Background: Hematopoietic Cell Transplantation – Comorbidity Index (HCT-CI) is possibly the most commonly used prognostic score to predict non-relapse mortality (NRM) in the setting of allogeneic hematopoietic cell transplantation (alloHCT). Traditionally, candidates to alloHCT with an HCT-CI >3 are discarded from transplant in consideration of a predicted 2-year NRM and overall survival (OS) of approximately 40%. However, its accuracy is not optimal, and its role in the reduced-intensity Conditioning (RIC) setting and post-transplant cyclophosphamide (PTCy) should be tested. In our study, we hypothesized that patients transplanted with an HCT-CI>3 in the current era would not have such dismal outcomes in terms of NRM and OS.

Methods: We performed a monocentric, retrospective study using clinical data from the Institut Català d'Oncologia (Barcelona, Spain). We included adult patients (> 18 years old) with oncohematological diseases, receiving a first alloHCT, using only RIC and peripheral blood stem cells as graft source, and a HCT-CI >3. The inclusion period was from January 2019 to June 2024. The primary endpoint was NRM. Secondary endpoints were progression-free survival (PFS), OS, relapse/progression, acute grade 2-4 graft-versus-host-disease (GVHD), chronic GVHD (all grades), neutrophil and platelets engraftment. Probabilities of OS and PFS were calculated using Kaplan-Meier estimator method, and NRM, relapse/progression, GVHD, and neutrophil/platelets engraftment as cumulative incidences. A univariate and multivariate analysis was performed to describe associations between main clinical variables, including HCT-CI, and survival outcomes.

Results: A total of 58 consecutive patients were included. Median age at transplant was 56 years (range 27-71 years). Female patients represented 46% of the population; chemosensitive disease (complete or partial response) at transplant was reached in 83% of cases; mean HCT-CI was 4 (range 3-7); donor type was matched-related in 28%, matched-unrelated in 24%, mismatched unrelated in 2% and haploidentical in 46%; median donor age was 37 years (range 21-74 years); sex mismatch (female donor to male recipient) was present in 14%; PTCy was used in 79% of cases. Acute myeloid leukemia represented 48% of patients followed by Hodgkin and non-Hodgkin lymphomas (24%), myelodysplastic syndromes (17%), and acute lymphoblastic leukemia (11%). Regarding the type of comorbidities, the most prevalent were pulmonary (89%) followed by diabetes (19%), hepatic (12%), infectious (10%), cerebrovascular (7%) and ischemic cardiac disease (7%), arrhythmia (5%), psychiatric 3%, obesity (2%). Median follow-up among survivors was 40 months (range 11-76 months). At +24 months, NRM was 22% (95%CI: 12-32%), OS was 58% (95%CI:44-69%), PFS was 55% (95%Ci: 41-67%),; relapse/progression was 23% (95%CI:13.33%), acute grade 2-4 graft-versus-host-disease (aGVHD) was 48% (95%CI:36%-60%), all grade chronic GVHD was 19% (95%CI:9-29%). At day +30 neutrophil and platelets engraftment were 90% (95%CI:81-99%) and 69% (95%CI: 59-79%), respectively.

On univariate analysis, an HCT-CI >4 (selected to stratify our cohort, being 4 the mean of HCT-CI score of the study population) was associated with decreased PFS (41% versus 66%, p=0.03), OS (40% versus 72%, p<0.01). When performing the multivariate analysis, such results were confirmed for PFS (HR=4.16, 95%CI:1.79-9.75, p<0.01) and OS (HR=6.79, 95%CI:2.61-17.7, p<0.01). Also, a worse NRM was observed (HR=5.86, 95%CI:2.02-17.01). However, the accuracy prediction of HCT-CI>4 was considered low with an area under the curve between 0.58 and 0.62.

Conclusions: Patients with a HCT-CI> 3 currently have a significantly improved NRM compared to the past era, especially when RIC Conditioning and PTCy are used. An HCT-CI cutoff of 4 could be a better predictor of higher NRM or OS. However, its accuracy is still low. A higher HCT-CI should not be solely used to exclude patients without a specific evaluation of comorbidities by a specialist and a comprehensive evaluation of the patient, including frailty.

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