Overview of HNC Epidemiology

There are more than 600,000 incident cases of HNC worldwide each year, including ∼263,020 oral cavity, 213,179 thyroid gland, 150,677 larynx, 136,622 pharynx, and 84,441 nasopharynx cases per year. Nasopharyngeal cancers are often considered separately because of their distinct cause, related largely to infection with Epstein-Barr virus (EBV). Thyroid cancers are also often distinguished from other HNCs because they are associated with separate risk factors and different causal pathways.

The incidence of oral cavity, larynx, and pharynx cancer varies widely by geographic region and gender ( Fig. 1 ).

• •
  

  HNC, ∼90% of which are squamous cell carcinomas (HNSCC), has a 2-fold to 9-fold higher incidence among men than women.

  
  
• •
  

  The age-standardized incidence of HNC is increased in South-Central Asia (14.1 per 100,000) and in Melanesia (a region in the western Pacific Ocean, 22.1 per 100,000), because of high rates of tobacco use in these areas.

  
  
• •
  

  HNC incidence rates in the United States (9.9 per 100,000) and Europe (10.8 per 100,000) are lower than that observed in South-Central Asia and Melanesia, but remain notable.

  
  
• •
  

  Compared with the United States, HNC incidence is significantly lower in Micronesia (1.8 per 100,000), Middle Africa (3.4 per 100,000), and Eastern Asia (3.5 per 100,000), as well as throughout Western Africa (3.9 per 100,000) and Central America (4.7 per 100,000).

World map from Globocan depicting age-standardized incidence rates (per 100,000 people) of cancers arising from the lip, oral cavity, and pharynx.

( Data from Ferlay J, Shin HR, Bray F, et al. GLOBOCAN 2008 v1.2, Cancer Incidence and Mortality Worldwide: IARC CancerBase No. 10 [Internet]. Lyon, France: International Agency for Research on Cancer; 2010. Available at: http://globocan.iarc.fr . Accessed August 14, 2011.)

In the United States, an estimated 49,260 new cases of HNC were diagnosed in 2010. Similar to worldwide differences in HNC by gender, the incidence of HNC in the United States is higher in men than women (15.9 vs 6.2 per 100,000). The median age at diagnosis of patients with HNC in the United States is in the seventh decade of life; however, the median age of onset is lower for cancers of the nasopharynx (55 years) and tonsil (57 years).

Temporal Trends in HNC Incidence

Much of the geographic variability in HNC incidence may be explained by differential exposure to the major risk factors for HNC. Tobacco and alcohol exposure remain the most important risk factors for HNC overall, and are responsible for ∼72% (95% confidence interval [CI]: 61%–79%) of HNC in the United States and Europe. As a result of successful tobacco control programs, many industrialized countries have observed a decrease in per-capita cigarette consumption over the past several decades, followed by a decrease in the overall incidence of HNC. In the United States, cigarette consumption peaked in the 1960s; current tobacco use has been declining among both men and women, although use remains more prevalent among men.

Following a decline in tobacco use, the incidence of cancer at many HNC sites has decreased in the United States. For example, between 1974 and 1999, there was a significant decrease in :

• •
  

  Lip cancer (from 2.7 to 1.1 cases per 100,000 people; P <.05)

  
  
• •
  

  Oral cavity cancer (from 3.6 to 2.7 per 100,000 people, P <.05)

  
  
• •
  

  Hypopharyngeal cancer (from 1.0 to 0.8 cases per 100,000 people; P <.05)

  
  
• •
  

  Laryngeal cancer (from 5.4 to 4.5 cases per 100,000 people; P <.05).

In contrast to these decreases, the overall incidence remained stable over the same period for :

• •
  

  Nasopharyngeal: ∼6 per 100,000 cancers

  
  
• •
  

  Oropharyngeal: ∼1.5 per 100,000 cancers.

