Search tips
Search criteria 


Logo of wtpaEurope PMCEurope PMC Funders GroupSubmit a Manuscript
Circulation. Author manuscript; available in PMC 2010 July 5.
Published in final edited form as:
PMCID: PMC2811869

Separating the mechanism-based and off-target actions of CETP-inhibitors using CETP gene polymorphisms

Reecha Sofat, MRCP,1, Aroon D Hingorani, PhD, FRCP,1,2, Liam Smeeth, MRCGP, PhD,3 Steve E Humphries, PhD, MRCP, FRCPath,4 Philippa J Talmud, PhD,4 Jackie Cooper, MSc,4 Tina Shah, PhD,1 Manjinder S Sandhu, PhD,5 Sally L Ricketts, PhD,5 S Matthijs Boekholdt, MD, PhD,5,31 Nicholas Wareham, MBBS FRCP,6 Kay Tee Khaw, MBBChir, FRCP,7 Meena Kumari, PhD,2 Mika Kivimaki, PhD,2 Michael Marmot, PhD, FRCP,2 Folkert W Asselbergs, MD, PhD,8,17 Pim van der Harst, MD, PhD,8 Robin P F Dullaart, MD, PhD,8 Gerjan Navis, MD, PhD,8 Dirk J van Veldhuisen, MD, PhD,8 Wiek H Van Gilst, PhD,8 John F Thompson, PhD,9 Pamela McCaskie, PhD,10 Lyle J Palmer, PhD,10 Marcello Arca, MD,11 Fabiana Quagliarini, MSc,11 Carlo Gaudio, MD,12 François Cambien, MD,13 Viviane Nicaud, MA,13 Odette Poirer, PhD,14 Vilmundur Gudnason, MD, PhD,15 Aaron Isaacs, PhD,16 Jacqueline C M Witteman, PhD,16 Cornelia M van Duijn, PhD,16 Michael Pencina, PhD,17 Ramachandran. S Vasan, MD,17 Ralph B D'Agostino, Sr, PhD,17 Jose Ordovas, PhD,17 Tricia Y. Li, MSc,18 Sakari Kakko, MD PhD,19 Heikki Kauma, MD, PhD,19 Markku J. Savolainen, MD PhD,19 Y.Antero Kesäniemi, MD, PhD, FAHA,19 Anton Sandhofer, MD,20 Bernhard Paulweber, MD,21 Jose V Sorli, MD, PhD,22 Akimoto Goto, MD, PhD,23 Shinji Yokoyama, MD, PhD, FRCPC,24 Kenji Okumura, MD, PhD,25 Benjamin D Horne, PhD, MPH,26 Chris Packard, DSc,27 Dilys Freeman, BSc, PhD,27 Ian Ford, PhD,28 Naveed Sattar, PhD, FRCPath,29 Valerie McCormack, PhD,3,30 Debbie A Lawlor, PhD,31 Shah Ebrahim, Dm, MSc, FFPHM,3 George Davey Smith, MD, DSc, FFPHM,31 John J P Kastelein, MD, PhD,32 John Deanfield, BA, BCh, MB, FRCP,33 and Juan P Casas, MD, PhD2,3



Cholesteryl ester transfer protein (CETP) inhibitors raise HDL-cholesterol but torcetrapib, the first-in-class inhibitor tested in a large outcome trial caused unexpected blood pressure elevation and increased cardiovascular events. Whether the hypertensive effect resulted from CETP-inhibition or an off-target action of torcetrapib has been debated. We hypothesised that common single nucleotide polymorphisms (SNPs) in the CETP-gene could help distinguish mechanism-based from off-target actions of CETP-inhibitors to inform on the validity of CETP as a therapeutic target.

Methods and Results

We compared the effect of CETP SNPs and torcetrapib treatment on lipid fractions, blood pressure and electrolytes in up to 67,687 individuals from genetic studies and 17,911 from randomised trials. CETP SNPs and torcetrapib treatment reduced CETP activity and had directionally concordant effect on eight lipid and lipoprotein traits (total-, LDL- and HDL-cholesterol, HDL2, HDL3, apolipoproteins A-I, -B, and triglycerides), with the genetic effect on HDL-cholesterol (0.13 mmol/L; 95% CI: 0.11, 0.14) being consistent with that expected of a 10 mg dose of torcetrapib (0.13 mmol/L; 0.10, 0.15). In trials, 60mg torcetrapib elevated systolic and diastolic blood pressure by 4.47mmHg (4.10, 4.84) and 2.08mmHg (1.84, 2.31) respectively. However, the effect of CETP SNPs on systolic 0.16mmHg (−0.28, 0.60) and diastolic blood pressure −0.04mmHg (−0.36, 0.28) was null and significantly different from that expected of 10 mg torcetrapib.


Discordance in the effects of CETP SNPs and torcetrapib treatment on blood pressure despite the concordant effects on lipids indicates the hypertensive action of torcetrapib is unlikely to be due to CETP-inhibition, or shared by chemically dissimilar CETP inhibitors. Genetic studies could find use in drug development programmes as a new source of randomised evidence for drug target validation in man.

Keywords: genetics, pharmacology, epidemiology


Higher concentrations of high-density lipoprotein (HDL)-cholesterol are associated with a lower risk of coronary heart disease (CHD) independent of LDL-cholesterol1. HDL particles have anti-atherogenic actions in vitro and experimental elevation of HDL-cholesterol concentration in some animal models attenuates atheroma formation 2-3. Inhibitors of cholesteryl ester transfer protein (CETP), which mediates exchange of lipids between HDL-particles and other lipoproteins, are a new class of drugs developed for their ability to raise HDL-cholesterol. However, when the combination of a CETP-inhibitor (torcetrapib) and a statin (atorvastatin) was compared with atorvastatin alone in the ILLUMINATE trial4, the Data Safety Monitoring Board terminated the trial prematurely because of an unexpectedly higher rate of both cardiovascular and non-cardiovascular events in the torcetrapib-treated patients.

