Introduction: Several risk scoring systems accurately predict the risk of non-relapse mortality (NRM) in patients undergoing allogeneic hematopoietic stem cell transplantation (HSCT). However, despite their wide use, these scores are currently applied at specific, and often late, time points after HSCT. The present study aimed to evaluate the ability of these scores to predict early NRM risk (within 30 and 100 days) post-HSCT among older adult HSCT recipients.

Methods: This retrospective single-center study included all consecutive patients aged ≥55 years who underwent allogeneic HSCT between 2013 and 2021. Data were retrieved from electronic medical records, encompassing demographics, hematological diagnoses, transplantation details, and information relevant to the following evaluated risk scores: the Disease Risk Index (DRI), the Hematopoietic Stem Cell Transplantation Comorbidity Index (HCT-CI), the European Society for Blood and Marrow Transplantation (EBMT) score, and the Endothelial Activation and Stress Index (EASIX). Correlations between variables (age, gender, conditioning intensity, female donor-to-male recipient status, risk scores) and NRM at 30 and 100 days post-HSCT were assessed. Associations between categorical variables were analyzed with a univariate chi-square test. Multivariate logistic regression analysis was used to examine the independent contribution of each variable to NRM risk. Kaplan-Meier analysis was performed to estimate the probability of overall survival (OS) and Cox proportional hazards regression was applied to assess an association between variables and OS. Statistical significance was set at p <0.05. All analyses were performed using SPSS version 28.0.1.1(14). The study was approved by the institutional review board (0325-21-RMB-D).

Results: This study included 229 patients at a median age of 64 (range 55-76) years. Nearly half (44.5%) were ≥65 years old; 38% were females. Most patients (52.8%) had acute myeloid leukemia (AML) and 21% had myelodysplastic syndrome (MDS). Two-thirds received an HSCT from a matched unrelated donor, with male patients receiving graft from female donors in 16% of cases. Myeloablative conditioning (MAC) was used in 62% of patients. According to the evaluated risk stratification tools, a substantial portion of patients was classified as high-risk for late post-HSCT NRM: 45% by DRI, 34.9% by HCT-CI >2, 85% by EBMT score >3, and 24.9% by EASIX (highest-risk quartile). The overall NRM rate was 14% at 30 days and 23.6% at 100 days. Sepsis was the leading cause of NRM (63%), followed by acute graft-versus-host disease-related complications (18%).

Univariate analysis identified non-MAC and a higher HCT-CI score (>2) as significant risk factors for NRM at both 30 days (p=0.022 and p=0.007, respectively) and 100 days (p=0.01 and p=0.004, respectively). Multivariate analysis confirmed that only the HCT-CI score >2 remained significantly associated with increased NRM risk at 30 days (OR=1.395, 95% CI [1.228, 13.27], p=0.022) and 100 days (OR 1.106, 95% CI [1.279, 7.135], p=0.012). None of other risk scores showed a significant association with early NRM risk at these time points. With a median follow-up of 47 (range 1-117) months for surviving patients, the Cox proportional hazards analysis revealed that reduced conditioning intensity (p=0.020), the female donor-to-male recipient status (p=0.026), and a higher HCT-CI score (score 1-2 category, p=0.026 and score >2 category, p=0.022) were associated with inferior OS. Similar to the NRM analysis, none of other assessed risk scores was found to significantly correlate with OS.

Conclusions: The current study has demonstrated the validity of the HCT-CI score in predicting increased short-term mortality within 30 and 100 days following HSCT. The application of this score could potentially improve decision-making by providing an additional assessment tool of early NRM risk for patients prior to HSCT. The impact of HCT-CI on short-term mortality warrants further assessment and confirmation in larger prospective studies.

Disclosures

No relevant conflicts of interest to declare.

This content is only available as a PDF.
2024
Sign in via your Institution