Treatment outcomes in adults with newly diagnosed B-cell acute lymphoblastic leukemia: A comparison between pediatric-inspired regimens and HyperCVAD at a Hispanic-serving cancer center.

Abstract

7048 Background: It is unknown if HyperCVAD is more favorable than pediatric-inspired regimens (PIRs) among adult Hispanic patients with newly diagnosed B-cell acute lymphoblastic leukemia. This analysis aims to determine the interaction of Hispanic ethnicity and treatment regimen on outcomes. Methods: Records of adults (≥ 18 years) treated with a PIR or HyperCVAD from January 2011 to November 2022 at our institution were reviewed. Patients were excluded if they received treatment modifications in a clinical trial or were lost to follow-up prior to 6 months. The primary endpoint was event-free survival (EFS) comprised of relapse, death, secondary malignancy, or failure to achieve complete remission. Secondary endpoints included overall survival (OS). Ph+ disease was not grouped into the poor-risk category. Results: 100 patients were included. Median age was 39 years, 57% were male, and 61% self-identified as Hispanic. 48 were included in the PIR and 52 in the HyperCVAD groups. 37 in the PIR group received CALGB 10403. There was no difference in race, ethnicity, rituximab use, WBC, LDH, bone marrow blast %, presence of CNS disease at diagnosis, or poor-risk genetics (Table) between the PIR and HyperCVAD groups. However, a lower rate of Ph+ disease (8.3% vs 58%) and younger median age (31 years vs 49 years) was observed in the PIR group. Allogeneic stem cell transplantation (ASCT) was pursued in 48% of PIR and 37% of HyperCVAD-treated patients. EFS at 5-years was lower among patients receiving a PIR compared to HyperCVAD (28% vs 41%; hazard ratio [HR], 1.89; P = .022). After censoring for ASCT there was no change (P = .03). Treatment group was the only significant predictor for EFS although a treatment-by-subgroup interaction (P < .10) with Hispanic ethnicity was found. OS at 5-years was 61% in PIR and 64% in HyperCVAD-treated patients (HR, 1.54; P = .25). In multivariable Cox regression analysis, Hispanic ethnicity (HR, 2.56; P = .041) and male sex (HR, 2.62; P = .027) were predictors for worse OS. Hispanic patients demonstrated worse EFS with a PIR compared to HyperCVAD (HR, 2.52; P = .008), which remained significant when evaluating only Ph- patients (HR, 2.20; P=.048). Non-Hispanics did not experience the same difference in EFS (HR, 0.88; P=.81). Conclusions: Treatment with a PIR resulted in worse EFS over HyperCVAD in adult Hispanic ALL patients. Future studies focusing on regimen intensity, adherence, and toxicities are needed to explore potential causes for this disparity. [Table: see text]