In Europe, the incidence of oral cavity and oropharyngeal cancers has varied greatly by country and gender, with some countries experiencing increases in the age-standardized incidence rates over the last decade :

• •
  

  Males: England, Wales, Czech Republic

  
  
• •
  

  Females: France, Netherlands, England, Wales, Denmark.

Other countries have observed decreasing incidences in rates over the last decade :

• •
  

  Males: Northern Ireland, Ireland, France, Finland, Norway, Slovenia, Spain

  
  
• •
  

  Females: Malta.

Laryngeal cancer has been decreasing or remained stable in males and females from most European countries, with the exception of Swedish women, among whom it increased from 0.2 to 0.3 per 100,000 between 1994 and 2005.

Changing Oral Cancer Incidence & Mortality in Selected Populations

Beginning in the 1980s, numerous reports from the United States and Europe began to document an increase in the incidence and mortality of tongue malignancies (both oral tongue and base of tongue) primarily among men younger than age 45 years. An increase in tongue carcinoma incidence has now been noted in multiple countries, including Scandinavian, Finnish, and US populations. Furthermore, these increases are consistently shown to be most prominent among young (age 30–50 years) individuals. Among 30-year-old to 39-year-old men in the United States, the rate of mortality from tongue cancer increased 3-fold (0.3 vs 0.9, per 100,000 P<. 001) between 1973 and 1984.

Similarly, between 1973 and 2007, the incidence of oral tongue SCC increased in both male (annual percent change [APC] = +1.6, P<. 05) and female (APC = +4; P<. 05) patients younger than 44 years, although it was stable when not stratified by age and sex (APC = +0.1; P = not significant). The increasing incidence of oral cancers has been observed among multiple countries, always in the younger age groups. These increases are in stark contrast to the decreasing incidence of some other HNC subsites.

Overview of HNC Epidemiology

There are more than 600,000 incident cases of HNC worldwide each year, including ∼263,020 oral cavity, 213,179 thyroid gland, 150,677 larynx, 136,622 pharynx, and 84,441 nasopharynx cases per year. Nasopharyngeal cancers are often considered separately because of their distinct cause, related largely to infection with Epstein-Barr virus (EBV). Thyroid cancers are also often distinguished from other HNCs because they are associated with separate risk factors and different causal pathways.

The incidence of oral cavity, larynx, and pharynx cancer varies widely by geographic region and gender ( Fig. 1 ).

• •
  

  HNC, ∼90% of which are squamous cell carcinomas (HNSCC), has a 2-fold to 9-fold higher incidence among men than women.

  
  
• •
  

  The age-standardized incidence of HNC is increased in South-Central Asia (14.1 per 100,000) and in Melanesia (a region in the western Pacific Ocean, 22.1 per 100,000), because of high rates of tobacco use in these areas.

  
  
• •
  

  HNC incidence rates in the United States (9.9 per 100,000) and Europe (10.8 per 100,000) are lower than that observed in South-Central Asia and Melanesia, but remain notable.

  
  
• •
  

  Compared with the United States, HNC incidence is significantly lower in Micronesia (1.8 per 100,000), Middle Africa (3.4 per 100,000), and Eastern Asia (3.5 per 100,000), as well as throughout Western Africa (3.9 per 100,000) and Central America (4.7 per 100,000).

World map from Globocan depicting age-standardized incidence rates (per 100,000 people) of cancers arising from the lip, oral cavity, and pharynx.

( Data from Ferlay J, Shin HR, Bray F, et al. GLOBOCAN 2008 v1.2, Cancer Incidence and Mortality Worldwide: IARC CancerBase No. 10 [Internet]. Lyon, France: International Agency for Research on Cancer; 2010. Available at: http://globocan.iarc.fr . Accessed August 14, 2011.)

In the United States, an estimated 49,260 new cases of HNC were diagnosed in 2010. Similar to worldwide differences in HNC by gender, the incidence of HNC in the United States is higher in men than women (15.9 vs 6.2 per 100,000). The median age at diagnosis of patients with HNC in the United States is in the seventh decade of life; however, the median age of onset is lower for cancers of the nasopharynx (55 years) and tonsil (57 years).

