Abstract 5298: Lung cancer risk and scarring on imaging and histology in the National Lung Screening Trial

Abstract Background: The contribution of pulmonary scars to lung cancer development and the degree to which lung cancers cause a scarring response are unclear. Also unknown is how lung scarring impacts lung cancer screening. Methods: We evaluated associations between scarring and lung cancer in the National Lung Screening Trial (NLST), a lung cancer screening trial among current or former, heavy smokers, 55-74 years-old. Baseline scarring (presence vs. absence) on screening low dose computed tomography (LDCT) scan was assessed at baseline (T0). Associations of T0 scarring with screen-detected lung cancers and with interval-detected lung cancers missed on screening within 3 years of T0 screen were analyzed using multinomial logistic regression. Cox proportional hazards models were used to analyze the relationship between T0 scarring and incident lung cancers diagnosed >3 years after T0. Regression models included age, sex, race, smoking history, chronic obstructive pulmonary disease, history of pneumonia, and family history of lung cancer. A thoracic pathologist (first author) evaluated lung cancer pathology slides from the Lung Screening Study (LSS) subset of NLST for scar grade (none, sparse, dense) and maturity (none, immature, intermediate, mature). Associations between T0 scarring on LDCT and histological scarring were examined by logistic regression. Results: NLST’s LDCT arm enrolled 26,722 participants (65% from the LSS). T0 scars were present in 132 (22%) screen-detected, 12 (29%) interval-detected, and 94 (26%) incident lung cancer cases. T0 scarring did not increase or decrease screen-detection of cancers [odds ratio (OR) 95% CI: 1.03 (0.84-1.26)]. However, scarring might increase the chance of an interval-detected cancer [OR (95% CI): 1.54 (0.76-3.12)]. After screening stopped, T0 scarring was associated with increased incident lung cancer risk [hazard ratio (HR) (95% CI): 1.27 (1.00-1.62); P=0.048]. Pathology slides were available for 258 (38%) lung cancers in LSS. Lung scarring was found in 172 (67%) of these cancers with 58 (22%) being characterized as mature scars. On microscopic review, scars were found in 80 (66%) ADC, 46 (82%) squamous cell carcinomas, and 20 (51%) bronchioloalveolar carcinomas. Microscopic scarring tended to be more frequent among cases with T0 scarring than those without T0 scarring (75% vs. 64%; P=0.10) [OR (95% CI): 1.89 (0.98-3.86)]. Conclusion: The association between T0 scarring and incident lung cancer over a period of more than 3 years is consistent with an etiologic contribution of scarring to development of lung cancer. The relationship between T0 scarring and scarring on microscopic evaluation suggests that scarring preceded the cancer, further supporting an etiologic relationship. Finally, the borderline association of T0 scarring and interval cancers suggests that scarring may decrease the sensitivity of screening. Citation Format: Alison L. Van Dyke, Christine D. Berg, Neil E. Caporaso, Hormuzd A. Katki, Anil K. Chaturvedi, Eric A. Engels. Lung cancer risk and scarring on imaging and histology in the National Lung Screening Trial [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 5298. doi:10.1158/1538-7445.AM2017-5298