CHICAGO A twofold to threefold increase in the incidence of tonsillar cancer over several decades was paralleled by a similar rise in the incidence of tumors positive for the human papillomavirus, the results of a Swedish cohort study indicate.
At the same time, however, survival rates among patients with tonsillar cancer also increased, possibly because of the higher proportion of HPV-positive cancers, which tend to have a much better prognosis than other oral cancers, Dr. Hanna Dahlstrand said at the annual meeting of the American Society of Clinical Oncology.
The reasons for the increase in HPV-positive tumors are not known, although they could be related to a possible increase in sexual behaviors, particularly in urban centers, Dr. Dahlstrand said.
Although Sweden is a relatively small nation, the results may be applicable to other countries, she noted. In the United States, for example, the incidence of HPV-related oropharyngeal squamous cell carcinomas has increased since 1973, while the incidence of squamous cell carcinomas at other oral sites has either remained constant or declined. In Finland, the incidence of tonsillar cancers doubled from 1956 through 2000, said Dr. Dahlstrand of the department of oncology-pathology at the Karolinska Institute in Stockholm.
HPV DNA has been shown to be present in 40%-75% of oropharyngeal cancers, compared with about 25% of all head and neck cancers.
"It is only the high-risk types of HPV that are found, with at least 90% dominance of HPV-16, and the oncogenes on HPV, E6 and E7 are transcribed," she said, "And importantly, there is a temporal connection: Exposure to HPV-16 precedes by at least 9 years the diagnosis, and has been shown to be a strong risk factor for tumor development."
HPV-positive oropharyngeal cancers tend to occur more often among nonsmokers and younger patients. Risk factors include multiple sexual partners, younger age at first intercourse, and oral sex. Several studies have shown that the presence of HPV positivity is associated with about a 50% reduction in 5-year mortality, Dr. Dahlstrand noted.
She and her colleagues at the Karolinska Institute conducted a nationwide cohort study using the exhaustive clinical and demographic databases available to Swedish investigators. Their goals were to see whether there has been an increase in the incidence of tonsillar cancer in Sweden; to determine whether such an increase, if present, could be linked to the proportion of HPV-positive tumors; and to see whether the incidence of HPV-positive tonsillar cancers would have an effect on survival.
They identified a total of 2,165 incident cases of tonsillar squamous cell carcinoma from 1960 through 2003, using the Swedish National Cancer Registry.
To determine survival, the investigators used records from the Swedish Causes of Death Register, and checked them against the Swedish Emigration Registry from 1960 to 2001 to ensure that cohort members were not lost to follow-up. They identified a total of 1,800 survivors as of 2003.
The investigators also assessed the incidence of tonsillar cancer from 1970 to 2001 and survival in a Stockholm cohort, and used this cohort to control for treatment, tumor-nodes-metastasis stage, and cause of death. They identified 515 cases in this cohort, and 337 survivors as of 2001.
They were able to obtain 203 biopsy samples and screened them for HPV using polymerase chain reaction testing; they then typed and sequenced the HPV to determine expression of the E6 and E7 oncogenes and expression of HPV-16.
In the nationwide cohort, there was a twofold increase in the incidence of tonsillar cancers, from 1.2/100,000 population to 2.4/100,000, from 1960 to 2003. There was no parallel increase in cancers of the oral cavity (for example, mobile tongue or floor of the mouth), however.
In the Stockholm cohort, tonsillar cancers increased from 1.3/100,000 to 3.6/100,000, or 2.8-fold, from 1970 to 2002. There was also a 2.9-fold increase in the proportion of HPV-positive tonsillar cancers during that same time period. This increase became significant for the 1990-1999 period, compared with 1970-1979, and remained significant for 2000-2002. During the 1970s, 23% of cases were HPV positive, which increased to 28% in the 1980s, 57% in the 1990s, and to 68% into the 21st century.
The mean 5-year relative survival rate among men with tonsillar cancer also increased in Sweden since the 1960s, from 32% to 53% in 1990-2001. In the Stockholm cohort, in a Cox multivariate analysis adjusted for age, gender, stage, and treatment, the investigators found a similar significant increase in relative 5-year survival, with a relative hazard ratio of 0.54 for 1990-2001, compared with 1970-1979. The presence of a lower proportion of stage I and II tumors in the 1990s, compared with the 1970s, suggests that the improvement in survival over the years cannot be explained by earlier diagnosis, Dr. Dahlstrand said.