Secondary Logo

Journal Logo

Kidney and Bladder Cancers Among People With AIDS in the United States

Layman, Annah B MPH*†; Engels, Eric A MD*

JAIDS Journal of Acquired Immune Deficiency Syndromes: July 1st, 2008 - Volume 48 - Issue 3 - p 365-367
doi: 10.1097/QAI.0b013e31817ae5da
Letters to the Editor

*Infections and Immunoepidemiology Branch, Division of Cancer Epidemiology and Genetics National Cancer Institute, National Institutes of Health Rockville, MD; dagger;Department of Health Science Brigham Young University Provo, UT

To the Editor:

Individuals with human immunodeficiency virus/acquired immunodeficiency syndrome (HIV/AIDS) have an elevated risk for certain cancers related to immunosuppression and infections with oncogenic viruses.1,2 Recent studies have suggested that the risk for renal cell carcinoma (RCC) is elevated in this population and that its incidence may be rising over time.1,2 Kidney diseases, such as HIV-associated nephropathy, could contribute to occurrence of RCC. Moreover, although HIV-infected people are living longer with the availability of highly active antiretroviral therapy (HAART) since 1996, some HIV medications are nephrotoxic.3 HIV-infected individuals might also be expected to have an increased risk for bladder cancer because many HIV-infected individuals smoke tobacco, an established risk factor for bladder cancer. Malignancies of the kidney and bladder are responsible for over 26,000 deaths each year in the United States.

We used data from the HIV/AIDS Cancer Match Study to systematically examine the epidemiology of these urinary tract cancers among people with AIDS in the United States. The HIV/AIDS Cancer Match Study links data from HIV/AIDS and cancer registries for 9 states and 5 metropolitan areas.1,4 A total of 553,040 persons with AIDS were identified from HIV/AIDS registries. After excluding children, individuals diagnosed with AIDS before 1980, and persons whose follow-up time did not overlap with years of cancer registry coverage, we analyzed data on 499,151 persons with AIDS diagnosed during 1980-2004.

We considered RCC, bladder cancer, and other urinary tract cancers arising in an overall follow-up period from 60 months before to 60 months after AIDS diagnosis.4 For certain analyses, we further restricted consideration to the 2-year incidence period from 4 to 27 months after AIDS onset (406,404 persons with AIDS).4 We calculated standardized incidence ratios (SIRs) to compare cancer risk with that expected in the general population. Expected counts were derived using registry-, age-, race-, sex-, and calendar year-specific incidence rates from the cancer registries.4 To evaluate cancer risk across subgroups, we calculated relative proportions (cancer cases per 100,000 persons with AIDS) and compared these proportions using logistic regression. We also used Poisson regression to model trends in SIRs with respect to calendar year of AIDS diagnosis. Likewise, we used Poisson regression to examine the relationship between CD4 count at AIDS onset and cancer incidence in the subsequent incidence period (4-27 months after AIDS onset).

During the overall follow-up period (−60 to +60 months), 130 RCCs, 109 bladder cancers, and 56 other urinary tract cancers were diagnosed. The risk of RCC and other urinary tract cancers was similar to that in the general population {SIR [95% confidence interval (CI)] = 1.1 (0.9 to 1.3) and 1.1 (0.8 to 1.4), respectively}. Bladder cancer risk was significantly lower than in the general population [SIR (95% CI) = 0.7 (0.6 to 0.9)]. When we focused on the incidence period (+4 to +27 months), 36 RCCs and 13 bladder cancers were identified (incidence 5.8 and 2.1 per 100,000 person-years, respectively). Eighteen other urinary tract cancers were diagnosed in the incidence period. Compared to the general population, the incidence of RCC and other urinary tract cancers was not significantly elevated [SIR (95% CI) = 1.2 (0.9 to 1.7) and 1.6 (0.9 to 2.5), respectively], whereas bladder cancer incidence was again reduced [SIR (95% CI) = 0.4 (0.2 to 0.7)].

Table 1 summarizes demographic characteristics associated with RCC and bladder cancer risk during the overall follow-up period (−60 to +60 months). For both cancers, males had a significantly higher risk than females, and risk increased with age, similar to patterns seen in the general population.5 Bladder cancer risk also differed significantly among racial/ethnic groups and HIV exposure groups; in contrast, RCC risk did not vary across these groups. RCC risk was not especially elevated among African Americans, despite higher rates of HIV-associated nephropathy in individuals of African ancestry compared with other racial/ethnic groups.3



RCC risk increased significantly across calendar time when we grouped years of AIDS diagnosis into 1980-1989, 1990-1995, and 1996-2004 (Ptrend = 0.0006, Table 1) or considered individual years of AIDS diagnosis (Ptrend = 0.01). However, when risk was measured relative to the general population, there was no clear increase over time, regardless of whether we considered SIRs for the 3 calendar periods (Ptrend = 0.45) or for individual calendar years (Ptrend = 0.95), suggesting that the rise in RCC risk reflects, at least in part, the changing demographics of the AIDS population. Also, the trend in risk for RCC over calendar year was weaker than we previously reported1 because with additional data the SIRs for the HAART era decreased, especially for recent years. Bladder cancer risk did not vary significantly across calendar periods (Table 1) or across individual calendar years (Ptrend = 0.20).

