Trends over Time in Overall Survival of Patients with Multiple Myeloma By Race/Ethnicity and By Comorbidity Status: A Surveillance, Epidemiology, and End Results (SEER)-Medicare Analysis

Introduction: New advances in multiple myeloma (MM) treatments have improved overall survival (OS), though there is uncertainty regarding the benefits of these treatments for all patients (pts) with MM. This study assessed OS trends over time in subgroups of pts with MM defined by race/ethnicity and comorbidity status.

Methods: Newly diagnosed pts with MM (2007-16) in the US were identified from the SEER-Medicare database. Medicare enrollment on initial diagnosis date was required and pts with other cancers were excluded. Pt characteristics at initial diagnosis and comorbidities during the 6-month baseline period were described for pts with continuous enrollment in the baseline period. OS was described using Kaplan-Meier analysis from initial MM diagnosis to death. Trends in OS over time were evaluated by creating cohorts based on year of diagnosis (2007-09, 2010-12, and 2013-16), and log-rank tests compared OS between the cohorts. All outcomes were described separately for the following race/ethnicity subgroups: non-Hispanic White (NHW), non-Hispanic Black (NHB), and Hispanic (HIS) and for subgroups defined by the presence of diabetes mellitus (DM) or hypertension (HTN), two of the most common comorbidities in pts with MM, in the baseline period.

Results: The study included 21,757 pts with MM. Median age at diagnosis was 75 years (range 27-101), 51% of pts were male, and mean Charlson Comorbidity Index (CCI) was 1.2. Overall, 14,077 (64.7%) were NHW, 4078 (18.7%) were NHB, and 2371 (10.9%) were HIS. At baseline, HTN (44.9%), DM (20.4%), renal disease (14.0%), lung disease (13.6%), and heart failure (11.4%) were common comorbidities not related to MM. Monoclonal gammopathy of undetermined significance was identified in 10.0% of pts overall, and in 7.6%, 9.8%, and 11.2% of the NHW, NHB, and HIS cohorts, respectively; 5.5% of pts overall were obese.

Median OS was 3.26 years, OS improved significantly over time (P < 0.001 for 2007-09 vs 2010-12, 2007-09 vs 2013-16, 2010-12 vs 2013-16), and median OS increased with each time period (2007-09, 2.78 years; 2010-12, 3.16 years; 2013-16, 3.64 years) (Table 1). The NHW subgroup had a median age of 75 years, 53% male pts, and a mean CCI of 1.1. Trends over time in OS were similar to the overall sample, with P < 0.001 for all-year cohort comparisons and increasing median OS over time. The NHB subgroup had a median age of 73 years, 43% male pts, and a mean CCI of 1.5. Improvements in OS were observed relative to the oldest subgroup (2007-09 vs 2010-12, P < 0.05; 2007-09 vs 2013-16, P < 0.001), but not in the most recent year subgroups (2010-12 vs 2013-16, P = 0.125). The HIS subgroup had a median age of 74 years, 50% male pts, and a mean CCI of 1.4. There was no significant improvement in OS across year cohorts (2007-09 vs 2010-12, P = 0.163; 2007-09 vs 2013-16, P = 0.053; 2010-12 vs 2013-16, P = 0.850), although median OS increased over time (2007-09, 2.92 years; 2010-12, 3.48 years; 2013-16, 3.53 years).

Among pts with MM and DM (N = 4410, 20.3%), OS improved over time relative to the oldest cohort (2007-09 vs 2010-12 and 2007-09 vs 2013-16, P < 0.01). No improvement was observed between the most recent cohorts (2010-12 vs 2013-16, P = 0.844) (Table 2). Median OS showed a similar trend (2007-09, 2.44 years; 2010-12, 2.90 years; 2013-16, 2.80 years) and was shorter than the overall population. In the HTN cohort (N = 9646, 44.3%), median OS increased over time (2007-09, 2.58 years; 2010-12: 2.79 years; 2013-16: 3.07 years), but OS showed statistically significant improvements only when comparing the oldest to the most recent cohort (2007-09 vs 2010-12, P = 0.058; 2007-09 vs 2013-16, P < 0.001; 2010-12 vs 2013-16, P = 0.081).

Conclusions: Variation in OS trends over time in pts with MM was observed across subgroups. While OS improved significantly across cohorts for NHW and NHB pts, significant improvement was not seen for HIS pts. Comorbidities were common and for pts with HTN and DM, no significant improvement in OS was observed in recent years. Median OS was lower in both comorbidity cohorts compared with the total MM population. Reasons for racial disparities need to be explored further; these differences in outcomes confirm the need for equitable access to effective treatments and tailored treatments for pts with comorbidities. Additional analyses are underway assessing trends over time in gaps in survival between pts with MM and the general population for these subgroups.

Anwer:Allogene Therapeutics: Research Funding; Janssen: Consultancy; BMS: Consultancy, Research Funding, Speakers Bureau. Goldschmidt:Analysis Group Inc.: Current Employment. Tang:Analysis Group Inc.: Current Employment; Bristol Myers Squibb: Other: Analysis Group Inc. received consultancy fees from Bristol Myers Squibb for this study. Steffen:Analysis Group Inc.: Current Employment; Bristol Myers Squibb: Other: Analysis Group Inc. received consultancy fees from Bristol Myers Squibb for this study. Yilma:Analysis Group Inc.: Current Employment; Bristol Myers Squibb: Other: Analysis Group Inc. received consultancy fees from Bristol Myers Squibb for this study.. Slaff:Merck (Spouse): Current Employment; Bristol Myers Squibb: Current Employment, Current equity holder in publicly-traded company.