This article discusses risk factors, incidence trends, and prognostic considerations for head and neck cancer (HNC). The primary causes of HNC are tobacco and alcohol use, and human papillomavirus (HPV). Tobacco-related HNC incidence rates are decreasing in countries where tobacco use has declined. HPV-HNC, which occurs primarily in the oropharynx and is associated with sexual behaviors, has been increasing over the past several decades, among white men in particular. The prognosis for HNC overall has improved slightly since the 1990s, and is influenced by site, stage, and HPV status. Prognosis for HPV-HNC is significantly better than for HPV-negative disease.
Key points
- •
The two major causes of head and neck cancer (HNC) are (1) tobacco and alcohol, and (2) human papillomavirus (HPV).
- •
HPV-HNC is strongly tied to oral sexual behaviors and is most common in middle-aged white men.
- •
The epidemiology of HPV-HNC can be explained in part by variations in oral sexual behaviors by race and age cohort.
- •
Globally, incidence of HPV-negative HNC is associated with patterns of tobacco use, and is falling in countries with declining rates of tobacco smoking.
- •
Prognosis for HNC is determined by anatomic site, stage, and HPV tumor status. HPV-positive HNC has better survival than HPV-negative disease.
Introduction
Head and neck cancer (HNC) encompasses a heterogeneous group of upper aerodigestive tract malignancies that together comprise the seventh most common cancer worldwide, and ninth most common cancer in the United States. The vast majority of HNCs are squamous cell carcinomas (SCCs). More than 54,000 new cases of HNC were projected in the United States in 2014, resulting in an annual incidence of 15.0 per 100,000, with 12,000 deaths attributable to the disease.
There are two primary causes of HNC: (1) tobacco and alcohol use, and (2) human papillomavirus (HPV), a sexually transmitted infection. Historically, most HNCs have been caused by tobacco and alcohol, but over the past decade, HPV has been recognized as another important cause of HNC. Both HPV-negative and HPV-positive HNC have a higher incidence among men than women and increasing incidence with age; however, they have otherwise unique risk factor profiles, prognostic considerations, and incidence trends ( Table 1 ). HPV-HNC patients are more likely to be white, have a higher socioeconomic status (SES), and are diagnosed at a younger age than HPV-negative HNC patients. HPV-HNC patients also have significantly improved prognosis compared with those diagnosed with HPV-negative HNC. In the United States, the epidemiologic landscape of HNC is changing dramatically as the incidence of HPV-HNC is rising, and declining tobacco-smoking prevalence rates are paralleled by a decreasing incidence of HPV-negative HNC. Globally, HNC incidence trends vary widely by country, reflecting differing trends in tobacco use, alcohol use, and sexual norms across countries.
| Parameter | HPV-Negative | HPV-Positive |
|---|---|---|
| Gender | 2–3-fold more common in men than women | 4–5-fold more common in men than women |
| Age | Median age late 50s and 60s | Median age early-mid 50s, with increasing incidence among younger cohorts |
| Race | Worse prognosis in blacks | Higher incidence in whites |
| Smoking | >90% have smoking history; risk increases with increasing tobacco use | 50%–65% have smoking history |
| Alcohol use | Synergistic with tobacco in increasing risk | Not a significant risk factor |
| Sexual history | Not a significant risk factor | Number of oral sex partners is a strong risk factor |
| Site | Larynx and oral cavity most common; others include oropharynx, hypopharynx, nasopharynx, nasal cavity and paranasal sinuses | Oropharynx, specifically lymphoid tissue of tonsils and tongue base; <20% HPV-positivity at other HNC sites |
| Presentation | Varies | Enlarged cervical lymph nodes common; also oropharyngeal pain, dysphagia, referred otalgia |
| Incidence trends | Decreasing | Increasing |
| Prognosis | All sites: 5-year survival 65%, 5-year recurrence 50% Oropharynx: 5-year survival 20%–25%, 5-year recurrence 50% | Oropharynx: 5-year survival 60%–90%, 5-year recurrence 10%–15% |
This article discusses trends in risk factors, incidence, and prognosis of both HPV-negative and HPV-positive HNC. Although HNC epidemiology literature from before 2005 did not distinguish between HPV-negative and HPV-positive HNC, limiting interpretation of epidemiologic trends across that time frame, recent research has greatly broadened our understanding of the role of HPV in HNC.
