A NOVEL classification model for multiple myeloma (MM) may provide personalised predictions of treatment outcomes for individual patients. This research has the potential to significantly improve current management of MM. Current prognostic tools, including fluorescence in situ hybridisation, “do not account for patient-to-patient variability, and ignore several prognostically relevant genomic and time-dependent features that could help physicians determine the best treatment strategy to boost overall patient survival,” according to study co-author, Ken Shain, Pentecost Family Myeloma Research Center, Tampa, Florida, USA.
Aiming to improve the specificity of MM treatment, Shain and team assembled a clinical, demographic, genomic, and therapeutic dataset from almost 2,000 newly diagnosed patients with MM. Extensive genomic profiling was then performed on the dataset using advanced sequencing techniques. This allowed the research team to identify recurrent genomic features and mutations associated with this rare form of blood cancer, which affects different individuals drastically differently.
The result of this research was the Individualized Risk Model in Multiple Myeloma (IRMMa). This predictive model uses deep neural networks, among other statistical methodologies, to generate individualised overall survival and event-free survival predictions for patients with MM. Rigorous validation of the IRMMa model using independent datasets was performed in order to ensure its accuracy and reliability in diverse patient populations. The research team also analysed treatment outcomes across therapeutic approaches, with the aim of identifying optimal treatment approaches tailored to individual patient needs.
Though IRMMa has limitations, it provides a useful model for personalising MM treatment. “To our knowledge this is the first individualised prediction model able to incorporate heterogeneous clinical and genomic information, to predict an individual patient’s response to given treatment options,” commented Shain. This initial technology can therefore be adapted over time to add new genomic drivers and therapies, eventually improving the approach to management of MM, for which there is currently no cure.
