Abstract 727: Prospective lung cancer diagnostic screening using whole, unstimulated saliva versus stool

Abstract Current screening for lung cancer (LC) reduces mortality but tests are not ideal. Low-dose CT scans are expensive, incur radiation exposure, and high false positive rates. Although sputum cytology is unhelpful, saliva biomarker testing is promising but no prospective data are available. Methods: In conducting a study using stool markers in a high risk colorectal neoplasia population we prospectively collected saliva and stool samples. Subsequently, 11 with available saliva; 8 with stool specimens but without significant personal or family history of GI neoplasia or symptoms, had contracted LC. We used 3 monoclonal antibodies to test reactivity in saliva (Adnab-9, BAC 18.1, COX-2). These are biomarkers of innate immune system, cell-mediated immunity, and inflammation, respectively. We used ELISA with either alkaline phosphatase (ALP) or immunoperoxidase (IMP) substrate and contrasted reactivity in LC patients and asymptomatic controls with no advanced polyps, and chemiluminescent dot blots with manual or Bel-blotter 96-well replicating tool. Results: Saliva IMP testing was positive in 73% of 11 LC patients and 50% of 8 controls contrasted with ALP ELISA for stool Adnab-9 in 75% of 8 LC patients and 32% in 34 controls (OR 6.27:CI1.09-36.25;p<0.05).Specificity was 68%. Sensitivity for Adnab-9 for IMP manual and bel-blotting was 73 and 9% respectively; specificities were 37 and 64% respectively. BAC 18.1 sensitivities were 73 and 55% respectively; specificities 13 and 29%. COX-2 sensitivity for bel-blotting only was 27% and specificity was 71%. Inherent salivary peroxidase activity (OD<0.1/1μg protein) was negative in all 7 LC versus 9 of 23 (39%) of non-LC patients (p=0.07). The peroxidase absorbance means[SD] were significantly different (0.077[0.014] versus (0.116[0.056];p<0.007). Equivalent inherent alkaline phosphatase of saliva samples was negative in both groups and means were not significantly different. The approximate time from saliva collection to diagnosis of LC was 3.76 years and 3.89 for stool. Conclusions: Adnab-9 sensitivity was moderate but promising due to the ability to make an early preclinical diagnosis. While this was only significantly different from controls in stool ALP ELISA, inherent IMP activity could be blocked to improve specificity. Significantly suppressed inherent peroxidase activity in LC saliva may explain the insensitivity of the Bac18.1 and Cox2 inflammatory biomarkers. The Bel-blotter volume capacity is 4-10μl/blot and may explain the lower sensitivity using this tool. A battery of tests, including Adnab-9 in an ALP ELISA format may allow for early disease intervention. Citation Format: Yosef Y. Tobi, Fadi Antaki, MaryAnn Rambus, Martin Tobi. Prospective lung cancer diagnostic screening using whole, unstimulated saliva versus stool [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 727. doi:10.1158/1538-7445.AM2017-727