Impact of natural killer cells on outcomes after allogeneic hematopoietic stem cell transplantation: A systematic review and meta-analysis

BACKGROUND: Natural killer (NK) cells play a vital role in early immune reconstitution following allogeneic hematopoietic stem cell transplantation (HSCT). METHODS: A literature search was performed on PubMed, Cochrane, and Clinical trials.gov through April 20, 2022. We included 21 studies reporting data on the impact of NK cells on outcomes after HSCT. Data was extracted following the PRISMA guidelines. Pooled analysis was done using the meta-package (Schwarzer et al.). Proportions with 95% confidence intervals (CI) were computed. RESULTS: We included 1785 patients from 21 studies investigating the impact of NK cell reconstitution post-HSCT (8 studies/1455 patients), stem cell graft NK cell content (4 studies/185 patients), therapeutic NK cell infusions post-HSCT (5 studies/74 patients), and pre-emptive/prophylactic NK cell infusions post-HSCT (4 studies/77 patients). Higher NK cell reconstitution was associated with a better 2-year overall survival (OS) (high: 77%, 95%CI 0.73-0.82 vs low: 55%, 95%CI 0.37-0.72; n=899), however, pooled analysis for relapse rate (RR) or graft versus host disease (GVHD) could not be performed due to insufficient data. Higher graft NK cell content demonstrated a trend towards a better pooled OS (high: 65.2%, 95%CI 0.47-0.81 vs low: 46.5%, 95%CI 0.24-0.70; n=157), lower RR (high: 16.9%, 95%CI 0.10-0.25 vs low: 33%, 95%CI 0.04-0.72; n=157), and lower acute GVHD incidence (high: 27.6%, 95%CI 0.20-0.36 vs low: 49.7%, 95%CI 0.26-0.74; n=157). Therapeutic NK or cytokine-induced killer (CIK) cell infusions for hematologic relapse post-HSCT reported an overall response rate (ORR) and complete response (CR) of 48.9% and 11% with CIK cell infusions and 82.8% and 44.8% with NK cell infusions, respectively. RR, acute GVHD, and chronic GVHD were observed in 55.6% and 51.7%, 34.5% and 20%, and 20.7% and 11.1% of patients with CIK and NK cell infusions, respectively. Pre-emptive donor-derived NK cell infusions to prevent relapse post-HSCT had promising outcomes with 1-year OS of 69%, CR rate of 42%, ORR of 77%, RR of 28%, and acute and chronic GVHD rates of 24.9% and 3.7%, respectively. CONCLUSION: NK cells have a favorable impact on outcomes after HSCT. The optimal use of NK cell infusions post-HSCT may be in a pre-emptive fashion to prevent disease relapse.