The Current Burden of Oropharyngeal Cancer: A Global Assessment Based on GLOBOCAN 2020

This article is featured in Highlights of This Issue, p. 1993

Oropharyngeal cancer (OPC) is a subtype belonging to the group of head and neck cancer (HNC) that affects the base of the tongue, palatine tonsils, lingual tonsils, soft palate, uvula, and posterior pharyngeal wall. There is growing evidence that OPC may be characterized by two different causal mechanisms, one related to the exposure of risk factors typically associated with HNC (e.g., tobacco smoking or chewing, alcohol consumption, pollutants, and oral hygiene), and the other related to the carcinogenic effect induced by human papillomavirus (HPV) infection, particularly HPV-16 (1), which has a better prognosis (2, 3). HPV is the most common sexually transmitted infection that causes cervical cancer and other anogenital cancers, and oral infection with high-risk HPV is strongly associated with OPC (4).

OPC subtypes are characterized by large geographical variability and diverging trends. Indeed, HPV-attributable fraction (AF) of OPC has been shown to vary between 19% and 70%, with significantly higher AF in developed countries (5–7). Despite the large decline in the use of tobacco, increases in the incidence rates of OPC have been reported in many high-income countries such as the United States and the United Kingdom. Indeed, the reason behind these unfavorable trends is likely due to the increased exposure to HPV among younger generations of individuals, starting from those born around the 1940s/1950s, following changes in sexual behaviors (8).

Given the heterogeneity and complexity of OPC, the objective of the current study was to provide an updated picture of the status of the burden of the disease worldwide, based on GLOBOCAN 2020 estimates of incidence and mortality, highlighting gender and geographical differences.

The number of new cases and deaths from cancers of the oropharynx was extracted from the GLOBOCAN 2020 database for 185 countries or territories by sex and 18 age groups (0–4, 5–9, …, 80–84, 85, and over; refs. 9–11). The anatomical sites in this GLOBOCAN analysis included the International Classification of Diseases (ICD)-10 C09–10, but excluded C01, C02, C05, and C14, which correspond to the base of the tongue, lingual tonsil, soft palate, uvula, and Waldeyer's ring. Corresponding population data for 2020 were extracted from the United Nations (UN) website (12). The data sources and hierarchy of methods used to compile cancer estimates have been described in detail elsewhere (9). In brief, the GLOBOCAN estimates were assembled at the national level using the best available sources of cancer incidence and mortality data within a given country. The methods used to derive the 2020 estimates correspond to those used to derive similar data for previous years (13–15); validity is dependent on the degree of representativeness and quality of the source information (9).

We present tables and figures based on the estimated new cases and deaths for all age groups, as well as two summary measures using direct standardization, namely the age-standardized (incidence or mortality) rate (ASR) per 100,000 person-years based on the 1966 Segi–Doll World standard population (16, 17) and the cumulative risk of developing or dying from cancer before the age of 75 expressed as a percentage, assuming the absence of competing causes of death (18). These measures allow comparisons between populations, adjusted for differences in age structures. The results are presented by country and aggregated across the 20 UN-defined world regions (12). For both incidence and mortality, the male-to-female incidence ratio was calculated using the male age-adjusted rate as the numerator and the female age-adjusted rate as the denominator. The incidence-to-mortality ratio (IMR) was also estimated for each geographical area and country, both overall, and for each sex as the ratio of the age-adjusted incidence rate to the age-adjusted mortality rate.

No ethical or legal approval was required for the present study.

Data used in the present study are published and available from the web-based platform of the Global Cancer Observatory (GCO; ref. 19) that includes also facilities that allow the tabulation and graphical visualization of the GLOBOCAN database, including exploration of the current and future (20) burden for 36 cancer types and all cancers combined, including non-melanoma skin cancer (ICD-10 C44 excluding basal-cell carcinomas).

In 2020, a total of 98,412 new OPC cases and 48,143 OPC deaths were estimated worldwide, with an ASR of approximately 1.1 per 100,000 for incidence and 0.51 per 100,000 for mortality. As shown in Fig. 1, OPC rates for both incidence and mortality varied substantially among countries for both sexes. High ASRs for incidence were estimated in most European countries in both women and men (Fig. 1A and B). In details, Denmark, France, Hungary, and Czechia showed the highest ASR of incidence among female whereas Romania, Belarus, Denmark, and Republic of Moldova lead in males. Outside the European area, the other countries that emerged with respect to high age-standardized incidence rates were Australia, Cuba, the United States, Brazil, the Russian Federation, India, Turkmenistan, and Bangladesh. Yemen and Namibia added this only to women. In contrast, low ASRs for incidence were estimated for most African countries, China, and many countries in Western Asia.

