Henry S Park1, Roy H Decker2, Lynn D Wilson2, James B Yu2. 1. Department of Therapeutic Radiology, Yale University School of Medicine, New Haven, CT. Electronic address: henry.park@yale.edu. 2. Department of Therapeutic Radiology, Yale University School of Medicine, New Haven, CT.
Abstract
BACKGROUND: Although there is no proven survival benefit of prophylactic cranial irradiation (PCI) for patients with locally advanced (LA) non-small-cell lung cancer (NSCLC), some speculate that PCI might be helpful for certain subpopulations at higher risk of brain metastases (< 60 years, adenocarcinoma, or stage IIIB). In this study we evaluated the effect of PCI on survival among these high-risk LA-NSCLC patients on a national scale. MATERIALS AND METHODS: Using the Surveillance, Epidemiology, and End Results database, we included all adult patients with primary stage III NSCLC, diagnosed from 1988 to 1997 (years during which PCI was recorded) with follow-up until 2008. The Kaplan-Meier estimator, log-rank test, and Cox proportional hazard regression were used to evaluate the survival effect of PCI. Sequential landmark analysis excluding patients from 1 to 6 months after diagnosis was used to account for immortal time bias. RESULTS: A total of 17,852 patients were included in the analysis, among whom 326 (1.8%) received PCI. Patients younger than 60 years and those with adenocarcinoma were significantly more likely to receive PCI. After adjustment for available covariates, there was no statistically significant survival difference between PCI and non-PCI patients (hazard ratio, 1.04; 95% confidence interval, 0.93-1.16). Similar results were found in all subgroup analyses of high-risk patients. Sequential landmark analysis suggested a potential survival detriment associated with PCI when analyzing only patients who survived at least 3 months after diagnosis. CONCLUSION: Our population-based analysis suggested no overall survival benefit of PCI for LA-NSCLC patients, even among a group of patients who were at higher risk for brain metastases.
BACKGROUND: Although there is no proven survival benefit of prophylactic cranial irradiation (PCI) for patients with locally advanced (LA) non-small-cell lung cancer (NSCLC), some speculate that PCI might be helpful for certain subpopulations at higher risk of brain metastases (< 60 years, adenocarcinoma, or stage IIIB). In this study we evaluated the effect of PCI on survival among these high-risk LA-NSCLCpatients on a national scale. MATERIALS AND METHODS: Using the Surveillance, Epidemiology, and End Results database, we included all adult patients with primary stage III NSCLC, diagnosed from 1988 to 1997 (years during which PCI was recorded) with follow-up until 2008. The Kaplan-Meier estimator, log-rank test, and Cox proportional hazard regression were used to evaluate the survival effect of PCI. Sequential landmark analysis excluding patients from 1 to 6 months after diagnosis was used to account for immortal time bias. RESULTS: A total of 17,852 patients were included in the analysis, among whom 326 (1.8%) received PCI. Patients younger than 60 years and those with adenocarcinoma were significantly more likely to receive PCI. After adjustment for available covariates, there was no statistically significant survival difference between PCI and non-PCI patients (hazard ratio, 1.04; 95% confidence interval, 0.93-1.16). Similar results were found in all subgroup analyses of high-risk patients. Sequential landmark analysis suggested a potential survival detriment associated with PCI when analyzing only patients who survived at least 3 months after diagnosis. CONCLUSION: Our population-based analysis suggested no overall survival benefit of PCI for LA-NSCLCpatients, even among a group of patients who were at higher risk for brain metastases.