Crystal City VI: Assay Criteria and Level of Validation Required for Biomarkers Will Depend Upon the Intended Purpose of the Assay

Written by Jennifer Zimmer, Alturas Analytics Inc

At the AAPS-FDA Crystal City VI (MD, USA) meeting, regulators and industry gathered to discuss the draft bioanaytical method validation guidance and how it relates to the assay of biomarkers, both by LC-MS and by immunoassay. One thing was clear throughout, biomarker assays are not PK assays; however, some of the tests used in the validation of PK assays are still relevant for the validation of biomarker assays. In addition, there is a wealth of knowledge that has been gathered over the last 25 years by the clinical chemistry community who operate under CLIA guidelines. While CLIA may not be entirely applicable to the bioanalytical validation of biomarker assays, there are portions of the CLIA regulations that can be adapted for use in this purpose.

The overall consensus at the meeting was that the assay criteria as well as the level of validation required for biomarkers will depend upon the intended purpose of the assay. The acceptance criteria for biomarker assays will not be as prescriptive as the criteria for PK assays and should take into account the biology of the biomarker, ensuring the assay can measure with precision any biologically relevant changes between disease and normal and/or between treatment groups.

In order to demonstrate assay performance, the following should be evaluated: precision, sensitivity, assay range, relative (or absolute, if possible) accuracy, parallelism and sample stability. The measurement of precision, sensitivity and assay range for biomarker assays does not vary much from the measurement of these parameters in the PK realm. As absolute accuracy is not possible without a true reference standard, most biomarker assays will rely upon relative accuracy. Parallelism is imperative to understanding the performance of your biomarker assay. In order to demonstrate parallelism, the response curve generated by diluting samples of the endogenous matrix and analyte should be parallel to the standard concentration response curve. Without a true reference standard that allows the creation of spiked standards that represent incurred samples, determining short-term and long-term stability becomes more challenging because there is no nominal concentration against which to measure. In addition, measuring a true t=0 timepoint presents a challenge when there is limited availability of fresh samples; regardless, best efforts must be made to assess sample stability from collection through analysis.

Once the biomarker assay has been sufficiently validated, the validation reports should go into much more descriptive detail than a typical PK validation report in order to allow the FDA reviewer to verify that the assay performance allows the validated assay to measure relevant changes in the disease biomarker.

Click here to view Alturas Analytics Inc company profile.