Bioanalytical Community

We are working on a PK assay that detects only free drug in human serum sample, where the drug is a tri-specific antibody that has specificity for BCMA.  The antibody is captured using the BCMA arm of the antibody. This complicates free drug detection in diseased patients such as Multiple Myeloma where soluble BCMA, which is often present in high concentrations, binds to the drug in solution thus making it less available for detection.  Validation testing has confirmed interference by BCMA at concentrations above 25ng/mL.  Our attempts at testing selectivity relative to a nominal spiked concentration of drug have not been successful, as we continually see significant low bias in drug quantitation that is consistent with BCMA interference.  Further, we have tested the reproducibility of drug quantitation by calculating the relative percent difference (RPD) between 2 observed values of a single sample analyzed over two days on basis of ISR but with acceptance criteria as applicable for selectivity, but still only 75% of the individuals tested had RPD within ± 25% at the LLOQ.  We believe the assay is able to accurately quantitate free drug despite the inconsistent drug quantitation we have seen, as we believe this inconsistency is due to variability of BCMA levels in individuals and is further confounded by the presence of BCMA's binding partners BAFF and April which is causing variable drug:BCMA binding and thus variable free drug quantitation.  We propose to accept the validation as-is, acknowledging the complexity of quantitation in the presence of BCMA interference, but wanted to get feedback from others who may have experience in this realm – what has your experience been and do you think regulators will accept this justification?