MRD at the end of induction and EFS in T-cell lymphoblastic lymphoma: Children’s Oncology Group trial AALL1231

Traditional variables, such as stage or radiologic response to therapy, have failed to correlate with event-free survival (EFS) in recent trials in T-lymphoblastic lymphoma (T-LL).^1-5 AALL1231 was a Children’s Oncology Group (COG) phase 3 clinical trial for newly diagnosed patients with T-cell acute lymphoblastic leukemia (ALL) or T-LL that randomized children and young adults (aged 1-30 years) to a modified COG-augmented Berlin-Frankfurt-Münster backbone to receive standard therapy (arm A) or with addition of bortezomib (arm B) during induction and delayed intensification (1.3 mg/m² × 4 doses per block).⁶ We previously reported the favorable results of patients with T-LL receiving bortezomib.⁶ We now report our analysis of a subgroup of participants with T-LL who voluntarily submitted bone marrow samples at the end of induction (EOI) to assess the correlation of minimal residual disease (MRD) at EOI on EFS and overall survival (OS). Identification of variables that correlate with EFS is essential to develop risk-based therapies. MRD has shown to be a powerful prognostic tool for both B-cell ALL and T-cell ALL.^7,8 Despite these advances in ALL, the relationship of EOI MRD to clinical risks in patients with T-LL is not known.

Newly diagnosed patients with T-LL, stage II to IV, were eligible for enrollment in COG ALL1231 (NCT02112916).⁶ Prior corticosteroid therapy was allowed if the administration was both <5 days within 7 days and <14 days in the 28 days before initiating induction therapy. Patients with T-LL were stratified as standard risk (SR) if they demonstrated <1% malignant cells in the bone marrow at diagnosis (minimally detectable disease [MDD]), had no central nervous system involvement, had no corticosteroid pretreatment, and demonstrated at least a partial response (PR) at the EOI. Intermediate risk (IR) patients had any of the following: corticosteroid pretreatment, >1% MDD, disease detectible in the central nervous system or testes at diagnosis, and still achieved at least a PR at the EOI. Very high-risk patients had any of the features of IR, but achieved no better than stable disease (SD) at the EOI. Bone marrow samples to assess MRD at the EOI were an optional submission for participants with T-LL, and these specimens were analyzed by flow cytometry, having previously demonstrated a validated sensitivity of 0.01% to assess its correlation to EFS.^9,10 

EFS was the primary outcome and defined as time from study enrollment to first event: death in induction or remission, refractory disease, relapse, second malignant neoplasm, or last contact date for those who were event free. OS was defined as time from study enrollment to death or last contact date. Proportions were compared using a χ² test or Fisher exact test. Survival rates were estimated using the Kaplan-Meier method and standard errors.^11,12 Multivariable analyses used Cox regression and included treatment arm and risk group. Per-protocol, subgroup analyses of overall outcomes, including by race, ethnicity, and sex, were performed. P < .05 was considered statistically significant for comparisons. Analyses were performed using SAS, version 9.4 (SAS Institute, Cary NC).

This study was conducted by COG under a National Cancer Institute–held Investigational New Drug application for bortezomib (NSC number 68129; Investigational New Drug number 58443). AALL1231 was approved by the Cancer Therapy and Evaluation Program, the Pediatric Central institutional review board, and participating center institutional review boards. Written informed consent and assent (if applicable) were obtained before study entry.

AALL1231 accrued 209 patients with T-LL from 2014 to 2017 (supplemental Figure 1, available on the Blood website). At the EOI, 43.6% of patients were in radiologic remission, 55.4% had a PR, and 1% had SD or no response. There were 86 patients (41%) for whom EOI samples for MRD assessment were submitted. Demographic characteristics in this subgroup did not significantly differ from the cohort with T-LL (Table 1). There were differences observed in percentage blasts observed in the bone marrow at diagnoses (P < .0001) and stage (P = .0004), although stage was unknown for 55.7% of the patients who submitted a sample for MRD assessment. There was a history of corticosteroid pretreatment in 25.6% of patients; 62.2% had <1% of MDD in the bone marrow at diagnosis, resulting in 30.2% and 67.5% of patients assigned SR and IR, respectively. Those who participated in the MRD assessment had a higher representation of IR patients than those patients who did not (67.5% vs 38.2%; P = .0003). There were no very high-risk patients in this cohort, and 2.3% of patients could not be classified in a specific risk group. Complete response rate was 51.6%, 48.4% had a PR, and none had SD (Table 1).

There was a significant difference in the 4-year EFS comparing arm A with arm B (78% + 8.1% vs 91.2% + 4.9%; P = .046). In addition, a significant difference was also observed in 4-year OS with those patients not receiving bortezomib (arm A, 78.8% + 8.1%) compared with those receiving bortezomib (arm B, 93.3% + 4.3%; P = .023), consistent with previously published results (Table 1). When examining MRD, there were 8 events in patients with MRD <0.1% (4 relapsed, 3 remission deaths, and 1 patient with progression) and 4 events in patients with MRD ≥ 0.1% (3 relapses, 1 remission death). Patients with MRD <0.1% (n = 75) at EOI had a superior EFS vs those with MRD ≥0.1% (n = 11) (89.0% ± 4.4% vs 63.6% ± 17.2%; P = .025). Analysis of the cohort above and below 0.01% failed to distinguish significant differences, possibly due to the small sample size (71 <0.01% vs 15 ≥0.01%). Furthermore, when examining the 4 patients with MRD <0.1 and >0.01, they are all free of disease. OS did not significantly differ between the 2 groups (88.9% ± 4.4% vs 72.7% ± 15.5%; P = .15) (Figure 1). IR and SR patients had similar EFS (arm A: 73.9% ± 7.5% vs 80.4% ± 6.7%; arm B: 87.2% ± 5.8% vs 90.5% ± 4.8%). Cox regression for EFS demonstrated inferior outcomes for those with MRD EOI ≥0.1% (hazard ratio [HR], 3.73; 95% confidence interval [CI], 1.12-12.40; P = .032), which was independent of treatment arm. OS failed to reach statistical significance for patients with MRD EOI ≥0.1% (HR, 2.714; 95% CI, 0.72-10.44; P = .14). Cox regression did not demonstrate a significant impact on EFS comparing arm A with arm B (HR, 0.57; 95% CI, 0.289-1.073; P = .080), or increasing MDD at diagnosis comparing <1% to 1% to 5% (HR, 0.830; 95% CI, 0.255-2.699) or >5% (HR, 2.67; 95% CI, 0.336-21.145; P = .141). Furthermore, MRD EOI ≥ 0.1% compared with EOI < 0.1% did not differ in complete response rates (55% vs 51%) or PR rates (45% vs 49%). In summary, MRD EOI was the only factor significantly associated with EFS.

