Circulating tumor DNA as a non-invasive tool to identify patients at risk for recurrence after chemoradiotherapy in stage III non-small cell lung cancer.

Lung Cancer—Non-Small Cell Local-Regional/Small Cell/Other Thoracic Cancers
Session Type and Session Title: 
Poster Session, Lung Cancer—Non-Small Cell Local-Regional/Small Cell/Other Thoracic Cancers
Abstract Number: 


Poster Board Number: 
Board #181
J Clin Oncol 34, 2016 (suppl; abstr 8553)
Steven H. Lin, Xiayu Rao, Kimberly C. Banks, Ting Xu, Jianzhong He, Richard Burnham Lanman, Amirali Talasaz, Jing Wang, Zhongxing X. Liao; Department of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX; Department of Bioinformatics and Comp Biology, The University of Texas MD Anderson Cancer Center, Houston, TX; Guardant Health, Inc., Redwood City, CA

Abstract Disclosures


Background: Outcomes for curative-intent chemoradiotherapy of unresectable stage III non-small cell lung cancer (NSCLC) remains poor. Heterogeneity in tumors is correlated with therapeutic resistance and poor prognosis. We hypothesize that tumor-specific alterations (alts) in circulating tumor DNA (ctDNA) quantifies tumor heterogeneity and can serve as a non-invasive means to determine prognosis and recurrence. Methods: Between 2009-2013, 156 patients (pts) with unresectable NSCLC receiving definitive radiotherapy (XRT) or chemo-XRT were consented to have blood drawn pre-XRT, 1-2 times during XRT, and 1-2 times in follow up ( > 670 samples). We used a ctDNA NGS assay to detect SNVs in 70 genes, amplifications in 16, and select fusions and indels. Statistical analyses were performed for association between OS/PFS and ctDNA serial samples results. Here we report the initial interim analysis. Results: Of 44 pts, 32 (73%) recurred (PFS range 0-26.6 months) and 12(27%) had no evidence of recurrence as of last contact (1.9-43.5 months post-XRT). We found 4 main patterns of ctDNA changes: specific alts persistent throughout XRT (n = 9); no alts detected in the post-XRT sample (n = 14), increased levels from baseline (n = 10), or fluctuates throughout therapy (n = 11). Alleles which persisted throughout XRT were found in TP53, ARID1A, BRCA2, NF1, KRAS, APC, AR, ERRB2, PIK3CA, ATM, MET, BRAF, PDGFRA, CRCA1, NOTCH1, RAF1, GATA3, KIT. While the 4 patterns of cfDNA fluctuations were not found to be prognostic in this interim analysis, the presence of specific mutations appears to correlate with outcomes, with reappearance of the driver mutations post-tx associated with shorter PFS. Using FDR cutoff of 0.1, NF1 mutations were associated with shorter OS (p = 0.07), and APC/ARID1Amutations associated with shorter PFS (p = 0.09) after adjustment for tumor histology or stage. Conclusions: In this interim analysis, we found that specific mutant alleles in cfDNA are correlated with potentially treatment resistance and early metastatic recurrence. Final analysis of the larger cohort may be required to achieve significance for additional prognostic patterns.