Temporal Trends in HNC Incidence

Much of the geographic variability in HNC incidence may be explained by differential exposure to the major risk factors for HNC. Tobacco and alcohol exposure remain the most important risk factors for HNC overall, and are responsible for ∼72% (95% confidence interval [CI]: 61%–79%) of HNC in the United States and Europe. As a result of successful tobacco control programs, many industrialized countries have observed a decrease in per-capita cigarette consumption over the past several decades, followed by a decrease in the overall incidence of HNC. In the United States, cigarette consumption peaked in the 1960s; current tobacco use has been declining among both men and women, although use remains more prevalent among men.

Following a decline in tobacco use, the incidence of cancer at many HNC sites has decreased in the United States. For example, between 1974 and 1999, there was a significant decrease in :

• •
  

  Lip cancer (from 2.7 to 1.1 cases per 100,000 people; P <.05)

  
  
• •
  

  Oral cavity cancer (from 3.6 to 2.7 per 100,000 people, P <.05)

  
  
• •
  

  Hypopharyngeal cancer (from 1.0 to 0.8 cases per 100,000 people; P <.05)

  
  
• •
  

  Laryngeal cancer (from 5.4 to 4.5 cases per 100,000 people; P <.05).

In contrast to these decreases, the overall incidence remained stable over the same period for :

• •
  

  Nasopharyngeal: ∼6 per 100,000 cancers

  
  
• •
  

  Oropharyngeal: ∼1.5 per 100,000 cancers.

In Europe, the incidence of oral cavity and oropharyngeal cancers has varied greatly by country and gender, with some countries experiencing increases in the age-standardized incidence rates over the last decade :

• •
  

  Males: England, Wales, Czech Republic

  
  
• •
  

  Females: France, Netherlands, England, Wales, Denmark.

Other countries have observed decreasing incidences in rates over the last decade :

• •
  

  Males: Northern Ireland, Ireland, France, Finland, Norway, Slovenia, Spain

  
  
• •
  

  Females: Malta.

Laryngeal cancer has been decreasing or remained stable in males and females from most European countries, with the exception of Swedish women, among whom it increased from 0.2 to 0.3 per 100,000 between 1994 and 2005.

Changing Oral Cancer Incidence & Mortality in Selected Populations

Beginning in the 1980s, numerous reports from the United States and Europe began to document an increase in the incidence and mortality of tongue malignancies (both oral tongue and base of tongue) primarily among men younger than age 45 years. An increase in tongue carcinoma incidence has now been noted in multiple countries, including Scandinavian, Finnish, and US populations. Furthermore, these increases are consistently shown to be most prominent among young (age 30–50 years) individuals. Among 30-year-old to 39-year-old men in the United States, the rate of mortality from tongue cancer increased 3-fold (0.3 vs 0.9, per 100,000 P<. 001) between 1973 and 1984.

Similarly, between 1973 and 2007, the incidence of oral tongue SCC increased in both male (annual percent change [APC] = +1.6, P<. 05) and female (APC = +4; P<. 05) patients younger than 44 years, although it was stable when not stratified by age and sex (APC = +0.1; P = not significant). The increasing incidence of oral cancers has been observed among multiple countries, always in the younger age groups. These increases are in stark contrast to the decreasing incidence of some other HNC subsites.

Evidence that HPV Causes a Subset of Oropharyngeal Cancers

Several different oncogenic HPV types have been identified in case series of HNSCC tumors ; however, greater than 90% of these HPV-HNSCC are associated with a single HPV type, HPV-16. Estimates of the proportion of HNSCC that are caused by HPV have varied widely, in part because of differences in case subsite composition, calendar year, country of study, and differences in tumor HPV detection methods. A systematic review of studies found that 26% of HNSCC have HPV DNA detected using polymerase chain reaction, although this prevalence differs substantially by head and neck subsite.

