A First for Multiple Myeloma
A new machine learning model gives a personalized prognosis for patients
By Rachel Tompa, Ph.D.
Illustration by Mariaelena Caputi
F
or those with a new diagnosis of cancer, the future can be dauntingly murky. That’s true not just in a philosophical sense, but in a statistical sense, as well — most methods of predicting patient outcome are based on probabilities and averages, some of them not very precise.
Now, C. Ola Landgren, M.D., Ph.D., and a team of researchers at Sylvester Comprehensive Cancer Center and collaborating institutions have unveiled a computational model that aims to reduce that uncertainty for people newly diagnosed with multiple myeloma. The model is the first to offer a personalized prognosis based on the patient’s tumor genomics and treatments.
The multiple myeloma field desperately needs better prediction tools, said Dr. Landgren, chief of the Division of Myeloma and director of the Sylvester Myeloma Institute. The number of treatments for the disease has dramatically expanded in the past two decades. That’s great news for people who are now living much longer with the disease than in decades past. But with so many options, clinicians need better ways to determine which treatment is going to work best for each patient.
“The future of the field will have to be focused on precision medicine,” Dr. Landgren said. “There’s no other way forward.”
To build the model, the Sylvester researchers and their collaborators used genetic, treatment and clinical data from nearly 2,000 patients newly diagnosed with multiple myeloma. From sequences of the patients’ DNA, the scientists identified 90 “driver genes” — genes bearing mutations in the cancer cells that appear to spur tumor growth. They then looked at the treatments each patient in their dataset received and how the patients fared with those treatments, matching treatment outcome to an individual’s tumor genetic sequences.
The resulting computational model, dubbed the Individual Risk Model for Myeloma, or IRMMa, does a better job predicting outcomes than previous prognostic tools due to its focus on individual tumor genetics paired with treatment outcomes. Because it’s built using machine learning, it can also learn and improve the more data it receives.
IRMMa improves on previous prognostic tools because it takes into account the biology of patients’ tumors, Dr. Landgren explained. It is also flexible — a patient’s prognosis from the model can be changed if, for example, they receive a transplant after a given treatment.
“More and more information will become available, and tools like this model are the future for optimized treatment and management,” Dr. Landgren said.
On April 12, following a presentation by Dr. Landgren and his team, the Food and Drug Administration’s Oncology Drugs Advisory Committee voted 12-0 to recommend adopting a clinical trial endpoint called “minimal residual disease” to replace current requirements for accelerated drug approval. Dr. Landgren called the decision “historic,” saying FDA adoption will speed new therapies to patients with multiple myeloma and other types of cancer. 
