WHO
CD4 Cell Criteria May Lead to Unnecessary Antiretroviral Therapy Changes in Resource-limited
Settings
By
Liz Highleyman In
industrialized regions such as North America and Europe, efficacy of antiretroviral
therapy (ART) is assessed using viral load testing. Having detectable HIV
RNA indicates that the virus is continuing to replicate and treatment is not working
optimally.
Due to cost and infrastructure requirements, however, viral
load monitoring is often not widely available in resource-limited regions, so
treatment success or failure -- and when to switch drugs -- is gauged based on
changes in CD4 cell count.
 The
World Health Organization (WHO) has set immunological failure criteria for use
in such settings, consisting of any of the following in the absence of other infections:
Persistent CD4 count < 100 cells/mm3;
More than a 50% drop in CD4 count from the peak level;
A CD4 decrease back to the baseline level or lower.
As
reported at the 16th Conference on Retroviruses and Opportunistic
Infections (CROI 2009) this month in Montreal, Steven Reynolds from Johns
Hopkins and colleagues with the Rakai Health Sciences Program evaluated the accuracy
of the WHO criteria for determining treatment failure compared with the "gold
standard" of viral load monitoring.
This analysis looked at 1133 participants
enrolled in an ART program in rural Rakai, Uganda, between June 2004 and September
2007 who had at least 6 months of follow-up data after starting treatment. Immunologic
monitoring was performed using CD4 cell counts every 3 months during the first
year, then every 6 months thereafter. HIV viral load was measured every 6 months.
Participants were followed for up to 44 (median 20) months.
Results
125 participants (11.0%) met the WHO immunological failure criteria during follow-up.
80 patients (7.1%) met the virological failure endpoint of HIV RNA > 10,000
copies/mL.
Only 18 participants (1.6%) met both immunological and virological (> 10,000
copies/mL) failure criteria.
Compared with viral load monitoring using a threshold of > 10,000 copies/mL,
the sensitivity of the WHO criteria was 23%, specificity was 90%, positive predictive
value (PPV) was 14%, and negative predictive value (NPV) was 94%.
Using a threshold of > 5000 copies/mL, the respective values were sensitivity
28%, specificity 90%, PPV 8%, and NPV 97%
Using a threshold of > 400 copies/mL -- the level used in many clinical trials
of antiretroviral drugs in industrialized countries, although today 50 copies/mL
is more common -- the values were sensitivity 23%, specificity 90%, PPV 21%, and
NPV 91%.
These
findings led the investigators to conclude, "Immunologic failure criteria
performed poorly" and would have resulted in a substantial proportion of
participants with suppressed HIV viral load switching drugs unnecessarily.
"Periodic
viral load measurements may be a better marker for treatment failure in our setting,"
they recommended.
Reynolds noted that using the WHO criteria would miss
a considerable number of individuals who were in fact experiencing virological
failure -- which typically occurs before immunological decline -- thereby allowing
disease progression to continue.
Conversely, relying on immunological criteria
alone could lead some people who are not actually experiencing virological failure
-- 107 in this analysis -- to switch therapy unnecessarily.
Because second-line
antiretroviral drugs are generally more expensive than first-line regimens, unnecessary
switching could potentially exceed the cost of periodic viral load testing. In
an attempt to address this problem, researchers are exploring several promising
lower-cost methods of viral load assessment, including measurement of HIV RNA
in dried blood spots.
2/27/09
Reference
S Reynolds, G
Nakigozi, K Newell, and others. Evaluation of the WHO Immunologic Criteria for
ART Failure among Adults in Rakai, Uganda. 16th Conference on Retroviruses and
Opportunistic Infections (CROI 2009). Montreal, Canada. February 8-11, 2009. Abstract
144. |
|
