Abstract
Human papillomavirus (HPV) causes oropharyngeal squamous cell carcinoma (OPSCC), although strongly divergent results have been reported regarding the prevalence of HPV16 in different countries, whether this represents important differences in etiology remains unclear. Applying rigorous protocols for sample processing, we centrally evaluated 1,420 head and neck tumors (533 oropharynx, 395 oral cavity and 482 larynx) from studies conducted in the US, Europe and Brazil for mucosal HPV DNA and p16INK4a expression to evaluate regional heterogeneity in the proportion of HPV16-associated OPSCC and other head and neck cancer, and to assess covariates associated with the risk of HPV16-positive OPSCC. While majority of OPSCC in the US (60%) were HPV16-positive, this proportion was 31% in Europe and only 4% in Brazil (p < 0.01). Similar differences were observed for other head and neck tumors, ranging from 7% in the US and 5% in Europe, to 0% in South America. The odds of HPV16-positive OPSCC declined with increasing pack years of smoking (OR: 0.75; 95% CI: 0.64–0.87) and drink years of alcohol use (OR: 0.64; 95% CI: 0.54–0.76). These results suggest that while the contribution of HPV16 is substantial for the oropharynx, it remains limited for oral cavity and laryngeal cancers.
Keywords: HPV16, oropharynx, head and neck cancer, smoking, alcohol, prevalence
Head and neck cancers (HNC) contribute nearly 600,000 new cases diagnosed, and over 300,000 deaths1 each year. Alongside traditional risk factors such as smoking and excessive alcohol consumption,2,3 human papillomavirus type 16 (HPV16) infection has been recognized to cause a subset of these cancers.4,5
Strongly divergent results have been reported regarding the extent of HPV16 infection in HNC in different countries.6–8 Studies in the US suggest that the majority of oropharyngeal squamous cell carcinoma (OPSCC) are now caused by HPV16,9–11 although proportions of <10% have been reported in the few studies completed in South America,12–14 with European estimates being in between.15–17 Further, the role of HPV16 in HNC outside the oropharynx remains unclear.8 A recent review has estimated that the probability of an HPV-attributable cancer of the oral cavity, larynx and hypopharynx could be up to five times lower than that of oropharyngeal cancer.18 Whether these divergent geographic results represent important differences in the etiology of HNC or whether they are explained by differences in laboratory practices is unknown. The recent publication on global HPV prevalence in HNC is of note where, based on a comparison of over 3,000 tumors tested for HPV DNA and subsequently, following triage for HPV16 E6*mRNA and p16INK4a, the authors report that nearly 22% of OPSCC could be attributed to HPV infection, while fewer than 5% of oral cavity and laryngeal cancers were HPV-positive.19 The study also reported strongly divergent results for the South and Central America (OPSCC HPV prevalence of 37%), and did not provide estimates for North America. Although the largest study to date, this report was unable to explore lifestyle factors associated with HPV-positivity.
We centrally tested a large series of HNCs from three different continents using a combination of HPV16 DNA detection and p16INK4a expression, our aims were to (i) evaluate regional heterogeneity in the proportion of HPV16-associated OPSCC and other HNC following rigorous protocols of sample processing; and (ii) evaluate covariates associated with the risk of HPV16-positive OPSCC.
Methods
The study was designed to include at least 400 HNCs, including 110 OPSCCs, from three HNC case-control studies conducted in the US, Europe and Brazil. All tumors were histologically confirmed diagnoses, and squamous cell in origin. HNC included cancers arising at the oral cavity (International Classification of Diseases for Oncology [ICD-O] C00.3-C00.9, C02.0–C06.9, C14.0-C14.9, excluding C02.4, C05.1, C05.2), oropharynx (ICD-O: C01, C02.4, C09, C10), hypopharynx and larynx (ICD-O: C13, C32), and non-specified and overlapping sites (ICD-O: C05.1-C05.2, C14.0, C014.2, C014.8).
The Carolina Head and Neck Cancer (CHANCE, NC) study recruited population-based incident HNCs and matched controls between 2002 and 2006.20 This analysis included 477 tumors comprising 123 oral cavity, 243 oropharynx, 107 laryngeal and 4 tumors of overlapping sites. The European Alcohol Related Cancers and Genetic susceptibility in Europe (ARCAGE) study was conducted across 10 countries in Europe.21 Tumor samples from 539 HNCs from Germany (n = 181), Italy (n = 289) and the UK (n = 69) were included that consisted of 165 oral cavity, 119 oropharynx, 251 larynx and 4 cancers of overlapping sites. The Brazilian Head and Neck Genome Project (GENCAPO, S~ao Paulo, Brazil) study recruited incident HNC patients and matched controls during 2002 to 2015.22,23 Four hundred and four HNCs included 107 oral cavity, 171 oropharyngeal, 124 larynx and two cancers of overlapping sites. Risk factor and lifestyle data were combined; briefly, tobacco use was categorized as ever or never smokers; ever smokers were defined as individuals who smoked any tobacco product (US), at least once a week for a year (Europe), or daily for a year (Brazil). Pack years were calculated for all types of tobacco smoking based on cigarette equivalents, where one cigar equaled five cigarettes, one pipe equaled four cigarettes and one hand-rolled cigarette equaled five store-bought cigarettes (Brazil only). Ever drinkers were those who reported ever consumption of any alcoholic beverage (US and Europe) or at least once a month (Brazil). A drink-year was defined as two drinks per week for 1 year, where a drink equaled 330 mL of beer, 125 mL of wine and 50 mL of liquor.
