Prediction of response to anti-PD-1 therapy in patients with non-small-cell lung cancer by electronic nose analysis of exhaled breath
2019
Abstract Background Immune checkpoint inhibitors have improved survival outcome of advanced non-small-cell lung cancer (NSCLC). However, most patients do not benefit. Therefore, biomarkers are needed that accurately predict response. We hypothesized that molecular profiling of exhaled air may capture the inflammatory milieu related to the individual responsiveness to anti-programmed death ligand 1 (PD-1) therapy. This study aimed to determine the accuracy of exhaled breath analysis at baseline for assessing nonresponders versus responders to anti-PD-1 therapy in NSCLC patients. Methods This was a prospective observational study in patients receiving checkpoint inhibitor therapy using both a training and validation set of NSCLC patients. At baseline, breath profiles were collected in duplicate by a metal oxide semiconductor electronic nose (eNose) positioned at the rear end of a pneumotachograph. Patients received nivolumab or pembrolizumab of which the efficacy was assessed by Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 at 3-month follow-up. Data analysis involved advanced signal-processing and statistics based on independent t-tests followed by linear discriminant and receiver operating characteristic (ROC) analysis. Results Exhaled breath data of 143 NSCLC patients (training: 92, validation: 51) were available at baseline. ENose sensors contributed significantly (P Conclusion ENose assessment was effective in the noninvasive prediction of individual patient responses to immunotherapy. The predictive accuracy and efficacy of the eNose for discrimination of immunotherapy responder types were replicated in an independent validation set op patients. This finding can potentially avoid application of ineffective treatment in identified probable nonresponders.
Keywords:
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
26
References
28
Citations
NaN
KQI