Cribriform pattern 4/intraductal carcinoma of the prostate and persistent prostate-specific antigen after radical prostatectomy

Abstract

Objectives: The objective of this study is to identify the effect of cribriform pattern 4 carcinoma/intraductal carcinoma of the prostate (CC/IDCP) on persistent prostate-specific antigen (PSA) levels after robot-assisted radical prostatectomy (RARP) in patients with localized prostate cancer (PCa).

Patients and methods: This retrospective study included 730 consecutive patients with localized PCa who underwent RARP at Mie University (n = 392) and Aichi Medical University (n = 338) between 2015 and 2021. Patients with clinically metastatic PCa (cN1 and cM1) and those who received neoadjuvant and/or adjuvant therapy before biochemical recurrence were excluded. We evaluated the effects of CC/IDCP on persistent PSA levels after RARP. Persistent PSA was defined as PSA level ≥0.2 ng/mL at 1 month postoperatively and consecutively thereafter. Using factors from logistic regression analysis, models were developed to predict persistent PSA levels.

Results: Approximately 6.3% (n = 46) of the patients had persistent PSA levels. Patients with biopsy CC/IDCP (bCC/IDCP) and pathological CC/IDCP (pCC/IDCP) based on RARP specimens were 11.6% (85/730) and 36.5% (267/730), respectively. Multivariate analysis of the prediction of persistent PSA levels using preoperative factors revealed that PSA density, percentage of positive cancer cores, biopsy grade group and bCC/IDCP were independent prognostic factors. Furthermore, multivariate analysis of the prediction of persistent PSA levels using postoperative factors, excluding pN1, revealed that pathological grade group, pCC/IDCP, seminal vesicle invasion and lymphovascular invasion were independent prognostic factors. In the receiver operating characteristic curve analysis for predicting persistent PSA after RARP, areas under the receiver operating characteristic curve for the model with preoperative factors, postoperative factors, including pN1, and postoperative factors, excluding pN1, were 0.827, 0.833 and 0.834, respectively.

Conclusions: bCC/IDCP predicted persistent PSA after RARP in the overall population, while pCC/IDCP predicted persistent PSA only when the pN1 population was excluded. This may be useful for predicting susceptible patients with worse outcomes.

Keywords: cribriform; intraductal carcinoma; persistent PSA; prostate cancer; radical prostatectomy.

FIGURE 1

Receiver operating characteristic curve analysis was used to evaluate the predictive value of persistent prostate‐specific antigen after robot‐associated radical prostatectomy. The area under the curve (AUC) of persistent prostate‐specific antigen after robot‐associated radical prostatectomy using the model with preoperative factors, postoperative factors, including pN1, and postoperative factors, excluding pN1, were 0.827, 0.831 and 0.834, respectively.

FIGURE 2

Distribution of scores and percentage of persistent prostate‐specific antigen (PSA) values after robot‐associated radical prostatectomy (RARP) for each model in low‐, intermediate‐, and high‐risk groups. (A) Using preoperative factors, (B) using postoperative factors, Model 1, and (C) using postoperative factors, Model 2. We defined a novel prognostic model in which patients with scores of ≥0 and <2 were defined as the low‐risk group, those with a total score of ≥2 and <4 were an the intermediate‐risk group, and those with a score of ≥4 were the high‐risk group. For all the models, the percentage of persistent PSA values after RARP was the highest in the high‐risk group. CC/IDCP, cribriform pattern 4 carcinoma/intraductal carcinoma of the prostate; GG, grade group; LVI, lymphovascular invasion; PSAD, PSA density; SV, seminal vesicle.