Tuberculosis-Associated Scar Carcinoma in Lung Cancer: Clinicopathological and Radiological Features of a Fibrotic-Cavitary Phenotype in a Retrospective Observational Cohort.

BACKGROUND: Scar carcinoma represents a distinct subtype of lung malignancy developing in areas of chronic pulmonary fibrosis, inflammation, and structural remodeling, frequently associated with previous pulmonary tuberculosis (TB). The present study aimed to evaluate the radiological, clinical, inflammatory, and histopathological characteristics associated with the scar carcinoma phenotype in patients with lung cancer (LC) and previous TB-related pulmonary abnormalities.

METHODS: A retrospective observational cohort study conducted between February 2020 and December 2025 included 844 patients diagnosed with lung cancer. The scar carcinoma phenotype was operationally defined by the coexistence of: (1) confirmed lung cancer, (2) post-tuberculous structural pulmonary abnormalities on thoracic imaging, and (3) clinical history compatible with prior pulmonary TB. Associations between the scar carcinoma phenotype and clinicopathological variables were evaluated using Pearson's Chi-square and Fisher's exact tests. Binary logistic regression analysis was performed to identify independent predictive factors associated with scar carcinoma. Receiver operating characteristic (ROC) curve and precision-recall curve analyses were additionally performed.

RESULTS: Post-TB sequelae were identified in 58.2% of patients, while active TB was present in 7.8% of cases. Adenocarcinoma represented the predominant histopathological subtype (63.3%). Fibrotic/interstitial/bronchial abnormalities (67.7%), cavitary/destructive lesions (69.0%), atelectatic/retractile changes (65.4%), and infectious/inflammatory pulmonary abnormalities (60.4%) were highly prevalent. Significant associations were identified between scar carcinoma and TB sequelae (χ2 = 811.850, p < 0.001), adenocarcinoma histology (χ2 = 655.545, p < 0.001), infectious/inflammatory changes (χ2 = 635.168, p < 0.001), cavitary/destructive lesions (χ2 = 508.347, p < 0.001), fibrotic/interstitial abnormalities (χ2 = 539.895, p < 0.001), and atelectatic/retractile changes (χ2 = 597.346, p < 0.001). Logistic regression identified haemoptysis (OR = 0.651 (95% CI: 0.486-0.871), p = 0.005) and pulmonary opacities and/or condensation (OR = 1.343 (95% CI: 1.014-1.779), p = 0.040) as independent predictive factors. ROC analysis demonstrated moderate predictive performance (AUC = 0.703).

CONCLUSIONS: TB-associated pulmonary remodeling was strongly associated with the scar carcinoma phenotype, consistent with an associative role of chronic inflammation, fibrosis, and post-TB structural damage in lung carcinogenesis; however, causal inferences cannot be drawn from this retrospective observational design.