In a groundbreaking collaboration, researchers at the Sylvester Comprehensive Cancer Center at the University of Miami Miller School of Medicine have developed the first computational model that predicts an individual’s personalized prognosis for newly diagnosed multiple myeloma.
Multiple myeloma is a cancer of blood cells in the bone marrow characterized by the abnormal proliferation of plasma cells, leading to bone damage, immunodeficiency, and the production of abnormal proteins. Depending on the cancer stage, only 40 to 80 percent of individuals diagnosed with the disease will survive their myeloma for five years or more after their diagnosis, revealing the need for a fast and precise classification of the condition as well as personalized treatment options.
Reporting in The Journal of Oncology, researchers have now created an innovative tool, known as the Prediction Model for Individualized Risk in Newly Diagnosed Multiple Myeloma (IRMMa), which takes into account the unique biology of patients’ tumors and treatment regimens, marking a significant leap forward in myeloma precision medicine.
Ola Landgren, Chief of the Division of Myeloma and Director of the Sylvester Myeloma Institute, emphasized the importance of considering the highly variable nature of multiple myeloma. According to the researchers, the IRMMa not only improves on existing prognostic tools but also identifies 12 distinct subtypes of the disease, providing a level of classification previously unseen in multiple myeloma.
“The future of the field will have to be focused on precision medicine. There’s no other way forward,” said Landgren, highlighting the necessity of tailoring treatments to the individual characteristics of each patient’s cancer.
Unlike traditional classification methods, which focused on the amount of cancer present, IRMMa leverages advanced machine learning to consider the specific genetic mutations in the tumor genome. This approach recognizes that the nature of cancerous cells and driver mutations significantly influences treatment outcomes. With the advent of new therapies, especially immunotherapies, the focus has shifted from the quantity to the quality of cancer cells.
Francesco Maura, assistant professor at Sylvester and the first author on the study, emphasized the need for precise prediction tools in the face of expanding treatment options for multiple myeloma. Existing prognostic tools often rely on population averages, categorizing patients into broad risk groups. IRMMa, however, focuses on predicting the risk of individual patients, offering a more tailored and accurate prognosis.
Built using deep learning, IRMMa is designed to learn and improve as it receives more data. This flexibility allows the model to adapt to emerging datasets with future treatment strategies, a crucial feature given the rapidly evolving landscape of cancer therapies. Currently accessible online, the model is geared towards researchers, providing a valuable resource for interpreting and designing new clinical trials.
The research team, in collaboration with Memorial Sloan Kettering Cancer Center, NYU Langone Health, Moffitt Cancer Center, and Heidelberg University Hospital in Germany, used genetic, treatment, and clinical data from nearly 2,000 patients to develop IRMMa. By identifying 90 “driver genes” responsible for tumor growth, the researchers matched treatment outcomes to individual patients’ genetic sequences, creating a robust and comprehensive model.
The researchers believe that as the field progresses, and with the increasing affordability of whole-genome sequencing, tools like IRMMa could become standard practice in optimizing treatment and management strategies for multiple myeloma patients.
