A Catalog of the Pathogenic Variants in ABCG5 and ABCG8 and Clinical Features in Sitosterolemia

Abstract

Aim: Since comprehensive data on the pathogenic variants ofABCG5 andABCG8 and clinical features in sitosterolemia remain limited, we aimed to compile a catalog through an extensive literature search of case reports from 2002 to 2024, as well as an evaluation of variants reported in review articles.

Methods: We compiled 155 cases of sitosterolemia from 133 families with nonsynonymous variants inABCG5 andABCG8, along with data on clinical information from case reports. Pathogenic variants were defined either as 1) protein-truncating variants, 2) classified as pathogenic or likely pathogenic variants according to the ACMG guidelines, or 3) serum sitosterol level of the case was measured at ≥ 1 mg/dL.

Results: Xanthoma was observed in 69.2% of patients, ischemic heart disease in 14.2%, and hematologic abnormalities in 57.9%. Fifty-three variants inABCG5 and 52 inABCG8 were evaluated for their pathogenicity, in which 33 inABCG5 and 29 inABCG8 were protein-truncating variants. Additionally, based on the ACMG criteria and serum sitosterol levels, 50 variants inABCG5 and 51 variants inABCG8 were finally classified as pathogenic. Among them, the frequently observed R446X and R389H inABCG5 were highly prevalent in East Asians, while W361X and S107X inABCG8 were predominantly found in Europeans.

Conclusion: We provided the largest catalog of clinical features and pathogenic variants ofABCG5 andABCG8 in the world. This study may help clarify the pathogenicity of variants inABCG5 andABCG8 and provide a valuable reference for the genetic diagnosis of sitosterolemia.

