According to a study published online May 9 in The American Journal of Pathology copy number variations (CNV) in both malignant and benign prostate tissue is predictive of prostate cancer relapse. The research performed at the University of Pittsburgh School of Medicine by Yan P. Yu, M.D. and colleagues evaluated whether CNV of the genomes of prostate cancer tumors, adjacent tissues, or blood samples could predict biochemical (prostate specific antigen or PSA) relapse and the kinetics of the relapse.

Dr. Yu analyzed 104 prostate tumor samples, 49 samples of benign prostate tissue adjacent to a tumor, and 85 blood samples from men with prostate cancer. He then used gene-specific CNVs to correctly predicted 73 percent of relapses and 75 percent of cases with a short PSA doubling time (PSADT) of less than four months. Using median-sized CNV from the tumors, the genome model correctly predicted 75 and 80 percent of cases for relapse and cases where the men had a short PSADT, respectively. In adjacent tissue samples, the gene-specific CNV models correctly identified 67 percent of relapses and 77 percent of cases with short PSADT. Using blood samples, the gene-specific CNV model correctly predicted 81 percent of relapses and 69 percent of short PSADT cases.
The researchers concluded that “CNV analysis on the genome of blood, normal prostate, or tumor tissues of the patients with prostate cancer holds promise to become a more efficient and accurate way to predict the behavior of prostate cancer.”

This genetic analysis is not yet ready for prime time as it needs to be replicated. However it can shed some light in prognosticating PSA relapse post treatment.

Joel T Nowak, M.A., M.S.W.