Markers May Indicate Occult High-Grade Prostate Cancer

Nick Mulcahy

February 21, 2013

Three "field-effect" biomarkers, which are used to detect prostate cancer in tissue adjacent to histologically negative biopsies, might also be indicative of occult aggressive cancers, according to a study presented at the 2013 Genitourinary Cancers Symposium (GUCS) in Orlando, Florida.

This study confirms what was previously found in a small cohort of subjects, according to the researchers, led by Sandra Gaston, PhD, from Harvard Medical School in Boston, Massachusetts.

The presentation follows the recent publication of a major study of 3 biomarkers that collectively make up the commercially available test known as ConfirmMDx for Prostate Cancer (MDxHealth), which is an epigenetic assay that helps urologists distinguish patients who have a true-negative biopsy from those at risk for occult cancer.

The results of that study, known as MATLOC (Methylation Analysis to Locate Occult Cancer), were first presented last year at the American Urological Association (AUA) annual meeting, and have now been published in the March issue of the Journal of Urology.

The MATLOC results show that the assay correctly confirmed negative biopsies in approximately 64% of men without prostate cancer. With its 90% negative predictive value, this assay could prevent unnecessary repeat prostate biopsies, according to the researchers. In addition, the assay correctly identified the presence of occult cancer in 68% of patients who had initial false-negative formalin-fixed paraffin-embedded (FFPE) biopsies.

In their poster presentation at GUCS, Dr. Gaston and colleagues report on a potential additional utility of the biomarkers — their ability to identify more aggressive occult cancers.

They explain that epigenetic changes induced by DNA methylation can serve as biomarkers for the presence of neighboring prostate cancer in tissue samples. "Gene methylation in histologically benign tissue adjacent to occult prostate cancer can provide field-effect biomarkers to detect cancer missed due to sampling errors," the researchers write in their abstract.

Dr. Gaston and colleagues used the methylation markers of prostate cancer (GSTP1, APC, and RASSF1) that comprise the ConfirmMDx test to compare histologically benign biopsy cores from patients diagnosed with no cancer, from those diagnosed with low-volume cancer (Gleason score [GS] 6), and from those diagnosed with higher-volume cancer (GS 7).

The study involved 12 tissue cores from each of the 76 biopsy patients: 34 diagnosed with GS 6 prostate cancer, 22 diagnosed with GS 7 prostate cancer, and 20 control subjects with no cancer.

In 19 of the 22 GS 7 cases (86%), epigenetic markers were positive for field effects in 1 or more adjacent benign cores, and 12 (54%) were positive for all 3 epigenetic markers. In contrast, only 9 of 34 (26%) GS 6 cases were positive for all 3 markers.

APC performed best as a single field-effect marker for adjacent high-grade prostate cancer, the researchers report.

They found that GSTP1, APC, and/or RASSF1 gene promoter methylation was more prevalent in histologically benign cores from biopsy patients who were later diagnosed with GS 7 prostate cancer than in those with GS 6 cancer. Thus, the assay could help predict whether occult cancers are higher grade and therefore more urgent candidates for treatment.

Published Study Results

In the MATLOC study, Grant D. Stewart, BSc, MB ChB, PhD, from the University of Edinburgh in the United Kingdom, and colleagues point out that an estimated 20% of all men will develop prostate cancer, even when the first biopsy is histopathologically negative. Clinicians and patients need a test to provide assurance that a negative biopsy for prostate cancer is truly negative, they write.

Dr. Stewart's team compared initially negative FFPE biopsies with subsequent biopsies conducted within 30 months. Prostate cancer cases were defined as patients with second biopsies that indicated prostate cancer.

On average, individual prostate biopsies consisted of approximately 10 separate cores, each processed using quantitative methylation-specific polymerase chain reaction technology with two 10-µm sections (i.e., ConfirmMDx for Prostate Cancer, the multiplexed methylated gene panel that contains GSTP1, APC, and RASSF1). A positive result was defined as a methylated value above the analytical cutoff in at least 1 core.

The researchers tested, in a blinded fashion, 4625 prostate biopsy cores collected from 483 men at 3 sites in Belgium and the United Kingdom.

The prevalence of prostate cancer in the repeat-biopsy cohort was 18%. In a multivariate model that corrected for age, prostate-specific antigen level (PSA), digital rectal exam, and histopathologic characteristics of the first biopsy, the assay was a significant independent predictor of patient outcome (odds ratio, 3.17; 95% confidence interval, 1.81 - 5.53).

"The addition of this epigenetic assay could improve the prostate cancer diagnostic process and reduce unnecessary repeat biopsies," Dr. Stewart and colleagues write.

When the results were first presented at the AUA meeting, an independent commentator put the study in clinical context.

"There are many more negative biopsies than positive, but then many repeat biopsies are needed when the PSA rises or the rectal exam changes," David I. Lee, MD, from the University of Pennsylvania in Philadelphia, told Medscape Medical News. "If there was a way to examine the normal tissue that could prove that a particular patient had an even lower risk of developing prostate cancer, this could certainly reduce the number of repeat biopsies," he said.

Dr. Lee added that this is "quite a novel test in the realm of prostate cancer diagnosis. There is no other way that I am aware of in which examination of the normal tissue removed from biopsy can lead to a better prediction of whether a patient will later develop prostate cancer."

Both studies were supported by MDxHealth. Some researchers from both studies are employees of MDxHealth. Dr. Lee has disclosed no relevant financial relationships.

2013 Genitourinary Cancers Symposium (GUCS): Abstract 116. Presented February 14, 2013.

J Urol. 2013;189:1110-1116. Abstract