Risk factors for head and neck cancer
The risk factors for HNC include some common risk factors associated with both HPV-positive and HPV-negative HNC, as well as other distinct causative factors. This section reviews risk factors for HNC overall, followed by a discussion of those specific to HPV-HNC.
Noninfectious Risk Factors for Head and Neck Cancer
Although the primary risk factors for HNC are tobacco, alcohol, and HPV, HNC etiology is multifactorial and many additional causes have been identified ( Box 1 ).
Major risk factors
Tobacco use
Alcohol
Human papillomavirus
Male gender
Older age
Betel quid
Other risk factors
Premalignant lesions, including leukoplakia and erythroplakia
Human immunodeficiency virus and/or immune suppression
Diet, including poor nutrition, low folate
Poor oral hygiene
Occupational exposures, including wood dust, asbestos, others
Epstein-Barr virus (nasopharynx; high incidence in Southeast Asia)
Sun exposure
Plummer-Vinson syndrome
Fanconi anemia
Li Fraumeni syndrome
Dyskeratosis congenita
Tobacco
Tobacco causes most HNC, and alcohol synergistically increases the risk of HNC conferred by tobacco use. Historically, approximately 90% of patients with HNC have a history of tobacco use, with a four-fold to five-fold increase in risk of developing oral cavity, oropharynx, and hypopharynx cancers, and a 10-fold increased risk of laryngeal cancer, among tobacco users. The carcinogenic effects of tobacco are dose-dependent, with the risk of HNC closely related to the frequency, duration, and intensity of cigarette smoking ( P trend <.001 for each parameter). Conversely, smoking cessation may reduce the risk of HNC, with evidence that risk of HNC decreases with increasing time since smoking cessation.
Smokeless tobacco, such as snuff or chewing tobacco, is also an important HNC risk factor, in particular for cancers of the oral cavity. There is an estimated 80% increase in risk of oral cavity cancer for individuals who have ever used smokeless tobacco, with a fourfold increase in odds of HNC among individuals who have used smokeless tobacco for at least 10 years compared with never-users. Although approximately 7% of oral cavity cancers in the United States are attributable to chewing tobacco, the attributable fraction is as high as 53% in India and 68% in Sudan, where the use of smokeless tobacco, including betel quid or areca nut with added tobacco, is popular.
Alcohol
Alcohol use independently increases the risk of HNC, with an estimated 1% to 4% of cases attributable to alcohol alone and a twofold increase in odds of HNC for drinkers who are never-users of tobacco. In particular, alcohol use increases risk of hypopharyngeal cancer when compared with other sites. However, the larger impact of alcohol is observed in the interaction of alcohol with tobacco use. The combined effects of alcohol and tobacco use have a greater than multiplicative impact in increasing the risk of cancer. Among individuals who smoke two or more packs of cigarettes and also drink more than four alcoholic drinks per day, the risk of HNC is increased greater than 35-fold.
Gender, age, and race
HNC overall is approximately twofold to threefold more common among men than women in the United States and although this ratio varies by geographic region globally, men in most countries have a twofold to fivefold greater risk of HNC than women. This difference is likely related to higher rates of substance abuse, particularly tobacco use, among men than women. The risk of HNC also increases with age, with the median age of diagnosis for most anatomic sites in the late sixth and seventh decades of life.
HNC incidence patterns by race have changed in the United States. Although the incidence of HNC has historically been higher among blacks than whites in the United States, HNC incidence among blacks has been declining since the 1990s and is currently lower than in whites (age-adjusted incidence rate of all HNCs is 14.8 per 100,000 for blacks compared with 15.5 per 100,000 for whites). This change is in part explained by the rising incidence of HPV-HNC among whites in the United States, which has not been observed among blacks.
