NEW YORK (GenomeWeb) – By unraveling the order in which mutations turn up in esophageal adenocarcinoma and the non-cancerous conditions that precede it, members of the International Cancer Genome Consortium have found recurrently mutated genes that may help in esophageal cancer prediction and early diagnosis.
As they reported online this weekend in Nature Genetics, researchers with ICGC's Esophageal Cancer Clinical and Molecular Stratification consortium and investigators from the University of Cambridge and elsewhere did whole-genome sequencing on nearly two dozen tumor-normal pairs from individuals with esophageal adenocarcinoma, along with targeted sequencing on additional tumor and pre-cancerous samples.
When the team followed the mutational transitions marking progression from non-malignant, non-dysplastic Barrett's esophagus to high-grade dysplasia to esophageal adenocarcinoma, it saw almost complete overlap between recurrently mutated genes at each stage.
But there were a few key stage-specific differences as well, including jumps in TP53 mutations in high-grade dysplasias and esophageal adenocarcinoma-related mutations in the SMAD4 gene. That information may prove useful for finding high-risk cases — a possibility the study's authors explored using a new, non-invasive cell sampling device called Cytosponge that collects cells in mucus from the esophagus, which can then be scrutinized for telltale TP53 mutations.
"The Cytosponge provides a representative sample of the entire esophageal mucosa and, coupled with high-throughput sequencing, is capable of sensitive detection of [high-grade dysplasia]," they wrote.
"This approach could be readily adapted as understanding of the genetic basis for the disease evolves," they added. "Furthermore, our systematic molecular approach to identify key mutations involved in the steps distinguishing pre-invasive from invasive disease has applicability to other epithelial cancers amenable to early detection."
In the past, the study's authors explained, the presence of Barrett's esophagus itself was considered a harbinger of possible esophageal adenocarcinoma development. But only a subset of those with Barrett's esophagus progress to the more serious high-grade dysplasia or cancer stages, prompting interest in more accurate disease markers.
The team took a genetic approach to finding such markers, starting with 22 esophageal adenocarcinoma samples nabbed by surgical resection or during endoscopic ultrasound tests.
Using Illumina's HiSeq 2000, the researchers did whole-genome sequencing on the tumors as well as matched normal blood or tissue samples from each individual. Matched tumor and normal sequence comparisons pointed to 17,000 or so possible somatic single nucleotide variants per tumor, on average, and nearly 1,000 small insertions and deletions.
From there, the team zoomed in on 26 particularly suspicious genes that were frequently mutated and/or belong to pathways of interest. With the help of amplicon sequencing, those genes were subsequently scrutinized in 90 more esophageal adenocarcinoma samples.
The researchers also did targeted gene sequencing on 66 never-dysplastic Barrett's esophagus (NDBE) biopsies (taken from 40 individuals who did not progress to subsequent stages of diseases) and 43 biopsies taken from individuals who went on to develop high-grade dysplasia.
Despite the benign nature of the NDBE tumors, they found that these samples often had mutations falling in the same genes that were recurrently mutated in tumors and in high-grade dysplasias, suggesting driver mutations appear early on and can occur in some tissues that never become cancerous.
The transition from NDBE to high-grade dysplasia was generally marked by a jump in TP53 mutations, researchers reported, while full-blown esophageal adenocarcinomas contained recurrent SMAD4 mutations.
In an effort to begin applying such insights, the team decided to focus on the TP53 mutations that mark high-grade dysplasia and subsequent tumor samples, using the Cytosponge device to collect esophageal mucosal cells from 44 individuals with NDBE, 22 individuals with high-grade dysplasia, and 23 unaffected controls.
Indeed, targeted TP53 sequencing on these samples uncovered mutations in 86 percent of samples from those with high-grade dysplasia, while the team did not detect TP53 mutations in samples from unaffected individuals or individuals with NDBE.
The study's authors noted that additional research is needed to track down the genetic or epigenetic alterations behind high-grade dysplasias and esophageal adenocarcinomas that don't contain TP53 mutations. Even so, they argued that the molecular approach might make it possible to more accurately discern Barrett's esophagus cases that are inching towards cancer from those that are not.