“Our findings suggest that genomic score and PSA density are risk factors for upgrading within 3 years of commencing active surveillance,” the investigators wrote.
There are currently no validated clinical tools to standardize the frequency of biopsies or clinical assessment for men in active surveillance for prostate cancer. With this in mind, investigators from the University of California San Francisco sought to determine predictors of biopsy reclassification at specific time points over a 10-year period using data from their active surveillance cohort.
The research was presented by Peter Eoin Lonergan, MD, at the 2020 European Association of Urology Virtual Congress.1
“Multi-parametric magnetic resonance imaging and genomic testing have been proposed to aid in tailoring surveillance intensity, but this remains to be determined. With this in mind, we sought to determine predictors of biopsy reclassification at specific time points up to 10 years after enrollment in a cohort of 1031 men at UCSF,” said Lonergan, urologic oncology fellow, Helen Diller Family Comprehensive Cancer Center, University of California San Francisco.
Wrote the investigators, “Our findings suggest that genomic score and PSA density are risk factors for upgrading within 3 years of commencing active surveillance, whereas PSA kinetics is associated with longer-term risk of upgrade. When used in tandem, genomic scores may identify a subset of men who could potentially have less intensive surveillance.”
The study included 1031 men enrolled in the active surveillance cohort at the University of California San Francisco between 2000 and 2016. At diagnosis, the men had a mean age of 62 years, 80% of the men were Caucasian, and 91% were in grade group 1.
Median PSA density was 0.13 ng/mL2. PSA kinetics were calculated using a linear mixed-effects model for log of PSA, adjusted for clinical characteristics. Median slope PSA kinetics at diagnosis was 0.24.
A genomic score was calculated as a composite of results from three commercially available genomic tests: Oncotype DX Genomic Prostate Score, Decipher, and Prolaris. Only 412 men (40% of the total cohort) had data available to calculate a genomic score. A total of 53 men were classified as having a “high” genomic score defined as a composite score greater than the mean plus 1 standard deviation of the cohort.
Only 315 men (30.6%) had a PI-RADS score at diagnosis, of which 126 men had a PI-RADS 1-3 lesion and 189 had a PI-RADS 4-5 lesion.
The primary outcome analyzed was biopsy reclassification to grade group 2 or higher on subsequent biopsy. A multivariable Cox proportional hazards regression model was used to identify factors associated with risk upgrade at first surveillance biopsy and at 3, 5, and 10 years after beginning active surveillance.
In the adjusted analysis, PSA density ≥0.15, percentage biopsy cores positive, and a high genomic score were associated with reclassification at first surveillance biopsy. A high genomic score and PSA density were risk factors for reclassification within 3 years of commencing surveillance. PSA kinetics was associated with a risk of reclassification at 5 and 10 years.
Having a PI-RADS 4-5 lesion was not an independent risk factor for biopsy reclassification on active surveillance at any of the follow-up intervals.
Lonergan noted there may be bias in the study considering that only about 40% of the active surveillance cohort had genomic testing.
“Whether or not a patient gets a genomic test is also at the discretion of the treating provider. Generally speaking, the tests may be done for men with intermediate risk who are trying to delay treatment,” he said.
In a comment submitted during the virtual session, an audience participant complimented the research because of its importance and stated that it should be validated in other active surveillance cohorts.
1. Lonergan PE, Washington SL, Zhao S, et al. Risk factors which predict biopsy upgrading over time in active surveillance for prostate cancer. 2020 European Association of Urology Virtual Congress. July 17-26, 2020. Abstract PT151