Two studies reveal genetic mutations often present in the most common form of head and neck cancer, offering a picture of how the cancer develops and how therapeutics could treat it.

Two independent studies by teams of Howard Hughes Medical Institute (HHMI) scientists have revealed genetic mutations often present in the most common form of head and neck cancer—squamous cell carcinoma. Both groups found an unexpected role for NOTCH1, a gene implicated in other cancer types, and discovered a number of mutations never identified before. Although some of the mutations were seen only rarely, many pointed to misregulation of one common pathway, which normally controls cell differentiation. The findings offer a picture of how head and neck squamous cell carcinoma (HNSCC) develops, and how therapeutics could treat it.

“The vast majority of the genes we’ve discovered here are totally new,” says HHMI investigator Todd R. Golub, who led one of the studies. “If you’d polled the world’s experts on head and neck cancer about what genes they thought were important to study, they wouldn’t have listed these. That’s why this unbiased, data-driven way of looking at a cancer is so important.”

If you’d polled the world’s experts on head and neck cancer about what genes they thought were important to study, they wouldn’t have listed these. That’s why this unbiased, data-driven way of looking at a cancer is so important.

Todd R. Golub

Head and neck cancers include cancer of the lips, mouth, sinuses, pharynx, and larynx. They are often not diagnosed until they spread to nearby lymph nodes, and the five-year survival rate after diagnosis is only 50 percent. The predominant risk factor for HNSCC is smoking—in Golub’s study, for example, 89 percent of the patients with HNSCC tumors had a history of tobacco use. Alcohol intake and infection with the human papilloma virus (HPV) also increase one’s odds of developing HNSCC.

Golub and colleagues, including senior scientists Levi Garraway and Jennifer Grandis, who looked at the genetics of tumors from 74 HNSCC patients. Golub and Garraway are at the Dana-Farber Cancer Institute and the Broad Institute of MIT and Harvard; Grandis is at the University of Pittsburgh. HHMI investigator Bert Vogelstein and a team at The Johns Hopkins University School of Medicine and the M.D. Anderson Cancer Center, led by Kenneth Kinzler and Jeffrey Myers, respectively, analyzed 32 HNSCC tumors and validated their results in an additional 88 tumors. The results of the studies appear in two separate papers published online in Science Express on July 28, 2011.

Both studies sequenced the exomes—the part of the genome that encodes proteins—in head and neck tumors. They compared each tumor genome with a non-cancerous sample from the same patient to identify mutations unique to the tumors. Then they compared the data across all the tumor samples.

“What’s becoming clear is that it’s very important to look not at a single tumor but across a collection of tumors to see recurrent patterns emerge from the complex data. Nowhere is this more needed than in head and neck cancer,” says Golub. “This classification covers a wide range of cancer sites. So to what extent are these tumors the same, despite arising in different parts of the oral cavity?”

The first similarity that both groups pinpointed was that a high percentage of the tumors had mutations in NOTCH1, which has been previously implicated in a number of human tumor types, including colon, breast, and pancreatic cancers. However, in those cancers, NOTCH1 is upregulated, expressed at higher levels than in normal tissue. This means it’s been classified as an oncogene, a gene that promotes tumor formation when it’s turned on. Inhibitors of the Notch pathway are currently being evaluated in clinical trials for their ability to treat many of these cancers.

Both Golub and Vogelstein’s work on HNSCC tumors showed NOTCH1 turned down in the cancerous cells. So in the case of HNSCC, it’s acting as a tumor suppressor gene, a gene that keeps HNSCC tumors from forming if it’s activated. HHMI Early Career Scientist Iannis Aifantis showed earlier this year that Notch1 plays a similar tumor suppressor role in preventing chronic myelomonocytic leukemia, a cancer of the white blood cells.

“This finding reinforces the point that in order to understand how cancer genes work, it is important to study them in the tumors in which they are mutated. The most basic functions of cancer genes vary from cell type to cell type and from cancer to cancer,” says Vogelstein. The startling finding that NOTCH1 works both ways suggests that therapeutics being developed for other cancers that aim to turn down NOTCH1 would not be helpful, and might even be harmful, in HNSCC.

While many of the other mutations discovered in the range of HNSCC tumors didn’t appear as frequently as NOTCH1, they were in genes that all funnel into one molecular pathway. The genes all had different functions in the differentiation of stem cells into specific types of squamous cells. Squamous cells include skin cells, but also the cells that make up the linings of the oral cavity, lungs, digestive system, and blood vessels.

Overall, both teams found a mutation rate similar to other smoking-related cancers. HNSCC associated with HPV had a lower mutation rate, something also seen in other tumor types. The results of both studies found no significant differences between any subtypes or locations of HNSCC.

The findings provide a place to start for developing therapeutics against HNSCC. In addition to treatments, however, knowing the mutations could help researchers develop ways to catch HNSCC before it spreads to lymph nodes. The Vogelstein lab is planning to study the possibility of a saliva test to catch HNSCC mutations.

“The reason to do this study is not simply to learn something interesting about cancer, but to find something useful,” says Vogelstein. “The next question is how do you use this vast amount of information to help patients?”

Scientist Profiles

For More Information

Jim Keeley 301.215.8858