Abstract
Purpose/Objective(s)To evaluate the incidence of severe radio induced toxicity on a nation-wide scale (in France).Materials/MethodsUsing data from the French national health insurance database, we identified patients treated for nonmetastatic prostate cancer with external beam radiation therapy (EBRT) from April 2010 to December 2011. In this population, we assessed the incidence of late severe toxicity, defined by the occurrence of a complete hospitalization for radiation proctitis or hemorrhage or ulcer of the anus or rectum due to radiation, up to December 2012. Predictive factors for late toxicity, including the use of intensity modulated radiation therapy (IMRT), age of the patient, hormonal therapy, antecedent of prostatectomy, and comorbidities, were studied. Multivariate analysis was performed using Poisson mixed models.ResultsThe study populations comprised 8064 patients treated with EBRT for prostate cancer in a public hospital and 12,657 patients treated in a private setting. The cumulative probability of occurrence of a complete hospital stay for radiation proctitis was estimated to be 0.6% (range 0.4%-0.7%) at 1 year and 1.7% (range 1.3%-2.0%) at 2 years in the cohort “EBRT in public hospitals” and 0.4% (range 0.3%-0.5%) at 1 year and 1.4% (range 1.2%-1.7%) at 2 years in the cohort “EBRT in private settings.” IMRT was used in 20.8% of the patients in a public hospital and 6.8% in those treated in private settings (P<.001). Use of antithrombotic treatment increased significantly the risk of toxicity, whereas use of IMRT decreased the risk. Patients treated after prostatectomy had also less toxicity, probably because of a lower dose delivered. Age, diabetes, and hormonal therapy were not significantly associated with late toxicity. We did not observe any significant geographical heterogeneity in the risk of toxicity inside the country.ConclusionThis population-based study showed a low incidence of severe late toxicity after EBRT of prostate cancer (1.5% at 2 year). IMRT was significantly associated with a lower incidence and should be considered as a standard of care.