TY - JOUR
T1 - 5hmC Level Predicts Biochemical Failure Following Radical Prostatectomy in Prostate Cancer Patients with ERG Negative Tumors
AU - Kristensen, Gitte
AU - Strand, Siri H
AU - Røder, Martin Andreas
AU - Berg, Kasper Drimer
AU - Toft, Birgitte Grønkær
AU - Høyer, Søren
AU - Borre, Michael
AU - Sørensen, Karina Dalsgaard
AU - Brasso, Klaus
PY - 2019/3/1
Y1 - 2019/3/1
N2 - This study aimed to validate whether 5-hydroxymethylcytosine (5hmC) level in combination with ERG expression is a predictive biomarker for biochemical failure (BF) in men undergoing radical prostatectomy (RP) for prostate cancer (PCa). The study included 592 PCa patients from two consecutive Danish RP cohorts. 5hmC level and ERG expression were analyzed using immunohistochemistry in RP specimens. 5hmC was scored as low or high and ERG was scored as negative or positive. Risk of BF was analyzed using stratified cumulative incidences and multiple cause-specific Cox regression using competing risk assessment. Median follow-up was 10 years (95% CI: 9.5⁻10.2). In total, 246 patients (41.6%) had low and 346 patients (58.4%) had high 5hmC level. No significant association was found between 5hmC level or ERG expression and time to BF (p = 0.2 and p = 1.0, respectively). However, for men with ERG negative tumors, high 5hmC level was associated with increased risk of BF following RP (p = 0.01). In multiple cause-specific Cox regression analyses of ERG negative patients, high 5hmC expression was associated with time to BF (HR: 1.8; 95% CI: 1.2⁻2.7; p = 0.003). In conclusion, high 5hmC level was correlated with time to BF in men with ERG negative PCa, which is in accordance with previous results.
AB - This study aimed to validate whether 5-hydroxymethylcytosine (5hmC) level in combination with ERG expression is a predictive biomarker for biochemical failure (BF) in men undergoing radical prostatectomy (RP) for prostate cancer (PCa). The study included 592 PCa patients from two consecutive Danish RP cohorts. 5hmC level and ERG expression were analyzed using immunohistochemistry in RP specimens. 5hmC was scored as low or high and ERG was scored as negative or positive. Risk of BF was analyzed using stratified cumulative incidences and multiple cause-specific Cox regression using competing risk assessment. Median follow-up was 10 years (95% CI: 9.5⁻10.2). In total, 246 patients (41.6%) had low and 346 patients (58.4%) had high 5hmC level. No significant association was found between 5hmC level or ERG expression and time to BF (p = 0.2 and p = 1.0, respectively). However, for men with ERG negative tumors, high 5hmC level was associated with increased risk of BF following RP (p = 0.01). In multiple cause-specific Cox regression analyses of ERG negative patients, high 5hmC expression was associated with time to BF (HR: 1.8; 95% CI: 1.2⁻2.7; p = 0.003). In conclusion, high 5hmC level was correlated with time to BF in men with ERG negative PCa, which is in accordance with previous results.
KW - 5hmC
KW - ERG
KW - Predictive biomarkers
KW - Prostate cancer
KW - Radical prostatectomy
UR - http://www.scopus.com/inward/record.url?scp=85062398558&partnerID=8YFLogxK
U2 - 10.3390/ijms20051025
DO - 10.3390/ijms20051025
M3 - Journal article
C2 - 30818754
SN - 1661-6596
VL - 20
JO - International Journal of Molecular Sciences
JF - International Journal of Molecular Sciences
IS - 5
M1 - 1025
ER -