Philadelphia Chromosome-Positive Acute Lymphoblastic Leukemia: Real-World Outcomes and Disparities in the Era of Novel Therapies, 2010-2021

INTRODUCTION Philadelphia chromosome (t[9;22])-positive B-cell acute lymphoblastic leukemia (Ph+ B-ALL) historically had dismal long-term survival, particularly in elderly patients (pts). Outcomes have substantially improved with the introduction of ABL-targeting tyrosine kinase inhibitors (TKI), starting with imatinib in 2001. Subsequent new agents, including ponatinib, a 3rd generation TKI introduced in 2012 which retains activity against all imatinib resistant mutations, and blinatumomab (Blin) and inotuzumab ozogamicin (Ino), both approved in 2017, have rapidly changed the treatment landscape of Ph+ B-ALL. In addition to improving outcomes in relapsed and refractory disease, these novel therapies provide deeper and more durable responses when incorporated in frontline regimens. Many recent clinical trials report 3-year (yr) overall survival (OS) >90% with novel chemotherapy-free frontline regimens, though small real-world studies suggest that outcomes are significantly worse in the general population. While disparities may be attributed to the selectivity of trial pts, limited treatment access in underserved populations may also play a role. Data on current outcomes of Ph+ B-ALL is critically needed to support clinical research in these areas. METHODS Adult Ph+ B-ALL pts (≥18 yrs) were identified in the National Cancer Database (NCDB), for all available yrs (diagnosis [dx] 2010-2021). Kaplan-Meier and Cox regression methods compared OS by time period of dx, corresponding with approved therapies (dasatinib, 2010-2011, ponatinib 2012-2016, Blin and Ino 2017-2021), and by age group (18-39, 40-64, 65-74, 75+ yrs). Finally, the effect of race, insurance status, and income of residential zip code on OS was evaluated. RESULTS 2,283 Ph+ B-ALL pts were identified, with median age of 54 yrs (IQR 43-66). The cohort was 49.9% male; 83.5% White, 8.5% Black, and 17.8% Hispanic. With median follow up of 30.1 months, median OS for the entire cohort was 62.1 months (95% CI 50.5-73.7), with 1-, 3- and 5-yr OS of 79%, 59%, and 51%, respectively. Age was a significant predictor of OS, with 3-yr OS for 18-39 yrs of 79%, 40-64 yrs of 63%, 65-74 yrs of 46%, and 29% for those ≥75 yrs; p<0.001. OS improved with each subsequent yr of dx (Hazard ratio [HR] 0.94 [95% CI 0.92-0.96]) for all pts. Though OS from dx yr 2012-2016 was not significantly different than 2010-2011 (HR 0.79 [95% CI 0.62-1.01], p=0.065), survival from 2017-2020 significantly improved (HR 0.63 [95% CI 0.49-0.81], p<0.001). Notably, pts who were 18-39 yrs did not have an improvement in OS over the observed yrs. Pts age 65-74 yrs and ≥75 yrs demonstrated numerical improvement (3 yr OS of 29% to 47%, p=0.126, and 20% to 36%, p=0.329, respectively, from 2010-2011 to 2017-2020), though this was not statistically significant. However, pts age 40-64 yrs had significant improvements in OS, with 3 yr OS 40% for dx 2010-2011, to 60% (p=0.003) for 2012-2016 (HR 0.61 [95% CI 0.44-0.85]), and 68% (p<0.001) for 2017-2020 (HR 0.49 [95% CI 0.35-0.68]). On multivariate cox regression model adjusted for age, yr of dx, and comorbidity score, neither minority race nor Hispanic ethnicity had a significant impact on OS. Residing in a low-income zip code (HR 1.35 [95% CI 1.06-1.71]), or being uninsured (HR 2.02 [95% CI 1.42-2.88]), insured through Medicaid (HR 1.44 [95% CI 1.13-1.83]), or Medicare (HR 1.42 [95% CI 1.10-1.85], reference private insurance) were negative predicters of OS. DISCUSSION To our knowledge, this is the largest study of Ph+ B-ALL to-date, evaluating current real-world outcomes in the era of novel therapies. We found significant improvements in survival in recent years, though real-world survival lags considerably behind outcomes reported in recent clinical trials. While age remains a significant predictor of OS, pts 40-64 yrs have seen a substantial improvement in OS in the last decade. Notably, elderly pts (≥ 65 yrs), have not benefited as strongly from the introduction of novel therapies, which may be due to high early mortality. We found considerable survival disparities in underinsured pts and pts from low-income areas, signaling that access to care may be a significant barrier to optimal outcomes. Efforts to optimize treatment regimens, such as novel chemotherapy-free regimens in elderly and frail pts, and improved access to novel therapies, are critically needed in Ph+ B-ALL to replicate the success observed in clinical trials in a real-world population.