Nomograms to Predict the Probabilities of Adverse Features Indicated for Adjuvant Radiotherapy for Patients with Prostate Cancer That May Help to Individualize Initial Treatment Options

Abstract

It is recommended that prostate cancer patients with positive margin(s)(PM) or extra-prostatic extension(EPE) after radical prostatectomy (RP) underwent adjuvant radiotherapy(RT). There is high level evidence supporting definitive RT as the primary treatment regimen as it was shown to be equivalent with RP. Predicting adverse features after surgery may help to individualize initial treatment options. The purpose of this study is to develop nomograms through the correlation of pretreatment PSA, clinical T stage and biopsy Gleason Score(GS) with adverse features such as EPE or PM that are commonly indicated for adjuvant RT. We analyzed 1921 RP patients between January, 2007 and December, 2019 at our hospital. Patients who had a history of neoadjuvant hormonal therapy, or transurethral resection of the prostate were excluded. All patients had a preoperative PSA level, biopsy GS, pelvic MRI and other examinations for staging. PSAmax level, clinical T stage (T2a/b, T2c, T3a, T3b), and biopsy GS (5 to 6, 3+4 = 7, 4 + 3 = 7, 8, 9 to 10) were recorded as preoperative predictors. These predictors were used in multivariable logistic regression analysis based nomograms to estimate the probabilities of EPE and PM after RP, respectively. The predictive accuracy and discriminative ability were determined by concordance index (C-index) and calibration curve. Analyses were performed with statistical software. A total of 1060 patients were eligible. Median age was 66(33-82) years.10.9% of the patients underwent open RP, 89.1% received laparoscopic RP, including 18 patients who received robot-assisted laparoscopic surgery. Preoperatively, 62.6% of patients had a PSA higher than 20 ng/mL; 83.1% were organ confined disease; and 76.6% had biopsy GS≤7. Postoperatively, 30.1% had PM, 54.3% had EPE. In addition, 48.7% (429/881) of patients who were clinically diagnosed with organ-confined disease(≤T2) were found to have EPE (≥T3) after surgery. As for the Gleason Score, 11.7%(95/812) of the patients who had biopsy GS< 8 had upgraded to 8 to 10 after surgery. Nomograms were developed for the predicted probabilities of having the indications of adjuvant radiation therapy (pathologically EPE disease or PM, respectively). The calibration curve for probabilities showed good agreement between prediction by nomogram and actual observation. The C-index of the nomograms for predicting for EPE disease, or PM were 0.70 and 0.76, respectively. The risk of having one of the indications of adjuvant radiation therapy increased with increases in predictors. Using clinic-pathological information, we produced nomograms which may accurately predict the probabilities of having indications for adjuvant radiation therapy after RP. Based on these nomograms, patients at high risk of having an indication of adjuvant RT may choose to receive definitive RT to avoid toxicities and therefore improve cost-effectiveness without compromising efficacy.