Improved Survival of Stage I Non-Small Cell Lung Cancer: A VA Central Cancer Registry Analysis.

INTRODUCTION: The combined impact of advances in diagnosis and treatment of stage I NSCLC has not been assessed comprehensively. To define the survival impact of modern staging and treatment techniques for clinical stage I NSCLC, the Veterans Administration Central Cancer Registry, a database of U.S. veterans in whom the disease was diagnosed in the Veteran's Health Administration, was queried. From this database, patients who had stage I NSCLC diagnosed from 2001 to 2010 and were treated with either surgery or radiation were identified. METHODS: Overall survival (OS) and lung cancer-specific survival were determined. Propensity score matching and Cox multivariate analysis were used to adjust for baseline patient characteristics. RESULTS: A total of 11,997 patients were identified. The 4-year OS rate increased from 38.9% to 53.2% from 2001 to 2010 for all patients. Positron emission tomography and endobronchial ultrasound did not improve OS. Survival of radiated patients improved from 12.7% to 28.5%. The introduction of stereotactic body radiation therapy (SBRT) significantly improved OS (hazard ratio [HR] = 0.60, 95% confidence interval [CI]: 0.54-0.68) and lung cancer-specific survival (HR = 0.39, 95% CI: 0.32-0.46) compared with conventionally fractionated radiation. The 4-year OS rate also improved after surgery (from 51.5% to 66.5%). This increase was associated with use of adjuvant chemotherapy, increased use of video-assisted thoracoscopic surgical procedures, and decreased pneumonectomy rates, with similar survival between open and video-assisted thoracoscopic surgical procedures. OS after lobectomy was superior to that after sublobar resection (HR = 0.82, 95% CI: 0.75-0.89). In the era of available SBRT (2008-2010), 4-year OS was not significantly different after sublobar resection or lobectomy for medically unfit patients (Charlson comorbidity index = 2) (55.4% and 58.1%, respectively; p = 0.69) but was significantly worse for fit patients (Charlson comorbidity index = 0-1) undergoing sublobar resection (55.5% and 68.0%, respectively; p < 0.001). OS (HR = 0.36, 95% CI: 0.35-0.38) and lung cancer-specific survival (HR = 0.31, 95% CI: 0.29-0.33) were improved after surgery as compared with after radiation, with the improvement maintained on matched comparison of lobectomy and SBRT. CONCLUSIONS: OS increased in veterans with a diagnosis of stage I NSCLC from 2001 to 2010; the increase was coincident with improved radiation and surgical techniques.

Duke Scholars

Author Joseph Kamel Salama Radiation Oncology

Author Matthew Jacob Boyer Radiation Oncology

Author Christina D Williams Medicine, Medical Oncology

Author David Harold Harpole Jr. Surgery, Cardiovascular and Thoracic Surgery

Author Mark William Onaitis Surgery, Cardiovascular and Thoracic Surgery

Author Michael John Kelley Medicine, Medical Oncology