There were no significant differences among the three study groups for duration of HIV, age, body mass index and CD4+ T cell count. Although the challenge of adherence to drugs is substantial in routine HIV care, we found a good tolerance to fenofibrate, and no complaint was recorded. As expected, only in fenofibrate treated patients the decrease in fasting plasma triglyceride concentration was statistically significant as well as the increase in plasma HDL-cholesterol levels (Figure 1A). A reduction in plasma total cholesterol and MCP-1/CCL2 concentrations were also observed but it did not reach statistical significance.
With the rationale that the target for fenofibrate, PPARα, is expressed in leukocytes  we explored the mRNA expression of selected chemokine genes in whole blood. We found that with respect to untreated patients there was no change in the expression of chemokine receptors CX3CR1 and CXCR4 but the expression of CCR2 was significantly decreased in patients treated with fenofibrate (U-Mann-Whitney, p = 0.016; Figure 1B). Of note, CCR2 gene expression was positively correlated with CD4+ T cell count only in those patients that were undergoing fenofibrate treatment (Spearman, ρ = 0.7; p = 0.02). The expression of ligands for these receptors was clearly different. We did not find any significant effect of fenofibrate in the MCP-1/CCL2 gene expression. However, there was a significant decrease in the gene expression of the CX3CL1/Fractalkine in patients treated with fenofibrate with respect to untreated hypertriglyceridemic patients (U-Mann-Whitney, p = 0.038; Figure 1B). This is probably accompanied by decreased CX3CL1 plasma levels and a similar effect in the contiguous endothelial cells, which, in combination with the CCR2 decreased expression, could be beneficial with respect to the risk of atherosclerosis considering their well documented role in the local accumulation of monocytes [8, 9]. However, fenofibrate had an intriguing effect in SDF-1 expression neutralizing the significant decrease in its expression observed in untreated hypertriglyceridemics (U-Mann-Whitney, p = 0.026; Figure 1B), presumably increasing SDF-1 values. SDF-1 plays a significant role in HIV infection because it is the natural ligand for CXCR4 which is used by HIV T-tropic strains to enter into the cells in advanced stages of the disease . Additionally, although the interpretation of SDF-1 levels is difficult we have previously shown that the allelic variant SDF1-3'A appears to be protective against the development of carotid atherosclerosis .
In multivariate analysis, only fenofibrate treatment significantly contributed to explain CCR2 [B = 0.18 (0.003 to 0.09); P = 0.04] and SDF-1 [B = 4.6 (1.1 to 18.0); P = 0.03] expression levels in hypertriglyceridemic patients. This relationship, however, was not maintained in CX3CL1/Fractalkine expression [B = 0.23 (0.04 to 1.1); P = 0.07].
Therefore, our results show less lipid disturbances with the addition of fenofibrate in patients with HIV, which is accompanied by significant changes in the pattern of chemokine gene expression in circulating leukocytes. The data suggest that PPARα activators may be useful in the attempt to decrease the risk of atherosclerosis and may act as modulators of systemic inflammation and the associated vascular response. However, further studies in other type of populations without lipid abnormalities should be needed to confirm the fenofibrate effect or any lipid-lowering drug effect on systemic inflammation. Moreover, a note of caution should be added before the indication of these active drugs and further studies are needed to disregard potential deleterious effects on the HIV infection itself.