Clinical Utility of Bile Acid Sequestrants in the Treatment of Dyslipidemia: A Scientific Review

Bile acid sequestrants (BAS) continue to command a position in the treatment of dyslipidemias 25 years after their introduction. Partial diversion of the enterohepatic circulation using BAS depletes the endogenous bile acid pool by approximately 40%, thus stimulating an increase in bile acid synthesis from cholesterol, which lowers low-density lipoprotein cholesterol (LDL-C) by 15 to 26%. Three BAS are currently used for treating hypercholesterolemia in the United States: the conventional sequestrants, cholestyramine and colestipol, and the specifically engineered BAS, colesevelam hydrochloride (HCl). Compared with conventional BAS, colesevelam HCl has enhanced specificity, greater affinity, and higher capacity for binding bile acids, due to its polymer structure engineered for bile acid sequestration. BAS are not absorbed by the intestine and thus have no systemic drug-drug interactions, but may interfere with the absorption of some drugs. Although BAS monotherapy effectively lowers LDL-C, combination therapy, especially with BAS and statins, is becoming increasingly common due to complementary mechanisms of action. Low-dose statin plus BAS combinations lead to greater or similar LDL-C reductions compared with high-dose statin monotherapy and may have a better safety profile. Combinations of BAS with nonstatin lipid-lowering agents, including niacin, fibrates, and cholesterol absorption inhibitors, may be useful in those patients who require intensive lipid-lowering, but are statin intolerant. BAS treatment can significantly reduce coronary artery disease (CAD) progression and the risk of CAD-associated outcomes. It is also becoming clear that BAS and other therapies that manipulate the bile acid synthetic pathway may have clinically useful therapeutic effects on other metabolic disorders including type 2 diabetes.

Key Points

* Partial diversion of the enterohepatic circulation using BAS depletes the endogenous bile acid pool by approximately 40%, stimulating an increase in bile acid synthesis from cholesterol, and consequently lowering serum LDL-C by 15% to 25%.

* Low-dose combination therapy with BAS and statins is becoming more common due to these agents' complementary mechanisms of action, and leads to greater or similar LDL-C reductions compared with high-dose statin monotherapy, perhaps with a better safety profile.

* BAS therapy, alone or in combination with other lipid-modifying therapies, can significantly reduce coronary artery disease progression and the risk of associated outcomes.

* Compared with the older BAS, cholestyramine and colestipol, the newer specifically engineered BAS, colesevelam HCl, has enhanced specificity, greater affinity, and higher capacity for binding bile acids, due to its unique polymer structure.

* BAS are not absorbed by the intestine, and thus have no systemic metabolism or drug-drug interactions, but may interfere with the absorption of some drugs.

Purpose and Rationale

The purpose of this review is to summarize how 25 years of controlled clinical trials of bile acid sequestrants (BAS) have established this drug class to safely and effectively treat hypercholesterolemia as part of both monotherapy and combination therapy, to reduce the risk of coronary artery disease (CAD), and to slow progression and induce regression of atherosclerosis. As the only nonsystemic lipid-lowering agents, BAS also have a demonstrated safety record. In addition a newer, specifically engineered BAS may have greater potency, limited binding of comedications due to selective absorption, and lower incidence of gastrointestinal side effects compared with older BAS. The practicing clinician should find that the data presented here support the use of BAS in patients with hypercholesterolemia.

Cholesterol and CAD

Maintaining cholesterol homeostasis is critical because it plays a role in the synthesis of biologic membranes, steroid hormones, vitamin D, and bile acids (BA). The cholesterol pool in a normal human is approximately 100 to 140 g; of this, about 10% is found in the circulation, with the remainder located predominantly in cell membranes.^1,2 Cholesterol balance is maintained by equilibrium between acquisition from de novo synthesis and dietary intake and loss by BA synthesis and fecal excretion. Preserving this balance avoids cholesterol accumulation in artery walls and the subsequent development of atherosclerosis.

Low-density lipoprotein cholesterol (LDL-C) is the primary target of lipid-lowering therapy, as the link between LDL-C and CAD is well established. The target LDL-C goals are dependent upon a patient's level of CAD risk. For those patients with CAD or a CAD-risk equivalent, the National Cholesterol Education Program (NCEP) recommends that LDL-C be lowered to <100 mg/dL in high-risk persons, but when risk is very high, a goal of <70 mg/dL is a reasonable option.³ For moderately high-risk persons, an LDL-C goal of <130 mg/dL is recommended, but <100 mg/dL is a therapeutic option. NCEP guidelines encourage lifestyle and dietary modifications to reduce LDL-C, but acknowledge that pharmacological treatment will be necessary for many patients to lower LDL-C to recommended targets.⁴

