Anales del Sistema Sanitario de Navarra
versión impresa ISSN 1137-6627
RINCON MAYANS, A. et al. Radical prostatectomy for high risk localized prostate cancer: Prognosis and study of influential variables. Anales Sis San Navarra [online]. 2012, vol.35, n.1, pp.9-18. ISSN 1137-6627. http://dx.doi.org/10.4321/S1137-66272012000100002.
Background. To study the biochemical progression-free survival (BPFS) achieved by a group of high risk patients in accordance with D´Amico´s classification treated with radical prostatectomy. To identify the clinical-pathological variables which are influential in biochemical progression-free survival and, if possible, use them to design a prognostic model. Material and methods. The study involves 232 patients, out of a series of 1,054, diagnosed with clinically localized prostate cancer, qualified as high risk on D´Amico´s classification (PSA>20 ng/ml or Gleason score 8-10 or T3) treated with radical prostatectomy. The BPFS is studied and the clinical-pathological variables obtained (PSA, Gleason score of the biopsy and of the piece, clinical and pathological study, unilateral or bilateral affectation, margins of the prostatectomy piece, Ki-67 expression) are analyzed to identify whether they influenced the BPFS. Contingency tables and tables for survival analysis: Kaplan-Meyer, log-rank and Cox models were used for the statistical study. Results. Descriptive study: PSA: 23.3 ng/ml (median); cGleason 2-6: 33%; 7: 13%; 8-10: 54%; T2: 58%; Bilateral affectation in the diagnostic biopsy: 59%; RNM T2: 60%; RNM T3: 40%. pGleason 2-6: 24%; 7: 28%; 8-10: 48%; pT2: 43%; pT3a: 30%; pT3b: 27%; Affected margin: 51%; N1:13%. Progression-free survival: with a mean and median follow-up of 64 months; 53% show biochemical progression. The median until progression: 42 months. Progression-free survival at 5 and 10 years is 43±3% and 26±7%. The multivariate study (Cox models) shows that the variables that are independently influential in the BPFS are the affectation of margins (HR: 3.5; 95% IC.1.9-6.7; p>0001); and Ki67 >10% (HR: 2.3; 95% IC: 1.2-4.3; P: 0.009). Risk groups: using the two influential variables and employing Cox models, three risk groups emerged as the best model: Group 1 (0 variables present); Group 2 (1 variable); Group 3 (2 variables). The progression-free survival is 69±8%; 27±6% and 18±11% at 5 years. The differences amongst the three groups are significant. Conclusion. The high risk group according to the D´Amico classification is heterogeneous in relation to biochemical progression and can be broken down into three risk groups using the two independently influential variables (affected margins and Ki67 percentage).
Palabras clave : Prostate cancer; Ki-67; Radical prostatectomy; High risk; D'Amico; Surgical margin.