In general, there is a good geographical correlation between incidence and mortality. Among females, Hungary, Denmark, Namibia, Montenegro, and Bangladesh showed the highest ASR for mortality, whereas Slovenia, Belarus, Romania, Slovakia, and the Republic of Moldova had the highest age-standardized mortality rates in males (Fig. 2A and B). For both sexes, the lowest values were observed for most of the countries in Africa and Western Asia. However, some exceptions were observed. In particular, Melanesia, South-Central Asia, and the Caribbean, despite having ASR for incidence lower than European regions, had higher mortality rates, whereas Northern Europe and Northern America that exhibited the second and third highest ASRs for incidence were not among the top five regions with the highest ASR for mortality.

With respect to world regions, European regions (except Southern Europe) were ranked among the top five areas with the highest ASR for incidence together with North America as well as Australia and New Zealand (with ASR varying from 2.3 to 2.8 per 100,000); Western and Eastern Asia, North-West Africa, and Central America showed the lowest ASR (Table 1). Central-East Europe, Western Europe, Melanesia, South-Central Asia, and the Caribbean lead for mortality rates with ASR values ranging from 0.85 to 1.10 per 100,000.

Nevertheless, in both sexes, a high degree of variability was estimated for countries within the same region. Among women, Northern and Western Europe showed the highest ASR in incidence, whereas the lowest was observed in Micronesia/Polynesia, Western and Eastern Asia, Central America, and Eastern Africa.

In European regions, age-standardized incidence rates in females also showed a marked variability among countries in the same region, with single countries leading (namely, Denmark and France) in both Northern and Western Europe, whereas the majority of the remaining countries showed ASR below the regional value (Fig. 3A; Supplementary Table S1).

The highest ASR incidence values for males were estimated in Central-East and Western Europe, and the lowest in Western Asia and North-West Africa (Fig. 3B, Supplementary Table S1). Moreover, within each region, variability of ASR for incidence was more pronounced among men, with countries in the same region exhibiting a wide range of values not only among European regions but also in South America, Melanesia, the Caribbean, and Southern Asia (Fig. 3B).

The worldwide ASR for mortality was 0.9 per 100,000 among males and 0.2 per 100,000 in females.

As detailed in Supplementary Table S2 and Fig. 4A, among females, the ASR for mortality was higher in North-West Europe and South-Central Asia. Marked within-region variability was observed.

Among men, Central-East Europe and Melanesia showed the highest ASR for mortality. The Republic of Moldova led mortality rates in Central-Eastern Europe, Bangladesh in Southern Asia, and Namibia in Southern Africa, and even among regions with some of the lowest ASR for mortality, such as Micronesia/Polynesia, Western Africa, and Western Asia, few or single countries exhibited rates similar to or above worldwide values, namely, French Polynesia, Cape Verde, Georgia, Brazil, and Cuba (Fig. 4B; Supplementary Table S2).

The IMR was 2.2 worldwide being 2.4 among females and 2 in males. Regarding age-standardized incidence and mortality rates, the IMR varied substantially among geographical areas. Indeed, Northern Europe, Australia, New Zealand, and Northern America had the highest IMR (from 3.6 to 4.6, respectively), whereas all the African areas (except North Africa), Central America, and South-Central Asia showed the lowest (range, 1.4–1.6; Table 1).

For both genders, Northern and Western Europe, Australia, New Zealand, and Northern America showed some of the highest values for IMR, whereas African regions (except Northern Africa), South-Central Asia, and Central America were ranked at the bottom for IMR values (Fig. 5). For each sex, the IMR also showed a certain degree of variability among countries within each geographical area. Among females, values in Northern Europe varied widely, with Norway and Sweden showing the highest values, and Lithuania and Estonia the lowest. Moreover, France and the Republic of Korea emerged because the IMR values were much higher than the average values for Western Europe and East Asia, respectively (Fig. 5A). Among males, IMR was much dispersed both in Western and Northern Europe, and few countries in these regions as well as in Southern Europe and South America exhibited IMR values much higher than the average value for the area (Luxembourg, Sweden, Finland, Cyprus, and French Guinea; see Fig. 5B).

The present study provides an up-to-date global overview of the current OPC burden based on estimates from the GLOBOCAN 2020 database. In 2020, a total of 98,400 new OPC cases and 48,100 OPC deaths were estimated worldwide. Northern America and North-West Europe are the geographical areas with the highest ASR for incidence; Western Melanesia and Central-East Europe lead to mortality. The results also highlighted a picture that is highly variable in different geographical regions and also a marked variability linked to gender. Among females, Northern America, Northern and Western Europe lead to incidence; a considerable degree of variability characterized the distribution of age-standardized incidence rates among males, with Northern America, Australia, and New Zealand as well as Europe (except Southern Europe) exhibiting the highest values. Melanesia and western and central-eastern Europe lead to mortality among females, although there is a high degree of variability between countries in the same region; in males, northern and western Europe showed the highest ASR values for mortality.

The most important risk factors for OPC are tobacco smoking, alcohol consumption, and oral HR-HPV infection. The marked geographic and gender variability in incidence and mortality from OPC highlighted in this study is thus likely to reflect a combination and/or interaction or synergy between these determinants, and in particular, the diverse prevalence of HPV-negative and HPV-positive cases (1, 21–23). Indeed, previous studies have shown that the worldwide HPV-AF varies in the range of 19% to 70%, with significantly higher AF in North America and Europe and lower elsewhere (5–7, 24).