BA Synthetic Pathway

BA are synthesized from hepatic cholesterol (Fig. 1) and stored in the gallbladder. During digestion, BA are secreted into the small intestine, where they solubilize dietary fats and fat-soluble vitamins to facilitate their absorption and transport. Enterohepatic circulation in a normal human (Fig. 2A) enables approximately 95% of BA to be reabsorbed in the distal ileum and transported back to the liver; this efficient recycling of a small pool of BA (2-4 g⁵) occurs 5 to 10 times a day.^5-7 Because BA exist as anions at physiologic pH, they require active transporters to cross cell membranes. Several classes of these transporters have been described, including sodium-dependent transporters that mediate BA reabsorption from the distal ileum.⁸ The 5% of intestinal BA that are not reabsorbed are excreted in feces, but subsequently replenished through de novo BA synthesis.⁵ The size of the BA pool is kept relatively constant through feedback inhibition of BA synthesis, and thus interruption of the enterohepatic circulation of BA is able to promptly influence the metabolism of cholesterol. Tight regulation of the BA synthetic pathway is necessary to prevent the accumulation of BA in the body to hepatotoxic levels and to control the influence of BA on cholesterol metabolism.⁶

Fig. 1

Fig. 2

Cholesterol is converted to the two primary bile acids in hepatocytes, including cholic acid (CA) and chenodeoxycholic acid (CDCA), via a cascade of 12 enzymatic reactions.^6,9 Under physiologic pH, BA are conjugated to glycine or taurine and exist in an anionic salt form. There are 2 pathways whereby BA are synthesized: the classic (or neutral) and the alternative (or acidic) pathway. BA are predominantly synthesized through the classic pathway, which produces CA and CDCA in roughly equal amounts; the initiating, and rate limiting, enzyme of this pathway is cholesterol 7α-hydroxylase (CYP7A1). After excretion into the bile, the primary bile acids CA and CDCA are partly 7α-dehydroxylated in the intestine by bacterial enzymes to create the secondary bile acids, deoxycholic acid and lithocholic acid, respectively (Fig. 1). The alternative pathway for BA synthesis produces mainly CDCA and requires the enzyme oxysterol 7α-hydroxylase (CYP7B1) rather than CYP7A1. Genetic knockout studies of CYP7A1 and CYP7B1 in mice suggest that the classic pathway is more regulated than the alternative pathway.¹⁰

Several factors may affect BA synthesis. Typically, synthesis is enhanced via transcriptional upregulation of the CYP7A1 gene.⁶ However, because HDL-C is the preferred source of cholesterol for BA synthesis,^11-13 enhanced synthesis might also involve an upregulation of reverse cholesterol transport, perhaps via an increase in lecithin cholesterol acyltransferase (LCAT) activity.¹⁴ Ultimately, much of the regulation of the BA synthetic pathway occurs through transcription factors, including nuclear hormone receptors.

Nuclear Hormone Receptors and Their Role in BA and Cholesterol Homeostasis

Recent studies have indicated that BA regulate the transcription of genes involved in their synthesis and cholesterol homeostasis through effects on nuclear hormone receptors. Consequently, these receptors may represent novel therapeutic targets for hypercholesterolemia and provide insight into the BA pathway's role in other metabolic processes.

One nuclear hormone receptor, liver X receptor (LXR), binds another nuclear hormone receptor, retinoid X receptor (RXR); this complex binds oxysterols (early metabolites of cholesterol), leading to enhanced transcription of CYP7A1.^9,15 The net result is that an increase in cholesterol levels promotes an increase in BA synthesis, suggesting that agonists of LXR might reduce serum cholesterol levels. However, activating LXR may increase triglyceride (TG) levels; thus any successful treatment would have to balance these two effects. One such LXR agonist, which inhibits atherosclerosis in mice but does not induce hypertriglyceridemia, has been described but has not been tested in humans.^16,17

The farnesoid X receptor (FXR) also binds RXR, but when activated this complex has the opposite effect of LXR/RXR on CYP7A1 transcription.^6,9 Bile acids activate FXR/RXR, which stimulates the negative feedback loop.^18-20 BA-activated FXR also decreases transcription of apolipoprotein (Apo) A-1.²¹ BA sequestration may decrease FXR activity, thus increasing ApoA-1 levels. Furthermore, these data imply that an antagonist of FXR could increase high-density lipoprotein cholesterol (HDL-C) via increasing ApoA-1. This theory is supported by data that a naturally occurring FXR antagonist, guggulsterone, can raise serum HDL-C levels in rats.²²

Nuclear hormone receptors may be affected in patients treated with agents that sequester BA or otherwise interrupt enterohepatic circulation. It is tempting to speculate that depletion of the BA pool by these agents might decrease the activity of FXR/RXR, which in turn would allow for an increase in CYP7A1 transcription. However, the effects of such agents on FXR/RXR activity have not been studied directly.

