INTRODUCTION: African Americans (AA) have higher incidence of multiple myeloma (MM), younger age at presentation, delays in starting therapy, and inadequate referral for auto-SCT compared to other races (Bhatnagar, V et al; Cancer 2015). However, prior studies are conflicted about AA patients' clinical outcomes if they have access to the same treatment as Caucasians. Herein, we evaluated survival endpoints for AA and Caucasian MM patients who underwent auto-SCT at a single center.

METHODS: We retrospectively analyzed patients with MM who underwent first auto-SCT between July 2013 and October 2016. July 2017 was the cut-off for survival data. Clinical features, MM therapies (pre- and post-transplant), treatment responses, and transplant outcomes were compared between AA and Caucasian patients. Continuous variables were compared using the t-test. Categorical variables were compared using the Fisher's exact test. Overall survival (OS) and progression-free survival (PFS) functions were estimated by the Kaplan-Meier method. Plausible risk factors for OS and PFS were assessed by the Cox regression model.

RESULTS: A total of 169 patients with MM were included (46% AA, 54% Caucasian). Patients' demographics are listed in Table 1. There were possible differences in race distribution. More AA were women (60% versus 45%, p=0.06), AA had intermediate cytogenetic risk based on fluorescent in situ hybridization (FISH) at diagnosis (47% versus 25%, p=0.06), and more AA had BMI > 25 kg/m2 (88% versus 77%, p=0.07). However, all p-values are of marginal statistical significance. AA patients were younger at auto-SCT with a median age of 57 years (range: 39-71) versus 60 (range: 42-74, p=0.01). AA had lower hemoglobin levels at transplant with 60% being at or below 10 g/dL versus 43% (p=0.03). Twenty-six percent of AA had progressive disease prior to transplant versus 11% (p=0.02) and subsequently had a longer time from diagnosis to auto-SCT with a median of 0.82 years (range: 0.39-4.03) versus 0.74 (range: 0.21-4.8, p=0.04). Among those receiving multiple induction regimens, 68% of AA had experienced progressive disease prior to transplant compared to 34% (p=0.02) and had slightly longer time from diagnosis to transplant with a median of 1.43 years (range: 0.57-4.03) versus 0.88 (range: 0.56-4.8, p=0.06). There were no racial differences in initial induction therapy type, best response prior to transplant, maintenance therapy, creatinine level at transplant, or ECOG score, and stage based on the ISS or isotype at diagnosis. Fifty-six of the patients progressed (36% of AA, 31% of Caucasian) and 22 expired (17% of AA, 10% of Caucasian). Univariate Cox regression showed significantly greater hazard of death (HR: 4.55, 95% CI: 1.93, 10.72, p=0.0005) for those with prior progressive disease. Those receiving multiple regimens prior to transplant had 2.4 times higher hazard of death (95% CI: 1.04, 5.54, p=0.04) compared to those receiving single induction regimen. Results of the univariable models did not indicate significantly different hazards of death (HR:1.69, p=0.23) or progression (HR:1.12, p=0.67) for AA when compared to Caucasians.

CONCLUSION: Even though AA patients experienced more disease progression prior to transplant and greater time to transplant than Caucasians, there were no significant differences in overall survival or progression-free survival between races.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.