Sputum Supernatant As An Alternative Source For Liquid Biopsy In Patients With Advanced Lung Cancer.

e21617 Background: With the advancements in the development of targeted therapy, the detection of actionable genes has become routine practice in diagnosing lung cancer. Plasma-based mutation profiling is widely used in clinical settings. Profiling using other body fluids is actively explored. In this study, we investigated the potential of induced sputum obtained from non-small cell lung cancer (NSCLC) patients for mutation profiling. Methods: We performed targeted sequencing on matched FFPE, plasma and induced sputum samples of 41 treatment-naïve advanced NSCLC using a panel consisting of 168 lung cancer-related genes. 10ml of whole blood and 8 ml of induced sputum were obtained from each patient. Induced sputum samples were fractionated into supernatant (sup) and precipitates (ppt). The cohort had a median age of 65 with a majority (32/41) diagnosed with lung adenocarcinoma. 7 of them were diagnosed with squamous cell carcinoma and the remaining 2 patients were diagnosed with neuroendocrine tumor. All patients had advanced disease (stage III or IV). Results: First, we investigated the optimal sputum fraction for mutation profiling by comparing the detection rates and maximum allelic fraction (maxAF) between sputum sup and ppt samples using matched tissue and plasma as references. The detection rate was defined as any mutation detected from the panel used; while maxAF was defined as the highest mutation allelic fraction among all mutations detected. Tissue, plasma, sputum sup and ppt achieved a rate of detection of 100%, 76.9%, 72.4% and 65.7% respectively, revealing comparable detection rates between plasma and sputum sup (p = 0.89). Sup had a significantly higher average maxAF than plasma (p = 0.03) and ppt samples (p < 0.01). Using driver mutations detected from tissue samples as references, sup, ppt and plasma samples achieved a sensitivity of 63.2%, 50% and 67.8%, respectively. Next, we investigated whether induced sputum obtained from smokers would be more informative than non-smokers. Our analysis revealed that sup from smokers had a significantly higher average AF than sup from non-smokers (p < 0.001). Furthermore, there was a statistical significant difference in average AF between sup and ppt obtained from smoker (p < 0.001) but such difference was non-existent in non-smokers (p = 0.46). Conclusions: Our study demonstrated that sputum sup demonstrates a comparable performance as plasma samples and is as a better sampling source than its ppt. Induced sputum from advanced stage NSCLC patients is an alternative media for mutation profiling.