A tumor sectioning protocol was established as described previously24 that was applied to tumors from Brazil and Europe, while in the US, pre-sectioned slides generated using a similar scheme were available and included. All tumors were subjected to pathology assessment completed centrally at IARC. Briefly, pathology evaluation determined 3.7% of all cases as ineligible due to insufficient tumor tissue or necrotic tissue, of which 0.8% were from the US, 5.4% from Europe and 4.4% from Brazil. The p16INK4a expression was evaluated using the CINtec Histology P16INK4a Kit (9511, mtmlabs) following manufacturer’s instructions. Staining and scoring was performed blinded with respect to tumor subsite and HPV DNA status. Expression was scored based on the percentage and intensity of nuclear and/or cytoplasmic staining, a composite score of 4 or greater was considered positive for p16INK4a overexpression.25 However, the proportion of p16 invalid tumors was under 1% and did not vary by study. DNA was extracted from tumor tissues26 and HPV genotyping was performed using the Type-Specific E7 PCR bead-based multiplex assay (TS-E7-MPG, IARC, Lyon, France).27–29 Briefly, reporter fluorescence was quantified using Luminex reader 200 (Luminex Corporation, Austin, TX) and cutoffs were computed by adding 5 to 1.1 × the median background value expressed as median fluorescence intensity (MFI). HPV16, HPV18, HPV31, HPV33, HPV35, HPV39, HPV58, HPV59 were considered carcinogenic.5,30 Since HPV16 accounts for over 95% of all HPV-positive HNC,8,12,31,32 the main results described in this manuscript are focused on this type. The largest proportion of betaglobin negative samples were noted in the Brazilian study (14%), compared to the US (2.4%) or European studies (4.6%). We further included several quality indicators at each stage of HPV testing including: (i) sectioning empty paraffin block every 10 tumors, (ii) one DNA extraction control per batch, (iii) one PCR control per amplification plate, (iv) one negative water hybridisation control per batch and (v) randomly selected tumors for retesting from each of the three studies (~5%). HPV testing was completed centrally at the International Agency for Research on Cancer (IARC). Twenty nine of 188 empty paraffin sections yielded a positive result for the betaglobin control gene (all were HPV DNA negative) reflecting a contamination rate of 15%. None of the other technical controls were positive for either the beta globin or HPV DNA. Among the 208 HNCs randomly retested, 8 tumors (3 each from Brazil and Europe and 2 from the US) were discordant for either HPV DNA (n = 6, 4 for HPV16 DNA) or betaglobin status (n = 2). The discordance rate in the duplicate testing was under 4%, and did not vary by region.
Characteristics of cases were compared using Pearson’s chi-squared statistic for categorical and Kruskal-Wallis test for continuous variables. The primary outcome was HPV16-positive tumors defined as being HPV16 DNA and P16INK4a-positive, with cases that were HPV16 DNA or P16INK4A-negative considered HPV16-negative. Unconditional logistic regression was used to estimate odds ratios (OR) and corresponding 95% confidence intervals (95% CI) of HPV16-positive OPSCC using covariates of age, sex, education level, stage, region, smoking (10 pack-year intervals) and alcohol consumption (10 drink-year intervals). The incidence rates of HPV16-positive and -negative OPSCC and Non-OPSCC were estimated by multiplying the incidence rate per 100,000 with the corresponding HPV16 prevalence observed in the present study. All statistical analyses were performed using STATA statistical software, version 11 (StataCorp, College Station, TX), and all reported p values are two sided. Statistical significance was set at p less than 0.05.
Results
This analysis included 1,420 HNCs, including 477 from the US, 539 from Europe and 404 from Brazil (Table 1). European subjects were only marginally older (median age of 60) than the US (median age = 56) and Brazilian subjects (median ages = 57; p < 0.01), current smoking and drinking were high in all three regions. More than two thirds of the HNC cases were diagnosed at late stages of the disease in Brazil (69% stage IV) compared to the less than half of cases in the US and Europe (~47% stage IV).
Table 1.
