Determinants of overall survival (OS) of elderly patients with multiple myeloma: A National Cancer Database (NCDB) analysis of years 2004-2017.

Abstract

e20035 Background: The survival of older adults with multiple myeloma (MM) continues to improve due to advancements in therapy but there is limited real-world data around the outcome of elderly (≥ 75 years old) MM patients. These patients are at risk due to comorbidities, frailty and potential for reduced tolerance to treatment. In this IRB approved retrospective analysis, the NCDB was used to evaluate the 30-day mortality and the determinants of overall survival (OS) of elderly MM patients who were treated at commission on cancer (CoC) accredited facilities across the USA. Methods: Using the NCDB, we identified N = 54,766 patients ≥ 75 years old diagnosed and treated for MM from 2004 to 2017. Multivariable cox regression analysis with backward elimination was utilized to identify the independent survival factors, using significance level of p < 0.05. Kaplan-Meier survival curves were produced, and SAS version 9.4 was used to analyze the data. Results: Overall median survival time was 23 months; while survival rates of 1, 3, and 5-year were 61.9%, 38.0%, and 23.6%, respectively. Multivariable cox regression analysis with backward elimination method revealed that there were 10 significant independent survival factors including age, sex, race, ethnicity, education level, Charlson-Deyo score, facility type, median income, year of diagnosis, and treatment-regimen. Male patients were more likely to die compared to female patients (HR = 1.08, p < 0.0001). Black patients were predicted to have less death events compared to White patients (HR = 0.90, p < 0.0001). In addition to that, patients of Hispanic ethnicity were more likely to die compared to non-Hispanic patients (HR = 1.16, p < 0.0001). Subjects who were treated in non-academic facilities were more likely to suffer death compared to the ones who received care in academic centers (HR = 1.15, p < 0.0001). Patients with median income < $38,000 (HR = 1.10, p < 0.0001), $38,000-$47,999 (HR = 1.06, p = 0.0011) and $48,000-$62,999 (HR = 1.04, p = 0.0031) were more likely to die compared to patients with higher median income ≥ $63,000. The hematopoietic stem cell transplantation utilization rate was 0.1%. The 30-day mortality rate was 8%. Detailed analysis will be presented. Conclusions: This large real-world analysis of elderly MM patients shows that White and Hispanic patients have inferior OS compared to Blacks and non-Hispanics. Patients with lower economic status and those treated at a non-academic facility also had inferior survival compared to those with higher income and with access to care at an academic facility. The 30-day mortality is strikingly high (8%) in the elderly. The unacceptably high mortality could be improved by facilitating access to high quality care and prompt use of novel agents. Prospective studies are needed to identify and address determinants of disparity in MM care in elderly patients.