Abstract 2993: The landscape of DNA allelic imbalance in the normal-appearing airway field of cancerization

• Published in: Cancer research (Chicago, Ill.) Ill.), 2015-08, Vol.75 (15_Supplement), p.2993-2993
• Main Authors: Jakubek, Yasminka, Lang, Wenhua, Vattathil, Selina, Garcia, Melinda, Huang, Lili, Lu, Wei, Chow, Chi-Wan, Weber, Zachary, Davies, Gareth E., Behrens, Carmen, Kalhor, Neda, Moran, Cesar, Fujimoto, Junya, Mehran, Reza J., Fowler, Jerry, Ehli, Erik A., Wistuba, Ignacio I., Scheet, Paul, Kadara, Humam
• Format: Article
• Language: English
• ISSN: 0008-5472; 1538-7445
• DOI: 10.1158/1538-7445.AM2015-2993

The phenomenon of field cancerization has been previously postulated and observed in various cancers, including those of the lung. We have recently demonstrated that normal-appearing airway cells carry expression profiles that are often characteristic of the adjacent tumor. A better understanding of the molecular mechanisms that drive these field changes may provide important biological insights into lung cancer pathogenesis. Loss-of-heterozygosity (LOH) and other forms of acquired chromosomal alterations (allelic imbalance; AI) have an established and profound role in oncogenesis. However, the relationship between AI and field cancerization has not been studied comprehensively across the genome. Here we address this void by interrogating a rich collection of normal-appearing airways from non-small cell lung cancer (NSCLC) patients. We applied Illumina 1M SNP arrays to characterize whole genome copy number alterations in 435 samples from 45 early-stage NSCLC patients [31 adenocarcinomas (ADCs), 14 squamous cell carcinomas (SCCs)]. Each patient set comprised samples from the primary tumor and adjacent normal-appearing airways paired with blood cells and/or uninvolved normal lung tissue. A subset of these included brushings from ipsilateral large airways (mainstem bronchi) and from the nasal cavities as well as multi-region tumor biopsies for intra-tumoral analysis (on 22, 27, and 20 patients, respectively). To characterize the field in normal-appearing airways at a genome-wide scale, we applied a haplotype-based computational program, hapLOH, to profile AI events (loss, gain, copy neutral LOH) in a paired mode contrasting signals in the blood or normal lung. We detected 198 AI events in normal-appearing airways of 24 of 45 patients; 92% of these events were represented in the paired tumor. Of the 24 patients, 23 had events in the adjacent airway, 3 had events in the large airway, and no events were observed in nasal brushings, indicating a pronounced AI field gradient. We detected AI in the airways of approximately 43% of ADCs regardless of smoking status (3 of 7 smokers, 10 of 24 non-smokers), and 79% (11 of 14) of SCCs, clearly indicating squamous histology as a greater predictor of observing a genomic field effect (P < 0.03). The most frequently observed airway alterations were in 9p and 9q, affecting 13 smoker patients; interestingly, these were not observed in the non-smokers. Finally, we note that AI events were present in the adjacent airways of 4/5