Description of the study group
| Description N cases | US 477 N (%) | Europe 539 N (%) | Brazil 404 N (%) | Total 1,420 N (%) | |
|---|---|---|---|---|---|
|
| |||||
| Years of diagnosis | 2002–2006 | 2002–2005 | 2002–2011 | ||
| Age | IQR | 64–50 | 68–54 | 64–50 | 66–51 |
| (in years) | Median | 56 | 60 | 57 | 58 |
| Sex | Men | 365 (76.5) | 426 (79.0) | 356 (88.1) | 1,147 (80.8) |
| Women | 112 (23.5) | 113 (21.0) | 48 (11.9) | 273 (19.2) | |
| Education | ≤Primary school | 144 (30.2) | 222 (46.5) | 314 (77.9) | 680 (50.1) |
| level1 | Secondary school | 142 (29.8) | 219 (45.8) | 61 (15.1) | 422 (31.1) |
| Any college | 191 (40.0) | 37 (7.7) | 88 (7.0) | 256 (18.8) | |
| Race2 | Caucasian | 346 (72.5) | 539 (100.0) | 290 (72.5) | 1,175 (83.0) |
| Black | 120 (25.2) | 0 | 39 (9.7) | 159 (11.2) | |
| Asian | 5 (1.1) | 0 | 3 (0.8) | 8 (0.6) | |
| Mixed | 2 (0.4) | 0 | 64 (16.0) | 66 (4.6) | |
| Other | 4 (0.8) | 0 | 4 (1.0) | 8 (0.6) | |
| Smoking3 | Never | 58 (12.2) | 55 (10.2) | 11 (2.7) | 124 (8.8) |
| Former | 129 (27.2) | 173 (32.2) | 89 (22.2) | 391 (27.7) | |
| Current | 288 (60.6) | 310 (57.6) | 301 (75.1) | 899 (63.6) | |
| Median pack years | 43.0 | 39.0 | 43.7 | ||
| Alcohol4 | Never | 46 (9.7) | 43 (8.1) | 42 (10.5) | 131 (9.3) |
| Former | 149 (31.3) | 63 (11.9) | 163 (40.7) | 375 (26.7) | |
| Current | 281 (59.0) | 423 (80.0) | 196 (48.9) | 900 (64.0) | |
| Median drink years | 49.6 | 23.3 | 100.3 | ||
| Stage5 | I | 70 (15.3) | 69 (16.9) | 18 (4.5) | 157 (12.4) |
| II | 80 (17.5) | 71 (17.4) | 49 (12.3) | 200 (15.8) | |
| III | 89 (19.5) | 77 (18.8) | 57 (14.4) | 223 (17.7) | |
| IV | 218 (47.7) | 192 (46.9) | 273 (68.8) | 683 (54.1) | |
| Cancer site | Oral cavity | 123 (25.8) | 165 (30.6) | 107 (26.5) | 395 (27.8) |
| Oropharynx | 243 (50.9) | 119 (22.1) | 171 (42.3) | 533 (37.5) | |
| Larynx6 | 107 (22.4) | 251 (46.6) | 124 (30.7) | 482 (33.9) | |
| Overlapping7 | 4 (0.8) | 4 (0.7) | 2 (0.5) | 10 (0.7) | |
62 cases missing education information; 61 from Europe, 1 case from Brazil.
Missing information for 4 cases (Brazil).
Data missing for 6 cases (2 each from the US,1 from Europe and 3 from Brazil). Median estimated among ever smokers.
Data missing for 14 cases (1 US, 10 Europe, 3 from Brazil). Median estimated among ever alcohol users.
Missing for 130 cases from Europe, 20 from the US and 7 from Brazil.
Includes hypopharynx and larynx cases.
Includes cancers of overlapping sites and non-specified sites.
Majority of the tumors were positive for a single HPV type (~95%), with HPV16 DNA contributing 92% of all HPV DNA positivity (Supporting Information Table 1), and HPV35 (3.5%), HPV18 (2.7%) and HPV33 (1.5%) were also observed. The proportion of HPV16-positive OPSCC differed dramatically by continent, US had the highest proportion (59.3%), Brazil the lowest (4.1%), and Europe reflected intermediate HPV16 prevalence (31.1%; Table 2). The proportion of HPV16-positive tumors did not vary among the three European countries. Regarding oral cavity cancer, 10.6% of tumors in the US and 6.1% in Europe were HPV16-positive (p = 0.20), whereas 2.8% of larynx tumors in the US and 5.2% in Europe were HPV16-positive (p = 0.30). None of the 231 oral or larynx cancer cases in Brazil were HPV16-positive. In order to clarify whether subsite misclassification could contribute to some HPV16-positive oral cavity cancers, we reassessed the sub-anatomic classification of these tumors based on ICD-O codes. Twelve of the 23 HPV16-positive oral cavity tumors were lesions overlapping the tongue, HPV16 prevalence when excluding these was 2.8% (95% CI: 1.19–5.55). When we examined the marginal incidence of HPV16-positive OPSCC, similar results were observed; the highest incidence of HPV-positive OPC were reflected in the US (2 per 100,000), followed by Europe (~1 per 100,000) and the least in Brazil (~0.2 per 100,000). Similarly, HPV16-positive non-OPSCC incident fraction was consistently under 1% in the US and Europe, while it was 0% in Brazil.