Whether the higher rate of cardiovascular events from torcetrapib treatment was a mechanism-based effect of CETP inhibition, which would be shared by other members of the same drug class, or an idiosyncratic (or off target) action of the torcetrapib molecule is uncertain. Distinguishing between the two is important because at least two other CETP-inhibitors, anacetrapib and dalcetrapib, are in advanced drug development5-6 7. Torcetrapib treatment has been associated with consistent and substantial elevations in blood pressure4, 8-10,perhaps secondary to a mineralocorticoid-like effect, that could have contributed to the increased risk of cardiovascular events 11. Although it has been proposed that the other CETP inhibitors do not share this blood pressure-elevating effect5 12, this is based on evidence from non-randomised animal experiments, and short-term dose-ranging studies in humans, both of which have limitations. Large, randomised outcome trials of anacetrapib or dalcetrapib would provide a definitive answer but could expose the trial participants to a potential hazard should the hypertensive effect be mechanism-based rather than off-target. On the other hand, the failure to further evaluate other members of this class in randomised trials runs the risk of abandoning a potentially valuable preventive therapy.

An alternative way of obtaining randomised evidence on the efficacy and safety of CETP-inhibition in humans, without the recruitment of new trial participants, prospective follow-up, or exposure to a drug, is to study the effect of carriage of common alleles of the human CETP gene associated with reduced CETP levels and activity13. Genetic association studies are a type of natural randomised trial, because maternal and paternal alleles assort at random at conception14,15. In effect, a study of alleles of the CETP gene that reduce CETP activity is akin to a very long-term randomised intervention trial of a “clean” CETP inhibitor, free from the off-target effects of individual drug molecules. We therefore compared the effect of torcetrapib and carriage of common CETP alleles on lipids and lipoproteins, blood pressure and other markers of cardiovascular risk in a large-scale international, collaborative analysis to ascertain if the increase in blood pressure seen in the clinical trials of torcetrapib was mechanism-based or off-target.


Search strategy and selection criteria

Randomised controlled trials

Randomised controlled trials evaluating the effect of torcetrapib on markers of cardiovascular risk or clinical outcomes were identified from PubMed and EMBASE up to the end of November 2007 using MeSH and free text terms “torcetrapib” or “CETP inhibitor” in combination with “randomised controlled trial”. For inclusion in the main analyses, studies had to be randomised, parallel design studies in adults that examined the effect of treatment with torcetrapib (alone or in combination) with a suitable comparator. Studies were included if published as full length articles or letters in peer reviewed journals in any language. Randomised studies were further sub-divided into shorter dose-finding studies of < 1year duration and longer clinical trials of > 1year duration and analysed separately.

Genetic studies

PubMed and EMBASE were searched up to November 2007 for studies in humans evaluating any polymorphism in the CETP gene. The search included the MeSH and free text terms “cholesteryl ester transfer protein” or “CETP” in combination with “polymorphism*”, “mutation*”, “allele*”, “gene*”, “Taq1B”, “−629C>A”, or “I405V” with no limits or restrictions. We supplemented information from published studies with unpublished genetic data obtained through a large collaborative network of investigators that allowed access to information on a wider range of traits of interest, more precise estimation of genetic effect sizes, and minimised the scope for reporting and publication bias. (For further details see Supplementary Methods)

Generation of tabular data

Two of the authors (ADH and RS) extracted data and disagreements were resolved by discussion with a third author (JPC). For randomised controlled trials, information was extracted on treatment regimen and comparator, and pre- and post-treatment measures of a wide range of cardiovascular risk markers (see supplementary methods for further details). The relationship between torcetrapib dose and effect on these variables, if available, was also recorded from dose-ranging studies. For genetic studies, study-level information was either extracted from published studies by two authors or requested from principal investigators (see supplementary methods).

Statistical analysis

Randomised clinical trials of torcetrapib

The effect of torcetrapib on different lipid fractions, blood pressure and other cardiovascular traits was assessed by calculation of the difference in the change in mean values between active and control arms. Study-specific estimates were weighted by the inverse of the variance and pooled by random effects meta-analysis to generate summary estimates.

Genetic studies

Primary analyses were based on the CETP gene variants commonly referred to as TaqI B (rs708272) and −629C>A (rs1800775), which were the most widely typed variants. The two are in linkage disequilibrium (r2 measure of association: 0.73 in European descent individuals16) (Figure S1), allowing information on the two variants to be treated jointly in a pooled analysis. Additional analyses involved the I405V variant (rs5882). For continuous outcomes, the mean difference and 95% confidence interval (CI) by genotype category was obtained from each study, and then pooled using a random effects model to obtain a summary mean difference and 95% CI. Individuals homozygous for the common TaqIB (or −629C) allele served as the reference group throughout, and this group was designated B1B1, with heterozygous individuals and individuals homozygous for either rare allele designated B1B2 and B2B2 respectively, to preserve the convention introduced in prior studies. For binary outcomes, results were expressed as an odds ratio and 95% CI. To assess the robustness of the findings, stratified analyses were conducted according to study level characteristics. In a subset of studies, predefined stratified analysis of individual level data was performed to investigate effect of CETP genotype on HDL-cholesterol by quartiles of systolic, diastolic and pulse pressure, and by LDL-cholesterol quartile, in order to gain insight on the potential for effect modification by blood pressure or cholesterol lowering medications. Deviation from Hardy Weinberg Equilibrium was assessed in each study. Heterogeneity was assessed using a chi-square test. The I2 measure17 and 95% CI was used to describe the extent of variability across studies. Additional information on the statistical analysis is provided in the supplementary methods section. All analyses were conducted using Stata 9.0.

Consistency between the CETP gene effects and equivalent torcetapib dose

To determine the consistency of the observed effect of CETP genotype on cardiovascular traits with the expected effects for a comparable dose of torcetrapib, the shape of the dose-effect relationship for torcetrapib was evaluated from dose ranging trials using the reported continuous outcomes HDL-, HDL2- and HDL3- as well as apo A1 and apo B. Despite careful searching no quantitative information on the relationship between torcetrapib dose and blood pressure from these trials was available in a form that could be utilised in the analysis. Then having confirmed a linear dose response relationship for the available variables (Figure 1a-c) we used the summary effect of a 60 mg dose of torcetrapib on HDL cholesterol (the measure with the most data) from the meta-analysis of randomised trials, and the summary effect of CETP genotype on HDL cholesterol from the meta-analysis of genetic studies, to: 1) express the effect of carriage of the B2 variant as a torcetrapib dose equivalent; and 2) to estimate the effect of this dose of torcetrapib on blood pressure and other traits. A simulation model that incorporated the variance in the effect estimates of the genotype and drug effects was used to obtain the confidence intervals (see supplementary methods for details). The observed gene effect was compared to the effect of a comparable dose of torcetrapib by means of a Z- test18. More details are provided in the supplementary methods.

