Page 45 - HIV/AIDS Guidelines

unclear. This benefit supports the Panel’s recommendation that ART should be initiated in patients with CD4
3
counts of 350 to 500 cells/mm (AII). Recent evidence demonstrating the public health benefit of earlier
intervention further supports the strength of this recommendation (see Prevention of Sexual Transmission).

Patients with CD4 counts >500 cells/mm 3
3
The NA-ACCORD study also observed patients who started ART at CD4 counts >500 cells/mm or after
CD4 counts dropped below this threshold. The adjusted mortality rates were significantly higher in the 6,935
3
patients who deferred therapy until their CD4 counts fell to <500 cells/mm than in the 2,200 patients who
3
11
started therapy at CD4 count >500 cells/mm (risk ratio: 1.94, 95% CI: 1.37–2.79). Although large and
generally representative of the HIV-infected patients in care in the United States, the study has several
limitations, including the small number of deaths and the potential for unmeasured confounders that might
have influenced outcomes independent of ART.

In contrast, results from 2 cohort studies did not identify a benefit of earlier initiation of therapy in reducing
6
AIDS progression or death. In an analysis of the ART-CC cohort, the rate of progression to AIDS/death
associated with deferral of therapy until CD4 count in the the 351 to 450 cells/mm range was similar to the
3
rate with initiation of therapy with CD4 count in the 451 to 550 cells/mm range (HR: 0.99, 95% CI: 0.76–
3
1.29). There was no significant difference in rate of death identified (HR: 0.93, 95% CI: 0.60–1.44). This
3
study also found that the proportion of patients with CD4 counts between 451 and 550 cells/mm who would
progress to AIDS or death before having a CD4 count <450 cells/mm was low (1.6%; 95% CI: 1.1%–2.1%).
3
In the CASCADE Collaboration, among the 5,162 patients with CD4 counts in the 500 to 799 cells/mm 3
12
range, compared with patients who deferred therapy, those who started ART immediately did not experience
a significant reduction in the composite outcome of progression to AIDS/death (HR: 1.10, 95% CI: 0.67–
1.79) or death (HR: 1.02, 95% CI: 0.49–2.12).

With a better understanding of the pathogenesis of HIV infection, the growing awareness that untreated HIV
infection increases the risk of many non-AIDS-defining diseases (as discussed below), and the benefit of
ART in reducing transmission of HIV, the Panel also recommends initiation of ART in patients with CD4
counts >500 cells/mm (BIII). However, in making this recommendation the Panel notes that the amount of
3
data supporting earlier initiation of therapy decreases as the CD4 count increases to >500 cells/mm and that
3
concerns remain over the unknown overall benefit, long-term risks, and cumulative additional costs
associated with earlier treatment.

When discussing starting ART at high CD4 cell counts (>500 cells/mm ), clinicians should inform patients
3
that data on the clinical benefit of starting treatment at such levels are not conclusive, especially for patients
with very high CD4 counts. The same is true for individuals with low viral load set points at presentation and
for “elite controllers”. Further ongoing research (both randomized clinical trials and cohort studies) to assess
the short- and long-term clinical and public health benefits and cost effectiveness of starting therapy at higher
CD4 counts is needed. Findings from such research will provide the Panel with guidance to make future
recommendations.

Effects of Viral Replication on HIV-Related Morbidity

Since the mid-1990s, measures of viral replication have been known to predict HIV disease progression.
Among untreated HIV-infected individuals, time to clinical progression and mortality is fastest in those with
15
greater viral loads. This finding is confirmed across the wide spectrum of HIV-infected patient populations
18
17
16
such as injection drug users (IDUs), women, and individuals with hemophilia. Several studies have
shown the prognostic value of pretherapy viral load for predicting post-therapy response. 19-20 Once therapy
24
has been initiated, failure to achieve viral suppression 21-23 and viral load at the time of treatment failure are
predictive of clinical disease progression.
More recent studies have examined the impact of ongoing viral replication for both longer durations and at

Guidelines for the Use of Antiretroviral Agents in HIV-1-Infected Adults and Adolescents E-4

Downloaded from http://aidsinfo.nih.gov/guidelines on 12/8/2012 EST.