Abstract 5027: Detection of soluble protein biomarkers in NSCLC serum samples by meso scale discovery electrochemiluminescence platform

Abstract

Abstract Identification of soluble biomarkers has become a critical non-invasive approach for disease diagnosis and monitoring in the treatment of solid tumors. Non-Small Cell Lung Cancer (NSCLC) is the most common form of lung cancer, accounting for approximately 85% of all lung cancers worldwide. The EGFR family of genes is known to be highly expressed in NSCLC tumors and play a critical role in the progression of the disease. Although patients initially respond well to EGFR TKI therapy, acquired resistance occurs through several mechanisms, including compensatory cMET pathway activation (Robinson et al. 2013). Previous reports have demonstrated that soluble forms of EGFR, Her2 and cMET are generated through either alternate splicing of mRNA or proteolytic cleavage of the full length receptors (Wilken et al. 2013). Serum concentrations of these receptors can be correlated with prognosis as well as treatment response (Gregorc, 2004, Kasahara, 2010 Heinmoller, 2003). Soluble receptor ligands, including Hepatocyte growth factor (HGF), have also been evaluated in NSCLC and a strong correlation has also been observed between the levels of serum HGF and disease outcome in patients treated with EGFR-TKIs (Kasahara et al. 2010). In this study, we evaluated concentrations of sEGFR, sMET, sHER2,, and HGF in NSCLC versus healthy normal serum samples via meso scale discovery (MSD) platform. Inventoried and custom MSD assays were optimized to measure these proteins in a training set consisting of 19 NSCLC and 16 normal healthy serum samples. Significantly lower concentrations of sEGFR and sHER2 were observed in the NSCLC serum samples compared to normal (p value = 0.0092 for both receptors), whereas, in contrast, the levels of HGF were significantly higher in the NSCLC patient samples (p value <0.0001). No significant difference in sMET concentration was observed between NSCLC and normal serum. Based on this data, it was determined that a ratio of protein concentration of each soluble receptor (EGFR, Her2 and cMET) to HGF yielded a greater differentiation between NSCLC and normal samples (p value <0.0001, = 0.0041,and 0.0327 respectively). Cut-off values to define NSCLC versus normal patients were established for each ratio and validated in a blinded independent cohort, consisting of 50 NSCLC and 12 healthy normal serum samples. PPV and NPV were then calculated to determine sensitivity and specificity for each protein ratio. Results from the validation study show an almost complete separation between healthy and NSCLC patient samples (p value <0.0001, AUC = 1)for all three ratios tested (sEGFR/HGF, sHER2/HGF and sMET/HGF), with a PPV = 100%, 98%, 100% and NPV = 100%, 100%, 85.7%, respectively. Overall, these findings indicate potential clinical applications for these protein pairs as non-invasive biomarkers of NSCLC diagnosis, prognosis and treatment monitoring to support emerging TKI therapies Citation Format: Dana S. Gaffney, Katherine Bell, Gabriela Martinez, Jayaprakash Karkera, Suso Platero. Detection of soluble protein biomarkers in NSCLC serum samples by meso scale discovery electrochemiluminescence platform. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 5027.