Findings from the present study show that OPC incidence rates in North America and Europe, particularly in Central, Eastern, and Western Europe, are among the highest worldwide for both men and women. Despite substantial efforts devoted to reducing alcohol and tobacco consumption that has led to a decrease in the incidence rates of HNC in many countries, the volume of (per capita) alcohol intake and smoking prevalence is still comparatively very high (25, 26). The prevalence of tobacco smoking and smokeless tobacco use (25, 27, 28) is also high among men in Central Asia, which may explain the high incidence of OPC in the region.

In countries such as Brazil, which currently have relatively low levels of alcohol and tobacco consumption, the high OPC prevalence could result from a historically high smoking prevalence, particularly among men of older generations (25, 26), but also from a widespread oral HPV infection (4, 29). In the US, Canada, Australia, and New Zealand, high OPC rates combined with high IMR values are likely to reflect the high prevalence of oral HPV infection, which reaches up to 44%–90% and is higher than that in other areas of the world (4, 8, 29, 30). Indeed, in the US and Europe, the rates of high-risk HPV-related OPC are on the rise (31).

IMR was low in most African regions as well as in Central America, whereas the highest values were found in Northern Europe, Australia, New Zealand, and Northern America, with considerable variability among countries in the same region, particularly among males, showing a more pronounced degree of variability among countries within European regions. Additional explanations for differences in mortality rates and low IMR values could be disparity in treatment access in low- and middle-income countries related to cultural and socio-economic barriers, as well as the availability of advanced treatment options (8, 30).

Despite depicting only a snapshot, the strength of our study is the use of worldwide data to provide a global overview of the OPC burden in 2020, thus driving future studies and offering countries and their policy-makers data that could help implement and enforce prevention policies, targeting and prioritizing interventions depending on the specific context, and allocating resources.

In particular, to what concern prevention, despite specific actions were identified to reduce tobacco and alcohol consumption (32, 33), and efforts undertaken accordingly in many countries, significant implementation gap remained, thus tobacco and alcohol consumption are still associated with significant disease burden, disability, and premature death.

This issue seems particularly challenging in low- and middle-income countries, where, due to the growing population and the widespread use of smokeless tobacco, the absolute number of deaths attributable to tobacco (smokeless and smoking) is increasing (25, 28, 34); On the other hand, alcohol use is still of concern, with the World Health Organization European Region having the highest proportion in the world of total ill health and premature death due to alcohol (35).

Empowerment of specific population groups is also required. On the one hand, in low- and middle-income countries, where HPV prevalence is higher, there is a need to increase awareness, particularly among females and disadvantaged people, to allow them access to prevention programs; on the other hand, in high-income countries, due to the increasing incidence of HPV-positive OPC in white and younger males (36–38) because of the different immunological susceptibility of males and sexual debut, lifetime number of sexual partners, even independently from other risk factors (2, 3, 31, 39–43) there is a need to design specific approaches addressing this group, foreseeing implementation of effective education programs to address changes in sexual behavior (41, 44, 45), and also considering extending vaccination programs (46).

Despite the strengths of this study, some limitations must be acknowledged. First, although it is one of the best sources available to describe cancer burden worldwide, it must be acknowledged that the accuracy and reliability of estimates from GLOBOCAN vary across countries according to the availability and validity of data available at the national level. For the same reason, caution is needed to interpret IMR values, particularly in some low- and middle-income countries, and the reliability of estimates may be relatively low because of the limited quality of data. Finally, GLOBOCAN does not allow separation of oral tongue cancers from the base of tongue cancers or separation of the hard palate from the soft palate/uvula as well as no information on disease stage is available; accordingly, specific analyses for the different anatomical sites or by disease stage could not be performed.

In summary, this study highlights the large geographical variability in the incidence and mortality of OPC and IMR. To further reduce disease incidence and lethality and to limit its impact in terms of costs and quality of life, evidence about the current burden of OPC could provide a stimulus to improve equity in access to advanced therapeutic options and/or surgical approaches (47).

No disclosures were reported.

Where authors are identified as personnel of the International Agency for Research on Cancer/World Health Organization, the authors alone are responsible for the views expressed in this article and they do not necessarily represent the decisions, policy or views of the International Agency for Research on Cancer/World Health Organization.

V. Lorenzoni: Conceptualization, formal analysis, visualization, writing–original draft. A.K. Chaturvedi: Writing-review and editing. J. Vignat: Data curation, software, formal analysis, visualization. M. Laversanne: Data curation, software, formal analysis, visualization. F. Bray: Conceptualization, writing–review and editing. S. Vaccarella: Conceptualization, supervision, writing–review and editing.

The publication costs of this article were defrayed in part by the payment of publication fees. Therefore, and solely to indicate this fact, this article is hereby marked “advertisement” in accordance with 18 USC section 1734.