Table 2.
Prevalence of HPV16-related cancer1: by region
| Region | ||||||||
|---|---|---|---|---|---|---|---|---|
| Cancer Site | Overall2 | US |
Europe |
Brazil |
||||
| N | Prevalence% (95% CI) | N | Prevalence% (95% CI) | N | Prevalence% (95% CI) | N | Prevalence% (95% CI) | |
|
| ||||||||
| Oropharyngeal cancer3 | 533 | 35.3 (31.2–39.3) | 243 | 59.3 (53.1–65.5) | 119 | 31.1 (22.7–39.5) | 171 | 4.1 (1.1–7.1) |
| Non-oropharyngeal cancers | 887 | 4.4 (3.0–5.7) | 234 | 6.8 (3.6–10.1) | 420 | 5.5 (3.3–7.7) | 233 | 0 |
| Oral cavity4 | 395 | 5.8 (3.5–8.1) | 123 | 10.6 (5.1–16.0) | 165 | 6.1 (2.4–9.7) | 107 | 0 |
| Larynx/hypopharynx4 | 482 | 3.3 (1.7–4.9) | 107 | 2.8 (0.0–6.0) | 251 | 5.2 (2.4–7.9) | 124 | 0 |
defined as joint positivity to HPV16 DNA and p16INK4a overexpression.
None of the 10 tumors of overlapping sites were HPV16-positive.
Chi-squared p values for the difference in prevalence between the three regions was <0.01.
-squared p values for the difference in prevalence between the US and Europe was 0.20, US and Brazil: <0.01 and Europe and Brazil was 0.01.
Chi-squared p values for the difference in prevalence between the US and Europe or Brazil was >0.10, Europe and Brazil was 0.01.
Regarding the discordance between p16INK4a and HPV16 DNA, there were 10.1% OPSCC (n = 54) and 14.5% non-OPSCC tumors (n = 127) that were p16INK4a-positive but HPV16 DNA negative with no regional difference (p > 0.20) (Supporting Information Table 2). Similarly, 9.8% of OPSCC and 14.5% of non-OPSCC were p16INK4a-negative but HPV16 DNA-positive. The highest proportion was observed in Europe compared to US and Brazil (p < 0.01). Concurrent overexpression of P16INK4a with oncogenic HPV types other HPV16 was rare (14 of 4,120, <1%), nearly all were from the US and were more often observed in OPSCC (11/529) compared to the oral cavity or larynx (3/887) (Supporting Information Table 3).
In multivariable analyses, the odds of being a HPV16-positive OPSCC was significantly associated with college education level (OR (95%CI): 3.60 (1.77–7.30)), pack years of ever smoking (OR: 0.75; 95% CI: 0.64–0.87), drink years of ever alcohol use (OR: 0.64; 95% CI: 0.54–0.76), and geographic region (Europe vs. US: OR = 0.42; 95% CI: 0.22–0.82; Brazil vs. US: OR: 0.04; 95% CI: 0.01–0.09; Table 3). Similar results were observed in univariable analysis (Supporting Information Table 4). We observed that ever smokers were less likely to be HPV16-positive (OR: 0.09; 95% CI: 0.04-0.18, Supporting Information Table 4), and the prevalence of HPV16 declined with increasing pack years (Supporting Information Table 5), similar results were noted for alcohol use. The observed inverse associations with smoking and alcohol were robust across geographic regions (p-heterogeneity: 0.10 and 0.66, respectively), while education was only significant in the US.
Table 3.
