In Reply We understand and welcome the comments of Little and colleagues, and we agree that it would require a meta-analysis to demonstrate the utility of minimal residual disease (MRD) as a trial-level surrogate, looking at predicting between-arm differences in survival using between-arm differences in MRD status.1 We do, however, believe that the magnitude of the MRD effects is sufficiently large as to encourage testing of its use as a surrogate. This is based not only on the evidence provided in the meta-analysis, but also on the fact that MRD seems to be specifically quantifying residual myeloma posttreatment, and so there is a biological underpinning to its use as a surrogate, because treatments are specifically designed to reduce this tumor burden. We have also previously demonstrated that it is not just MRD positivity or negativity that predicts subsequent outcome but the MRD level itself, 2 which would be a consequence of the fact that it is quantifying posttreatment residual disease. If it was possible to also measure MRD regrowth rates easily, progression-free survival (PFS) would then be uniquely determined.
Munshi N, Samur MK, Gregory W. Myeloma Minimal Residual Disease and Surrogacy—Reply. JAMA Oncol. Published online May 11, 2017. doi:10.1001/jamaoncol.2017.0404