https://orcid.org/0000-0002-7614-9931
-
Views
-
Cite
Cite
Jeffrey J Tosoian, Active surveillance for prostate cancer: How active is too active?, JNCI: Journal of the National Cancer Institute, Volume 117, Issue 4, April 2025, Pages 583–585, https://doi.org/10.1093/jnci/djae342
Close - Share Icon Share
Extract
The first prospective studies of active surveillance (AS) for low-risk prostate cancer were initiated in two large North American centers in 1995.1,2 In contrast to watchful waiting, which offers palliative treatment of symptomatic disease in patients unlikely to benefit from curative treatment, active surveillance involves monitoring low-risk cancers with an intention to offer curative treatment if higher-risk features are detected during follow-up. Indeed, active surveillance is intended for patients that are both eligible and appropriate for curative treatment, with aims to avoid or delay treatment and the associated risk of side effects unless proven to be necessary.3 Although proposed at a time of mounting evidence that low-grade cancers posed minimal threat to quantity or quality of life,4,5 the concept of monitoring cancer was initially met with some scrutiny.
Now 30 years later, AS is the standard of care for low-risk prostate cancer,6,7 and the conversation has shifted to the other end of the spectrum. Instead of questioning the concept of surveillance, the field is now asking—with good reason—whether the intensity of monitoring can be reduced. Although monitoring protocols are not standardized, the majority include early repeat biopsy (ie, confirmatory biopsy) to rule out the presence of high-grade disease potentially missed by diagnostic biopsy, followed by biannual serum prostate-specific antigen (PSA) and scheduled biopsies approximately every 2-4 years.6-8 Despite middling data in the setting of AS,9,10 surveillance magnetic resonance imaging (MRI) has been variably incorporated at most centers, largely in an effort to reduce the frequency of follow-up biopsies. Understanding that the prospective multi-arm studies needed to best inform the intensity of monitoring are unlikely to occur, data-driven modeling studies offer unique insight on the question.
