Abstract

22216 Background: Early detection is crucial to reducing cancer mortality. Currently, there is no single test available to identify early stage non-small cell lung cancer (NSCLC). Previously, we reported a panel of auto-antibodies used to detect early-stage and occult NSCLC prior to discovery on computed tomogram. From serum samples of NSCLC patients drawn 0 to 5 years before radiographic detection, we achieved 91.3% sensitivity and 91.3% specificity to detect stage I NSCLC and 82.6% of sensitivity and 87.5% specificity to discover occult NSCLC. In this study, we repeated the biopanning process to further confirm and potentially increase the sensitivity and specificity of the test. Methods: Using 5-pooled NSCLC sera from one NSCLC T7 cDNA in-frame phage library and after 4 cycles of biopanning enrichment, 2000 phage clones were randomly picked and spotted onto nitrocellulose coated slides along with 400 empty phage clones and healthy human serum as negative and positive controls, respectively. Results: After comparative screening with 28 stage I NSCLC patient sera and 28 healthy control serum samples, 54 clones were found to have significant difference (P<0.05), while screening with 20 stage II-IV NSCLC patient serum and 20 healthy control serum samples, 162 clones were found to have significant difference (P<0.05). Among these selected clones, 15 clones were found to be common between the two groups. A panel consisting of as few as 22 clones appeared to achieve 100% sensitivity and 100% specificity using discriminant analysis differentiating 48 patients and 48 matched controls. We sequenced the top 20 of the most predictive clones and found 9 of them are well characterized tumor-associated proteins. Conclusions: Currently, this panel of markers is under evaluation and validation with additional samples to assess the predictive accuracy. Improved accuracy should allow earlier intervention and superior results in the treatment of NSCLC. No significant financial relationships to disclose.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call