About 10% of people living with myositis—a rare inflammatory disease in which the immune system attacks healthy muscle cells—also are at risk for developing cancer. Determining which patients need screening and close follow-up is difficult.

“We screen a lot of people aggressively, and possibly unnecessarily,” says rheumatologist Christopher Mecoli, M.D., M.H.S., director of research operations and physician lead for the Myositis Precision Medicine Center of Excellence. This can lead to potential harms for patients such as radiation exposure, anxiety and false-positive test results that indicate a person has cancer when they actually do not.

Precision Medicine at Work

In a precision medicine project, Mecoli and colleagues are working on a computerized cancer-prediction tool that incorporates features of patients with myositis to better pinpoint which ones should be screened aggressively for cancer versus those who can follow routine cancer screening recommendations.

“We have discovered and validated several novel autoantibodies, proteins that target one’s own body, in patients with myositis and other rheumatic diseases,” Mecoli says. “It turns out these are pretty good risk-stratifying tools. Depending on which ones are present, and how many of them are present, and in what combination they are present, we could predict who is at higher risk for developing cancer.”

To look for associations between these autoantibodies and cancer development, Mecoli’s team worked with scientists at the Johns Hopkins University Applied Physics Laboratory on advanced biostatistical analyses using the Wonder Engine computing platform. Now, a multisite study of the cancer-prediction model is in the planning stage with investigators at Stanford University and the University of Pittsburgh. Mecoli hopes the cancer-prediction tool will be available for clinical use within two to three years.