Baseline Viral Load Predicts HIV Disease Progression & Baseline CD4 Count Predicts CD4 Decline in ART-Naives
Reported by Jules Levin
Much ado about nothing in the end, Jules Levin
So I guess it's been a controversy regarding the mix of using baseline CD4 and viral load in predicting CD4 decline and disease progression in ART-naives. Mellors said his analysis confirms Rodriguez's finding that HIV RNA at baseline explains very little of the variability in CD4 slope, 3% in Mellors analysis and 4-6% in Rodriguez's analysis. Predicting CD4 decline from baseline can be difficult. In a broad sense VL at baseline can predict CD4 decline but there is a lot of intrapatient variability in CD4 decline so predicting CD4 decline by baseline VL becomes difficult & baseline CD4 appears more predictive, this has not changed by this new information. Lau found baseline CD4 predicts CD4 decline. Baseline VL & CD4 still should and are both considered in deciding when to begin therapy: baseline VL is a better predictor of AIDS and death than baseline CD4 count in naives but together they are better predictors of AIDS & death than either marker alone, both Lau & Mellors said this in their presentations at CROI. But, it appears there are other unknown variables in CD4 decline & disease progression, which we are uncertain about but some of which are discussed in the JAMA Commentary by Keith Henry et al (Sept 2006).
In the next study reported on this subject, which was the very next presentation after Mellors, Lau reported these findings. But it appears the followup for this study was 2 years so the finding by Lau that baseline VL predicts AIDS & death holds for a 2-year length of time which is shorter than what Mellors said; he said that VL predicts AIDS & death when looking out over the longer term of about a decade.
LAU CONCLUDED
--HIV RNA had poor predictive value for CD4
--any explanation for rate of CD4 slope should account for initial CD4
--although HIV RNA did not predict CD4 slope, high predictive value for clinical disease events even after adjusting for CD4 levels was seen --these data reinforces the importance of HIV RNA as an independent marker of HIV disease progression>
As David Margolis summarized after these presentations:
"I think the important thing to get across is that really nothing has changed:
-clinically you look at both CD4 and VL to assess risk
-dropping CD4 over time is bad
-rising VL over time is probably bad
-better surrogate markers would be good to have
-Viral replication contributes to pathogenesis but does not explain everything
-we do not understand everything"
