Clinical Impact of Next Generation Flow in Bone Marrow Vs Qip-Mass Spectrometry in Peripheral Blood to Assess Minimal Residual Disease in Newly Diagnosed Multiple Myeloma Patients Receiving Maintenance As Part of the GEM2014MAIN Trial

Introduction: GEM2014MAIN is a phase 3, multicenter, open-label, randomized trial exploring maintenance treatment with lenalidomide and dexamethasone (Rd) vs lenalidomide, dexamethasone and ixazomib (IRd) after autologous stem cell transplantation in newly diagnosed patients with multiple myeloma (MM). The study includes yearly analysis of minimal residual disease (MRD) in bone marrow (BM) by Next Generation Flow (NGF) and the results at year 2 guide further management. Mass spectrometry methods measuring low-level immunoglobulins have shown potential for peripheral-blood (PB) based MRD assessment in MM. Aim: In this study, we analyzed and compared the results and clinical impact of MRD evaluation in BM by NGF and in PB by quantitative immunoprecipitation mass-spectrometry (QIP-MS) during maintenance in patients enrolled in the GEM2014MAIN trial. Patients and Methods: Patients achieving at least stable disease at the end of the GEM2012MENOS65 trial were included in the GEM14MAIN, and randomized to receive Rd (lenalidomide 15mg po od days 1-21 and dexamethasone 20mg po od days 1-4 and 9-12) or IRd (Rd plus ixazomib 4mg po od days 1, 8 and 15) in cycles of 28 days. After two years of treatment, BM samples were analyzed by NGF (sensitivity of 3x10-6) as per protocol: if MRD was undetectable, maintenance was discontinued; if detectable, Rd with reduced dexamethasone at 20mg po od only days 1-4 was continued for 3 additional years. At the same time point, MRD was assessed in PB by QIP-MS with anti IgG/A/M, total k and total l beads using the EXENT system (The Binding Site, in development). The association of MRD detection by either method with progression-free survival (PFS) was assessed by the Kaplan-Meier method. Landmark analyses of PFS at 2 years post-maintenance were performed to minimize lead-time bias. Additional survival analyses were carried out to evaluate the clinical impact of MRD evolution during the study period. Results: 156 patients with paired BM/PB samples available after 2 years of maintenance were included in the analysis. NGF detected residual disease in 29% (45/156) of cases and EXENT in 24% (37/156). The results of both techniques were concordant in 85% of cases and discordances were due to 16 patients (10%) NGF+/EXENT- and 8 (5%) NGF-/EXENT+. Of note, 8/16 cases NGF+/EXENT- corresponded to light-chain MM (anti-free light chain beads were not used in this study). Therefore, overall, and considering NGF as a reference, the negative predictive value (NPV) of EXENT was 87% and the positive predictive value (PPV) of 78% (p<0.0001). With a median follow-up of 66 months from the start of the GEM14MAIN, MRD detection by either method associated with a shorter PFS at 5 years as compared to patients with undetectable MRD: NGF: 57% vs 89%, respectively; HR= 6.54 (2.92-14.67); p<0.0001; EXENT: 62%% vs 86%, respectively; HR= 4.79 (1.99-11.53); p=0.0005. We also analyzed the evolution of MRD between enrollment and after 2 years of maintenance, since we already had MRD results by both methods at the end of the GEM2012. Patients were grouped into 1) those with persistent MRD positivity at the two time points or converting from MRD- to + by either method and 2) those with sustained MRD negativity at the two time points or converting from MRD+ to - by either method. Both NGF and EXENT showed that persistent MRD+ or conversion from MRD- to MRD+ were associated with a shorter PFS at 66 months, compared to patients who remained MRD- or converted from MRD+ to MRD-: NGF: 54% vs 89%, respectively; HR = 8.9 (3.55-22.08); p <0.0001; EXENT: 61% vs 86%, respectively; HR: 4.95 (1.91-12.9); p=0.0010. Conclusions: Our study shows that, during maintenance, the identification of MRD by NGF and EXENT and its evolutionary pattern are both associated with a comparable clinical value in terms of PFS. Supported by a NPV of 87% as compared to NGF, mass spectrometry could be used to define the optimal time-point for BM-based MRD assessments as well as, alongside with standard methods, to introduce MRD adapted treatment approaches.