CT Deep Learning Model Predicts Response to Immunotherapy Plus Chemo After EGFR-TKI Resistance

CT-based 2.5D deep learning model predicted response to immunotherapy+chemo after EGFR-TKI resistance: AUCs 0.885 (training), 0.819, 0.863 (validation cohorts). Prospective validation needed. (n=490)

• Population: 490 patients with EGFR-mutant lung adenocarcinoma after first-generation TKI resistance and without T790M mutation.
• Model: 2.5D axial CT deep learning model yielded AUCs of 0.885 (training), 0.819, and 0.863 in independent validation cohorts.
• Limitation: Lack of prospective validation and comparison with clinical variables (authors note).

Automated summary

RadPigeon summaries are original and for information only. They are not clinical advice.