Trimodality Therapy For Locally Advanced Non-Small Cell Lung Cancer (Nsclc): Predictors Of Surgical Drop Out

The standard of care for stage III NSCLC patients is concurrent chemotherapy and radiation therapy (bimodality). The addition of surgery (trimodality) improves local control but a significant portion of patients fails to make it to surgery. We sought to determine if any pretreatment characteristics predicted surgical drop out. Between January 2000 and December 2010, 230 patients were treated curatively for stage III NSCLC. A multi-disciplinary team prior to treatment evaluated all patients where 114 of the 230 patients were recommended to undergo surgery following the completion of chemoradiation. Almost all patients received a definitive dose of radiation therapy with a median dose of 61.2 Gy. Concurrent radiation was administered with weekly carboplatin/paclitaxel most commonly utilized. Trimodality patients underwent restaging studies to confirm (1) the absence of disease progression and (2) the clearance of mediastinal lymph nodes prior to their surgery. After concurrent therapy +/- surgery, patients received 2 additional cycles of systemic chemotherapy. The Kaplan-Meier product limit method provided estimates of overall and disease-free survival; the log-rank test statistic indicated whether the stratifications of these estimates by selected prognostic factors were significant. Fisher's exact test was used to test contingency tables. Sixty of the 114 (53%) initially recommended to undergo trimodality therapy actually underwent surgical resection (48 lobectomies, 11 pneumonectomies, and 1 incomplete record). Surgery was deferred due to the development of distant metastases in 17 patients based on restaging studies, persistent mediastinal disease based on mediastinoscopy in 9 patients, 5 patients remained unresectable, co-morbid condition prohibited surgery in 5 patients, surgery refusal in 4 patients, treatment toxicities in 4 patients, poor patient compliance in 2 patients, and unknown in 8 patients. The 3 and 5-year overall survival (OS) in the trimodality patients who had surgery were 47 and 38%, as compared to 22 and 17% for those who did not undergo surgery, respectively. This was similar to the 116 patients who were initially recommended to undergo bimodality therapy, 21 and 8%. No pretreatment variables including T stage, N stage, race, or age predicted which patients would eventually have surgery. Trimodality patients have an improved overall survival however roughly 50% of patients thought to be trimodality candidates do not make it to surgery. We were unable to find any pretreatment characteristics that predicted surgical drop out. However, the most common reasons for surgical drop out were disease progression or persistent mediastinal disease.