Recent studies have illustrated complete response (CR) rates >50% for transplant-eligible patients with multiple myeloma (MM) treated with optimized induction followed by high-dose therapy (HDT) and autologous stem-cell transplantation (ASCT). However, many patients relapse early. Patients who relapse are generally thought to have very low survival rates.
This has not been explored using the most recent therapies for treating patients with MM. Survival could be improved if it is understood which patients are most likely to relapse. Hence, this analysis attempted to define practical guidelines to predict patients at risk of unsustained CR after patients received induction therapy with bortezomib, lenalidomide, and dexamethasone (VRD), HDT/ASCT, and VRD consolidation.
To accomplish this aim, a subanalysis of phase 3 PETHEMA/GEM2012MENOS65 trial data was completed by the authors. Progression-free survival (PFS) was compared in patients receiving 1 of 2 pretransplant conditioning regimens, BUMEL or MEL-200.
A total of 262 patients who were in CR after consolidation were included in this subanalysis. Patients were treated with 6 cycles of 28-day VRD induction during the trial and received conditioning treatment (either BUMEL or MEL-200). Three months after transplantation, patients received 2 cycles of consolidation treatment with VRD. Patients then were enrolled in the PETHEMA/GEM2014MAIN clinical trial, which was designed to compare maintenance treatment with lenalidomide and dexamethasone (RD) versus RD and ixazomib after autologous hematopoietic stem-cell transplantation for 2 years. Patients then continued with RD for 3 years if they were positive for minimal residual disease (MRD) or stopped therapy if they were negative for MRD. Unsustained CR was defined as disease progression within 12 months of HDT/ASCT.
A significant effect on PFS in CR after consolidation occurred because of the presence of baseline anemia (<10 g/dL) and the International Staging System. In a multivariate analysis for PFS, Eastern Cooperative Oncology Group performance score (0 vs ≥1) and plasma cell counts (≤20% vs >20%) were independently useful for prognosis. When other tests typically performed on negative immunofixation were analyzed, the authors found that plasma cell enumeration by morphology is not a valuable prognostic analysis; median percentage of plasma cells by morphology was 1.7% (range, 0%-5%). Of the patients with negative immunofixation, only 4 (1.5%) had >5% bone marrow plasma cells and not classified as in CR. Approximately one-fourth (56/248, 23%) of the patients in CR displayed an abnormal serum-free light chain ratio, but PFS was not affected by this compared with patients with normal serum-free light chain, indicating that this stringent CR criterion is not a good predictor of relapse. In 29% of patients (76/257), however, persistent MRD was identified and was highly correlated with inferior PFS (3-year rates, 49% vs 83%; P <.00001). Despite maintenance therapy, 14 (6%) patients had unsustained CR and low survival (only 3 patients remain alive). After consolidation, many patients with unsustained CR had persistent MRD (77%), whereas only 27% of patients with sustained CR were characterized by persistent MRD (P <.00001). After HDT/ASCT, later CR achievement also varied by whether CR was sustained (68%) or unsustained (36%; P = .0085). Notably, relapse or unsustained CR was not predicted by high-risk cytogenetics and other baseline characteristics.
Results from this study determined that assessment of MRD combined with response kinetics is useful for predicting patients who may relapse after optimized induction, HDT, and ASCT. Morphologic and serum-free light chain measurements did not prove to be predictors of unsustained CR.
Abstract and Poster EP936. EHA 2020. June 12, 2020. How to define and predict unsustained CR in transplant-eligible multiple myeloma: a sub analysis of the GEM2012MENOS65 trial.