A central question is: If you have an elevated PSA, what is your risk of having clinically significant disease? To help with the answer, additional serum or urine testing and magnetic resonance imaging (MRI) can be used, and they are complementary. Biomarkers such as percentage free PSA, the Prostate Health Index, or the 4Kscore, as well as many others, may be used to increase specificity and to reduce the number of unnecessary biopsies. Additionally, prebiopsy MRI can potentially prevent unnecessary biopsies in some individuals; if biopsy is required, MRI allows for a fusion biopsy, a technique that decreases the risk of failing to detect clinically significant disease. MRI alone has a negative predictive value of only approximately 85%, so it should not be used alone to determine whether that biopsy is unnecessary. You would use another marker to help refine that. Even PSA density will bring the predictive value up to a higher level. With such approaches, we have been able to avoid probably 30% to 40% of biopsies without significantly compromising the detection of high-grade cancer.

I think that this is a good paradigm shift. You are avoiding biopsies in men who do not have cancer and in those with low-risk (indolent) disease while detecting the vast majority of those with higher-risk disease, disease for which treatment should be considered. We have been big proponents of active surveillance for low-risk patients, but you still see a lot of men with low-risk disease being treated unnecessarily. I used to say that the antidote to overdetection was active surveillance. Now I think that the antidote is to not detect the very low–risk disease to begin with, because we continue to see a lot of unnecessary intervention.

The other thing to remember is that, once you identify prostate cancer, you can better characterize it using contemporary imaging (ie, MRI, prostate-specific membrane antigen [PSMA] positron emission tomography) and genomic profiling. Both give patients and their physicians more confidence that nothing clinically significant has been missed. Ultimately, I believe that MRI and genomic profiling have allowed us to refine the treatment paradigm and to identify the patients who we previously thought needed to be treated who may instead actually be good candidates for surveillance.

Screening has an important role in prostate cancer; when we do detect this disease early, we can cure it. When we do not detect it and it develops before PSA screening, patients can present with metastatic disease and diminished rates of cure. Although screening can make a tremendous difference, overdetection and overtreatment are huge risks for patients, and it is important for them to recognize that.

Multispecialty clinics are especially helpful in generating the different perspectives needed to advise patients about all of their options, including active surveillance. Historically, we have relied on markers including the PSA level, the Gleason score, and the size of the cancer. We have come to realize, however, that there is a lot beyond just those features. In fact, there is a whole biology underlying these Gleason scores that may not be represented. That is where I think some of these tools, such as Decipher (Decipher Biosciences, Inc) or genomic health predictors, have really helped us to understand some of the biological and molecular bases of Gleason scores. These tools should be integrated more regularly, particularly for patients who are contemplating active surveillance.

Incorporating imaging is also critical. Traditionally, prostate cancer has differed from other solid tumors such as lung cancer, breast cancer, and colon cancer, where we are biopsying the tumor, not the whole organ. This has changed more recently with the use of MRI fusion biopsies—these are really developing nicely but are, by no means, perfect imaging techniques. I think that the more we can layer additional imaging modalities—even, potentially, PSMA imaging on top of MRI—the better we will be able to know not only where the cancer is but also the type of cancer and its aggressiveness. Ideally, this should enable us to biopsy just that area of suspicion rather than the entire gland. And I think that this would cut down on some of the overtreatment and excessive biopsies that we are seeing.

I am in agreement with my colleagues regarding the importance of avoiding overtreatment. We are grateful that Dr Carroll has been very influential in helping patients and physicians to understand that active surveillance is perfectly appropriate. And now he is helping to lead the charge in avoiding biopsies altogether in individuals who are very likely to have low-risk disease, so I will continue to follow that work.

The luminal-basal characterization is something that can be done through the transcriptome with RNA, and we have noted the increasing utility of that in patients with low-risk disease. Recently, an article led by one of Dr Carroll’s colleagues at the University of California, San Francisco, Felix Y. Feng, MD, examined luminal-basal phenotyping in concert with genomic testing, looking at how it may shape prognosis, even in patients with advanced disease. And this is really quite intriguing, as it may have some predictive value with regard to the durability of hormonal responses as well. So, while we are looking hard at the genome in traditional ways (ie, the DNA), the transcripts (ie, the RNA) can also be very informative.