Allogeneic haematopoietic stem cell transplantation (HSCT) provides the possibility of cure for patients with haematological malignancies. Nonetheless, the high risks of potentially lethal graft-versus-host disease (GVHD), infections and other adverse outcomes limit the effectiveness of this intervention and are difficult to predict in advance. Now, data from a large-cohort observational study provide prospective evidence of the intestinal microbiota as a predictor of mortality in patients undergoing HSCT.

A total of 1,362 patients were enrolled at four centres; data from patients enrolled at MSKCC (cohort 1) and in other locations (cohort 2) were analysed separately. Baseline stool samples were obtained from patients undergoing HSCT for haematological malignancies (predominantly acute myeloid leukaemia), within 30 days of treatment. Weekly stool samples were obtained over the 3 weeks after HSCT. Each sample was assessed using 16S RNA sequencing.

A more diverse intestinal microbiota after HSCT (defined as above the median inverse Simpson diversity index) was associated with a lower risk of death in cohort 1 (HR 0.71; 95% CI 0.55–0.92) and in cohort 2 (HR 0.49; 95% CI, 0.27–0.90). Comparisons of post-HSCT and pre-HSCT intestinal microbiota diversity revealed a lower risk of death (HR 0.41, 95% CI 0.24–0.71) and transplantation-related death (HR 0.44, 95% CI 0.22–0.87) in patients with higher pre-HSCT diversity when assessed as a continuous variable in cohort 1. These observations were not validated in cohort 2, although survival data were immature at the cutoff used. A risk score was compiled based upon common taxonomic features of the pre-HSCT microbiota of patients in cohort 1, and this score was predictive of risk of death in cohort 2 (HR 1.39, 95% CI 1.02–1.91).

These findings demonstrate that a more diverse intestinal microbiota is associated with a lower risk of mortality following HSCT. Associations between specific taxa and outcomes will require further validation.