Use of Active Surveillance in Prostate Cancer on the Rise, but Highly Variable

While use of active surveillance for the management of low-risk prostate cancer has increased in the U.S., it still varies widely both at the practitioner level and even within urology practices, according to a cohort study.

Looking at over 20,000 patients included in a quality reporting registry, the rate of active surveillance increased “sharply and consistently” from 26.5% in 2014 to 59.6% in 2021, reported Matthew R. Cooperberg, MD, MPH, of the University of California San Francisco Helen Diller Family Comprehensive Cancer Center, and colleagues.

However, use of active surveillance was variable at the urology practice level, ranging from 4% to 78%, and from 0% to 100% at the practitioner level, over this time period, they noted in JAMA Network Open.

Despite the increase, active surveillance rates remain “suboptimal,” suggested Cooperberg and colleagues, adding that while the optimal rate hasn’t been defined, it is likely greater than 80% based on reports from the U.S. Veterans Affairs system and other health systems.

On multivariable analysis, year of diagnosis was most strongly associated with active surveillance use, with an odds ratio (OR) per year of 1.25 (95% CI 1.24-1.27). The ORs for individual years increased progressively to 4.48 (95% CI 4.31-4.65) for 2021 relative to 2014, the authors said.

In addition, older patient age and lower prostate-specific antigen (PSA) level were associated with active surveillance. Black men were less likely to receive active surveillance than white men (OR 0.87, 95% CI 0.75-1.00).

Neither practice volume nor indicators of practitioner density affected the odds of active surveillance. However, higher urologist density per capita was associated with slightly lower odds of active surveillance (OR 0.92, 95% CI 0.81-1.03), Cooperberg and team reported, “suggesting that more competition in a given area tends to create more opportunities for overtreatment.”

They said that variation in the use of active surveillance has been identified previously, but within smaller sets of practices. Moreover, the variation in the use of active surveillance “is not unique to [active surveillance] for prostate cancer,” and variations in practice exist in every area of clinical medicine for which multiple clinical options are available.

Nevertheless, “continued progress on this critical quality indicator is essential to minimize overtreatment of low-risk prostate cancer and by extension to improve the benefit-to-harm ratio of national prostate cancer early detection efforts,” they concluded.

For this study, Cooperberg and colleagues used data collected from the American Urological Association Quality (AQUA) Registry, a quality reporting registry that collects data from 349 practices across the U.S., representing 1,945 urology practitioners, and including more than 8.5 million patients.

Among 298,801 patients newly diagnosed with prostate cancer from 2014 to 2021, 27,289 were diagnosed with low-risk disease (defined as a PSA level less than 10 ng/mL, Gleason grade group 1, and clinical stage T1c or T2a).

Of these patients, 20,809 had known primary treatment. Median age was 65 years, 40.1% were white, and 8.9% were Black; 49.3% were missing information on race or ethnicity.

Cooperberg and colleagues noted that electronic health record data are often incomplete, and race data are self-reported, which were limitations of their study. Furthermore, access to data on social determinants for patients in the AQUA Registry are limited, and findings on racial disparities in active surveillance use should be interpreted with caution.