Figure 1 a-c
Relationship between torcetrapib dose and HDL cholesterol and HDL2 and HDL3 sub-fractions. P values refer to the results of a meta- regression and N refers to the total number of individuals in the three dose-ranging studies contributing to this analysis ...


Randomised controlled trials of torcetrapib

Dose-response relationship of torcetrapib on HDL

Three studies (median size 40 participants; range 19 to 162), mean study duration of 5.3 (SD 3.1) weeks allowed exploration of the effect of different doses of torcetrapib on HDL-cholesterol and its sub-fractions (HDL2 and HDL3)19-21. Over the dose range studied (10 to 240 mg daily), torcetrapib produced a linear, dose-dependent increase in HDL-cholesterol (p<0.001 from meta-regression), HDL2 (p=0.03) and HDL3 (p=0.003), with no evidence of a threshold effect (Figures 1a-c).

Effect of torcetrapib on lipid profile, blood pressure and biomarkers

Four randomised trials (range 752 to 15067 participants) with mean duration 21 (SD 6) months involving 17,911 participants in aggregate, with mean age 55.4 (SD 6.9) years evaluated the effect of torcetrapib 60mg daily (in combination with atorvastatin) versus atorvastatin alone and were included in the main analysis4, 8-10. Torcetrapib 60 mg daily increased HDL-cholesterol by 0.78 mmol/L (0.68, 0.87), apolipoprotein A-I (apoA-I) by 0.30 g/L (0.30, 0.31) and total cholesterol by 0.18mmol/L (0.10, 0.25). The same dose reduced LDL-cholesterol by 0.54 mmol/L (−0.64, −0.43), triglycerides by 0.12 mmol/L (−0.18, −0.07) and apolipoprotein B (apoB) by 0.11 g/L (−0.11, −0.10) (Table 1 and Figure S2a). A pooled analysis of all 17,911 participants from four trials indicated that torcetrapib 60 mg daily led to a mean increase in systolic blood pressure of 4.47 mmHg (4.10, 4.84) and an increase in diastolic blood pressure of 2.08 mmHg (1.84, 2.31). In the ILLUMINATE trial the elevation in blood pressure was accompanied by a decrease in plasma potassium, an increase in sodium and an increase in aldosterone concentration4 (Table 2,). In three trials4, 9-10 including 17,159 participants there was no effect of torcetrapib on C- reactive protein (CRP) concentration (Table S1).

Table 1
Effect of torcetrapib (60mg) and CETP genotype on lipids and lipoproteins. Differences between continuous traits are for values reported at the end of the randomised trials unless otherwise indicated.
Table 2
Effect of torcetrapib (60mg) and CETP genotype on blood pressure, and circulating and urinary electrolytes and creatinine. Differences between continuous traits are at end of the randomised trials unless otherwise indicated

Genetic studies

Study details and CETP polymorphisms evaluated

A total of 31 studies (supplementary references 1-39) and 67,687 individuals of mean age 55.8 (SD 9.6) years contributed information on at least one continuous outcome. Twenty-three studies with 60,316 individuals provided previously unpublished data. Of the unpublished studies, 21 studies (50,908 individuals) provided data on the rs708272 (Taq1B) polymorphism, two studies (8,535 participants) provided data only on the rs1800775 (−629C>A) polymorphism. Where studies provided data on both −629C>A and Taq1B, the latter was used for the primary analysis. Seven studies (21,353 individuals) also provided data on the rs5882 (I405V) polymorphism (S8,S10,S15,S16,18-S20,S22,S25,S32,S33) and these results are provided in the supplementary section. Study details are provided in Table S2 and Table S3 respectively.

Effect of CETP genotypes on CETP concentration, CETP activity and lipids

Six studies in individuals of European ancestry (5,340 participants) provided information on the effect of CETP genotype on CETP concentration (S7,S8,S15,S28,S30,S31), and two studies (858 participants) (S15,S18-20) on the effect on CETP activity. A further five studies (1,867 participants) contributed data from individuals of Japanese origin (Figure S3, references S34-S39). A graded effect of genotype on CETP concentration and activity was evident in both populations. People of European ancestry homozygous for the B2 allele had lower CETP concentration (−0.47 μg/ml; −0.67, −0.26) and lower CETP activity (−17.00 nmol/ml/hr; −18.52, −15.49) compared to people homozygous for the B1 allele (Figure S3a and S3b). In 31 studies with 67,687 participants, B2 homozygous individuals had higher concentrations of HDL-cholesterol (0.13 mmol/L; 0.11, 0.14) (Figure 2). The link between genotype and HDL-cholesterol was consistent in analyses stratified by study size, gender, presence of CHD, ancestry, and across quartiles of LDL-cholesterol, systolic and diastolic blood pressure and pulse pressure (Figure S3c, Figure S4). In addition, B2 homozygous individuals exhibited higher concentrations of total cholesterol (0.05 mmol/L; 0.03, 0.07) and apo A-I (0.06 g/L; 0.05, 0.08) and lower concentrations of LDL-cholesterol (−0.03 mmol/L; −0.05, −0.01) and triglycerides (−0.06 mmol/L; −0.10, −0.02), and apo B (0.02 g/L; −0.03,−0.01). In two studies, individuals homozygous for the B2 allele had higher circulating concentrations of both the larger HDL2 particles (0.03 mmol/L; 0.01, 0.04) and smaller HDL3 particles (0.06 mmol/L; 0.02, 0.11) (Table 1). Heterozygous subjects exhibited lipid and lipoprotein concentrations approximately intermediate between those found in homozygous subjects, consistent with an additive effect of each copy of the variant allele (Table 1 and per-allele data available on request). The effect of variant CETP alleles on lipid and lipoprotein profile thus reproduced the direction of effect of treatment with torcetrapib in clinical trials for eight separate lipid and lipoprotein traits (Table 1, Figures Figures3a,3a, S2a, S2b).Using a simulation model and assuming a linear dose-response relationship (Figure 1) we estimated that the effect on HDL in B2 homozygous individuals corresponds to a dose of torcetrapib of 9.7 mg (8.18, 11.41), and for heterozygous individuals to a dose of 4.5 mg (3.71, 5.38), i.e. to a torcetrapib dose of approximately 10 mg and 5mg respectively (Figure3b).

