Case-control study of prophylactic cranial irradiation in nonmetastatic non-small cell lung cancer.

Abstract

7050 Background: Prophylactic cranial irradiation (PCI) reduces the incidence of brain metastases in NSCLC patients after primary therapy, but its impact on survival is uncertain. We report on the largest study of survival in patients treated with and without PCI for non-small cell lung cancer (NSCLC). Methods: We reviewed 17 Surveillance, Epidemiology and End Results (SEER) registries for a retrospective study on patients who had PCI as part of their primary treatment for NSCLC from 1988 - 97. Cases were limited to those with non-metastatic (Stage I-III) NSCLC. To balance the cohorts, we matched each PCI patient with four non-PCI patients on stage, histology, race and sex. Associations between treatment type, clinical factors, and demographics were assessed using the Chi-squared test. Survival time was calculated as the number of months from diagnosis to the date of death. Survival was censored as of the last month when patients were known to be alive. Overall (OS) and cancer cause-specific survival (CSS) were investigated using the Kaplan-Meier, competing risks, Cox proportional hazards, and log-rank tests. Results: We found 472 PCI matched to 1,888 non-PCI patients. Characteristics were balanced across groups: race (p = 1.00), sex (p = 0.95), histology (p = 1.00), stage (p = 1.00), and surgery (p = 0.81). PCI group was younger, median age 64 vs 68 (p < 0.01). PCI vs no PCI median OS was 8 vs 10 months (p < 0.01). OS was 14% vs 28% at 2 years and 5% vs 12% at 5 years, PCI vs no PCI respectively (p < 0.01). Stage III OS was also different; 10% vs 21% at 2 years, PCI vs no PCI respectively (p < 0.01). Median CSS was the same at 9 months in both groups. Median follow-up was 14 years. Conclusions: PCI was not associated with improved OS or CSS in these NSCLC patients, and PCI may have a detrimental effect on OS. In limited-stage small cell lung cancer, a retrospective SEER analysis during the 1988 – 97 period showed a survival benefit in treated patients with PCI, which has been confirmed with prospective studies. To date, 4 prospective trials examining PCI for NSCLC have shown a reduced incidence of brain metastases, but the effect on survival is unclear. Further investigation is needed to determine whether PCI for NSCLC increases the risk of other causes of death.