Have Barrett’s esophagus? NSAID to the rescue

From the Reid Lab, Human Biology Division
Decorative Image
Image by Lori Sanchez

Non-steroidal anti-inflammatory drugs (NSAIDs), widely used to target inflammation, generally have low toxicity and are inexpensive. Numerous experimental, epidemiologic and clinical studies in a variety of cancers suggest that NSAIDs have potential as anticancer agents because they stimulate cancer cell death while inhibiting the formation of new blood vessels that fuel tumor growth. Another anticancer strategy could be to reduce mutations occurring in neoplastic evolutionary processes that lead to cancer, particularly single nucleotide variants (SNVs) and small insertions/deletions (indels) that can be easily identified using current sequencing technologies. However, it is not known if the use of NSAIDs has any effect on these mutations. 

To investigate the impact of NSAID usage on rates of acquisition of somatic mutations before detection of cancer, Dr. Xiaohong Li and his colleagues in the Reid Laboratory (Human Biology Division), together with their collaborators at the University of Washington and the Fred Hutch Genomics Shared Resource, turned to Barrett’s esophagus (BE), a condition of the esophagus that increases a patient’s risk of developing esophageal adenocarcinoma (EA). Although the risk of progression from BE to EA is small, instabilities in the genome have been reported to be more common in BE patients that progress to EA compared to non-progressors. Importantly, these genomic aberrations can be detected several years before EA diagnosis, and can serve to inform disease progression.  Use of aspirin and other NSAIDs in BE patients has been reported to decrease the risk of tetraploidy, aneuploidy and to reduce chromosomal alterations over time.  However, evidence of NSAIDs’ influence on somatic SNVs and indels in BE has not been reported. To remedy this, the authors tapped into a valuable resource known as The Seattle Barrett’s Esophagus Study, which was initiated in 1983, to investigate the temporal course of somatic genomic evolution from an epithelial precursor to cancer and its relationship to NSAID usage. 

The authors extracted DNA from 82 epithelial tissue samples and paired blood samples, from a cross-sectional study of 41 NSAID users and 41 non-users matched by sex, age, smoking status, and continuous time on or off NSAIDs, and generated whole exome sequences from these samples.  In this study, published recently in Genome Medicine, the authors found that patients who used NSAIDs had fewer point mutations overall at the time of sampling.  This lower frequency of mutations held true even when adjusted for both TP53 mutation and smoking status.  NSAID use also decreased the prevalence of clones with high variant allele frequency.  Patients who never smoked and used NSAIDs consistently also had fewer point mutations with a cancer mutational signature prevalent in EA, suggesting that NSAIDs may hinder the survival or growth of specific mutant cell populations.


Decorative Image
NSAID use selects against diversity of functional mutations across key cancer-associated pathways and suggests that the mechanism by which NSAID use protects against cancer incidence and mortality is in part through preventing mutations that drive cells toward cancer. Figure provided by Dr. Xiaohong Li

The authors hypothesized that mutations in gene regulatory regions or within coding sequences may be more prevalent due to selection, and therefore defined “functional” mutations as those likely to impact gene function, including non-synonymous coding mutations, those within 5’ or 3’UTR exons, and those affecting splicing (within 2 bps of a splice junction). Dr. Xiaohong Li further explained: “This became very important in our pathway analysis, where 8 out of 9 pathways with lower functional mutations in NSAID users were found to have no significant difference in non-functional mutations. This result suggests there is selection by NSAIDs against development or expansion of cells with functional mutations in these pathways.”

The majority of patients with BE neither progress to nor die of esophageal cancer. Paradoxically, the majority of esophageal adenocarcinoma patients have no prior diagnosis of BE, despite evidence suggesting that it was present and undetected.  Members of the Reid lab who contributed to this study were also co-authors on a recent Nature Communications article published at the same time as this study.  In that article, they utilized The Seattle Barrett’s Esophagus Study to investigate the dynamics of BE progression using single-crypts and epithelium across two time points per patient, critical for understanding somatic evolution in BE and consequences for clinical surveillance. The Nature Communications report indicated that BE segments are mostly clonal; the phylogenetic analyses also provided insight into the tempo and mode of somatic evolution in BE, allowing a deeper understanding of how BE develops. These findings nicely compliment the current study, and the authors further described how they embarked on this current research: “We were in the unique position to test how environmental factors such as aspirin use and smoking contribute to altering the somatic mutational profile within neoplastic tissue in the human body in the absence of cancer. Our NSAID study directly compared single base pair, insertion/deletion, and chromosomal mutation profiles between NSAID users and nonusers in a single sample at a single time-point after a period of NSAID use or nonuse. This will allow us to address how NSAID use decreases the risk of cancer at the level of the somatic genome and to be able to provide insight into cancer prevention efforts.”

These findings that NSAID use can reduce overall mutations and limit the growth of specific mutant cell populations can inform future studies to integrate precision cancer prevention approaches.  As for where the Reid lab is taking this, the authors enthused: “We are currently conducting a multi-region, multi-time-point, whole genome sequencing study exploring the evolution of the full somatic genome in BE individuals who progress to esophageal adenocarcinoma compared to those who never progressed and remain stable over time. This is a highly unique study of the natural history of the somatic genome over time and will provide key evolutionary measures of diversity, selection, and spread of altered genomes, including measures of complex chromosomal structural variation that could not be assessed in the previous exome study.”

Galipeau PC, Oman KM, Paulson TG, Sanchez CA, Zhang Q, Marty JA, Delrow JJ, Kuhner MK, Vaughan TL, Reid BJ, Li X. 2018. NSAID use and somatic exomic mutations in Barrett’s esophagus. Genome Medicine. Feb 27; 10(1):17

Funding was provided by the National Institutes of Health.

This study was a Cancer Consortium collaboration between Brian Reid (Fred Hutch) and Thomas Vaughan (UW/FHCRC).