Disrupting Enterohepatic Circulation

Interruption of enterohepatic circulation consists of partial diversion of bile, or a constituent of bile, to fecal excretion (Fig. 2B). Four different methods have been used to interrupt enterohepatic circulation, thereby lowering plasma cholesterol levels. Two of these methods are surgical: chronic partial ileal bypass and short-duration partial drainage of hepatic bile to outside the body by a T-tube in the bile duct, and two are medical: BAS and cholesterol absorption inhibitors (CAI).^1,5,23,24 Substantial reduction of serum cholesterol results from partial diversion of bile out of the body via a T-tube in the bile duct. An average daily loss of 1.5 g of cholic acid (CA) and 300 mg of cholesterol resulted in a decrease in serum total cholesterol (TC) of about 40% in 4 days, with this reduction maintained for up to 11 days of continued drainage.²³ CA was estimated to account for 71.5% of total BA loss.²⁴ This indicates the potential for similar effects to be obtained by medical interruption of enterohepatic circulation. The medical methods partially duplicate the effects of ileal bypass by diverting only one of the two steroid elements of bile, BA or cholesterol. Disruption of BA enterohepatic circulation by BAS to treat hypercholesterolemia is the focus of this review.

BAS

BAS are positively charged indigestible resins that bind to negatively charged BA in the intestinal lumen and are then excreted with bound BA in the feces. BAS disrupt enterohepatic circulation of BA, depleting the endogenous BA pool by approximately 40%.⁷ As BA are chronically depleted, CYP7A1 and hepatic LDL receptors are upregulated, increasing BA synthesis and reducing plasma LDL-C concentrations.^7-25 BAS typically lower LDL-C by 15 to 26% at the maximum approved doses.^26-29 BAS can induce a slight increase (4-8%) in HDL-C levels.³⁰ ApoA-1, the major lipoprotein constituent of HDL, also increases with BAS treatment.³¹ Small increases in TG may occur in patients taking BAS.^32-34

Two types of bile acid sequestrants:

* Conventional○ cholestyramine○ colestipol

* Specifically engineered○ colesevelam hydrochloride

Available BAS

The available BAS include the conventional sequestrants, cholestyramine and colestipol, and the specifically engineered BAS, colesevelam hydrochloride (HCl) (Table 1).

Table 1

Cholestyramine is an anion exchange resin consisting of trimethylbenzylammonium groups in a long-chain polymer of styrene and divinylbenzene, and colestipol is a long-chain polymer of diethylenetriamine and 1-chloro-2,3-epoxypropane.³⁵ These 2 agents are typically milled into powders for oral administration as a suspension with water or juice.³⁵ Colestipol is also formulated as a tablet, of which up to 16 may be administered daily.³³

Cholestyramine and colestipol have a greater affinity for dihydroxy than for trihydroxy BA and therefore preferably bind to CDCA and deoxycholic acid.⁷ Over time, this creates an imbalance in the BA pool, increasing the proportion of trihydroxy BA.³⁶ This may limit the efficacy of these BAS because, as the hydrophilic content of the BA pool increases, there are fewer BA for these agents to sequester.

Colesevelam HCl is a polyallylamine cross-linked with epichlorohydrin and alkylated with 1-bromodecane and (6-bromohexyl)-trimethylammonium bromide.^37,38 Unlike the conventional BAS, the backbone of colesevelam HCl was specifically engineered to contain long hydrophobic side chains, which maximize hydrophobic interactions with BA, thus enhancing affinity, specificity and capacity compared with conventional BAS (Fig. 3).^37,39 Colesevelam HCl binds BA via both hydrophobic interactions and ionic bonding with primary amines, while quaternary amine side chains stabilize the structure.³⁷ Thus, a larger number of BA can bind to, and remain bound by, the colesevelam HCl polymer, allowing for an increase in fecal excretion of BA per dose. Colesevelam HCl binds to both dihydroxy and trihydroxy BA^30,36 and binds CA with greater affinity than cholestyramine or colestipol in vitro.³⁹ Thus, colesevelam HCl should not result in an imbalance in the BA profile, nor should it be subject to the reduced efficacy seen over time with cholestyramine and colestipol.

Fig. 3

Safety and Tolerability of BAS

To determine the safety of BAS therapy, a review of the published literature and the prescribing information for each BAS was conducted. In February 2005, the PubMed database was searched using the terms BA sequestrants, BA resins, BA binding, cholestyramine, colestipol, and colesevelam, in conjunction with the terms safety, adverse events, adverse effects, and side effects. In addition, these terms were used to search Reactions Weekly, which includes published case reports, incidence, and clinical studies discussing adverse events (January 2000-February 2005).

