Objectives: We intend to analyze the prognostic value of positive surgical margins depending on their number and location in pT2 patients.
Methods: We analyze 448 (34.3%) patients with positive surgical margins from a series of 1,310 T1-T2 patients treated with radical prostatectomy between 1989-2012. Of them 164 are pT2 (+). 119 (72.6% ) have unifocal affectation (41 (34.5%) unifocal in right lobe; 35 (29.4%) unifocal in left lobe, 40 (33.6%) unifocal in apex, 3 (2.5% ) unifocal proximal) and 45 (27.4%) multifocal involvement.
Results: Unifocal and multifocal pT2(+)patients have not evidenced significant differences in any of the clinicopathologic variables compared. However the BPFS at 5 and 10 years is significantly worse in the multifocal group, (p<0.000) In the BPFS multivariate study of 164 pT2(+ )influential variables are: multifocal involvement (HR: 3.4; 95%IC 1.7-6.9 p<0.000) and PSA (HR: 1.03; 95%IC 1.02-1.05 p<0.000), being PSA >15 ng/ml )HR: 3.7; 95%IC 2.1-6.6 p<0.000 ( the best cut-off point. Risk groups: Using the independent influence variables, the best model (using Cox models ) includes two risk groups: Group 1 (0 variables): They are pT2(+) with unifocal affectation and PSA<15 ng/ml, (63%). Their BPFS are 81±4% and 77±4% (5 and 10 years). Grupo 2 (1-2 variables): They are pT2 (+) with multifocal involvement, PSA> 15 ng/ml or both of them, (37%). Their BPFS are 46±6% and 26±7% (5 and 10 years). The BPFS differs significantly between the two groups (p<0.000). The Group 1 BPFS is similar to the pT2 (-) patients, (p:0.242). The Group 2 BPFS is similar to the pT3(+) patients, (p:0.637). The model explained significantly better the BPFS than any of the individual variables analyzed.
Conclusions: In pT2(+) patients the prognosis is significantly worse in multifocal involvement. In addition two groups of patients can be clearly distinguished from the BPFS point view according to their influential variables. The data suggest that since the prognostic point view the second group is understaged while the first is overstaged.