OA-53: Peripheral Blood and Bone Marrow Residual Disease Negativity to Identify a Subgroup of Newly Diagnosed Multiple Myeloma Patients with Long Term Disease Control

Introduction: In patients (pts) with myeloma (MM), analysis of minimal residual disease (MRD) in bone marrow (BM) by NGF has shown a solid prognostic value. Mass spectrometry (MS) has potential for Peripheral-blood (PB) based Residual Disease (PRD) assessment. However, the value of PRD in the context of standard and MRD methods of response assessment is limited.

Methods: In this study, we compare MRD and PRD in 1019 paired BM and PB samples from newly diagnosed MM pts in the GEMCESAR, GEM2014MAIN and GEM2012MENOS65 trials. PRD was analyzed by MS with anti IgG/A/M, total  and  beads using the EXENT®Solution. MRD was analyzed following the recommendations of the IMWG. MRD and PRD were assessed in parallel at the pre-specified time-points common to the three trials: post-induction, post-ASCT, post-consolidation and after 2 ys of maintenance.

Results: Both PRD and MRD segregated two groups of pts with significantly different progression-free survival (PFS): not reached (nr) in cases PRD or MRD negative and with a mPFS of 5.52 ys in PRD+ cases (p < 0.0001) and 4.99 ys in MRD+ cases (p < 0.0001) (Fig. 1A). 78.1% of the results were concordant (326/1019 PRD+/MRD+; 471/1019 PRD-/MRD-) and 22.1% discordant (86/1019 PRD+/MRD-; 136/1019 PRD-/MRD+). Taking the results of NGF as a reference, the negative predictive value of EXENT®Solution was 77% overall and higher by MRD levels: 94% in ≥10-4 (p < 0,0001), 91% in ≥10-5 (p < 0,0001) and 92% in ≥10-6 (p < 0,0001).
PRD and MRD at the pre-determined time points common to the 3 trials (post-ind [n=242], post-ASCT [n=230], post-cons [n=237] and 2nd year of maintenance [n=192]) was associated with a significant clinical value in terms of PFS (data not shown).
Focusing in the 695 samples in complete response or better (>CR), 21% were PRD+ and 29.5% MRD+. PRD and MRD were able to segregate two groups of pts with similar and very significantly different PFS: nr in cases PRD or MRD negative and with a mPFS of 4.65 ys in PRD+ cases (p < 0.0001) and 4.45 ys in MRD+ cases, respectively (p < 0.0001) (Fig1B).
Finally, we analyzed the clinical impact of the combination of PRD plus MRD. Importantly, we observed that including the 1019 samples, the mPFS has not yet been reached in the samples PRD-/MRD- but is of 5.56 ys if reaching negativity by any of the two techniques and 4.92 ys if PRD+/MRD+. Also, in samples >CR, the mPFS was not reached in for PRD-/MRD- but it is of of 5.07 ys if achieving negativity by any of the two techniques and 3.38 ys in PRD+/MRD+ cases.

Conclusions: We confirm that MRD and PRD associate with a very significant clinical value, both independently of standard responses as well as pts in >CR, the latter suggesting the need for the definition of a deeper degree of serological response to correctly stratify pts evolution. We also show that the combined analysis of PRD and MRD, specially in CR cases, allows to better stratify pts outcome by identifying a very favorable subgroup with long term disease control.