Over the past decade, it has become clear that the increasing incidence of oropharyngeal cancers in developed nations is being driven by a new cause. There is clear molecular and epidemiologic evidence supporting a causal role for HPV in a subset of HNSCC. HPV is detected in the tumor cell nuclei, where it is transcriptionally active, clonal, and not found in the surrounding benign tissue. In addition, case-control studies have shown that cases are more likely than matched controls to have current HPV DNA detected in their exfoliated oral cells, to have a higher number of lifetime oral sex partners (a surrogate for oral HPV exposure), and to have antibodies for HPV oncogenes.

Studies have consistently shown that most HNSCC with HPV detected in tumor are from the oropharynx. The virus has an affinity for the lymphoepithelium of the Waldeyer ring, and in studies of HNSCC patients has been detected most frequently in oropharyngeal cancers. Most oropharyngeal cancers in the United States and Europe are now caused by HPV infection, although there is wide variation in the proportions reported by different US (36%–73%) and European (14%–93%) studies.

The proportion of oropharyngeal cancers caused by HPV seems to have increased over the past 30 years, suggesting HPV is driving the increasing incidence of oropharyngeal cancer recently observed. A large Swedish study reported the prevalence of HPV detected in tonsil cancer increased from 23% in the 1970s, to 68% in 2000 ( P <.001). Additional increases of oropharyngeal cancers in Sweden being associated with HPV have been observed in the past 10 years :

• •
  

  68%: 2000 to 2002

  
  
• •
  

  77%: 2003 to 2005

  
  
• •
  

  93%: 2006 to 2007.

Similar trends have been reported in several other studies, including a study of US SEER tissues which showed that only 16% of US oropharyngeal cases had HPV detected in 1984 to 1989 compared with 73% during 2000 to 2004, a 4-fold increase in only 2 decades. This study further suggested that the annual number of HPV-related oropharyngeal cancers in the United States is projected to surpass the number of incident invasive cervical cancer cases in the United States by 2020.

Does HPV Cause Cancer at Nonoropharyngeal HNC Sites?

A causal role for HPV among oropharyngeal cancers is now established, but it is less clear whether HPV may cause HNSCC at other sites. Case-control studies evaluating the association of HPV DNA in oral exfoliated cells and odds of oral cavity or other nonoropharyngeal HNSCCs have most commonly been null or small in magnitude, unlike the strong associations observed for oropharyngeal cancers. However, a role for HPV in a small subset of these cancers cannot be excluded because the etiologic heterogeneity of these nonoropharyngeal HNSCC sites would be more difficult to observe and some studies have suggested an association. Initial studies have reported HPV detection in a subset of oral cavity, laryngeal, and nasopharyngeal cancers, but the consistency and strength of evidence implicating HPV at these sites are notably less than in the oropharynx.

HPV in oral cavity cancers

HPV DNA has been detected in ∼25% of oral cavity cancers, although the results of individual studies have varied greatly ( Fig. 2 ). It is likely that these are overestimates, because some HPV-related oral cavity cancers show clear oropharyngeal involvement and some of these oral cavity cancers are extensions from base-of-tongue primaries ; thus the proportion of oral cavity cancers that are HPV-related is currently unknown. Several case-control studies have evaluated the association of HPV antibodies, biopsy specimens, or HPV DNA in oral epithelial cells. The results of these studies have been varied, with some reporting a significant association between HPV and increased odds of oral cavity cancer, whereas others studies found no significant association. Therefore, it remains unclear whether HPV is a cause of a subset of oral cavity carcinomas.

Prevalence of HPV in tumors derived from the oropharynx, oral cavity, larynx, and hypopharynx. Restricted to studies with greater than 25 subjects (oropharynx, oral cavity, larynx) or 15 subjects (hypopharynx) and which began enrollment after 1990. The size of each circle is proportional to the sample size of cases in that study.

