MTMarc Thelen

Marc Thelen

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Marc Thelen has a total of 358 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Garbage in, (mis)diagnosis out: making laboratory measurement bias visible and actionable across clinical algorithm outputs

    by Niels van Noort, Marith van Schrojenstein Lantman, Mathie P.G. Leers, Remy J.H. Martens, Volkher Scharnhorst, Marc H.M. Thelen, Arjen-Kars Boer

    Objectives: Clinical algorithms increasingly combine multiple laboratory measurands to support diagnosis, prognosis, and treatment decisions. Yet, measurement bias arising from differences in calibration, reagent lots, or analytical platforms can propagate through such algorithms in non-linear ways, shifting outputs and altering clinical classifications. Existing external quality assessment (EQA) programmes evaluate measurands individually and do not capture the cumulative effect of bias on algorithm behaviour. We present a framework that extends EQA to quantify and communicate the impact of laboratory-specific measurement bias on clinical algorithm outputs. Methods: Bias estimates derived from EQA results are injected into a representative patient cohort and propagated through the algorithm; when clinical outcomes are available, output shifts are translated into changes in clinically interpretable performance characteristics. Measurand-specific what-if analyses estimate the expected benefit of correcting individual bias. The framework is illustrated using the Fibrosis-4 (FIB-4) index and scalability is demonstrated on the ten-input CoLab COVID-19 severity score. Results: Participant-facing reports were developed that summarise deviation from a zero-bias reference, position each laboratory within the peer distribution, and identify which input measurands offer the greatest potential for performance recovery. Applied to FIB-4, the reports demonstrate how laboratories can locate themselves relative to clinically desired performance thresholds and receive actionable, measurand-specific guidance to prioritise corrective efforts. Conclusions: By translating bias-propagation analysis into interpretable laboratory reports, this framework bridges the gap between measurand-level analytical performance and clinical algorithm performance. It supports post-deployment monitoring, cross-laboratory comparability and verification, targeted quality improvement, and is applicable to any transparent formula-based clinical algorithm.

  • Open Access

    Evaluating the predictive value of multiple rounds of faecal haemoglobin measurement for the prediction of advanced neoplasia and colorectal cancer: an analysis of data from the Dutch national colorectal cancer screening programme

    by Lucy Wairimu Mwangi, Chiara Brück, Alexandra Katsara, Esther Toes-Zoutendijk, Anneke van Vuuren, Marc Thelen, Iris Nagetaal, Evelien Dekker, Manon Spaander, Monique van Leerdam, et al.

    Objective: Faecal haemoglobin (f-Hb) concentrations from two screening rounds improve colorectal cancer (CRC) risk prediction, but the value of longer longitudinal data remains uncertain. This study aimed to assess whether incorporating f-Hb data from three prior screening rounds improves risk stratification for CRC and advanced neoplasia (AN). Methods: Using data from four rounds of the Dutch national CRC screening programme (1 January 2014 and 31 December 2022), prognostic models combining age, sex, and f-Hb concentrations were developed to predict CRC and AN. Multiple representations of longitudinal f-Hb were evaluated. Model performance was assessed using the area under the receiver operating characteristic curve (AUC), calibration, and information criteria. Risk reclassification was quantified using net reclassification improvement (NRI) and integrated discrimination improvement (IDI). The policy impact of risk-based referral, expressed as additional colonoscopies, was examined. External validation used data from a Dutch faecal immunochemical test pilot study. Results: Among 1 171 070 participants included, 2357 (0.2%) had CRC and 14 543 (1.2%) had AN. Models incorporating categorised f-Hb values from three prior rounds, age, and sex achieved the highest discrimination (AUC: AN 0.76 (0.75–0.77); CRC 0.73 (0.72–0.74)) and significantly improved reclassification (NRI and IDI, p<0.001). Participants in the highest 15% predicted risk had a 6.8-fold (95% CI 6.4 to 6.9) increased risk of AN and a 5.9-fold (95% CI 5.4 to 6.6) increased risk of CRC. Risk-based referral enabled earlier detection with minimal additional colonoscopy for both CRC and AN. Conclusion: Including f-Hb data from three prior screening rounds improves CRC and AN risk prediction. Targeted referral of the highest risk individuals enables earlier detection with limited impact on colonoscopy capacity, supporting personalised CRC screening strategies.

