The Off-Target Effects of Torcetrapib

Abstract

Cholesterol ester transfer protein (CETP) is responsible for the transfer of lipid from high-density lipoprotein (HDL; so-called good cholesterol) to low-density lipoprotein (LDL; socalled bad cholesterol). From a number of epidemiological studies, there is wide assent that subjects with high HDL are relatively protected in terms of cardiovascular risk, whereas a high LDL is a recognized risk factor. Inhibition of CETP thus presented an attractive therapeutic target, and the CETP inhibitor torcetrapib was shown in preliminary human studies to elevate HDL levels by 60–100%, and to lower LDL by up to 25%. On this basis, a number of large-scale trials were launched in which ultrasonography and other imaging techniques were used; in these trials, however, torcetrapib was found to have no significant effect on coronary atheroma burden or on carotid intima-media thickness (1–3). At the same time, a very large outcome trial [the Investigation of Lipid Level Management to Understand Its Impact on Atherosclerotic Events (ILLUMINATE)] began, comparing torcetrapib plus atorvastatin with atorvastatin alone, in more than 15,000 high-risk subjects (4). On December 4, 2006, the trial was terminated on the advice of the Data Safety Monitoring Board, reflecting higher all-cases mortality and major cardiovascular events in the torcetrapib arm. This outcome is of interest to endocrinologists for several reasons. The torcetrapib subjects, in addition to a 72% elevation in HDL and a 25% lower LDL, showed differences in their plasma electrolytes consistent with a so-called off-target effect of this drug to elevate plasma aldosterone. In addition, they showed an average of 4.5 mm Hg increase in systolic blood pressure; post hoc plasma analysis indeed confirmed elevated plasma aldosterone concentrations in the torcetrapib arm. That torcetrapib is indeed a potent stimulus of aldosterone secretion is confirmed by a series of elegant in vitro and in vivo studies by Hu et al. in this current issue (5). Using human adrenal carcinoma cell lines, they showed that torcetrapib induces secretion of both aldosterone and cortisol at 24 and 48 h to levels equivalent to those seen with angiotensin II. In keeping with these results, they showed torcetrapib to increase expression of both CYP11B1 and CYP11B2 by 6 h to levels equal or higher than those seen with forskolin, and maintenance of stimulation in contrast with the transient effect of angiotensin. Transcriptional profiling showed that of the 10 genes most induced by torcetrapib, seven were similarly induced by angiotensin, evidence for considerable commonality in final mechanism of action. Initial mechanistic differences, however, are clear: angiotensin rapidly increases intracellular [Ca ], whereas torcetrapib does not. Although both elicit a secondary increase in intracellular [Ca ], this can be shown only at relatively high torcetrapib concentrations, above those for maximal steroidogenic effects. There is thus no doubt that torcetrapib is a potent aldosterone secretagogue; studies on other CETP inhibitor molecules clearly showed that activity to inhibit CETP was not necessary for activity as an aldosterone secretagogue, which was thus a true off-target effect. On the basis of these studies, plus complementary studies from elsewhere (6, 7), the default explanation for the increased mortality in the torcetrapib arm is hyperaldosteronism, responsible for the electrolyte differences and the blood pressure elevation. It is worth considering the extent of these changes over the course of the study. At 3 months, the average increase in plasma [Na ] was 0.58 mEq/liter; [HCO3 ], 0.27 mEq/liter; and blood pressure, 4 mm Hg (systolic) and 2 mm Hg (diastolic); plasma [K ] on average fell 0.14 mEq/liter. At 12 months, the equivalent values were 0.61 and 0.35 for [Na ] and [HCO3 ], 4.5/2.1 for blood pressure, and 0.14 for [K ]. Although aldosterone levels were significantly (P 0.001) higher in the torcetrapib group, the post hoc analysis on residual plasma samples, where available, makes accurate quantitation across the two groups impracticable. Higher death rates were seen in patients with more profound lowering of plasma [K ] and higher elevation of [HCO3 ]. At this stage, the smoking gun points to aldosterone as the culprit. That aldosterone can have very deleterious effects at levels excessive for sodium status has been well described, experimentally and clinically, effects that may be independent of blood pressure elevation. In uninephrectomized rats infused with aldosterone, blood pressure can be clamped at control levels by the