Comparison of Outcomes of Donor Lymphocyte Infusions with or without Lenalidomide in Patients with Hematological Malignancies Post Allogeneic Transplant

Introduction

Outcomes of patients (pts) with relapsed hematological malignancies post allogeneic stem cell transplant (ASCT) are dismal. Donor lymphocyte infusion (DLI) is one of the treatment options; however outcomes with DLI are also poor especially in acute leukemias. Addition of immunomodulatory drugs to DLI may augment graft-versus-leukemia effect and may improve outcomes. Lenalidomide (len) is an immunomodulatory drug and has several effects on the immune system. Addition of len to DLI has been variably practiced at our centre. In this study, we retrospectively compare the outcomes of DLI with or without len.

Methods

All pts who received DLI from January 2010 to January 2020 were included in this retrospective analysis. No immunosuppressant prophylaxis was administered and ongoing immunosuppression (if any) was stopped prior to DLI. DLI was defined as therapeutic if it was given for hematological relapse (with or without cytoreductive chemotherapy); pre-emptive if there was cytogenetic / flow cytometric / molecular relapse or slipping chimerism. DLI was prophylactic if it was given to prevent a relapse in the absence of any of the above features. Len was given on a continuous schedule at a starting dose of 2.5-25 mg/day. Len was started along with 1st or subsequent DLI as per discretion of treating clinician. For the purpose of this study, pts were divided into two groups – those who received DLI alone (no len group) versus those who received DLI with len (len group). In both groups, cytoreductive chemotherapy was given for some pts with hematological relapse at discretion of treating physician. Event free and overall survival were calculated from the date of 1^st DLI. Event was defined as hematological relapse / progression or death. Complete response (CR) was defined as attainment of full donor chimerism in those with chimerism slippage; attainment of cytogenetic/ flow/molecular negativity in those with any of these positive at time of DLI; attainment of morphological remission in those with hematologic relapse. The primary objective was to compare the overall survival (OS) in no len group versus len group. Secondary endpoints were event free survival (EFS), CR rates, grade II-IV acute GVHD, len related toxicities and therapy related mortality. Patient and donor age, gender, diagnosis, type of ASCT, disease risk index (DRI), time to relapse (or slipping chimerism), type of DLI, pre-DLI morphological disease status, and HLA-A*24 or B*40 in pts were evaluated as factors affecting outcomes with or without len. (HLA-A*24 and B*40 were selected because of prior data from our institute showing benefit of len in pts with AML who had these alleles.) Statistical analysis was done using standard methods. Survival was assessed by Kaplan Meier method. All p values were 2 sided; p value of <0.05 was considered significant.

Results

Total 61 pts received DLI. Table 1 shows the baseline characteristics, transplant details, & details of DLI & len. The 2 groups were comparable in all aspects except for younger age in len group. Median OS (Fig 1a) and EFS (Fig 1b) were not different in len vs no len group (11 vs 8 months, p=0.66 and 6 vs 2 months, p=0.42). Len was not associated with improvement in OS in pts with myeloid malignancies (n=47, median OS 8 vs 3 months, p=0.80) or in AML pts (n=25, median OS 9 months vs 2 months, p=0.47). Among pts with HLA-A*24 or B*40 (n=26), there was an improvement in OS (median not reached vs 8 months, 4 year OS 62% vs 32%, p=0.1) (Fig 1c) and EFS (median 9 vs 1 month, 4 year EFS 50% vs 22%, p=0.1) (Fig 1d) with len. In this subgroup, hazard ratio for both OS and EFS was 0.47 (Fig 1e and 1f respectively). 49 pts had measurable disease / slipped chimerism at DLI. Among these, CR rate was not different between the 2 groups (41% vs 40% with len, p=0.9). Median time to CR was similar (31 vs 38 days, p= 0.36). Post DLI acute grade II-IV GVHD developed in 8 (19%) in no len group vs 4 (22%) in len group respectively, p=0.75. Median duration of len was 36 days. In the len group, 17 discontinued len; 8 (44%) due to disease progression, 5 (28%) due to GVHD, 3 (17%) due to cytopenias and 1 due to unrelated cause. Death due to DLI related GVHD was seen in 1 (6%) and 4 (9%) in len and no len group respectively, p=0.6.

Conclusions

There was no difference in outcomes in patients who received DLI with or without len. However, in subset of patients who had HLA*24 or B*40, an improvement in OS and EFS was seen. The relation of above HLA alleles with len needs further exploration.