The EASIX scoring system should not be used to predict outcomes following allogeneic stem cell transplantation (alloSCT) in patients with myelofibrosis.
A new study has found that the Hematopoietic Cell Transplantation Comorbidity Index (HCT-CI) and the augmented version of the index were both useful predictors of overall survival (OS) and nonrelapse mortality (NRM) in people with myelofibrosis (MF) undergoing allogeneic stem cell transplantation (alloSCT). However, the same was not true for the Endothelial Activation and Stress Index (EASIX).
These findings could help clinicians better assess mortality risk among people with MF) who undergo alloSCT, wrote the study authors in Transplantation and Cellular Therapy.
They noted that alloSCT is the only curative option for patients with MF, but that there remains no clear way to predict who is most likely to achieve long-term survival following therapy. “Given the rarity of MF, the identification of predictors of post-alloSCT survival in patients with this disease has remained an area of highly unmet need,” they said.
The HCT-CI and EASIX scoring systems were developed to assess the likelihood of NRM in patients undergoing stem cell transplantation. In addition, an augmented version of HCT-CI (aHCT-CI) was developed and incorporates albumin, ferritin, and platelet counts.
“However, patients with myeloproliferative neoplasms were significantly underrepresented in the studies validating these scoring systems,” the investigators said. In particular, they noted that the seminal study evaluating the HCT-CI system did not include any patients with a reported myeloproliferative neoplasm, so they decided to evaluate the scoring systems by retrospectively scoring a cohort of patients with MF treated at Mayo Clinic and then comparing their scores to the patients’ real-world outcomes.
The team identified 87 patients with MF who underwent alloSCT at the hospital between 2012 and 2020. Patients were scored based on electronic medical record data and, in the case of EASIX scores, based on log2-transformed values calculated prior to the start of conditioning therapy.
The authors found meaningful connections between HCT-CI scores and patient outcomes. After a median follow-up of 5 years post transplantation, patients in the high HCT-CI category had a significantly higher rate of NRM at 3 years (35.5% vs 11.6%; P = .011). Similarly, patients in the high or very high aHCT-CI categories had higher 3-year NRM than those in the low- and intermediate-risk categories (31.9% vs 6.52%; P = .004).
Yet, EASIX scores did not have the same predictive quality. When divided into quartiles, the investigators found no significant correlation between score and 3-year NRM.
In terms of OS, 3-year survival was significantly worse in patients with high HCT-CI (HR, 4.41; 95% CI, 1.97-9.87; P < .001) or a high or very-high aHCT-CI score (HR, 3.99; 95% CI, 1.56-10.22; P = .004). Again, though, EASIX scores showed no such relationship.
“Most importantly, our study demonstrates that the EASIX score was not able to predict posttransplantation outcomes in patients with MF,” the authors wrote. They said the scores should therefore not be used in a clinical setting to predict patient outcomes.
They also made a larger point about the usefulness of such scoring systems. They said it is important that clinicians understand the makeup of the cohorts of patients used to validate such systems.
“Our study also stresses the idea that although the ‘validated’ scoring systems can predict outcomes in a majority of patients, studies are needed to evaluate their predictive ability in selected malignancies,” they said.
Acosta-Medina AA, Baranwal A, Johnson IM, et al. Comparison of pretransplantation prediction models for nonrelapse mortality in patients with myelofibrosis undergoing allogeneic stem cell transplantation. Transplant Cell Ther. Published February 9, 2023. doi:10.1016/j.jtct.2023.02.002