^†The Committee on Primary Dyslipidemia under the Research Program on Rare and Intractable Disease of the Ministry of Health, Labour and Welfare of Japan: Mariko Harada-Shiba (Cardiovascular Center, Osaka Medical and Pharmaceutical University, Osaka, Japan), Hayato Tada (Department of Cardiovascular Medicine, Kanazawa University Graduate School of Medical Sciences, Ishikawa, Japan), Shinji Yokoyama (Institute for Biological Functions, Chubu University, Aichi, Japan), Hitoshi Shimano (Department of Internal Medicine (Endocrinology and Metabolism), Faculty of Medicine University of Tsukuba, Ibaraki, Japan), Koutaro Yokote (Chiba University, Chiba, Japan), Hideaki Bujo (Department of Clinical-Laboratory and Experimental-Research Medicine, Toho University Sakura Medical Center, Chiba, Japan), Shizuya Yamashita (Department of Cardiology, Rinku General Medical Center, Osaka, Japan), Kazuhisa Tsukamoto (Department of Internal Medicine, Teikyo University, Tokyo, Japan), Katsunori Ikewaki (Division of Neurology, Anti-Aging, and Vascular Medicine, Department of Internal Medicine, National Defense Medical College, Saitama, Japan), Takanari Gotoda (Department of Metabolic Biochemistry, Faculty of Medicine, Kyorin University, Tokyo, Japan), Kazushige Dobashi (Department of Pediatrics, School of Medicine, University of Yamanashi, Yamanashi, Japan), Tomohiro Fujisaka (Department of Cardiology, Osaka Medical and Pharmaceutical University, Osaka, Japan), Misa Takegami (Department of Preventive Medicine and Epidemiology, National Cerebral and Cardiovascular Center, Osaka, Japan), Yoshiki Sekijima (Department of Medicine (Neurology & Rheumatology), Shinshu University School of Medicine, Nagano, Japan), Yasushi Ishigaki (Division of Diabetes, Metabolism and Endocrinology, Department of Internal Medicine, Iwate Medical University, Iwate, Japan), Hiroaki Okazaki (Division of Endocrinology and Metabolism, Department of Internal Medicine, Jichi Medical University, Tochigi, Japan), Atsushi Nohara (Ishikawa Prefectural Central Hospital, Ishikawa, Japan), Shingo Koyama (Division of Neurology and Clinical Neuroscience, Department of Internal Medicine III, Yamagata University Faculty of Medicine, Yamagata, Japan), Kyoko Inagaki (Division of Diabetes, Endocrinology, and Metabolism, Department of Medicine, Nippon Medical School, Tokyo, Japan), Koh Ono (Department of Cardiovascular Medicine, Kyoto University Graduate School of Medicine, Kyoto, Japan), Masahiro Koseki (Division of Cardiovascular Medicine, Department of Medicine, Osaka University Graduate School of Medicine, Osaka, Japan), Hiroyuki Daida (Faculty of Health Science, Juntendo University, Juntendo University Graduate School of Medicine, Tokyo, Japan), Tetsuji Wakabayashi (Division of Endocrinology and Metabolism, Department of Internal Medicine, Jichi Medical University, Tochigi, Japan), Kimitoshi Nakamura (Department of Pediatrics, Kumamoto University Graduate School of Medical Sciences, Kumamoto, Japan), Takashi Miida (Department of Clinical Laboratory Medicine, Juntendo University Graduate School of Medicine, Tokyo, Japan), Masa-aki Kawashiri (Department of Cardiology, Kaga Medical Center, Ishikawa, Japan), Tetsuo Minamino (Department of Cardiorenal and Cerebrovascular Medicine, Faculty of Medicine, Kagawa University, Kagawa, Japan), Sachiko Okazaki (Division for Health Service Promotion, The University of Tokyo, Tokyo, Japan), Jun Wada (Department of Nephrology, Rheumatology, Endocrinology and Metabolism, Okayama University Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, Okayama, Japan), Masatsune Ogura (Department of Clinical Laboratory Technology, Faculty of Medical Science, Juntendo University, Chiba, Japan), Hiroshi Yoshida (Department of Laboratory Medicine, Jikei University Kashiwa Hospital, Chiba, Japan), Yu Kataoka (Department of Cardiovascular Medicine, National Cerebral and Cardiovascular Center, Osaka, Japan), Hirotoshi Ohmura (Department of Cardiovascular Medicine, School of Medicine, Juntendo University, Tokyo, Japan), Mika Hori (Department of Endocrinology, Research Institute of Environmental Medicine, Nagoya University, Nagoya, Japan), Kota Matsuki (Department of Endocrinology and Metabolism, Hirosaki University Graduate School of Medicine, Aomori, Japan), Takashi Yamaguchi (Center for Diabetes, Metabolism and Endocrinology, Toho University Medical Center Sakura Hospital, Chiba, Japan), Hirofumi Okada (Department of Cardiovascular Medicine, Kanazawa University Graduate School of Medical Sciences, Ishikawa, Japan), Masashi Yamamoto (Division of Diabetes, Metabolism and Endocrinology, Matsudo City General Hospital, Chiba, Japan), Yasuo Takeuchi (Division of Nephrology, Kitasato University School of Medicine, Ishikawa, Japan), Atsuko Nakatsuka (Department of Nephrology, Rheumatology, Endocrinology and Metabolism, Okayama University Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, Okayama, Japan), Daisaku Masuda (Department of Cardiology, Health Care Center, Rinku Innovation Center for Wellness Care and Activities (RICWA), Rinku General Medical Center, Osaka, Japan), Satoshi Hirayama (Department of Clinical Laboratory Medicine, Juntendo University Graduate School of Medicine, Tokyo, Japan), Masayuki Kuroda (Center for Advanced Medicine, Chiba University Hospital, Chiba University, Chiba, Japan), Manabu Minami (Department of Data Science, National Cerebral and Cardiovascular Center, Osaka, Japan), Hisashi Makino (Division of Diabetes and Lipid Metabolism, National Cerebral and Cardiovascular Center, Osaka, Japan), Masaki Matsubara (Department of General Medicine, Nara Medical University, Nara, Japan), Sayaka Funabashi (Department of Cardiovascular Medicine, Kyorin University School of Medicine, Tokyo, Japan), Keiko Goto (Department of Clinical Genetics, Juntendo University, Graduate School of Medicine, Tokyo, Japan), Masa-aki Hoshiga (Department of Cardiology, Osaka Medical and Pharmaceutical University, Osaka, Japan), Shinpei Fujioka (Department of Cardiology, Osaka Medical and Pharmaceutical University, Osaka, Japan), Manabu Takahashi (Division of Endocrinology and Metabolism, Department of Internal Medicine, Jichi Medical University, Tochigi, Japan), Kayoko Sato (Department of Cardiology, Tokyo Women’s Medical University, Tokyo, Japan), Daisuke Shishikura (Department of Cardiology, Osaka Medical and Pharmaceutical University, Osaka, Japan), Yasuaki Takeji (Department of Cardiovascular Medicine, Kanazawa University Graduate School of Medical Sciences, Ishikawa, Japan).

Introduction

Sitosterolemia is a rare autosomal recessive disorder characterized by elevated serum concentrations of plant sterols such as sitosterol, campesterol, and stigmasterol ^{1 - 3)} . Patients with sitosterolemia primarily present with tendinous and tuberous xanthomas, elevated serum low-density lipoprotein cholesterol (LDL-C) levels, and premature atherosclerotic cardiovascular diseases. These symptoms often overlap with other primary dyslipidemia such as familial hypercholesterolemia (FH) ⁴⁾ . In addition, patients with sitosterolemia have a wide variety of clinical presentations such as hemolysis, platelet abnormalities, splenomegaly, and arthralgia/ arthritis ⁵⁾ . Thus, it is not easy to make a differential clinical diagnosis of sitosterolemia just based on physical manifestations ⁶⁾ .

