Myelofibrosis (MF) is a myeloproliferative neoplasm (MPN) with heterogeneous clinical course. Allogeneic hematopoietic cell transplantation remains the only curative therapy, but its morbidity and ...mortality require careful candidate selection. Therefore, accurate disease risk prognostication is critical for treatment decision‐making. We obtained registry data from patients diagnosed with MF in 60 Spanish institutions (N = 1386). These were randomly divided into a training set (80%) and a test set (20%). A machine learning (ML) technique (random forest) was used to model overall survival (OS) and leukemia‐free survival (LFS) in the training set, and the results were validated in the test set. We derived the AIPSS‐MF (Artificial Intelligence Prognostic Scoring System for Myelofibrosis) model, which was based on 8 clinical variables at diagnosis and achieved high accuracy in predicting OS (training set c‐index, 0.750; test set c‐index, 0.744) and LFS (training set c‐index, 0.697; test set c‐index, 0.703). No improvement was obtained with the inclusion of MPN driver mutations in the model. We were unable to adequately assess the potential benefit of including adverse cytogenetics or high‐risk mutations due to the lack of these data in many patients. AIPSS‐MF was superior to the IPSS regardless of MF subtype and age range and outperformed the MYSEC‐PM in patients with secondary MF. In conclusion, we have developed a prediction model based exclusively on clinical variables that provides individualized prognostic estimates in patients with primary and secondary MF. The use of AIPSS‐MF in combination with predictive models that incorporate genetic information may improve disease risk stratification.
...the potential improvement of our ML model's prognostic accuracy by incorporating molecular data on additional somatic mutations could not be adequately evaluated because a significant proportion ...of patients did not have this information available at that time. For OS prediction, the ML model considering VAF proved superior to those based solely on the presence/absence of mutations or the total mutation count per gene (Supporting Information S1: Table ). In parallel, the primary model for LFS prediction was based on the VAF of 20 genes (c-index, 0.702; Supporting Information S1: Table ) and showed slight improvement when mutational data of the CALR gene or U2AF1 Q157 mutation was incorporated. ...although we mitigated the absence of an external dataset by cross-validating our findings, the intrinsic limitations of internal validation loom.