Risk factors for relapse and post-relapse outcomes in patients with Philadelphia chromosome-negative B-cell acute lymphoblastic leukemia receiving frontline blinatumomab-based regimens

Abstract Introduction: Blinatumomab (Blina) is a CD19/CD3 bispecific T-cell engager with demonstrated efficacy in B-cell acute lymphoblastic leukemia (B-ALL). It is now a part of the standard of care frontline treatment of newly diagnosed (ND) Philadelphia chromosome-negative (Ph-neg) B-ALL in combination with multiagent chemotherapy in the consolidation phase. There are very limited data on the outcomes of patients (pts) who relapse after frontline treatment with Blina-based regimens. Methods: Retrospective study including pts with ND Ph-neg B-ALL who received Blina in the frontline setting as a part of the multiagent HCVAD/HCMAD (Intensive chemotherapy; IC) or mini-HCVD (low intensity therapy; LIT) regimen in the induction, consolidation, or maintenance phase. Relapse was defined by the presence of >5% blasts in the bone marrow (BM) or extra-medullary involvement. Hazard of relapse was assessed through univariate and multivariate analysis via a competing risk regression model (death in remission as competing risk). Results: 239 pts received frontline Blina-based chemotherapy from 8/2013 - 5/2025. Composite complete response (CRc=CR+CRi) was attained in 236 pts (99%). For these responders, the median age at diagnosis was 43.5 years (range 18.3 - 87.7), and 69 pts (29%) had age≥ 60 years. 111 pts (47%) were obese (BMI>30 Kg/m2) at baseline. Median white blood cell (WBC) count (x 109/L) was 5.1 (0.2 - 553), with WBC>50k in 33 pts (14%), while median BM blasts were 84% (1 - 99), with BM blasts >50% in 197/230 pts (86%). Low hypodiploid/near triploid (HoTr) karyotype, CRLF2 overexpression (CRLF2 high), and KMT2A rearrangement (KMT2Ar) occurred in 11%, 19% and 5% pts, respectively, while TP53 mutations were noted in 25% of pts. Frontline regimens were IC in 135 pts (57%) and LIT in 101 pts (43%). Median time from diagnosis to Blina initiation was 2.8 months (mos) (0.1 – 38.6). Inotuzumab (Ino) was used frontline in 173 pts (73%). 54 pts (23%) underwent ASCT in 1st remission. After a median follow up of 37.2 mos from Blina initiation, 28 pts (12%) relapsed, and 36 pts (15%) died in remission. The 3-year cumulative incidence of relapse (CIR) was 13% (95% CI: 9% - 19%). On univariate analysis, age (p=0.06), obesity (p<0.01), WBC>50k (p=0.02), BM blasts>50% (p=0.04), and TP53 mut (p=0.08) were identified as potential risk factors for relapse. Notably, HoTr karyotype (p=0.12), KMT2Ar (p=0.17), CRLF2-high (p=0.34), chemo intensity; IC vs LIT (p=0.68), and use of frontline Ino (p=0.60) were not significantly associated with relapse risk. On multivariate analysis, age (HR 1.02, 95% CI: 1.003 - 1.05, p=0.02), obesity (HR 3.7, 95% CI: 2.5 - 5.2, p<0.01), and WBC>50k (HR 3.4, 95% CI: 2.4 - 4.8, p<0.01) were independently associated with increased risk of relapse. For the full relapsed cohort (n=28), the median age at initial diagnosis was 56 years (22 – 74), with age≥60 years in 12 pts (43%). The median time to relapse from the 1st dose of Blina was 6.1 mos (range 1.8 - 53.8). Relapse site was bone marrow (BM) alone in 10 pts (36%), extramedullary disease (EMD) alone in 9 pts (32%), and BM + EMD in 9 pts (32%). The most common EMD sites were lymph node/soft tissue in 9 pts (32%), and central nervous system (CNS) in 8 pts (29%). CNS only relapse occurred in 5 pts (18%). At relapse, loss of CD19 expression occurred in 3/25 pts (12%) with available testing. A total of 27 pts (96%) received at least one line of salvage therapy, with 2nd CRc attained in 18 pts (67%). 1st salvage regimens included chemotherapy backbone (+/- venetoclax, Blina/Ino) in 24 pts (89%), CAR-T in 2 pts, and Ino alone in 1 pt. Among those in 2nd CRc, 12/18 pts (67%) received subsequent consolidation with CAR-T and/or ASCT. After a median follow-up of 25.1 mos, median overall survival (OS) for the full relapsed cohort from the date of relapse was 8.6 mos (95% CI: 4.3 - 17.5), with 2-year OS 30%. The median OS for pts with age <60 and ≥60 years was 9.5 mos and 3.6 mos, respectively. The median OS from the date of relapse for pts who received CAR-T and/or ASCT as a consolidation in 2nd CRc was not reached (5-year OS 51%). Conclusions: Relapse after frontline Blina-based regimens in Ph-neg B-ALL is infrequent, with a 3-year CIR of 13% in our cohort. We identified older age, obesity, and baseline WBC>50k as key risk factors for relapse. Post-relapse outcomes are dismal, with a median OS of only 8.6 mos (2-year OS 30%), underscoring the need for newer therapies.