Association Between Travel Distance and Use of Postoperative Radiation Therapy for Incompletely Resected Non-Small Cell Lung Cancer

Abstract

Postoperative radiation (PORT) is recommended by consensus guidelines for incompletely resected stage II-III non-small cell lung cancer (NSCLC) to optimize local control. Prior data across disease sites has suggested disparities exist in receipt of beneficial adjuvant therapies based on sociodemographic barriers including geography. We sought to explore if receipt of PORT in margin positive locally advanced NSCLC was related to travel distance from the treating facility.We identified patients with pathologic stage II-III NSCLC within the National Cancer Data Base who underwent treatment with upfront surgery (lobectomy or pneumonectomy) found to have microscopic (R1) or macroscopic (R2) residual tumor. Only patients coded as receiving adjuvant external-beam PORT (50-74 Gy) or no radiation were included. Multivariable logistic regression was utilized to determine factors associated with PORT receipt. Cox regression was used to identify predictors of survival.We identified 7,270 pathologic stage II-III NSCLC patients who underwent R1/R2 resections from 2004-2015; of these, 36% received PORT. Median overall survival in these patients was significantly improved with the addition of PORT, from 22.3 months to 32.8 months and this persisted when stratified by distance traveled (see table 1). Of the patients who did not receive PORT, 43.4% did receive adjuvant chemotherapy and < 1% received immunotherapy. Reason for why PORT was not given was largely unknown, but stated reasons included: contraindicated due to risk factors (3.6%) and patient refusal (4.2%). In contrast, 79.1% of PORT patients also received chemotherapy. Half of PORT patients resided within 10 miles of treatment facility. Progressively father travel distance was associated with decreased likelihood of PORT (OR 0.89, P = 0.002). Other factors associated with decreased PORT receipt included older age and decreased level of education. On Cox regression, higher co-morbidity score was predictive of death, while receipt of PORT was protective (HR 0.87, P < 0.001).For NSCLC patients with margin positive disease after surgery, increasing distance was associated with decreased likelihood of receiving PORT, despite being routinely recommended in this setting and proven survival benefit irrespective of geography in our patient cohort. Decreased likelihood to receive PORT with increasing travel distance suggests location may play a role in the decision to undergo PORT for these patients with adverse pathology.