Researchers Aim to Find New Pathway Interactions in Colorectal Cancers

Colorectal cancer mainly exists in people older than 50, but it also can occur in young adults who have a genetic condition called familial adenomatous polyposis (FAP). Polyps can begin to form inside the intestinal tract during their teen years. If FAP is not diagnosed and treated, they have almost a 100 percent chance of developing colorectal cancer.

While FAP is uncommon, understanding its genetic basis could lead researchers at The Children’s Hospital of Philadelphia to better approaches to several kinds of pediatric cancers. In a new study funded by the National Cancer Institute, Andrei Thomas-Tikhonenko, PhD, chief of the Division of Cancer Pathobiology, and colleagues aim to identify subsets of patients who potentially would benefit the most from emerging drugs targeting pathways that are important to colorectal cancer development.

“It’s a study that harnesses the power of cancer genomics and links it to cancer biology and cancer therapeutics,” Dr. Thomas-Tikhonenko said.

Previously, scientists had identified mutations occurring in the APC gene that could lead to FAP and other types of colorectal cancer. An important target of the APC pathway is an oncogene called MYC that as a result becomes hyperactive. Oncogenes play major roles in initiating and maintaining tumor growth. In addition to colorectal cancers, MYC has been implicated in cancers of the cervix, breast, lung, stomach, and pediatric neuroblastoma. MYC has a broad reach, and it is especially good at helping tumors to make blood vessels that are needed for oxygen and energy.

Dr. Thomas-Tikhonenko’s laboratory has dedicated many years of research to studying how MYC works in the context of cancer cell biology and how it interacts with the unique combinations of genetic alterations that appear in colorectal tumors. The Cancer Genome Atlas (TCGA) project, a publicly available catalog of genetic mutations and deletions discovered by scientists worldwide, reported in 2012 that no less than two dozen genes were mutated in a significant number of colorectal cases.

The study team is leveraging data from these cancer genome studies to determine whether or not some of these mutations are redundant, or epistatic. Basically, epistasis is the idea that one gene depends on a second gene because the two genes control a common function. The researchers hypothesize that when a cancer mutation occurs, it picks only one of the genes at work in the same pathway.

Think of epistasis as a series of landmarks along several avenues that all point to the same destination. A roadblock appears at one of the landmarks, diverting traffic. It would be redundant to erect more roadblocks at landmarks further down the same route. Instead, by strategically placing roadblocks on other avenues, the cancer can block normal organ development more efficiently and in doing so direct different aspects of tumor formation and growth.

“What we’re saying is cancer is parsimonious by nature: It doesn’t tolerate redundant mutations; they all happen for a reason,” Dr. Thomas-Tikhonenko explained. “There is stringent selection for useful mutations.”

In the case of a colorectal tumor, the researchers propose that late-stage mutations in the gene called transforming growth factor beta (TGFβ) might be influenced by seemingly unrelated earlier mutations in MYC, typically found in early-stage cancers. They have shown that TGFβ loss helps tumors to build blood vessels, just as MYC hyperactivation does. Since both drive the same tumor trait, could they be epistatic? Indeed, extensive TCGA data mining revealed that MYC and TGFβ mutations almost never occur together. Conversely, some mutations always seem to co-exist in colorectal tumors, which suggests that the mutations likely act on different, complementary pathways. This analysis was performed in collaboration with Pichai Raman, a bioinformatics scientist from the new Department of Biomedical and Health Informatics at CHOP, and Adam Bass, MD, from the Dana-Farber Cancer Institute in Boston.

Identifying which mutations occur in the same pathway and which occur in parallel is useful because it could help researchers to preselect patients who are likely to respond to certain drug therapeutics. For example, Dr. Thomas-Tikhonenko and his study team will investigate how to maximize impact of drugs within two subtypes of colorectal cancer, called chromosomal instability (CIN) and microsatellite instability (MSI). FAP falls under the CIN class.

Patients with MSI frequently have TGFβ mutations, but they are rare in patients with CIN; MYC mutations generally follow the opposite pattern. Since TGFβ and MYC mutations do not usually appear in the same tumors, the researchers predict that drugs that effectively target MYC could be helpful for FAP but would not have the same impact on patients with MSI.

“The idea is to develop an algorithm that would allow you to use this information in a way that predicts response to therapy,” said Dr. Thomas-Tikhonenko, who also is a professor of Pathology and Laboratory Medicine at the Perelman School of Medicine at the University of Pennsylvania.

By the completion of the five-year grant, he expects that scientists will have a new vantage point for colorectal cancer development and treatment. Dr. Thomas-Tikhonenko is looking forward to collaborating on the project with Struan Grant, PhD, a human genetics researcher at CHOP and an associate professor of Pediatrics at the Perelman School of Medicine at the University of Pennsylvania.