The molecular basis of sitosterolemia lies in pathogenic variants in the ATP-binding cassette transporter subfamily G member 5 (ABCG5) and ABCG8, which are located on chromosome 2p21. These genes encode for ABCG5 and ABCG8 transporters, which form heterodimers that play a key role in facilitating the excretion of dietary plant sterols and cholesterol into bile and the intestinal lumen ^{2 , 7)} . Loss-of-function variants in either ABCG5 or ABCG8 lead to impaired sterol excretion and increased absorption of plant sterols, resulting in their accumulation in plasma and tissues.

The measurement of serum sitosterol levels is useful for diagnosing sitosterolemia. However, due to its structural similarity to cholesterol, it requires analysis using gas chromatography-mass spectrometry (GC-MS) or liquid chromatography-mass spectrometry (LC-MS), which are difficult to routinely perform. In contrast, recently, advancements in genetic sequencing technologies have significantly improved our ability to identify disease-causing variants for various genetic disorders ⁸⁾ , and genetic analysis of sitosterolemia using next-generation sequencing (NGS) is also becoming increasingly widespread. Nevertheless, utilizing the results of genetic testing for clinical diagnosis poses a challenge, since it requires judgement whether each identified variant is pathogenic or not ⁹⁾ . This is especially true for sitosterolemia, where population-specific variants are likely, given the rarity of the disease and its underrepresentation in large genomic databases. Functional assays and familial segregation analyses are often necessary to establish the pathogenicity of candidate variants, but these approaches are labor-intensive and not always feasible. We have previously highlighted the importance of comprehensive genetic catalogs in guiding diagnosis, risk stratification, and management of FH ¹⁰⁾ . However, such efforts for sitosterolemia are currently lacking. Given that sitosterolemia is rare and often underdiagnosed, especially in specific populations, building a detailed catalog of pathogenic variants in ABCG5 and ABCG8 is critical to advancing clinical understanding and improving diagnostic accuracy. In this study, we aim to systematically catalog clinical features as well as pathogenic variants in ABCG5 and ABCG8 by comprehensively collecting information from previous case reports and evaluating their effects on proteins, assessing pathogenicity based on the American College of Medical Genetics and Genomics (ACMG) guidelines, and linking those variants to serum sitosterol levels.

Methods

1. Study Population

We retrospectively conducted a comprehensive literature search for case reports of sitosterolemia published between 2002 and 2024, in which mutational analyses in ABCG5 and ABCG8 were performed. Data on age, sex, nationality, and the presence of xanthoma, atherosclerotic disease, valvular heart disease, and hematologic abnormalities, as well as serum total cholesterol (T-Cho), LDL-C, and sitosterol levels under untreated conditions, were also collected from the literature. Serum T-Cho and LDL-C levels expressed in mmol/L were converted to mg/dL by multiplying by 38.67, while serum sitosterol levels expressed in µmol/L were converted to mg/dL by multiplying by 0.04147. We collected ethnicity information directly from case reports where such details were explicitly mentioned. For cases where ethnicity was not mentioned, we recorded the country of the authors’ affiliations instead.

In addition, we evaluated the pathogenicity of gene variants in the reviews on genetic variants reported by Lee et al. ¹¹⁾ and Heimerl et al. ¹²⁾ , in which the cases were diagnosed with sitosterolemia. We focused on 155 cases from 133 unrelated families who had nonsynonymous variants in ABCG5 or ABCG8 in this study.

2. Evaluation of the Pathogenicity of Genetic Variants

At first, we filtered out any variants whose minor allele frequency was 5% or greater. We defined nonsynonymous variants as those that result in any alterations to the amino acid sequence of a protein, as is commonly recognized. Effects on proteins were determined through the reference for the Franklin website developed by Genoox, where we can obtain multiple sources of information from a variety of similar sources ¹³⁾ . In addition, we referred to the ACMG guideline to determine whether the variants met the criteria for pathogenicity. A variant was considered pathogenic if it fulfilled any of the following criteria: 1) protein-truncating variants, 2) classified as pathogenic or likely pathogenic variants according to the ACMG guidelines, or 3) the serum sitosterol level of the case was measured at more than 1 mg/dL , which is a threshold that aligns with the diagnostic criteria for sitosterolemia in Japan ¹⁴⁾ . Otherwise, it was classified as a variant of unknown significance (VUS).

