A study of 92 men with HIV in New York has found that nine per cent had high—grade anal intraepithelial neoplasia (AIN), a condition that carries a risk of developing into invasive anal cancer without treatment.
Receptive anal intercourse was the strongest risk factor for infection with cancer-causing types of human papillomavirus (HPV), whilst the factor most strongly predictive of AIN (both low and high grade) was a low nadir CD4 cell count (the lowest level ever reached).
Lower CD4 cell count nadir was associated with an increased risk of AIN, even after controlling for current treatment.
The investigators argue that while receptive anal intercourse is a strong indicator of HPV and virus-associated cell damage, a relatively intact immune system can contain further progression. Conversely, they also suggest that whilst immune reconstitution on highly active antiretroviral therapy (HAART) in individuals with a low CD4 cell count nadir may be sufficiently robust to stop prophylaxis against opportunistic infections, deficits in immune cell proliferation may persist that continue to expose these people to a higher risk of developing AIN.
Nearly a quarter of the cases of AIN were in men who had never had receptive anal intercourse and who were heterosexual. The investigators argue that sexual orientation or behaviour should not be used as entry criteria for programmes that screen for anal cancer or its precursors.
Being on HAART or having an undetectable viral load were both strongly protective factors, reducing the risk of AIN at least tenfold. A previous study did not find HAART protective, but did find more anal cancer in people with higher viral loads.
The present study recruited men from the Columbia Presbyterian Medical Center Infectious Diseases Clinic in New York. This serves an almost entirely Latino and Black population, and only seven individuals (8%) of the study group were white. 53% were Latino and 36% African-American.
Half of the study population self-identified as gay and 11% as bisexual; 60% reported ever having had receptive anal intercourse.
Eighty-two per cent were receiving HAART and were reasonably well-controlled virologically, with two-thirds of the whole group having an HIV viral load under 400 copies/ml.
Several clinical investigations were made. Cells from the anal canal were tested for the presence of HPV DNA and genotyped. Microscopy was used to determined the presence of atypical cells, and trial participants had a biopsy to determine AIN.
Sixty-one per cent of participants were found to have at least one high-risk type of HPV, while 18% had multiple high-risk types.
Forty-seven per cent of the participants whose cells were analysed by microscopy had abnormal results, with 30% having the microscopic precursors of neoplasia termed squamous intraepithelial lesions.
Forty per cent of participants had AIN when biopsies were examined, though only 9% had the high grade version which is normally seen as an indicator for topical or surgical intervention.
When risk factors for these results were determined it was found that the only significant risk factor for infection with high-risk HPV was a history of receptive anal sex (odds ratio 7.1, p < 0.001), though age under 40 showed a trend to significance (odd ratio 2.4, p = 0.12)
For abnormal squamous cells, receptive anal intercourse was still a risk factor (odds ratio 5.7, p = 0.03), though nadir CD4 count was almost significant (odds ratio 1.7 for a low nadir, p = 0.06).
For anal intraepithelial neoplasia, however, the CD4 nadir was the only significant predictor (odds ratio 2.0 for a low nadir, p = 0.03), though age under 40 was almost significant (odds ratio 3.6, p = 0.06)
Current use of HAART was strongly protective both against abnormal cells (odds ratio 0.06 or an 18-fold reduction in frequency compared to untreated patients, p < 0.01) and AIN (odds ratio 0.09 for an elevenfold reduction, p = 0.03).
Only a minority of cases of high-grade AIN ever turn into invasive anal cancer. Previous studies found its incidence to be 65-70 cases in 100,000 person years, or about one case in 1500 patients a year (Goedert 1998).
The cross-sectional design of the study was unable to show whether HAART stops progression of cancer or promotes regression. Another limitation of the study was that it did not distinguish between risk factors for high and low grade AIN.
Wilkin TJ et al. Anal intraepithelial neoplasia in heterosexual and homosexual HIV-positive men with access to antiretroviral therapy. J Infect Dis 190: 1685-1691, 2004.
Goedert JJ et al. Spectrum of AIDS-associated malignant disorders. Lancet 351: 1833-1839, 1998.