Use of Active Surveillance Increasing for Low-Risk Prostate Cancer

Matthew R. Cooperberg, MD, MPH

The utilization of active surveillance and prostate-specific antigen (PSA) screening has continued to shift in the recent years, particularly following call-to-actions with the updated recommendations from the US Preventive Services Task Force (USPSTF), ASCO prostate cancer screening guidelines, and the collaborative AUA/ASTRO/SUO guidelines.

“Those guidelines now state that active surveillance is the preferred management strategy for low-risk prostate cancer,” states Matthew R. Cooperberg, MD.

It is now recommended for physicians to have decision-making conversations with their patients regarding the benefits and risks of screening, according to the USPSTF recommendations, a draft of which was released in April 2017. According to Cooperberg, active surveillance rates have increased from 10% to 40% and will continue to increase following these revised recommendations.

OncLive: Can you please provide an overview of your presentation of prostate cancer screening?

In an interview during the 2017 OncLive ^® State of the Science SummitTM on Genitourinary Cancers, Cooperberg, an associate professor of urology and Helen Diller Family Chair in Urology at the University of California, San Francisco (UCSF) School of Medicine, discussed the benefits associated with active surveillance for patients with low-risk prostate cancer, as well as the development of personalized medicine.Cooperberg: It is increasingly clear that prostate cancer screening saves thousands of lives but at the expense of overdiagnosis and overtreatment. There is a growing consensus that screening makes sense but in the context of targeting treatments only for the patients who are going to benefit.

We can risk-stratify prostate cancer consistently based on standard clinical parameters, such as PSA, Gleason score, and their tumor biopsy. Those alone get us to about 80% accuracy and we are then able to refine risk-stratification further with factors such as PSA density, the PSA kinetics, and better assessment of histology. We can do even better with the incorporation of imaging tests, such as MRI, and immersion biomarkers for selected patients.

All of this together gives us the ability to risk-stratify and separate the cancers from those that are likely to progress to a clinically meaningful state from those who are not. We have a consensus including clear guideline statements from ASCO and AUA/ASTRO/SUO, who put out a joint statement at the 2017 AUA Annual Meeting. Those guidelines now state that active surveillance is the preferred management strategy for low-risk prostate cancer. It’s good to have these guidelines, which make it clear that most low-risk prostate cancer should not be treated upfront.

We have recent data suggesting that rates of active surveillance have increased from 10% to 40%, which has been verified in a few different data sources including data from the AUA Quality (AQUA) Registry. Forty percent is still too low, but it is tremendous progress in the right direction.

It was cited by the USPSTF that they have changed the recommendation from “don’t screen anyone” to having a shared decision-making conversation with patients about the risks and benefits of screening. This is what other guidelines are saying, as well.

The role for biomarkers and MRIs is evolving rapidly. There is no adjuvant test that is formally endorsed by the guidelines. Additionally, no adjuvant test—either imaging or biomarker—needs to be used every time for every patient to help make a decision about active surveillance. These tests have a growing role for selective patients, such as those who are particularly anxious about the decision.

There are 3 tissue tests that are on the market: Prolaris^®, Decipher^® Prostate Cancer Test, and Oncotype DX^®. All of these can improve our ability to predict the likelihood of the cancer progressing. None of them have been shown to make clear decisions about whether the patient should be treated or not. These tests can help with clinical decision making, but they don’t replace the interpretation of the clinical information and the discussion that the urologist must have with the patient.

MRI is being advocated as a replacement to biopsy and to facilitate decision making but we are not there yet at the community level. This test requires a great deal of some specialty expertise. It has its place and can be useful in select circumstances, but does not need to be used for every patient.

For patients with high-risk prostate cancer, we are seeing one of the greatest stage migrations since the start of the PSA era, and that is being driven by novel imaging tests. Until recently, we had bone-scanning CT scans, which were our staging modalities for men with high-risk prostate cancer. We have a large body of evidence for men with high-risk but clinically localized prostate cancer, stating that local therapy is more effective than watchful waiting or systemic therapy alone in terms of minimizing prostate cancer mortality.

We also have controversial evidence from several trials suggesting that surgery should be the [in the] first-line setting of therapy for high-risk prostate cancer. This is often followed by radiation, but surgery should be part of the paradigm. We have novel imaging tests, such as prostate-specific membrane antigen (PSMA)—based PET and fluorocyclobutane-1-carboxylic acid (FACBC) PET scans. These are showing us smaller volumes of metastatic disease than we would have detected before. We are often seeing them in locations that we would not have expected based on classical teaching on the spread patterns for prostate cancer. Now, we are reclassifying men based on changing the threshold of detection. The question becomes, “Should we take a man who has a positive lymph node or small bone metastasis on a PSMA scan, and tell him he’s ineligible for local therapy and should receive hormonal therapy?”

