The validation of simplified comorbidity index for predicting non-relapse mortality after allogeneic haematopoietic cell transplantation

OBJECTIVES: A simplified comorbidity index (SCI) was recently introduced and demonstrated effective risk stratification for non-relapse mortality (NRM) in cohorts from MSKCC and Sheba Medical Center. METHODS: We retrospectively analysed registry data from 1,677 patients with haematologic malignancies or bone marrow failure syndromes who underwent allogeneic HCT with unmanipulated grafts between 2008 and 2019 at Asan Medical Center, Seoul, Korea. RESULTS: In univariate analyses, the 5-year NRM rates by HCT-CI scores (0, 1, 2, 3, ≥4) were 8.3%, 22.9%, 15.1%, 17.2%, and 23.4%, respectively (P<0.001); by SCI scores (0, 1, 2, 3, ≥4), they were 17.9%, 15.5%, 16.0%, 24.2%, and 28.5%, respectively (P<0.001). Notably, SCI scores of 0-2 did not effectively stratify NRM risk. In multivariate analysis, an SCI score of ≥3 was independently associated with increased NRM (hazard ratio [HR], 1.68; P<0.001) and decreased overall survival (OS) (HR, 1.39; P<0.001). An exploratory Revised SCI incorporating clinically anchored threshold adjustments showed enhanced beta coefficients across comorbidity components. CONCLUSIONS: SCI score of ≥3 predicted increased NRM and decreased OS following HCT. The Revised SCI provided a hypothesis-generating basis for future prospective validation in independent cohorts.