Socioeconomic status
Indicators of low SES, such as low educational attainment and income, are strongly associated with increased risk of HNC globally. In a recent pooled analysis of 23,964 HNC cases and 31,954 controls across 27 countries, the lowest socioeconomic stratum experienced more than twofold higher risk of HNC than the highest socioeconomic stratum (education: odds ratio [OR] 2.50, 95% confidence interval [CI] 2.02–3.09, and income: OR 2.44, 95% CI 1.62–3.67). Although this risk was attenuated after adjustment for tobacco smoking and alcohol use, which are associated with low SES, approximately one-third of the risk associated with low SES remained unexplained by known risk behaviors. Even among never-users of tobacco and never-drinkers, individuals with low education had significantly increased risk of HNC (OR 1.61, 95% CI 1.13–2.31). These associations vary by country, with greater effect of SES observed in South and Central America than in North America and Europe. Reasons for the increased risk of HNC associated with low SES are unclear, but may include occupational exposures, psychosocial variables, and other unrecognized behavioral factors.
Immune suppression
Immune suppression secondary to solid organ transplantation or HIV infection is associated with an increased risk of HNC. After solid organ transplantation, incidence of lip cancer increases by 10-fold or more, with more modest 2-fold to 5-fold increases at other head and neck sites. HIV-infected patients are similarly at a twofold to fivefold increased risk for HNC, which has risen in the post-HAART (highly active antiretroviral therapy) era as patients survive longer and have the “opportunity” to develop non-AIDS–defining malignancies, including HNC.
Risk Factors for Human Papillomavirus-Positive Head and Neck Cancer
Human papillomavirus
HPV is a sexually transmitted infection that causes HPV-positive HNC (HPV-HNC), a distinct subset of HNCs that occur primarily in the oropharynx and arise from the lymphoid tissues of the palatine and lingual tonsils. The proportion of oropharyngeal cancers caused by HPV is growing, with current prevalence estimates in the United States near 80%. In contrast, HPV is detected in fewer than 20% of tumors at other anatomic sites of the head and neck, with no apparent increase in prevalence over time. Before widespread testing of oropharyngeal cancers for HPV, the oropharyngeal predominance of HPV-HNC allowed for early epidemiologic studies to use an oropharyngeal site as a surrogate for HPV-positive tumor status.
Oral HPV infection is the putative precursor to HPV-HNC. Although there are more than 100 types of HPV, only a small number of these are considered high risk or carcinogenic. One type, HPV16, is responsible for the vast majority (more than 90%) of HPV-HNCs. In the United States, the prevalence of oral HPV infection of any type in the general population is approximately 7%, and the prevalence of oral HPV16 infection is 1%. Although the natural history of progression from oral HPV infection to HPV-HNC is not well understood, it is clear that the vast majority of infections do not progress to cancer. The amount of time from first oral HPV infection to the development of cancer is unknown but is believed to be more than a decade.
Although HPV is the clear cause of HPV-HNC, other risk factors influence odds of exposure to oral HPV, as well as persistence of HPV infection after it is acquired.
Sexual behaviors and head and neck cancer
Sexual behaviors are a proxy for exposure to oral HPV infection, and as such are strongly associated with odds of HPV-HNC. In contrast, sexual behaviors are not associated with HPV-negative HNC, reflecting the distinct etiology of these 2 diseases. As sexual behaviors are highly correlated, not only higher number of oral sex partners, but also other measures of sexual behavior, such as higher number of vaginal sex partners, younger age at sexual debut, and history of genital warts, have been associated with HPV-HNC. Generational changes in sexual behavior, including the “sexual revolution” of the 1960s in the United States, are thought to have contributed to the increased incidence of HPV-HNC now observed.
Oral HPV infection has likewise been associated with various measures of sexual exposure, including number of oral and vaginal sex partners, number of rimming partners, anogenital warts, herpes simplex virus-2 seropositivity, and possibly deep kissing.
Although it is difficult to delineate which specific behaviors confer the greatest risk of oral HPV infection and HPV-HNC, oral sexual behaviors are believed to be the primary mode of transmission.