We are going to gradually evolve into a paradigm of aggressive multimodal, cytoreductive local treatment for men with aggressive high-risk local cancers and limited volume of metastatic disease. This needs to be validated in prospective trials, some of which are ongoing. In kidney cancer, the paradigm for more than 20 years for a patient with a large renal primary and a limited volume of metastatic disease with good performance status, is cytoreductive nephrectomy as the lead treatment. This is not because cytoreductive therapy cures anybody but it extends survival. We are gradually realizing that the same paradigm applies in prostate cancer. For patients with high-risk cancers, the correct approach is multimodal and often involves surgery, radiation therapy, and systemic therapies, both traditional androgen deprivation—based therapies and immunotherapy, which are coming down the pipeline.

Do oncologists have a preference between the 3 tissue tests?

The remaining questions are what sequence should a patient receive, what combinations, and how those combinations can be tailored to individual patient biology and clinical situation.There have been no head-to-head tests comparing the 3 tests. We’ve done validation work with all 3 companies at UCSF. There is overlap in terms of the biology that the tests are investigating. In terms of approved situations, Prolaris® is approved for both the biopsy and radical prostatectomy situations. Oncotype DX^® is focused on the biopsy. Decipher^® Prostate Cancer Test has focused mostly on the post-prostatectomy setting, but they are now available in biopsy, as well.

They do get different pathways and different aspects of cancer biology, although they have similar accuracy in terms of improving the prognostic classification of patients. If we did do a head-to-head comparison, my expectation is that it wouldn’t vary by more than a couple percentage points.

From the academic standpoint, we tend to favor the Decipher^® Prostate Cancer Test platform, not because the score itself is better, but because of what the test provides. It is not just reverse transcription polymerase chain reaction results on a fixed gene set, but rather the entire Affymetrix expression array. Coming from an academic standpoint, we are getting a great deal more information about the cancer than we would get with the other tests.

In the community, I am not sure how much is known about the uptake and use of the different tests. A lot of it may come down to the accuracy of the test itself and the biology, but also how the tests are presented to patients. The test report for Oncotype DX^® was misleading to the point of being dangerous. They’ve finally updated it in the last couple of months but for the first couple of years, we were not ordering it because the report was so poorly formatted.

None of these tests are perfect; none of them function as a binary classifier for treatment versus no treatment. In many cases, it comes down to how the urologist uses that information in the context of an overall clinical decision-making conversation. That can be complex.

Are there any other trials that you are excited about?

There is a trial ongoing at the University of Michigan that Dr Todd Morgan is running that we are very excited about. They are cluster-randomizing community-based practices across the state to use Decipher® Prostate Cancer Test after prostatectomy versus not using the test after prostatectomy. Since this is a randomized trial, we will finally get a real readout on how the test affects decision making and outcomes. There are 3 trials of adjuvant versus early salvage radiation therapy. There are already 3 randomized trials published that support the use of secondary radiation therapy after surgery for men with adverse pathology. None of those trials got to the question of whether secondary radiation needs to be adjuvant, based on the pathology report alone versus early salvage.

Our practice pattern at UCSF tends to favor early salvage rather than adjuvant. Therefore, even if a patient has an adverse pathology report with an undetectable sensitive PSA, they can often avoid radiation therapy. This is where some of the best data for the biomarkers and Decipher® Prostate Cancer Test has been published. We are still not quite there as far as being a true predicative test, but they are getting closer.

Are there any other takeaways that you want to mention for the audience?

The field seems to be barreling toward this concept of cytoreductive local treatment for limited metastatic disease but that concept does need to be validated in randomized trials. There is a trial that The University of Texas at MD Anderson Cancer Center is leading with a few other centers, including UCSF, the results of which should be out in another year. There are similar trials in Germany, Australia, and Sweden, for which we might see results sooner. The existing state of biomarkers is not the end of the story of how we are going to personalize treatment for men with prostate cancer. They certainly represent a very important first step down this path to personalizing treatment. The next phase is going to be digging a lot deeper into the biology and making predictions about which cancers will benefit most from surgery versus other treatments.

We have some interesting data from UCSF that was presented at the 2017 AUA Annual Meeting, which will hopefully be published soon. Finding that even low-risk prostate cancers by the standard clinical definition can be stratified into very different types of biology. Those different subtypes from a biologic standpoint look similar to subtypes of breast cancer, such as luminal A, luminal B, and basal. It is our hope that we’ll be able to stratify men to immediate treatment versus delayed treatment and what the types of treatment are. Importantly, how can we start to tailor active surveillance for the men who we think do not need immediate treatment?

Active surveillance is still burdensome to men. It involves multiple biopsies, anxiety with subsequent testing, and has significant cost associated with it. We know biologically that many men who get classified to active surveillance could just go home and forget all about it. It would be nice to be able to stratify them to watchful waiting rather than active surveillance. Tailoring the intensity of active surveillance to an individual patient’s clinical risk is an important next step with biomarker research and personalized medicine research.