1247P - A detection model for EGFR mutations in lung adenocarcinoma patients based on volatile organic compounds

Background

Volatile organic compounds (VOCs) are known to be associated with lung cancer, and EGFR mutations are common in lung adenocarcinoma. In this study, we aimed to investigate the potential of VOCs analysis as a non-invasive method for early detection of EGFR mutation.

Methods

We collected breath samples from 114 EGFR wild-type lung adenocarcinoma patients and 132 EGFR mutant lung adenocarcinoma patients and analyzed them using High-Pressure Photon Ionization Time-of-Flight Mass Spectrometry (HPPl-TOFMS) to detect VOCs. A detection model for EGFR mutation based on VOC analysis was established using the eXtreme Gradient Boosting (XGB) algorithm. For building the EGFR mutation detection model, all the samples were randomly split into a training set, validation set, and test set with a ratio of 5:2:3.

Results

The top ten VOC ions which were significantly different between the EGFR mutant lung adenocarcinoma patients and patients without EGFR mutation were displayed in the table (P < 0.05). The discrimination power of each VOC was relatively limited. Among them, 2-Ethyl-4,5-dimethylthiazole had the highest accuracy with an area under the receiver operating characteristic curve (AUC) of 0.829 (95%CI: 0.744-0.974). With the combination of these 10 VOC ions, the model performs well. The AUC for the combination model was 0.895 (95% CI: 0.810-0.980) in the validation cohort, while the AUC in the test set was 0.849 (95% CI: 0.769-0.930), indicating good diagnostic accuracy, with sensitivity and specificity of 70.7% (95% CI: 56.6%, 84.4%) and 71.4% (95% CI: 56.2%, 86.6%). Table: 1247P

The top 10 VOC ions related to EGFR mutation in lung adenocarcinoma

Conclusions

Our study suggested that VOCs analysis might be useful as non-invasive methods for early detection of EGFR mutations in lung adenocarcinoma. Further studies are needed to validate the accuracy of our model.

Clinical trial identification

Legal entity responsible for the study

The authors.

Funding

The Chinese National Natural Science Foundation Project; China Postdoctoral Science Foundation.

Disclosure

All authors have declared no conflicts of interest.