Gastric cancer prediction model could help guide therapy

Last Updated: 2017-08-01

By David Douglas

NEW YORK (Reuters Health) - Analysis of multiple data sets has helped to elucidate how four recently revealed molecular subsets of gastric cancer might best be managed, according to researchers in the United States and Korea.

Dr. Ju-Seog Lee told Reuters Health by email that the study provides "a foundation for a personalized medicine approach for treatment of patients with gastric cancer."

In an article published in Clinical Cancer Research, online July 26, Dr. Lee of the University of Texas MD Anderson Cancer Center, Houston, and colleagues note that The Cancer Genome Atlas (TCGA) project has pinpointed four molecular subtypes of gastric cancer: Epstein-Barr virus (EBV), microsatellite instability (MSI), genomically stable (GS), and chromosomal instability.

To examine how subtypes might influence clinical outcomes, the team developed a genomic prediction model using a statistically defined set of multiple genes and gene-expression data from 262 patients in the TCGA project. The model was then tested in two separate cohorts totaling 699 patients.

The EBV subtype was associated with the best prognosis and the GS with the worst. Patients with MSI and CIN subtypes had poorer overall survival than those with EBV subtype but better than those with the GS subtype.

The investigators note that the findings for the EBV subtype are "consistent with a previous report showing that positive immunostaining for EBV was associated with a good prognosis."

The subtype-specific genetic signatures used in the model, say the researchers, "predicted not only survival outcomes, but also the relative benefit of adjuvant chemotherapy."

Patients with the CIN subtype showed the greatest benefit from adjuvant chemotherapy (hazard ratio for recurrence-free survival, 0.39) and those with GS showed the least (HR, 0.83).

"Frequently altered genes in the GS subtype," the researchers write, "might account for chemoresistance. They also note that "further development of the prediction model will be necessary before it can be implemented into routine clinical practice."

In an email, Dr. Lee said, "The new classification of gastric cancer will help determine which patients will have benefit from different treatments such as immunotherapy or chemotherapy."

SOURCE: http://bit.ly/2vkH86f

Clin Cancer Res 2017.