Mortality of Patients with Multiple Myeloma after the Introduction of Novel Therapies in the United States

• Published in: Blood 2019-11, Vol.134 (Supplement_1), p.72-72
• Main Authors: Binder, Moritz, Nandakumar, Bharat, Gertz, Morie A, Dispenzieri, Angela, Lacy, Martha Q., Maurer, Matthew J., Hayman, Suzanne R., Buadi, Francis K., Dingli, David, Kapoor, Prashant, Go, Ronald S., Hwa, Yi L., Fonder, Amie, Hobbs, Miriam A., Lin, Yi, Leung, Nelson, Kourelis, Taxiarchis, Warsame, Rahma, Kyle, Robert A., Rajkumar, S. Vincent, Kumar, Shaji K.
• Format: Article
• Language: English
• ISSN: 0006-4971; 1528-0020
• DOI: 10.1182/blood-2019-124936

Introduction: Advances in the understanding of disease biology, the introduction of new drugs, and better supportive care have improved outcomes in multiple myeloma (MM). Most improvements have been observed in clinical trial and tertiary referral center populations but questions remain about the generalizability of these findings to patients treated in the community. Methods: We studied all patients diagnosed with MM between 01/01/2001 and 12/31/2015 who had complete demographic and overall survival (OS) data available and were seen at Mayo Clinic (MAYO) or followed in the Surveillance, Epidemiology, and End Results Program (SEER, 18 registry data 2000 - 2016, 11/2018 submission). We retrieved age at diagnosis, sex, date of diagnosis, date of last follow-up, and OS for all patients. OS was defined as the time from diagnosis to death from any cause. Patients who were alive at the end of follow-up (12/31/ 2016) were censored. OS estimates were calculated using the Kaplan-Meier method. Age- and sex-adjusted Cox models were used to assess the association between the 5-year interval of diagnosis and OS. Expected OS estimates were calculated based on United States general population rate tables (Human Mortality Database) using a conditional approach. Standardized mortality ratios (SMR) were calculated by dividing the number of observed deaths by the number of expected deaths in age- and sex-matched general United States population controls. Linear regression models were fit to test for linear trends in early mortality and SMR over time (per 5-year interval). P-values below 0.05 were considered statistically significant. Results: The median age at diagnosis was 3 years lower in patients at MAYO (64 years, 15% ≥ 75 years, 60% male, n = 3293) compared to SEER (67 years, 29% ≥ 75 years, 55% male, n = 61779). After a median follow-up of 2.8 years the median OS was longer in MAYO compared to SEER (5.4 versus 4.0 years, HR 0.82, 95% CI 0.78 - 0.86, p < 0.001) and remained statistically significant after adjusting for age and sex (HR 0.91, 95% CI 0.86 - 0.95, p < 0.001). Early mortality (1-year mortality) decreased between 2001-2005 and 2011-2015: From 20% to 11% at MAYO and from 26% to 19% in SEER. When grouping OS by year of diagnosis (in 5-year-intervals) improvements were seen in both populations (A) and remained statistically significant after adjusting for age and sex. The relative improvements were similar comparing the 5-year period after the introduction of th