For many men, prostate cancers are extremely slow-growing — so much so that they often can be monitored periodically through blood tests, biopsies and MRI without necessarily needing treatment. Johns Hopkins experts have directed an extensive active surveillance program for 2,000 patients with these favorable-risk cancers since the mid-1990s.

Precision Medicine at Work

More recently, urologists have had a precision medicine tool at their disposal to help identify any changes in these cancers that might necessitate treatment. The model, called Active Care, is a software algorithm that employs sophisticated statistical modeling to incorporate all of the patient’s prostate-specific antigen (PSA) protein level tests, biopsy results and other clinical information to estimate the chance that there is a more aggressive cancer in the prostate, explains urologist Christian Pavlovich, M.D., director of the Johns Hopkins Precision Medicine Center of Excellence for Prostate Cancer and the Active Surveillance Program.

“It gives us a prediction. For example, one patient may have a 75% chance that their cancer is truly low-grade where continued surveillance would be the right thing, and a 25% chance there might be something more aggressive lurking within the prostate,” he says. “Then, if we do another biopsy and don’t find anything more aggressive, the model’s prediction of aggressive cancer may decrease to, say, a 19% chance because of the new information.”

The predictive model is run weekly, and uses the power of all other patients in the program to keep fine-tuning its accuracy. “It helps men on surveillance understand where their cancer is going,” Pavlovich says. “Everyone’s an individual. That’s why we have to keep pursuing a personalized approach to these patients. Every cancer is different, and every treatment has specific side effects and its own pros and cons.”

Pavlovich has been working with Aki Nishimura, Ph.D., a biostatistician at the Johns Hopkins Bloomberg School of Public Health, to expand Active Care by incorporating MRI findings into the model, which should enhance its accuracy even further. “The goal is one day to have an online platform where we could share this with other patients and programs,” he says.

More information on prostate cancer.