798
Long-term Consequences of the Delay between Virologic Failure of HAART and Regimen Modification: A Prospective Cohort Study
Maya Petersen*1, M van der Laan2, S Napravnik3, J Eron3, R Moore4, and S Deeks1
1Univ of California, San Francisco, US; 2Univ of California, Berkeley, US; 3Univ of North Carolina at Chapel Hill, US; and 4Johns Hopkins Univ, Baltimore, MD, US
Background: Current treatment guidelines recommend
immediate modification of combination ART (cART) in HIV-infected individuals
exhibiting evidence of incomplete viral suppression. These recommendations have
not been tested in either observational studies or randomized clinical trials.
Methods: We used prospective data from 2 U.S.-based
clinical cohorts followed prospectively between 1990 and 2006 to estimate the
effect of delay until regimen modification following first HAART failure on
all-cause mortality. We also compared the effect of early vs delayed switch
among subjects failing a protease inhibitor (PI) -based vs a non-PI-based
regimen, the vast majority of whom received a NNRTI-based regimen. The effect
of delayed switching can be confounded if subjects with a poor prognosis modify
therapy earlier than those with a good prognosis. We used a statistical
method—marginal structural models—to control for such time-dependent confounding,
which is not amenable to standard statistical techniques.
Results: A total of 982 subjects contributed a total
of 3414 person-years of follow-up following first regimen failure. The majority
(76%) of first HAART failures occurred among subjects treated with at least 1
PI; 32% of these received ritonavir-boosting. After controlling for
time-dependent confounding by HIV RNA level, CD4 T cell count, treatment
experience, risk group and other factors, delay until treatment modification
was associated with an elevated hazard of all-cause mortality among subjects
failing a non-PI regimen (hazard ratio per additional 3-month delay = 1.24, 95%CI
1.10 to 1.39), but appeared to have a small protective effect among subjects
failing a PI-based regimen (hazard ratio per additional 3-month delay = 0.93,
95%CI 0.87 to 0.99). The effect of a delay in modification following second
regimen failure was similar to that observed following first HAART failure.
Conclusions: Delay in modification after failure of
NNRTI-based regimens is associated with increased risk in disease progression. PI-based
regimens are less dependent on early vs delayed switching strategies. Efforts
should be made to minimize delay until treatment modification in resource-poor
regions, where the vast majority of patients are starting a NNRTI-based regimen
and plasma HIV RNA level monitoring may not be available.
|
