Background/Objectives: Allogeneic Hematopoietic stem cell transplantation (Allo-HSCT) for multiple myeloma (MM) increases the risk of Graft-versus-host disease (GvHD), resulting in greater morbidity and mortality. Genetic, socioeconomic, and healthcare access factors have been reported to contribute to racial disparities in the prevalence and incidence of GvHD. Developing targeted interventions requires an in-depth understanding of population disparities in outcomes among individuals with MM. This study identified racial disparities in the incidence of GvHD and mortality among patients with MM undergoing Allo-HSCT.

Methods. We identified hospitalizations for patients with MM who underwent Allo-HSCT as frontline or salvage therapy between 2015 and 2021 from the national inpatient sample database using the ICD-9 and ICD-10 diagnostic and procedure codes.Patients under the age of 18, being in remission, had incomplete or missing race, history of autologous HSCT, multiple HSCT, primary graft failure or other autoimmune disorders were excluded. Baseline characteristics were compared between hospitalizations with and without GvHD using χ² and linear regression for categorical and continuous variables. The primary endpoints of this study were the incidence and odds of GvHD by race. Secondary outcomes included all-cause in-hospital mortality, resource utilization, and incidence of posttransplant cytomegalovirus infection/reactivation by race. Illness severity, baseline risk of mortality, and comorbidity burden were adjusted using all patient-refined diagnosis-related groups (APR-DRG) metrics and the Charlson comorbidity index (CCI). Prolonged hospitalization was defined as the length of hospital stays in the top decile for all hospitalizations in the study (≥ 44 days). Multivariate regression analysis was performed to assess the odds of the primary and secondary endpoints.

Results: Approximately 7004 MM admissions were included in the analysis. Approximately 75.3% (5,274) of the cohort were white Americans compared with black Americans (469; 6.7%), Hispanics (539; 7.7%), Asians/Pacific islanders (322; 4.6%), Native Americans (14; 0.2%), and other races (392; 5.6%). The incidence of GvHD was 11% (765). There were 397 (51.6%) males and 373 (48.4%) females. The incidence of GvHD by race was 11.1% (585) among white Americans, 12.1% (57) among blacks, 13.5% (73) among Hispanics, 6.5% (21) among Asians/Pacific Islanders, and 33.3% (5) among native Americans in the study cohort. Race was not correlated with higher odds of GvHD in the multivariate regression analysis (aOR: 0.83;95% CI: 0.69-1.01; p=0.065). There were 200 deaths in the study; 2.5% (132) white, 3.3% (15) black, 2.9%(16) Hispanic, 3.2% (10) Asian/Pacific Islanders, and 50% (7) native Americans. The mortality rate was higher among GvHD hospitalizations (35 [4.6%] vs. 165 [2.6%]; p=0.015). Asian/Pacific Islander race (aOR:6.7; 95% CI: 4.32-7.01; p=0.043) and native American race (aOR: 7.5; 95% CI: 5.40-8.82; p=0.008) were correlated with greater odds of mortality. There were 445 cases of CMV infection/reactivation (6.4%). Hispanic (aOR: 3.63; 95% CI: 1.40-4.04; p=0.041) and Asian/Pacific Islander descent (aOR: 5.10; 95% CI:1.14-8.45; p=0.033) were correlated with greater odds of post-transplant CMV reactivation/infection on multivariate regression. In total, 630 (9%) hospitalizations were prolonged. Hospitalizations of Hispanics (aOR: 2.16; 95% CI: 1.03-4.03; p=0.032) and other mixed races (aOR: 4.45; 95% CI: 1.12-6.38; p=0.014) had a higher likelihood of prolonged hospitalization. Hispanic (Adjusted mean difference [AMD]: $US 147,188; p=0.005) and Asian/Pacific Islander race (AMD: 138,686; p=0.006) were correlated with higher mean hospital costs. Asian/Pacific Islander descent was correlated with a higher mean length of hospital stay than other races (AMD: 3.13; p=0.007).

Conclusion: Although race did not independently predict the risk of GvHD, mortality rates were higher among Asian/Pacific Islanders and Native Americans. Additionally, Hispanics and Asian/Pacific Islanders were more likely to experience posttransplant cytomegalovirus infections/reactivations and prolonged hospitalization, leading to higher healthcare costs.

Limitations: Related and unrelated Allo-HSCT included. Limited out-of-hospital follow-up data.

Disclosures

No relevant conflicts of interest to declare.

This content is only available as a PDF.
2024
Sign in via your Institution