AUTHORS
Farzeen Fatma Syed, MD, Jennifer Elizabeth Collins, PhD2, Talia Alexander3, Amir Kamran, MD4
AFFILIATIONS
1 Charleston Area Medical Center, Charleston, West Virginia, United States
2 Charleston Area Medical Center, Charleston, West Virginia, United States
3 Charleston Area Medical Center, Charleston, West Virginia, United States
4 Charleston Area Medical Center, Charleston, West Virginia, United States
ABSTRACT
Introduction
Survival in multiple myeloma (MM) and diffuse large B-cell lymphoma (DLBCL) has improved markedly since the early 1990s, driven by immunomodulatory agents, proteasome inhibitors and anti-CD38 antibodies in MM, and by rituximab-based chemoimmunotherapy in DLBCL. Realizing those gains requires that patients reach subspecialty care and receive systemic therapy, which depends on where patients live and what coverage they carry. West Virginia is an informative setting: it is the only state lying entirely within the Appalachian Region, its population is dispersed across 55 counties, and specialized hematologic care is available at very few sites. Using 31 years of statewide registry data, we examined which patient and geographic factors were associated with chemotherapy receipt and overall survival, and whether the gains of successive therapeutic eras were shared across the state.
Methods
This retrospective population-based cohort study included adults aged 18–89 years with a first primary MM or DLBCLreported to the West Virginia Cancer Registry between January 1, 1993, and December 31, 2023 (IRB 26-1441). Geographic variables included rurality based on RUCA4 classification, county of diagnosis, and straight-line distance from patient ZIP code centroid to the nearest in-state treatment center, categorized as <10, 10–24, 25–49, and ≥50 miles. Outcomes were chemotherapy receipt and overall survival. Patients were categorized into disease-specific treatment eras (MM: 1993–2002, 2003–2015, and 2016–2023; DLBCL: pre-rituximab [1993–2001], rituximab [2002–2016], and contemporary [2017–2023]). Chemotherapy receipt was assessed using multivariable logistic regression adjusted for age, sex, race, and insurance. Survival was evaluated using Kaplan-Meier methods with log-rank testing and multivariable Cox regression. DLBCL survival models additionally adjusted for SEER Summary Stage.
Results
A total of 8,594 patients were included (DLBCL 4,666; MM 3,928). Age, race and insurance were most consistently associated with chemotherapy receipt and survival. Increasing age was associated with lower odds of chemotherapy receipt in both diseases (OR 0.97 per year, p<0.001) and higher mortality (HR 1.04 per year, p<0.001). Black patients were less likely to receive chemotherapy than White patients (DLBCL OR 0.52, p=0.008; MM OR 0.73, p=0.038), as were uninsured or self-pay patients (DLBCL OR 0.60, p=0.013; MM OR 0.62, p=0.040), while private insurance was associated with higher receipt in DLBCL (OR 1.60, p=0.012). Insurance also tracked survival: Medicaid was associated with worse DLBCL survival (HR 1.76, p<0.001) and private insurance with improved MM survival (HR 0.78, p=0.007).
Geographic associations were confined to MM. Rural residence was associated with worse MM survival after adjustment (HR 1.11, 95% CI 1.01-1.22, p=0.032), and patients living 25-49 miles from a treatment center had higher mortality (HR 1.12, 95% CI 1.02-1.23, p=0.019), although this did not hold at ≥50 miles (HR 1.02, 95% CI 0.87-1.18). Neither rurality nor distance was associated with chemotherapy receipt in either disease, or with survival in DLBCL.
MM chemotherapy receipt rose across eras from 45.1% to 56.3% to 65.6% (p<0.001), with improved survival in the novel agent (HR 0.83, 95% CI 0.74-0.93) and modern eras (HR 0.86, 95% CI 0.74-0.99) vs the pre-novel agent era. In DLBCL, chemotherapy receipt was higher in the contemporary era (OR 1.34, p=0.025), but after adjustment treatment era was not associated with survival (rituximab HR 0.93, p=0.18; contemporary HR 1.04, p=0.61).
Conclusions
Across 31 years and 8,594 patients in an entirely Appalachian state, older age, Black race and lack of insurance were most consistently associated with lower chemotherapy receipt, in both diseases despite their distinct biology. Geographic factors were associated with survival only in MM, where rural residence and intermediate distance were associated with worse survival, with no corresponding association in DLBCL. Improved survival across treatment eras was observed in MM but not DLBCL, where limited follow-up in the most recent era and mortality during 2020-2023 may have obscured benefit. Improving outcomes in rural hematologic malignancy may require greater attention to insurance-related disparities and referral of older and uninsured patients, in addition to travel burden.
DISCLOSURES
Farzeen Fatma Syed, MD: No relevant conflicts of interest to declare.