Comparative Analysis of the Representative Value of Somatic Mutation Profiles Derived from Ctdna and Other Updated Lung Tumor Biopsy Alternative in Non-Resectable Non-Small Cell Lung Cancer Patients.

Xiaozheng Kang,Ting Xu,Ting Zhou,De-Qun Tian,Grace Q. Zhao,Shengrong Lin,Kang Ying,Guo-Bing Xu,Ke-Neng Chen

DOI: https://doi.org/10.1200/jco.2016.34.15_suppl.e23034

IF: 45.3

2016-01-01

Journal of Clinical Oncology

Abstract:e23034 Background: Genomic profiling is critical for optimal treatment selection in non-small cell lung cancer (NSCLC) patients. Unfortunately, a number of these patients are ill-suited for tissue biopsy. While tumors under formalin fixation and paraffin embedding (FFPE) from prior sampling or lymph node biopsy may provide alternatives to update lung tumor biopsy, genomic profiles generated from these sources may not best represent a patient’s current condition. Here, we compare circulating tumor DNA (ctDNA) to DNA acquired from either tumor biopsy or a biopsy alternative to assess their comparative representative value. Methods: Plasma samples were collected from 20 non-resectable NSCLC patients and subject to NGS ctDNA analysis using Accu-Act, a panel of 61 genes with implications in cancer treatment. Parallel sequencing was performed on tumor tissue derived from update lung tumor biopsy or alternatives. Of the 20 tumor tissue samples, 7 came from old FFPE samples, 4 from recent lymph node biopsy, 8 from recent tumor biopsy, and 1 from recent needle biopsy. Concordance of ctDNA and tumor/lymph node gDNA was determined and mutation profiles were correlated with the patient’s treatment progression. Results: Of the 7 FFPE samples, 3 showed good concordance (90%) with cfDNA mutation profiles. No concordance was found for gDNA derived from recent lymph node biopsy. Among the 14 patients undergoing EGFR-TKI treatment, their respective ctDNA mutation profiles correlated with their current status including 2 patients determined as partial response had clearly suppressed EGFR mutations; 2 out of 6 patients determined as progressive disease, had known EGFR drug-resistant mutations. Conclusions: From this small comparative analysis, we demonstrate the superiority of cfDNA profiling in providing a more accurate representation of a patient’s cancer progression relative to genomic profiling performed on alternative genomic sources in patients with non-resectable NSCLC. In addition, our finding suggests treatment prediction using ctDNA profiling is superior to profiling performed on gDNA from alternatives to biopsy.