Hyper-CVAD versus hyper-CVAD plus blinatumomab for adult patients with newly diagnosed Philadelphia chromosome-negative acute lymphoblastic leukemia: A propensity score analysis

Abstract

Abstract Background While the phase II trial of Hyper-CVAD (HCVAD) with sequential blinatumomab (blina) demonstrates the benefit of blina in the frontline setting for adults with newly diagnosed (ND) Philadelphia-negative B-cell acute lymphoblastic leukemia (B-ALL) (Kantarjian AJH 2025), the pivotal phase 3 ECOG-ACRIN 1910 did not assess its impact in an intention-to-treat manner, as less than half of enrolled pts were evaluable. We, therefore, conducted a propensity matched analysis comparing the outcomes of pts with ND Ph-negative (-) B-ALL who received HCVAD vs. HCVAD and sequential blina. Methods Pts with ND Ph- B-ALL 18-60 years who received HCVAD with/without an anti-CD20 antibody (rituximab/ofatumumab) (NCT01363128) or HCVAD and sequential blina with/without inotuzumab (NCT02877303) were included. In HCVAD/blina, the number of intensive chemotherapy cycles was reduced from 8 to 4, with 4 cycles of blina replacing these. The duration of maintenance was reduced from 30 cycles of POMP (6-mercaptopurine, vincristine, methotrexate, prednisone) to 12 cycles with 1 course of blina every third course for a total of 3 courses of blina. We assessed responses and survival in pre-matched and matched cohorts through logistic regression analysis for propensity score matching among baseline covariates (age, gender, race, ethnicity, smoking status, BMI, ECOG, baseline laboratory values, CD 19/20/22 expression, cytogenetics, Ph-like phenotype, CD20 antibody). Results Two hundred and forty one pts were treated with HCVAD (November 2002-April 2021; n=161) or HCVAD/blina (November 2016-January 2025; n=80). Propensity score matching identified 55 pts in each cohort. With matching, the median age was 37 years (range, 18-59) for HCVAD and 35 years (20-60) for HCVAD/blina with AYA pts 56% vs. 60% of the pts (p=0.7), respectively. A similar number of pts in both groups had high-risk cytogenetic abnormalities including low-hypodiploidy/near triploidy, KMT2Ar, and complex karyotype (p=0.7). Twenty-six percent of pts were Ph-like in HCVAD vs. 18% in HCVAD/blina (p=0.3). Pre-matched, the overall response rate (ORR) was 151/155 (97%) in HCVAD vs. 67/68 (99%) in HCVAD/blina (p=0.99). Measurable residual disease (MRD)-negativity by multiparameter flow cytometry (FCM) overall was achieved in 120/137 (87%) of HCVAD vs. 76/77 (99%) of HCVAD/blina (p=0.002). The rate of MRD-negativity by next generation sequencing (NGS) in the HCVAD/blina cohort was 76% overall, with 26% after induction. The ORR among the matched pts was 98% in both groups. Among HCVAD, MRD-negativity by FCM was achieved in 80% of pts vs. 91% of HCVAD/blina (p=0.15). Pre-matched, after a median follow-up of 153 months for HCVAD and 36 months for HCVAD/blina, the median EFS was 50 months vs. not reached (NR) (p<0.001), with a median OS of 140 months vs. NR (p<0.001), respectively. With propensity score matching, the median EFS was 140 months with HCVAD and NR with HCVAD/blina (p<0.001). The 3- and 5-year EFS rates with HCVAD were 62% and 55%, vs. 91% and 91% with HCVAD/blina. The median OS was NR for either cohort. The 3- and 5-year OS rates with HCVAD vs. HCVAD/blina were 69% and 63%, vs. 90% and 90%, respectively (p=0.027). Despite a reduction in chemotherapy, only 3 pts (4%) relapsed with CNS disease in HCVAD/blina vs. 7 (4%) with HCVAD (p=0.99). For matched AYA, the 3- and 5-year EFS rates were 77% and 66% for HCVAD vs. 89% and 89% for HCVAD/blina, respectively (p=0.116), with 3- and 5-year OS rates of 77% and 74% vs. 96% and 96%, respectively (p=0.042). For matched Ph-like, the 3- and 5-year EFS rates were 63% and 50% vs. 83% and 83% for HCVAD vs. HCVAD/blina, respectively (p=0.271), with 3- and 5-year OS rates of 63% vs. 100% respectively (p=0.099). MVA identified age (EFS HR 1.02 p=0.003 OS HR 1.024 p=0.013), smoking (EFS HR 1.8 p=0.003 OS HR 1.8 p=0.01), BMI (EFS HR 1.06 p<0.001 OS HR 1.04 p=0.017), ECOG (EFS HR 2.1 p=0.01), KMT2Ar (EFS HR 5.9 p<0.001 OS HR 3.3 p=0.001), Ph-like (EFS HR 1.9 p=0.008 OS HR 1.9 p=0.012), and blina (EFS HR 0.2 p=0.002 OS HR 0.3 p=0.002) as significant predictors for survival. Conclusion Through propensity score analysis, we demonstrate that the addition of blina to HCVAD significantly improves survival in the frontline setting for adults with ND Ph-negative B-cell ALL. This regimen results in promising outcomes with no increased CNS relapses despite a significant reduction in chemotherapy; ongoing randomized trials are underway to confirm our findings.