Although squamous cell carcinomas and adenocarcinomas of the esophagus differ histologically, they are often lumped together in clinical trials. Genomic analysis shows that the two cancers are molecularly distinct. Esophageal adenocarcinomas are more similar to stomach adenocarcinomas, whereas esophageal squamous cell carcinomas are closer to head and neck squamous cell carcinomas.

Esophageal adenocarcinomas (EAC) more closely resemble certain stomach cancers than they do other esophageal tumors, according to the largest genomic analysis of the cancers (Nature 2017;541:169–75).

The relationships among esophageal tumors and their connections to stomach tumors have been unclear. Histology suggests that esophageal squamous cell carcinomas (ESSC) and EACs are distinct. However, clinical trials often include patients with both types of tumors. What distinguishes EACs from gastric adenocarcinomas has also been a matter of debate.

To try to resolve these issues, Adam Bass, MD, of Dana-Farber Cancer Institute in Boston, MA, and colleagues with The Cancer Genome Atlas Research Network obtained samples from 164 esophageal tumors: 90 ESSCs, 72 EACs, and two undifferentiated carcinomas. The researchers performed whole-exome sequencing on the samples, checked for copy-number alterations, profiled DNA methylation patterns, and sequenced mRNAs and microRNAs. The scientists also compared the samples to 359 gastric adenocarcinomas and 36 adenocarcinomas from the gastroesophageal junction.

The results bolster the case that ESSCs and EACs are separate. For example, CCND1 and SOX were often amplified in ESSCs, whereas EACs typically had amplifications of ERBB2 and VEGFA. The Wnt, syndecan, and p63 pathways tended to be upregulated in ESSCs, but EACs typically showed upregulation of the ARF6 and FOXA pathways. The researchers found that ESSCs most closely resembled squamous cell cancers of the head and neck. In contrast, EACs were very similar to a subtype of gastric adenocarcinoma with high levels of chromosome instability.

“Not only are esophageal squamous cell carcinoma and adenocarcinoma biologically distinct, but we show that esophageal adenocarcinoma and gastric adenocarcinoma should be viewed more as a single entity,” says Bass. In colorectal cancer, a gradient of carcinoma subtypes spans the colon and rectum. Bass says that a similar gradient of adenocarcinoma subtypes stretches across the esophagus and stomach.

Recognizing the overlap between stomach and esophageal tumors could lead to new opportunities for treating patients, the researchers note. For instance, the clinical trials that led to the approval of trastuzumab (Herceptin; Genentech) for metastatic adenocarcinomas of the stomach and gastroesophageal junction excluded patients with EACs. Given the striking overlap in targets, grouping patients with EAC and gastric cancer for trials could speed clinical development, Bass says.

The paper “is a step further in attempting to help us understand the immense heterogeneity of this disease,” says Daniel Catenacci, MD, of the University of Chicago Medical Center in Illinois, who wasn't connected to the study.

The results should have an impact on the design of clinical trials for esophageal cancer, says Prateek Sharma, MD, of the University of Kansas Medical School in Kansas City, who also wasn't connected to the study. “For a long time, all esophageal cancers were enrolled in chemotherapy and radiotherapy trials and lumped together. Although there was growing evidence that they should be separated out, this really proves it,” he says.

Catenacci cautions, however, that the implications for treatment aren't clear. Other studies have suggested different ways of classifying EACs, he notes. Further research is needed to determine which classification provides “demonstrable clinical relevance and predictive ability,” he says. –Mitch Leslie

For more news on cancer research, visit Cancer Discovery online at http://cancerdiscovery.aacrjournals.org/content/early/by/section.