Multivariable factors associated with HPV16-positive oropharyngeal cancer
| Overall (N = 533) |
US (N = 243) |
Europe (N = 119) |
Brazil (N = 171) |
||||
|---|---|---|---|---|---|---|---|
| Risk factor | N | % HPV16 positive2 | OR (95% CI)3 | OR (95% CI)3 | OR (95% CI)3 | OR (95% CI)3 | p-heterogeneity1 |
|
| |||||||
|
Age (10 year increase) |
533 | – | 0.95 (0.71–1.27) | 0.75 (0.49–1.16) | 0.98 (0.53–1.81) | 1.21 (0.39–3.78) | 0.64 |
| Sex | |||||||
| Men | 447 | 33.8 | 1.0 | 1.0 | 1.0 | 1.0 | |
| Women | 86 | 43.0 | 0.56 (0.27–1.15) | 0.65 (0.22–1.96) | 0.34 (0.07–1.68) | 0.42 (0.02–8.19) | 0.79 |
| Education level | |||||||
| ≤Primary school | 252 | 15.9 | 1.0 | 1.0 | 1.0 | 1.0 | |
| Secondary school | 144 | 40.3 | 2.31 (1.23–4.35) | 2.68 (0.88–8.13) | 1.76 (0.54–5.72) | 17.59 (0.56–555.11) | 0.45 |
| Any college | 131 | 67.9 | 3.60 (1.77–7.30) | 2.23 (0.82–6.06) | 0.29 (0.01–7.69) | 25.24 (0.45–1,410.25) | 0.24 |
| Race | |||||||
| Caucasian | – | – | – | 1.0 | – | – | |
| Black | – | – | – | 0.04 (0.01–0.13) | – | – | |
| Mixed | – | – | – | 0.15 (0.01–3.16) | – | – | |
| Lifetime sex partners | |||||||
| 0–3 | – | – | – | 1.0 | – | – | |
| 4–5 | – | – | – | 4.21 (1.36–13.0) | – | – | |
| Stage4 | |||||||
| I | 25 | 32.0 | 1.0 | 1.0 | 1.0 | 1.0 | |
| II | 44 | 27.3 | 0.79 (0.20–3.11) | 0.91 (0.16–5.25) | – | – | |
| III | 35 | 39.8 | 1.62 (0.49–5.32) | 1.93 (0.38–9.77) | 1.21 (0.11–13.78) | – | 0.75 |
| IV | 120 | 35.7 | 2.49 (0.83–7.45) | 1.73 (0.41–7.37) | 3.20 (0.35–28.89) | – | 0.65 |
| Smoking status | |||||||
| Never | 57 | 82.5 | 1.0 | 1.0 | 1.0 | 1.0 | |
| Pack years5 | 472 | 29.8 | 0.75 (0.64–0.87) | 0.84 (0.66–1.06) | 0.59 (0.41–0.84) | 0.36 (0.13–0.99) | 0.10 |
| Alcohol status | |||||||
| Never | 39 | 53.8 | 1.0 | 1.0 | 1.0 | 1.0 | |
| Drink years5 | 478 | 33.9 | 0.64 (0.54–0.76) | 0.64 (0.48–0.84) | 0.62 (0.44–0.87) | 0.40 (0.15–1.05) | 0.66 |
| Region | |||||||
| US | 243 | 59.3 | 1.0 | – | – | – | |
| Europe | 119 | 31.1 | 0.42 (0.22–0.82) | – | – | – | |
| Brazil | 171 | 4.1 | 0.04 (0.01–0.09) | – | – | – | |
chi-squared test for heterogeneity across the studies.
HPV16 positivity was define as tumor positive for HPV16 DNA and overexpressing P16INK4a protein.
Estimates were adjusted for age, sex, education level, stage, region, smoking (10 pack year intervals) and alcohol consumption (10 drink year intervals) as appropriate.
Seven HPV16 positives in Brazil were all of advanced stage.
Denotes increments of 10 pack years or drink years until 40 or more as appropriate.
Discussion
In this study of 1,420 HNCs from three world regions, we demonstrate dramatic differences in the prevalence of HPV16-positive HNC. While nearly 60% of OPSCCs in the US are HPV16-positive, this proportion is only about 4% in Brazil, and OPSCCs in Europe have intermediate HPV16 prevalence (31%). <4% of laryngeal and oral cavity cancers are HPV16-positive. That similar results were noted for the marginal incident fractions of HPV16-positive OPSCC and non-OPSCC lends support to these conclusions.
We show that among OPSCC cases, ever smokers and ever drinkers are less likely to be HPV16-positive, and the prevalence of HPV16-positive OPSCC declines with increasing pack years of smoking and drink years of alcohol use. It is noteworthy that these conclusions are based on case comparisons, therefore, a higher HPV attributable fraction is expected among never smokers/drinkers as these individuals developed the disease in the absence of the most common risk factors- smoking and alcohol. It also remains to be noted that these results do not represent incidence rate interpretations.
Our results are consistent with the few studies that have examined both HPV DNA and p16INK4a expression in OPSCC, both for estimates in the US.32–34 as well as Europe.15–17,35 Fewer studies have been reported from South America, our observations are similar to the earlier report describing HPV16 prevalence in OPSCC as 4.4%.14 In this context, the recent pooled analysis on HPV prevalence in HNC is of note where, based on 158 tumors from South and Central America tested for HPV DNA and subsequently, following triage for HPV16 E6*mRNA and p16INK4a, the authors report a prevalence of 37% for HPV-positive OPSCC.19 Our estimates are substantially lower than this report even though the number of OPSCC included is comparable. Similarly, our estimates for HPV16-positive OPSCC in Europe are higher than the reported 16%. It is possible that these differences could reflect distinct geographic locations of sampling; while majority of the tumors in the published pooled analysis were drawn from southern Europe, nearly 70% of our European OPSCC subjects were from western and northern Europe. Similarly, South and Central America were represented by nine countries of the region (Argentina, Chile, Colombia, Ecuador, Guatemala, Honduras, Mexico, Paraguay and Venezuela), it is noteworthy that Brazil was not among these. It is also possible that the differences could be due to differing periods of sampling, the pooled analysis retrieved tumors diagnosed over a longer period (1990–2012), with almost 60% of the cases diagnosed after 2005 while most OPSCC tumors included in this analysis were recruited before 2005. Our study consistently showed that a low proportion (~4%) of non-OPSCC were HPV-positive across all three continents. A marginally higher proportion of HPV16-positive tumors were observed in the oral cavity (5.8%) compared to the larynx (3.3%). Closer inspection of the subanatomic sites revealed that almost half all the HPV16-positive oral cavity tumors may have been potentially misclassified and suggest that the true prevalence of HPV16 in oral cavity cancer is likely to be much lower (~3%), and similar to that of laryngeal tumors.
