Urinary DNA methylation biomarkers for prediction of prostate cancer upgrading and upstaging.

Bakavicius A1,2,3, Daniunaite K2,4, Zukauskaite K2,4, Barisiene M1,3, Jarmalaite S5, Jankevicius F1,2.

Author information

1: Institute of Clinical Medicine, Faculty of Medicine, Vilnius University, Vilnius, Lithuania.
2: National Cancer Institute, Vilnius, Lithuania.
3: Vilnius University Hospital Santaros Klinikos, Vilnius, Lithuania.
4: Institute of Biosciences, Life Sciences Center, Vilnius University, Vilnius, Lithuania.
5: National Cancer Institute, Vilnius, Lithuania. sonata.jarmalaite@nvi.lt.

Abstract

BACKGROUND:

Significant numbers of prostate cancer (PCa) patients experience tumour upstaging and upgrading in surgical specimens that cause serious problems in timely and proper selection of the treatment strategy. This study was aimed at the evaluation of a set of established epigenetic biomarkers as a noninvasive tool for more accurate PCa categorization before radical prostatectomy (RP).

METHODS:

Quantitative methylation-specific PCR was applied for the methylation analysis of RARB, RASSF1, and GSTP1 in 514 preoperatively collected voided or catheterized urine samples from the single-centre cohort of 1056 treatment-naïve PCa patients who underwent RP. The rates of biopsy upgrading and upstaging were analysed in the whole cohort.

RESULTS:

Pathological examination of RP specimens revealed Gleason score upgrading in 27.2% and upstaging in 20.3% of the patients with a total misclassification rate of 39.0%. DNA methylation changes in at least one gene were detected in more than 80% of urine samples. Combination of the PSA test with the three-gene methylation analysis in urine was a significant predictor of pathological upstaging and upgrading (P < 0.050), however, with limited increase in overall accuracy. The PSA test or each gene alone was not informative enough.

CONCLUSIONS:

The urinary DNA methylation assay in combination with serum PSA may predict tumour stage or grade migration post-RP aiding in improved individual risk assessment and appropriate treatment selection. Clinical utility of these biomarkers should be proven in larger multi-centre studies.