NK cell adoptive transfer in acute myeloid leukemia: a systematic review and meta-analysis

Acute myeloid leukemia (AML) remains associated with high relapse rates and poor long-term survival, particularly in refractory patients or those ineligible for hematopoietic stem cell transplantation (HSCT). Adoptive transfer of natural killer (NK) cells has emerged as a promising immunotherapeutic strategy due to intrinsic cytotoxicity, low risk of graft-versus-host disease (GVHD), immune restoration capacity, and feasibility of allogeneic off-the-shelf manufacturing.

A systematic review of phase 1-2 clinical trials evaluating non-genetically modified NK-cell adoptive transfer was performed using PubMed and EMBASE (2000-2024). Study selection and data extraction were conducted independently by two reviewers in Rayyan^®. Study characteristics, interventions, and outcomes were extracted. Response rate in relapsed/refractory (R/R) AML receiving NK monotherapy was pooled using single-proportion meta-analysis. One-year disease-free survival (DFS) in other settings was pooled by inverse variance using logarithm-transformed rates. Heterogeneity (I²), publication bias, and study quality (NIH before-after tool) were assessed.

Of 790 records identified, 29 studies were included, predominantly single-arm phase I/II trials (one randomized phase II). In R/R AML (11 studies; 217 patients), pooled response rate was 35% (95% CI 29-42; I²=30%), with 20.2% proceeding to HSCT. Higher responses were observed in strategies enhancing NK activation or persistence, including expanded multi-infusion platforms and CIML-NK. In low-/intermediate-risk AML in remission (3 studies; 38 patients), pooled 1-year DFS was 82% (95% CI 56-100; I²=0). In high-risk AML in remission and ineligible for HSCT (5 studies; 59 patients), pooled 1-year DFS was 26% (95% CI 12-55; I²=84). In studies combining NK transfer with haploidentical HSCT (9 studies; 303 patients), pooled 1-year DFS was 40% (95% CI 27-57; I²=94). Toxicities were generally mild, GVHD rates were low (0-8%), and severe events were uncommon. NK-cell persistence was typically short, improving with multiple infusions and CIML approaches.

Adoptive allogeneic NK-cell transfer appears safe and clinically promising, particularly for patients unfit for intensive therapy or HSCT. At present, its use should remain limited to clinical trials. Future studies should define the optimal approach that maximizes clinical activity while maintaining low toxicity and achieving durable persistence of allogeneic NK cells.