AUTHORS
Yael Morgenstern, MD, PhD, Dawn Maze Sr, MD2, Karen W.L. Yee Sr, MD FRCPC3, Aniket Bankar, MD, MSc, DM, MBBS4, Marta Davidson, PhD MD5, Guillaume Richard-Carpentier Sr, MD6, Steven Chan, MD PhD7, Aaron Schimmer, MD PhD8, Andre C. Schuh Sr, BSc, MD, FRCPC9, Mark Minden Sr, MD PhD10, Vikas Gupta, MD FRCP, FRCPath11, Mark Gower, PhD12, Anastasia N. Tikhonova, PhD13, Hassan Sibai, MD, Msc, BSc14, Maria Perusini, MD, Msc15
AFFILIATIONS
1 Princess Margaret Cancer Center, UHN, Toronto, Ontario, Canada
2 Princess Margaret Cancer Center, UHN, Toronto, Ontario, Canada
3 Princess Margaret Cancer Center, UHN, Toronto, Ontario, Canada
4 Princess Margaret Cancer Center, UHN, Toronto, Ontario, Canada
5 Princess Margaret Cancer Center, UHN, Toronto, Ontario, Canada
6 Princess Margaret Cancer Center, UHN, Toronto, Ontario, Canada
7 Princess Margaret Cancer Center, UHN, Toronto, Ontario, Canada
8 Ontario Institute of Cancer Research, Toronto, Ontario, Canada, Princess Margaret Cancer Center, UHN, Toronto, Ontario, Canada
9 Princess Margaret Cancer Center, UHN, Toronto, Ontario, Canada
10 Princess Margaret Cancer Center, UHN, Toronto, Ontario, Canada
11 Princess Margaret Cancer Center, UHN, Toronto, Ontario, Canada
12 Princess Margaret Cancer Center, UHN, Toronto, Ontario, Canada
13 Princess Margaret Cancer Center, UHN, Toronto, Ontario, Canada
14 Princess Margaret Cancer Center, UHN, Toronto, Ontario, Canada
15 Princess Margaret Cancer Center, UHN, Toronto, Ontario, Canada
ABSTRACT
Introduction
T-cell acute lymphoblastic leukemia (T-ALL) is a biologically heterogeneous malignancy associated with limited therapeutic options beyond intensive chemotherapy. In contrast to B-ALL, effective immunotherapies and molecularly targeted treatments remain largely unavailable for T-ALL, and treatment in the relapsed setting continues to rely predominantly on multi-agent chemotherapy. Patients with suboptimal response represent particularly high-risk groups with an unmet need for treatment strategies.
Aims
To evaluate the prognostic significance among patients requiring re-induction and to characterize clinical outcomes following relapse in adults with T-ALL.
Methods
We performed a retrospective cohort study of adult pts (≥18 years) diagnosed with T-ALL between 2001 and 2024, followed at Princess Margaret Hospital.
Results
Among 184 patients with T-ALL, 159 achieved complete remission (CR) after first induction and 25 required re-induction. Re-induction rates were similar among patients initially treated with hyper-CVAD and those receiving a pediatric-inspired regimen containing peg-asparaginase (15.4% vs 13.1%; p=0.79). Of the 25 patients who required re-induction, 15 (60.0%) achieved CR1, whereas 10 (40.0%) remained refractory. During follow-up, 11/15 (73%) patients who achieved CR1 after re-induction underwent allogeneic hematopoietic stem-cell transplantation (allo-HSCT), including 10 patients in CR1 and 1 patient in CR2. Among these patients who went through allo-HSCT, 6/11 (55%) relapsed post allo-HSCT, with a median time to relapse of 5.6 months (range, 3.8-15.2). Overall survival (OS) was inferior among 15 patients who required re-induction to achieve CR compared with those who achieved CR after first induction with a median OS of 37.5 months in the re-induction group (95% CI, 12.8 months–not reached), compared to median OS not reached among patients achieving CR after first induction (95% CI, 75.3 months–not reached), p<0.001. Nevertheless, outcomes were poorest among patients who remained refractory after re-induction, with a median OS of only 7.5 months (95% CI, 3.6–15.8). Patients requiring re-induction had a significantly higher 3-year cumulative incidence of relapse (CIR) than those achieving CR after first induction (61.9%, 95% CI, 30.0-82.6 vs 22.1%, 95% CI, 15.2-29.8, p<0.001). In multivariable analysis, requirement for re-induction (HR, 4.61; 95% CI, 2.22-9.56; p<0.001) and age >40 years (HR, 3.10; 95% CI, 1.50-6.39; p=0.002) independently predicted relapse, whereas extramedullary disease, central nervous system involvement, and induction regimen were not associated with relapse risk.
Overall, 52 patients (28.3%) presented with relapsed disease. Median duration of CR1 before relapse was 13.1 months (range, 1.0-166.6). Among patients with relapse, 32 (76.2%) patients received salvage therapy. Salvage regimens included hyper-CVAD in 50.0%, FLAG-IDA in 15.6%, a pediatric-inspired ALL induction regimen in 12.5%, and other regimens in 21.9%. Of the 32 patients who received salvage therapy, 15 (46.9%) patients achieved CR2, of whom 12/15 (80%) patients subsequently underwent allo-HSCT. Patients with relapsed disease had poor outcomes with median OS post-relapse of 4.9 months (95% CI, 3.7-7.4), and median 1-year OS 25.6% (95% CI, 13.9-38.9). Post-relapse OS did not differ significantly between patients with early relapse (CR1 duration ≤12 months) and those with late relapse (CR1 duration>12 months). Among 12 patients who had undergone allo-HSCT post-relapse, median OS was 12.4 months from the time of allo-HSCT (95% CI, 1.9 months-not reached) with 3-year OS of 50.0% (95% CI, 20.8%-73.6%).
Conclusions
Requirement for re-induction identified a high-risk subgroup of T-ALL, even among patients who subsequently achieved CR, and was independently associated with high risk of relapse. Outcomes following relapse were poor despite salvage therapy, with fewer than half of treated patients achieving CR2 and a median post-relapse OS of less than 5 months. Notably among patients undergoing allo-HSCT, survival outcomes were substantially improved, suggesting a meaningful benefit of allo-HSCT in patients requiring re-induction and in the relapsed setting. These findings support early identification of patients with suboptimal induction response and other high-risk features and highlight the importance of considering allo-HSCT in the upfront setting for these patients.
DISCLOSURES
Yael Morgenstern, MD, PhD: No relevant conflicts of interest to declare.