Defining functional high-risk multiple myeloma in the era of modern induction therapy Free

Treatment of multiple myeloma (MM) has advanced dramatically over the past decade with median overall survival (OS) for newly diagnosed MM patients (pts) now exceeding 10 years. However, outcomes in pts with high-risk disease remain poor. While most pts with high-risk disease are identified through baseline cytogenetic risk stratification, presence of extramedullary disease, or R-ISS stage, there is also a population of pts who may or may not have these traditional high-risk factors who relapse earlier than expected and have similarly poor outcomes. These pts are often delineated as “functional high-risk,” but a standardized definition of this population in the era of modern myeloma therapy is lacking.

Using the ASH Research Collaborative (ASH RC) Data Hub, we retrospectively searched 7,249 pts across 10 participating institutions for those with newly diagnosed multiple myeloma (NDMM) treated with a triplet or quadruplet induction regimen (with or without autologous stem cell transplant) who had response data available. We evaluated duration of response from first line therapy (PD1) on post-PD1 clinical outcomes such as: time to next treatment (TTNT), time to subsequent progression and/or death (PFS2), and OS. We also analyzed characteristics of those who did vs. did not progress at prespecified time points of interest (12-, 18-, 24-, and 36-months) from the start of induction. We utilized recursive partitioning analyses to iteratively explore all possible cutpoints and identify which one(s) provide the greatest differentiation in the risk of subsequent progression and/or death.

We identified 1,305 pts who met the eligibility criteria. Of these, 264 had PD1 data including 201 who had response follow-up (f/u) after PD1. Median f/u after PD1 in this population was 3.2 years, and median PFS2 f/u after PD1 (n=201) was 4.4 years. Months to PD1 was significantly associated with PFS2 (p=0.038). All prespecified time points examined showed inferior PFS2 compared to those with PD1 >36 mos, although each subgroup was not statistically significant, likely due to limited numbers in each group: PD1 <12 mos HR=1.44 (n=43; 95% CI: 0.90 – 2.30; p=0.126), PD1 12-24 mos HR=1.65 (n=30; 95% CI: 0.95 – 2.85; p=0.073), and PD1 24-36 mos HR=1.58 (n=52; 95% CI: 0.98 – 2.53; p=0.059). Similarly, timing of PD1 was also associated with OS post-PD1; pts with PD1 <12 mos had significantly lower OS compared to those with PD1 >36 mos (HR=2.35, 95% CI: 1.27 – 4.32; p=0.006). Our recursive partitioning analysis identified 30-months at the cutoff that produced the greatest differential in outcomes between those above vs. below the landmark. Approximately 23% of the pts analyzed had PD1 within 30 mos, and these pts had significantly lower OS compared to those with PD1 ≥30 mos (HR=1.99, 95% CI: 1.21 – 3.27; p=0.006). Similarly, pts who progressed within 30 mos of starting 1L therapy (n=98) had significantly shorter PFS2 (median=12.5 mos vs 22.8 mos; HR=1.55, 95% CI: 1.09 – 2.20; p=0.014) than those who did not (n=103). Those pts who progressed within 30 mos were more likely to have a higher R-ISS stage at diagnosis (p=0.048), but were not more likely to have high-risk cytogenetics (p=0.66) compared to pts with PD1 ≥30 mos. Interestingly, receiving SCT as part of 1L therapy was not significantly associated with improved PFS2 (HR=0.89, 95% CI: 0.62 – 1.29; 0.54) but was significantly associated with improved OS (HR=0.53, 95% CI: 0.31 – 0.90; 0.018) by univariate analysis.

Our study is the first to use a large, multicenter, cohort to evaluate the definition of functional high-risk MM in the era of novel induction therapies. Our data suggest that pts who progress within 30 mos of induction on triplet or quadruplet regimens (irrespective of whether they underwent stem cell transplant) have poorer outcomes than those who progress at 30 mos or later, and that progression prior to 30 mos represents functional high-risk disease. These pts represent a clear unmet medical need, and additional focus should be taken to identify them early with the goal of designing trials to optimize their subsequent therapies.