Abstract Background With intensified, multiagent chemotherapy protocols, outcomes for childhood T-cell acute lymphoblastic leukemia (T-ALL) have improved with most patients (pts) achieving long term cure. However, identification of prognostic variables that can be used to assign risk-adapted therapy has lagged behind that for B-ALL. Here we present results of the T-ALL cohort enrolled on Dana-Farber Cancer Institute (DFCI) ALL Consortium Protocol 16-001. Methods DFCI 16-001 enrolled children with ALL ages 1 to 21 years from Mar 2017 to Sept 2022. Pts with T-ALL were classified as initial high-risk (HR) and received multiagent induction chemotherapy, including doxorubicin and dexamethasone. Pts with adverse biology (KMT2A-R, low hypodiploidy, IKZF1 deletion) were upstaged to very high risk (VHR); those with Ph+ ALL were removed from protocol during induction. Induction Failure (IF) was defined as M3 marrow at end of the first month of treatment (TP1), or M2/M3 marrow approximately 10 weeks after starting therapy (TP2). For pts achieving complete remission (CR), Final Risk Group was assigned as follows: 1) Final Intermediate Risk (IR): Absence of adverse biology and low TP1 MRD (10-4); 2) Final HR: Absence of adverse biology, high TP1 MRD but low TP2 MRD (10-3); 3) Final VHR: Adverse biology regardless of MRD or high TP2 MRD (10-3). Pts assigned to Final VHR received 2 intensified Consolidation cycles, including nelarabine, high-dose cytarabine, and etoposide. MRD was assessed using next generation sequencing (NGS) by ClonoSEQ (Adaptive, Seattle WA). Early T-cell precursor (ETP) and near-ETP status was centrally determined by flow cytometry, but was not used to assign risk group. Cranial radiation (18 Gy) was administered to pts with CNS3 status. Protocol therapy continued until 24 months from date of CR. Overall survival (OS) and event-free survival (EFS) were estimated with the Kaplan-Meier method and compared between groups with a log-rank test. Results Eighty-six pts with T-ALL enrolled; 77 pts were classified as Initial HR and 9 as Initial VHR based on biology (all with KMT2A-R). Twelve pts were identified as ETP-ALL and 6 as near-ETP. Eight pts were classified as CNS3 at diagnosis. Of the 83 pts evaluable for induction events, 76 (91.6%) achieved CR; 4 HR and 3 VHR pts were classified as having IF and removed from protocol therapy. There were no induction deaths. Of the 76 pts who achieved CR, 33 (38%) were assigned to Final IR, 24 (28%) to Final HR, and 19 (22%) to Final VHR. Of the pts achieving CR, 3 relapses have been observed (2 isolated CNS relapses, 1 marrow-involved relapse). With a median follow up of 4.9 yrs, 4-yr EFS and OS for the T-ALL cohort was 84.8% and 91.5%, respectively. The 4-yr EFS was 96.3% for Final IR/HR pts (Final IR:93.4%; Final HR:100%), and 79% for Final VHR pts (p=0.024). There was no significant difference in outcome by MRD at TP1, however, pts with high TP2 MRD (n=13) had a significantly worse outcome than those who achieved low MRD either at TP1 or TP2 (n=62), 4-yr EFS 69.2% vs 95%, (p=0.0032). Pts with ETP/near-ETP ALL (n=18) had a 4-yr EFS of 66.7% compared with 88.9% for non-ETP ALL (p=0.034), largely due to IF events (n=4); 2 pts with ETP/near-ETP experienced SMN; there were no relapses in this group. The 4-yr OS for ETP/near-ETP was 93.8%, compared with 90.5% for non-ETP. Pts with KMT2A-R (n=9) had an inferior 4-yr EFS of 55.6%. On a univariate Cox model for EFS, significant predictors of adverse outcome were high TP2 MRD (HR 6.36, p=0.035), KMT2A-r (HR 4.07, p=0.018) and ETP or near-ETP status (HR 2.93, p=0.047); age, presenting white blood cell count, CNS status and TP1 MRD were not significant. Conclusions Overall outcomes for pts with T-ALL enrolled on DFCI 16-001 were favorable, and comparable to prior DFCI Consortium trials despite omission of cranial radiation for all but CNS3 pts. Pts with absence of adverse biology who achieved low MRD at either TP1 or TP2 had an excellent 4-yr EFS of 96.3%, comparable to the most favorable B-ALL subtypes, supporting future studies to investigate treatment de-intensification for this pt subset. Pts with ETP/near-ETP experienced high rates of IF, but OS remained comparable to non-ETP T-ALL pts. Novel treatment strategies are needed for VHR T-ALL pts, including those with high TP2 MRD and/or KMT2A-rearrangements.
Burns et al. (Mon,) studied this question.