We found no association between CD4 count at AIDS onset and risk of either RCC or bladder cancer during the incidence period (Ptrend = 0.77 and 0.79, respectively). This result and the absence of a decline in risk during the HAART era argues against a major role for immunosuppression in the development of these cancers.1

Although RCC risk was not significantly elevated in our study, the statistical uncertainty in our results does not exclude some excess risk in people with AIDS. An elevated risk for RCC was suggested by a recent meta-analysis of registry-based studies,2 which included earlier data from our HIV/AIDS Cancer Match Study.1,4 When we recalculated the meta-analysis results using the updated data in the present report, the overall SIR was lower than that reported by Grulich et al2 (SIR 1.3 vs 1.5 previously), but the overall SIR remained significantly elevated [SIR (95% CI) = 1.3 (1.1 to 1.6)]. This result suggests that a modest elevation in RCC incidence may indeed be present in persons with HIV/AIDS.

The deficit in bladder cancer risk in people with AIDS was somewhat unexpected, given high rates of smoking in this population. Although it is possible that the low bladder cancer risk is an artifact (eg, reflecting underdiagnosis resulting from lack of evaluation of urinary symptoms among ill persons or lack of access to medical care), the decreased risk for bladder cancer has been reported in other studies.1,2 In their meta-analysis, Grulich et al2 found the risk of bladder cancer to be low but not significantly so [SIR (95% CI) = 0.8 (0.4 to 1.3)]. When we added the updated data from this study to the meta-analysis, this deficit became statistically significant [SIR (95% CI) = 0.7 (0.5 to 0.8)].

Strengths of the HIV/AIDS Cancer Match Study include the large and representative sample of the US population with AIDS, which allowed us to systematically evaluate trends in incidence and demographic risk factors for urinary tract malignancies. A limitation is that information on some important risk factors of interest (eg, presence of renal disease, smoking status, and antiretroviral medication use) was unavailable. Further, our study only considered individuals with AIDS, who represent a minority of the HIV population, especially in the HAART era. As a result, we could not evaluate the relationships of cancer risk with CD4 count at levels observed in less immunocompromised persons.

In summary, we did not find a statistically elevated risk for RCC in people with AIDS, although consideration of our results with those from other studies suggests that RCC risk might still be modestly elevated in this population. For bladder cancer, we found a decreased risk among people with AIDS. Additional epidemiologic research regarding risk factors for these cancers, especially renal disease and tobacco, is needed to better understand their etiology among HIV-infected people. Continued monitoring of trends in cancer incidence will be useful, particularly, as new therapies improve life expectancy among this population.

Back to Top | Article Outline


We gratefully acknowledge the contributions of the many staff members at the HIV/AIDS and cancer registries who collected the data, prepared data files for the matches, and facilitated the record linkages. We thank Brigham Young University for helping facilitate an internship, which allowed this research to be conducted. Supported by the Intramural Research Program of the National Cancer Institute.

Annah B. Layman, MPH*†

Eric A. Engels, MD*

*Infections and Immunoepidemiology Branch, Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health Rockville, MD

†Department of Health Science, Brigham, Young University Provo, UT

Back to Top | Article Outline


1. Engels EA, Pfeiffer RM, Goedert JJ, et al. Trends in cancer risk among people with AIDS in the United States 1980-2002. AIDS. 2006;20:1645-1654.
2. Grulich AE, van Leeuwen MT, Falster MO, et al. Incidence of cancers in people with HIV/AIDS compared with immunosuppressed transplant recipients: a meta-analysis. Lancet. 2007;370:59-67.
3. Izzedine H, Deray G. The nephrologist in the HAART era. AIDS. 2007;21:409-421.
4. Frisch M, Biggar RJ, Engels EA, et al. Association of cancer with AIDS-related immunosuppression in adults. JAMA. 2001;285:1736-1745.
5. Schottenfeld D, Fraumeni JF Jr, eds. Cancer Epidemiology and Prevention. 3rd ed. New York: Oxford University Press; 2006.
© 2008 Lippincott Williams & Wilkins, Inc.