The large molecular size of BAS preparations prevents them from being absorbed by the intestinal mucosa, which leads to fewer systemic adverse effects.^7,35,40 The major adverse effects reported with conventional BAS are gastrointestinal, including constipation and flatulence. For example, in the Lipid Research Clinics Coronary Primary Prevention Trial (LRC-CPPT), 39% of cholestyramine-treated patients reported moderate-to-severe constipation, versus 10% of placebo-treated patients.²⁶ In addition, titration of cholestyramine to higher doses (>8 g/d) to enhance lipid-lowering may further increase adverse effects and decrease compliance.^41,42 Adverse effects, as well as complicated dosing schedules and poor palatability, result in discontinuation rates of 40 to 60% for patients taking conventional BAS.^26,43,44

Because of its enhanced BA binding, colesevelam HCl can be administered at lower doses than cholestyramine and colestipol, thus reducing adverse effects.^{28,29, 37} However, constipation and dyspepsia are reported in more patients taking colesevelam HCl versus placebo (11% versus 7% and 8% versus 3% for constipation and dyspepsia, respectively).³² Although the lower incidence of gastrointestinal adverse effects for colesevelam HCl may increase patient compliance and adherence to therapy as compared with cholestyramine and colestipol, this has not been studied directly.

Major side effects associated with cholestyramine and colestipol include constipation and flatulence, leading to high patient discontinuation rates, between 40 to 60%. Due to enhanced bile acid-binding capacity, colesevelam HCl is administered at lower doses, thus reducing GI side effects. This may lead to better patient compliance.

Effects on intestinal mucosa and tumorigenesis

Cholestyramine and colestipol treatment caused disruptive changes in colonic morphology in rats,⁴⁵ and cholestyramine increased the number of colon tumors in rats also receiving carcinogen.⁴⁶ However, clinical experience indicates that even if similar changes occur in humans, they do not lead to increases in tumorigenesis with long-term use of these BAS.^33,34 Cholestyramine treatment did not cause an excess of gastrointestinal tumors during the 10-year duration of the LRC-CPPT trial²⁶ or in the 6 years of follow-up.⁴⁷ Colesevelam HCl did not increase tumorigenesis in mice at doses up to 3 g/kg/d (approximately 50 times the maximum recommended daily dose for humans); however, in rats, an increase in the incidence of pancreatic acinar cell adenoma was seen at doses approximately 20 times the maximum recommended daily dose for humans with no differences in colon tumorigenesis observed.³²

Drug interactions and effects on bioavailability

Because BAS are not metabolized, there are no incidences of systemic drug-drug interactions. However, BAS are positively charged and therefore may nonspecifically bind to coadministered drugs, particularly those that are acidic, and thus reduce the bioavailability of these drugs.

BAS, in reducing the intestinal content of BA, may impair the solubilization and absorption of fat-soluble substances, including fat-soluble vitamins.³⁵ Therefore, it is recommended that these vitamins be supplemented in patients receiving long-term BAS therapy. Decreases in vitamin absorption may occur to a lesser extent in patients treated with colesevelam HCl compared with the conventional BAS, as one study found that colesevelam HCl did not significantly change serum levels of vitamins A and E.²⁹

Cholestyramine can affect the bioavailability of coadministered drugs; valproic acid, warfarin, cardiac glycosides (eg, digoxin), quinidine, glipizide, thiazide diuretics, propranolol, phenobarbital, penicillin G, estrogens, progestins, and phenylbutazone have been reported.^34,48-50 Similarly, interactions of colestipol with gemfibrozil, propranolol, furosemide, tetracycline, and penicillin G have been reported.³³ Colestipol has also been reported to interfere with absorption of thiazide diuretics, but to a lesser extent than cholestyramine.⁵¹ Absorption of digoxin is not affected by colestipol.⁵² Fewer interactions of colestipol with some coadministered drugs, as compared with cholestyramine, may occur because cholestyramine is more highly charged.

Colesevelam HCl, unlike cholestyramine, did not decrease the absorption of valproic acid or the common cardiovascular agents warfarin, digoxin, and quinidine.⁵³ In addition, colesevelam HCl did not affect the bioavailability of metoprolol.⁵³ Coadministration of colesevelam HCl with sustained-release (SR) verapamil reduced the C_max and AUC_(0-∞) of SR verapamil by 31% and 11%, respectively.⁵³ However, the clinical significance of this finding is unclear because even in the absence of colesevelam HCl, significant inter-individual variability in SR verapamil pharmacokinetic properties is observed.⁵³

Colesevelam HCl's apparent lack of effect on the bioavailability of coadministered agents is an important consideration when using combination therapies. Colesevelam HCl does not adversely affect the pharmacokinetics of the 3-hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) reductase inhibitor lovastatin, which is frequently administered in combination with colesevelam HCl to additively lower LDL-C.⁵⁴ In addition, colesevelam HCl, when administered either concomitantly or 4 hours following fenofibrate, had no significant effect on the bioavailability of fenofibrate. Furthermore, when the two drugs were coadministered, colesevelam HCl did not significantly affect the total drug exposure as measured by the AUC_(0-t) and AUC_(0-∞) of fenofibric acid.⁵⁵ Interactions of colesevelam HCl with other drugs have not been studied.

Contraindications and precautions

BAS are contraindicated in patients with bowel obstruction. Because BAS may modestly increase TG levels and have not been studied in patients with significantly elevated TG levels (>300 mg/dL), they should not be used in these patients or in those with familial abetalipoproteinemia or genetic hypobetalipoproteinemia.