The comparison of a large number of tumors tested for HPV16 DNA and p16INK4a shows that the methods are discordant in 8–13% of the cases. The lack of specificity of p16INK4a appears to vary by subsite but not by region and the specificity of p16INK4a outside the OPSCC appears to be low.18,36,37 Some authors have suggested the use of p16INK4a testing as replacement for HPV DNA,38,39 however, our results argue against this approach. For oncogenic HPV types other than HPV16, concurrent P16INK4A expression occurred more often in OPSCC compared to other HNC sites. Whether this reflects oncogenicity awaits further functional data.
Geographic differences in the proportion of HPV16-positive OPSCC may in part be explained by differences in tobacco use (i.e. the varying proportion of smoking-related OPC in each region). This hypothesis is supported by higher age-standardized incidence rates of OPSCC in Brazil (ASRSão Paulo: 4.0) compared to other regions (ASRNorth Carolina, US: 3.4 and ASRGermany, Italy, Scotland: 2.4).40 However, data on smoking prevalence in the general population of these regions does not provide clear support. While tobacco smoking was lower among controls in the US study compared to Europe (median smoking pack-years: 6.5 and 7.6 in US and Europe), Brazil had the lowest smoking prevalence (median smoking pack-years: 2.6), suggesting the differences in tobacco use do not fully explain the observed HPV prevalence differences in OPSCC. As performing oral sex is the primary risk factor for HPV-positive OPSCC, differences in oral sexual behavior likely contribute to differences in incidence, but unfortunately sexual data was not available in two of the studies. It is important to note however that these risk factors are subject to strong birth cohort effects. In addition, smoking and alcohol are additionally impacted by strong interaction effects.
Despite the dramatic differences in HPV prevalence, the three studies were similar in terms of the risk factor profiles, which was reflected in the consistency of the factors associated with HPV-positivity. The limitations of our study arise from utilizing previously collected survey data from studies that did not collect identical lifestyle data. For instance, sexual behavior questions were not included in the European and South American study, while the data collected in the US study was limited. In addition, although we present results for three world regions, it must be noted that the US is represented by a single study conducted in North Carolina where smoking prevalence is among the highest in the country. Similarly, Europe is represented by three countries only while South America is represented by a single study that was conducted in Brazil with all cases being recruited from public hospitals in São Paulo.
We conclude that the proportion of OPSCC now caused by HPV16 varies by geographic region with low proportions in Brazil, moderate proportions in Western Europe, and the majority of OPSCC diagnosed in the US being HPV16-positive. Our results suggest that the effect of prophylactic HPV vaccination on OPSCC would be largest in the US.
Supplementary Material
Whaťs new?
Where might HPV vaccine have the most impact on head and neck cancer? To find out, these authors tested over 1,400 HNCs from three different continents—North America, South America and Europe—looking for HPV-16. When they focused specifically on oropharyngeal squamous cell carcinoma, the results were striking: 60% of the OPSCCs in the US were HPV-16 positive, compared with 30% in Europe and just 4% in Brazil. Thus, the variation in the number of cases of OPSCC in different regions may not be entirely due to lifestyle differences and the proportion of smokers in the population.
Acknowledgements
The authors acknowledge the contribution of GENCAPO (Brazilian Head and Neck Genome Project) for clinical samples and for clinical and pathological data collection (complete list of members and affiliations presented at http://www.gencapo.famerp.br).
Grant sponsor: European Commission’s 5th Framework Program (ARCAGE study; P.B., PI); Grant number: QLK1–2001–00182; Grant sponsor: European Commission’s 7th Framework Program (M.T., PI); Grant number: FP7-HEALTH-2011–282562; Grant sponsor: Health General Directorate of the French Social Affairs and Health Ministry; Grant sponsor: Italian Association for Research on Cancer (AIRC; S.B., PI); Grant number: IG 2013 N.14220; Grant sponsor: Fondazione Veronesi (S.B., PI); Grant sponsor: National Cancer Institute (U.S. CHANCE study; A.O., PI); Grant number: NCI R01-CA90731; Grant sponsor: FAPESP (São Paulo Research Foundation; GENCAPO study); Grant number: 04/12054–9 and 10/51168–0
Footnotes
Additional Supporting Information may be found in the online version of this article.
