This change is timely, and particularly notable in contrast to the NCCN guideline, which last year not only reiterated the “very low risk” category but transiently removed “preferred” from its listing of surveillance for “low risk” disease, endorsing prostatectomy and radiation as equivalent alternatives. The NCCN reversed this decision a few months later, and now once again states that surveillance is “preferred,” though with multiple caveats and exceptions. So which guideline is more relevant for contemporary practice?
The NCCN (and former AUA) “very low risk” definition recapitulates the Epstein criteria for indolent cancer, published nearly 30 years ago—barely even into the era of widespread PSA screening—at a time when biopsies were performed on sextant templates, with minimal imaging guidance. The criteria—requiring not only stage T1c, PSA <10 ng/ml, and grade group (GG) 1 but also PSA density <0.15 ng/ml/g, no more than 2 cores positive, and no more than 50% of any single core involved—have always been cumbersome to apply in practice and excessively conservative, intended as they originally were to predict not only low-grade, organ confined disease at prostatectomy, but also a tumor volume <0.5 cc. In the past decade, we have recognized that the 0.5 cc threshold is needlessly low; but more importantly, in the era of extended templates and targeted biopsies, applying the last two criteria without modification makes little clinical sense, and the proportion of diagnosed cases meeting the “very low” definition will likely dwindle to irrelevance.
In at least some health care systems, surveillance is already used for >80% of patients with low-risk disease, and in the US the Veterans Affairs system has also achieved that threshold. Rates in broader community practice have lagged but are trending rapidly upward. Fairly extreme variation does persist, however, across individual urology practices and providers. To be clear, the “optimal” rate of surveillance is not 100%; there will always be situations—e.g., high-volume disease in young patients or those with germline mutations and/or strong family history of lethal disease, those with concurrent severe obstructive symptoms who prefer definitive surgery to outlet procedures combined with surveillance, among others—for which immediate treatment is warranted in context of shared decision making. We do need to sub-stratify “low-risk” disease more consistently, and while “very low” was never optimal, the continued use in both the AUA and NCCN guidelines of coarse risk groups rather than more contemporary and reliable systems like nomograms or the well-validated CAPRA score is disappointing. Genomic tests, while still too expensive for routine use in every low-risk case, are clearly able to identify occult high-risk biology and are appropriately endorsed by the new guidelines for select cases.
For many men, active surveillance denotes delayed rather than avoided treatment, which is a critical point. The window of opportunity to cure prostate cancer is generally measurable in years and decades, and in most cases progression can be identified well within this window. Just because a GG1 cancer in, e.g., a younger man with a BRCA mutation, is more likely to progress than a GG1 cancer in the general population does not mean the cancer needs to be treated immediately at diagnosis. Indeed, we cannot promote early baseline PSA testing among young patients with or without mutations—those most likely to benefit from screening—unless we are prepared to offer surveillance nearly universally when GG1 is identified.
To be clear, surveillance must be performed properly, including regular PSA and periodic biopsy, augmented selectively with imaging and biomarkers. A confirmatory biopsy within the first year is particularly important, imaging results notwithstanding, but downstream surveillance intensity can be very much individualized. However, data from MUSIC and elsewhere suggests men frequently do not undergo sufficient follow-up, and the ProtecT trial clearly showed higher rates of progression for those treated with PSA-based monitoring only.
Looking forward, we still have work to do in reducing overdiagnosis of GG1 tumors through smarter screening and liberal use of imaging and biomarkers—and perhaps we will make progress toward removing the “cancer” label from at least a subset of GG1 lesions. We should continue to refine and personalize surveillance protocols to minimize the burden and costs for most men, while striving to identify incipient progression early. Consistently identifying GG2 tumors (e.g., those with small proportions of fused / poorly formed subtype pattern 4) which are as indolent as GG1 is another key frontier of ongoing investigation and progress.
As a specialty, we need to continue to work in academic and community practice alike to address both overdiagnosis and overtreatment of low-risk prostate cancer. The 2022 AUA guideline is a major step in the right direction.
Written by: Matthew R. Cooperberg, MD, MPH, Professor of Urology, Epidemiology & Biostatistics, Helen Diller Family Chair in Urology, UCSF Department of Urology, University of California, San Francisco, California
Related Content:
AUA 2022: AUA Guidelines: Localized Prostate Cancer