For continuity with above formatting, we suggest you insert a subheading here with the following title “HPV in nasopharyngeal carcinomas.” Recent studies have also begun to explore a potential role of HPV in nasopharyngeal carcinomas (NPCs). NPCs have classically been associated with EBV, but HPV has also been detected in a subset (5%–83%) of NPC. However, the few studies that have investigated HPV infections in normal nasopharyngeal epithelium have reported similar HPV estimates in benign nasopharyngeal tissues. A recent study by Maxwell and colleagues found that among 5 patients with NPC from a North American population, 4 harbored detectable HPV DNA and high p16 expression but lacked EBV-encoded RNA. These findings are consistent with a potential role for HPV in a subset of NPC from this population, but do not prove a causal relationship. In another recent investigation, Huang and colleagues performed a case-control study and found no association between HPV and NPC in Taiwan. Therefore, there is insufficient evidence to support a causal association between HPV and NPC. Further studies are necessary to investigate this potentially important association. Nonetheless, given the potential for tonsillar carcinomas to extend into the nasopharynx, misclassification remains a challenge for these studies.

Role of HPV in the Changing Incidence of HNSCC

Given the strong role of HPV in oropharyngeal cancer and the small role of HPV in other HNSCC subsites, several studies have evaluated incidence trends after stratifying head and neck sites into those sites most likely to be HPV-related (subsites of the Waldeyer ring, including soft palate, palatine, and lingual tonsils) and those HNSCC sites where an HPV cause is unlikely (oral cavity, oral tongue). Chaturvedi and colleagues found that between 1973 and 2004, the US incidence of HPV-HNSCC increased significantly (APC = +0.80, P<. 001), especially among younger males (US incidence trends are depicted in Fig. 3 ). In contrast, the incidence of HPV-unrelated HNSCC decreased significantly during this same period (APC = –1.85, P<. 001).

Age-adjusted incidence of HNSCC that were diagnosed between 1973 and 2006 in males and females at HPV-related ( A : males and C : females) and HPV-unrelated sites ( B : males and D : females). HPV-related sites defined by International Classification of Disease for Oncology version 3 (ICD-3) codes for lingual tonsil, palatine tonsil, base of tongue, oropharynx, and Waldeyer ring. HPV-unrelated sites were defined by ICD-3 codes for unspecified and other sites of the oral cavity, including tongue, gum, floor of mouth, and palate.

( From Marur S, D’Souza G, Westra WH, et al. HPV-associated head and neck cancer: a virus-related cancer epidemic. Lancet Oncol 2010;11(8):782; with permission.)

In a similar study from 15 population-based European cancer registries, incidence trends of HPV-related HNC increased significantly between 1988 and 2002 (APC = +3.37, P<. 05). HPV-unrelated HNC incidence in Europe also increased from 1988 until 1998 (APC = +1.73; 95% CI 1.2, 2.3), but seemed to plateau after 1998 (APC = –0.8; 95% CI –3.0, 1.5). Studies using Canadian and Australian cancer registries also have observed increases in oropharyngeal cancer incidence.

Oral HPV Infection

There are more than 100 different types of HPV, distinguished by variations in their genetic sequence, and more than 15 HPV types that have been associated with human cancers. HPV types are separated into high-risk oncogenic HPV types, including HPV-16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 68, and 73; the remaining types are classified as low-risk or nononcogenic. The carcinogenicity of HPV has been well described in the lower genital tract, where HPV is a necessary, but not sufficient, cause of invasive cervical cancer. Although it is clear that HPV is an important cause of oropharyngeal cancer, infection with HPV is neither a necessary nor a sufficient cause of oropharyngeal cancer (oropharyngeal cancer can occur in the absence of HPV and not all oral HPV infections lead to malignant transformation).