Declaration of Interests: G.D. received previous research support from Merck Inc.
References
- 1.Ferlay J, Shin HR, Bray F, et al. Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008. Int J Cancer 2010;127:2893–917. [DOI] [PubMed] [Google Scholar]
- 2.Anantharaman D, Marron M, Lagiou P, et al. Population attributable risk of tobacco and alcohol for upper aerodigestive tract cancer. Oral Oncol 2011;47:725–31. [DOI] [PubMed] [Google Scholar]
- 3.Hashibe M, Brennan P, Chuang SC, et al. Interaction between tobacco and alcohol use and the risk of head and neck cancer: pooled analysis in the International Head and Neck Cancer Epidemiology Consortium. Cancer Epidemiol. Biomarkers Prev 2009;18:5401–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Human Papillomaviruses. IARC Monogr Eval Carcinog Risks Hum 1995;64:1–378. [PMC free article] [PubMed] [Google Scholar]
- 5.Human Papillomaviruses. IARC Monogr Eval Carcinog Risks Hum 2007;90:1–636. [PMC free article] [PubMed] [Google Scholar]
- 6.Kreimer AR, Clifford GM, Boyle P, et al. Human papillomavirus types in head and neck squamous cell carcinomas worldwide: a systematic review. Cancer Epidemiol Biomarkers Prev 2005;14:467–75. [DOI] [PubMed] [Google Scholar]
- 7.Marur S, D’Souza G, Westra WH, et al. HPV-associated head and neck cancer: a virus-related cancer epidemic. Lancet Oncol 2010;11:781–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Ndiaye C, Mena M, Alemany L, et al. HPV DNA, E6/E7 mRNA, and p16INK4a detection in head and neck cancers: a systematic review and meta-analysis. The Lancet Oncology 2014;15: 1319–31. [DOI] [PubMed] [Google Scholar]
- 9.Chaturvedi AK, Anderson WF, Lortet-Tieulent J, et al. Worldwide trends in incidence rates for oral cavity and oropharyngeal cancers. J Clin Oncol 2013;31:4550–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Chaturvedi AK, Engels EA, Anderson WF, et al. Incidence trends for human papillomavirus-related and -unrelated oral squamous cell carcinomas in the United States. J Clin Oncol 2008;26: 612–9. [DOI] [PubMed] [Google Scholar]
- 11.Nasman A, Attner P, Hammarstedt L, et al. Incidence of human papillomavirus (HPV) positive tonsillar carcinoma in Stockholm, Sweden: an epidemic of viral-induced carcinoma? Int J Cancer 2009;125:362–6. [DOI] [PubMed] [Google Scholar]
- 12.Herrero R, Castellsague X, Pawlita M, et al. Human papillomavirus and oral cancer: the International Agency for Research on Cancer multicenter study. J Natl Cancer Inst 2003;95: 1772–83. [DOI] [PubMed] [Google Scholar]
- 13.Lopez RV, Levi JE, Eluf-Neto J, et al. Human papillomavirus (HPV) 16 and the prognosis of head and neck cancer in a geographical region with a low prevalence of HPV infection. Cancer Causes Control: CCC 2014;25:461–71. [DOI] [PubMed] [Google Scholar]
- 14.Ribeiro KB, Levi JE, Pawlita M, et al. Low human papillomavirus prevalence in head and neck cancer: results from two large case-control studies in high-incidence regions. Int J Epidemiol 2011;40: 489–502. [DOI] [PubMed] [Google Scholar]
- 15.Hoffmann M, Ihloff AS, Gorogh T, et al. p16(INK4a) overexpression predicts translational active human papillomavirus infection in tonsillar cancer. Int J Cancer 2010;127:1595–602. [DOI] [PubMed] [Google Scholar]
- 16.Reimers N, Kasper HU, Weissenborn SJ, et al. Combined analysis of HPV-DNA, p16 and EGFR expression to predict prognosis in oropharyngeal cancer. Int J Cancer 2007;120:1731–8. [DOI] [PubMed] [Google Scholar]
- 17.Smeets SJ, Hesselink AT, Speel EJ, et al. A novel algorithm for reliable detection of human papillomavirus in paraffin embedded head and neck cancer specimen. Int J Cancer 2007;121:2465–72. [DOI] [PubMed] [Google Scholar]
- 18.Combes JD, Franceschi S. Role of human papillomavirus in non-oropharyngeal head and neck cancers. Oral Oncol 2014;50:370–9. [DOI] [PubMed] [Google Scholar]
- 19.Castellsague X, Alemany L, Quer M, et al. HPV Involvement in Head and Neck Cancers: Comprehensive Assessment of Biomarkers in 3680 Patients. J Natl Cancer Inst 2016;108. [DOI] [PubMed] [Google Scholar]
- 20.Divaris K, Olshan AF, Smith J, et al. Oral health and risk for head and neck squamous cell carcinoma: the Carolina Head and Neck Cancer Study. Cancer Causes Control: CCC 2010;21:567–75. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Lagiou P, Georgila C, Minaki P, et al. Alcohol-related cancers and genetic susceptibility in Europe: the ARCAGE project: study samples and data collection. Eur J Cancer Prev 2009;18:76–84. [DOI] [PubMed] [Google Scholar]