Results

Patient Demographics and Clinical Features

The characteristics of patients are shown in Fig.1 . Among the 155 patients, we identified the sex of 139 individuals, including 52 males (37.4%) and 87 females (62.6%) ( Fig.1A ) . The median age was 14 years old, with the first quartile (Q1) at 8 years old and the third quartile (Q3) at 26 years old ( Fig.1B ) . Among the cases with available data, xanthoma was observed in 99 (69.2%) of 143 cases. Ischemic heart disease was identified in 14 (14.2%) of 98 cases, while atherosclerotic change was observed in 13 cases (13.2%). Valvular heart disease was observed in 11 (14.2%) of 77 cases. Hematologic abnormalities were observed in 81 (57.9%) of 140 cases, including anemia with thrombocytopenia in 47 cases (33.6%), anemia alone in 10 cases (7.1%), and thrombocytopenia alone in 24 cases (17.1%) ( Fig.1C-1F ) . The median serum sitosterol level was 20.3 mg/dL (Q1: 8.2 mg/dL, Q3: 28.5 mg/dL). Meanwhile, the median serum T-Cho level was 354 mg/dL (Q1: 220mg/dL, Q3: 520mg/dL), and the median LDL-C level was 286 mg/dL (Q1: 188mg/dL, Q3: 404mg/dL), respectively. No correlation was observed between serum sitosterol levels and T-Cho or LDL-C levels ( Fig.1G ) .

Fig.1. Patient characteristics enrolled in this study.

A-B: Distribution of sex (A) and age (B). C-F: Proportion of patients with xanthoma (C), atherosclerosis (D), valvular heart disease (E), and hematologic abnormalities (F) comorbidity. G: Distribution and correlation of serum sitosterol, T-Cho and LDL-C levels.

Genotypic configurations and genetic variants in ABCG5 and ABCG8

Next, we summarize the genotype configurations identified in the 155 cases from 133 families. Among them, 47 were homozygous for ABCG5, and 59 were compound heterozygous for ABCG5, whereas 21 cases were homozygous for ABCG8, and 27 were compound heterozygous for ABCG8. Furthermore, 1 case was identified as a double heterozygote for both ABCG5 and ABCG8 ¹⁵⁾ ( Fig.2 ) . A total of 50 different variants in ABCG5 and 41 different variants in ABCG8 were identified among these patients. By incorporating variants in the reviews reported by Lee et al. ¹¹⁾ and Heimerl et al. ¹²⁾ , a total of 53 variants in ABCG5 and 52 variants in ABCG8 were evaluated for their pathogenicity.

Fig.2.

Genotype configurations identified in 155 cases of sitosterolemia, collected from the literature and analyzed in this study

Pathogenic variants in ABCG5 and ABCG8

The 53 variants in ABCG5 are illustrated in Table 1 and Fig.3A . Thirty-three of the 53 variants in ABCG5 were protein-truncating variants (including nonsense, insertion or deletion, frameshift, or splice-site variants) and were classified as pathogenic variants. Next, using the ACMG criteria, we identified 5 of the 20 missense variants in ABCG5 as pathogenic variants. Furthermore, among the remaining missense variants, those were considered pathogenic if the serum sitosterol levels of the cases carrying the variants were 1 mg/dL or higher. As a result, 12 of the 15 missense variants in ABCG5 were classified as pathogenic variants. Thus, 50 variants were classified as pathogenic, while 3 missense variants were designated as VUS ( Fig.3A ) .

Table 1. Variants in ABCG5 assessed for pathogenicity in this study.