Several cross-sectional and case-control studies have sought to characterize the prevalence of oral HPV infections in the general population. In a recent systematic review of 18 published studies, 4.5% (95% CI 3.9, 5.1) of 4070 healthy individuals had HPV DNA detected in oral exfoliated cells. The oral HPV prevalence was similar in men (4.6%) and women (4.4%). High-risk oral HPV was detected in 3.5% of participants and oral HPV-16 infection was detected in 1.3% of all study individuals, representing approximately 28% of all oral HPV infections that were detected. A recent large multinational study of healthy men reported slightly lower prevalence of :

• •
  

  Any high-risk oral HPV infection (1.3%; 95% CI 0.8%, 2.0%)

  
  
• •
  

  HPV-16 (0.6%; 95% CI 0.2%, 1.1%) infection.

Higher oral HPV prevalence has been reported among select groups, such as:

• •
  

  Individuals infected with the human immunodeficiency virus (HIV) (33%)

  
  
• •
  

  Current smokers (10%)

  
  
• •
  

  People with more than 5 lifetime sexual partners (7.4%).

Natural History of Oral HPV Infection

Although the natural history of anogenital HPV infections and its transmission are well understood, there have been few studies investigating the natural history of oral HPV infection. Initial studies suggest that some oral HPV infections may persist after more than 2 years, although the studies had limited power or did not include analysis by HPV type. D’Souza and colleagues investigated the 6-month natural history of oral HPV infections in a high-risk longitudinal cohort of 136 HIV-positive and 63 at-risk HIV-negative individuals. In this population, prevalent oral HPV infections detected at baseline were as likely as cervical infections to persist to 6 months among HIV-negative and HIV-positive individuals. Factors that were associated with increased oral HPV persistence included :

• •
  

  Current smoking status

  
  
• •
  

  Older age

  
  
• •
  

  Lower current CD4 cell count

  
  
• •
  

  Use of highly active antiretroviral therapy (HAART) therapy

  
  
• •
  

  Duration of HAART therapy.

These factors differed from those that predicted cervical HPV persistence at 6 months. For example, tobacco use was found to significantly increase the odds of oral HPV persistence, but not cervical persistence. Although this study population differs from the general population, it suggests that the oral HPV natural history or risk factors for persistence may differ in some ways from cervical HPV infection.

The long-term natural history of oral HPV infection has not been evaluated; however, initial studies suggest that, as with genital HPV infection, many infected people may clear their oral HPV infections quickly. Further studies on oral HPV transmission, median time to clearance, and risk factors for persistence and progression are needed.

Mischaracterization of Patients with HPV-HNSCC

Patients with HPV-HNSCC are sometimes mischaracterized by the press as highly sexual or promiscuous. Although the average number of sexual partners is larger among those with HPV-related versus HPV-unrelated HNSCC, the histories of individual patients vary greatly, and it is common for patients with HPV-related tumors to report a low number (<6) of lifetime sexual partners. It is similarly important to note that only some patients with HPV-related tumors are nonsmokers/nondrinkers. Given evidence suggesting that smoking predicts higher risk of disease recurrence and distant metastasis in HPV-HNSCC, counseling for smoking cessation should remain a priority for clinicians.

Race

Racial disparities in HNSCC incidence and survival remain striking. Because HPV-HNSCCs have been shown to have improved survival compared with HPV-unrelated carcinomas, and a lower proportion of black patients diagnosed with HNSCC are HPV-related, racial differences in HNSCC survival may be in part or entirely explained by this difference in cause. Although initial studies support this possibility, suggesting survival of black and white patients with HPV-unrelated HNSCC is similar, further investigation of these disparities is needed to understand whether there are differences in oral HPV acquisition or clearance.

Tumor Biology and Classification

There is some limited evidence to suggest that HPV may be associated with a small subset of HNSCC outside the oropharynx, although the results of these studies are far from clear. Furthermore, although HPV tumor status is clearly associated with improved survival among patients with oropharyngeal carcinomas, consensus around optimal biomarker profiles for HNSCC survival has not been established. Even more unclear is how to define patients with tumors who are HPV-negative but p16-positive. Because p16 overexpression is used by some institutions as a surrogate marker for HPV infection, are these tumors HPV-related despite the lack of detection of HPV DNA with the current assays? Further understanding of the biology of these tumors and how to classify them needs to be addressed.