Despite the diversity inherent to cancer, they all share a common theme: many things tend to go wrong. This can make it difficult to understand which factors are driving the cancer and which ones are consequences of this aberrant growth. Such distinctions are important to understand when considering disease mechanisms or developing treatment strategies. In esophageal adenocarcinoma, “one characteristic Barrett’s esophagus patients have is a very unstable genome, which can take a lot of different forms. This includes inactivation of tumor suppressor gene TP53, where loss of two copies leads to genomic instability which can result in different types of gene amplification,” stated Dr. Thomas Paulson, a Senior Staff Scientist at Fred Hutch in the Clinical Research and Translational Science and Therapeutics Divisions. Dr. Paulson has been dedicated to understanding how Barrett’s esophagus, a pre-cancerous metaplastic condition that affects the esophageal lining, can sometimes develop into a highly lethal esophageal adenocarcinoma. Paulson had been conducting his research in the lab of now retired Dr. Brian Reid, and continues his work in Dr. William Grady’s Lab. Paulson went on to explain that genomic instability can lead to gene amplification within a chromosome, multiplying a small number of genes, but can also develop when a segment of the chromosome “pops out and circularizes, allowing it to replicate and segregate independently of chromosomes, leading to very high increases in gene copy number.” Patricia Galipeau, the research project manager of the Reid Lab, added “if you happen to get the wrong gene pop out, such as a cancer-driving gene, it can lead to big problems.” This DNA that pops out is known as extrachromosomal DNA and is known to drive tumor evolution, resistance to treatment and is associated with poor outcomes for cancer, including esophageal adenocarcinoma. However, it was unclear whether extrachromosomal DNA was a later manifestation of genomic instability or a contributor to the transition of Barrett’s esophagus to cancer. Through collaborative efforts, Paulson and other researchers at Fred Hutch, Stanford University, the University of California San Diego, Cambridge University and the University of Texas Southwestern Medical Center, analyzed patient samples from Fred Hutch and Cambridge University to get to the bottom of this chicken or the egg extrachromosomal DNA question. Dr. Xiaohong Li, a Senior Staff Scientist at Fred Hutch who contributed to this work explained that in their recent Nature study, “the two independent data sets in this paper, including our longitudinal samples, show extrachromosomal DNA significantly increased over time in those patients who progressed to cancer later.” He added that overall, “extrachromosomal DNA is strongly related with cancer outcome in the study population,” suggesting it may indeed be a driver of this cancer.
The basis for this study actually began almost 30 years ago when physician-scientist Dr. Reid developed the “Seattle Protocol”, which involves high density biopsy sampling every 1-2 cm of the Barrett’s esophagus segment. This surveillance practice is repeated every 6-12 months for Barrett’s esophagus patients with dysplastic tissue – an abnormal tissue phenotype that can precede progression to cancer. While most biopsies were used to see if cancer had developed, others were saved for research to better understand why only some Barrett’s esophagus patients progress to cancer. The patients in the published study were being treated at Fred Hutch, with the average length of follow up being about 10 years, making this a very powerful longitudinal study to investigate the factors driving cancer progression. In total, this study included 40 Barrett’s esophagus patients who eventually developed cancer, and 40 who never developed cancer during the length of the study. To increase the number of study samples and add data from different sites, the Fred Hutch researchers teamed up with Dr. Rebecca Fitzgerald, a clinician-scientist at Cambridge University who manages one of the largest cohorts of esophageal adenocarcinoma patients. While the Cambridge cohort was not longitudinal, it examined a larger number of patients at different stages of disease. This cross-sectional cohort included nearly 70 Barrett’s esophagus patients having low- or high-grade dysplasia, as well as ~50 early-stage esophageal adenocarcinoma patients and ~90 late-stage patients. The biopsies from each institution’s cohorts were analyzed by histology and whole genome sequencing to identify mutations and genomic alterations, including extrachromosomal DNA. The methods for detecting and characterizing extrachromosomal DNA, developed by collaborators Jens Luebeck (UCSD), Vineet Bafna (UCSD) and Paul Mischel (Stanford), were applied to the WGS sequence data from these cohorts. In the Cambridge samples, extrachromosomal DNA was not detected in non-dysplastic or low-grade Barrett’s esophagus samples, but was found in 25% of patients with early-stage esophageal adenocarcinoma and 43% of those with advanced disease stages. Interestingly, extrachromosomal DNA was found at higher levels in those with this late-stage cancer. Focusing on the Fred Hutch longitudinal cohort, the researchers narrowed in on those patients who eventually developed cancer and asked when they could first detect extrachromosomal DNA. Notably, in 18% of these patients, extrachromosomal DNA was detected in dysplastic tissue biopsies years prior to the onset of cancer and in 32.5% of the progressing patients overall, indicating a strong association with extrachromosomal DNA and disease progression. Furthermore, the researchers found associations with extrachromosomal DNA and biallelic loss of TP53, consistent with previous findings that loss of TP53 enables extrachromosomal DNA formation.