Exon / Intron	Nucleotide change	Effect of protein	Mode of mutation	ACMG	Number of patients	Sitosterol (mg/dL)	Judgement
Exon 1	c.46C>T	Q16X	Nonsense	Pathogenic	1	2.4	Pathogenic
Exon 1	c.64C>T	Q22X	Nonsense	Pathogenic	7	33.1±14.1	Pathogenic
Exon 1	c.130C>T	S44A	Missense	VUS	1	8	Pathogenic
Exon 1	c.130T>C	S44P	Missense	VUS	1	6.5	Pathogenic
Intron 1	c.144-1G>A	IVS1-1G/A	Splicing	Likely Pathogenic	1	68.1	Pathogenic
Exon 2	c.161G>A	W54X	Nonsense	Likely Pathogenic	1	21.4	Pathogenic
Exon 2	c.229G>T	E77X	Nonsense	Pathogenic	4	37.1±17.9	Pathogenic
Exon 3	NA	exon 3 I/D	Large deletion	Pathogenic	1	22.3	Pathogenic
Exon 3	c.296T>G	M99R	Missense	VUS	1	11.6	Pathogenic
Exon 3	c.332G<A	G111E	Missense	Likely Pathogenic	1	NA	Pathogenic
Exon 3	c.335dupA	V113fs	Frameshift	Pathogenic	3	NA	Pathogenic
Exon 3	C336-337insA	C336-337insA	Frameshift	Pathogenic	NA	≥ 8.0 11)	Pathogenic
Exon 4	c.436G>C	E146Q	Missense	VUS	NA	≥ 8.0 11)	Pathogenic
Exon 4	c.436G>T	E146X	Nonsense	Likely Pathogenic	2	92.9±0.6	Pathogenic
Exon 6	c.727C>T	R243X	Nonsense	Pathogenic	3	NA	Pathogenic
Exon 6	c.751C>A	Q251X	Nonsense	Pathogenic	10	9.6±7.7	Pathogenic
Exon 7	c.850G>A	G269R	Missense	VUS	1	6.1	Pathogenic
Exon 7	NA	288X (19bp insert)	Frameshift	Pathogenic	2	24.5±3.0	Pathogenic
Intron 7	c.904+1G>A	IVS7+1G/A	Splicing	Likely Pathogenic	12	15.8±7.2	Pathogenic
Intron 7	c.904+2G>A	IVS7+2G/A	Splicing	Pathogenic	2	53.9±51.4	Pathogenic
Intron 7	c.904+5G>C	IVS7+5G/C	Splicing	VUS	1	18.8	Pathogenic
Exon 8	c.914C>G	T305R	Missense	Likely Pathogenic	1	27.7	Pathogenic
Exon 8	c.985T>C	Y329H	Missense	VUS	1	13.7	Pathogenic
Exon 8	c.987C>A	Y329X	Nonsense	Likely Pathogenic	1	NA	Pathogenic
Exon 8	c.1108_1118+2del	c.1108_1118+2del	Splicing	Likely Pathogenic	1	13.2	Pathogenic
Intron 8	c.1118+3A>C	IVS8+3A/C	Splicing	VUS	1	13.9	Pathogenic
Exon 9	c.1166G>A	R389H	Missense	Likely Pathogenic	23	10.1±7.8	Pathogenic
Exon 9	c.1217G>A	R406Q	Missense	VUS	1	NA	VUS
Exon 9	c.1222C>T	R408X	Nonsense	Pathogenic	4	31.1	Pathogenic
Exon 9	c.1255C<T	R419C	Missense	VUS	1	NA	VUS
Exon 9	c.1256G>A	R419H	Missense	VUS	8	36.6±30.1	Pathogenic
Exon 9	c.1256G>C	R419P	Missense	VUS	NA	≥ 8.0 11)	Pathogenic
Exon 9	c.1292C>T	P431L	Missense	VUS	1	19.4	Pathogenic
Exon 9	c.1311C>G	N437K	Missense	Likely Pathogenic	2	7.4	Pathogenic
Intron 9	c.1324+2A>G	IVS9+2A/G	Splicing	Pathogenic	4	41.0±17.7	Pathogenic
Intron 9	c.1324+1delG	IVS9+1delG	Splicing	Likely Pathogenic	4	NA	Pathogenic
Intron 9	c.1325-3_	IVS9-3_	Splicing	Pathogenic	2	12.5	Pathogenic
	c.1325-2delCA	IVS9-2/delCA
Exon 10	c.1336C>T	R446X	Nonsense	Pathogenic	28	20.5±15.7	Pathogenic
Exon 10	c.1337G>A	R446Q	Missense	Likely Pathogenic	2	3.3	Pathogenic
Exon 10	c.1375C>T	Q459X	Nonsense	Likely Pathogenic	1	24.3	Pathogenic
Intron 10	c.1464-1G>T	IVS10-1G/T	Splicing	VUS	2	13.1±10.5	Pathogenic
Exon 11	c.1523delC	c.1523delC	Frameshift	Pathogenic	1	68.1	Pathogenic
Exon 11	c.1528C<A	H510N	Missense	Likely Benign	3	16.6±7.6	Pathogenic
Exon 11	c.1528C>G	H510D	Missense	VUS	1	NA	VUS
Exon 11	c.1621G>A	G541R	Missense	VUS	1	13.7	Pathogenic
Intron 11	c.1649+3insT	IVS11+3insT	Splicing	Pathogenic	2	92.9±0.6	Pathogenic
Exon 12	c.1650A>C	R550S	Missense	VUS	1	24.3	Pathogenic
Exon 12	c.1673_1677delCTTTT	P558Qfs＊14	Frameshift	Likely Pathogenic	4	9.5±4.3	Pathogenic
Intron 12	c.1762+1G>A	IVS12+1G/A	Splicing	Likely Pathogenic	1	95.4	Pathogenic
Intron 12	c.1763-1G>A	IVS12-1G/A	Splicing	Likely Pathogenic	1	18.2	Pathogenic
Exon 13	c.1769C>A	S590X	Nonsense	Likely Pathogenic	1	19.7	Pathogenic
Exon 13	c.1890delT	F630Lfs＊8	Frameshift	Pathogenic	1	27.7	Pathogenic
Exon 13	c.1915delA	I639X	Nonsense	Likely Pathogenic	1	4.2	Pathogenic

Fig.3. Variant filtering of this study.