Tobacco
Individuals with HPV-HNC are significantly less likely than their HPV-negative counterparts to be smokers, however tobacco exposure remains common among patients with HPV-HNC. Indeed, one-half to two-thirds of patients with HPV-HNC report a history of tobacco use. Furthermore, tobacco smoking is associated with significantly higher oral HPV prevalence, and oral HPV16 prevalence increases in a dose-dependent fashion with higher measures of tobacco-smoking intensity. In addition to being carcinogenic in its own right, tobacco smoking is hypothesized to increase either chances of oral HPV infection when exposed, or HPV persistence, or both, and therefore may increase risk of HPV-HNC.
Demographic risk factors
HPV-HNC patients are predominantly male (85%–90%) and white (92%–95%), with a median age of 50 to 56 years. Compared with HPV-negative HNC, patients with HPV-HNC are more likely to be male and white, have higher SES, and be married (see Table 1 ). The median age of patients with HPV-positive HNC is lower than that of patients with HPV-negative HNC by 4 to 10 years in most studies. Furthermore, the incidence of HPV-HNC among younger individuals ages 36 to 44 years continues to increase. Initial research suggests differences in HPV-HNC and oral HPV infection by gender, race, and age cohort are explained by patterns of oral sexual behaviors in these groups.
Oropharyngeal cancer is nearly 5 times more common among men than women (age-adjusted incidence ratio 4.71, 95% CI 4.42–5.02), and oral HPV16 infection is 7 times more common among men than women in the United States (prevalence ratio 6.79, 95% CI 2.07–22.26). These gender differences are not entirely explained by differences in measures of oral sexual behavior, and may be due to higher risk of oral HPV infection when performing oral sex on a woman than a man.
The proportion of oropharyngeal cancers caused by HPV is higher among whites than blacks. This racial discrepancy may be explained both by a higher rate of tobacco use and thus tobacco-related oropharyngeal cancer among blacks, as well as differences in oral sexual behavior patterns by race. For example, whites are more likely to perform oral sex, and to do so at a younger age, than blacks.
Trends in head and neck cancer incidence
Head and Neck Cancer Incidence in the United States and Globally
Although HNC as whole is the seventh most common cancer worldwide, the incidence varies considerably by anatomic tumor site ( Table 2 ) and geographic region. Oral cavity and laryngeal SCC are the most common HNCs worldwide (age-adjusted standardized incidence rate [SIR] 3.9 per 100,000 and 2.3 per 100,000, respectively), with nasopharynx and other pharynx cancers less common (SIR 1.2 and 2.0 per 100,000, respectively). Notably, currently available HNC incidence statistics are limited by inconsistencies in categorization by site, with broad and/or ill-defined categories used in many cancer registries. Many registries do not currently categorize by sites that would be useful in distinguishing HPV-positive compared with HPV-negative disease (eg, oropharynx compared with nonoropharynx). Therefore, interpretation of incidence and mortality rates must be approached with caution.
| Site | Incidence: Age-Adjusted Standardized Incidence Rate per 100,000 | Mortality: Age-Adjusted Standardized Death Rate per 100,000 | 5-year Survival, a % |
|---|---|---|---|
| United States | |||
| Larynx | 3.3 | 1.1 | 60.0 |
| Tongue | 3.2 | 0.6 | 62.7 |
| Tonsil | 1.8 | 0.2 | 70.8 |
| Lip | 0.7 | 0.0 | 89.5 |
| Gum and other oral cavity | 1.5 | 0.4 | 59.7 |
| Nasopharynx | 0.7 | 0.2 | 59.2 |
| Floor of mouth | 0.6 | 0.0 | 51.4 |
| Hypopharynx | 0.6 | 0.1 | 31.9 |
| Oropharynx | 0.4 | 0.2 | 41.7 |
| Other oral cavity and pharynx | 0.2 | 0.5 | 35.4 |
| Worldwide | |||
| Lip and oral cavity | 3.9 | 1.9 | |
| Larynx | 2.3 | 1.2 | |
| Other pharynx | 2.0 | 1.4 | |
| Nasopharynx | 1.2 | 0.8 | |
a Five-year relative survival. Only available for US statistics.
