Worldwide trends in incidence rates for oral cavity and oropharyngeal cancers

Abstract

Purpose: Human papillomavirus (HPV) has been identified as the cause of the increasing oropharyngeal cancer (OPC) incidence in some countries. To investigate whether this represents a global phenomenon, we evaluated incidence trends for OPCs and oral cavity cancers (OCCs) in 23 countries across four continents.

Methods: We used data from the Cancer Incidence in Five Continents database Volumes VI to IX (years 1983 to 2002). Using age-period-cohort modeling, incidence trends for OPCs were compared with those of OCCs and lung cancers to delineate the potential role of HPV vis-à-vis smoking on incidence trends. Analyses were country specific and sex specific.

Results: OPC incidence significantly increased during 1983 to 2002 predominantly in economically developed countries. Among men, OPC incidence significantly increased in the United States, Australia, Canada, Japan, and Slovakia, despite nonsignificant or significantly decreasing incidence of OCCs. In contrast, among women, in all countries with increasing OPC incidence (Denmark, Estonia, France, the Netherlands, Poland, Slovakia, Switzerland, and United Kingdom), there was a concomitant increase in incidence of OCCs. Although increasing OPC incidence among men was accompanied by decreasing lung cancer incidence, increasing incidence among women was generally accompanied by increasing lung cancer incidence. The magnitude of increase in OPC incidence among men was significantly higher at younger ages (< 60 years) than older ages in the United States, Australia, Canada, Slovakia, Denmark, and United Kingdom.

Conclusion: OPC incidence significantly increased during 1983 to 2002 predominantly in developed countries and at younger ages. These results underscore a potential role for HPV infection on increasing OPC incidence, particularly among men.

Fig 1.

Incidence trends for oropharyngeal (OP) cancers (shown in gold) and oral cavity (OC) cancers (shown in blue). Results are shown for (A) countries with significantly increasing OP cancer incidence among men. For all parts of Figure 1, filled circles represent observed incidence rates, and solid lines represent fitted incidence rates. The key in each graph shows estimated annual percent changes (EAPCs) in incidence during 1983 to 2002, which were calculated using weighted least squares log-linear regression. (*) EAPC statistically significant at P < .05. NE, not estimable. Results are shown for (B, C) countries with nonsignificant trends or significant declines in OP cancer incidence among men. Results are shown for (D) countries with significantly increasing OP cancer incidence among women. Results are shown for (E, F) countries with nonsignificant trends or significant declines in OP cancer incidence among women.

Fig 1.

Incidence trends for oropharyngeal (OP) cancers (shown in gold) and oral cavity (OC) cancers (shown in blue). Results are shown for (A) countries with significantly increasing OP cancer incidence among men. For all parts of Figure 1, filled circles represent observed incidence rates, and solid lines represent fitted incidence rates. The key in each graph shows estimated annual percent changes (EAPCs) in incidence during 1983 to 2002, which were calculated using weighted least squares log-linear regression. (*) EAPC statistically significant at P < .05. NE, not estimable. Results are shown for (B, C) countries with nonsignificant trends or significant declines in OP cancer incidence among men. Results are shown for (D) countries with significantly increasing OP cancer incidence among women. Results are shown for (E, F) countries with nonsignificant trends or significant declines in OP cancer incidence among women.

Fig 1.

Incidence trends for oropharyngeal (OP) cancers (shown in gold) and oral cavity (OC) cancers (shown in blue). Results are shown for (A) countries with significantly increasing OP cancer incidence among men. For all parts of Figure 1, filled circles represent observed incidence rates, and solid lines represent fitted incidence rates. The key in each graph shows estimated annual percent changes (EAPCs) in incidence during 1983 to 2002, which were calculated using weighted least squares log-linear regression. (*) EAPC statistically significant at P < .05. NE, not estimable. Results are shown for (B, C) countries with nonsignificant trends or significant declines in OP cancer incidence among men. Results are shown for (D) countries with significantly increasing OP cancer incidence among women. Results are shown for (E, F) countries with nonsignificant trends or significant declines in OP cancer incidence among women.

Fig 1.

Incidence trends for oropharyngeal (OP) cancers (shown in gold) and oral cavity (OC) cancers (shown in blue). Results are shown for (A) countries with significantly increasing OP cancer incidence among men. For all parts of Figure 1, filled circles represent observed incidence rates, and solid lines represent fitted incidence rates. The key in each graph shows estimated annual percent changes (EAPCs) in incidence during 1983 to 2002, which were calculated using weighted least squares log-linear regression. (*) EAPC statistically significant at P < .05. NE, not estimable. Results are shown for (B, C) countries with nonsignificant trends or significant declines in OP cancer incidence among men. Results are shown for (D) countries with significantly increasing OP cancer incidence among women. Results are shown for (E, F) countries with nonsignificant trends or significant declines in OP cancer incidence among women.

Fig 2.

The net drifts and 95% CIs from age-period-cohort models are shown for oropharyngeal (OP) cancers (gold squares), oral cavity (OC) cancers (blue squares), and lung cancers (gray circles). Results are shown separately for (A) men and (B) women. The net drift represents the net sum of the linear trend in period and cohort effects. P values for the comparison of net drifts for OP versus OC cancers, as well as for comparisons of OP versus lung (L) cancer, are also shown in each panel. Results are presented for countries with significant increases in OP cancer incidence. Results for all other countries are presented in the Data Supplement. (*) Countries with significant differences between OP versus OC cancers. UK, United Kingdom.