Genomic Alterations Characterized in Squamous Cell Cancers of the Head and Neck

Oncology & Biotech News, June 2013, Volume 7, Issue 6

Squamous cell carcinomas of the head and neck have different patterns of genetic alterations, some of which may be actionable or druggable with available agents or drugs in development.

D. Neil Hayes, MD, MPH

Squamous cell carcinomas of the head and neck (SCCHN) have different patterns of genetic alterations, some of which may be actionable or druggable with available agents or drugs in development. These were the findings of a report from The Cancer Genome Atlas (TCGA) on SCCHN presented at the 2013 AACR Annual Meeting. The SCCHN report is the 8th TCGA report on cancer types, explained D. Neil Hayes, MD, MPH, University of North Carolina Lineberger Comprehensive Cancer Center in Chapel Hill, North Carolina. Hayes said that this study should be useful as a landmark research tool for SCCHN.

The comprehensive genomic analysis was based on tumor tissue and healthy tissue obtained from 279 patients with previously untreated SCCHN. Tobacco use and HPV infection are major risk factors for SCCHN; 80% of these patients had a history of smoking, and 13% had HPV-positive tumors.

Smoking-related cancers typically are diagnosed when patients are node-positive and 5-year survival is about 50%. People with HPV-positive SCCHN have a better prognosis, with 5-year survival rates of up to 90%. However, standard treatment for HPV-positive SCCHN is highly toxic, and there is an incentive to identify targeted therapy with fewer side effects.

Hayes and his team identified more than 30 sites of significant alterations in genes or genetic regions in patients with SCCHN, and 15 significantly mutated genes that included CDKN2A, TP53, PIK3CA, NOTCH1, HRAS, and NFE2L2. Most of the genetic aberrations overlapped with findings in squamous cell lung carcinoma.

“This is not surprising. Both of these cancers are related to tobacco. The shared genetic alterations between SCCHN and squamous cell lung cancer suggest that these properties go beyond the type of cancer, and squamous cell cancers could be studied at a model systems level,” Hayes said.

Key observations of the TCGA study were related to HPV status. Patients with HPV-positive SCCHN have infrequent EGFR gene amplification, and also have a high rate of PIK3CA gene mutations, but these patients almost never have TP53 alterations. PIK3CA mutations were activated in about 21% of HPV-positive tumors, which was the highest frequency of a druggable mutation in this study. PIK3CA inhibitors are being developed for breast cancer and may have role in treating SCCHN with these mutations, Hayes suggested.

Other investigators have previously shown that SCCHN tumors organize themselves into four groups based on genetic alterations, and Hayes’s report confirms this observation. These subtypes are:

  • Atypical—with no amplification of EGFR, HPV-positive, and a high rate of PI3 kinase (PIK3CA)
  • Classical—also seen in squamous cell lung cancer, and associated with two key mutations, KEAP1 and NFE2L2
  • Mesenchymal—mostly mutations of FGFR1 and FGFR2
  • Basal—highly associated with SOX2 amplifications and overexpression

In general, HPV-positive patients have a better outcome than HPV-negative patients. However, Hayes and his colleagues identified several druggable mutations in HPV-negative patients, including EGFR and FGFR.

“There is a wealth of data, and now we will struggle with making sense of it. Our observations are an advance but make things more complex. We are able to recognize patterns in genetic alterations, and some of these patterns may turn out to be druggable, or actionable in the future,” Hayes said.

He added that the findings about infrequent EGFR amplification in HPV-positive patients raise questions about the effectiveness of EGFR inhibitors approved by the FDA in this setting. “The findings also suggest that PIK3CA inhibitors in development may be of benefit for these patients. However, any treatment conclusions should be based on treatment data, and the TCGA study did not include treatment data,” he noted.

Hoadley KA, Fan C, Wilkerson MD, et al. Multi-tumor analysis of TCGA data identifies expression commonalities across tumor types. Presented at: the AACR Annual Meeting 2013; April 6-10, 2013; Washington, DC. Abstract SY12-03.