Immune Deficiency Linked to Non-AIDS-defining Cancers with Infectious Causes

By Liz Highleyman

Over the past few years, evidence has accumulated that immune deficiency increased the risk for several types of cancer that are not traditionally classified as AIDS-defining (i.e., Kaposi's sarcoma, non-Hodgkin lymphoma, and cervical cancer). Further data along these lines were presented at the 16th Conference on Retroviruses and Opportunistic Infections (CROI 2009) last week in Montreal.

Kaiser California

Michael Silverberg and colleagues reported rates of infection-related cancers at Kaiser Permanente health facilities in California. Like the 3 AIDS-defining cancers, several other malignancies are associated with infectious agents, and therefore might be expected to increase when the immune system is compromised. Anal cancer, for example, is caused by the same types of human papillomavirus (HPV) as cervical cancer, though only the latter is classified as opportunistic.

The investigators identified 18,890 adult HIV patients, pairing each one with 10 age-, sex-, and year-matched HIV negative Kaiser members. The HIV positive members contributed a total of 81,831 person-years (PY) of follow-up, while the 189,804 HIV negative control subjects contributed 971,675 PY. Cohort members were followed from first enrollment after January 1996 until the earliest non-AIDS-defining cancer diagnosis, last Kaiser enrollment, or December 2006. A large majority of the HIV positive patients were gay/bisexual men.

Cancers were classified as infection-related or infection-unrelated. Infection-related cancers included anal cancer, head and neck cancer (often linked to HPV, but also to non-infectious causes such as smoking), liver cancer (a possible consequence of chronic hepatitis B or C), and Hodgkin's lymphoma (associated with Epstein-Barr virus).

Incident (newly occurring) non-AIDS-defining cancers were identified from Kaiser Permanente cancer registries based on the National Cancer Institute's Surveillance, Epidemiology and End Results (SEER) Program. Age- and sex-adjusted rate ratios (RR) were estimated and compared between HIV positive and HIV negative participants. Changes in RRs over time were evaluated for the periods 1996-1999, 2000-2003, and 2004-2006.

Results

482 non-AIDS-defining cancers were identified among the HIV positive patients, 220 infection-related and 269 infection-unrelated (7 people had both).

3065 non-AIDS-defining cancers were identified among the HIV negative participants, 398 infection-related, and 2698 infection-unrelated (31 had both).

The rate of infection-related non-AIDS-defining cancers was 29.7 per 10,000 PY in the HIV positive group, compared with 4.4 per 10,000 PY in the HIV negative group (RR 6.8, or nearly 7 times greater risk; P <0.001).

The rates of infection-unrelated non-AIDS-defining cancers were 36.4 for HIV positive group and 30.6 for the HIV negative group (RR 1.2; P = 0.002).

The RR for infection-unrelated cancers was significant only for the most recent time period: 1.2 in 1996-1999 (P = 0.26), 1.2 in 2000-2003 (P = 0.12), and 1.3 in 2004-2006 (P = 0.02) (P for trend = 0.85).

Looking at infection-related non-AIDS-defining cancers, significant rate differences between HIV positive and HIV negative participants were observed for anal cancer (RR 81.4, or 81% greater risk; P <0.001), Hodgkin's lymphoma (RR 17.4; P <0.001), and head and neck cancers (RR 2.1; P <0.001).

Looking at infection-unrelated cancers, HIV positive participants' rates were significantly higher for kidney cancer (RR 1.8; P = 0.045), lung cancer (RR 1.7; P = 0.004), and melanoma skin cancer (RR 1.7; P = 0.002).

However, HIV positive patients had a lower rate of prostate cancer (RR 0.7; P = 0.007).

The latter result agrees with a previous analysis that also found that a reduced prostate cancer rate in HIV positive men. But this study did not find an elevated risk of liver cancer in people with HIV, as others have]

"HIV positive persons have an elevated risk of non-AIDS-defining cancers, particularly infection-related, which comprised almost half of all the non-AIDS-defining cancers in this population," the investigators concluded.

They noted, however, that the risk of developing an infection-related non-AIDS-defining cancer has declined for HIV positive individuals since the advent of effective combination antiretroviral therapy, while remaining the same for HIV negative individuals, suggesting that treatment restores immune function and can protect against development of certain cancers.

EuroSIDA

A related analysis of the EuroSIDA cohort supported the Kaiser findings. Joanne Reekie and colleagues investigated whether current CD4 count was independently associated with risk of non-AIDS-defining cancers after accounting for traditional and other HIV-associated risk factors. The study included 12,865 EuroSIDA participants with a known CD4 count prior to enrollment, contributing a total of 75,234 PY of follow-up.

Results

A total of 317 non-AIDS-defining cancers were diagnosed in 309 patients during the follow-up period (58 anal, 24 lung, 60 hematological, 48 Hodgkin's lymphoma, 40 genitourinary, 55 digestive, and 80 other), for an incidence rate of 4.2 per 1000 PY.

After adjusting for potential confounding factors, the excess risk of developing a non-AIDS-defining cancer, compared to participants with more than 500 cells/mm3, was:

CD4 count > 50 cells/mm3: 76% higher risk (RR 1.76; P = 0.0408);
51-200 cells/mm3: 82% higher risk (RR 1.82; P = 0.0014);
201-350 cell/mm3: 43% higher risk (RR 1.43).
351-500/mm3: no significant difference (P = 0.89).

Each doubling of current CD4 count was associated with an 11% decrease in the risk of non-AIDS-defining cancers, including anal cancer (RR 0.83), hematological cancers (RR 0.94), genitourinary cancers (RR 0.87), and digestive cancers (RR 0.84).

Due to the small numbers of events, however, this finding only reached statistical significant for anal cancer (P = 0.0038) and digestive cancers (P = 0.0233).

In addition to CD4 count, other significant predictors of non-AIDS-defining cancer were older age, prior AIDS diagnosis, previous history of non-AIDS-defining cancer, time on antiretroviral therapy, and hepatitis B (but not hepatitis C) coinfection.

Based on these findings, the researchers concluded, "Immunosuppression was associated with an excess risk of [non-AIDS-defining cancers]."

"One possible explanation is enhanced oncogenic potential by pro-oncogenic viruses (e.g., human papillomavirus and anal cancer)," they hypothesized. "An infectious oncogenesis has only been established for a few [non-AIDS-defining cancers] linked with immunosuppression in this study and other mechanisms may also contribute. As the immunosuppression may be reversed by [combination antiretroviral therapy], HIV is a suitable candidate model for improving our understanding of how immunosuppression affects oncogenic transformation."

2/20/09

References

M Silverberg, W Leyden, C Chao, and others. Infection-related Non-AIDS-defining Cancer Risk in HIV-infected and -uninfected Persons. 16th Conference on Retroviruses and Opportunistic Infections (CROI 2009). Montreal, Canada. February 8-11, 2009. Abstract 30.

J Reekie, A Mocroft, F Engsig, and the EuroSIDA Study Group. Relationship between Current Level of Immunodeficiency and Non-AIDS-defining Malignancies. 16th Conference on Retroviruses and Opportunistic Infections (CROI 2009). Montreal, Canada. February 8-11, 2009. Abstract 860a.

A Grulich. HIV, Immune Deficiency, and Malignancy. 16th Conference on Retroviruses and Opportunistic Infections (CROI 2009). Montreal, Canada. February 8-11, 2009. Abstract 178.