Interval Progression Serves As a Predictor of Adverse Outcomes in Patients with Multiple Myeloma

Background: Multiple myeloma (MM) remains an incurable disease, but deep, lasting remission can be achieved with induction therapy and autologous stem cell transplant (ASCT). The current ASCT treatment timeline includes a gap between the end of induction and the day of ASCT to permit correct mobilization and collection of stem cells. During this interval, some patients experience increases in their MM markers. We specifically evaluated if patients with interval progression (IP) have different outcomes post-ASCT compared to patients without IP.

Methods: We completed a retrospective analysis of 482 patients who underwent ASCT at The Ohio State University Wexner Medical Center between 2011 and 2016. MM labs, including protein electrophoresis with immunofixation and free light chains, were obtained at day of diagnosis, end of induction, and day of ASCT. A 20% increase in the index protein between the end of induction and the day of ASCT was defined as IP. The primary outcomes evaluated were time-to-progression (TTP), progression-free survival (PFS), and overall survival (OS). TTP was defined as time from ASCT to disease progression, treating death without progression as a competing risk. PFS was defined as the time from ASCT until either disease progression or death, while OS was defined as the time from ASCT until death or last follow-up. Poisson regression models with robust variance were used to evaluate the associations between patients' characteristics and IP, while competing risk regression models were used for TTP analysis, and Cox regression models for OS and PFS analysis.

Results: 482 patients who underwent ASCT for MM between 2011 and 2016 were studied. The patient population was primarily male (n = 296, 61.4%), Caucasian (n = 420, 87.1%), and had IgG disease (n = 269, 55.8%). MM stage at diagnosis was equally represented. Cytogenetic analysis showed 108 (22.4%) patients with 1q21 amplification, 60 (12.4%) patients with t(11;14) translocation, 33 (6.8%) patients with t(4;14) translocation, and 40 (8.3%) patients with 17p deletion. More than 80% of patients underwent maintenance therapy following ASCT.

One hundred and fifty-six (32.3%) patients were defined as having IP. Patients in the IP group were significant more likely to have LC disease (p = 0.0018), but no other patient characteristics varied significantly between the two groups. LC disease (relative risk (RR _adj) = 2.12, 95% CI 1.37-3.30, p = 0.001) and deletion 17p (RR _adj = 1.62, 95% CI 1.08-2.42, p = 0.02) were associated with increased risk of IP on multivariate analysis (MVA).

Five-year PFS rates were 47.2% and 36.8% in the non-IP and IP groups, respectively. Five-year cumulative progression rates were 49.0% in the non-IP group and 57.8% in the IP group. Five-year OS rates were 76.8% in the non-IP group and 69.6% in the IP group. On univariate analysis (UVA), IP correlated with inferior outcomes in both TTP (hazard ratio (HR) = 1.39, p = 0.01) and OS (HR = 1.39, p = 0.04). In agreement with previous literature, other factors independently correlated with inferior TTP and OS outcomes, including MM stage, cytogenetics (amp1q, t(4;14), and del(17p)), and absence of maintenance therapy post-ASCT. On MVA, IP was no longer significantly associated with a shorter TTP (HR = 1.16, p = 0.35), but approached statistical significance in conferring inferior OS (HR = 1.40, p = 0.07). Notably, once deletion 17p was removed from the MVA, the presence of IP once again conferred significantly lower OS (HR = 1.45, 95% CI 1.01-2.07, p = 0.04), suggesting that deletion 17p, which is a significant risk factor for both IP and OS, may mediate the effect of IP. Combined, all data suggest that uncontrolled disease prior to ASCT trends towards worse long-term outcomes.

Conclusion: In this study, we report the first data regarding patients who experienced rapid disease progression in the interval between induction therapy and ASCT. These patients have inferior TTP, PFS, and OS, likely due to the presence of LC disease or deletion 17p. IP may provide a valuable predictor of robustness of response to ASCT and prompt the need for additional induction cycles or alternative regimens. This is especially relevant for patients with deletion 17p. Indeed, these data support a prospective study of concepts to improve outcomes for patients with IP prior to ASCT.