Thyroid mimetic as an option for lowering low-density lipoprotein

In this issue of PNAS, Berkenstam et al. (1) demonstrate that the synthetic thyroid hormone mimetic KB2115 produces a marked reduction in low-density lipoprotein (LDL) cholesterol. This mimetic is designed to have a selective effect on the β-isoform of the thyroid hormone receptor. Activation of this receptor has been reported to modify cholesterol metabolism in the liver but not to affect the heart (2, 3). The β-isoform of the receptor is to be distinguished from the α-isoform, which mediates the cardiogenic actions of thyroid hormone.

Thyroid Hormone Mimetics

Selective thyroid hormone mimetics have shown potential in animal models. In these models, β-selective mimetics have markedly reduced serum cholesterol levels and produced weight loss without causing tachycardia (3). Because the thyroid-hormone receptor activates a multitude of pathways, the exact mechanisms whereby cholesterol levels are lowered and weight loss occurs are uncertain. One mechanism likely is through activation of the LDL-receptor pathway (4).

As shown by Berkenstam et al. (1), in a 2-week study, LDL-cholesterol levels were reduced by KB2115 by ≈40%. No consistent changes were found for high-density lipoproteins (HDL) or triglycerides. Indirect markers of cholesterol synthesis showed no change, although there was a suggestion that bile acid synthesis was increased. KB2115 produced no changes in heart rate, blood pressure, or electrocardiogram.

The current results raise the question of whether KB2115 or other selective thyroid hormone mimetics might become viable candidates for cholesterol-lowering drugs to be used in clinical practice. If so, they must be viewed in the context of currently available cholesterol-lowering drugs. The success of the latter builds a high hurdle for the introduction of any new category of cholesterol-lowering drug. But at the same time, the marked clinical benefit derived from cholesterol reduction highlights the potential to the pharmaceutical industry for new agents to augment or supersede currently available drugs.

Statins

Clearly, the discovery of statins opened a new era for prevention of atherosclerotic cardiovascular disease (ASCVD) (5). Although several lines of evidence underpin the hypothesis that LDL and other lipoproteins containing apolipo protein B constitute a major risk factor for ASCVD (6), the repeated demonstration that LDL reduction with statin therapy reduces ASCVD events provided the confirmation needed for introduction and acceptance of LDL-lowering therapy in clinical practice (6, 7). Reduction of LDL levels with statins can reduce risk for ASCVD events between 35% and 45% over a 5-year period (6–10). Moreover, epidemiological and genetic studies suggest that maintaining very low LDL levels over a lifetime will largely eliminate ASCVD even when other risk factors are present (11, 12). Other risk factors undoubtedly accelerate atherosclerosis, but only when atherogenesis has been initiated and sustained by some elevation of plasma LDL.

Hyperthyroidism is well known to be accompanied by low levels of LDL.

Statins as a class are amazing drugs. They specifically target hydroxymethylglutaryl-CoA reductase in the liver to inhibit cholesterol synthesis; this in turn raises LDL receptor levels and lowers plasma LDL levels (13). Statins generally are well tolerated; for most patients, they are free of side effects (14). Now that most statins have gone off patent, expense of therapy is no longer an issue. But statins do have their limitations. A small percentage of persons cannot tolerate them because of either associated muscle discomfort or other perceived or real side effects. In others, LDL lowering is insufficient to achieve treatment goals. For these reasons, statins are not the final answer to treatment of an elevated LDL. The pharmaceutical industry thus is still searching for new drugs that can either supplement statins or replace them in intolerant patients. One promising example is ezetimibe, which blocks cholesterol absorption. In combination with low-dose statins, ezetimibe will reduce LDL levels to the range observed with high dose statins (15). Recently another potential target for LDL-lowering therapy has been identified, namely PCSK9 (16). This protein promotes the degradation of LDL receptors in the liver. Mutations in PCSK9 impair its function, which blocks receptor degradation, thereby lowering LDL levels. Subjects who carry these mutations are virtually devoid of ASCVD throughout life, even when other risk factors are present (12). If a drug could be developed that would inhibit the action of PCSK9, it could prove to be highly effective for the treatment of hypercholesterolemia.

Thyroid hormone appears to be another modulator of LDL-receptor activity. Hyperthyroidism is well known to be accompanied by low levels of LDL. There has been interest for many years in the treatment of hypercholesterolemia with thyroid hormone or one of its derivatives. Because l-thyroxine has too many side effects to be used long-term for LDL lowering, other derivatives have been sought. In the 1960s, dextrothyroxine was considered to be a candidate because of a perceived lesser cardiotoxicity. However, in the Coronary Drug Project, dextrothyroxine was found to cause too many cardiovascular complications to be an acceptable therapy (17).

The initial results with KB2115 as a selective thyroid hormone mimetic are encouraging. If such a mimetic could be shown to be devoid of adverse effects on the cardiovascular system, bone, or muscle, it might be useful either as an adjunct to statin therapy or as an alternate LDL-lowering drug in patients who are intolerant to statins. However, because LDL-lowering drugs presumably are prescribed as lifetime therapy, they must have a minimum of side effects to ever be acceptable. Particular scrutiny must be given to the possibility of cardiovascular side effects. The long-term effects on HDL levels must also be monitored carefully in view of the known action of thyroid hormone to reduce HDL concentration. Should a thyroid mimetic significantly reduce HDL levels, its viability as an LDL-lowering drug would be cast into doubt.

Cautions

Thyroid hormone mediates its effects through nuclear receptors. A word about nuclear receptors in general as drug targets for chronic diseases is in order. On the one hand, they are tempting targets because they have multiple actions that could favorably affect several metabolic pathways to mitigate a disease process. On the other, some of the activated pathways could produce adverse consequences. The peroxisome proliferator-activated receptor (PPAR) nuclear receptors are a case in point. Activation of PPAR α by fibrates affects several pathways in the liver that regulate lipid metabolism. Fibrates lower plasma triglycerides and LDL cholesterol and raise HDL cholesterol. At the same time, they increase biliary cholesterol (18), which predisposes to cholesterol gallstones. In animals, activation of PPAR α may have adverse effects on myocardial function (19). A partner PPAR is the γ-isoform, which appears to be most active in adipose tissue. Activation of PPAR γ by thiazolidinediones improves insulin resistance through several pathways and reduces hyperglycemia. At the same time, thiazolidinediones predispose to fluid retention, which in some patients can lead to heart failure. Certainly if selective activation of thyroid-receptor β in the liver by thyroid hormone mimetics could reduce LDL levels without systemic side effects, they could have potential as LDL-lowering drugs. Nevertheless, thyroid hormone receptor β is widely distributed throughout the body (20). For this reason, extensive and long-term studies in a large number of patients will be required to document that benefits of LDL lowering with these agents outweigh their side effects.