Figure 2
Effect of CETP genotype on HDL Cholesterol in individuals of European ancestry. The B1B1 genotype is used as the reference group
Figure 3
Figure 3a The effect of torcetrapib and CETP gene variants on six lipid traits evaluated in both genetic studies and randomised trials

Effect of CETP genotypes on blood pressure and electrolytes

Twenty-two studies (58,948 individuals) provided information on CETP genotypes and systolic and diastolic blood pressure, including previously unpublished information from 20 studies (54,936 individuals). CETP genotype had no effect on systolic and diastolic blood pressure; the mean differences in comparisons between homozygous subjects were 0.16 mmHg (−0.28, 0.60) and −0.04 mmHg (−0.36, 0.28) for systolic and diastolic blood pressure respectively. Mean differences in systolic and diastolic blood pressure between heterozygous individuals (B1B2) and those homozygous for the B1 allele were −0.27 mmHg (−0.64, 0.10) and −0.23 mmHg (−0.43, −0.04) respectively (Figure 4a). The null findings were again consistent in analyses stratified by study size, gender, presence of pre-existing CHD, ancestral origin, and allele typed (Figure 4a, 4b, S5a, S5b). The expected effect on blood pressure of a 10 mg daily dose of torcetrapib was estimated to be 0.72 mmHg (0.60, 0.87) and 0.33 mmHg (0.27, 0.41) for systolic and diastolic blood pressure respectively, assuming a linear relationship between torcetrapib dose and blood pressure, and this was significantly different from the observed genetic effect on blood pressure (Figure 5a-b). Unlike torcetrapib treatment, CETP genotype was not associated with serum sodium, potassium or creatinine concentration, nor with urinary sodium or potassium concentration (Table 2, Figure 5c-d). Individuals with variant CETP alleles were also no more likely to receive anti-hypertensive medications (OR 0.98; 0.80, 1.21) (Table S1).

Figure 4
Effect of CETP genotype on (a) systolic and (b) diastolic blood pressure in populations of European descent. Weighted mean difference is given, with the B1B1 genotype used as the reference genotype.
Figure 5 a-d
Observed effect of the CETP gene and expected effects of a 5 and 10 mg dose of torcetrapib on (a) systolic and (b) diastolic blood pressure, and (c) serum potassium and (d) sodium levels.

Effect of CETP genotypes on variables unrelated to CETP inhibition

There was no link between CETP genotypes and variables unrelated to CETP function including age, body-mass index, or smoking habit (Table S1). There was also no consistent association with blood glucose or with CRP concentration consistent with data from clinical trials of torcetrapib (Table S1).


Main findings and interpretation

We found concordance in the effect of common variants in the CETP gene and pharmacological inhibition of CETP by torcetrapib on eight continuous lipid and lipoprotein markers evaluated in both randomised trials and genetic studies (HDL-C, HDL2, HDL3, LDL-C, triglycerides, total cholesterol, apo A-I, apo B). The only continuous traits for which the effect of genotype and drug were consistently discordant were systolic and diastolic blood pressure, and the electrolytes sodium and potassium. This large-scale randomised evidence in humans supports the interpretation that the blood pressure elevating effect of torcetrapib (and the connected effect on electrolytes) is mechanistically unrelated to CETP inhibition. The findings have important implications specifically for the development of other CETP inhibitors and more generally for the potential use of genetic variants to inform drug development.

Other sources of evidence on the same question

Our interpretation that the hypertensive effect of torcetrapib is off target receives additional support from other lines of evidence. First, treatment with the CETP inhibitors anacetrapib and dalcetrapib has not been associated with blood pressure elevation, though the studies thus far have been relatively small in size and of short duration5, 7. Second, torcetrapib (but not anacetrapib) has been reported to cause a blood pressure increase in several animal models12 a species which do not express CETP. Third, a recent study 22 indicated that torcetrapib treatment elevates aldosterone concentration, with corresponding effects on sodium and potassium concentration, and these electrolyte changes were not observed in a short-term dose-ranging study of anacetrapib7. These findings, from the separate lines of investigation, each with differing limitations and sources of error, provide reassurance that the hypertensive effect of torcetrapib is off-target and therefore unlikely to be shared by other CETP-inhibitors.

CETP inhibition and prevention of CHD

The higher blood pressure among individuals in the torcetrapib arm of the ILLUMINATE trial might explain the higher rate of cardiovascular events, but there may also be other explanations. CETP inhibition might interfere with reverse cholesterol transport and generate an HDL particle of abnormal size and function23, a mechanism-based adverse effect. Prior small mechanistic studies have suggested torcetrapib treatment increased the concentration of both large HDL2 and small HDL3 particles, but that the effect on HDL2 was proportionately greater. However this differential effect was only seen at a dose of torcetrapib four times as large as the dose used in the large scale clinical trials 19. Genetic data on the effect of CETP genotype on HDL subtype were limited, but in our analysis there was no clear evidence of a differential effect of CETP genotype on HDL subclasses. Although we have focussed here on the effect of CETP genotypes on lipid, lipoproteins and blood pressure, to make direct comparison of the effect of pharmacological CETP inhibition and carriage of CETP alleles, a recent meta-analysis of studies including 27 196 coronary cases and 55 338 controls, and a genome wide analysis from the Women's Genome Health Study both provide support for the CETP variants studied here being protective from CHD events 24-25.Although this protective effect has not been consistent across all studies 26, there has been no consistent signal for an increase in CHD risk from carriage of these alleles.

