Social vulnerability by neighborhood of association correlates with CAR T referrals.
Abstract
7025 Background: Chimeric antigen receptor T (CAR T) cell therapies are effective for relapsed/refractory (R/R) diffuse large B-cell lymphoma (DLBCL) and multiple myeloma (MM). However, factors influencing referrals remain unclear, particularly among the denominator of all eligible candidates. This study examines patterns of CAR T referral in the VA health system. Methods: A retrospective review identified veterans with R/R DLBCL or MM eligible for CAR T per FDA label during approval periods. Multivariable logistic regression assessed associations of CAR T referral with demographics, comorbidities, and social deprivation index (SDI). SDI is a composite social determinants of health measure at the zip code level, with higher SDI (range: 0-100) indicating greater social disadvantage. Results: Of 1,474 eligible patients across 112 VA hospitals, 25% (153/606) of DLBCL and 7.5% (65/868) of MM patients were referred for CAR T. Multivariable analysis showed that higher SDI (OR 0.90 per 10 points, 95% CI 0.84-0.96, p = 0.001) and older age (OR 0.58 per 10 years, 95% CI 0.49-0.69, p < 0.0001) were associated with a lower chance of referral, while DLBCL diagnosis (OR 3.22, 95% CI 2.28-4.59, p < 0.0001) and Hispanic ethnicity (OR 1.964, 95% 1.09-3.45, p = 0.02) were more likely to have referrals. Marital status, sex, race, psychiatric history, substance abuse, heart disease, kidney disease, cirrhosis, and rural/urban residence were not significant. Among referred patients, older age (OR 0.98, 95% CI 0.95-0.99) and substance abuse (OR 0.42, 95% CI 0.19-0.88) were linked to lower CAR T administration. Median lines of therapy for CAR T were 4 (range: 2-6) for DLBCL and 7 (range: 5-10) for MM. Common reasons for not receiving CAR T included death (23%), age/performance status (21%), alternative therapies (18%), adequate disease control (13%), comorbidities (8%), and patient preference (5%). Conclusions: Social vulnerability by neighborhood of association is associated with fewer CAR T referrals. Once referred, most variables minimally impacted CAR T administration. Efforts should focus on improving referral rates and addressing access barriers across all patients. Baseline characteristics of DLBCL and multiple myeloma cohorts. DLBCL (n = 606) Multiple Myeloma (n = 868) Median Age (range) 70 (30 – 96) 73 (36 – 97) Male Sex (%) 585 (97%) 829 (96%) Race/Ethnicity (%) Non-Hispanic White Non-Hispanic Black Hispanic Other 402 (66%)94 (16%)56 (9%)17 (3%) 458 (53%)320 (37%)53 (6%)14 (2%) Rural-Urban Residence (%) Urban Rural 410 (68%)188 (31%) 645 (74%)219 (25%) Marital Status (%) Married Never Married Divorced/Separated/ Widowed 361 (60%)64 (11%)178 (29%) 495 (57%)83 (10%)287 (33%) > 3 FACT-Accredited Centers in State (%) 342 (56%) 493 (57%) Median SDI (range) 52 (1-99) 55 (1-100) Comorbidities (%) Heart Disease Severe CKD Hepatic Cirrhosis Psychiatric Diagnosis Substance Abuse 226 (37%)55 (9%)36 (6%)336 (55%)100 (17%) 315 (36%)171 (20%)39 (4%)442 (51%)135 (16%)
Article Details
Journal Info
Journal of Clinical Oncology
Lippincott Williams & Wilkins
Authors (8)
Clare E. Anderson
Durham VA Health Care System and Duke University, Durham, NC
Micaela R. Scobie
National Oncology Program Office, Department of Veterans Affairs, Washington, DC
Robin Nanda Baidya
National Oncology Program Office, Department of Veterans Affairs, Washington, DC
Daniel McSkimming
National Oncology Program Office, Department of Veterans Affairs, Washington, DC
Thomas David Rodgers
Durham Veterans Affairs Medical Center, Durham, NC
Daphne R. Friedman
Durham VA Health Care System and Duke University School of Medicine, Durham, NC
Michael J. Kelley
National Oncology Program Office, Department of Veterans Affairs, Durham VA Health Care System, Duke University, Durham, NC
Chenyu Lin
Department of Earth and Planetary Sciences, Harvard University