  • Open Access

    The Dutch colon cancer screening programme as an example of adopting indirect-outcome-based analytical performance specifications using a virtual analyser concept

    by Marith Van Schrojenstein Lantman, Anja De Jong, Petra Van Der Groep, Evelien A. Kemper, Monique De Groot, Renée Vossen, Karin Mohrmann, Joop Ten Kate, Marieke A.M. Frasa, Marc H.M. Thelen

    To ensure reported laboratory results are of adequate quality, it is paramount that the set analytical performance specifications (APS) are tailored to the intended uses of the measurand. For measurands with a pivotal position in the clinical pathway, APS should be set based on clinical outcome. This paper, written as a result of an outcome-based-APS-summit in Prague, gives practical guidance on how to set outcome-based APS. Two approaches are mentioned with distinct starting points of the required performance (how much clinical misclassification do we allow, and which APS fulfils that premise) or the current performance (what is the effect of the current analytical performance characteristics (APC) on how much misclassification occurs). Importantly, APS should not just be set, but also requires monitoring, which method depends on the utilization of the test in clinical practice. The laboratory can monitor APS for so-called mother-daughter systems, individual analysers or a virtual-analyser system. This paper discusses the pros and cons of these approaches. To highlight the scope of clinically consequential outcomes for which APS can be set and monitored, the Faecal Immunochemical Testing for human haemoglobin (FIT), used for the Dutch Colorectal Cancer Screening programme (CRC-Screening), is used as an example. Here, APS were based on a clinically consequential intermediate outcome, i.e. analytically induced variability in colonoscopy referral rate (±0.5 %), and implemented as a virtual analyser system, dividing the APS budget of 10 % equally across four sources of FIT variation: reagent/calibrator lot variation, sampling buffer lot variation, within-analyser variation, and between-analyser variation.


  • Open Access

    Evaluating feasibility of ferritin assay harmonisation with commutable proof-of-concept reference materials and EQA-based monitoring

    by Moniek M. Bioch, Rogier M. Schoeman, Cas W. Weykamp, Hans van der Vuurst, Aldy W.H.M. Kuypers, Fokke A. Lindeboom, Marieke A.M. Frasa, Antonius E. van Herwaarden, Dorine W. Swinkels, Marc H.M. Thelen

    Objectives: Clinical interpretation of ferritin is hindered by between-assay differences, persisting despite the introduction of different generations of WHO standard. To aid harmonisation efforts, this study investigates the commutability of External Quality Assessment (EQA) materials, proof-of-concept reference materials (cRMs), and the newest generation of WHO standard (19/118). Methods: Four multi-level serum-based cRMs, 50 clinical samples, four EQA materials, and WHO 19/118 were measured with Abbott Alinity, Beckman Coulter Access DxI, Roche Cobas and Siemens Atellica. Commutability was analysed using the prediction interval approach for EQA materials and the difference in bias and calibration effectiveness approaches for cRMs and WHO standard. Results: Differences in assay selectivity did not allow for formal commutability assessment of EQA materials. Nevertheless, all four investigated EQA materials fell within the 95 % prediction interval around clinical samples in all six method comparisons, indicating their suitability for monitoring harmonisation status. Calibration effectiveness demonstrated commutability for cRM1 and cRM3, in contrast to poor commutability for cRM2, cRM4 and WHO 19/118. Recalibration with cRM1 decreased inter-assay CV four-fold from 34 % to 8.2 %. Conclusions: Improving harmonisation among current ferritin assays is possible, and success of such efforts can be monitored with EQA materials. Although limited by differences in selectivity, the harmonisation potential of the current assays is substantial, and therefore we encourage harmonisation based on a ISO15194-proof material produced like our cRM1 or cRM3. True standardisation will need much more time, requiring prior measurand characterisation, identification of a clinically relevant ferritin measurand, and subsequent assay redesign to eliminate differences in selectivity.

  • Open Access

    Validation of synthetic M-protein materials for external quality assessment: homogeneity, commutability, and stability

    by Moniek M. Bioch, Theo Rispens, Corrie M. de Kat Angelino, Anke J. Ten Haaken-Meijer, Miranda van Berkel, Marc H.M. Thelen, Joannes F.M. Jacobs

    Objectives: Accurate quantification of M-proteins is important for clinical management of monoclonal gammopathies. Reliable assessment of trueness on M-protein quantification in External Quality Assessment (EQA) schemes is difficult due to the lack of reference methods and materials for target value assignment. We propose a strategy to produce targeted synthetic M-protein EQA materials by spiking known amounts of human(ised) therapeutic monoclonal antibodies (t-mAb) into null serum. This paper validates homogeneity, commutability, and stability of these materials. Methods: Homogeneity was assessed on two aliquoted synthetic M-protein EQA materials comparing within-vial variation with between-vial variation for sodium, total protein, IgG and M-protein quantification. Commutability was assessed by comparing between-method differences in 51 patient-derived and nine synthetic M-protein EQA materials among various M-protein quantification methods. Stability was checked by reanalysis of two previously distributed synthetic M-protein EQA materials. Results: Within- and between-vial variation were similar in both samples, and therefore homogeneity criteria were met. Commutability was observed for eight out of nine samples, whereas one hypergammaglobulinemic sample was non-commutable. Long-term storage stability at −80 °C was found acceptable in both samples. Conclusions: Due to their proven homogeneity, commutability, and stability, targeted synthetic M-protein EQA materials are suitable for use in M-protein EQA schemes. This allows for the assessment of laboratory performances based on target values, and opens the way for harmonisation and exchangeability of M-protein diagnostic methods. Additionally, it provides a new source for the production of M-protein EQA materials, alleviating the demand for large volumes of patient material needed to organise M-protein EQA schemes.


