Abstract

268 Background: Invasive cribriform and intraductal carcinoma (CF/IDC) portends an unfavorable prognosis for patients diagnosed with prostate cancer (CaP). Limited studies with small sample sizes have explored whether genomic classifiers are associated with IDC and/or CF status. We investigated the correlation between Decipher genomic risk score and IDC/CF status and assessed PCa transcriptomic features. Methods: We performed a retrospective review of CaP patients who had Decipher testing at a single high volume center between 2009-2020. The highest grade index lesion from radical prostatectomy specimens was identified by GU pathologists and used for Decipher testing. Genitourinary pathologists reviewed prostatectomy specimens for the presence of CF and IDC features. Patients were divided into three groups based on pathologic features, absent CF/IDC (CF-/IDC-), CF positive only (CF+/IDC-), and CF/IDC positive (CF+/IDC+). Categorical clinical, genomic, and pathologic variables were assessed using the Pearson Chi-Square test, quantitative with the Kruskal-Wallis test. Multivariable logistic regression was used to identify predictors of high-risk Decipher GC scores. The Kaplan-Meier method with log-rank was used to compare biochemical recurrence free survival. Differential gene expression and gene network analysis was used to identify genes and pathways associated with IDC/CF features. Results: 463 patients were included with a median follow-up of 25 months. Patients who were CF+/IDC+ had higher GC scores (CF+/IDC+: 0.77 vs. CF+/IDC-: 0.71 vs. CF-/IDC-: 0.61, p<0.001). Patients who were CF+/IDC+ had a higher percentage of Gleason grade group >3 (CF+/IDC+: 79% vs. CF+/IDC-: 52% vs. CF-/IDC-: 52%, p<0.001). On multivariate logistic regression, predictors of high-risk GC score were presence of CF+/IDC+ features on final pathology (OR: 3.94, p<0.001) and pathologic Gleason grade group >3 (OR: 1.58, p=0.04). Transcriptomic analysis revealed that the hallmark androgen response pathway was significantly upregulated in CF+/IDC+ patients (Log fold change: 15.7, p<0001). Conclusions: This is the largest series investigating the association of a clinically validated genomic classifier and pathologic features such as cribriform and intraductal carcinoma. These findings have implications for the use of genomic classifiers in settings where expert GU pathology is not readily available and in potentially unmasking adverse histology at the time of biopsy.[Table: see text]

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call