Abstract
OBJECTIVES: To investigate variables associated to not receiving adjuvant therapy in patients with resected pathological IB-IIIA NSCLC from a national registry. METHODS: Patients who had lung resections were analyzed retrospectively for stages IB-III from 2009-2023, sourced from a multicentre registry. Uni/Multivariable logistic regression was performed, and a predictive tool was created to predict adjuvant therapy probability and cut-off values were determined through ROC analysis. The model's discrimination, performance and fit were evaluated using AUC-ROC, sensitivity/specificity and Hosmer-Lemeshow test respectively. RESULTS: Of the 427 patients analyzed, the mean age was 65.4 years and 83.1% were treated under public insurance. 38.4% of patients received adjuvant chemotherapy after surgery. Multivariable analysis identified older age (OR = 1.06, p < 0.001), histological subtype (OR = 21.3, p < 0.001), public insurance (OR = 2.65, p = 0.006), stage I-B (OR = 8.10, p = 0.002), and negative lymph node status (OR = 3.34, p < 0.001) as independent factors associated with not receiving adjuvant therapy. Using the final equation, a calculated probability > 42.5% categorizes a patient for exclusive surgical intervention. The model's performance yielded an AUC of 0.833, with a sensitivity of 90.2%. The Hosmer-Lemeshow test resulted in a p ≥ 0.05. CONCLUSIONS: Underutilization of adjuvant therapy in NSCLC is influenced by age, insurance type, and pathological factors. Public insurance status indicates healthcare access but does not fully capture social disparities. The absence of race, income, and geographic data limits broader analysis. Our model identifies patients at risk of treatment omission, aiding targeted interventions.