FH is an inherited disease of lipid metabolism characterized by persistently elevated levels of LDL-C and premature CVD (Goldstein, Hobbs, and Brown 2001).
Clinical diagnosis is most often based on Dutch Lipid Clinic Network (DLCN) criteria (Table 1), less frequently on Simon Broone or MEDPED criteria. (Austin et al.
2004; Haase and Goldberg 2012; Nordestgaard et al. 2013; Williams et al. 1993).
A score of over 8 points based on the DLCN criteria is diagnostic for FH. If the score is between 6 to 8 points FH is probable and with 3 to 5 points possible. Definitive diagnosis is achieved by identification of a pathogenic variant in one of the three genes (APOB, LDLR and PCSK9) known to be associated with FH.
Clinical manifestations of FH vary widely depending on the mutations. Typical clinical signs include a varying extent of hypercholesterolemia (especially elevated LDL-C), cholesterol deposits in tendons called xanthomas, cholesterol deposits around the eye called xantelasmas and arcus cornea which is a deposition of lipids in the cornea of the eye. The life-long and persistent elevation of LDL-C, if untreated, leads to premature and accelerated atherosclerotic cardiovascular disease. The risk of cardiovascular disease is 2.5 to 10-fold increased in FH but when diagnosed and treated early in life, is reduced by up to 80%. This highlights the importance of early diagnosis and initiation of cholesterol lowering treatment for prevention of future cardiovascular complications.
Cascade screening is a systematic process of identifying individuals at risk for FH based on the identification of a person with the disease and/or a pathogenic variant, and then testing the at-risk biological relatives of this person. As new cases or pathogenic variants are identified more rounds of screening are carried out. Cascade screening has proven effective in identifying new cases when a monogenic background has been verified for the index case. As FH is still vastly underdiagnosed, some countries like the Netherlands have adopted active screening programs to improve the discovery of new FH cases. (Louter, Defesche, and Roeters van Lennep 2017; Umans-Eckenhausen et al. 2001).
While on the one hand there is a definite need for more intensive screening to identify FH patients for treatment to reduce future CVD risk, on the other hand in 25-30% of clinically diagnosed FH patients no pathogenic variants can be identified (Taylor et al. 2010). In these cases, a likely polygenic background explains the hypercholesterolemia and cascade screening might not be cost-effective.
Table 1. Dutch Lipid Clinic Network diagnostic criteria for familial hypercholesterolemia
Criteria Points
Family History
degree relative with known premature* coronary and vascular disease OR
First-degree relative with known LDL-C level over the 95th percentile 1 First-degree relative with tendon xanthomata and/or arcus cornealis OR
Children aged less than 18 with LDL-C level over the 95th percentile 2 Clinical History
Patient with premature* coronary artery disease 2
Patient with premature* cerebral or peripherial vascular disease 1 Physical Examination
Tendinous xanthomata 6
Arcus cornealis prior to age 45 years 4
Cholesterol levels (mmol/l)
Functional mutation in the APOB, LDLR or PSCK9 gene 8
Diagnosis (based on the total number of points obtained)
Definitive Familial Hypercholesterolemia >8
Probable Familial Hypercholesterolemia 6-8
Possible Familial Hypercholesterolemia 3-5
Unlikely Familial Hypercholesterolemia <3
*premature = <55 years in men; <60 years in women.
2.3.2 Genetics and prevalence
Defects in three genes, LDLR, APOB or PCSK9, are known to lead to familial hypercholesterolemia (Marks et al. 2003; Nordestgaard et al. 2013). Most of the mutations (60-80%) causing the FH phenotype arise from mutations in the LDLR gene. Overall, an excess of over 2000 disease-causing variants for familial hypercholesterolemia have been identified to date (Leigh et al. 2017). These include exonic substitutions, exonic small rearrengements, large rearrangements, promoter variants, intronic variants and one variant in the 3´untranslated sequence. Out of the identified variants 81% have been classified as pathogenic, 12% as non-pathogenic and 7% as variants of unknown significance (Leigh et al. 2017). Continued effort to find new mutations behind FH have been encouraged by the fact that with every
FH according to the DLCN criteria, a mutation can only be found in about 25-30% of cases. It has been suggested that most of these cases are actually polygenic, where patients have inherited a greater-than-average number of cholesterol-raising variants. An LDL-C SNP-score has been developed and tested and indeed these SNPs have been estimated to account for 80% of the cases where a clinical diagnosis of possible FH has been made. (Futema et al. 2015; Talmud et al. 2013). Since the phenotype of FH is similar irrespective of genetic background, is there benefit in the knowledge of a monogenic or polygenic background? It has been shown that the prevalence of CHD is higher in monogenic compared to polygenic hypercholesterolemia (Humphries et al. 2006; Khera et al. 2016); more severe intimal thickening in the carotid artery of monogenic vs. polygenic FH patients has been demonstrated despite similar cholesterol levels, and higher calcium scores of lesions from monogenic patients has been demonstrated. (Sharifi et al. 2017).
