Skip to main content
Premium Trial:

Request an Annual Quote

VA Launches Prostate Cancer Screening Study to Improve Early Detection, Personalize Recommendations

Premium

NEW YORK – Enrollment recently began for a prostate cancer screening program sponsored by the Veterans Administration aimed at determining whether genetic testing can improve prostate cancer screening.

Conducted in collaboration with Broad Clinical Labs (BCL), the Prostate Cancer, Genetic Risk, and Equitable Screening Study (ProGRESS) aims to recruit up to 5,000 veterans, who will be randomly assigned to receive either standard of care or precision prostate cancer screening recommendations tailored to their individual genetic profile.

Prostate cancer screening is generally done by testing an individual for elevated levels of prostate-specific antigen (PSA), and while this has reduced deaths among people with the disease, its imprecise nature has also contributed to overdiagnosis and overtreatment.

"[PSA] is not a great test for prostate cancer screening," said Jason Vassy, associate professor of medicine at Harvard Medical School and the study's principal investigator.

Vassy explained that numerous non-cancer conditions that cause inflammation and enlargement of the prostate lead to elevated PSA levels. "And some people just genetically have higher levels of PSA," Vassy said.

To assist in ProGRESS, Vassy and an international group of colleagues developed a prostate cancer risk prediction model based on a polygenic score, family history, and genetic principal components, which is described in a paper currently available as a preprint on medRxiv.

The model, called Prostate CAncer integrated Risk Evaluation, or P-CARE, includes 601 single-nucleotide polymorphisms derived from literature that are associated with prostate cancer as well as other potentially confounding prostate cancer-related phenotypes such as benign prostate enlargement, benign prosthetic hypertrophy, and benign PSA elevation.

P-CARE was validated using data from the Million Veterans Program, as well as four cohorts from the Prostate Cancer Association Group to Investigate Cancer Associated Alterations in the Genome (PRACTICAL) Consortium. The P-CARE model was then used to develop a clinical laboratory assay that incorporated a novel blended genome-exome (BGE) sequencing platform invented by BCL to streamline sequencing for both common and rare variants related to monogenic and polygenic risk calculations.

Notably, an individual with a pathogenic or likely pathogenic variant in one of the 12 prostate cancer-associated genes tested for via the BGE assay is categorized as high-risk, regardless of P-CARE results.

The BGE platform is also currently being used by the state of Alabama as part of a genetic testing public health initiative.

P-CARE also integrates a patient's genetic ancestry and family history into the resulting risk score, which is delivered to patients and their physicians.

The study showed that P-CARE performed well in distinguishing men who developed prostate cancer from those who didn't across both the MVP and PRACTICAL datasets, and when applied to any, metastatic, and fatal cancer populations. Overall, men in the lower and upper 20 percent of the model's values had slightly under half and nearly three times the average risk of prostate cancer, respectively.

"It [shows] a pretty robust association across different kinds of prostate cancer," Vassy said.

Rebecca Graff, associate professor of epidemiology and biostatistics at the University of California, San Francisco, who is not involved in the ProGRESS study, called the P-CARE model a "promising endeavor" for combining information about genetic factors that contribute to prostate cancer risk with other clinical risk factors that are important for decision making around screening.

"P-CARE has the potential to improve prostate cancer screening because it consolidates varying risk factors into a streamlined tool," she said. "Rather than having to pull data from varying sources to describe a patient's prostate cancer risk, clinicians will be better equipped to easily offer an overall picture of risk that will allow for an informed conversation about screening."

Graff cautioned, however, that while the manuscript describing the model does demonstrate that P-CARE predicts metastatic and fatal prostate cancer, it does not describe whether those predictions are stronger than those for less aggressive disease.

"Because prostate cancer screening has the potential to result in overdiagnosis, it is important to demonstrate that the tool will not indicate more screening for disease that will go on to be indolent," Graff said. "Disentangling the genetic factors that influence PSA levels independently of prostate cancer could be important for refining models for aggressive disease specifically."

Vassy likened the ProGRESS study to the WISDOM study of breast cancer risk, in which 100,000 women received either standard screening by mammogram or a more personalized screening based on genetic risk factors, including family history and a PRS.

"WISDOM … was a large, pragmatic breast cancer screening trial, that was very informative as I was planning ProGRESS," Vassy said.

Although not an endpoint of the study, Vassy said that investigators would collect data on how P-CARE reports impact clinical decision making, such as whether providers order more follow-on tests for people with high-risk results and fewer tests for those deemed low-risk.

"We're also going to survey the participants … to ask them what that conversation with their providers looked like," Vassy said. "And we've got future plans to develop a way to ask primary care doctors what it was like to receive these results."

Along those lines, but separate from the ProGRESS study, Vassy mentioned that his lab recently obtained a VA-funded grant to study primary care providers' needs with respect to precision prostate cancer screening. Results of that study, called ScreenShare, will inform the development of tools tailored to those needs.

Vassy also leads the Genomic Medicine at Veterans Affairs (GenoVA) study, which is tasked with assessing the clinical utility of PRS in screening exams for atrial fibrillation, coronary artery disease, type 2 diabetes, colorectal cancer, breast cancer, and prostate cancer. That study aims to determine whether inclusion of these PRS shortens time to diagnosis and impacts the tests that physicians order.

Other studies are also evaluating how well patients and providers understand genetic reports as part of efforts to implement these in clinical practices.

Earlier this year, for example, UK-based Genomics Ltd. published the results of a study evaluating the feasibility of implementing a cardiovascular PRS to clinical care routines in England and whether its use changed treatment decisions. That study determined that in this setting, PRS reports were generally well-received and understood by both patients and providers, easily incorporated into clinical routines, and did impact clinical care decisions, particularly in the cases of patients with the highest estimated polygenic risk.

The ProGRESS study is projected to be funded through 2030, and Vassy expects to be able to present an interim data analysis in approximately two years.