Individualized nomogram for predicting ALK rearrangement status in lung adenocarcinoma patients

摘要: To develop a nomogram to identify anaplastic lymphoma kinase (ALK) mutations in lung adenocarcinoma patients using clinical, CT, PET/CT, and histopathological features. This retrospective study included 399 (129 ALK-rearranged 270 ALK-negative patients) that were randomly divided into training cohort an internal validation (4:1 ratio). Clinical factors, radiologist-defined CT features, maximum standard uptake values (SUVmax), features used construct predictive models with stepwise backward-selection multivariate logistic regression (MLR). The then evaluated the AUC. integrated model was compared clinico-radiological DeLong test evaluate role of An associated individualized established. reached AUC 0.918 (95% CI, 0.886–0.950), sensitivity 0.774, specificity 0.934 0.857 0.777–0.937), 0.739, 0.810 cohort. MLR analysis showed younger age, never smoker, lymph node enlargement, presence cavity, high SUVmax, solid or micropapillary predominant histology subtype, local invasiveness strong independent predictors ALK rearrangements. calculated risk harboring mutation for exhibited good generalization ability. Our demonstrates added value imaging characteristics-based model. imaging, can serve as supplementary non-invasive tool probability rearrangement adenocarcinoma. • developed accurately predict ALK-fused gene. Pathological is important adenocarcinoma. Lung lepidic growth pattern TTF-1 negativity unlikely have rearrangement.