A total of (A) 53 variants in ABCG5 and (B) 52 variants in ABCG8 were defined as pathogenic if they fulfilled either of criteria 1) protein truncated variants (including, nonsense, insertions or deletions, frameshift, or splice-site variant), 2) classified as pathogenic or likely pathogenic variants according to the ACMG criteria, or 3) the serum sitosterol level of the case was measured at more than 1 mg/dL, otherwise, it was classified as VUS.

ACMG, American College of Medical Genetics and Genomics

VUS, variant of unknown significance

Regarding the 50 pathogenic variants in ABCG5, the three most frequent variants, which are R446X (N = 28), R389H (N = 23), and IVS7+1G/A (N = 12), accounted for more than one-third of the total variants ( Fig.4A ) . The majority of cases with these three variants were observed in East Asian cases ( Fig.4B ) .

Fig.4. Distribution of variants in ABCG5 among 107 cases of sitosterolemia collected from the literature and analyzed in this study .

(A) Summary of 50 distinct types and a total of 162 variants identified among 107 cases. (B) Ethnic characteristics of the three most frequent variants (R446X, R389H, and IVS7+1G>A) in ABCG5.

We also evaluated the pathogenicity of 52 variants in ABCG8 ( Table 2 and Fig.3B ) . Twenty-nine of the 53 variants were protein-truncating variants and were classified as pathogenic. Moreover, using the ACMG criteria, we identified 5 of the 23 missense variants as pathogenic variants. Furthermore, according to the serum sitosterol levels, 17 of the 18 missense variants in ABCG8 were classified as pathogenic. As a result, 51 variants were classified as pathogenic, while 1 missense variant was designated as VUS ( Fig.3B ) .

Table 2. Variants in ABCG8 assessed for pathogenicity in this study .

