Context A previous randomized trial demonstrated that adding bevacizumab to carboplatin and paclitaxel improved survival in advanced non–small cell lung cancer (NSCLC). However, longer survival was not observed in the subgroup of patients aged 65 years or older.
Objective To examine whether adding bevacizumab to carboplatin and paclitaxel chemotherapy is associated with improved survival in older patients with NSCLC.
Design, Setting, and Participants Retrospective cohort study of 4168 Medicare beneficiaries aged 65 years or older with stage IIIB or stage IV non−squamous cell NSCLC diagnosed in 2002-2007 in a Surveillance, Epidemiology, and End Results (SEER) region. Patients were categorized into 3 cohorts based on diagnosis year and type of initial chemotherapy administered within 4 months of diagnosis: (1) diagnosis in 2006-2007 and bevacizumab-carboplatin-paclitaxel therapy; (2) diagnosis in 2006-2007 and carboplatin-paclitaxel therapy; or (3) diagnosis in 2002-2005 and carboplatin-paclitaxel therapy. The associations between carboplatin-paclitaxel with vs without bevacizumab and overall survival were compared using Cox proportional hazards models and propensity score analyses including information about patient characteristics recorded in SEER-Medicare.
Main Outcome Measure Overall survival measured from the first date of chemotherapy treatment until death or the censoring date of December 31, 2009.
Results The median survival estimates were 9.7 (interquartile range [IQR], 4.4-18.6) months for bevacizumab-carboplatin-paclitaxel, 8.9 (IQR, 3.5-19.3) months for carboplatin-paclitaxel in 2006-2007, and 8.0 (IQR, 3.7-17.2) months for carboplatin-paclitaxel in 2002-2005. One-year survival probabilities were 39.6% (95% CI, 34.6%-45.4%) for bevacizumab-carboplatin-paclitaxel vs 40.1% (95% CI, 37.4%-43.0%) for carboplatin-paclitaxel in 2006-2007 and 35.6% (95% CI, 33.8%-37.5%) for carboplatin-paclitaxel in 2002-2005. Neither multivariable nor propensity score–adjusted Cox models demonstrated a survival advantage for bevacizumab-carboplatin-paclitaxel compared with carboplatin-paclitaxel cohorts. In propensity score–stratified models, the hazard ratio for overall survival for bevacizumab-carboplatin-paclitaxel compared with carboplatin-paclitaxel in 2006-2007 was 1.01 (95% CI, 0.89-1.16; P = .85) and compared with carboplatin-paclitaxel in 2002-2005 was 0.93 (95% CI, 0.83-1.06; P = .28). The propensity score–weighted model and propensity score–matching model similarly failed to demonstrate a statistically significant superiority for bevacizumab-carboplatin-paclitaxel. Subgroup and sensitivity analyses for key variables did not change these findings.