Geneva, Geneva, Switzerland; ³School of Pharmaceutical Sciences, Universities of Geneva and Lausanne, Geneva, Switzerland; ⁴Department of Internal Medicine of Specialties, University of Geneva, Geneva, Switzerland

While the methodology for analyzing sperm parameter is well established, the biochemical composition of the semen remains poorly investigated.

The goal of this work is to determine whether xenobiotics like paracetamol, THC or bisphenol could affect the sperm quality or the endocannabinoid levels to identify samples affected by potential disrupting agents in the evaluation of new putative biomarkers of human semen quality.

We analyzed samples of 500 young men randomly selected from a large cohort of Swiss conscripts. Sperm parameters were determined according to WHO criteria. Endocannabinoids were measured in seminal fluid and blood serum and bisphenol S in urine by LC-MS/MS. An immunological screening was performed to determine the presence, among others, of paracetamol and metabolites of THC in urine. Samples positive for THC metabolites were then quantified by GC-MS.

We found that individuals positive for THC have significantly higher seminal pH, those positive for paracetamol have higher sperm count and those positive for the endocrine disruptor bisphenol S have lower sperm count. Moreover, THC positivity was linked to lower seminal anandamide concentrations. These results must be taken into account in the evaluation of new biomarkers of human sperm quality.