EXPERT ANALYSIS FROM THE NCCN ANNUAL CONFERENCE
HOLLYWOOD, FLA. (FRONTLINE MEDICAL NEWS) – Current evidence suggests that molecular tests for prostate cancer are prognostic and can help clinicians and patients with difficult treatment decisions. In the not-too-distant future, gene tests could also guide choice of therapies.
“I think that the largest impact is going to come in areas of both the greatest treatment uncertainty and areas where we can be predictive about the response to treatment,” said Dr. Ashley Ross , a urologic oncologist and pathologist at the Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins University, Baltimore.
Multigene panels may soon be able help identify which patients might benefit more from radical prostatectomy or radiation therapy, whether radiation therapy effects could be enhanced with the addition of androgen-deprivation therapy, and whether early use of docetaxel might add therapeutic benefit, he said at the annual conference of the National Comprehensive Cancer Network.
The 2016 iteration of the NCCN guidelines for the treatment of prostate cancer include a note stating that “men with clinically localized disease may consider the use of tumor-based molecular assays. Retrospective case cohort studies have shown that molecular assays performed on biopsy or prostatectomy specimens provide prognostic information independent of NCCN risk groups.”
The use of molecular assays may inform treatment decisions by helping to predict the likelihood of death if a patient is managed conservatively, risks for biochemical progression after radical prostatectomy or external beam therapy, and the likelihood that a patient could develop metastatic disease after radical prostatectomy or salvage radiotherapy, the guidelines say.
Dr. Ross reviewed the molecular biology of localized prostate cancer and the benefits and risks of currently available molecular tests.
“We’ve had an increased ability to get molecular information or genomic information from very limited amounts of routinely-stored pathologic tissue, and that’s resulted in the generation of many molecular-based tissue tests in prostate cancer. With the emergence of those tests and a lot of aggressive marketing, there has been a lot of confusion for patients and providers about whether we should use them or not and in what context,” he said.
Prostate cancer is genomically complex, even in the localized stage, with copy number alterations, deletions, and amplifications; chromosomal rearrangements; and point mutations, he said.
One of the best characterized genomic events is the early loss of the tumor suppressor gene PTEN (phosphatase and tensin homolog). This gene works within the PI3 kinase (PI3K)/AKT pathway. PI3K pathway mutations have been identified in up to 40% of all primary prostate cancers and 100% of mutations, Dr. Ross explained.
Loss of PTEN itself has been detected in about 15%-40% of primary prostate cancers and 50% of metastases, and the loss correlates with disease stage and tumor grade.
The NCCN guidelines list six available tissue-based tests for prostate cancer prognosis, including tests based on general cancer features such as cell-cycle proliferation, and those based on specific molecular features of cancer.
An example of the general type of test is the Ki-67 immunohistochemistry (IHC) test, which looks for a cellular marker of proliferation, and has been shown to have independent prognostic significance after radiation therapy or radical prostatectomy. This test is not currently recommended by the Medicare Molecular Diagnostic Services (MolDx) program, however.
Another general-type test is the Polaris quantitative reverse-transcriptase polymerase chain reaction (RT-PCR) panel, which tests for 31 cell-cycle-related genes and 15 “housekeeping” controls. This test is recommended for post-biopsy evaluation of men with very-low-risk or low-risk prostate cancer who at the time of diagnosis have at least 10 years of life expectancy. It has been shown to independently predict prostate cancer–specific mortality, biochemical failure/recurrence, and metastasis, the guidelines say.
Tests based on molecular features include:
• PTEN/ERG, an IHC or fluorescent in situ hybridization test that has been shown to predict prostate cancer–specific mortality, upgrading to Gleason pattern 4 on radical prostatectomy, and biochemical recurrence (not MolDx recommended).
• Decipher, a whole-transcriptome 1.4M RNA expression oligonucleotide microarray shown to predict biochemical failure, metastasis, and prostate cancer–specific mortality (recommended for postradical prostatectomy for patients with pT2 tumors with positive margins, and pT3 disease, and rising PSA above nadir);
• Oncotype DX, an RT-PCR assay for 12 prostate cancer genes and five housekeeping controls (recommended for post-biopsy evaluation of men with very-low-risk or low-risk prostate cancer who at the time of diagnosis have at least 10 years of life expectancy).
• ProMark, multiplex immunofluorescent staining of eight proteins, which has been shown to independently predict non–organ-confined pT3 disease or Gleason pattern 4 disease on radical prostatectomy (not reviewed).
Dr. Ross said that in his practice, he generally does not order molecular testing for surveillance of men older than 65 who have very-low-risk disease. For men with low-risk disease, however, molecular testing may help in clinical decision making to predict upgrading or disease progression.
“There’s limited data from surveillance populations, but these tests can be used in this context with retrospective data available, realizing that in most cases the tests will be confirmative, or another way of thinking about it is ‘noninformative,’ so there are some considerations about cost in that context,” he said. For men with intermediate or high-risk disease, however, currently available tests are not good at predicting what an individual patient’s response would be to a specific type of therapy, whether surgery, radiation, androgen deprivation, chemotherapy, or a combination.
“This is an area where predictive biomarkers would be very informative. There is ongoing research, and I think this is an area of potentially large advancement in how we risk-stratify our patients,” Dr. Ross said.
