Importantly, in our patients who received a darunavir-containing regimen, we observed a sustained and steady increase in trunk fat tissue, with a median increase of 1 kg over the 96-week study period. Indeed, this fat accumulation, which represented a 12% increase
in the monotherapy arm, was consistent with peripheral fat gain within this group. Therefore, it could be hypothesized that there is an overall increase in fat content, which is slowed by the maintenance of NRTIs in triple-drug strategies. Several factors may explain fat accumulation in the trunk. PIs were associated, in the MK-1775 cell line late 1990s, with central adiposity in long-term HIV-infected patients . In our study, the vast majority of patients were receiving a PI regimen at entry to the study and all received darunavir/r during the study. However, when we looked at the potential impact of prior antiretroviral drug history, we could not
find any impact of drug class on fat accumulation. To varying degrees, the majority of PIs have been associated with metabolic disturbances and lipodystrophy syndrome. Recently, Ferrer et al.  reported that a switch from lopinavir/r to atazanavir/r was associated with an increase in subcutaneous and visceral trunk fat. Several studies have also shown that lipohypertrophy is Ganetespib not restricted to patients receiving a PI [11, 28, 31]. In the study ROS1 by Cameron et al., which evaluated a maintenance strategy where patients received standard triple therapy with efavirenz or lopinavir/r monotherapy, trunk fat content increased similarly in patients receiving efavirenz or lopinavir/r combined with NRTIs . In the ACTG 5142 study, which investigated metabolic outcomes over 96 weeks in patients treated with
efavirenz or lopinavir/r plus two NRTIs vs. an NRTI-sparing regimen with lopinavir/r plus efavirenz, trunk fat was significantly increased from 8.2 kg at entry to 10.4 kg by week 96, with no difference between PI and non-PI treatments . The centralized blinded use of DEXA and quality control is important in fat distribution evaluation in order to reduce disparities in measurements. However, there are several limitations to our study. First, DEXA scans have the advantage of overall quantification of limb and trunk fat contents, but only the addition of a computed tomography (CT) scan with an L4 slice allows differentiation between visceral and subcutaneous compartments within the trunk fat content . Indeed, we cannot rule out the possibility that the overall increase in trunk fat was partially attributable to increased subcutaneous abdominal fat.