Exon / Intron	Nucleotide change	Effect of protein	Mode of mutation	ACMG	Number of patients	Sitosterol (mg/dL)	Judgement
Intron 1	c.64-2A>G	IVS1-2A/G	Splicing	Likely Pathogenic	NA	NA	Pathogenic
Exon 3	c.225G>A	W75X	Nonsense	Likely Pathogenic	1	22.4	Pathogenic
Exon 3	g.12283-4insC	L86PfsTer185	Frameshift	Pathogenic	1	67	Pathogenic
Exon 3	c.320C>G	S107X	Nonsense	Pathogenic	5	29.3±12.9	Pathogenic
Exon 4	c.361C>T	R121X	Nonsense	Pathogenic	NA	≥ 8.0 11)	Pathogenic
Exon 4	c.490C>T	R164X	Nonsense	Pathogenic	3	23.4±7.9	Pathogenic
Exon 4	c.514C>T	Q172X	Nonsense	Likely Pathogenic	NA	≥ 8.0 11)	Pathogenic
Exon 4	c.547del	Q183Sfs＊9	Frameshift	Pathogenic	NA	≥ 8.0 11)	Pathogenic
Exon 4	c.551G>A	R184H	Missense	VUS	NA	≥ 8.0 11)	Pathogenic
Exon 5	c.584T>A	L195Q	Missense	Likely Pathogenic	1	20.3	Pathogenic
Exon 5	c.683T>C	L228P	Missense	VUS	1	8.8	Pathogenic
Exon 5	c.691C>A	P231T	Missense	VUS	NA	≥ 8.0 11)	Pathogenic
Intron 5	c.694+5G>C	IVS5+5G/C	Splicing	VUS	1	4.6	Pathogenic
Exon 6	c.788G>A	R263Q	Missense	Likely Pathogenic	3	40.6±41.1	Pathogenic
Exon 6	c.811C>T	Q271X	Nonsense	Pathogenic	2	97.0±14.1	Pathogenic
Exon 6	c.904C>T	Q302X	Nonsense	Pathogenic	1	NA	Pathogenic
Intron 6	c.965-1G>C	IVS6-1G/C	Splicing	Pathogenic	2	NA	Pathogenic
Intron 6	c.965-1G>A	IVS6-1G/A	Splicing	Pathogenic	1	1.3	Pathogenic
Exon 7	c.1083G>A	W361X	Nonsense	Pathogenic	10	21.9±36.2	Pathogenic
Intron 8	c.1212-1G>A	IVS8-1G/A	Splicing	Likely Pathogenic	2	97.0±14.1	Pathogenic
Exon 9	c.1214G>A	R405H	Missense	Likely Pathogenic	3	29.5±15.0	Pathogenic
Exon 9	c.1225A>G	N409D	Missense	VUS	1	5.4	Pathogenic
Exon 9	c.1234C>T	R412X	Nonsense	Pathogenic	1	22.1	Pathogenic
Exon 9	c.1256T>A	I419N	Missense	VUS	2	13.5±12.6	Pathogenic
Exon 9	c.1267G>A	E423K	Missense	VUS	1	7.4	Pathogenic
Exon 9	c.1269G>T	E423D	Missense	VUS	1	12.9	Pathogenic
Exon 9	c.1285A>C	M429L	Missense	VUS	1	1.6	Pathogenic
Exon 9	c.1405_1406del	S469Qfs＊12	Frameshift	Pathogenic	1	17.6	Pathogenic
Exon 10	c.1418C>G	S473X	Nonsense	Likely Pathogenic	1	1.3	Pathogenic
Exon 10	c.1444del	L482Wfs＊40	Frameshift	Pathogenic	1	19.7	Pathogenic
Exon 10	c.1476T>A	Y492X	Nonsense	Likely Pathogenic	2	15.1	Pathogenic
Exon 10	c.1487insA	I497Dfs＊105	Frameshift	Likely Pathogenic	1	7.4	Pathogenic
Exon 11	g.41193-202 ins	E500DfsTer604	Frameshift	Pathogenic	1	87.2	Pathogenic
Exon 11	c.1502T>C	L501P	Missense	VUS	NA	≥ 8.0 11)	Pathogenic
Exon 11	c.1528_1530delATC	I510del	Small deletion	VUS	1	9.6	Pathogenic
Exon 11	c.1534G>A	G512R	Missense	VUS	1	1.2	Pathogenic
Exon 11	c.1608G>A	W536X	Nonsense	Likely Pathogenic	1	115.9	Pathogenic
Exon 11	c.1629G>C	R543S	Missense	VUS	NA	≥ 8.0 11)	Pathogenic
Exon 11	c.1640T>C	L547P	Missense	VUS	1	19.7	Pathogenic
Exon 11	c.1709_1711del	F570del	Small deletion	Likely Pathogenic	2	37.2±9.5	Pathogenic
Exon 11	c.1715T>C	L572P	Missense	VUS	4	1.5±1.2	Pathogenic
Exon 11	c.1720G>A	G574R	Missense	Pathogenic	3	28.0±14.8	Pathogenic
Exon 11	c.1721G>A	G574E	Missense	Likely Pathogenic	NA	≥ 8.0 11)	Pathogenic
Exon 11	c.1724G>A	G575D	Missense	VUS	1	8.8	Pathogenic
Intron 11	c.1756+2T>A	IVS11+2T/A	Splicing	Pathogenic	1	5.4	Pathogenic
Exon 12	c.1762G>C	A588P	Missense	VUS	2	4.3±0.5	Pathogenic
Exon 12	c.1787T>G	L596R	Missense	VUS	NA	≥ 8.0 11)	Pathogenic
Exon 12	g.43683-727del	M614-K628del	small deletion	Pathogenic	1	78.3	Pathogenic
Exon 12	c.1877G>T	G626V	Missense	VUS	1	NA	VUS
Exon 13	g.43866-7indelsT	V642SfsTer662	Frameshift	Pathogenic	1	78.3	Pathogenic
Exon 13	c.1949T>G	L650R	Missense	VUS	1	24.1	Pathogenic
Exon 13	c.1974C>A	Y658X	Nonsense	Likely Pathogenic	NA	≥ 8.0 11)	Pathogenic

Among these pathogenic variants in ABCG8, the three most frequent variants were W361X (N = 10), S107X (N = 5), and L572P (N = 4) ( Fig.5A ) . In contrast to the variants in ABCG5, these pathogenic variants in ABCG8 were predominantly observed in cases of European descent ( Fig.5B ) . These results highlight the geographical and ethnic differences in the distribution of pathogenic variants in ABCG5 and ABCG8.

Fig.5. Distribution of variants in ABCG8 among 49 cases of sitosterolemia collected from the literature and analyzed in this study .

(A) Summary of 41 distinct types and a total of 72 variants identified among 49 cases. (B) Ethnic characteristics of the three most frequent variants (W361X, S107X, and L572P) in ABCG8.

The diagrams illustrate the location and type of variants responsible for sitosterolemia in ABCG5 ( Fig.6A ) and ABCG8 ( Fig.6B ) . Additionally, Table 3 comprehensively demonstrates the information collected for each case.

Fig.6. Diagram depicting the ABCG5 (A) and ABCG8 (B) genes, illustrating the location and type of variants responsible for sitosterolemia .

Exons are represented as vertical boxes, while introns are indicated by the connecting lines.

