Efficacy of Adoptive Immunotherapy By Donor Lymphocyte Infusions after Allogeneic Stem Cell Transplantation from Related an Unrelated Donors: A Single Center Retrospective Analysis

Introduction: Allogeneic stem cell transplantation (alloSCT) has become an integral part in the therapy of patients with malignancies of the lympho-hematopoietic system. One of the main reasons for treatment failure after alloSCT is relapse of the underlying disease, which, in the majority of cases, is associated with a poor prognosis. Adoptive immunotherapy by the use of donor lymphocyte infusions (DLI) was shown to be effective in this setting. However, the conditions and the optimal timing of DLI administration for prophylaxis or treatment of (impending) relapse remains controversial.

Patients and Methods: We retrospectively analyzed 160 consecutive patients (median age: 48 (range: 17-69) years) who received DLI after previous alloSCT performed at our center between 1998 and 2014. Indications for alloSCT were: acute myeloid leukemia (AML) (N=68), acute lymphoblastic leukemia (ALL) (N=49), myelodysplastic syndrome/myeloproliferative neoplasia (MDS/MPN) (N=26), or myeloma/lymphoma (N=17). The disease risk index (DRI) was low (N=1), intermediate (N=101), high (N=43), or very high (N=6) (unknown: N=9). Comorbidities, as specified by the hematopoietic cell transplantation-specific comorbidity index (HCT-CI), were low (N=38), intermediate (N=79), or high (N=38) (unknown: N=5). In N=71 patients a 10/10 human leukocyte antigen (HLA) matched-related donor was chosen, whereas N=89 patients were transplanted from an unrelated donor, either matched (N=73) or mismatched (N=16). Conditioning was either myeloablative (MAC) (N=71) or reduced-intensity (RIC) (N=89). The median interval from alloSCT to first DLI was 7.1 (range: 1.0-93.2) months. Indication for DLI was prophylactic (e.g. high-risk of relapse or active disease at the time of alloSCT) (N=28), pre-emptive (e.g. persistent or increasing mixed chimerism/molecular relapse) (N=86), or hematologic relapse (N=46). Pre-treatment before DLI was none/cessation of immunosuppression (N=129), lymphodepleting chemotherapy (N=16), or other (N=15). The median number of DLI units given was 2 (range: 1 - 6) and the median cumulative CD3+ cell dose/kg body weight given was 1.1 x 10E7 (range: 5.0 - 16.0 x 10E7).

Results: The median follow-up of all patients from day of alloSCT was 37.3 (range: 3.0 - 202.6) months, whereas the median follow-up from day of first DLI administration was 21.2 (range: 0.3 - 200.5) months. Overall survival (OS) of the entire cohort at 1, 3, and 5 years after alloSCT was 80.5%, 63.8%, and 57.7%. Calculated from the day of first DLI OS at the same time points was 68.8%, 61.0%, and 55.8%. At five years after alloSCT OS in the group of patients with AML or ALL was significantly lower as compared to patients with MDS/MPN or myeloma/lymphoma, i.e. 52.0% versus 66.2% (p=0.043). Furthermore, OS in the group of patients receiving pre-emptive DLI was virtually identical to patients who received prophylactic DLI, i.e. 70.4% versus 69.8% at 5 years. In contrast, patients with hematologic relapse prior to DLI had an inferior outcome, i.e. an OS of 23.3% at 5 years. In addition to indication for DLI administration, i.e. prophylactic or pre-emptive versus therapeutic, the occurrence of chronic graft-versus-host disease (cGvHD) was the strongest predictor for outcome, i.e. long-term survival.

Conclusions: Taken together, our data indicate that adoptive immunotherapy by the use of DLI is capable of inducing long-term remissions in patients after alloSCT. As pre-emptive and prophylactic treatment yielded virtually identical results, latter may be reserved for selected patients with (very) unfavorable disease characteristics, e.g. AML or ALL with active disease at the time of transplant. The optimal type of pre-treatment needs to be determined by investigating larger patient cohorts.