Individuals who start HAART with higher CD4 cell counts paradoxically appear to be at higher risk of subsequent immunological failure than patients with lower CD4 cell counts, according to the results of a EuroSIDA study published in the July 1st issue of The Journal of Infectious Diseases.
The study found that both higher viral loads as therapy continued and injection drug use (IDU) were also independent predictors of immunological failure, which occurred in 11.1% of the 2372 in the cohort within twelve months of an initial immunological response to HAART.
This pan-European study sought to determine the predictors of immunological failure after an initial response to HAART. Of the 9803 patients in the EuroSIDA database in the spring of 2003, 6472 were on HAART. Of these, 5450 had a CD4 cell count available within six months of starting HAART, of which 2372 had a CD4 cell increase of 100 cells or more between six and twelve months after starting HAART. These 2372 “immunological responders” were included in the analysis.
Initial response to HAART was defined as an increase of at least 100 cells/mm3 from the pre-HAART CD4 cell count. Immunological failure was defined as CD4 counts dropping to equal or less than their pre-HAART levels.
Eighty-eight percent of individuals in the cohort were white; 78% were men; 48% acquired HIV through gay sex, 25% through IDU and 20% through heterosexual sex; the median age was 37.4 years. The median pre-HAART CD4 cell count was 200 cells/mm3 (interquartile range [IQR] 84-315 cells/mm3) and the median plasma viral load, measured in the six months prior to starting HAART, and available for 80% of those included in the analysis, was 4.57 log10 copies/ml (IQR 3.79-519 log10copies/ml).
The majority of individuals in the cohort (74%) used an unboosted PI regimen, and 12% used an NNRTI-based regimen. Seven percent took HAART that contained both a PI and an NNRTI, and only 38 patients (less than 2% of total) started with a triple nucleoside regimen containing abacavir, which is not now recommended due to the increased risk of treatment failure.
Of the 2372 initial immunological responders, 555 (23%) experienced immunological failure during the study period, 11.1% of which occurred within twelve months of initial immunological success. As time progressed, however, there was a gradual and significant decrease in the rate of immunological failure, reducing 28% each additional year after immunological success (95% CI, 22%-33%; p
In the multivariate model that used time-updated co-variates, three factors were found to be significantly associated with an increased risk of immunological failure: pre-HAART CD4 count (RH, 2.05; 95% CI, 1.83-2.31; p
The authors suggest several reasons why starting HAART with a higher CD4 count led to the highest risk of immunological failure after initial immunological success. It is possible that those with lower CD4 counts had their HAART modified faster and/or more often, since clinicians would be more acutely aware that virological failure for people who began HAART with CD4 counts below 200 cells/mm3 is more clinically relevant than those with higher pre-HAART CD4 counts. This argument is supported by the study’s finding that those with lower pre-HAART CD4s had a significantly shorter time to treatment modification.
Another theory is that individuals with higher CD4 cell counts may have had lower rates of adherence because of their lower risk of an opportunistic infection (OI). This argument is supported by the study’s finding that the median CD4 count at the time of immunological failure was 240 cells/mm3, which is above the threshold of increased OI risk.
The study authors also used various other, stricter, definitions of immunological failure and repeated their calculations. The results were found to be very similar. Interestingly, when they defined failure as a CD4 cell decline of at least 100 cells below baseline (as opposed to simply reaching baseline levels or below), the association between a higher pre-HAART CD4 cell count and failure was even stronger.
The implications of this study, they suggest, may be particularly relevant if current treatment strategies, which are focused more on virological suppression, are changed in the future to focus more on CD4 counts.
Further information on this website
Risk of HIV rebound lessens over time in EuroSIDA cohort - news story
Dragsted UB et al. Predictors of immunological failure after initial response to HAART in HIV-1-infected adults: a EuroSIDA study J Inf Dis 190: 148-55, 2004.