- 22.Boing AF, Antunes JL, de Carvalho MB, et al. How much do smoking and alcohol consumption explain socioeconomic inequalities in head and neck cancer risk? J Epidemiol Community Health 2011;65:709–14. [DOI] [PubMed] [Google Scholar]
- 23.Severino P, Alvares AM, Michaluart P Jr., et al. Global gene expression profiling of oral cavity cancers suggests molecular heterogeneity within anatomic subsites. BMC Res Notes 2008;1:113. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Anantharaman D, Gheit T, Waterboer T, et al. No causal association identified for human papillomavirus infections in lung cancer. Cancer Res 2014;74:3525–34. [DOI] [PubMed] [Google Scholar]
- 25.Koo CL, Kok LF, Lee MY, et al. Scoring mechanisms of p16INK4a immunohistochemistry based on either independent nucleic stain or mixed cytoplasmic with nucleic expression can significantly signal to distinguish between endocervical and endometrial adenocarcinomas in a tissue microarray study. J Transl Med 2009;7:25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Gheit T, Vaccarella S, Schmitt M, et al. Prevalence of human papillomavirus types in cervical and oral cancers in central India. Vaccine 2009; 27:636–9. [DOI] [PubMed] [Google Scholar]
- 27.Gheit T, Landi S, Gemignani F, et al. Development of a sensitive and specific assay combining multiplex PCR and DNA microarray primer extension to detect high-risk mucosal human papillomavirus types. J Clin Microbiol 2006;44:2025–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Schmitt M, Bravo IG, Snijders PJ, et al. Bead-based multiplex genotyping of human papillomaviruses. J Clin Microbiol 2006;44:504–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Schmitt M, Dondog B, Waterboer T, et al. Abundance of multiple high-risk human papillomavirus (HPV) infections found in cervical cells analyzed by use of an ultrasensitive HPV genotyping assay. J ClinMicrobiol 2010;48:143–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Biological Agents. Volume 100 B. A review of human carcinogens. IARC Monogr Eval Carcinog Risks Hum 2012;100:1–441. [PMC free article] [PubMed] [Google Scholar]
- 31.Anantharaman D, Gheit T, Waterboer T, et al. Human papillomavirus infections and upper aero-digestive tract cancers: the ARCAGE Study. J Natl Cancer Inst 2013;105:536–45. [DOI] [PubMed] [Google Scholar]
- 32.Gillison ML, Koch WM, Capone RB, et al. Evidence for a causal association between human papillomavirus and a subset of head and neck cancers. J Natl Cancer Inst 2000;92:709–20. [DOI] [PubMed] [Google Scholar]
- 33.Ang KK, Harris J, Wheeler R, et al. Human papillomavirus and survival of patients with oropharyngeal cancer. N Engl J Med 2010;363:24–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Fakhry C, Westra WH, Li S, et al. Improved survival of patients with human papillomavirus-positive head and neck squamous cell carcinoma in a prospective clinical trial. J Natl Cancer Inst 2008;100:261–9. [DOI] [PubMed] [Google Scholar]
- 35.Licitra L, Perrone F, Bossi P, et al. High-risk human papillomavirus affects prognosis in patients with surgically treated oropharyngeal squamous cell carcinoma. J Clin Oncol 2006;24:5630–6. [DOI] [PubMed] [Google Scholar]
- 36.Chaturvedi AK. Global burden of human papillomavirus-positive head and neck cancers. Lancet Oncol 2014;15:1282–3. [DOI] [PubMed] [Google Scholar]
- 37.Lingen MW, Xiao W, Schmitt A, et al. Low etiologic fraction for high-risk human papillomavirus in oral cavity squamous cell carcinomas. Oral Oncol 2013;49:1–8. [DOI] [PubMed] [Google Scholar]
- 38.Kuo KT, Hsiao CH, Lin, et al. The biomarkers of human papillomavirus infection in tonsillar squamous cell carcinoma-molecular basis and predicting favorable outcome. Mod Pathol 2008; 21:376–86. [DOI] [PubMed] [Google Scholar]
- 39.Mellin Dahlstrand H, Lindquist D, et al. P16(INK4a) correlates to human papillomavirus presence, response to radiotherapy and clinical outcome in tonsillar carcinoma. Anticancer Res 2005;25:4375–83. [PubMed] [Google Scholar]
- 40.Curado MP, Edwards B, Shin HR, et al. , eds. Cancer Incidence in Five Continents, Volume IXed, Vol. IX. Lyon: International Agency for Research on Cancer, 2007. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.