The effect of smoking on outcome following external radiation for localized prostate cancer

Purpose: We investigated whether a smoking habit affects biochemical and survival outcome after curative external beam radiation therapy (EBRT) for localized prostate cancer. Materials and Methods: The study population comprised 601 men treated with curative EBRT between 1994 and 1997 who had a smoking history available. Pretreatment prognostic factors were examined and high risk was defined as any of prostate specific antigen greater than 20, Gleason greater than 7 or stages T3-4. Biochemical outcome (bNED) was assessed by American Society for Therapeutic Radiology and Oncology, and Houston criteria. Biochemical, clinical, prostate cancer and nonprostate cancer death rates were examined by univariate and multivariate statistics. Results: Of the men 15% were current smokers, 55% were former smokers and 31% were nonsmokers. Current smokers were younger than former smokers or nonsmokers by a, mean of 2.5 years (p<0.001). Current smokers had higher risk cancers than former smokers or nonsmokers (high risk 60%, 40% and 43%, respectively, p=0.017). Five-year bNED rates for smokers were significantly worse than for former smokers or nonsmokers (55%, 69% and 73%, p=0.01 and 0.0019, respectively). Median followup was 59 months. Multivariate analysis confirmed smoking as an independent adverse factor for bNED (p=0.013) even when controlling for prostate specific antigen (p<0.0001), Gleason score (p<0.0001), stage (not significant), radiation dose (not significant) and neoadjuvant hormone use (p=0.0014). Local and metastatic failure did not differ among the groups. Prostate cancer specific mortality was nonsignificantly worse in smokers but overall mortality was much greater. Conclusions: Smokers present with higher risk prostate cancers. Outcomes following EBRT are poorer, even when accounting for differences in known pretreatment factors.