Oral Cancer More Common Among HIV-Infected North Americans
As an attendee of CROI 2012 this year, I really had no intention of stumbling into the "Cancer and HIV" session, but I was glad I did.
One presentation in particular, by Alison Abraham, Ph.D., of the Johns Hopkins University School of Public Health, reinforced what we providers have known for some time: That although oral cavity and pharyngeal cancers are uncommon, approximately 75% of them are attributable to tobacco in its many forms, as well as alcohol.
These data are from the North American AIDS Cohort Collaboration on Research and Design (NA-ACCORD) study, which is a collaboration of more than 20 HIV research studies with more than 110,000 subjects. However, only participants from 14 North American cohorts were utilized for this analysis.
Using the data, the prevalence and incidence of oral cavity and pharyngeal cancers were calculated. Patients were stratified by smoking history, age, calendar time, cancer subsite (oropharyngeal vs. other sites in or around the mouth) and whether their CD4+ cell count was above or below 350 cells/ml. There were no shocking differences in demographics between the cohorts.
The results of the study indicated that out of 51,151 HIV-infected participants, there were 7 cancers at baseline, for a prevalence rate of 14 per 100,000 patient-years. Among participants who were cancer-free at baseline, there were 43 incidents, for a crude incidence rate of 16.4 per 100,000 patient-years and an age-standardization incidence rate (SIR) of 21.7 per 100,000 patient-years, which looks similar to other studies on these cancers.
Since this NA-ACCORD analysis was conducted by chart review with clinical confirmation or cancer registry-linkage, a direct-age standardization of rates was utilized. The Surveillance Epidemiology and End Results (SEER) cohort was used to match the NA-ACCORD cohort with general population data of those persons older than 20 years.
Among the general population, we have seen indications of a decline in these cancers. Not so in the HIV-infected population. When the NA-ACCORD cohort was compared to the SEER cohort, there was a statistically higher trend over time between 2006 and 2010 in the NA-ACCORD than in SEER (SIR 30.1 vs. 14.9 per 100,000 patient-years).
Overall, the 40- to 49-year-old age group in NA-ACCORD had an SIR of 1.6 compared to the same group in SEER, indicating a moderately increased risk for oral-pharyngeal cancers among HIV-infected patients. Patients 50 and older had variable rates over all time periods.
Also notable was stratification by age and CD4+ cell count between the cohorts. Patients 50 and older had a higher incidence in NA-ACCORD vs. SEER (45.7 compared to 24.3), especially if they had a CD4+ cell count lower than 350.
Overall, however, the incidence of cancers in the oral cavity or the pharynx were not associated with a lower baseline CD4+ cell count, although the data indicate that the association may be influenced by the cancer subsite and the patient's age.
Abraham noted that a smoking history was also present in all of those who were diagnosed with oropharyngeal cancer, but warned that smoking history data were limited overall, and that a lack of data on alcohol use or sexual behavior also prevented a deeper understanding of potential contributing factors to cancer incidence.
Subsite analysis indicated that oropharyngeal cancers in particular are on the rise, whereas other oral cancers (such as those of the anterior epiglottis and vallecula) are trending down. In the 2006-2010 period, the overall SIR for oropharyngeal cancers was 2.8 for NA-ACCORD as compared to 1.1 for SEER.
What was truly of interest to me, as a clinician who performs oral cavity assessments, was the number of study patients with oral human papillomavirus (HPV) lesions. As we know, patients are having a study diet of HPV, with subsequent HPV persistence possibly due to the HIV-related immunosuppression. This HPV persistence can lead to oral, pharyngeal and perhaps esophageal cancer. Oral HPV, tobacco use disorders and alcohol use may just be more common in individuals with HIV.
In my humble opinion, providers should remain diligent in patient education on risk factors for developing oral pharyngeal cancers, should provide HPV vaccinations as indicated and should offer their patients referral for dental care.