Abstract
CLSI-C62A – A new standard for clinical mass spectrometry
K Lynch
ClinChem. Epub ahead of print 1 October 2015. pii: clinchem. 2015.238626
The Clinical Laboratory Standards Institute (CLSI) liquid chromatography-mass spectrometry (LC-MS)-approved Guideline (C62-A), published November 2014, provides guidance to clinical laboratories for bio-analytical method validation. Like other useful documents from CLSI, this one comes at a price of $140 and comes as a companion to the excellent general document on MS in clinical laboratories (C50 – a further $140). The article by Lynch is valuable because it compares standards of the CLSI with the free guidelines on bio-analytical method validation from the Food and Drug Administration (FDA) Center for Drug Evaluation and Research, and the European Medicines Agency (EMA). The CLSI assay verification according to Lynch needs reference to EP5 (precision), 6 (linearity), 7 (interference testing), 9 (comparison and bias with patient samples), 10 (preliminary quantitative procedure), 14 (commutability of processed samples), 15 (precision and bias), 17 (detection capability), 19 (using CLSI documents for clinical laboratory measurement procedures), 21 (total analytical error) and 32 (metrological traceability), each costing $180, so in all beyond the budgets of hospital laboratories – although they are general methods, not confined to MS. Lynch did not include validation of reference ranges (CLSI C28) and did not need to include a specific guideline for androgens and estrogens (C57), that is of interest to a selected group of analysts.
There are many similarities between the three standards in imprecision, sensitivity, interferences, stability and recovery of clinical assays. Carryover was not addressed by the FDA guideline, and linearity was not addressed by FDA or EMA. The CLSI specification for accuracy needs definition of biological variation, clinical guidelines established by expert groups and local or regulatory requirements. For linearity, CLSI requires non-linearity to be assessed for clinical significance. On balance, the FDA and EMA documents will be the choices of clinical laboratories unless directed differently.
For the reporting of measurements of steroid hormone and other analytes, some journals will require that all assays meet minimal analytical validity standards of accuracy, precision, specificity, sensitivity, reproducibility and stability to FDA standard, hence looking for data derived with MS methods. Lynch emphasises that only one LC-MS/MS method has FDA approval, all other methods are laboratory-developed tests (LDT), and the FDA is developing a risk-based system for review of LDT. More MS manufacturers therefore need to provide turnkey solutions for laboratories, and more certified reference materials (CRM) are needed if harmony is to be achieved. The immense volume of work to fully validate a new method will be beyond the resources of some laboratories, but it is essential that MS is exploited wherever possible to improve the quality of service to the users (ultimately the patients). Currently, MS methods use selected reaction monitoring but methods based on full-scan mode for ion response plotting and spectral recognition will become more common in quantitative and qualitative tests. Standards will also be needed for proteomic tests.