Clinical utility of the 5-point visual Deauville score for preoperative evaluation in stage I non-small cell lung cancer: A multicenter retrospective study.

Abstract

Reliable prognostic tools are essential for improving the management of stage I non-small cell lung cancer. Whereas fluorodeoxyglucose positron emission tomography has proven valuable, metrics such as maximum standardized uptake value are limited by interinstitutional variability. This study evaluates the Deauville score for its clinical utility in predicting pathological invasiveness and survival outcomes. Data were retrospectively collected from 495 patients with stage I non-small cell lung cancer who underwent pulmonary resection across 4 institutions in Japan. Tumors were classified by a visual 5-point scale Deauville scores into groups: 1 and 2, 3, and 4 and 5. Associations between these scores, pathological invasiveness (defined as lymphatic invasion, vascular invasion, pleural invasion, and/or nodal involvement), and survival outcomes were analyzed. High Deauville scores (4 and 5) were significantly associated with pure solid tumors (85%; P < .0001) and pathological invasiveness (66%; P < .0001). Multivariable analysis showed the Deauville score to be an independent predictor of invasiveness (odds ratio, 10.145; P < .0001 for scores 4 and 5 vs 1 and 2). Survival analyses revealed poorer outcomes for patients with scores of 4 and 5, with a 5-year recurrence-free survival of approximately 60% (P < .0001). Even in part-solid tumors, scores of 4 and 5 were linked to significantly worse recurrence-free survival compared with scores of 1 and 2 or 3 (P < .0001). The Deauville score provides a straightforward and consistent method to predict pathological invasiveness and survival in stage I non-small cell lung cancer. Its reliability and reproducibility across institutions highlight its potential as a valuable tool for both clinical practice and large-scale research.