Genomic analysis identified favorable treatment option after prostate cancer surgery

A genomic classifier may be effective in predicting the risk of biochemical failure and distant metastasis in post-surgery radiation in patients with prostate cancer who have undergone radical prostatectomy, according to recent findings.

These findings suggest utility for a genomic classifier in deciding a patient’s course of treatment after radical prostatectomy, particularly in terms of identifying which patients will benefit from additional radiation.

“We are moving away from treating everyone the same,” researcher Robert Den, MD, assistant professor of radiation oncology and cancer biology at Thomas Jefferson University in Philadelphia, said in a press release. “Genomic tools are letting us gauge which cancers are more aggressive and should be treated earlier with radiation, and which ones are unlikely to benefit from additional therapy.”

Den and colleagues identified 139 patients with prostate cancer who underwent radical prostatectomy and subsequent radiation therapy at Thomas Jefferson University’s Kimmel Cancer Center between 1999 and 2009. All study participants had pT3 or margin-positive disease and were able to provide tumor specimens for analysis.

The researchers conducted a histopathologic analysis of formalin-fixed, paraffin-embedded tumor blocks and chose the highest Gleason grade lesion for RNA extraction. Microarray expression data was produced for each patient sample, and this data underwent 22-marker genomic classification.

The researchers evaluated the accuracy of genomic classification in predicting biochemical failure and distant metastases after postoperative radiation therapy using receiver operating characteristic, calibration, cumulative incidence and Cox regression analyses. The performance of genomic classification was compared with that of clinical nomograms, including the Stephenson 5-year survival probability nomogram.

The researchers found that the Stephenson model had an area under the receiver operating curve (AUC) of 0.7 for both biochemical failure and distant metastases, and that the addition of genomic classification significantly improved AUC to 0.78 for biochemical failure and 0.8 for distant metastases.

The 8-year cumulative incidence of biochemical failure, categorized by genomic classification risk groups, was as follows: low-risk patients, 21%; intermediate-risk patients, 48%; and high-risk patients, 81% (P<.0001).

For distant metastases, the genomic classification risk stratified 8-year cumulative incidence was the following: low-risk patients, 0%, intermediate-risk patients, 12%, and high-risk patients, 17% (P=.032). Multivariate analysis revealed that high genomic classification was associated with a hazard ratio of 8.1 for biochemical failure and 14.3 for distant metastases.

Among intermediate or high genomic classification patients who underwent radiation, those with undetectable PSA (≤0.2 ng/mL) before the start of radiation had median biochemical failure survival of more than 8 years vs. less than 4 years for those with detectable PSA (>0.2 ng/mL) prior to radiation.

The researchers found that there was a 3% cumulative 8-year incidence of distant metastases in patients with high genomic classification and radiation therapy whose PSA was undetectable vs. 23% distant metastases incidence in patients with detectable PSA (P=.03).

“Our analysis suggests that genomic analysis scores could be used, in concert with other diagnostic measures such as PSA testing, to help determine which patients would benefit from additional radiation therapy and more aggressive measures, and which are less likely to benefit,” Den said in the release.

Disclosure: The researchers reported employment relationships with GenomeDx, manufacturer of the genomic classifier.