Impact of Social Determinants of Health on Survival Outcomes in Multiple Myeloma Patients at a Major Cancer Center

Introduction: Social determinants of health (SDOH), including ethnic background, education, housing, employment, healthcare access, and economic stability, have been shown to impact outcomes in cancer, however the influence on outcomes of multiple myeloma (MM) patients is not well documented. Area deprivation index (ADI) is a tool that categorizes residential areas based on 17 US Census block indicators of poverty, education, housing, and employment. Here we aimed to investigate the influence of SDOH, income and ADI on clinical outcomes including overall survival (OS), progression free survival (PFS), overall response rate (ORR), and tolerability.

Methods: This retrospective study at a tertiary cancer center included newly diagnosed patients with MM from 2013-2021. Patients who received treatment at an outside facility or those with solitary plasmacytoma were excluded. SDOH data was gathered from electronic medical records, 2020 U.S. Census (USC) data, and ADI based on patients’ listed addresses at diagnosis. Income was categorized into lower class ($0-50,000), middle to upper middle class ($50,001- 250,000), and upper class (>$250,000) groups. ADI was categorized into groupings: ADI 1-3 (least disadvantaged areas), ADI 4 (moderate), and ADI 5 (most). Additional data analyzed included stratification using a revised international staging system (R-ISS) criteria, treatment, response based on the International Myeloma Working Group (IMWG), and toxicities. Endpoints were assessed using univariate and multivariable Cox regression models, and Kaplan-Meier estimators. Associations with categorical outcomes were assessed using Fisher’s exact tests.

Results: 139 patients were followed for a median of 88.6 months, (95% CI: 83.1-95.5) of which 51.8% were male with a median age of 65 years (IQR 56-73). 30.5% held at least a bachelor’s degree and 59.9% were employed. Most patients were white (81%), married (61.2%), and spoke English (95.7%). According to USC, 20.1% were categorized as lower class, 79.9% as middle to upper class, and none as upper class. Most patients resided in deprived areas; 29.5% lived in ADI 5, 35.3% in ADI 4, and 35% resided in ADIs 1-3.

Initial chemotherapy treatment included doublets (16.5%), triplets (78.4%; 12.2% were cyclophosphamide-containing), quads (2.2%) and 2.9% others. The median number of induction cycles was 4, with 79% complete remission (CR) or very good partial response (VGPR), while 18% had less than VGPR, and 3% had progressive disease during induction therapy. Post induction, 59% underwent autologous transplants, 28.8% were ineligible, 10.8% refused/ delayed, and 1.4% did not undergo transplant for unknown reasons.

Median overall survival for the whole population was not reached, and the median PFS was 42.1 months (IQR:35-55.6). OS analysis showed worse outcomes for patients who don’t speak English (P=0.02), those in ADI 5 (P=0.02), and lower class patients (p=0.02). Worse OS was also associated with ECOG scores 2-3 (p <0.01), advanced R-ISS stages 2 and 3 (p=0.01, p <0.01, respectively), and comorbidity score > 1 (p=0.02). In contrast, younger age (≤65 years) (p= 0.01), actively working (p<0.01), achieving a CR/VGPR response (p=0.01), and undergoing transplant (p <0.01) were associated with better OS. Lower class income status was associated with fewer transplants received (p=0.02) while being married (p=0.04), stage II disease (P <0.001), and triplet therapy (p <0.001) were associated with increased likelihood of transplant. The univariate analysis revealed a shorter PFS in patients with ADI 5 (p=0.02), advanced R-ISS stage at diagnosis (Stage III HR 5.93, p<0.01), and adverse cytogenetics (p=0.02). Undergoing transplantation (p<0.01) and achieving CR/VGPR (p<0.01) were associated with better PFS. Factors such as gender, race, religion, distance, education, smoking, and illicit drug use did not show significant associations with OS or PFS. None of the SDOH variables were associated with treatment toxicities.

Conclusions: High ADI was associated with worse clinical outcomes in MM; patients living in the most disadvantaged areas, being lower class, or non-English speaking, fared poorly. Additionally, there were disparities associated with SDOH in transplant eligibility, impacting outcomes. These findings highlight the impact of SDOH and access on MM prognosis and the importance of addressing these factors in clinical practice.