Liao G, Huang L, Wu S, et al. - Findings showed a substantial association of peritumoral characteristics with the spread through air space (STAS) status in clinical stage I lung adenocarcinoma. Observations unveiled the unit of radiomic signature and clinical signatures could have a better performance in STAS status.

• In this retrospective study of patients with pathologically confirmed STAS status (N=256), the development cohort included 191 patients (74.6%) selected randomly in a 7:3 ratio, whereas the validation group comprised 65 patients (25.4%).

• STAS positive and negative status was detected in 33.2% and 66.8% of the patients, respectively.

• AUCs (areas under the receiver operating characteristics curve) of 0.854 and 0.870 in the development cohort and validation cohort, respectively, were generated by the combined tumor radiomic signature (TRS) and peritumor radiomic signature (PRS)-15mm model.

• This suggests that the TRS model and different PRS models were employed to construct the optimal gross radiomic signature (GRS) model.

• The radiomic nomogram displayed superior discriminatory performance, relative to GRS and clinical and radiological signatures, with an AUC of 0.871 in the development and 0.869 in the validation cohorts, respectively.

• Greater clinical predictive capacity was achieved with the radiomic nomogram vs clinical or any radiomic signatures alone.