Pediatric-inspired regimens for ALL are not associated with improved overall survival in US hispanic adult patients

Abstract Background: Adult ALL outcomes have significantly improved with the adoption of pediatric inspired regiments (PIR). However, these regimens are associated with unique toxicities, particularly hepatotoxicity and pancreatitis related to incorporation of asparaginase and intensified steroid therapies. While national studies show a clear survival advantage with PIR compared to adult regimens (AR), it is unclear if treatment disruptions due to toxicities compromise this survival benefit. We previously noted that Hispanic patients (pts) with ALL receiving PIR had high rates of toxicity and treatment changes compared to Hispanic pts receiving AR (Nassar et al., Blood, 2024; 144:5873). Here we report the results of a multicenter, retrospective ALL study comparing outcomes of Hispanic and non-Hispanic pts treated with PIR or AR in the United States. Methods:We retrospectively analyzed a cohort of consecutive patients with ALL aged 16+, treated between Jan 2010 – Apr 2024 with either PIR or AR at the University of Illinois Hospital and Northwestern Memorial Hospital in Chicago, Illinois. The primary outcome was overall survival (OS). Secondary outcomes were > grade 3 hepatotoxicity (HPTX) per CTCAE v5.0 criteria and treatment change due to toxicities, defined as changing, dose-reducing, or discontinuing a protocol agent during treatment. Patients who died within two months of treatment initiation were excluded. Baseline characteristics were compared by group using Pearson's Chi-squared and Wilcoxon rank sum tests. OS was examined using Kaplan-Meier and Cox proportional hazards analyses, with censorship beyond 120 months of treatment. HPTX and treatment changes were analyzed with multivariable logistic regression models using ethnicity and regimen type as independent predictors. Our multivariable models were adjusted for age, sex, obesity, and treatment site as covariates. Sensitivity analyses were independently conducted by ethnicity and regimen subgroups. Results: 202 patients with ALL were identified with a median age of 40 yrs (IQR 27–53 yrs), of whom 118 (58%) were male, 125 (62%) self-reported as Hispanic pts, and 128 (63%) received PIR. Hispanic pts were younger (age 34 vs 44 yrs, P<0.001), but without significant differences in rates of obesity, hyperlipidemia, diabetes mellitus and hypertension (HTN) (P>0.3) compared to non-Hispanic pts. Patients receiving PIR were also younger (age 30 vs 51 yrs, P<0.001) with lower rates of HTN (18% vs 36%, P=0.003) than those receiving AR. PIR use was more common in Hispanic pts (71% vs 58%, P=0.06). In the overall cohort, as expected, patients receiving PIR versus AR had improved OS (HR 0.59, 95% CI 0.35–0.98, P=0.04). When analyzing survival outcomes by ethnicity, however, an OS benefit of PIR versus AR was observed in non-Hispanic pts (HR 0.52, 95% CI 0.27–0.98, P=0.04) but, surprisingly there was no survival benefit of PIR versus AR in Hispanic pts (HR 0.84, 95% CI 0.33–2.13, P=0.7). We next analyzed HPTX and treatment change as potential contributors to the lack of benefit for PIR in Hispanic pts. HPTX occurred in 40% of Hispanic and in 26% of non-Hispanic pts (P=0.011). Use of PIR independently predicted for HPTX (OR 8.3, P<0.001), but Hispanic ethnicity did not predict for HPTX in either treatment subgroup (OR 1.3, P=0.5). Treatment change due to toxicities was significantly more frequent in Hispanic than in non-Hispanic pts (20 vs 9.6%; P=0.013). Hispanic ethnicity independently predicted treatment change (OR 2.5, P=0.05); on sensitivity analysis by regimen, the association of Hispanic ethnicity and treatment change was only observed in PIR (OR 5.1, P=0.01), not in AR (OR 0.7, P=0.6). HPTX independently predicted treatment change (OR 4.0, P<0.005). Neither HPTX nor treatment change independently predicted OS in multivariable analysis, including in subgroup analysis by ethnicity. Conclusions: Our overall cohort findings are consistent with national studies showing survival benefits for PIR versus AR in adult ALL pts, however our subgroup analysis suggests this association may be attenuated or absent in Hispanic pts. In exploring this discrepant survival benefit, we found that while Hispanic pts have higher rates of HPTX and treatment changes due to toxicities, these secondary outcomes are not independently associated with OS in the full cohort. Future studies are needed to confirm these findings and to address the mechanisms of inequitable outcomes faced by Hispanic pts with ALL.