BAS should only be used in pregnant women if clearly needed; cholestyramine and colestipol carry a pregnancy category C rating, and colesevelam HCl carries a more favorable pregnancy category B rating. Colesevelam HCl administered at approximately 30 times the approved dose did not show any significant adverse reproductive or fertility effects in rats.⁵⁶ In addition, colesevelam HCl did not lead to developmental toxicity in rats or rabbits or in pre- or postnatal toxicity in rats.⁵⁷

Efficacy of BAS Therapy

Controlled clinical trials have evaluated the efficacy of BAS for both lipid alterations and atherosclerotic regression.

Monotherapy with BAS: lipid effects

BAS monotherapy has been used to treat hypercholesterolemia since the 1960s.^58-60 All BAS are indicated as first-line therapy for primary hypercholesterolemia and lower LDL-C by 15 to 26%,^26-29 with maximum effects observed within 2 weeks.³⁵ Following discontinuation of BAS treatment, cholesterol levels typically return to pretreatment levels within a month, without a transitory rebound to higher levels.³⁵ Studies employing BAS monotherapy are summarized in Table 2.

Table 2

Two trials of the effects of cholestyramine therapy conducted in the 1970s, the LRC-CPPT study^26,61 and the National Heart, Lung, and Blood Institute (NHLBI) Type II Coronary Intervention Study,^27,62 demonstrated that BAS significantly lowered TC and LDL-C.

LRC-CPPT was a double-blind, placebo-controlled trial of the effects of cholestyramine on the prevention of CAD in 3,806 men with primary hypercholesterolemia.^{26, 61} Cholestyramine treatment reduced TC and LDL-C by 13.4% and 20.3%, respectively, which was 8.5% and 12.6% greater than the reductions with placebo (P < 0.001). The aforementioned data are for all enrolled patients; however, some patients took less cholestyramine than the prescribed 24 g/d. When specific dosages were analyzed, a dose response was observed for both TC and LDL-C lowering.

The NHLBI Type II Coronary Intervention Study treated 143 patients with cholestyramine or placebo and found that cholestyramine treatment reduced LDL-C by 26% versus 5% with placebo (P < 0.001).^27,62 Although HDL-C levels increased slightly with cholestyramine, the difference was not significant versus placebo. TG levels increased in both groups, with no significant difference between treatments.

The development of the specifically engineered BAS colesevelam HCl provoked further studies of the effects of BAS on lipid levels. A double-blind placebo-controlled study with colesevelam HCl (1.5, 2.25, 3.0, or 3.75 g/d) was conducted in 137 patients for 6 weeks.²⁸ Colesevelam HCl decreased LDL-C levels dose-dependently by up to 19% (P < 0.001 versus baseline). At the 2 highest doses, colesevelam HCl also significantly increased HDL-C by 8 to 9%. No significant changes were observed in TG levels in any treatment group.

Another double-blind study compared colesevelam HCl (2.3, 3.0, 3.8, or 4.5 g/d) and placebo for 24 weeks in 494 patients with primary hypercholesterolemia.²⁹ Colesevelam HCl decreased LDL-C levels by 18% at the highest dose; all doses reduced LDL-C significantly more than placebo (P < 0.001). LDL-C reductions were achieved within 2 weeks and sustained over the 6 months of treatment. A 3 to 4% increase in HDL-C was observed; although this was significantly better versus baseline or placebo, it is a smaller increase than was observed by Davidson and colleagues.²⁸ This may be due to differences in study design and patient populations. ApoB levels also decreased in a dose-dependent manner in response to colesevelam HCl. Although TG levels increased relative to baseline in all groups, none of the increases were statistically significant versus placebo.

Small, dense LDL particles are emerging as a risk factor for CAD,⁶³ and colestipol has been shown to reduce LDL particle size, creating more atherogenic particles.⁶⁴ Therefore, the effect of colesevelam HCl on LDL particle size was examined.⁶⁵ Colesevelam HCl 3.0 and 3.75 g/d reduced LDL particle number by 6.8% and 13.7%, respectively (P < 0.05 versus baseline). Furthermore, colesevelam HCl 3.75 g/d increased LDL particle size by 1.1% (P < 0.05 versus baseline).

Combination therapy with BAS: lipid effects

Combination therapy is being used more widely because of more stringent recommendations for lowering LDL-C, especially in high-risk patients,⁴ and the recognition that improvements in HDL-C and TG are also important.⁶⁶ Ideal combinations will provide additive or synergistic efficacy with complementary mechanisms of action, safety profiles similar to or better than each drug as monotherapy, and no adverse drug-drug interactions.

BAS and statin combinations.

