Background: Allogeneic hematopoietic cell transplantation (HCT) is the only curative option for most adults with acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS). Although older age increases the risk of non-relapse mortality (NRM), reduced-intensity conditioning (RIC) has made HCT more accessible to elderly patients. Age adjusted HCT comorbidity index (HCT-CI), developed from a heterogeneous cohort, evaluates physiological reserve based on age and comorbidities. However, changes in practice and variable HCT-CI performance limits its predictive value in this growing older population. We present additional predictors of NRM and overall survival (OS) and introduce Refined Elderly Comorbidity Index (RECI) for patients ≥60 years (yrs) receiving HCT for AML/MDS.

Methods: In this single-center retrospective study, we evaluated predictive markers and survival outcomes in patients aged ≥60 yrs who underwent HCT (2005-2023) for AML or MDS. Combining European LeukemiaNet (ELN) for AML and the Revised International Prognostic Scoring System (R-IPSS) for MDS, we risk stratified patients as high risk (adverse risk ELN; very high- or high-risk R-IPSS) and low risk (favorable or intermediate risk ELN; intermediate or low or very-low risk R-IPSS). The primary endpoints were 2-year NRM and OS. Adjusted hazard ratios (HR) of significant variables from multivariate Cox model of NRM were converted to an integer score to develop the RECI. HR of 1.3-2, 2.1-3, and 3-4, were assigned weights of 1, 2, and 3, respectively. RECI, the sum of integer weights, was used to stratify patients according to 2-year NRM. Prognostic significance of RECI was also determined for OS.

Results: Total of 941 (530 AML; 411 MDS) patients were included: 672 (60-69 years) and 269 (>70 years). Median follow up for the entire cohort was 24 months (range: 7.53-24). At time of HCT, median age was 67 yrs (60-78),61% were male, 62% recipients were CMV seropositive (CMV R+), 29% had KPS <80, 61% had HCT-CI >3, 12% had eGFR <60 mL/min per 1.73 m2, 9.8% had albumin <3.5 g/dL, 69% had ferritin >700 ng/mL, 50% had high risk AML/MDS, 70% received matched unrelated donor HCT, 89% received RIC and 37% post-HCT cyclophosphamide.

At 2-year post-HCT, NRM was 22% (95% CI: 20-25) and OS was 59% (95% CI: 56-62) for the entire cohort. In multivariate analysis, NRM and OS were similar in patients aged 60-69 and >70 years. NRM was higher with HCT-CI >3 (HR 1.61, 95% CI 1.19- 2.18, p=0.002), eGFR <60 (HR 1.83, 95% CI 1.26- 2.67, p=0.002), albumin <3.5 (HR 1.83, 95% CI 1.25- 2.65, p=0.002), and CMV R+ (HR 1.36, 95% CI 1.01- 1.83, p=0.04). OS was worse with HCT-CI >3 (HR 1.46, 95% CI 1.15- 1.86, p=0.0018), albumin <3.5 (HR 1.54, 95% CI 1.1- 2.16, p=0.01), ferritin >700 (HR 1.43, 95% CI 1.10- 1.87, p=0.007), and high risk AML/MDS (HR 1.58, 95% CI 1.25- 1.99, p= <0.0001).

Components of RECI and their weights included HCT-CI >3 (1), eGFR <60 (1), albumin <3.5 g/dL (1), and CMV R+ (1). RECI ranged from 0 to 4 and stratified patients into three risk groups with score of 0 (33%), 1 (53%), and >2 (14%). RECI was identified as a strong prognostic factor with of 2-year NRM ranging from 15% (95% CI: 11-19) in RECI 0, 22% (95% CI: 19-26) in RECI 1 (HR=1.6, 95% CI 1.13-2.25, p=0.009) and 38% (95% CI: 29-46) in RECI >2 (HR=3.0, 95% CI 1.99-4.46, p<0.001). The prognostic significance of RECI was also determined for OS ranging from 66% (95% CI: 61-72) in RICI 0, 58% (95% CI: 54-63) in RECI 1 (HR=1.3, 95% CI 1.04-1.68, p=0.02) and 44% (95% CI: 36-53) in RECI >2 (HR=2.2, 95% CI 1.60-2.91, p<0.0001).

Conclusions: RECI applied at time of HCT strongly predicts NRM and OS of patients aged >60 receiving HCT for AML/MDS. No difference in survival outcomes was observed in patients aged 60-69 and >70 yrs. While elderly patients should be considered candidates for HCT, predictive factors that affect NRM and OS should be carefully assessed prior to HCT. If validated in large independent HCT cohort, RECI as a novel prognostic tool can aid in the appropriate selection of elderly patients for HCT.

This content is only available as a PDF.
2025
Sign in via your Institution