Reply to “Is this quantitative test fit-for-purpose?”

Sarah J. Lord; Andrea Rita Horvath; Phillip J. Monaghan; Sverre Sandberg; Patrick M. Bossuyt; on behalf of the EFLM Committee, Clinical and Analytical Performance Specifications (C-CAPS)

doi:10.1515/cclm-2025-0291

Article Open Access

Reply to “Is this quantitative test fit-for-purpose?”

Sarah J. Lord , Andrea Rita Horvath , Phillip J. Monaghan , Sverre Sandberg , Patrick M. Bossuyt and on behalf of the EFLM Committee, Clinical and Analytical Performance Specifications (C-CAPS)

Published/Copyright: March 24, 2025

Published by

Become an author with De Gruyter Brill

Submit Manuscript Author Information Explore this Subject

From the journal Clinical Chemistry and Laboratory Medicine (CCLM) Volume 63 Issue 7

Keywords: cut-off limit; likelihood ratio; pretest probability; posttest probability; sensitivity; specificity

To the Editor,

We thank Asberg and Bolann [1] for their interest in our paper “Is this test fit-for-purpose? Principles and a checklist for evaluating the clinical performance of a test in the new era of in vitro diagnostic (IVD) regulation” [2]. The In Vitro Diagnostic Medical Devices Regulation (IVDR) 2017/746 defines clinical performance as “the ability of a device to yield results that are correlated with a particular clinical condition or a physiological or pathological process or state in accordance with the target population and intended user” [3]. The purpose of our paper is to present the principles for evaluating this ability to detect or predict a clinically defined target condition. We explain this entails enrolling a study group that represents the target population and cross-referencing the test results with the presence of the target condition. The distribution of test results in those with and without the target condition expresses its clinical performance. The receiver operating characteristic (ROC) curve and the area under the curve (AUC) are transformations of these distributions, displaying the ability of the test to correctly classify or identify members of the target population, as do sensitivity and specificity for a specific threshold. We refer to these measures as the primary measures of clinical performance and focus our paper on explaining how the same principles apply to tests intended for other purposes, such as prognosis, where the target condition is determined in the future, and key elements for study design for valid estimation. We also describe how clinical performance can be evaluated for any clinically defined criteria, groupings, or risk category relevant to the intended purpose of the test.

In contrast, as Asberg and Bolann explain in their letter [1], likelihood ratios (LR) allow clinicians to estimate the posttest probability of the target condition for an individual with a given test result by applying the LR for that result to an estimate of the pretest probability and thereby aid the interpretation of diagnostic and screening tests results in clinical practice. While we do not wish to detract from Asberg and Bolann’s call for reporting functions or tables of LR when reporting clinical performance data, including in package inserts, we do not regard these ratios as primary measures of clinical performance. A likelihood ratio can be calculated for a single result, while our suggested measures of performance apply to the whole target population for whom testing is considered.

We hope our paper [2] will help foster a shared understanding of how clinical performance is defined in this new era of IVDR and the essential elements for study design to provide sound evidence for regulatory assessment.

Corresponding author: Sarah J. Lord, National Health and Medical Research Council (NHMRC) Clinical Trials Centre, University of Sydney, Sydney, Australia, E-mail: sally.lord@sydney.edu.au

Funding source: The European Federation of Clinical Chemistry and Laboratory Medicine (EFLM)

Research ethics: Not applicable.
Informed consent: Not applicable.
Author contributions: All authors have accepted responsibility for the entire content of this manuscript and approved its submission.
Use of Large Language Models, AI and Machine Learning Tools: None declared.
Conflict of interest: The authors state no conflict of interest.
Research funding: The European Federation of Clinical Chemistry and Laboratory Medicine (EFLM) provided financial support to the Working Group on Test Evaluation (now the Committee, Clinical and Analytical Performance Specifications) to attend meetings. Roche Diagnostics provided an independent educational grant to EFLM.
Data availability: Not applicable.

References

1. Asberg, A, Bolann, B. Is this quantitative test fit-for-purpose? Clin Chem Lab Med 2025;63:e165.10.1515/cclm-2025-0194Search in Google Scholar PubMed

2. Lord, SJ, Horvath, AR, Sandberg, S, Monaghan, PJ, Cobbaert, CM, Reim, M, et al.. Is this test fit-for-purpose? Principles and a checklist for evaluating the clinical performance of a test in the new era of in vitro diagnostic (IVD) regulation. Crit Rev Clin Lab Sci 2025:1–16. https://doi.org/10.1080/10408363.2025.2453148.Search in Google Scholar PubMed

3. Regulation (EU) 2017/746 of the European Parliament and of the Council of 5 April 2017 on in vitro diagnostic medical devices. https://eur-lex.europa.eu/eli/reg/2017/746/2022-01-28 [Accessed 11 Mar 2025].Search in Google Scholar

Received: 2025-03-11

Accepted: 2025-03-12

Published Online: 2025-03-24

Published in Print: 2025-06-26

This work is licensed under the Creative Commons Attribution 4.0 International License.

Articles in the same Issue

https://doi.org/10.1515/cclm-2025-0291

Keywords for this article

cut-off limit; likelihood ratio; pretest probability; posttest probability; sensitivity; specificity

Creative Commons

BY 4.0