Impact of Social Determinants of Health and Novel Therapies on the Outcomes of Relapsed and Refractory Diffuse Large B Cell Lymphoma

Introduction:

Outcomes for patients with relapsed and refractory diffuse large B cell lymphoma (R/R DLBCL) have improved dramatically with the recent approvals of chimeric antigen receptor (CAR) T cell therapies and novel immunotherapies. Unfortunately, studies show persistent disparities in DLBCL that impact survival, including patient race, income, and rural residence. The U.S. Veterans Health Administration (VA) is a single-payer healthcare system that provides coverage for cancer therapies and travel support, which could address barriers to treatment. This national VA study examines the influence of social determinants of health (SDoH) and treatment patterns on the outcomes of veterans with R/R DLBCL in the era of novel therapies.

Methods:

Patients who received frontline anthracycline at any time and any systemic therapy for R/R DLBCL between 2019–2022 were identified through the VA Corporate Data Warehouse and retrospective chart review. The Social Deprivation Index (SDI) was estimated at the zip code level using the 2019 American Community Survey (Butler et al, 2013). Median household income was estimated from federal census tract data. Logistic regression was used to evaluate binary outcomes, and Cox regression for survival analyses.

Results:

225 patients across 70 VA medical centers met eligibility (Table 1). The median age was 69, and most (96%) were male, reflecting the VA population. The median follow-up was 32.7 months (range, 2.3–217.5). Patients received a median of 3 lines of therapy (range, 2–8). Novel immunotherapies were administered to 98 patients (44%): 10 (4%) received loncastuximab, 23 (10%) tafasitamab, and 83 (37%) polatuzumab. 52 (23%) and 26 (12%) patients received CAR T and autologous stem cell transplant (auto-SCT), respectively.

The estimated 2-year overall survival (OS) was 65.8% (95% CI, 59.1–71.7). In univariate analyses, age < 70 (HR 0.58, 95% CI 0.41–0.82), use of CAR T (HR 0.64, 95% CI 0.40–0.99), and auto-SCT (HR 0.39, 95% CI 0.19–0.73) were associated with improved OS. Interestingly, the use of novel immunotherapies was associated with worse OS (HR 1.43, 95% CI 1.01–2.01), but among those who received it, using it in earlier lines had better survival (HR 0.71, 95% CI 0.54–0.94). Notably, Hispanic white had worse survival than non-Hispanic white patients (HR 1.97, 95% CI 1.13–3.23), while African Americans had similar survival to non-Hispanic white patients. Higher stage showed worse survival, but this did not reach significance (HR 1.12, 95% CI 0.93-1.40). Other SDoH (sex, income, rural residence, SDI), disease traits (transformed, LDH, cell of origin, double-hit), and medical comorbidities did not impact survival. In multivariate analysis, the impacts of CAR T, auto-SCT, and Hispanic ethnicity persisted (Figure 1). Among patients receiving at least 3 lines of therapy (n = 143), the only significant covariate in multivariate regression for OS was CAR T (HR 0.57, 95% CI 0.32–0.96).

Referrals for SCT and CAR T were sent for 65 (29%) and 81 (36%) patients, respectively. The most advantaged SDI quartile had higher odds of getting an SCT referral than the least advantaged quartile (OR 5.00, p = 0.0015). Similarly, household income ≥ $75,000 had higher odds of SCT referral vs. lower incomes (OR 2.88, p = 0.004). Race did not affect SCT referral rates, and SDoH in general did not impact CAR T referrals. SDoH was not associated with whether patients actually received novel immunotherapies, CAR T, or SCT (only affecting referral rates in the latter). The most common reasons for not pursuing auto-SCT after referral were comorbidities (n = 23, 33%), pursuance of other cell therapies (n = 17, 25%), and patient decision (n = 12, 17%). The most common reasons for deferring CAR T were comorbidities (n = 15, 28%), death (n = 15, 28%), and patient decision (n = 6, 11%). At last follow-up, of the 47 patients in remission for > 1 year, 27 (57%) had CAR T or SCT as their last therapy.

Conclusion:

Cellular approaches were the most impactful factors in improving OS. Hispanic ethnicity remained an independent poor prognostic factor, but other SDoH did not impact survival in this cohort. Broad drug coverage by a single-payer system, front-line anthracycline, and equal receipt of cell therapies may have mitigated some of the impact of SDoH on DLBCL survival. Nonetheless, cell therapies had a high referral but low administration rate, highlighting the challenge of getting patients to definitive treatment.