Advertisement

Machine Learning Program May Enhance Transplantation Risk Assessment in Patients With Myelofibrosis


Advertisement
Get Permission

A novel machine learning model may outperform standard statistical models in identifying and stratifying transplantation risk among patients with myelofibrosis, according to a recent study published by Hernández-Boluda et al in Blood.

Background

Although there are several therapies available to treat myelofibrosis, allogeneic hematopoietic cell transplantation is currently the only curative option in this patient population.

“The decision to perform a transplant in a patient with myelofibrosis is frequently very complex. Only about 10% of patients with myelofibrosis ultimately receive one,” explained lead study author Juan Carlos Hernández-Boluda, MD, PhD, a hematologist at the Hospital Clínico of Valencia and Lead of the Myeloproliferative Neoplasms Committee within the European Society for Blood and Marrow Transplantation’s (EBMT) Chronic Malignancies Working Party. “Unfortunately, given the lack of risk assessment tools, providers typically have to rely on personal experience rather than well-defined criteria to help make this challenging decision,” he added.

Study Methods and Results

In the study, researchers sought to improve risk assessment among patients eligible for allogeneic hematopoietic cell transplantation. They used the data from 288 EBMT registered centers reporting to the EBMT registry to identify 5,183 adult patients who underwent allogeneic hematopoietic cell transplantation between 2005 and 2020. Among the patients, 3,887 of them were used to train the machine learning model’s algorithm, and 1,296 of them were used to assess and validate the novel model. The median follow-up and overall survival for the training cohort was 58.2 months and 79.4 months, respectively; for the test cohort, the median follow-up and overall survival was a respective 60 months and 73.7 months.

The researchers then used the Kaplan-Meier method to estimate overall survival and progression-free survival. Nonrelapse mortality was defined as the time from the date of transplantation to the date of death (uncensored) or to the date of disease relapse (censored). Independent statisticians used separate methodologies such as Cox regression methods or machine learning techniques to evaluate the factors influencing overall survival within the same data set, aiming to develop prognostic models to stratify the patients into different risk groups of posttransplantation mortality.

The resulting risk classifications were compared and contextualized to assess their clinical relevance. The researchers developed a prognostic model using random survival forests, which achieved higher concordance indices for predicting both overall survival and nonrelapse mortality compared with three alternative machine learning methods. As a result, the random survival forests method was selected as the primary approach for subsequent analyses. The machine learning model produced more reproducible results compared with the other models. Additionally, it outperformed standard models in accuracy and identified a subset of high-risk patients with a 40% risk of mortality within 1 year following allogeneic hematopoietic cell transplantation—25% of the patients with myelofibrosis. The researchers emphasized that the data were helpful to contrast with disease-related risk factors, making it important for patients deciding whether to undergo transplantation.

The researchers translated their random forests survival model into an open-access, web-based calculator for health-care providers to reference. The final tool was found to be capable of predicting overall survival in patients posttransplantation based on 10 key patient characteristics—including patient age, performance status, comorbidity index, hematologic parameters, donor type, conditioning intensity, and type of graft-vs-host disease prophylaxis.

Conclusions

“Although there are many models available to identify patients with high-risk myelofibrosis, we are still lacking tools to determine the risk of transplant for these patients,” suggested Dr. Hernández-Boluda. “Our prognostic tool comprehensively and effectively identifies high-risk patients with myelofibrosis at high risk for mortality after transplantation, enabling better strategic planning and potentially improving outcomes,” he continued.

“[Physicians] can use this calculator to enhance shared decision-making with their patients,” highlighted senior study author Donal McLornan, MRCP, PhD, FRCPath, a consultant in Hematology and Stem Cell Transplantation at the University College London Hospitals National Health Service Foundation Trust, Co-Chair of the EBMT Scientific Council, and Chair of the Chronic Malignancies Working Party. “This is a practical, easy-to-use tool that takes into account data any transplant [physician] will already have on hand,” he noted.

The researchers cautioned that the study had limitations, including its reliance on a patient registry, which lacked data for some variables like the degree of bone marrow fibrosis and the presence of additional somatic mutations at the time of allogeneic hematopoietic cell transplantation. They hope to collate real-world data entered into the application and incorporate other specific disease factors to refine the applicability of the model.

“From a machine learning perspective, this model truly meets what I call the ‘three A’s’ for effectively integrating artificial intelligence into medicine,” underscored co–study author Adrián Mosquera Orgueira, MD, PhD, a hematologist and machine learning expert at the University Hospital of Santiago de Compostela. “First, it is broadly applicable: the variables that make up the model are easily obtainable regardless of the health-care system. Second, it is highly accessible, thanks to a simple computational tool freely available on the [Internet]. Finally, it is clinically actionable: this tool supports multidisciplinary teams in deciding whether to pursue cell therapy or to evaluate other medical options and supports both [physicians] and patients in making more informed decisions,” he concluded.

Disclosure: For full disclosures of the study authors, visit ashpublications.org.

The content in this post has not been reviewed by the American Society of Clinical Oncology, Inc. (ASCO®) and does not necessarily reflect the ideas and opinions of ASCO®.
Advertisement

Advertisement




Advertisement