MSACL 2020 US Abstract

A Suggested Standard for Validation of LC-MS/MS-based Analytical Series (batches) in Diagnostic Laboratories

Judy Stone (1) Michael Vogeser (2)
(1) University of California, San Francisco Health Dept. of Clinical Laboratories (2)Hospital of the University of the Ludwig-Maximilians-University Munich, Germany (Institute of Laboratory Medicine)

Warning: Undefined variable $headshot in /var/www/html/view_abstract/view_abstract_in_program.php on line 704

Judith Stone (Presenter) Univ of California, San Francisco Health

Presenter Bio: Judy Stone served as Asst. Professor and Assoc. Director of Clinical Chemistry at the University of Connecticut Health Center, as Asst. Professor and Director of the Core, Immunology and Toxicology Laboratories at the University of Massachusetts Medical Center; and has worked with mass spectrometry at Specialty Laboratories, San Francisco General Hospital, Kaiser Regional Laboratories, Northern California and the Center for Advanced Laboratory Medicine, Univ. of Calif. San Diego Health System. She currently works in the UC San Francisco, Parnassus Automated Chemistry STAT laboratory. She is a member of AACC and ASMS and of the team at MSACL meetings that teaches an introductory short course on quantitative clinical mass spectrometry and programs the Practical Training track/Troubleshooting Poster rounds. She was the chair for the 2009 AACC mass spectrometry online certificate course.

Relevant Financial Disclosures (within past 24 months) Honorarium/Expenses MSACL Salary Thermofisher Scientific Inc. Discussing Products? NO, presenter will NOT present educational content related to products/services of the ineligible company(ies) listed above. Conflict mitigation NOT REQUIRED.

INTRODUCTION
There has been extensive effort devoted to standardizing method validation for diagnostic LC-MS assays. There has been less attention directed to standardizing the criteria used to validate each production batch (also described as runs, series) from which results are reported to patient charts. Although initial method validation is critically important, the goal of delivering higher quality results through use of LC-MS cannot be achieved unless every batch from which patient results are reported is reviewed and approved with the same rigor that is applied to the method validation.

Issues encountered in routine production that are not addressed during initial method validation include testing by multiple analysts; larger batches with more variance in specimen matrix; degradation of instrument performance over time; use of multiple lots of reagent, extraction media, consumables, mobile phase, calibrators, internal standard; calibration drift; the pressure to meet turn-around time targets; and the low frequency occurrence of bizarre outlier patient samples.

OBJECTIVE
Present a set of parameters in table format to use for quantitative LC-MS production batch validation. Rather than defining acceptance limits, our approach is to list items to include in standard operating procedures (SOP) and batch acceptance checklists. Resources cited for users to consult in order to set their own acceptance criteria for those parameters include consensus documents (e.g. CLSI C62) and recommendations for in-house statistical derivations.

METHODS
Literature and consensus guidelines for method and batch validation were reviewed. Experience from the authors’ practice in low and high workload diagnostic LC-MS laboratories in North America and Germany was incorporated in the recommendations.

RESULTS
A table with 31 recommended validation parameters was created. Categories addressed include calibration, quality control, batch structure and timing, internal standards, branching ratios (qualifier ratios), retention times, chromatography, general, and longitudinal review of instrument metrics and LC-MS metadata. Key aspects of the table will be reviewed in the presentation and the full table will be made available for download by attendees.

CONCLUSION
We provide a set of recommendations for diagnostic LC-MS batch validation. There is little consensus and some controversy surrounding this topic. The table presented is an initial effort, designed to promote discussion and additional refinement.