Acute Lymphoblastic Leukemia (ALL) Outcomes after Allogeneic Blood or Marrow Transplantation (alloBMT) with High-Dose Post-Transplantation Cyclophosphamide (PTCy) in the Era of More Effective Pre-Transplant Therapy

Background: The therapeutic landscape in ALL changed dramatically with the approval of blinatumomab (BLIN) for relapsed/refractory B ALL in December 2014. BLIN has subsequently moved earlier in therapy, while ponatinib; inotuzumab ozogamicin (INO); and brexucabtagene autoleucel have been approved. AlloBMT remains an important option for ALL consolidation in high-risk CR1 and ≥CR2. PTCy facilitates HLA-mismatched transplants and reduces GVHD without compromising survival in HLA-matched transplants compared to standard GVHD prophylaxis. We analyzed outcomes of alloBMT with PTCy for ALL to ascertain the impact of changes in therapy.

Methods: The alloBMT database at Johns Hopkins was queried for adult patients (age ≥18) undergoing a first transplant with PTCy from January 2008-June 2022. ERA1 was from 2008-2014 before BLIN approval, and ERA2 was from 2015-2022. Patient characteristics were compared using Fisher's exact test for categorical variables and the Student t test for continuous variables. Estimators of overall survival (OS) and relapse-free survival (RFS) were reported using the Kaplan-Meier method. Differences in time-to-event outcomes were estimated using Cox proportional hazards model, or Fine and Gray's model for cumulative incidence of relapse (CIR)/non-relapse mortality (NRM) considering competing risks.

Results: There were 102 and 149 patients in ERA1 and ERA2, respectively. ERA1 patients were younger (median age 45 vs. 50, p=0.03) and more often had an HCT-CI of 0 (37% vs. 21%, p=0.006). Overall, 85% of patients had B ALL and 15% had T ALL. The majority of patients (74%) were transplanted in CR1, 26% in CR after salvage for primary refractory disease (6%) or relapse (20%), and one patient (0.4%) had refractory disease. B ALL patients were more frequently Ph+ in ERA1 (62% vs. 46%, p=0.03). The incidence of persistent MRD by flow cytometry (sensitivity 1/10,000) was higher in ERA1 (20% vs. 9%, p=0.01). The majority of ERA1 patients received myeloablative conditioning (56%) compared to 3% of ERA2 patients (p<0.0001). ERA1 patients were more likely to receive HLA-matched sibling (28% vs. 15%, p=0.02) or unrelated (25% vs. 9%, p=0.001) grafts, and less likely to receive haploidentical (48% vs. 68%, p=0.003) or HLA-mismatched unrelated (0% vs. 9%) grafts. Among Ph+ B ALL in CR1 in ERA1, 56% had received imatinib (IM) at diagnosis, whereas just 7% of such ERA2 patients received IM (p=0.0001). None of the ERA1 patients with B ALL received BLIN or INO prior to transplant. In contrast, 33% of ERA2 patients with B ALL in CR1 had received BLIN, while 86% and 31% of ERA2 patients undergoing transplant following salvage had received BLIN and/or INO, respectively.

The 5-year OS was 56% and 72% (HR 1.85, p=0.005), while RFS was 44% and 65% (HR 1.93, p=0.001) in ERA1 and ERA2, respectively. Five-year NRM was 13% in both ERA1 and ERA2 (HR 1.13, p=0.73), while CIR was 44% in ERA1 and 23% in ERA2 (HR 2.24, p=0.0005). In T ALL, OS and RFS were similar between eras. In B ALL, OS (HR 2.03, p=0.004) and RFS (HR=2.17, p=0.0004) were improved in ERA2 due to decreased CIR (HR 2.65, p=0.0003). These effects persisted when B ALL patients with MRD were removed from the analysis. B ALL patients transplanted after salvage had improved OS (HR 2.69, p=0.02), RFS (HR 1.97, p=0.09), and CIR (HR 2.22, p=0.09) in ERA2. Similarly, B ALL patients transplanted in CR1 had improved OS (HR 2.04, p=0.02), RFS (HR 2.56, p=0.0004), and CIR (HR 3.56, p=0.0003) in ERA2. Among Ph+ B ALL patients in CR1, OS (HR 3.36, p=0.01); RFS (HR 3.71, p=0.001); and CIR (HR 6.42, p=0.003) were improved in ERA2. Regardless of era, Ph+ ALL patients transplanted in CR1 after IM at diagnosis had a 5-year CIR of 39% compared to 12% after a 2^nd or 3^rd generation TKI (HR 3.70, p=0.005), leading to a 5-year RFS of 42% after IM compared to 78% in the latter group (HR 3.03, p=0.002). For B ALL patients in CR1, 5-year CIR was 10% after receipt of pre-transplant BLIN compared to 27% without BLIN (HR 2.65, p=0.11), leading to a 5-year RFS of 84% with BLIN compared to 60% without it (HR 2.53, p=0.05).

Conclusions: AlloBMT outcomes with PTCy for ALL in ERA2 were improved due to reduced relapse in spite of older age, increased co-morbidities, and less intensive conditioning. Improved outcomes were restricted to B ALL and may be driven by changes in pre-transplant therapy. With dramatic improvements in non-transplant outcomes, the selection of ALL patients for alloBMT is an important research question.