[The significance of PSA density as predictive factor in Afro-Caribbean patients eligible for active surveillance by the French protocol criteria]. / Intérêt de la densité du PSA comme facteur prédictif chez les patients afro-caribéens éligibles à la surveillance active du cancer de prostate selon les critères du protocole français.

ABSTRACT

OBJECTIVES: Study the interest of addition of PSA density (PSAD), to the selection criteria of the French protocol for inclusion patients Afro-Caribbean on active surveillance prostate cancer. METHODS: A retrospective analysis of 1505 patients who had, in turn, a radical prostatectomy for cancer between 2000 and 2012, in a single reference center. One hundred and forty-one patients was eligible, at the time of their diagnosis, for active surveillance by the criteria of the French protocol. This population was divided into 2 groups according to the histological analysis of the prostatectomy specimen confirmed indolent cancer or overturned. The median PSAD of each group was calculated to be compared. Secondarily, the most discriminating PSAD was investigated by the method of ROC after constitution tables intrinsic validity in this population. This threshold has secondary conducting a comparative analysis of the underestimation of cancer in terms of aggressiveness and/or extension between patients selected according to the criteria of the French protocol and "on-selected" patients according to these criteria and their PSAD. RESULTS: Of the 141 patients identified for analysis, histological examination of the prostatectomy specimen has to show that 42 patients (29.7 %) were actually more aggressive cancer (20.6 % of Gleason ≥ 7), wider (4.2 % ≥ pT3) or larger and more aggressive (4.9 %) than foreshadowed criteria French protocol. The median PSAD these 42 patients were significantly higher than the median PSAD patients correctly estimated (0.18 vs. 0.14, p-value=0.046). The application of the most discriminating threshold 0.15 ng/ml/cm(3) in this population allowed to significantly improve the selection of candidates of the 79 "on-selected" patients, six (20.2 %) were actually more aggressive cancer (13.9 % of Gleason ≥ 7), wider (2.5 % ≥ pT3) or larger and more aggressive (3.8 %). CONCLUSION: The criteria for the French protocol for active surveillance, applied to the Caribbean population underestimate 29 % of non-latent cancers. Adjuvants criteria that must be inexpensive, sensitive and specific seem necessary in this population. A PSAD<0.15 ng/ml/cm(3) could be one of these criteria.