New support for active surveillance of prostate cancer

Science Spotlight

New support for active surveillance of prostate cancer

From the Etzioni Group, Public Health Sciences Division

Jan. 15, 2018

Prostate cancer is one of the most common cancers diagnosed in men. Although one in seven men will be diagnosed with prostate cancer in his lifetime, nearly 90% of prostate cancer diagnoses occur at an early stage, with the cancer confined locally or regionally. The prognosis for patients with prostate cancer diagnosed in these early stages is excellent, with a 5-year survival rate of nearly 100%. Because many prostate cancers are considered ‘low-risk’, with very slow growth and progression over time, management of the cancer may not require treatment immediately upon diagnosis. One such management approach that has gained popularity in recent years is ‘active surveillance’. During this approach, the cancer is typically monitored up to two times per year by measuring the level of prostate-specific antigen (PSA) using a blood test and a digital rectal exam, while prostate biopsies are conducted less frequently, commonly every or every other year. Treatment is initiated only after cancer progression is detected. An advantage of this management approach is that it may help reduce prostate cancer overtreatment and associated morbidity. Dr. Ruth Etzioni and colleagues in the Public Health Sciences Division recently published a paper in Annals of Internal Medicine that reports results from a cross-cohort analysis of risks for prostate cancer progression during active surveillance.

Tissue section of prostate cancer.

Tissue section of prostate cancer.

Image by Robert Hood/Fred Hutch

Although active surveillance is a routine management approach for low-risk prostate cancers, a standardized implementation protocol has not yet been established. Thus, there is substantial variation in determination of patient eligibility, follow-up monitoring schedule and tests, definition of cancer progression, and timing of treatment initiation. In clinical research studies, such variation precludes the ability to identify the ideal active surveillance protocol and develop the best possible management guidelines. The authors sought to compare the risks for cancer progression across four large active surveillance cohort studies conducted in North America that varied in their implementation protocols. To standardize the cohorts, cancer progression was defined as an upgrade in biopsy Gleason score to seven or higher, a commonly used cut-point to distinguish low-risk prostate cancers from those that are of intermediate- or high-risk. The authors also controlled for variation in the follow-up schedule. The study included data collected from a total of 2,576 individuals from across the four cohorts.

The risk for Gleason score upgrading during active surveillance was estimated using regression modeling separately for each of the cohorts. In the absence of competing treatments, the cumulative probability of biopsy upgrading differed by more than two-fold across the four cohorts. The ten-year cumulative risk for biopsy upgrade ranged from 25% to 65%. In contrast, PSA velocity, the annual rate of increase in circulating PSA, among the cohorts was more consistent and ranged from 5% to 8%. The authors then compared the risk for upgrade by the level of surveillance intensity, as determined by biopsy frequency. As expected, a biennial biopsy schedule initiated at either the time of enrollment or one year after enrollment following a confirmatory biopsy significantly reduced the number of biopsies as compared to an annual biopsy schedule. However, the time to detection of biopsy upgrade was delayed by six to eight months when the biennial biopsies were started at enrollment but by only three to five months when started one year after enrollment. The authors used an algorithm previously established to estimate risk for death from prostate cancer when detection of biopsy upgrade is delayed. The risk of death increased by approximately 1.5% to 2.5% across the cohorts. The highest risk for death in each cohort was associated with longer delays in detection of upgrading.

Taken together, the results from this study support the recommendation to use the active surveillance management method for low-risk patients. In addition, the finding that a biennial biopsy is an acceptable monitoring schedule is consistent with clinical practice guidelines that advise less frequent biopsies during active surveillance. 


Also contributing to this project from the Fred Hutch were Amy Leonardson and Roman Gulati and Drs. Lurdes Inoue, Daniel Lin, and Lisa Newcomb.

Research reported in the publication is a collaboration between Cancer Consortium members Ruth Etzioni (Fred Hutch), Lurdes Inoue (UW), and Daniel Lin (UW).

This research was supported by the National Cancer Institute.

Inoue LYT, Lin DW, Newcomb LF, Leonardson AS, Ankerst D, Gulati R, Carter HB, Trock BJ, Carroll PR, Cooperberg MR, Cowan JE, Klotz LH, Mamedov A, Penson DF, Etzioni R. Comparative Analysis of Biopsy Upgrading in Four Prostate Cancer Active Surveillance Cohorts. Annals of Internal Medicine. 2017. doi: 10.7326/M17-0548.