Radical cystoprostatectomy (RC) is one standard treatment for muscle-invasive bladder cancer (MIBC) in male patients. Another therapeutic option is trimodal therapy. Including the prostate in the trimodal therapy radiation therapy volume is based on MIBC surgical series showing prostatic stromal (PS) involvement. Our aim was to establish the rate of pathologic PS involvement by preoperative T stage in men treated with RC for MIBC.

We conducted a retrospective review of men with MIBC treated with RC between 2006 and 2019. Electronic medical records were reviewed, and preoperative clinical staging data were collected. χ test was done to test for a statistically significant difference in the rate of prostatic involvement between clinical tumor (cT) stages. Preoperatively identified carcinoma in situ, lymph node involvement, tumor location, and urethral involvement were also analyzed to see if they conferred a higher risk of PS involvement. Multivariate analysis using multiple logistic regression was performed.

We identified 283 men with bladder cancer treated with RC. Patients with non-MIBC or incomplete medical data were excluded (n = 72). We analyzed 211 patients, and 46 (22%) had pathologic PS involvement. PS involvement by preoperative T stage was cT2 = 18%, cT3 = 23%, and cT4 = 48%. Twenty-nine (12%) patients had clinical lymph node involvement, of whom 19 (76%) had PS involvement. Thirty-four (16%) had urethral involvement, of whom 17 (50%) had PS involvement. Sixteen percent and 17% of percent of clinical T2 and T3 patients had bladder neck/trigone tumors, of whom 57% and 50% had prostatic involvement. Clinical T2 and T3 were not statistically different with regards to PS involvement (P = .385). Preoperative urethral involvement, lymph node involvement, cT4, and bladder neck/trigone location were statistically significant predictors of pathologic PS involvement (all P < .05). On multivariate analysis, only clinical urethral involvement was significant (P < .0001).

The high rate of pathologic PS involvement seen in cT2 patients and the lack of ability of cT stage to predict PS involvement support routinely treating the prostate in trimodal therapy. Patients with preoperatively identified bladder neck/trigone tumors, urethral involvement, positive lymph nodes, or prostatic involvement are a subset at even higher risk of having pathologic PS involvement.