Anal cancer in men with HIV much more common in post-HAART era: could HPV vaccination help?

The annual incidence of anal cancer amongst people with HIV has continued to increase in recent years and now stands at 128 cases per 100,000 or one case in 784 people, the Fifth IAS Conference was told on Wednesday by Nancy Crum-Cianflone of the US Infectious Disease Clinical Research Program. This is nearly 100 times the rate in the general population (1.4 per 100,000).

In a separate presentation, Stephen Berman of the Veterans Administration Healthcare System told the conference that findings from a study of HIV/HPV co-infected army veterans implied that up to 89% of men with HIV could potentially derive at least some benefit from one of the currently-licensed HPV vaccines and 53% substantial benefit, providing that HIV infection did not compromise the normal immune response to the vaccine.

In the first study, researchers investigated the incidence of anal cancer in HIV-positive individuals between 1985 and 2008. The figures were obtained from the U.S. Military Natural History Study, a cohort study which followed participants from 1985 till 2008.

A total of 4,901 HIV-infected participants, 55% with documented seroconversion dates, contributed 40,951 person years of follow-up. Researchers investigated the association between CD4 cell count, use of antiretroviral therapy, AIDS-defining illnesses and sexually transmitted infections, and the risk of developing anal cancer.

Anal cancer was diagnosed in 20 patients, all men and 55% Caucasian. At cancer diagnosis the patients’ median age was 42 years, with 40% of cases in men below 40.

The rate of cancer was five times higher in the post-HAART era than in the pre-HAART era with an annual incidence of 11 per 100,000 before 1996 and 55 per 100,000 from 1996-2008.

The incidence increased during the HAART era from 13.4 per 100,000 between 1996 and 2000 to 51 per 100,000 between 2001 and 2005 and 128 per 100,000 in 2006 to 2008.

The incidence of anal cancer was also considerably higher in people with longer duration of HIV infection. Up to 10 years post-infection the rate was 28 per 100,000, from 10-15 years it was 63 per 100,000 and in patients diagnosed for more than 15 years it was 348 per 100,000 - one case per 288 patients per year.

In multivariate analysis only AIDS diagnosis and nadir CD4 count were associated with anal cancer, with having had an AIDS-defining illness raising the risk 3.4 times and the risk decreasing by 15% for every 50 cells/mm³ higher CD4 nadir. Gonorrhoea was also marginally associated with anal cancer with a hazard ratio of 2.33 (p=0.08). Being on HAART was not protective against the development of the condition (p = 0.19).

In another study of patients in members of former members of the armed forces, Dr Stephen Berman investigated infection with human papilloma virus (HPV) and its subtypes in 62 HIV-positive men, 90% of them gay and 90% on HAART with undetectable HIV.

“Little or no data has been published on the ability of HIV positive individuals to respond to [the quadrivalent HPV vaccine] Gardasil,” he commented. “Can they make an appropriate serological response, and which subgroups will not benefit?”

The study therefore took serological tests of study participants to see if they had antibodies to HPV types 16 and 18. These, the two most common ‘high risk’ varieties of HPV, cause 70% of cases of anal and cervical cancer, and the available HPV vaccines Gardasil and Cervarix produce a high level of protection against these strains for individuals who are not already infected with them. In addition Gardasil protects against infection with HPV types 6 and 11, which cause 95% of non-cancerous warts.

The patients also had an anal smear for precancerous cells and an anal screen for HPV DNA, which indicates actively reproducing virus. The patients were then given an initial dose of Gardasil.

The second and third doses of Gardasil were given at month two and month six, and tests taken again at month seven for antibodies to HPV, HPV DNA and precancerous cells.

Results showed that 34% of the men in the study had antibodies to HPV-16, 6.4% to HPV-18 and 3.2% to both. The majority of individuals (56.5%) therefore did not have antibodies to the two strains of HPV with the highest risk of anal and genital cancers.

Sixty-three per cent of men had HPV DNA (of any high-risk type) detectable in anal samples, and 22.6% of men had HPV DNA from types 16, 18 or both. This means that 40% of men had active infections with other high-risk types of HPV (although types other than 16 and 18 are associated with much slower progression to cancer).

Berman commented that the men with detectable DNA could either be men with relatively recent infection and a normal antibody response or men with prolonged infection and a delayed or non-existent antibody response. The latter are at most risk of developing cancer. In HIV negative people antibodies to the infecting subtype of HPV develop within 6-9 months of infection and then clear the infection from the body. This process is delayed and attenuated in people with HIV, which is the probable cause of the higher rates of anal cancer seen.

Twelve individuals had abnormal anal cells, six of whom had squamous intraepithelial lesions, or precancerous cells. Of the twelve, ten had detectable HPV DNA, three of whom had HPV-16, three were co-infected with HPV-16 and HPV-18, and four had types other than 16 or 18.

Of the 25 individual with antibodies to HPV-16, ten (40%) had detectable DNA from this strain detected in samples, in addition to three of the individuals who did not have antibodies to HPV-16, and two men without antibodies to HPV-18 had detectable DNA from this strain. Altogether 11% of patients had both antibodies to HPV16 or 18 and evidence of the genetic material of these strains, indicating a complete or ineffective antibody response.

In answer to a question, Berman commented that it was possible that giving such men the HPV vaccine might stimulate a much stronger response which might result in clearance of infection “but Merck (Gardasil’s manufacturers) have absolutely no data on the efficacy of giving their vaccine to people with an existing infection.”

Altogether, 53% of men had evidence of neither exposure nor infection with HPV-16 or HPV-18 and were therefore likely to derive substantial benefit from Gardasil as long as HIV infection does not impair the immune response, while another 35% had one of the strains and would therefore benefit at least from vaccination against the other one.

Becker commented that the proportion of men with antibodies to HPV16 was higher than reported previously. However since 40% of men had indications of active infections with subtypes other than HPV-16 or 18, second-generation polyvalent vaccines should be developed.

Powerpoint presentations by Nancy Crum-Cianflone and Stephen Berman, and a webcast of the conference session in which they were presented are available on the IAS 2009 website.