Prognostic Biomarkers for Prostate Cancer Move Forward

WASHINGTON—Novel prognostic biomarkers for prostate cancer are moving toward the clinic and may eventually join Gleason score and other predictors of relapse to help with treatment decisions, according to data on two candidate markers presented at the 97th Annual Meeting of the American Association for Cancer Research.

Clinicians now rely primarily on a tumor's Gleason score, along with other variables such as stage and PSA levels, to estimate a man's risk of recurrence after prostatectomy. However, these predictive markers are not useful in all cases. For instance, men with an intermediate Gleason score, around 6 or 7, have about a 50/50 chance of recurrence.

Universal Bead Array

The lack of a more accurate prognostic tool for these men is frustrating, said Tracy Downs, MD, a surgeon at the University of California, San Diego, Medical Center, who is collaborating with Illumina, Inc. of San Diego on the development of one of the markers. "There has to be a molecular signature that's predicting the biological nature and aggressiveness of these cancers," he said.

In search of that marker, Illumina has identified a group of 16 genes whose expression pattern appears to correlate with relapse rates (abstract 1203). Of the 16 genes, 10 are positively correlated with cancer progression, and six are negatively correlated.

This microarray uses Illumina's "universal bead array" technology, which requires only tiny amounts of DNA and RNA and can run up to 96 different tests on a single tray. The microarray was able to identify two very distinct groups, among 71 prostate cancer patients, said Jian-Bing Fan, PhD, director of scientific research at Illumina. One group had a 75% chance of relapse in 5 years, while the other group had just a 20% chance.

The researchers are now planning validation studies with other clinical centers and data sets, Dr. Fan said.

Methylated PITX2

The other prognostic marker for prostate cancer highlighted at AACR is the methylation of a single gene, called PITX2, which correlated with poor prognosis in a validation study with samples from 600 prostatectomy patients from three clinical centers (abstract LB-226). After 5 to 6 years of follow-up, 64 of the patients had a cancer recurrence, defined as a rise in PSA levels. Of these, 49 were PITX2 positive, using median methylation level as the cut-off point; 15 of the relapsed patients had methylation levels below the median.

Patients positive for the marker were five times more likely to have a recurrence within 10 years, said lead author Susan Cottrell, PhD, a senior scientist at Epigenomics, Inc., in Seattle. "You can really see that this marker is pulling out those who are most likely to have a PSA recurrence," she said.

The investigators then looked at subgroups of patients, such as those with different stages of disease. "In all cases, PITX2 within these subgroups was able to further subdivide the patients," Dr. Cottrell said. For instance in patients with organ-confined disease, those who were PITX2 positive had a 32% chance of recurrence, compared with 9% among those who were PITX2 negative.

Likewise, in patients with a Gleason score of 7, those with methylation levels below the median had a lower risk of recurrence. The Epigenomics researchers used Cox proportional hazard analysis to demonstrate that PITX2 methylation significantly added information to Gleason score, presurgical PSA, staging, and a nomogram, a computerized tool that combines all available indicators, Dr. Cottrell said.

Next steps in the development of PITX2 are "making this into a validated clinical test so we can get it into the clinic," Dr. Cottrell said. The test could be available in 2 or 3 years, she said, and could help determine which patients should be treated with adjuvant therapy after radical prostatectomy.

Epigenomics is also planning to test its marker in prostate biopsy samples, to see if it could be used in making decisions about initial treatment vs watchful waiting for newly diagnosed tumors, Dr. Cottrell said.

Both Dr. Cottrell and Dr. Downs said that, in clinical practice, they would expect their prognostic markers to be used in conjunction with Gleason score and other indicators.