They’re addressing a challenge posed by the complexity of colon and rectal cancer, which leads to major differences in responses to treatment and patient survival.
“Some cancers have two or three very well-defined biomarkers that, if found in a patient’s tumor, can be reliably used to guide treatment,” says project co-leader Jeffrey Morris, Ph.D., professor of Biostatistics. “For many cancers, it’s not that simple, and colorectal cancer is one of those.”
Morris, co-leader Dipen Maru, M.D., professor of Pathology, and colleagues are working to impose some clarity on the problem of heterogeneity — a fancy word for the genetic and molecular diversity found among cancer cells in the same tumor, across tumors in the same patient, and in the results of various genetic tests that seek to characterize cancer.
The team is working with the Moon Shots Program’s Cancer Genomics Laboratory platform to apply an integromic approach — an integrated analysis of genetic variations, expression of genes and regulation of genes by nongenetic factors — in 200 tumor samples.
Morris says the researchers have whittled down the 500-gene beginning set to a 100-gene test that accurately places tumors in one of the four categories 94% of the time. They’re working to get the number of genes down much lower, which would make a final test much less expensive.
While hints about potential treatment are emerging from the characteristics of the four categories, Morris says much work remains to get to the point of treating patients based on tumor category.
“This information could help us do a better job choosing among standard therapies. We have some preliminary ideas there to pursue,” he says. “And we’ll try to identify and validate some targeted-therapy strategies.”
Preclinical research, in conjunction with the Moon Shots Program Center for Co-Clinical Trials, has developed cell lines and mouse models from patient-derived tumor samples to better understand each category’s mechanisms and to apply the gene test to these tumors.
The new categories were described by MD Anderson investigators and five other research groups, which make up the Colorectal Cancer Subtyping Consortium, in a paper that was published in Nature Medicine late in 2015. The investigators folded six independent classification systems into the four consensus subgroups.
The consensus molecular subtypes of colorectal cancer, which come from their analysis of 18 studies involving more than 4,000 patients, are:
CMS1 – Genetic/immune, hypermutation with a strong immune response, 14% of tumors
CMS2 – Canonical, classic tumor of epithelial tissue, the lining of the organ, with specific tumor-promoting pathways activated, 37% of tumors
CMS3 – Metabolic, epithelial tumors with metabolic defects, 13%
CMS4 – Mesenchymal, with prominent transforming growth factor-beta activation, invasion of supportive tissue and generation of new blood vessels, 23%
About 13% of tumors in the study had mixed features, which the authors note could represent a transition from one category to another or major molecular heterogeneity within the tumors.
Colorectal cancer in general has resisted immunotherapies, but those in CMS1 may be vulnerable to this approach.
Interestingly, a drug that targets a vital metabolic pathway in some tumors that’s under development at the Institute for Applied Cancer Science, the moon shot small-molecule drug development platform, has potential against CMS3.
And the researchers note that the prime characteristics of CMS2 and CMS4 lend themselves to possible targeted therapy.
“It won’t be as simple as one target per group,” Morris says, “These classifications provide a foundation for improved prognostics, and for identifying new targets in well-defined groups of patients that we can move into clinical trials quickly.”
Read about the latest progress in Making Cancer History® at Cancer Frontline.