PurposeTo estimate the diagnostic accuracy of multiparametric MRI (mpMRI) for the detection of locally advanced prostate cancer (T-stage 3–4) prior to radical prostatectomy, in a multicenter cohort representing daily clinical practice. In addition, the radiologic learning curve for the detection of locally advanced disease is evaluated.MethodsPreoperative mpMRI findings of 430 patients (2012–2016) were compared to pathology results following radical prostatectomy. The diagnostic accuracy (sensitivity, specificity, PPV, and NPV) for the detection of locally advanced disease was calculated and compared for all years separately, to evaluate the presence of a radiological learning curve.ResultsOf all 137 patients with locally advanced disease, 62 patients were preoperatively detected with mpMRI [sensitivity 45.3% (95% CI 36.9–53.6%), specificity 75.8% (CI 70.9–80.7%), PPV 46.6% (CI 38.1–55.1%), and NPV 74.7% (CI 69.8–79.7%)]. The diagnostic accuracy did not improve significantly over time (sensitivity p = 0.12; specificity p = 0.57).ConclusionsIn daily clinical practice, the diagnostic accuracy of mpMRI for the detection of locally advanced prostate cancer remains limited. It, therefore, seems questionable whether mpMRI is adequate to guide preoperative decision-making. No significant radiologic learning curve for the detection of locally advance disease was observed.

SamenvattingIn een multicenter, retrospectieve analyse werd de diagnostische accuratesse van multiparametrische MRI (mpMRI) onderzocht voor de lokale stadiëring van prostaatcarcinoom. 430 patiënten die een preoperatieve mpMRI ondergingen tussen 2012 en 2016, en vervolgens een robotgeassisteerde laparoscopische radicale prostatectomie (RARP), werden geïncludeerd. De uitslag van het histologiepreparaat werd vergeleken met de lokale stadiëring in het mpMRI-verslag van de radioloog. Bij 15,1% (65/430) van de patiënten leverde de mpMRI geen nieuwe informatie op (rT0 of rTx). De sensitiviteit van de mpMRI voor (minimale) kapseldoorbraak (pT3a/b-pT4) was 45,6% en de specificiteit 75,5%. Bij 17,2% van de patiën-ten werd onterecht een orgaanbeperkt prostaatcarcinoom (rT2a-c) beschreven, terwijl in het histologiepreparaat minstens sprake was van pT3a-prostaatcarcinoom. Bij 16,7% van de patiënten werden op de mpMRI onterecht aanwijzingen voor kapseldoorbraak (rT3a) gevonden, terwijl deze er histologisch niet was (pT2a-b). De conclusie luidt dat de mpMRI als diagnostische test voor de lokale stadiëring van prostaatkanker, in deze studie, matig presteert.Trefwoorden mpMRI · kapseldoorgroei · prostaatkanker · radicale prostatectomie · stadiëring Diagnostic accuracy of multiparametric magnetic resonance imaging (mpMRI) for local staging of prostate cancer in a multicenter setting Abstract The diagnostic accuracy of mpMRI for local staging was assessed. This study concerns a multicenter, retrospective analysis of 430 patients who underwent mpMRI preoperatively and robot assisted radical prostatectomy thereafter between 2012-2016. Histology of the radical prostatectomy specimen was compared to local staging as assessed on mpMRI. The mpMRI did not contribute in 15.1% (65/430) (rT0 or rTx). The mpMRI respectively showed a sensitivity of 45.6% and specificity of 75.5% with regard to (minimal) extracapsular extension (pT3a/b-pT4). In 17.2% of all patients mpMRI reported an organ confined disease (rT2a-c), while the histological specimen showed a pT3a prostate cancer. For 16.7% mpMRI's showed an false indication of extraprostatic growth (rT3a), while it was histologically not demonstrated (pT2a-b).In conclusion, as a diagnostic examination tool of local staging for prostate cancer mpMRI performs moderate.

Therefore, to determine the risk of disease and thus trying to Stratify it is essential in these tumors. Thus, various tools for predicting postoperative evolution have been developed and subsequently validated in several studies. In 2011, the University of California, San Francisco proposed the Cancer of the Prostate Risk Assessment Score (CAPRA-S), which uses clinical and pathological information in order to predict progression of an easy and reasonable way. The aim of the present study is to perform external validation to determine whether the Cancer of the Prostate Risk Assessment score (CAPRA-S) predicts biochemical recurrence, metastasis and death for prostate cancer after radical prostatectomy in Latin American population.METHODS: 216 patients were studied. Probability of score to predict biochemical recurrence after radical prostatectomy was analyzed by Cox proportional method. Biochemical recurrence, metastasis and cancer specific free survival was determinate by Kaplan method. Exactitude of CAPRA-S score to predict biochemical recurrence, metastasis and death by prostate cancer was made in accordance with Harrells concordance index.RESULTS: median follow up was 74 months. Biochemical recurrence index increase proportionally with the increment of CAPRA-S score. In the stratification of patients in low, intermediate or high risk, biochemical recurrence free rate were 85%, 54% and 4% respectively. Concordance index (C-Index) to biochemical progression, metastasis and death by prostate cance were 0.85, 0.90 and 0.90 respectively.CONCLUSIONS: CAPRA-S score is an easily applicable tool and had high predictive exactitude to determinate biochemical recurrence, metastasis and death by prostate cancer in our population. Concordance Index in these variables was higher than 0.85.