From the Journals

Esophageal cancers: Apples and oranges wrongly lumped together

View on the News

Location, location, location – may not be as important

This article published in Nature summarizes an integrated genomic analysis of esophageal cancer with careful comparisons to other cancers in the neighborhood (head and neck, lung, and gastric cancer). While clinically apparent to physicians taking care of esophageal cancer throughout the world, these analyses confirm that esophageal squamous cell cancer and esophageal adenocarcinoma are essentially two different diseases with distinct genomic characteristics. This has an important implication in clinical trial design: These pathologies should not be analyzed together, but instead should be studied distinctly.

In addition to the above major conclusion, several other features deserve to be noted. First, esophageal cancer does not seem to be associated with HPV as the HPV transcript levels in these tumors resemble those in HPV-negative head and neck cancers. Second, there are significant differences in the genomic characteristics of esophageal squamous cell cancer depending on geographic location. Third, esophageal adenocarcinoma is most like one particular molecular variant of gastric cancer (chromosomal instability type) and as one moves from the gastric antrum to the esophagus, there is an enrichment of this type of cancer. Such a gradient is found in methylation patterns as well, suggesting a similar cell of origin between gastric and esophageal cancers.

This study brings into focus the overarching theme that cancers may soon be treated based on molecular characteristics rather than anatomic location and clinical trials may have to be grouped based on genetic changes rather than organ systems.

Sai Yendamuri, MD, FACS, is an attending surgeon at the department of thoracic surgery, and director, Thoracic Surgery Research Laboratory, and an associate professor of oncology at Roswell Park Cancer Institute, Buffalo, N.Y.


 

FROM NATURE

Genomic analysis suggests that esophageal adenocarcinoma (EAC) and esophageal squamous cell carcinoma (ESCC) are two separate diseases that should not be combined in clinical trials and may benefit from different treatments, according to the results of a molecular study of 559 esophageal and gastric carcinoma tumors obtained from around the world.

The comprehensive molecular analysis comprised 164 esophageal tumors, 359 gastric adenocarcinomas, and 36 additional adenocarcinomas spanning the gastroesophageal junction.

esophagus Eraxion/Thinkstock
Overall, squamous cell carcinomas showed amplification of different genes than those amplified in adenocarcinomas, according to the report published in Nature (2017;541:169-75).

The results of their analysis “call into question the premise of envisioning esophageal carcinoma as a single entity” and “argue against approaches that combine EAC and ESCC for clinical trials of neoadjuvant, adjuvant, or systemic therapies,” wrote the members of The Cancer Genome Atlas Research Network under the coordination of the National Cancer Institute and the National Human Genome Research Institute project.

The researchers evaluated the 164 esophageal carcinomas using integrated clustering of somatic copy number aberrations, DNA methylation, mRNA, and microRNA expression.

Gene expression analysis showed EACs had increased E-cadherin (CDH1) signaling and upregulation of ARF6 and FOXA pathways, which regulate E-cadherin. In contrast, ESCCs showed upregulation of Wnt, syndecan; p63 pathways, which are essential for squamous epithelial cell differentiation, were also upregulated. “These data suggest the presence of lineage-specific alterations that drive progression in EACs and ESCCs,” according to the researchers.

Somatic genome alterations showed that many of the same genetic pathways were altered in both EAC and ESCC, but the specific genes affected were dissimilar, suggesting distinct pathophysiologies between the two types of cancer. This could signal the need for different treatment approaches and led the researchers to caution against lumping EAC and ESCC in the same clinical trials.

Molecular subtype analysis of the ESCC cancers showed three molecular subtypes: ESCC1 (50 tumors), ESCC2 (36) and ESCC3 (4), distinguished by their mutation types. ESCC1, for example, was characterized by alterations in the NRF2 pathway, mutations in which are associated with poor prognosis and resistance to chemotherapy.

The three subtypes also showed trends for geographic associations, with Vietnamese patients (the only Asian population studied) showing a predominance of ESCC1 (27/41), and all 4 ESCC3 tumors being derived from United States patients.

The researchers also evaluated the molecular association between ESCC and human papillomavirus (HPV), which has been shown to have a pathogenic role in cervical SCC and head and neck (HN)SCC. They found that ESCC mRNA sequencing showed that ESCC-HPV transcript levels were similar to HPV-negative HNSCC tumors, diminishing the likelihood of an etiological role for HPV in ESCC.

In evaluating EACs in comparison to chromosomal instability (CIN) gastric cancers, the researchers found “clear similarity between chromosomal aberrations” in the two cancer types, with a stronger similarity between EAC and CIN gastric cancers than between EAC and ESCC, further differentiating the two esophageal cancers.

“The notable molecular similarity between EACs and CIN gastric cancers provides indirect support for gastric origin of Barrett’s esophagus and EAC and indicates that we may view GEA [gastroesophageal adenocarcinoma] as a singular entity, analogous to colorectal adenocarcinoma,” the authors added.

A notable anatomic gradient showed up in the progression of DNA methylation as seen from proximal to distal GEA-CIN tumors, with the most frequent hypermethylation seen in EACs, compared with gastric CIN cancers, a significant difference.

“These molecular data show that EAC and ESCC are distinct in their molecular characteristics across all platforms tested. ESCC emerges as a disease more reminiscent of other SCCs than of EAC, which itself bears striking resemblance to CIN gastric cancer,” the researchers concluded.

The authors reported that they had no competing financial interests.

Next Article: