Poster Session I: Myeloma and other monoclonal gammopathies - Clinical
Despite improvements in the treatment (Tx) of transplant-ineligible (TNE) newly diagnosed multiple myeloma (NDMM), many patients (pts) remain at risk of relapse within the first year of therapy, which may be driven by pt characteristics and disease biology. Conversely, continuous therapy has been shown to prolong time to relapse, as demonstrated in the pivotal FIRST trial studying lenalidomide + low-dose dexamethasone until disease progression (Rd continuous), Rd for 18 cycles (Rd18), and melphalan + prednisone + thalidomide (MPT) in TNE pts with NDMM. Indeed, in FIRST, ≈26% of pts continued to receive Rd continuous > 3 years.
To identify factors associated with early vs late relapse and further assess the impact of continuous therapy.
In FIRST (NCT00689936), pts with TNE NDMM were randomized 1:1:1 to Rd continuous, Rd18, or MPT. Rd18 and MPT were both 72 weeks in duration. Rd continuous vs MPT was the primary comparison. Predictors of early relapse were identified by univariate and multivariate analyses using a binary outcome (relapse <vs ≥ 12 mos) in 11 pt-, 18 disease-, and 2 Tx-specific baseline variables.
Of the 1623 pts in the intent-to-treat population, 345 who were censored (lost to follow-up or discontinuation or death without progression) before 12 mos, and 32 who did not meet definitions of relapse were excluded. The remaining pts were categorized as those who relapsed early (<12 mos; n = 271) and those who relapsed or were censored ≥ 12 mos (n = 975).
Five factors were associated with risk of early progression (LDH ≥ 200 U/L, ISS Stage III, high-risk cytogenetics, EORTC My20 ≥ 50, and baseline platelet count ≤ 150 × 109/L). Due to infrequent use of EORTC in clinical practice, it was replaced with ECOG PS ≥ 2 in the final model without impact on predictability. As expected, there was a lower percentage of pts with these factors in late relapse which continued to decrease in pts who relapsed after 2 years. The model was validated using MM-015 data (Palumbo, NEJM 2012).
Independent of Tx arm, pts with early relapse had a lower overall response rate (ORR) vs those with late relapse (59.8% vs 95.6%, with a ≥ very good partial response (VGPR) rate of 13.3% vs 61.6%). Responses deepened over time, and in the primary comparison (Rd continuous vs MPT), the depth of response was higher in pts with later relapse: ≥ VGPR was 86.8% vs 60.9% in pts who relapsed ≥ 3 years treated with Rd continuous (n = 159) vs MPT (n = 87), respectively. As reported previously, the median time to ≥ VGPR was 5.8 cycles, and in those pts who entered cycle 6 with ≥ VGPR, the ability to maintain responses with Rd continuous resulted in prolonged remission (median progression-free survival [PFS] not reached vs 25.8 mos with MPT). While high-risk cytogenetics is associated with poor outcomes, pts with favorable hyperdiploidy (t(11;14) or gain of chromosome 5, 9, or 15) had prolonged remissions (median PFS 44.6 vs 27.4 mos with Rd continuous vs MPT). Early relapse (<12 mos) was associated with shorter overall survival (OS) vs late relapse in Rd continuous (median 27.9 vs 67.5 mos) and MPT (18.7 vs 54.0 mos).
These results highlight the importance of achieving and maintaining a response to frontline Tx to delay relapse. Pts with early relapse had lower ORR and ≥ VGPR rates and poor OS. The predictive factors identified suggest that the individual risk of progression appears to be a combination of disease biology, genetics, and pt-specific factors and could be used to help optimize Tx strategies to improve outcomes in pts with TNE NDMM.