Computer Algorithm Can Distinguish Dangerous From Manageable Prostate Cancer Interview with:

Davide Pellacani Ph.D. Postdoctoral Fellow, Eaves' Lab Terry Fox Laboratory, BC Cancer Research Centre Vancouver, BC

Dr. Pellacani

Davide Pellacani Ph.D.
Postdoctoral Fellow, Eaves’ Lab
Terry Fox Laboratory,
BC Cancer Research Centre
Vancouver, BC What is the background for this study? What are the main findings?

Response: Prostate cancer is characterized by frequent DNA methylation changes compared to normal tissue. Nevertheless, understanding which of those changes lead to the acquisition of malignant proprieties is complicated by the predominance of cells with luminal features in prostate cancers.

In this study we generated DNA methylation maps of two distinct cell populations found within prostate cancer samples (called basal and luminal) and their normal counterparts. These datasets clearly showed that many of the common DNA methylation changes found in prostate cancer are present in luminal cells from both cancer and normal tissues. These changes are not necessarily cancer-specific, and are likely due to the bias associated with analyzing tissues in bulk, where most cancer cells have luminal-like features.

We used these datasets to derive two cancer-specific and phenotype-independent DNA methylation signatures: one specific to prostate cancer luminal cells, and one composed of changes measured in both luminal and basal cancer cells.

We then validated the potential clinical utility of these signatures by testing their ability to distinguish prostate cancer from normal samples, and tumours that have already escaped the prostate from those that have not, using the publicly available dataset from The Cancer Genome Atlas.

Continue reading