Thus, in this phase 3 clinical trial, MRD <0.1% in the bone marrow at EOI for T-LL was associated with improved EFS, regardless of treatment arm for both univariate and multivariate analyses. These findings are consistent with a previous report examining MRD at the end of induction.¹³ Race, age, gender, risk group, MDD, and radiologic response to therapy were not prognostic. No chromosomal or molecular characterization of the disease was available. To our knowledge, this is the first report demonstrating that MRD at EOI is an independent risk factor correlating with EFS using a uniform means of assessing MRD. The findings most likely reflect a greater and more rapid reduction of disease burden, perhaps reflecting greater sensitivity to therapy, consistent with results from other pediatric lymphoma and leukemia trials. The study was limited as submission of EOI bone marrow specimens was voluntary and, thus, only 41% of the patients with T-LL enrolled had specimens available for MRD analysis. Larger numbers of patients would have permitted better analysis of MRD levels (0.01%-0.1%) and differences in treatment assignments due to risk stratification. Recent trials have failed to identify clear prognostic variables that would aid in risk stratifying patients for treatment.^6,14 Given the paucity of available prognostic factors in this disease, incorporation of MRD at the EOI in large clinical trials will establish its value in risk stratification for future therapeutic trials to clarify the significance of this variable.

The authors thank the Cancer Therapeutic Evaluation Program for its support of AALL1231, the study's data safety monitoring committee, and the Pediatric Central Institutional Review Board. The authors especially acknowledge all participating Children’s Oncology Group investigators and the subjects and their families for participating in the study, without whom this report would not have been possible. Figures were generated using R version 2.13.1 (http://www.r-project.org).

This study was supported by National Institutes of Health (NIH), National Cancer Institute (NCI) grants U10CA180886 (M.L.L., D.T.T.), U10CA180899 (M.L.L.), U24CA196173 (M.L.L., D.T.T.), R01CA193776 (D.T.T., M.L.H., T.M.H., B.L.W., M.D.), and NIH, Eunice Kennedy Shriver National Institute of Child Health and Human Development grant X01HD100702 (D.T.T., M.L.L., S.P.H., M.D., S.S.W., K.P.D.), the Leukemia & Lymphoma Society (D.T.T.), NIH/NCI grant R03CA256550 (D.T.T., M.L.L., S.P.H., M.D., S.S.W., K.P.D.), St. Baldrick’s Foundation funding (M.L.L., S.P.H.), and the American Lebanese Syrian Association of Charities. M.L.L. is the Aldarra Foundation Endowed Chair, Seattle Children’s Research Institute, Seattle Children’s Hospital. S.P.H. is the Jeffrey E. Perelman Distinguished Chair in Pediatrics at The Children’s Hospital of Philadelphia. E.A.R. is a KiDS of NYU Foundation Professor at NYU Langone Health.

The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Contribution: R.J.H., D.T.T., M.D., Z.C., M.L.H., P.H.-M., R.D.A., J.L.F., B.L.A., K.A., K.P.D., P.J.G., A.L., Y.H.M., S.P., E.S.S., S.L.V., S.S.W., P.A.Z.-M., C.M.B., M.L.L., S.P.H., and E.A.R. all contributed to the design of the study; D.T.T., A.L., S.S.W., and C.W. provided administrative support; D.T.T., M.D., B.L.W., Z.C., M.L.H., S.C., M.O., A.S., T.M.H., and R.R.M. participated in collection and assembly of data; and R.J.H., D.T.T., M.D., B.L.W., Z.C., M.L.H., K.A., K.P.D., J.L.F., P.H.-M., T.M.H., A.I.J., R.R.M., E.S.S., T.S., K.A.S., N.S., M.L.S., S.S.W., C.M.B., E.A.R., and C.E.A. performed data analysis and interpretation. All authors contributed to manuscript writing and final approval of manuscript.

Conflict-of-interest disclosure: S.P.H. received honoraria from Amgen; received consulting fees from Novartis; and owns common stock in Amgen. M.L.L. received consulting fees from MediSix Therapeutics. M.L.H. served on advisory boards for Novartis and Sobi. D.T.T. serves on advisory boards for Amgen, La Roche, Janssen, and Sobi. C.E.A. serves on advisory boards for Sobi and OPNA; and receives research support from Genentech. The remaining authors declare no competing financial interests.

Correspondence: Robert J. Hayashi, Division of Pediatric Hematology/Oncology, Department of Pediatrics, Washington University School of Medicine, 660 S Euclid Ave, St. Louis, MO 63110; email: hayashi_r@wustl.edu.