  • A roadmap for standardization of serum albumin measurements

    by Mauro Panteghini, Guy Auclair, Marc Thelen, Liqing Wu, Robert Wielgosz

    Albumin measurements in serum have important diagnostic and prognostic roles. However, non-standardized measurement procedures are still used in medical laboratories. An agreed approach for the implementation of metrological traceability to the International System of Units (SI) of serum albumin measurements is lacking and the use of non-selective assays in medical laboratories is still predominant. This paper provides an analysis of the current status of serum albumin measurements, including discussion on how to improve standardization and highlighting some important aspects that still require adequate solutions, such as the abandonment by medical laboratories of non-selective bromocresol green methods based on risk of clinical harm. The discussion also includes considerations about the analytical performance specifications that should be fulfilled when using serum albumin measurements for patient management decisions.


  • Open Access

    Model 1b analytical performance specifications – report on the Prague 2025 workshop

    by Andrea R. Horvath, Sverre Sandberg, Marc H.M. Thelen, Mauro Panteghini, Hikmet Can Çubukçu, Emmi Rotgers, Eric Kilpatrick, Katy J.L. Bell, Graham R.D. Jones, Elvar Theodorsson

    The main purpose of establishing APS is to ensure that test performance aligns with their specific clinical goals. Analytical performance specifications serve multiple functions: as acceptance criteria to verify that in vitro diagnostic medical devices (IVD MD) meet regulatory and quality standards before market release; for external quality assurance assessments; for quality control, monitoring reagent lot-to-lot variability, and allowing preanalytical variation assessments in labs; and for guiding purchasing decisions by healthcare providers, reimbursement agencies, and both public and private insurers. Over the years, various systems and projects have been established to evaluate the current state of the art (Model 3) of IVD MD through external quality assurance (EQA). Different national and international EQA providers often use distinct acceptance criteria tailored to their organisational goals and local healthcare settings. Model 2, which relies on biological variation data, has been effectively supported by the European Federation of Clinical Chemistry and Laboratory Medicine’s database and its interactive tools for calculating APS ( www.biologicalvariation.eu ). Nonetheless, debates continue about optimal APS setting methods and the most suitable models for different measurands and scenarios. To explore these issues further, workshops were held in Prague in 2023 and 2025. This report presents the expert presentations from the 2025 workshop, which exclusively focused on APS based on Model 1b.

  • The severity of harm score database: an evidence-based foundation for laboratory test risk analysis

    by Hikmet Can Çubukçu, Oswald Sonntag, Nathan Lorde, Charles R. Lefèvre, Chiara Cosma, Annelien Van Dalem, Panagiotis N. Kanellopoulos, Marija Kocijancic, Guilaine Boursier, Pika Meško Brguljan, et al.

    Background: Patient risk management has become central to clinical laboratory quality standards, including the Clinical and Laboratory Standards Institute (CLSI) EP23 2nd edition (CLSI EP23-ED2), International Organization for Standardization (ISO) 15189:2022, ISO 22367:2026, and ISO 14971:2019. A key component of risk analysis is estimating the severity of harm (SoH) that may result from erroneous laboratory results. We present here the Severity of Harm Evidence Database, a structured, multi-study resource that integrates survey-based SoH assessments from the published independent studies. The database provides ordinal distributions and pooled SoH for 196 unique laboratory tests. Methods: A canonical five-level scale (negligible, minor, serious, critical, catastrophic) was adopted, aligned with ISO 14971:2019. For the 19 tests common to both studies, direct count pooling was performed by summing category-level respondent counts. Wilson score confidence intervals were calculated for all proportions. The database is organized into nine relational sheets and is accompanied by an interactive web application. Results: Out of the 196 tests covered, 19 have pooled estimates from both studies (total n=781 respondents). Pooled analysis confirms that cardiac biomarkers, blood gases, coagulation parameters, electrolytes (potassium, sodium), and therapeutic drug concentrations (digoxin, lithium) consistently receive the highest severity ratings, while certain endocrine markers receive lower scores. Conclusions: The SoH evidence database offers the first publicly available, structured repository of SoH assessments for clinical laboratory tests. It supports compliance with risk management requirements, informs risk-based strategies per CLSI EP23-ED2, and provides an evidence-based foundation for laboratory risk analysis.