Although statins (HMG-CoA reductase inhibitors) are currently the most prescribed lipid-lowering therapy, titration to higher doses typically results in only small additional LDL-C lowering: with each doubling of the statin dose, LDL-C levels fall by an additional 6%.⁴ Statin use can be associated with muscle toxicity, including myopathy and rhabdomyolysis, which is more likely to occur at higher doses.⁶⁷ Thus, patients who require more aggressive lipid-lowering can benefit from combination therapy. Because HMG-CoA reductase is upregulated in response to depletion of BA, blocking this enzyme with a statin results in complementary and additive effects on the lipid profile. Thus, a BAS and statin combination is particularly efficacious. Numerous clinical studies have examined this combination (Table 3).

Table 3

Table 3

Combination therapy with cholestyramine and pravastatin or fluvastatin has additive effects on LDL-C and TC lowering, compared with either drug alone.^42,68,69 Pan and colleagues found that pravastatin (5, 10, or 20 mg twice daily) plus cholestyramine (24 g/d) reduced LDL-C by 47 to 56% and increased HDL-C by 11 to 18%.⁶⁹ Another study showed that pravastatin (20 mg twice daily) combined with cholestyramine (24 g/d) reduced LDL-C levels by 51%.⁶⁸ HDL-C levels in this study increased by 5%, regardless of the treatment group. Furthermore, a pharmacoeconomic study reported that treatment with pravastatin 20 mg/d plus cholestyramine 10 g/d had greater efficacy in lowering LDL-C and was more cost-effective than monotherapy with pravastatin 20 or 40 mg/d.⁷⁰ Analogous to the pravastatin studies, low-dose fluvastatin plus cholestyramine reduced LDL-C, more than doubling the fluvastatin dose.⁴²

All 3 BAS have been studied in combination with lovastatin.^71-74 Lovastatin 20 mg/d plus colestipol 10 g/d produced greater reductions in LDL-C than did high-dose lovastatin (40 mg/d) monotherapy.⁷³ Moreover, low-dose combinations were more than 25% more cost-effective than high-dose statin monotherapy. Another low-dose combination, lovastatin 5 mg/d plus cholestyramine 8 g/d, lowered LDL-C to a similar extent compared with lovastatin 20 mg/d monotherapy.⁷⁴ A 4-week placebo-controlled study in 135 patients compared low-dose colesevelam HCl (2.3 g/d) and lovastatin (10 mg/d) alone and in combination and found that the combination reduced LDL-C levels by 32 to 34%, a significantly greater reduction than seen with either therapy alone (P < 0.05),⁷¹ and equivalent to the standard statin dose response.³

Patients on a 3-drug regimen consisting of niacin, colestipol and lovastatin for 10 years had a significant reduction in cardiovascular events compared with patients taking the standard regimen.

Low-dose colestipol (5-10 g/d) plus simvastatin (20-40 mg/d) reduced LDL-C by 45 to 50%, a significantly greater reduction than with either therapy alone.⁷⁵ Similarly, colesevelam HCl (2.3 or 3.8 g/d) plus simvastatin (20 or 10 mg/d) reduced LDL-C by 42%, which was more than with either agent alone.⁷⁶

Colesevelam HCl (3.8 g/d) plus low-dose atorvastatin (10 mg/d) lowered LDL-C by 48%, which was greater than the reductions observed with either therapy alone (12% with colesevelam HCl and 38% with atorvastatin; P < 0.01), and similar to reductions achieved with high-dose atorvastatin (80 mg/d; 53% reduction).⁷⁷ ApoB decreases mirrored those seen with LDL-C. Interestingly, ApoA-1 levels increased significantly from baseline in response to combination therapy or colesevelam HCl alone, but not with atorvastatin monotherapy.

In summary, BAS have been studied in combination with several statins, and these combinations consistently appear to have additive efficacy. Generally, low-dose combinations lead to greater or similar LDL-C reductions compared with high-dose statin monotherapy and are safe and well tolerated.

BAS and nonstatin combinations.

Combinations of BAS and nonstatin lipid-lowering agents may be useful in those patients who require intensive lipid lowering but have a contraindication to statins, are statin intolerant, or refuse statin therapy.

The combination of a BAS (cholestyramine or colestipol) and niacin appears to be safe and effective for lowering LDL-C by approximately 20 to 40% and increasing HDL-C by up to 40%.^78-82 Thus, this combination may be especially useful in patients who also require an increase in HDL-C. To date, no studies using colesevelam HCl plus niacin have been published.

Studies with BAS and fibrates show that, in general, the combination effectively reduces LDL-C and TG but may be less reliable for increasing HDL-C.^83-89 For example, colesevelam HCl plus fenofibrate decreased LDL-C and TG by 17% and 32%, respectively, and increased HDL-C by 12%.⁸⁹

Another strategy may be to combine a BAS with a cholesterol absorption inhibitor (CAI), such as ezetimibe.¹ One study assessed the additive effects of ezetimibe and a BAS by performing a chart review of patients in whom ezetimibe was added to a BAS-based regimen.⁹⁰ The addition of ezetimibe further reduced TC, LDL-C, and TG by 18%, 19%, and 14%, respectively. Another study found that ezetimibe plus colesevelam HCl reduced LDL-C by 37 to 41% in patients who were intolerant to or refused statin therapy.⁹¹ Further studies are needed to determine whether the combination of a BAS and a CAI is additive or synergistic.

