The new Artificial Intelligence Prognostic Scoring System for Myelofibrosis (AIPSS-MF) prognostic score stratifies patients with myelofibrosis (MF) at diagnosis better than standard models, according to research published in Cancer Reports.
Researchers conducted a retrospective observational study to validate the AIPSS-MF in patients with MF who started ruxolitinib treatment compared with the standard prognostic models at diagnosis, including the International Prognostic Score System (IPSS), the Myelofibrosis Secondary to PV and ET-Prognostic Model (MYSEC-PM), and the Response to Ruxolitinib after 6 months (RR6) score after the start of treatment. The discriminative ability of the models was evaluated with the concordance index (C-index).
The study cohort was based on 103 adult patients (60.3% male) with MF who were ineligible for allogeneic stem cell transplantation. The median age was 68.4 years (range, 38-84 years) at diagnosis and 69.4 years (range, 38-83 years) at the start of ruxolitinib treatment.
Continue Reading
The study demonstrated the AIPSS-MF performs better at diagnosis than the IPSS for primary MF (C-index, 0.636 vs 0.596) and MYSEC-PM for secondary MF (C-index, 0.616 vs 0.593). It also showed that RR6 outperformed the AIPSS-MF at predicting an inadequate response by these patients to JAKi therapy during ruxolitinib treatment (C-index, 0.682 vs 0.571); similar results were observed when subdividing the patients by primary and secondary MF.
“The development of individualized prognostic models represents the future,” the study authors noted in their report. “Patient’s specific characteristics (disease’s features, concurrent comorbidities) pave the way to personalized prognosis, ensuring the best management. Machine learning represents a visionary approach to identifying crucial information to frame the patient and the best treatment plan, promising improvements in sight with shocking speed.”
Limitations of the study included small numbers of patients in some of the groups and and no consideration of molecular and genetics variables.
Disclosure: Some study authors declared affiliations with biotech, pharmaceutical, or device companies. Please see the original reference for a full list of authors’ disclosures.
Reference
Duminuco A, Mosquera-Orgueira A, Nardo A, Di Raimondo F, Palumbo GA. AIPSS-MF machine learning prognostic score validation in a cohort of myelofibrosis patients treated with ruxolitinib. Cancer Rep (Hoboken). Published online August 8, 2023; e1881. doi:10.1002/cnr2.1881
