A PCF study presented at the 2017 American Society of Clinical Oncology (ACSO) Annual Meeting found that using a 22-gene test along with standard prostate cancer risk group features vastly improves predictions of how aggressive a patient’s cancer is – enabling clinicians to choose more appropriate therapies and better avoid over or under-treatment.
Prostate cancer is a highly variable disease. Some patients have very aggressive cancer that requires aggressive treatment, while in other patients, the cancer grows so slowly, it will never cause them any health problems and would be best left untreated. Clinicians have historically used a combination of Gleason grading (how messed-up prostate cancer tissues look under a microscope), PSA levels, and tumor extent to place patients with localized prostate cancer into low, intermediate, and high risk groups. Typically, low-risk patients are recommended active surveillance, intermediate-risk patients undergo radical therapy (surgery or radiation), and high-risk patients undergo surgery or radiation therapy plus hormone therapy. Unfortunately, this risk grouping system is far from perfect and many patients are inaccurately characterized and over or under-treated. In fact, the system is nearly 20 years old and a lot of what we understand about prostate cancer biology and how to treat it has changed.
Dr. Daniel Spratt, a PCF Young Investigator and radiation oncologist at the University of Michigan, conducted a study to determine whether considering tumor molecular features would improve the accuracy of prostate cancer risk assessments. The Decipher test examines the expression of 22 cancer-related genes in tumor samples to predict how aggressive a tumor is. Almost 7,000 patients from four multi-center test groups underwent the Decipher 22-gene test on prostatectomy or prostate biopsy samples. Patients were then grouped into “clinical-genomic” risk groups that incorporated the 22-gene test score along with standard tumor grading system scores. With the addition of the 22-gene test, over a third of patients were reclassified into risk groups that were far better matched to their long-term risk of metastasis or death. The new clinical-genomic risk groups were 40% more accurate in detecting if a man will develop metastatic disease. With the new classification, patients with “low-risk” prostate cancer went from having an 8% chance of metastasis after 10 years, to 3.7%; while patients with “high-risk” disease went from having a 41% chance of metastasis after 10 years, to a 58% chance. In fact, in one of the test groups, 0% of men in the new clinical-genomic “low risk” group developed metastatic disease or died of prostate cancer.
According to Spratt, this new clinical-genomic risk system would provide confidence to clinicians to use active surveillance for the low risk group, while suggesting a more intensified treatment regimen for high-risk patients, as surgery or radiation therapy plus hormone therapy was not sufficient for controlling cancer in the patients in this study. Prospective clinical trials are necessary to validate this new system of defining prostate cancer risk groups.