 |
|
|
| |
Duration of PIs or NNRTIs Has No Consistent Impact on Non-AIDS Cancer Rates
|
| |
| |
18th Conference on Retroviruses and Opportunistic Infections, February 27-March 2, 2011, Boston
Mark Mascolini
Duration of treatment with a protease inhibitor (PI) or nonnucleoside (NNRTI) had no significant impact on incidence of major non-AIDS cancers in a large analysis of patients seen in California's Kaiser Permanente system [1]. There were trends to lower prostate cancer incidence with each year of PI therapy, and to higher anal cancer incidence with each year of PI therapy. But Kaiser investigators cautioned that those trends must be examined in larger and longer studies. Longer duration of both PIs and NNRTIs was associated with lower incidence of two AIDS cancers, Kaposi sarcoma and non-Hodgkin lymphoma.
Rates of certain non-AIDS cancers appear to be rising in the combination antiretroviral era. But because few studies have analyzed the potential impact of antiretroviral duration on new cancer diagnoses (incidence), Kaiser researchers planned this analysis of the database from their integrated healthcare system.
The study group included 12,872 HIV-positive Kaiser members seen from 1996 through 2008. The researchers excluded anyone under 18 years old and anyone without complete pharmacy records for PIs and NNRTIs. Using Kaiser HIV and cancer registries, they identified 313 cases of Kaposi sarcoma, 159 cases of non-Hodgkin lymphoma, 52 anal cancers, 42 prostate cancers, 40 lung cancers, and 29 cases of Hodgkin lymphoma. Other cancers occurred too infrequently to support statistical analysis.
The Kaiser team tracked patients from enrollment in the Kaiser system until cancer diagnosis, death, withdrawal from the Kaiser system, or the end of 2008. They computed incidence and rate ratios according to duration of PI or NNRTI use: 0, 1, 2, 3, 4, or 5 or more years. Potential confounders considered in their statistical analysis were age, gender, race/ethnicity, calendar year, smoking, alcohol/drug abuse, overweight/obesity, initial CD4 count, HIV transmission risk, and duration of HIV infection.
Kaiser entry age of the study group averaged 40 years, 90% were men, 76% were gay men, and 56% were white. Known HIV duration averaged 2.7 years, 44% ever smoked, 39% were overweight or obese, and 22% abused alcohol or drugs. These people took a PI for an average 3.6 years or an NNRTI for an average 2.8 years. Total person-years of antiretroviral use were 21,249 for PIs and 15,643 for NNRTI. Total follow-up in the untreated comparison group was 19,479 years.
Kaposi sarcoma incidence dropped 31% with each year of PI therapy (rate ratio [RR] 0.69, 95% confidence interval [CI] 0.62 to 0.77, P < 0.001) and 32% with each year of NNRTI therapy (RR 0.68, 95% CI 0.58 to 0.59, P < 0.001). Incidence of the other AIDS cancer considered, non-Hodgkin lymphoma, also fell at an equivalent rate per year of PIs (16%, RR 0.84, 95% CI 0.78 to 0.91, P < 0.001) and NNRTIs (15%, RR 0.85, 95% CI 0.77 to 0.95, P = 0.004).
Duration of NNRTI therapy had no impact on incidence of anal cancer. Each year of PI therapy was associated with a 9% increase in anal cancer incidence, but that increase was not statistically significant (RR 1.09, 95% CI 0.97 to 1.23, P = 0.15). Each year of PI therapy lowered prostate cancer incidence 9%, but again this change lacked statistical significance (RR 0.91, 95% CI 0.81 to 1.02, P = 0.10). Duration of NNRTI therapy had no impact on prostate cancer incidence.
Duration of PIs or NNRTIs had no measurable effect on incidence of two other non-AIDS cancers, lung cancer and Hodgkin lymphoma.
The Kaiser investigators cautioned that the findings may not apply to women, who made up only 10% of their study population. They cited four strengths of their study: (1) direct evaluation of PI and NNRTI exposure, (2) analysis of a large population-based cohort, (3) comprehensive HIV and cancer ascertainment in the Kaiser system, and (4) statistical adjustment for several cancer risk factors.
Although this study did not directly assess antiretroviral adherence, other analyses of this HIV population found about 80% to 85% adherence measured by pharmacy refill records. The researchers suggested that the borderline associations between PI duration and anal and prostate cancer incidence require further study with longer follow-up. They agreed that differing use of anal Pap smears from center to center could confound the anal cancer results.
Reference
1. Michael Silverberg M, Leyden W, Xu X, et al. ART use and risk of cancer in HIV patients. 18th Conference on Retroviruses and Opportunistic Infections. February 27-March 2, 2011. Boston. Abstract 81.
|
| |
|
|
|
|
|
