Toniv,

I think you got it right. Apabetalone increases apolipoprotein-AI (apoAI) synthesis, this apoAI serves as the building block for the production of new HDL particles. Additionally, as you mentioned, apabetalone affects many other cardioprotective pathways. So what may be happening is a synergistic effect of multiple cardioprotective pathways being simultaneously modulated by apabetalone. 

The failure of CETP inhibitors has been surprising from both genetic and epidemiological studies. Patients with loss of function mutations in CETP have decreased risk of cardiovascular disease. HDL-cholesterol levels are correlated with risk of cardiovascular disease. 

The discussion in the evacetrapib NEJM article offers some potential explanations (see below). 

"Several explanations are possible as to why evacetrapib did not result in a lower risk of cardiovascular events than placebo in this trial. Epidemiologic associations between the HDL cholesterol level and the risk of cardiovascular events have been observed primarily in patients who were healthy initially. Beneficial vascular effects of HDL particles, such as cholesterol efflux capacity, may be attenuated in patients who have coronary artery disease or acute coronary syndromes, as compared with healthy trial participants, and a number of mechanisms have been proposed regarding the possible causes of such HDL dysfunction.24,25 Alternatively, some clinicians have expressed concern that the inhibition of CETP pathways may produce HDL particles that are dysfunctional,26 although the protective effects of genetic loss-of-function polymorphisms of CETP do not support this hypothesis. The profile of change in HDL lipid particles and the enhancement of cellular cholesterol efflux with evacetrapib suggest that the HDL cholesterol produced by this agent should be functional, but this concept of functionality27,28 has yet to be validated as a predictor of therapeutic benefit. Moreover, HDL particles possess other properties that are thought to be vasoprotective,24 and the effect of CETP inhibition on these properties is unknown.

Even if HDL cholesterol is dismissed as a modifiable risk factor in patients with vascular disease, it is surprising that the decrease of 37 percentage points in the LDL cholesterol level that was observed with evacetrapib, as compared with placebo, in this trial did not result in a beneficial effect on cardiovascular events. This magnitude of reduction in the LDL cholesterol level is commensurate with the magnitude that has been observed with moderate-intensity statin therapy and that would be expected to produce an approximately 15% lower risk of major coronary events.2 It is conceivable that mechanisms of reduction in the LDL cholesterol level that are specific to CETP inhibition, in contrast to the LDL cholesterol–receptor up-regulation that is induced by statins and ezetimibe, affect LDL cholesterol in ways that do not influence cardiovascular risk. Although treatment with evacetrapib has resulted in reductions of 60 to 70% in the levels of small dense LDL particles,29 effects on the total number of LDL particles and on apolipoprotein B levels (reductions of 22% and 20%, respectively) are considerably less pronounced. The effect on the atherogenicity of the polydisperse LDL cholesterol pattern in association with CETP deficiency or inhibition remains unknown.30

We cannot exclude the possibility that CETP inhibition or evacetrapib itself produced an unmeasured toxic effect that offset the beneficial lipid effects. The observed mean increase in systolic blood pressure that was associated with evacetrapib in this trial (1.2 mm Hg) was small as compared with the increase of 5.4 mm Hg that was induced by torcetrapib in an earlier trial9 and was unlikely to be of sufficient magnitude to worsen cardiovascular outcomes directly. Nevertheless, this finding may be a marker of more profound adverse neuroendocrine or vasomotor effects. Similarly, patients who received evacetrapib had a slight increase in the level of C-reactive protein, which, although small, contrasts with the effects of statins31 and may signal heightened inflammatory responses to CETP inhibition."