HNC incidence trends over time and across countries are strongly influenced by patterns of tobacco use. As a general rule, tobacco use in most countries rises first among men, followed by an increase in smoking among women. In the United States, tobacco use prevalence peaked in the 1940s to 1960s, first in men and then in women, before declining after the 1964 Surgeon General’s report identifying tobacco as a causative agent in lung cancer. Because of a concerted public health effort, from 1965 to 2011 smoking prevalence declined from 51.9% to 19.0% among men, and from 33.9% to 17.3% among women. The impact of these trends on HNC incidence became clear when, after years of increasing numbers of HNCs per year, the incidence rate for HNCs overall and those that are particularly tied to tobacco began a slow decline in the 1990s. For example, the age-adjusted annual percentage change from 1992 to 2005 was –1.4 for oral cavity and pharynx cancer, and from 1992 to 2009 was –2.8 for larynx cancer.
Although tobacco use has declined in the United States, global patterns of tobacco use are more heterogeneous. Several economically developed countries, such as Canada and Australia, have observed decreased smoking prevalence rates over the past several decades, similar to the United States, first among men and then among women, and are experiencing significant decreases in HNC incidence. However, other countries, including many that are less economically developed, appear to be in earlier stages of the tobacco epidemic, with high smoking prevalence rates that have not yet substantially decreased. For example, Eastern Europe has some of the highest cigarette consumption rates in the world among both men and women, and in many Eastern European countries, HNC incidence is on the rise. In China, smoking prevalence among men is greater than 50%, and smoking prevalence in Thailand is 40% to 50%, whereas prevalence among women in these countries is at least 10 times lower than among men ; a dramatic rise in HNC has not yet been observed in these countries but may be anticipated, along with a rise in other tobacco-related diseases.
Risk for HPV-negative HNC is not limited to tobacco smoking, reflected by a variety of incidence trends in different countries and for different anatomic sites that do not follow smoking prevalence patterns. For example, France has much higher incidence of HNC among men than neighboring countries, which has been attributed to patterns of combined cigarette and alcohol use. The United Kingdom has had falling smoking prevalence rates among both men and women ; however, oral cavity incidence rates are significantly rising in the United Kingdom in all age groups. It is unclear whether this increase may be explained by trends in nutrition and socioeconomic factors. In the United States, although oral cavity cancers as a whole are decreasing in incidence, oral tongue cancers are increasing in a specific subset of the population, that is, young white women ages 18 to 44. Many of these individuals are also never-smokers and never-drinkers. The cause of this trend remains unknown, and no viral etiology has been described.
Increasing Incidence of Human Papillomavirus-Positive Head and Neck Cancers
In contrast with decreases in incidence of HPV-negative HNC, which is primarily tobacco-related, the incidence of HPV-HNC has risen dramatically in the United States since the 1970s, primarily oropharyngeal cancer among early middle-aged white men. Analysis of the Surveillance, Epidemiology and End Results database and tissue repository demonstrated a 225% increase in HPV-positive oropharyngeal cancer incidence from 0.8 per 100,000 in 1988 to 1990, to 2.6 per 100,000 in 2003 to 2004, compared with a 50% decrease in HPV-negative oropharyngeal cancer incidence from 2.0 to 1.0 per 100,000 in the same time period. Concurrently, the proportion of oropharyngeal cancers attributable to HPV rose from 16% in the 1980s to near 80% in the past decade. HPV-HNC incidence is now expected to exceed cervical cancer incidence by 2020.
Several other developed countries have observed a similar rise in HPV-related HNCs in men. The incidence of oropharyngeal cancer significantly increased among men from 1983 to 2002 in Australia, Canada, and Slovakia despite stable or decreasing rates of oral cavity cancers. As in the United States, this increase was primarily among men younger than 60 years. In Sweden, there was a 2.8-fold increase in the incidence of tonsil cancer from 1970 to 2002, with a parallel rise in proportion of HPV-positive cases from 23.3% to 68.0%. In Europe overall, the estimated prevalence of HPV in HNC has increased from 35% before 2000 to 73% after 2005 ( P = .004). Similar trends have not yet been observed in low-income countries.
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