Morbidity, mortality, and plaque regression trials with BAS

The ultimate test of the efficacy of BAS has come from studies designed to determine whether BAS, alone or in combination with other lipid-lowering therapies, can reduce cardiovascular outcomes, slow plaque progression, or induce plaque regression (Table 4).

Table 4

One of the earliest trials to study the effects of BAS on CAD outcomes was reported by Dorr and colleagues in 1978.⁹² This early trial had limitations⁹³ but indicated that colestipol treatment was safe and effectively reduced CAD outcomes.

LRC-CPPT, which compared cardiovascular outcomes in 3,806 men treated double-blind with BAS or placebo, was one of the first large studies to definitively associate a reduction in LDL-C with a reduction in CAD risk. Patients treated for a mean of 7.4 years with cholestyramine had a 19% reduction in CAD risk (P < 0.05 compared with placebo)²⁶ that was associated with a 13% decrease in LDL-C and TC levels.⁶¹ Follow-up for 6 years posttrial indicated that the reduction of CAD risk was attenuated after cessation of treatment, suggesting that active lipid-lowering therapy should be continued to maintain improvement in cardiovascular outcomes.⁴⁷ In addition, adverse events did not significantly increase with 6 years of follow-up.

The NHLBI Type II Coronary Intervention Study evaluated the effects of cholestyramine versus placebo on the progression of CAD as assessed by coronary angiography after 5 years of treatment in 116 patients.²⁷ LDL-C levels were reduced by 26% with cholestyramine and 5% with placebo (P < 0.001). Overall, CAD progressed in 49% of placebo-treated patients versus 32% of cholestyramine-treated patients (P < 0.05). When evaluating lesions causing 50% or greater stenosis at baseline, lesion progression was observed in 33% of placebo-treated patients versus 12% of cholestyramine-treated patients (P < 0.05). Increases in HDL-C, decreases in LDL-C or TC, and increases in the ratio of HDL-C to LDL-C or TC were associated with lower rates of CAD progression.⁶²

The St. Thomas' Atherosclerosis Regression Study (STARS) compared, in 90 men with CAD, the effects of dietary intervention or diet combined with cholestyramine treatment on angiographic progression after 39 months.⁹⁴ LDL-C was reduced by 16% with diet and 36% with diet plus cholestyramine. The mean absolute lumen diameter of the coronary segments evaluated increased by 0.1 mm more in those patients treated with cholestyramine and diet versus those treated with diet alone (P < 0.05); however, there was no significant difference in cardiovascular events between the 2 groups (3 events occurred in the group treated with diet alone versus 1 in the group treated with diet and cholestyramine). The change in coronary lumen diameter correlated with LDL-C levels and the LDL-C/HDL-C ratio.

Combination therapy including BAS also reduces progression of atherosclerotic lesions. Studies using a combination of niacin and colestipol or lovastatin and cholestyramine have shown that aggressive lowering of LDL-C (<100 mg/dL) in patients with coronary venous⁷⁸ or coronary artery⁹⁵ bypass grafts, respectively, reduces progression in the grafts and native arteries. In fact, niacin is specifically indicated with any of the 3 BAS for slowing progression or promoting regression of atherosclerotic disease in patients with a history of CAD.⁹⁶

The Familial Atherosclerosis Treatment Study (FATS) examined angiographic progression and cardiovascular outcomes in response to combination therapy including BAS. After 6 years of treatment, 2-drug (niacin and colestipol or lovastatin and colestipol) combination therapy reduced both the frequency of angiographic progression of coronary lesions and cardiovascular events in men with documented CAD who were at high risk for cardiovascular events, although the differences in progression were not significant compared with usual care.⁸⁰ At the conclusion of FATS, patients were invited to continue with nonrandomized 3-drug (niacin, colestipol, and lovastatin) therapy or return to usual care and were followed up for a mean of 8 and 10 years, respectively. Of those patients who chose to continue with triple therapy, 5.3% experienced a cardiovascular event, and the overall mortality rate was 1.3%, compared with 18.8% and 19.8% of patients who returned to usual care (P = 0.05 and P = 0.001, respectively).⁹⁷ These results highlight the benefit of long-term lipid-lowering treatment for improving cardiovascular outcomes.

