Incidence Trends for Human Papillomavirus-Related (HPV-R) and Unrelated (HPV-U) Head and Neck Squamous Cell Carcinoma (HNSCC) in the United States (US)
Abramson Cancer Center of the University of Pennsylvania
Last Modified: June 2, 2007
Presenter: A. Chaturvedi
Presenter's Affiliation: National Cancer Institute
Type of Session: Scientific
- Approximately 405,000 cases of head and neck cancers are diagnosed annually worldwide and approximately 40,000 are diagnosed in the U.S. annually.
- The majority are related to tobacco and alcohol use (75%).
- However, head and neck squamous cell carcinoma (HNSCC) is a heterogeneous disease.
- Certain subsets of HNSCC have been related to HPV (HPV-R). These include SCC of the base of tongue (BOT), tonsil, and the oropharynx. HPV-16 is the most common type of HPV associated with HNSCC.
- Those which are not related to HPV (HPV-U) include lip, tongue, gum, floor of mouth, palate, and pharynx.
- Changes in the rate of HPV infection may affect HNSCC incidence:
- The incidence of oropharyngeal cancer has been increasing and due to oropharyngeal cancer’s association with HPV, this implies an increasing rate of HPV infection.
- In comparison, oral cavity cancers, which are associated with tobacco and alcohol use are declining, which may reflect a decrease in tobacco use.
- The present study examined the incidence of HNSCC and the two year survival of patients with HNSCC in both HPV-R and HPV-U anatomical sites. The purpose of this study was to investigate the potential influence HPV infection has on the incidence and survival of patients with HNSCC.
- Patients with HNSCC from the Surveillance, Epidemiology and End Results (SEER) database from nine regions were collected. A total of 58,158 patients from 1973 to 2003 with HNSCC were collected.
- These patients were then divided based on anatomical site into those with HPV-R and HPV-U HNSCC based on anatomical site.
- Cancers coded as BOT, tonsil and oropharynx were classified as HPV-R HNSCC (N=16,712).
- Cancers coded as lip, tongue, gum, floor of mouth, palate, other mouth, hypopharynx, ill-defined sites of the lip, oral cavity, and pharynx were classified as HPV-U HNSCC (N=41,446).
- Patients were stratified according to radiation treatment (whether radiation was given during the initial treatment or not) and stage (local, regional, distant).
- Incidence trends were determined using joinpoint regression. The difference in two year survival between the HPV-R and HPV-U groups was determined using the life-table method.
- In HPV-R patients the age adjusted incidence of HNSCC increased significantly. In HPV-R patients with HNSCC, between 1973 and 2003 the annual percent change (APC) was 0.65. This increase in APC was most pronounced in males (APC=1.02), whites (APC=0.89), and patients of younger age (30-39 APC=1.46; 40-49 APC=1.92; 50-59 APC=0.61; 60 APC=-0.66). The APC decreased in black males and African American and Caucasian females.
- In patients with HPV-U HNSCC the incidence was found to be stable from 1973-1983 followed by a significant decrease in incidence between 1983 and 2003 (APC=-2.42). They found that this decrease occurred in both males and females as well as African American and Caucasian patients.
- Mean age of diagnosis was younger for HPV-R compared with HPV-U patients (61.1 versus 64.5; p<0.001). From 1973-2003 the age of diagnosis decreased significantly for the HPV-R HNSCC, but increased in HPV-U HNSCC patients.
- Two year overall survival (OS) improved with radiation treatment in patients with HPV-R with any stage of HNSCC between 1993 and 2003. There was not a significant difference between the radiation and non-radiation groups from 1973 to 1982 and from 1983 to 1992. However, when patients with HPV-U HNSCC were examined from 1973-2003 radiation appeared to have an adverse effect on overall survival.
- The authors note that this study is limited by the use of retrospective data. Furthermore, anatomical site was used to divide patients into HPV-R and HPV-U HNSCC rather than actual assaying for HPV.
- However, despite these limitations, the authors suggest that the younger age of onset and increasing incidence of HPV-R HNSCC may be related to changing sexual behaviors.
- The decrease in HPV-U HNSCC may be related to decreases in tobacco use.
- They suggest that improved outcomes and better prognosis seen in more recent HNSCC studies may be related to a greater proportion of HPV-R patients which are more radiosensitive.
- This study provides some interesting data demonstrating an increased incidence of BOT, tonsillar and oropharyngeal HNSCC in the United States between 1973 and 2003. As noted by the authors this study was limited by the inability to check pathological specimens for expression of HPV related proteins. Furthermore, there are other variables, such as HIV status which may have contributed to changes in HNSCC incidence during the time period between 1973 and 2003 in younger men, which were not accounted for. Another variable of interest would be previous alcohol and tobacco use of the patients. The improved outcome associated with HPV-R HNSCC may be related to an inherent radiosensitivity of these anatomical sites, regardless of HPV infection. HPV has also been associated with more advanced HNSCC, however, this may be due to a higher rate of nodal involvement in these anatomical sites.
- None the less, this study provides an interesting basis to further explore whether this increased incidence is, in fact, due to increasing rates of oral HPV infection. This study supports the use of assays for HPV expression to help direct treatment and determine prognosis in HNSCC patients. Furthermore, it provides a baseline that can be used to determine the effects that the HPV vaccine may have on the incidence of HNSCC, particularly the HPV-R HNSCC. It also suggests that differences in treatment outcomes in older studies compared with newer series may be confounded by an increasing incidence of HPV. These results also suggest that there is potential benefit to vaccinating men to reduce their risk of HNSCC, however clearly further studies are needed.