Background:  A 1996 case control study from New York showed an HIV prevalence of 4.5% among head and neck squamous cell carcinoma (HNSCC) patients. Patients with HIV and HNSCC were younger, had more advanced disease, and poorer survival than those without HIV. We assessed the prevalence of HIV in HNSCC patients at an inner city hospital and the effect of HAART on the outcome of these patients.

Methods:  This was a retrospective study of patients with oral, oropharyngeal, or laryngeal squamous cell cancer, identified by pathological diagnosis at Grady Memorial Hospital from 1991 to 2006. Individual demographics, tumor stage, HIV status, and survival data were collected. Prevalence of HIV was determined in HNSCC patients diagnosed before and after 1999. This year was chosen because of increased HIV testing in our health care system since that year.

Results:  There were 204 patients diagnosed with HNSCC since 1991. Alcohol use and smoking were reported in 88% of HIV⁺ and 84% of HIV^– patients. Before 1999, 23 of 103 HNSCC patients were tested for HIV:  only 1 was HIV⁺ (prevalence = 0.97%). After 1999, 43 of 101 HNSCC patients were tested, and 16 were HIV⁺ (prevalence = 15.84%). There was no significant difference between the HIV⁺ and HIV^– groups with regard to age, gender, initial tumor stage, or survival. Of the total 17 HIV⁺ patients, 8 were not on HAART at diagnosis. All untreated patients presented with stage 3 or 4 disease, whereas half of those on HAART presented with stage 3 or 4 (p = 0.0253). More HAART-treated patients survived at 1 year than those not on HAART (29% vs 100%, p = 0.0047). There was no significant difference in 1-year survival between HIV patients on HAART and HIV^– patients (100% vs 81%, p = 0.1788). Interestingly, a greater percentage of HIV^– patients presented with stage 3 or 4 disease than HIV⁺ patients on HAART (77% vs 50%, p = 0.028).

Conclusions:  In our population, the prevalence of HIV among HNSCC patients is increasing. Our prevalence of 15.84% is higher than the baseline HIV prevalence of 3% in a cohort from our urgent care clinic in 1999. We may be underestimating the true prevalence of HIV in HNSCC, since fewer than half of patients were tested. All patients with HNSCC should be offered testing for HIV. Despite the limitation of small number and the retrospective nature of the study, our data showed that HIV⁺ and HIV^– patients now have equivalent tumor staging at presentation and 1-year survival, and that HAART has changed the presentation and outcome of HNSCC in HIV patients.