Potential limitations

Though the findings are robust, our interpretation requires consideration in the light of certain theoretical and practical limitations of the genetic approach we have used. CETP alleles are of much smaller effect than the most widely studied dose of torcetrapib, so it might be argued that the failure to detect an association between genotype and a continuous marker such as blood pressure could have arisen because of inadequate power, or if the effect on blood pressure requires a supra-threshold degree of CETP inhibition. We attempted to maximise power and minimise the potential for a type II error by establishing a large genetic collaboration that included a substantial amount of previously unpublished information. Blood pressure was an outcome which had been widely recorded in the studies included in this analysis (22 studies and 59,948 individuals) but was not widely reported, and so the findings should not be prone to bias. Although the investigation of the effect of CETP polymorphism on BP was not the primary aim of any of the studies included here, BP measurement was carried out using validated devices and widely accepted methods. The study was also sufficiently powered to detect a BP signal of the size expected of a 5-10 mg dose of torcetrapib (see supplementary methods). Indeed 3 of these studies (14,109 individuals) contributed to the recent whole genome analysis of BP loci which identified SNPs that altered BP by around 1/0.5 mmHg, close to the effect size being sought in the current analysis27-28. With the available sample size, we also detected an effect of CETP genotype on triglycerides which was similar in size to that which would have been expected for blood pressure were this effect mechanism-based (Figure S2a). We also triangulated the findings from RCTs with the genetic data (i.e. we compared the expected effect of a 5 and 10mg dose of torcetrapib with the observed genetic effect) rather than focusing solely on statistical tests in the genetic associations. Taken together these analyses suggest that the null findings in relation to blood pressure are neither biased nor explained by inadequate sample size. Although we were unable to exclude a hypothetical non-linear (threshold) relationship between CETP-inhibition by torcetrapib and blood pressure because none of the dose ranging studies of torcetrapib reported quantitative data on the dose-response effect in a form that could be extracted for analysis, the effect of torcetrapib on all lipid and lipoprotein traits evaluated was linear over the dose range studied. We therefore made the assumption that this was also true for blood pressure.

The randomised allocation of alleles in genetic studies differs from the randomised drug intervention in a clinical trial in that assignment of genotype occurs at conception producing an effect across a lifetime, rather than in mid- to late-adulthood, when most randomised controlled trials are conducted. It is conceivable, therefore, that an adverse effect of a common genetic variant on blood pressure from early life may have led to developmental compensation by other systems15. If this were the case, a null association of CETP genotype with blood pressure seen in genetic studies might lead to unreliable inference on the likely effect of modification of CETP activity by a drug. However, there was no evidence that such developmental compensation operated in the case of any of the eight lipid traits we studied for which both the lifelong effect of the genetic exposure and shorter term effect of the drug were consistent.

Although the precise functional alleles at the CETP locus have yet to be identified with certainty, the −629C>A (rs1800775) and I405V (rs5885) alleles are either likely to be functional themselves or be in sufficiently strong linkage disequilibrium (LD) with functional variant(s) as to be valid tools for this type of analysis. The −629C>A variant has been shown to alter binding of Sp transcription factors29. The Taq1B allele (rs708272) is intronic and less likely to be functional itself, but is in strong LD with several promoter polymorphisms (including the −629C>A variant) and as our analyses show, exhibits very strong association with multiple lipid traits. It is important to be clear, however, that for the analyses we have conducted it is not necessary for functional alleles to have been precisely delineated so long as an effect of the alleles studied on the traits of interest can be robustly demonstrated30. Although there are likely to be other variants in and around the CETP gene that are also associated with CETP activity and lipids, some because they are causal and some because they are simply associated with causal SNPs by linkage disequilibrium, the use of a single SNP in this region does not compromise the analysis provided it can be demonstrated that it provides a reliable index of CETP activity and differences in the lipid traits of interest (which we have demonstrated), and that, on the assumption that the SNP is in linkage disequilibrium (LD) with a causal SNP rather than causal itself, that the main analyses are grouped according to subjects of similar ancestry to ensure that the LD relationships are consistent across studies. Moreover, SNPs at the CETP locus, including rs1800775 (−629C>A) and rs708272 (Taq1B) studied here, have emerged as among the strongest associated signals with HDL cholesterol in recent genome wide association studies25, 31-33 (Figure S1).

Wider implications of this work

We utilised the principle that allelic variants in a gene encoding a specific drug target can be used to model the mechanism-based effect of modifying the same target pharmacologically. In the current analysis this was applied to help distinguish the mechanism-based from off-target actions of a drug molecule in advanced development. However, further research should now address whether this principle could be exploited at earlier phases in the drug development pathway to help, for example, with the validation of a promising new target, or to assemble a panel of biomarkers of efficacy to test in clinical trials. The directional concordance of the effect of HMGCR SNPs in genetic studies and HMG-coA reductase (statin) treatment on LDL-cholesterol and CHD risk in clinical trials, lends additional support to the potential utility of this approach. There is likely to be wide availability of genetic tools for this purpose, because the majority of drug targets are proteins, and regulatory genetic variants acting in cis, located within a 100kb of genes, appear to be a common feature of the human genome34.


In summary, a novel large-scale genetic approach has provided evidence that the hypertensive effect of torcetrapib is likely an off-target action. This provides reassurance that this particular adverse effect of torcetrapib is unlikely to be shared by other chemically-dissimilar CETP-inhibitors, but further drug development will be required to assess if these other agents, and the CETP inhibitor class of drugs in general, are likely to be efficacious in the prevention of CHD events with an acceptable risk-benefit profile. Further research should investigate whether genetic studies could find use in drug development programmes as a new source of randomised evidence for drug target validation in man.

Short Commentary: Clinical Perspective

The inverse relationship between HDL-cholesterol and risk of coronary heart disease (CHD) suggests that therapeutic elevation of HDL-cholesterol may provide an effective means of CHD prevention. Pharmacological inhibition of cholesteryl ester transfer protein (CETP) leads to elevation in HDL-cholesterol, but torcetrapib (the first-in-class CETP inhibitor) increased the risk of cardiovascular events in the ILLUMINATE trial, which may have resulted from an unexpected blood pressure elevating effect of this agent. We utilised common genetic polymorphisms in the CETP gene to distinguish whether the hypertensive action of torcetrapib was mechanism-based or off-target, because a genetic study of these variants can be considered to be a type of natural randomised trial of a ‘clean’ low dose CETP-inhibitor with no off-target actions. Common CETP gene polymorphisms and torcetrapib treatment had concordant effects on eight lipid and lipoprotein markers including HDL-cholesterol, but CETP gene variants had no effect on blood pressure. The blood-pressure elevating effect of torcetrapib appears to be an off-target action that is unlikely to be shared by chemically-dissimilar CETP-inhibitors. Genetic studies could find use in drug development programmes as a new source of randomised evidence for drug target validation in man.

