Researchers have developed a novel computational model for personalized prognosis prediction in patients with newly diagnosed multiple myeloma, according to a new study published by Maura et al in the Journal of Clinical Oncology. Their model for individualized risk in multiple myeloma, or IRMMa, proved significantly more accurate than existing prognostic models for the disease.
Background
Multiple myeloma is known to be a highly variable disease. The original method for classifying multiple myeloma, developed in the 1970s, was based on staging for solid tumors and relied on the amount of cancer present. With the treatments available at the time, the method was fairly accurate; however, with new advancements in multiple myeloma therapeutics such as immunotherapies, the amount of cancer is often less important than the nature of the tumor cells.
Different kinds of driver mutations in the tumor genome may affect the cancer’s growth. Therefore, certain subtypes of multiple myeloma could have a positive outcome if the appropriate treatment is matched to the patient, even after the cancer metastasizes.
The number of treatment options for multiple myeloma has expanded over the past 20 years, significantly improving survival rates. Nonetheless, the number of therapeutic choices has increased the need for a more accurate strategy for predicting which treatment will be most effective.
Although prognostic tools have been updated through the years, the field of multiple myeloma has historically lacked precise prediction. Some tools included tumor genomic features, but recent findings about genomic risk factors had not yet been included in these prediction models and all of the models to date relied on population averages. As a result, patients were categorized into groups such as “standard risk” or “high risk,” and prognoses for the groups didn’t take into account individualized risk and how distinct treatments could modify risk.
Study Methods and Results
In the new study, the researchers built the novel IRMMa model using genetic, treatment, and clinical data from nearly 2,000 patients who were newly diagnosed with multiple myeloma. From sequences of the patients’ DNA, the researchers identified 90 driver genes bearing mutations in the cancer cells that appeared to spur tumor growth. They then examined the treatments each patient received and how the patients responded, matching treatment outcomes to the patients’ individual tumor genetic sequences. Further, the researchers identified 12 distinct subtypes of multiple myeloma, a classification system that had not previously been devised.
The researchers explained that the IRMMa model was designed to improve on previous prognostic tools by factoring in patients’ tumor genomics and treatments. Because it was created using deep learning, the novel model can receive emerging datasets with future treatment strategies—allowing the model to learn and improve its predictive accuracy.
The IRMMa model is also flexible. A patient’s prognosis from the model can be changed if, for example, they receive a transplant after a given treatment. The researchers highlighted that when new therapies become available—as long as there are data from at least a few hundred patients—the model can be updated to incorporate those treatments.
Conclusions
“Our model is based on the idea of predicting the risk of the individual patient rather than that of the group,” explained lead study author Francesco Maura, MD, Assistant Professor at the Sylvester Comprehensive Cancer Center at the University of Miami Miller School of Medicine. “This model can only grow with the help of the research community. The next challenge is to keep feeding it with the right datasets so at a certain point it will be usable for clinical purposes,” he noted.
The researchers plan to integrate additional data sets from patients treated with newer antibody-based multiple myeloma therapies to enhance the model’s capabilities. Although its current iteration is aimed at researchers rather than physicians, the IRMMa model is currently available online. The researchers hope it may be useful in interpreting or designing new clinical trials, for instance, to provide a large set of comparisons to the experimental treatment.
The researchers underscored that the field isn’t quite at the point of sequencing entire tumor genomes for every patient newly diagnosed with multiple myeloma. However, they hope this might come in the near future as whole-genome sequencing becomes more economical.
“The future of the field will have to be focused on precision medicine. There’s no other way forward,” emphasized senior study author C. Ola Landgren, MD, PhD, Chief of the Division of Myeloma and Director of the Myeloma Institute at the Sylvester Comprehensive Cancer Center at the University of Miami Miller School of Medicine. “More and more information will become available, and tools like this model are the future for optimized treatment and management,” he concluded.
Disclosure: For full disclosures of the study authors, visit ascopubs.org.