Further studies have shown the benefits of 3-drug therapy including BAS. In patients with familial hypercholesterolemia, a controlled clinical trial of colestipol, niacin, and lovastatin triple therapy resulted in angiographic regression; levels of LDL-C were found to be the best single predictor of outcome in these patients.⁹⁸ Another study found that gemfibrozil, niacin, and cholestyramine triple therapy significantly reduced cardiovascular events versus placebo.⁹⁹

Thus, treatment of patients who have documented CAD or who are at risk for CAD with BAS alone or as part of combination therapy can significantly reduce CAD progression and the risk of CAD-associated outcomes. A recent meta-analysis of 8 randomized controlled BAS trials found that BAS monotherapy significantly reduced cardiac mortality.¹⁰⁰ It is also clear that the reduction in cardiovascular outcomes correlates with reductions in LDL-C levels across the range of LDL-C levels studied.¹⁰¹

Clinical Experience

BAS are effective for primary and secondary prevention of CAD and for treating atherosclerotic plaques. Clinical experience indicates that BAS can be beneficial for a range of patient populations, including those of different ages and genders.

Despite the widespread use of statins to lower LDL-C, a 2000 study found that only 40% of patients treated with statins alone achieved NCEP-specified LDL-C targets.¹⁰² Therefore, there is an unmet need for lipid-lowering therapies that are not only effective in lowering LDL-C but can also provide additional LDL-C lowering when combined with a statin. There is also a need for agents that increase HDL-C and decrease TG levels when combined with a fibrate.

Patients who are not able to tolerate treatment with statins due to adverse effects,⁶⁷ and patients in whom statin therapy is contraindicated, are candidates for treatment with a BAS. Combination therapy with statins also allows for lower doses of both the statin and the BAS to be administered, leading to fewer adverse effects.

Bile acid sequestrants are also beneficial as part of a combination therapy in patients with very elevated cholesterol levels or in very high-risk patients, as the combination of a BAS and a statin can lower LDL-C more effectively than simply raising the dose of the statin.^42,70,73 A semilog linear relationship between LDL-C levels and CAD has been demonstrated, and suggests that for every 30 mg/dL change in LDL-C (for levels between 40 and 200 mg/dL), the relative risk for CAD is changed by about 30%.³

Because BAS are not systemically absorbed, they may be useful for treating patients who require or prefer nonsystemic therapy. Therefore, they may be useful for pregnant women or adolescents; however, these patients should only take BAS if clearly needed. To date, no evidence has suggested that special considerations should be taken when treating elderly patients with BAS.

Although fewer adverse effects appear to occur with BAS specifically engineered for selective BA absorption compared with conventional BAS, lower compliance may still occur with some patients. To achieve better efficacy with BAS therapy, clinicians should counsel patients on appropriate use and consider using compliance aids, such as telephone and computer-based reminders.⁴

Future Developments

The successful LDL-C lowering, reduction in CAD events, atherosclerotic regression, and safety observed with 2-drug combination therapy including a BAS argues that therapy with 3 or even 4 drugs might be more efficacious when acting by complementary mechanisms. Typically, the triple combination that has been employed is a BAS, a statin, and niacin (Table 3).^{103, 104} This combination appears to be more effective than monotherapy or 2-drug therapy, but more studies are needed to prove this a valid treatment alternative. A BAS and CAI combination may also prove to be additive or synergistic, and more potent bile acid sequestrants may be developed. Such additional treatment options may increase the capability to lower LDL-C to <70 mg/dL in very high-risk patients, as recommended by the NCEP.³

BAS and other treatments that target the BA pathway may have additional, nonlipid-lowering therapeutic effects. For example, BAS may indirectly affect nonlipid parameters that play a role in CAD development. The BA synthetic pathway and the nuclear hormone receptors LXR and FXR play roles in glucose metabolism and thus are interesting potential targets for diabetes therapies. Treatment of patients with dyslipidemia and type 2 diabetes with cholestyramine significantly reduced mean plasma glucose levels and median urinary glucose excretion.¹⁰⁵ Activated LXR inhibits hepatic gluconeogenesis and lowers serum glucose levels, thus improving glucose tolerance.^106-108 Therefore, an LXR agonist might be beneficial for diabetics. However, this is coupled with an increase in circulating lipids and TG,¹⁰⁹ and therefore, the beneficial and deleterious effects must be balanced. It remains to be seen how BAS effects LXR and FXR.

Conclusions

Twenty-five years of controlled clinical trials have demonstrated that BAS therapy is both safe and effective for treating dyslipidemia, alone and in combination with other lipid-lowering therapies such as statins. In addition to lowering LDL-C and slightly raising HDL-C levels, bile acid sequestrants reduce the progression of atherosclerotic plaques and the incidence of clinical cardiovascular events. Bile acid sequestrants, as the only nonsystemic lipid-lowering agents, have a demonstrated safety record. Newer, specifically engineered BAS have greater potency, limited binding of comedications due to selective absorption, and a lower incidence of gastrointestinal side effects compared with older BAS. It is also becoming clear that BAS and other therapies that manipulate the BA synthetic pathway may have therapeutic effects on other metabolic disorders such as type 2 diabetes.

Acknowledgments

The preparation of this manuscript was supported in part by a grant from Sankyo Pharma Inc. Thank you to Sarah Seton-Rogers, PhD for assistance in conducting the literature search and the development of this manuscript. The opinions expressed in this report are those of the author.