Supplementary Material

Supplementary Methods, Results, Tables, Figs, Refs


Funding Source

Reecha Sofat is supported by a British Heart Foundation (Schillingford) Clinical Training Fellowship (FS/07/011). Aroon Hingorani holds a British Heart Foundation Senior Fellowship (FS 05/125). Liam Smeeth holds a Wellcome Trust Senior Research Fellowship. Steve Humphries holds a British Heart Foundation Chair in Cardiovascular Genetics he and Philippa Talmud are supported by British Heart Foundation PG04/110/17827. Michael Marmot is supported by a MRC Research Professorship. Debbie Lawlor is part funded by a UK Department of Heath Career Scientist Award. John Deanfield holds the British Heart Foundation Vandervell Chair in Cardiology of the Young.

The Whitehall II study has been supported by grants from the Medical Research Council; Economic and Social Research Council; British Heart Foundation; Health and Safety Executive; Department of Health; National Heart Lung and Blood Institute (HL36310), US, NIH: National Institute on Aging (AG13196), US, NIH; Agency for Health Care Policy Research (HS06516); and the John D and Catherine T MacArthur Foundation Research Networks on Successful Midlife Development and Socio-economic Status and Health. Samples from the English Longitudinal Study of Ageing (ELSA) DNA Repository (EDNAR), received support under a grant (AG1764406S1) awarded by the National Institute on Aging (NIA). ELSA was developed by a team of researchers based at the National Centre for Social Research, University College London and the Institute of Fiscal Studies. The data were collected by the National Centre for Social Research. The developers and funders of ELSA and the Archive do not bear any responsibility for the analyses or interpretations presented here. EPIC-Norfolk is supported by the Medical Research Council and Cancer Research UK. The British Women's Heart and Health Study is funded by the Department of Health Policy Research Programme and the British Heart Foundation. The UK Medical Research Council provide support for the MRC Centre where DA Lawlor & G Davey Smith work. The views expressed in the publication are those of the authors and not necessarily those of any funding bodies.



ADH is a member of the editorial board of Drug and Therapeutics Bulletin, has provided non-remunerated advice to GSK and London Genetics and has received honoraria for speaking at educational meetings on cardiovascular risk which have been donated in whole or part to charity. Marcello Arca was on the Pfizer advisory board for Torcetrapib.


