A retrospective analysis of the results of p(65) + Be neutrontherapy for the treatment of prostate adenocarcinoma at the cyclotron of Louvain-la-Neuve. Part I: Survival and progression-free survival

PURPOSE: To retrospectively evaluate survival, progression-free survival (PFS) and biological response in a series of patients irradiated with mixed neutron/photon beams for locally advanced prostate cancer in our institution. PATIENTS AND METHODS: Three hundred and eight patients were treated between January 1990 and December 1996. Fifty-five of these were recruited for pT3 or pN1 tumors after radical prostatectomy. Neoadjuvant androgen deprivation was given in 106 patients. The treatment protocol consisted of a mixed photon/neutron irradiation in a two-to-three proportion, up to a total equivalent dose of 66 Gy (assuming a clinical RBE value of 2.8). Pre- and post-treatment PSA determinations were available in practically all cases. Study endpoints were overall survival (OAS) and progression-free survival (PFS). The Cox proportional hazard regression model was used to investigate the prognostic value of baseline characteristics on survival and progression-free survival were a progression was defined as local, regional, metastatic or biological progression. Mean age was 69 years (49-86); mean pretreatment PSA was 15 (0.5-330) in all patients and 14 (0.5-160) in those receiving neoadjuvant hormonotherapy; seven patients only had an initial PSA < or = 4 ng/mL; 15% were T1, 46% were T2, 28% were T3 or pT3 and 4% were T4 (7% unspecified); WHO grade of differentiation was I in 38%, II in 38% and III in 14% (5% unspecified). RESULTS: The median follow-up was 2.8 years (0-7.8). Five-year overall survival (OAS) was 79% (95% CI: 71-87%) and 5-year progression-free survival (PFS) was 64% (95% CI: 54-74%) for the entire series. PFS in patients with an initial PSA > or = 20 ng/mL was the same. PFS could be predicted by two optimal Cox regression models, one including histological grade (p = 0.003) and initial PSA (p = 0.0009) as cofactors, the other including histological grade (p = 0.003) and T stage (p = 0.02). The main prognostic factors for overall survival were PSA and age. Biological responses with PSA < 1.5 ng/mL, < 1 ng/mL and < 0.5 ng/mL at any time after treatment were documented in 70%, 61% and 47% of the patients, respectively. CONCLUSION: Five-year OAS was 79%, PFS was 64%, and biological response was 70% for prostate cancer patients treated with mixed photon/neutron beams as applied at Louvain-la-Neuve, which are good results as compared with the literature. The usual prognostic factors were confirmed.