Low-risk prostate cancer can and should often be managed with active surveillance and selective delayed intervention

Laurence Klotz  About the author

Low-risk prostate cancer constitutes 50–60% of all newly diagnosed prostate cancers.¹ Patients with low-risk prostate cancer have a long window of curability and a relatively low risk of prostate cancer mortality at 20 years or more. Evidence indicates that, among populations heavily screened for prostate cancer, treating all patients radically results in overtreatment.² Active surveillance of patients with low-risk prostate cancer followed by intervention when pre-specified criteria are fulfilled is an approach that could prevent unnecessary overtreatment.

This strategy is a safe and feasible option in patients with low-risk prostate cancer for several reasons. First, these patients can be identified with reasonable accuracy. Second, no treatment for clinically insignificant prostate cancer exists that has minimal adverse effects and cost. Third, patients who are initially classified as having low-risk prostate cancer, and are later reclassified as having high-risk prostate cancer, can be treated radically and are cured in most cases. Finally, the psychological burden of living with untreated cancer has less effect on quality of life than unnecessary but curative therapy. To implement active surveillance and delayed intervention, patients must be identified who have a low likelihood of disease progression. They must be closely monitored for changes in their cancer, and reasonable criteria for intervention must be fulfilled. Reducing the psychological burden of living with untreated cancer should also be addressed.

Common clinical parameters used to identify patients with low-risk prostate cancer include a Gleason score <6, a serum PSA level <10 ng/ml, and clinical stage T1c–T2a disease. The Epstein criteria³ for clinically insignificant prostate cancer, which, in addition to the parameters described, also include no more than one third of cores involved, and no core with greater than 50% tumor involvement, can be used to identify the group of low-risk patients with the most favorable prognoses. For patients with a short life expectancy based on age and comorbidity, the risk of prostate cancer mortality is low; for these patients, the parameters for active surveillance can be relaxed. Our approach is to offer surveillance to patients aged >65 years with a Gleason score of 6, regardless of the extent of disease, and to those aged >75 years with a Gleason score of 7 (3 + 4). These criteria are reasonable parameters for identifying patients who are suitable for active surveillance; however, they clearly require validation. Concerns about undergrading are justifiable, and the prevalence and effect of undergrading should be assessed in surveillance cohorts with a long follow-up period (i.e. 15 years).

In our group, we perform a confirmatory biopsy at 1 year to identify higher-grade or more-extensive cancer missed on the initial biopsy, and then every 3–5 years thereafter to identify biological progression. In the context of the 30-year natural history of most prostate cancers, this interval seems reasonable. We recommend that patients on active surveillance have their PSA level measured and receive a digital rectal examination every 3 months for 2 years, then every 6 months thereafter, assuming PSA level is stable; have 10–12 core biopsies at 1 year, and then every 3–5 years thereafter until the age of 80 years; and undergo optional transrectal ultrasonography on alternate visits. A summary of our patient eligibility criteria for active surveillance, the follow-up strategy, and the triggers for intervention are described in Supplementary Box 1 online.⁴

The effects of active surveillance, and the attendant 'cancer anxiety', have been studied by several groups. The rates of anxiety, depression and psychological functioning in patients managed with surveillance seem to be no different to those who have received treatment.⁵ This finding indicates that the major effect of prostate cancer on patients' psychological functioning is related to the diagnosis, rather than whether or not the patient has curative therapy. Our approach is to reassure patients as to the safety of active surveillance during their clinic appointment, review the data supporting its use, and provide literature to those who are interested.

Several groups have now reported the results of an active surveillance approach in low-risk patients, although different groups have used different eligibility criteria and triggers for intervention. In contrast to the 'watchful waiting' approach of an earlier era, these groups have offered definitive radical treatment to patients who have been reclassified to a higher-risk category. Klotz et al.⁴ and Choo et al.⁶ were the first to report on a prospective active surveillance protocol incorporating selective delayed intervention for a subset of patients with rapid PSA progression or grade progression on repeat biopsy. The cohort comprised 331 patients with a median follow-up period of 84 months (range 24–132 months); 101 patients (31%) came off active surveillance because criteria for intervention were fulfilled, 32 patients (10%) received radical treatment although they did not fulfill the criteria for intervention, and 198 patients (60%) remained on active surveillance. The indication for intervention was rapid PSA progression in 15%, clinical local progression (unequivocal and substantial increase in palpable disease) in 3%, Gleason score progression in 4%, patient preference in 10%, and more than one indication in 3% of patients. The overall survival was 85% and the disease-specific survival was 99%. Three patients died of prostate cancer; all three had a PSA doubling time <2 years and a very rapid disease progression after diagnosis, which indicated that these patients' outcomes would not likely have been altered by earlier treatment. In this Toronto series, among 18 patients with a PSA doubling time <3 years, only 7 (39%) had positive margins. This finding indicated that, even among patients with a PSA doubling time <3 years, the majority of patients remained curable by delayed therapy.^{4, 6}

An update of a large group of patients in Connecticut treated with watchful waiting reported a 20-year follow-up.⁷ In this pre-PSA screening cohort, 23% of untreated patients with Gleason score 6 died of prostate cancer within 20 years, although these results likely represent a 'worst case' scenario for the expected mortality because of stage migration. Number-needed-to-treat analyses have been performed to investigate whether delayed intervention might cause physicians to miss the chance of cure. In a Swedish trial, the number needed to treat for each death avoided at 10 years was 20 patients.⁸ This population of patients was not screened, was at intermediate risk of disease progression, and had no opportunity for delayed radical intervention. After adjusting conservatively for lead time, grade shift, stage migration, and the effect of delayed curative therapy, it was estimated that about 80 radical prostatectomies would be required for each prostate cancer death averted at 20 years in patients with low-risk prostate cancer. This figure translates to about 6 weeks of survival benefit for each radical prostatectomy performed. Other series report similar, favorable outcomes. Parker et al.⁹ has modeled the likelihood of prostate cancer mortality based on several large datasets. This modelling analysis predicted no prostate cancer deaths at 15 years in patients with Gleason score 6 prostate cancer on active surveillance.

PSA screening results in the detection of aggressive disease at a more curable stage; however, a high percentage of the prostate cancer that is detected is clinically insignificant. Physicians have a social and personal responsibility to minimize the adverse effects of unnecessary treatment in these low-risk patients. A period of active surveillance, during which time PSA kinetics and repeat biopsies can be used to identify patients at high risk of disease progression and thereby trigger intervention, represents a reasonable solution to this problem. This approach is effective and appears to be safe.

Competing interests

The author declared no competing interests.

ARTICLE NAVIGATION -ARTICLE TOOLS

Full text
• Download PDF
• Send to a friend
• Export citation
• Rights and permissions
• Order commercial reprints

ISSUE NAVIGATION - THIS ISSUE

• Table of Contents
• Previous article
• Next article