AUTHORS
Karthik Chetlapalli, MS, Ding Quan Ng, PhD2, Noffar Bar, MD3, Yuxin Liu, MD4, George Goshua, MD, SM5
AFFILIATIONS
1 Yale School of Medicine, New Haven, Connecticut, United States
2 Yale School of Medicine & Yale Cancer Center, New Haven, Connecticut, United States
3 Yale School of Medicine & Yale Cancer Center, New Haven, Connecticut, United States
4 Dana Farber Cancer Institute, Boston, Massachusetts, United States
5 Yale School of Medicine & Yale Cancer Center, New Haven, Connecticut, United States
ABSTRACT
*KC and DQN are co-first authors
Introduction: BCMA-targeted T-cell engager (TCE) and chimeric antigen receptor T-cell (CAR-T) therapies are being considered in earlier lines of treatment for multiple myeloma via landmark MAJESTEC-3 and CARTITUDE-4 trials. Historically, BCMA-targeted TCEs have been less efficacious and durable than CAR-T therapies, but are readily available, off-the-shelf therapies with less severe adverse effects, including CRS and ICANS. CAR-T therapies typically require monitoring at FACT centers during the apheresis, conditioning, and cell therapy, with at least a month lag between treatment election and receipt. Prior studies demonstrated geographic disparities in access to CAR-T, but the impact of the rapid adoption of TCEs as an alternative has yet to be explored. We sought to fill this gap by investigating the association of patient-level demographic, socioeconomic, and clinical characteristics with real-world administration of second-line or later TCE versus CAR-T therapy.
Methods: Using Epic COSMOS electronic health record data, we identified multiple myeloma patients who received second-line or later BCMA-targeted TCEs (teclistamab, elranatamab, or linvoseltamab) or CAR-T therapy (ciltacabtagene autoleucel, or idecabtagene vicleucel) from 2022 – 2026, after exposure to one to three previous treatment lines (daratumumab, , bortezomib, lenalidomide, pomalidomide, carfilzomib, selinexor, or isatuximab) in the two years prior. We analyzed age, sex, race/ethnicity, marital status, research participation, state of residence, social vulnerability index (SVI), and rural-urban commuting area (RUCA) codes at time of treatment exposure and extracted insurance status and the most recent serum free kappa/lambda levels in the 6 months prior. We performed multivariate logistic regression to evaluate the association between patient-level characteristics and treatment exposure, presenting effect sizes as adjusted odds ratios (OR) with 95% confidence intervals. OR > 1 denote higher odds of TCE treatment over CAR-T therapy.
Results: We identified 4,976 patients who received TCE or CAR-T therapy (2,972 TCE and 2,004 CAR-T), with median age of 71 and 67 years, respectively. Overall, 54.9% of patients were male; 55.2% were white, 17.5% were non-Hispanic black, and 13.6% had multiple races; 64.6% were married/partnered; 49.4% had private insurance, 39.7% had Medicare, and 5.0% had Medicaid.
Compared to younger patients (aged <65), patients aged 65-69, 70-75, and 75+ were more likely to receive TCE as seen from adjusted ORs of 1.19 (0.99-1.43), 1.49 (1.24-1.79) and 4.53 (3.76-5.46) respectively. Compared to white patients, multiple race patients had an adjusted OR of 1.51 (1.24-1.84). Compared to married/partnered patients, single patients had an adjusted OR of 1.35 (1.17-1.55). Compared with private insurance, patients with Medicaid and Medicare had an adjusted OR of 1.49 (1.11-2.02) and 1.06 (0.90-1.19), respectively. Compared to patients living in the West, patients in the Midwest, Northeast, and South had an adjusted OR of 1.32 (1.08-1.61), 2.42 (1.95-3.01), and 1.69 (1.40-2.04). Compared to patients living in urban areas, patients from rural regions had an adjusted OR of 0.88 (0.73 -1.05). Compared to a reference group of patients with an SVI score of 0.0-0.24 (i.e., least vulnerable quartile), patients with an SVI of 0.25-0.49, 0.50-0.74, 0.75-1.00 had an adjusted OR of 1.07 (0.88-1.30), 1.05 (0.87-1.27), and 1.3 (1.07- 1.58) respectively.
Among 1,082 patients with kappa-dominance (kappa/lambda ratio >3), patients who received TCE reported a higher median kappa light chain of 286.1 mg/L (interquartile range: 101.0-867.8) compared to patients who received CAR-T of 84.3 mg/L (33.1- 256.9). For 727 patients with lambda dominance (kappa/lambda ratio <0.3), patients who received TCE, reported a higher median lambda light chain of 433.3 mg/L (154.6-1559.2) compared to patients who received CAR-T of 183.1 mg/L (53.8-496.0).
Conclusion: BCMA-targeted TCE recipients with multiple myeloma tended to be older, have more burden of disease, with less social support, and be from communities of higher social vulnerabilities compared to BCMA-targeted CAR-T recipients. Given the high cost for both treatments, the tradeoff between bridging socioeconomic gaps and providing the most durable/efficacious treatment for a patient is an evolving ethical and clinical discussion for multiple myeloma experts and health systems.
DISCLOSURES
Karthik Chetlapalli, MS: No relevant conflicts of interest to declare.