European Radiology: Radiomics enables non-invasive prediction of EGFR mutation status in GGO lung adenocarcinoma!

According to statistics, adenocarcinoma is the most common histological subtype of lung cancer, accounting for more than 50% of all case.

Currently, tissue biopsy-based genetic testing is the gold standard for EGFR mutation detectio.

Therefore, there is a need to develop a noninvasive, time-saving method to predict the EGFR mutation status of early-stage lung adenocarcinoma lesions that appear as GGOs on C.

Currently, CT has been used as a routine examination for lung cancer screenin.

Recently, a study published in European Radiology established an EGFR mutation prediction model specifically for early-stage lung adenocarcinoma GGO lesions, which is an early, rapid and accurate clinical assessment of the EGFR mutation status of such lesions and personalized treatment plan.

Clinicopathological information and preoperative CT images of 636 lung adenocarcinoma patients (464, 100, and 72 in training, internal, and external validation sets, respectively) who underwent GGO lesion resection were included in this stud.

The established radiomics model containing 102 features showed strong discriminative power for EGFR mutation status (mutant or wild-type), with better predictive power than clinical models (AUC: 838 v.

Unusually, this model was validated in a cohort of lung adenocarcinoma patients who had received adjuvant EGFR-TKI therapy and had unresectable GGOs during the treatment period, resulting in significantly increased potency of EGFR-TKIs (Response rate: 29% v.

This study developed a CT image-based radiomics model for predicting EGFR mutation status in patients with GGO-characterized lung adenocarcinom.

Original source:

Bo Cheng, Hongsheng Deng, Yi Zhao, et a.