To better understand the potential relationship between extrachromosomal DNA and the transition from high-grade dysplasia to esophageal adenocarcinoma, the research team focused on a single patient in the Fred Hutch cohort who had biopsies collected from four endoscopies over a seven-year period. Here they were able to uncover that extrachromosomal DNA was first detected in high-grade dysplastic tissue, a time when TP53 loss had also been detected. Esophageal adenocarcinoma was diagnosed 6.5 months later, at which point a second extrachromosomal DNA derived from a different chromosomal location was identified, suggesting that multiple amplification events can occur in dysplastic tissue and enhance the transformation to malignancy.
When beginning this study, Paulson had a hunch that extrachromosomal DNA would develop before the progression to cancer; however, he admits it was exciting to see how well the Fred Hutch cohort data was replicated by the Cambridge cohort. Paulson added, “the sequencing and sample processing were done differently at each institution, so it was very gratifying in the end to see the consistent results.” Collectively, this study conclusively demonstrated that extrachromosomal amplification can precede cancer and can contribute to cancer evolution in Barrett’s esophagus patients. More broadly, Paulson pointed out that this work adds to the growing evidence of how “our genomes can be rearranged to change the function of our cells and, importantly, we were able to document this process in patients in vivo.” Galipeau added that, “inherent in the multiple samples per patient that we were able to collect, and the longitudinal aspect of this work enabled us to accurately characterize these events during the transition” from Barrett’s esophagus to cancer. Paulson stated that making the vast amount of data produced by this study and from previous studies from this group publicly available has already aided and will continue to aid researchers in studying these esophageal diseases specifically and in cancer progression more generally. Clinically, “given the high specificity of ecDNA for progression to esophageal adenocarcinoma, it has potential to be a diagnostic tool for identifying patients at risk of progressing,” Paulson explained. He added that there are also certain ways to eliminate extrachromosomal DNA, which although may not stop the cancer progression, may weaken cells and increase the susceptibility of cancer to other treatments and hopefully eliminate it. Paulson and Galipeau concluded by thanking the many individuals who donated their time and tissue to scientific research, without whom this study could not have been possible.
This work was supported by Cancer Research UK, the National Cancer Institute, the National Institutes of Health, the National Brain Tumor Society, the Cancer Prevention and Research Institute of Texas, the Brain Korea 21 Four Project, the Korean Ministry of Food and Drug Safety, Korean Ministry of Science and the PreCancer Genome Atlas.
UW/Fred Hutch/ Seattle Children’s Cancer Consortium member Brian Reid (retired) and Carissa Sanchez (Reid lab manager, retired) contributed to this work.
Luebeck J, Ng AWT, Galipeau PC, Li X, Sanchez CA, Katz-Summercorn AC, Kim H, Jammula S, He Y, Lippman SM, Verhaak RGW, Maley CC, Alexandrov LB, Reid BJ, Fitzgerald RC, Paulson TG, Chang HY, Wu S, Bafna V, Mischel PS. Extrachromosomal DNA in the cancerous transformation of Barrett's oesophagus. Nature. 2023 Apr;616(7958):798-805. doi: 10.1038/s41586-023-05937-5. Epub 2023 Apr 12. PMID: 37046089; PMCID: PMC10132967.