1. Lewington S, Whitlock G, Clarke R, Sherliker P, Emberson J, Halsey J, Qizilbash N, Peto R, Collins R. Blood cholesterol and vascular mortality by age, sex, and blood pressure: a meta-analysis of individual data from 61 prospective studies with 55,000 vascular deaths. Lancet. 2007;370(9602):1829–1839. [PubMed]
2. Sugano M, Makino N, Sawada S, Otsuka S, Watanabe M, Okamoto H, Kamada M, Mizushima A. Effect of antisense oligonucleotides against cholesteryl ester transfer protein on the development of atherosclerosis in cholesterol-fed rabbits. J Biol Chem. 1998;273(9):5033–5036. [PubMed]
3. Whitlock ME, Swenson TL, Ramakrishnan R, Leonard MT, Marcel YL, Milne RW, Tall AR. Monoclonal antibody inhibition of cholesteryl ester transfer protein activity in the rabbit. Effects on lipoprotein composition and high density lipoprotein cholesteryl ester metabolism. J Clin Invest. 1989;84(1):129–137. [PMC free article] [PubMed]
4. Barter PJ, Caulfield M, Eriksson M, Grundy SM, Kastelein JJ, Komajda M, Lopez-Sendon J, Mosca L, Tardif JC, Waters DD, Shear CL, Revkin JH, Buhr KA, Fisher MR, Tall AR, Brewer B. Effects of torcetrapib in patients at high risk for coronary events. N Engl J Med. 2007;357(21):2109–2122. [PubMed]
5. Krishna R, Anderson MS, Bergman AJ, Jin B, Fallon M, Cote J, Rosko K, Chavez-Eng C, Lutz R, Bloomfield DM, Gutierrez M, Doherty J, Bieberdorf F, Chodakewitz J, Gottesdiener KM, Wagner JA. Effect of the cholesteryl ester transfer protein inhibitor, anacetrapib, on lipoproteins in patients with dyslipidaemia and on 24-h ambulatory blood pressure in healthy individuals: two double-blind, randomised placebo-controlled phase I studies. Lancet. 2007;370(9603):1907–1914. [PubMed]
6. Kuivenhoven JA, de Grooth GJ, Kawamura H, Klerkx AH, Wilhelm F, Trip MD, Kastelein JJ. Effectiveness of inhibition of cholesteryl ester transfer protein by JTT-705 in combination with pravastatin in type II dyslipidemia. Am J Cardiol. 2005;95(9):1085–1088. [PubMed]
7. Bloomfield D, Carlson GL, Sapre A, Tribble D, McKenney JM, Littlejohn TW, 3rd, Sisk CM, Mitchel Y, Pasternak RC. Efficacy and safety of the cholesteryl ester transfer protein inhibitor anacetrapib as monotherapy and coadministered with atorvastatin in dyslipidemic patients. Am Heart J. 2009;157(2):352–360. e352. [PubMed]
8. Bots ML, Visseren FL, Evans GW, Riley WA, Revkin JH, Tegeler CH, Shear CL, Duggan WT, Vicari RM, Grobbee DE, Kastelein JJ. Torcetrapib and carotid intima-media thickness in mixed dyslipidaemia (RADIANCE 2 study): a randomised, double-blind trial. Lancet. 2007;370(9582):153–160. [PubMed]
9. Kastelein JJ, van Leuven SI, Burgess L, Evans GW, Kuivenhoven JA, Barter PJ, Revkin JH, Grobbee DE, Riley WA, Shear CL, Duggan WT, Bots ML. Effect of torcetrapib on carotid atherosclerosis in familial hypercholesterolemia. N Engl J Med. 2007;356(16):1620–1630. [PubMed]
10. Nissen SE, Tardif JC, Nicholls SJ, Revkin JH, Shear CL, Duggan WT, Ruzyllo W, Bachinsky WB, Lasala GP, Tuzcu EM. Effect of torcetrapib on the progression of coronary atherosclerosis. N Engl J Med. 2007;356(13):1304–1316. [PubMed]
11. Neal B, MacMahon S. An overview of 37 randomised trials of blood pressure lowering agents among 270,000 individuals. World Health Organization-International Society of Hypertension Blood Pressure Lowering Treatment Trialists' Collaboration. Clin Exp Hypertens. 1999;21(5-6):517–529. [PubMed]
12. Forrest MJ, Bloomfield D, Briscoe RJ, Brown PN, Cumiskey AM, Ehrhart J, Hershey JC, Keller WJ, Ma X, McPherson HE, Messina E, Peterson LB, Sharif-Rodriguez W, Siegl PK, Sinclair PJ, Sparrow CP, Stevenson AS, Sun SY, Tsai C, Vargas H, Walker M, 3rd, West SH, White V, Woltmann RF. Torcetrapib-induced blood pressure elevation is independent of CETP inhibition and is accompanied by increased circulating levels of aldosterone. Br J Pharmacol. 2008;154(7):1465–1473. [PMC free article] [PubMed]
13. Boekholdt SM, Thompson JF. Natural genetic variation as a tool in understanding the role of CETP in lipid levels and disease. J Lipid Res. 2003;44(6):1080–1093. [PubMed]
14. Hingorani A, Humphries S. Nature's randomised trials. Lancet. 2005;366(9501):1906–1908. [PubMed]
15. Davey Smith G, Ebrahim S. ‘Mendelian randomization’: can genetic epidemiology contribute to understanding environmental determinants of disease? Int J Epidemiol. 2003;32(1):1–22. [PubMed]
16. Thompson JF, Wood LS, Pickering EH, Dechairo B, Hyde CL. High-density genotyping and functional SNP localization in the CETP gene. J Lipid Res. 2007;48(2):434–443. [PubMed]
17. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. Bmj. 2003;327(7414):557–560. [PMC free article] [PubMed]
18. Bautista LE, Smeeth L, Hingorani AD, Casas JP. Estimation of bias in nongenetic observational studies using "mendelian triangulation". Ann Epidemiol. 2006;16(9):675–680. [PubMed]
19. Brousseau ME, Schaefer EJ, Wolfe ML, Bloedon LT, Digenio AG, Clark RW, Mancuso JP, Rader DJ. Effects of an inhibitor of cholesteryl ester transfer protein on HDL cholesterol. N Engl J Med. 2004;350(15):1505–1515. [PubMed]
20. Davidson MH, McKenney JM, Shear CL, Revkin JH. Efficacy and safety of torcetrapib, a novel cholesteryl ester transfer protein inhibitor, in individuals with below-average high-density lipoprotein cholesterol levels. J Am Coll Cardiol. 2006;48(9):1774–1781. [PubMed]
21. Clark RW, Sutfin TA, Ruggeri RB, Willauer AT, Sugarman ED, Magnus-Aryitey G, Cosgrove PG, Sand TM, Wester RT, Williams JA, Perlman ME, Bamberger MJ. Raising high-density lipoprotein in humans through inhibition of cholesteryl ester transfer protein: an initial multidose study of torcetrapib. Arterioscler Thromb Vasc Biol. 2004;24(3):490–497. [PubMed]
22. Vergeer M, Bots ML, van Leuven SI, Basart DC, Sijbrands EJ, Evans GW, Grobbee DE, Visseren FL, Stalenhoef AF, Stroes ES, Kastelein JJ. Cholesteryl ester transfer protein inhibitor torcetrapib and off-target toxicity: a pooled analysis of the rating atherosclerotic disease change by imaging with a new CETP inhibitor (RADIANCE) trials. Circulation. 2008;118(24):2515–2522. [PubMed]