Traditionally, the functional consequences of LDLR mutations have been classified into five categories: class I null, class II transport defective, class III binding defective, class IV internalization defective and class V recycling defective (Hobbs, Brown, and Goldstein 1992). However, better understanding of the different mechanisms that may affect normal LDLR function suggest that additional classes may be necessary (Koivisto, Hubbard, and Mellman 2001; Strøm et al. 2014; Strøm, Leren, and Laerdahl 2015).
The prevalence of heterozygous FH (heFH) has previously been cited as 1:500 but emerging data from white/European populations suggest that the prevalence of heFH may be as high as 1:200-1:250 (Nordestgaard et al. 2013). Homozygous FH (HoFH) is rare with a prevalence of 1:1000000, although based on the new studies estimates of 1:160000-300000 have been suggested (Cuchel et al. 2013). FH is more common in selected populations, such as the Finnish population, due to founder effects. (Lahtinen et al. 2015; Vuorio et al. 2001). In Finland eight founder mutations (Helsinki, Pohjois-Karjala, Turku, Pori, Pogosta, Keuruu, FH-Espoo and FH-Pro84Ser) account for approximately 80% of the FH cases (Vuorio et al. 2001).
2.3.3 Current treatment of familial hypercholesterolemia
Treatment focuses on lowering LDL-C via diet, exercise and cholesterol lowering drugs. Current guidelines for LDL-C targets for FH patients are <2.3 mmol/l for adults, <1.8mmol/l for adults with CHD or diabetes, and <3.5mmol/l for children.
These targets are achievable in most FH patients with statins, ezetimibe or the new PCSK9 inhibitors. In the most severe cases lomitapide or mipomersen alone or in combination with apheresis is required.
Statins are the cornerstone of treatment. They function via inhibiting the HMG-CoA reductase in the liver affecting cholesterol biosynthesis. Because the product of the HMG-CoA reductase reaction, mevalonate, is a precursor to many non-steroidal isoprenic compounds in addition to cholesterol, benefits beyond cholesterol lowering have been discovered. (Bellosta et al. 2000; Oesterle, Laufs, and Liao 2017). In heFH
statins reduce LDL levels and CHD mortality but in homozygous patients, especially patients who are null-mutants, the efficacy of statins is modest due to the lack of functioning LDL receptors.
A major addition to the treatment of elevated LDL cholesterol has been the introduction of PCSK9 inhibitors, in form of PCSK9-directed antibodies (He et al.
2017). Currently there are two commercially available anti-PCSK9 monoclonal antibodies Alirocumab and Evolocumab (Navarese et al. 2015). In clinical trials responses have been good in patients with receptor defective mutations but those with one receptor negative allele had lower reductions in cholesterol and one patient with two receptor negative alleles had no response to therapy highlighting that current therapies are still not sufficient for all patients. (Bergeron et al. 2015; Stein et al. 2012). Monoclonal antibodies reduce also lipoprotein (a) in addition to apoliprotein-B, non-HDL cholesterol and triglycerides (TGs). Levels of HDL and apolipoprotein A1 are increased as with statin therapy.
Other FH treatments include mipomersen which is an anti-sense oligonucleotide that binds the coding region of human ApoB mRNA and triggers its degradation, resulting in reduction of all ApoB containing atherogenic lipoproteins. An added benefit for hoFH is that the mipomersen-mediated reductions in LDL-C are not dependent on LDLR expression. In clinical studies elevations in liver transaminases and liver steatosis without inflammation or fibrosis were seen but were resolved after discontinuation of the drug. Mipomersen was also shown to reduce major cardiac events by 85%. (McGowan et al. 2012; Raal et al. 2010).
