Impact of Consolidative Cellular Therapy with DLI or Second Transplant in Patients with AML Who Achieve Complete Remission Following Post-Transplant Relapse

Introduction:

Relapse of acute myeloid leukemia (AML) after allogeneic hematopoietic cell transplantation (HCT) portends a dismal prognosis. Donor lymphocyte infusions (DLI) and 2^nd HCT (consolidative cellular therapy [CT]) have been deployed in relapsed leukemias. However, the impact of these interventions in post-HCT relapsed AML is not clear, particularly in those patients achieving morphologic CR (mCR). There are no randomized studies to guide which patients benefit from consolidative CT, thus we evaluated the outcome of these interventions for post-HCT AML relapse at our institution.

Methods:

All patients who underwent HCT for AML between 2010 and 2022 at Dana Farber Cancer Institute and experienced morphologic relapse post-HCT were included in this consecutive series (N = 464). 110 patients received DLI, 30 underwent 2^nd HCT, and 324 received no post-relapse CT.

Results:

Transplant characteristics were similar between groups, except that patients receiving 2^nd HCT were younger than those receiving DLI or no subsequent CT (median recipient age at 1^st HCT 46.7, 57.3, and 61.4 years, respectively). For the entire cohort, 2-year overall survival (OS) was 16% (95% confidence interval [CI]: 13% – 19%), and median survival was 3.8 months (95% CI: 3.2 – 5.2). In a multivariable Cox model for OS, higher HCT-CI (≥3 vs. 0-2, Hazard Ratio [HR] 1.27, 95% CI: 1.03 – 1.58, p=0.03), being on steroid immunosuppression at relapse (HR 2.31, 95% CI: 1.54 – 3.47, p <0.0001), and shorter time to relapse (≤3mo vs. >12mo, HR 4.19, 95% CI: 2.49 – 7.04, p <0.0001) were associated with inferior OS. In contrast, DLI treated as a time-dependent variable vs. no consolidative CT was associated with improved OS (HR 0.69, 95% CI: 0.5 – 0.94, p=0.02). 2^nd HCT had the same HR (0.69) but did not reach statistical significance, p=0.18. In a landmark analysis at 4 months (selected as median time to receive CT to account for individuals who did not survive long enough to receive CT), 2-year OS was 47% (95 CI: 34% – 60%), 53% (95% CI: 23% – 75%), and 25% (95% CI: 18% – 32%) for patients receiving DLI, 2^nd HCT, or no consolidative CT, respectively (p=0.0055).

Overall, 137/464 (30%) patients achieved mCR (62% [68/110] prior to DLI, 93% [28/30] prior to 2^nd HCT, and 13% [41/324] of patients who did not receive consolidative CT). mCR was achieved with intensive chemotherapy in 51/137 (37%) and with non-intensive therapy in 86/137 (63%). Median time from mCR to consolidative CT was 1.38 months for DLI and 1.71 months for 2^nd HCT. Among patients receiving DLI, 2-year OS and median survival from DLI were 36% (95% CI: 27% – 45%) and 15.1 months (95% CI: 8 – 19). mCR prior to DLI was associated with survival in a multivariable model (HR 0.48, 95% CI: 0.26 – 0.9, p=0.023), with 2-year OS from time of DLI of 45% for those in mCR (95% CI: 33% – 57%) vs. 21% (95% CI: 10% – 35%) for those not in mCR, p = 0.0019. No other baseline characteristics were associated with improved survival post-DLI.

For the entire cohort of patients in mCR, 2-year OS from mCR was 40% (95% CI: 32% – 49%). In a multivariable Cox model for survival from mCR treating DLI and 2^nd HCT as time dependent variables, neither DLI (HR 1.06, 95% CI: 0.6 – 1.86, p=0.85) nor 2^nd HCT (HR 1.13, 95% CI: 0.55 – 2.31, p=0.75) was associated with OS.

Conclusions:

There is no clear standard of care for patients with relapsed AML after HCT. Selection of therapy depends on patient fitness, genomic profile, donor availability, and the status of co-morbidities including GVHD. Retrospective analyses comparing these approaches risk confounding by selection and guaranteed-time bias – however, in our large single-institution cohort, DLI was associated with improved survival (and 2^nd HCT with similar HR) when treated as time-dependent variables in the entire cohort and in a landmark analysis. In contrast, multivariable models did not find improved long-term survival of DLI/2^nd HCT compared to no CT specifically in the subset of patients who achieved mCR, though this analysis is limited by small sample sizes. These data raise questions regarding which patients derive most benefit from consolidative CT. Larger retrospective studies and prospective trials may more clearly establish the role of DLI or 2^nd HCT for those in mCR. Future work will include assessing whether the type of therapy used to achieve mCR, genomic subgroups, or patients who do not achieve mCR differentially benefit from consolidative cellular therapies.