icon-folder.gif   Conference Reports for NATAP  
  51th ICAAC
Chicago, IL
September 17-20, 2011
Back grey_arrow_rt.gif
Under 10% in Large Canadian Cohort Reach Normal CD4/CD8 Ratio
  51st Interscience Conference on Antimicrobial Agents and Chemotherapy (ICAAC), September 17-20, 2011, Chicago

Mark Mascolini

In the current antiretroviral era, only 7% of a 4600-person Canadian cohort reached a normal CD4/CD8 ratio after starting treatment [1]. Reaching a normal ratio did not independently lower the risk of a new AIDS diagnosis or death, but the study did not analyze the potential impact of a normal ratio on non-AIDS diseases.

Untreated HIV infection inverts the CD4/CD8 ratio as CD4 counts fall and CD8 counts rise. Before the current antiretroviral era, the CD4/CD8 ratio received some attention as a predictor of HIV disease progression. A study of 424 gay men in the Multicenter AIDS Cohort Study who became infected with HIV before the end of 1995 found a significant association between longer time to CD4/CD8 ratio inversion after HIV infection and longer time to AIDS, even after statistical adjusting for CD4 count and viral load at the first seropositive visit and over the first 3 seropositive visits [2].

To see how often current antiretroviral regimens restore a normal CD4/CD8 ratio and whether a normal ratio affects risk of HIV disease progression or death, researchers from the Canadian Observational Cohort Collaboration (CANOC) analyzed ratios in 4588 adults who started combination antiretroviral therapy (cART) between 2000 and 2010. The CANOC researchers defined a normal ratio as at or above 1.2 (a little more than 1 CD4 cell for every CD8 cell) on two consecutive measures after starting cART. All study participants had a ratio below 1.2 before starting cART and had at least two CD4/CD8 measurements after starting therapy.

CANOC currently includes people from 8 cohorts in British Columbia, Ontario, and Quebec. Of the 4588 cohort members analyzed, 81% were men, 41% were men who have sex with men (MSM), and 22% had HCV infection. Median year of starting cART was 2005, median initial age 40, median pretreatment CD4 count 190, and median pretreatment viral load 80,000 copies. The largest proportion of cohort members (47%) was from British Columbia, while 31% were from Ontario and 22% from Quebec.

During a median follow-up of 2.86 years, 314 cohort members (6.8%) reached a normal CD4/CD8 ratio. In an analysis that adjusted for gender, injection drug use, province, and number of CD4/CD8 measurements, several factors independently lowered chances of reaching a normal CD4/CD8 ratio, at the following hazard ratios (HR) (and 95% confidence intervals [CI]):

-- Initial CD4 count below 200 (vs over 350): HR 0.07 (0.05 to 0.10), P = 0.0001

-- Initial CD4 count 200 to 350 (vs over 350): HR 0.35 (0.25 to 0.48), P < 0.0001

-- Initial CD8 count 500 to 750 (vs under 500): HR 0.31 (0.23 to 0.44), P < 0.0001

-- Initial CD8 count 750 to 1150 (vs under 500): HR 0.19 (0.13 to 0.27), P < 0.0001

-- Initial CD8 count above 1150 (vs under 500): HR 0.07 (0.05 to 0.12), P < 0.0001

-- Every 10-fold higher time-updated viral load: HR 0.65 (0.55 to 0.78), P < 0.0001

-- MSM: HR 0.68 (0.50 to 0.93), P = 0.01

Starting cART in 2006-2007 (compared with 2000-2001) more than doubled chances of reaching a normal CD4/CD8 ratio (HR 2.25, 95% CI 1.42 to 3.56, P = 0.0005). Having more CD4/CD8 calculations per year of follow-up also independently raised chances of reaching a normal ratio.

After cART began, 4481 cohort members had 260 new AIDS diagnoses (5.8%), and 329 people (7.3%) died. Univariate analysis determined that a time-updated normal CD4/CD8 ratio lowered the risk of a new AIDS diagnosis and death combined. But after statistical adjustment for age, gender, HIV risk factors, province, and cART regimen, attaining a normal CD4/CD8 ratio no longer independently lowered the risk of AIDS or death (HR 0.59, 95% CI 0.19 to 1.89, P = 0.37).

Every 10-fold higher time-updated viral load upped the risk of AIDS or death 67% (HR 1.67, 95% CI 1.51 to 1.84, P < 0.0001). Age 35 to 45 (versus under 35) raised the progression risk about 50% (HR 1.52, 95% CI 1.09 to 2.11, P = 0.01), and age over 45 more than doubled the progression risk (HR 2.13, 95% CI 1.48 to 3.07, P < 0.0001).

Several classic risk factors lowered the risk of progression at the following hazard ratios (and 95% CIs):

-- MSM: HR 0.61 (0.43 to 0.87), P = 0.006

-- Starting cART in 2008-2010 (vs 2000-2001): HR 0.36 (0.17 to 0.75), P = 0.006

-- Initial CD4 count 500-750 (vs under 500): HR 0.61 (0.42 to 0.87), P = 0.006

-- Initial CD4 count 750-1150 (vs under 500): HR 0.66 (0.46 to 0.96), P = 0.03

To these researchers' surprise, starting cART with a regimen built around a nonnucleoside or protease inhibitor (PI) raised AIDS or death risk when compared with starting a triple-nucleoside combination, as 3% of this cohort did. They proposed that this finding reflects selection bias--clinicians may have been more likely to prescribe less potent triple-nucleoside regimens for healthier people. When the investigators eliminated triple-nuke regimens from the analysis and used an unboosted PI regimen as the reference for comparison, boosted PI regimens and nonnucleoside regimens were no longer associated with a higher risk of AIDS or death.

Much recent work demonstrates ongoing inflammation in people who reach an undetectable viral load while taking cART, which would be reflected in a persistently high CD8 count [3]. But this study did not try to figure whether reaching a normal CD4/CD8 ratio lowers the risk of inflammation-associated non-AIDS diseases.


1. Leung, V, Gillis J, Raboud J, Ndumbi P, Tsoukas C, Klein M. Predictors of CD4/CD8 ratio normalization and its affect on health outcomes in the era of combination antiretroviral therapy (cART). 51st Interscience Conference on Antimicrobial Agents and Chemotherapy (ICAAC). September 17-20, 2011. Chicago. Abstract H1-1409.

2. Margolick JB, Gange SJ, Detels R, et al. Impact of inversion of the CD4/CD8 ratio on the natural history of HIV-1 infection. J Acquir Immune Defic Syndr. 2006;42:620-626.

3. Deeks SG. HIV infection, inflammation, immunosenescence, and aging. Annu Rev Med. 2011;62:141-155.