Adaptive NK Cell Expansion Is Associated with Reduced Relapse of Lymphoid Malignancies after Autologous Hematopoietic Cell Transplant

Introduction: Our group previously reported that latent or reactivated cytomegalovirus (CMV) after hematopoietic cell transplantation (HCT) leads to expansion of a CD56^dimCD57⁺NKG2C⁺ "adaptive" natural killer (NK) cell subset that displays characteristics of immune memory. Here we report on the impact of these adaptive NK cells on outcomes after autologous (auto) HCT in patients with lymphoid malignancies.

Methods: Based on our prior studies in allogeneic transplant, we hypothesized that patients with high absolute numbers of CD56^dimCD57⁺NKG2C⁺ "adaptive" NK cells at day 28 post auto HCT ("high group") would have lower risk of relapse compared to those with low numbers ("low group"). Secondary outcomes included overall survival (OS) at 2 years and incidence of bacterial, fungal and viral infections 3 months post HCT. Patients with CMV reactivation were excluded because of low incidence (5%). Recursive partitioning was used to differentiate a high (≥1.529 adaptive NK cells/µL, n=88) from low (<1.529 cells/µL, n=25) group based on the largest Gray test statistic at day 28, which predicted the risk of relapse. Kaplan-Meier curves were used to estimate the probability of relapse and disease free survival (DFS) through 2-year post-HCT, and the log-rank test was used for comparisons. Fine-Gray and Cox regression was used to examine the independent effect of factors on relapse and DFS, respectively.

Results: Overall, mean age at HCT was 55.8 years, 52% were males and 64% were CMV seropositive. Diagnoses included Hodgkin (13%) and non-Hodgkin lymphoma (36%) and multiple myeloma (51%) [Table 1]. Almost all patients (96%) in the low group vs. 77% in the high group had Karnofsky performance scores (KPS) of 90 or higher at the time of HCT, p=0.04.There were no other differences between the baseline characteristics of the groups. Cumulative incidence of relapse at 2-years was 34% (95% confidence interval [CI] 23-44%) in the high group compared with 70% (95% CI 49-91%) in the low group (p=0.0012). 2-year DFS rates were 63% (95% CI 51-72%) and 30% (95% CI 12-50%), respectively (p=0.0032) [Figure 1]. Overall survival (OS) at 2-years was 87% (95% CI 77-92%) in the high group vs. 84% (95% CI 63-94%) in the low group, p=0.70. Cumulative incidence of bacterial infections (25% vs. 24%, p=0.91), viral infections (2% vs. 0%, p=0.9) or fungal infections at 90 days (4% each, p=0.9) were similar in high and low groups. After adjusting for CMV serostatus, diagnosis, disease status, age, gender, KPS and comorbidity index in multivariate analysis, the high group had significantly lower risk of relapse (hazard ratio [HR] 0.22, 95% CI 0.11-0.46, p<0.0001) and improved DFS (HR 0.28, 95% CI 0.13-0.56, p<0.001) compared with the low group. The protective effect of CD56^dimCD57⁺NKG2C⁺ NK cells on relapse (HR 0.12, 95% CI 0.05-0.30, p<0.0001) and DFS (HR 0.14, 95% CI 0.06-0.33, p<0.0001) was noted only in CMV seropositive patients, but not in seronegative individuals.

Conclusion: We show that patients with lymphoid malignancies with high absolute counts of "adaptive" CD56^dimCD57⁺NKG2C⁺ NK cells at day 28 post auto HCT had significantly superior DFS and lower risk of relapse compared with patients with low absolute "adaptive" NK cell counts. There was no difference in the risk of bacterial, fungal or viral infections or in OS between the groups. These provocative data suggest that interventions to increase numbers of these adaptive NK cells might provide a novel means of enhancing protection against relapse following auto HCT.

View largeDownload slide

View largeDownload slide

Close modal

View largeDownload slide

View largeDownload slide

Close modal