HoFH can also be treated with lomitapide which is a microsomal triglyseride transfer protein (MTP) inhibitor (Rader and Kastelein 2014). MTP is localized in the endoplasmic reticulum of hepatocytes and enterocytes playing a key role in the assembly and secretion of ApoB-containing lipoproteins. Loss-of-function mutations lead to hypocholesterolemia and reduced levels of ApoB containing lipoproteins. The reductions in LDL-C by lomitapide, similarly to mipomersen, are independent of LDLR expression.
Both lomitapide and mipomersen have been shown to cause accumulation of fat in the liver, although in recent trials no inflammation or fibrosis have been seen.
However, the possibility of the development of cirrhosis as a result of continuous, long-term fat accumulation in the liver can not be ruled out. An observational registry to monitor patients receiving lomitapide (LOWER: Lomitapide Observational Worldwide Evaluation Registry) has been set up and will follow lomitapide-treated FH patients for up to 10 years.
LDL apheresis is the physical removal of lipoproteins from the blood. It is indicated in FH patients who are unresponsive or have an inadequate response to cholesterol lowering drugs. It is mostly used in hoFH patients but occasionally is necessary also in severe heFH patients unresponsive to statin treatment. Reductions
The only curative treatment for hoFH is liver transplantation by which normal LDLR function can be restored in the liver (Bilheimer et al. 1984; Starzl et al. 1984).
However, this does not reverse existing cardiovascular complications, and for maximal benefit would need to be performed prior to the development of these. In addition, due to the poor availability of transplants, technical difficulty of the procedure and the life-long immunosuppression required, it is an option only for the most severe cases of hoFH.
2.3.4 Animal models of familial hypercholesterolemia
Several animal models have been available to study hypercholesterolemia and atherosclerosis in vivo including mouse, hamster and rabbit. Mice and rabbits with mutations in the LDLR gene recapitulate the disease and are the most appropriate for use in FH research. While the lipid metabolism of mice differs from humans significantly (Getz and Reardon, 2012), the development of human chimeric mouse models has improved congruity with human FH (Bissig-Choisat et al., 2015).
However, due to the small size of mice and differences in the development and progression of atherosclerosis, the WHHL rabbit can be considered to be the most useful in modeling FH in vivo. (Emini Veseli et al., 2017).
2.3.5 Rabbit models
Rabbits are naturally resistant to the development of atherosclerosis, even on a high-fat diet. Therefore the discovery of a rabbit with significant, spontaneous hypercholesterolemia in a colony of Japanse white rabbits in the early 70’s led to the generation of the WHHL rabbit by selective in-breeding of this mutant (Watanabe, 1980). Later it was shown that the cause of the hypercholesterolemia was deficiency of LDL receptors in the liver and adrenal gland (Kita et al. 1981; Yamamoto et al.
1986). The rabbits show increased levels of total cholesterol, LDL-C and TG at various ages. They develop spontaneous aortic atherosclerosis that progresses in extent and severity with age, and xanthoma of digital joints, both also features of human FH.
Cultured fibroblasts from homozygous animals express less than 5% of the expected number of LDLR.
In WHHL rabbits, the first detectable lesions are seen in the aortic arch at about 2 months of age. Intimal lesions containing CE in SMCs and macrophage foam cells predominate. Raised lesions in all parts of the aorta are visible by age of 6 months, and advanced atherosclerotic plaques, resembling those seen in humans, are present by the age of 10-12 months. Advanced lesions have a necrotic cholesterol-filled core, calcification and a fibrous cap. (Buja et al., 1990). Despite recapitulating the human disease extremely well, differences still exist. In the WHHL rabbit early lesions develop in the thoracic aorta, followed by the abdominal aorta. This relates most likely to the hemodynamic differences between four-legged animals and humans.
Also, aortic stenosis typically present in hoFH is not seen in the WHHL rabbits.
Finally, while both humans and rabbits display elevated plasma cholesterol levels, only WHHL rabbits show marked elevations in TG levels.
A recent publication by Lu and co-workers describes the generation of LDLR KO-rabbits using CRISPR/Cas9 to induce biallelic mutations to exon 7 of the LDLR, with subsequent development of spontaneous hypercholesterolemia and atherosclerosis on a normal chow-diet. While there was great variation in lesion area in this model, it may provide more opportunities to study the heterogenous molecular defects underlying human FH in the future (Lu et al. 2018).
2.4 LIPOPROTEIN METABOLISM