Tumors from patients with gastric cancer can be divided into subgroups, based on their gene expression pattern. This information can be used to select the best treatment, according to Patrick Tan and colleagues in the August issue of Gastroenterology.
Gastric tumors have large, inter-individual differences in aggressiveness, histopathology features, and responses to therapy, yet most patients are managed alike, resulting in diverse outcomes. Approaches to classify tumors into biologically and clinically similar subgroups are needed. Several histopathologic classifications have been developed, but none of these can identify which patients are most likely to respond to specific therapies.
Tan et al. analyzed gene expression profiles for 37 gastric cancer cell lines to identify subtypes. They then looked to see if these subtypes existed among gastric tumors from 521 patients, in 4 independent cohorts. The authors also analyzed subtype-specific immunohistochemical markers (LGALS4, CDH17). They compared the resulting profiles with sensitivity of gastric cancer cells to the chemotherapeutic agents 5-fluorouracil (5-FU), cisplatin, and oxaliplatin.
Based on the expression pattern of 171 genes, the authors identified 2 major subtypes, which they named G-INT and G-DIF. Genes upregulated in the G-INT subtype included those that regulated carbohydrate and protein metabolism (FUT2) and cell adhesion (LGALS4, CDH17). Genes associated with the G-DIF subtype included those that regulated cell proliferation (AURKB) and fatty acid metabolism (ELOVL5).
Tan et al. then identified 270 patients with gastric tumors of one of these subtypes. They found that patients with G-DIF tumors had shorter survival times than patients with G-INT tumors; they validated this finding in a separate cohort of 65 patients. “This difference in survival remained significant even after adjusting for other clinical parameters, such as disease stage” said Tan in a video abstract that accompanies the article.
In vitro, the G-INT cell lines were significantly more sensitive to 5-FU and oxaliplatin, but more resistant to cisplatin, than the G-DIF cell lines. The magnitude of different sensitivities was modest (3- to 5-fold differences in responses to the drugs), but Tan et al. thought that these differences might be significant in patients, given the relatively small therapeutic window associated with cytotoxic chemotherapy.
In support of the clinical relevance of these differences, they found that significantly more patients with the G-INT subtype survived following 5-FU–based therapy than patients with the G-DIF subtype.
Why were Tan et al. able to classify gastric tumors into subgroups associated with survival and response to treatment when previous studies were not? They propose that other classification systems focused on the characterization of entire primary tumors, which contain many different cell types (tumor cells, fibroblastic or desmoplastic stroma, blood vessels, and immune cells). Given the high level of tumor tissue complexity, Tan et al. reasoned that subtle differences in cells, among and within tumors, could lead to differences in classification between observers, and ultimately pose difficulties for standardization across different centers. They therefore used an alternate strategy of analyzing gastric cancer cell lines, rather than primary tumor samples.
Tan et al. acknowledge that this approach could miss rare subtypes, and that analysis of cell lines excludes factors in the tumor microenvironment that affect progression and treatment response. In the video abstract (above), Tan says that the group has initiated a prospective trial in which patients with gastric cancer will be profiled before treatment; treatment will then be determined based on the resulting gene expression patterns.
More Information on Gastric Cancer:
- NCI Websites on Gastric Cancer and Treatment
- National Library of Medicine Website on Gastric Cancer
Read the article online.
Tan IB, Ivanova T, Lim KH, et al. Intrinsic subtypes of gastric cancer, based on gene expression pattern, predict survival and respond differently to chemotherapy. Gastroenterology 2011;141:476-485.e11.