23. Tall AR. CETP Inhibitors to Increase HDL Cholesterol Levels. N Engl J Med. 2007;356(13):1364–1366. [PubMed]
24. Thompson A, Di Angelantonio E, Sarwar N, Erqou S, Saleheen D, Dullaart RP, Keavney B, Ye Z, Danesh J. Association of cholesteryl ester transfer protein genotypes with CETP mass and activity, lipid levels, and coronary risk. Jama. 2008;299(23):2777–2788. [PubMed]
25. Ridker PM, Pare G, Parker AN, Zee RYL, Miletich JP, Chasman DI. Polymorphism in the CETP Gene Region, HDL Cholesterol, and Risk of Future Myocardial Infarction: Genomewide Analysis Among 18 245 Initially Healthy Women From the Women's Genome Health Study. Circ Cardiovasc Genet. 2009;2(1):26–33. [PMC free article] [PubMed]
26. Anand SS, Xie C, Pare G, Montpetit A, Rangarajan S, McQueen MJ, Cordell HJ, Keavney B, Yusuf S, Hudson TJ, Engert JC. Genetic Variants Associated With Myocardial Infarction Risk Factors in Over 8000 Individuals From Five Ethnic Groups: The INTERHEART Genetics Study. Circ Cardiovasc Genet. 2009;2(1):16–25. on Behalf of the II. [PubMed]
27. Levy D, Ehret GB, Rice K, Verwoert GC, Launer LJ, Dehghan A, Glazer NL, Morrison AC, Johnson AD, Aspelund T, Aulchenko Y, Lumley T, Kottgen A, Vasan RS, Rivadeneira F, Eiriksdottir G, Guo X, Arking DE, Mitchell GF, Mattace-Raso FU, Smith AV, Taylor K, Scharpf RB, Hwang SJ, Sijbrands EJ, Bis J, Harris TB, Ganesh SK, O'Donnell CJ, Hofman A, Rotter JI, Coresh J, Benjamin EJ, Uitterlinden AG, Heiss G, Fox CS, Witteman JC, Boerwinkle E, Wang TJ, Gudnason V, Larson MG, Chakravarti A, Psaty BM, van Duijn CM. Genome-wide association study of blood pressure and hypertension. Nat Genet. 2009 [PMC free article] [PubMed]
28. Newton-Cheh C, Johnson T, Gateva V, Tobin MD, Bochud M, Coin L, Najjar SS, Zhao JH, Heath SC, Eyheramendy S, Papadakis K, Voight BF, Scott LJ, Zhang F, Farrall M, Tanaka T, Wallace C, Chambers JC, Khaw KT, Nilsson P, van der Harst P, Polidoro S, Grobbee DE, Onland-Moret NC, Bots ML, Wain LV, Elliott KS, Teumer A, Luan J, Lucas G, Kuusisto J, Burton PR, Hadley D, McArdle WL, Brown M, Dominiczak A, Newhouse SJ, Samani NJ, Webster J, Zeggini E, Beckmann JS, Bergmann S, Lim N, Song K, Vollenweider P, Waeber G, Waterworth DM, Yuan X, Groop L, Orho-Melander M, Allione A, Di Gregorio A, Guarrera S, Panico S, Ricceri F, Romanazzi V, Sacerdote C, Vineis P, Barroso I, Sandhu MS, Luben RN, Crawford GJ, Jousilahti P, Perola M, Boehnke M, Bonnycastle LL, Collins FS, Jackson AU, Mohlke KL, Stringham HM, Valle TT, Willer CJ, Bergman RN, Morken MA, Doring A, Gieger C, Illig T, Meitinger T, Org E, Pfeufer A, Wichmann HE, Kathiresan S, Marrugat J, O'Donnell CJ, Schwartz SM, Siscovick DS, Subirana I, Freimer NB, Hartikainen AL, McCarthy MI, O'Reilly PF, Peltonen L, Pouta A, de Jong PE, Snieder H, van Gilst WH, Clarke R, Goel A, Hamsten A, Peden JF, Seedorf U, Syvanen AC, Tognoni G, Lakatta EG, Sanna S, Scheet P, Schlessinger D, Scuteri A, Dorr M, Ernst F, Felix SB, Homuth G, Lorbeer R, Reffelmann T, Rettig R, Volker U, Galan P, Gut IG, Hercberg S, Lathrop GM, Zelenika D, Deloukas P, Soranzo N, Williams FM, Zhai G, Salomaa V, Laakso M, Elosua R, Forouhi NG, Volzke H, Uiterwaal CS, van der Schouw YT, Numans ME, Matullo G, Navis G, Berglund G, Bingham SA, Kooner JS, Connell JM, Bandinelli S, Ferrucci L, Watkins H, Spector TD, Tuomilehto J, Altshuler D, Strachan DP, Laan M, Meneton P, Wareham NJ, Uda M, Jarvelin MR, Mooser V, Melander O, Loos RJ, Elliott P, Abecasis GR, Caulfield M, Munroe PB. Genome-wide association study identifies eight loci associated with blood pressure. Nat Genet. 2009 [PMC free article] [PubMed]
29. Dachet C, Poirier O, Cambien F, Chapman J, Rouis M. New functionalpromoter polymorphism, CETP/-629, in cholesteryl ester transfer protein (CETP) gene related to CETP mass and high density lipoprotein cholesterol levels: role of Sp1/Sp3 in transcriptional regulation. Arterioscler Thromb Vasc Biol. 2000;20(2):507–515. [PubMed]
30. Lawlor DA, Harbord RM, Sterne JA, Timpson N, Davey Smith G. Mendelian randomization: Using genes as instruments for making causal inferences in epidemiology. Stat Med. 2008;27(8):1133–1163. [PubMed]
31. Kathiresan S, Melander O, Guiducci C, Surti A, Burtt NP, Rieder MJ, Cooper GM, Roos C, Voight BF, Havulinna AS, Wahlstrand B, Hedner T, Corella D, Tai ES, Ordovas JM, Berglund G, Vartiainen E, Jousilahti P, Hedblad B, Taskinen MR, Newton-Cheh C, Salomaa V, Peltonen L, Groop L, Altshuler DM, Orho-Melander M. Six new loci associated with blood low-density lipoprotein cholesterol, high-density lipoprotein cholesterol or triglycerides in humans. Nat Genet. 2008;40(2):189–197. [PMC free article] [PubMed]
32. Kooner JS, Chambers JC, Aguilar-Salinas CA, Hinds DA, Hyde CL, Warnes GR, Gomez Perez FJ, Frazer KA, Elliott P, Scott J, Milos PM, Cox DR, Thompson JF. Genome-wide scan identifies variation in MLXIPL associated with plasma triglycerides. Nat Genet. 2008;40(2):149–151. [PubMed]
33. Willer CJ, Sanna S, Jackson AU, Scuteri A, Bonnycastle LL, Clarke R, Heath SC, Timpson NJ, Najjar SS, Stringham HM, Strait J, Duren WL, Maschio A, Busonero F, Mulas A, Albai G, Swift AJ, Morken MA, Narisu N, Bennett D, Parish S, Shen H, Galan P, Meneton P, Hercberg S, Zelenika D, Chen WM, Li Y, Scott LJ, Scheet PA, Sundvall J, Watanabe RM, Nagaraja R, Ebrahim S, Lawlor DA, Ben-Shlomo Y, Davey-Smith G, Shuldiner AR, Collins R, Bergman RN, Uda M, Tuomilehto J, Cao A, Collins FS, Lakatta E, Lathrop GM, Boehnke M, Schlessinger D, Mohlke KL, Abecasis GR. Newly identified loci that influence lipid concentrations and risk of coronary artery disease. Nat Genet. 2008;40(2):161–169. [PubMed]
34. Melzer D, Perry JR, Hernandez D, Corsi AM, Stevens K, Rafferty I, Lauretani F, Murray A, Gibbs JR, Paolisso G, Rafiq S, Simon-Sanchez J, Lango H, Scholz S, Weedon MN, Arepalli S, Rice N, Washecka N, Hurst A, Britton A, Henley W, van de Leemput J, Li R, Newman AB, Tranah G, Harris T, Panicker V, Dayan C, Bennett A, McCarthy MI, Ruokonen A, Jarvelin MR, Guralnik J, Bandinelli S, Frayling TM, Singleton A, Ferrucci L. A genome-wide association study identifies protein quantitative trait loci (pQTLs) PLoS Genet. 2008;4(5):e1000072. [PMC free article] [PubMed]