 |
|
|
| |
Varying Risk of AIDS and Non-AIDS Cancers in Diverse CD4 Brackets
|
| |
| |
5th IAS Conference on HIV Pathogenesis, Treatment and Prevention, July 19-22, 2009, Cape Town
Mark Mascolini
The trend toward a higher AIDS cancer risk at a CD4 count above 350 reflects the experience of many clinicians who see these cancers in people with a CD4 count higher than the current threshold for starting antiretroviral therapy.
A CD4 count under 500 raised the risk of AIDS-defining cancers in a large US naval medical center study [1]. CD4 counts under 700 made non-AIDS cancers more likely in these people, but those correlations held true only for skin cancers in the pre-HAART era.
This retrospective analysis involved 4963 HIV-infected people cared for at the San Diego Naval Medical Center from 1984 through 2008. The investigators used multivariate statistical analysis considering age, race, gender, and year of HIV diagnosis to estimate the risk of AIDS cancers and non-AIDS cancers according to CD4 bracket just before the cancer diagnosis. AIDS cancers are Kaposi sarcoma, non-Hodgkin lymphoma, and invasive cervical cancer.
Records showed that 501 people (10%) got a new cancer diagnosis during the study period, including 21 people with more than one cancer. The researchers counted 336 AIDS cancers (64%) and 186 non-AIDS cancers. Among AIDS cancers, there were 234 Kaposi sarcomas (70% of 336), 101 lymphomas (30%), and 1 cervical cancer. Most non-AIDS cancers were skin cancers (96 of 186, 52%), followed by anal carcinoma (27, 14.5%), Hodgkin lymphoma (17, 9%), prostate cancer (14, 7.5%), and others (32, 17%).
Of the 501 people diagnosed with cancer, 482 (96%) were men in this largely male cohort. Most people with a new cancer were white (57.9%), and 167 (33.6%) were African American, whereas 43.5% were white and 44.7% African American in the entire cohort. At cancer diagnosis, median age was 37.2 years (interquartile range [IQR] 31.6 to 44.1). While 53% had a positive HBV test, 7% had a positive HCV test. The investigators counted 315 deaths (63%) during follow-up.
Median viral load at cancer diagnosis stood at 4000 copies, but the researchers did not have viral load data on one third of the study group. Median lowest-ever CD4 count at cancer diagnosis was 139 (IQR 29 to 314). Median CD4 count at cancer diagnosis was 194 (IQR 37 to 443). Almost half of those diagnosed with cancer (46.5%) had a count under 200 when diagnosed, while 28% had 200 to 500 CD4s, 18% had more than 500, and 8% did not have CD4 data.
Median nadir CD4 count was much lower in people with an AIDS cancer than without an AIDS cancer (70 versus 286), but not much different in people with a non-AIDS cancer than without a non-AIDS cancer (291 versus 273). Median CD4 count in the year before cancer diagnosis or screening was much lower with an AIDS cancer than without an AIDS cancer (53 versus 449) but again not much different with or without a non-AIDS cancer (454 versus 431).
Rates of AIDS cancers fell from 44.7 per 1000 person-years with a most recent CD4 count below 200 and to 7.0 with 200 to 349 CD4s, 3.7 with 350 to 499 CD4s, 1.6 with 500 to 699 CD4s, and 1.3 with more than 699 CD4s. Rates were lower in the HAART era (17.5 at a CD4 count under 200, 4.6 at 200 to 349, 2.0 at 350 to 499, 0.5 at 500 to 699, and 0.6 above 699). For all non-AIDS cancers, rates remained stable across CD4 brackets below 700 (5.0 to 6.0 per 100 person-years), falling only at a count above 699 (3.6 per 100 person-years). Even that difference disappeared in the HAART era.
Looking at AIDS cancers in the HAART era, the researchers calculated almost a 30 times higher risk at a CD4 count under 200 and more than a 7 time higher risk with 200 to 349 CD4s than with a count above 499. The more-than-tripled risk of an AIDS cancer at 350 to 499 CD4s versus 500 or more CD4s stopped short of statistical significance:
· Under 200 CD4s: hazard ratio (HR) 29.8, 95% confidence interval [CI] 5.2 to 171.8, P < 0.001
· 200 to 349 CD4s: HR 7.6, 95% CI 1.4 to 42.2, P = 0.020
· 350 to 499 CD4s: HR 3.6, 95% CI 0.6 to 21.8, P = 0.165
For all non-AIDS cancers, a CD4 count under 500 raised the risk 70% compared with a count above 699 (HR 1.7, 95% CI 1.1 to 2.7, P = 0.023), and a count of 500 to 699 upped the risk 90% compared with higher counts (HR 1.9, 95% CI 1.1 to 3.1, P = 0.016). But those difference disappeared during the HAART era. For skin cancers, the risk was 80% higher with fewer than 500 CD4s and twice higher with 500 to 699 CD4s than with more than 699 CD4s, but again that difference vanished in a HAART era analysis. For nonskin non-AIDS cancers, CD4 bracket had no significant impact on risk before or during the HAART era.
The lack of a link between most recent CD4 and non-AIDS nonskin cancers may reflect the small numbers of such cancers analyzed. The trend toward a higher AIDS cancer risk at a CD4 count above 350 reflects the experience of many clinicians who see these cancers in people with a CD4 count higher than the current threshold for starting antiretroviral therapy.
Unlike this study, research by others has tied lower nadir or recent CD4 count with non-AIDS cancer risk. A study by the same Naval Medical Center investigators that focused only on anal cancer found that every 50-cell higher nadir CD4 count independently cut the risk of this non-AIDS cancer 15% [2]. Another study presented at this conference correlated low CD4 count just before cancer diagnosis with occurrence of non-AIDS cancers of infectious origin [3]. (NATAP reviewed each of these studies separately.) A recent Australian study discerned a link between compromised immunity and non-AIDS cancer risk in people with HIV infection and in organ transplant patients [4].
From Jules: In addition, yesterday I emailed report on a study by INSERM French group that found cumulative exposure to detectable viral load was associated with increased NHL risk and independently cumulative exposure to CD4 <350 was also associated with increased NHL risk.
References
1. Crum-Cianflone N, Huppler Hullsiek K, Marconi V, et al. What CD4 cell count levels are associated with a reduced risk of cancer? 5th IAS Conference on HIV Pathogenesis, Treatment and Prevention. July 19-22, 2009. Cape Town. Abstract WEPEB249.
2. Crum-Cianflone N, Huppler Hullsiek K, Weintrob A, et al. Anal cancers among HIV-infected persons: HAART is not slowing rising incidence. 5th IAS Conference on HIV Pathogenesis, Treatment and Prevention. July 19-22, 2009. Cape Town. Abstract WEAB101.
3. Kesselring A, Gras L, Smit C, et al. Longer duration of exposure to immunodeficiency and detectable viremia both are risk factors for non-AIDS defining malignancies in HIV-1 infected patients on combination antiretroviral therapy. 5th IAS Conference on HIV Pathogenesis, Treatment and Prevention. July 19-22, 2009. Cape Town. Abstract WEAB104.
4. Grulich AE, van Leeuwen MT, Falster MO, Vajdic CM. Incidence of cancers in people with HIV/AIDS compared with immunosuppressed transplant recipients: a meta-analysis. Lancet. 2007;370:59-67.
|
| |
|
|
|
|
|