Table 3. Patients with sitosterolemia enrolled in this study.

Case No	age	sex	nationality/ ethnicity	Xanthoma	Atherosclerosis, Ischemic heart disease	Valvular heart disease	Hematologic abnor malities	Sitosterol (mg/dL)	T-Cho (mg/ dL)	LDL-C (mg/ dL)	ABCG5 variant1	ABCG5 variant2	ABCG8 variant1	ABCG8 variant2	Reference
1	NA	NA	German	NA	2	NA	NA	31.1	305.0	246.0	c.1222C>T (R408X)	NA			12)
2	NA	NA	Caucasian	NA	NA	NA	NA	20.3	210.0	129.0			c.584T>A (L195Q)	NA	12)
3	NA	NA	Caucasian	NA	NA	NA	NA	22.1	218.0	166.0			c.1083G>A (W361X)	c.1234C>T (R412X)	12)
4	NA	NA	Caucasian	NA	NA	NA	NA	20.5	247.0	155.0			c.1083G>A (W361X)	homo	12)
5	NA	NA	Swiss	NA	NA	NA	NA	17.5	214.0	145.0			c.1720G>A (G574R)	homo	12)
6	7	M	Chinese	1	0	0	NA	95.4	709.0	636.0	c.1256G>A (R419H)	c.1762+1G>A (IVS12+1G/A)			18 - 20)
7A	5	F	Canadian	1	1	1	NA	NA	483.4	NA			c.320C>G (S107X)	homo	21)
8A	4	F	Canadian	0	NA	NA	NA	28.2	417.6	348.4			c.320C>G (S107X)	homo	21)
9A	3	M	Canadian	0	NA	NA	NA	47.7	456.3	390.6			c.320C>G (S107X)	homo	21)
10A	8	F	Canadian	0	NA	NA	NA	18.4	357.3	288.1			c.320C>G (S107X)	homo	21)
11	75	F	Japanese/ Canadian	1	2	1	NA	22.3	278.4	193.4	NA (exon 3 I/ D)	homo			21)
12B	15	F	Asian	0	NA	NA	3	61.0	197.2	NA	c.229G>T (E77X)	homo			5 , 22)
13B	12	M	Asian	0	NA	NA	3	40.2	143.1	NA	c.229G>T (E77X)	homo			5 , 22)
14B	11	M	Asian	0	NA	NA	3	25.9	NA	NA	c.229G>T (E77X)	homo			5 , 22)
15B	20	F	Asian	0	NA	0	3	21.1	139.2	NA	c.229G>T (E77X)	homo			5 , 22)
16C	11	M	Caucasian	0	NA	NA	3	92.5	193.4	NA	c.1649+3insT (IVS11+3insT)	c.436G>T (E146X)			22)
17C	9	M	Caucasian	0	NA	NA	3	93.3	255.2	NA	c.1649+3insT (IVS11+3insT)	c.436G>T (E146X)			22)
18D	29	M	Caucasian	0	NA	NA	2	87.1	158.5	NA			c.811C>T (Q271X)	c.1212-1G>A (IVS8-1G/A)	22)
19D	27	F	Caucasian	0	NA	0	2	107.0	162.4	NA			c.811C>T (Q271X)	c.1212-1G>A (IVS8-1G/A)	22)
20	14	F	Caucasian	0	NA	NA	1	29.7	228.2	NA			c.1083G>A (W361X)	homo	22)
21	26	F	Caucasian	1	NA	NA	0	76.5	212.7	NA			c.1083G>A (W361X)	homo	5 , 22)
22	19	M	Finn	0	0	0	2	12.9	NA	65.7			c.1083G>A (W361X)	c.1269G>T (E423D)	23)
23E	25	NA	Asian	1	0	NA	3	27.6	205.7	154.7	c.64C>T (Q22X)	homo			24)
24E	24	NA	Asian	1	0	NA	3	48.5	220.0	150.8	c.64C>T (Q22X)	homo			24)
25E	23	NA	Asian	1	0	NA	3	36.0	134.6	88.9	c.64C>T (Q22X)	homo			5 , 24)
26	10	F	Iranian	1	0	1	0	15.8	380.0	270.0	c.1336C>T (R446X)	homo			19 , 25)
27	56	F	Japanese	1	NA	NA	2	27.5	389.0	NA	NA (288X(19bp insert))	homo			26)
28	33	F	Japanese	1	NA	NA	3	21.4	414.0	NA	NA (288X(19bp insert))	c.1336C>T (R446X)			26)
29	23	F	Japanese	1	2	NA	1	24.3	458.0	NA	c.1650A>C (R550S)	c.1256G>A (R419H)			19 , 27)
30	18	M	Japanese	1	1	0	0	8.9	353.0	NA	c.1166G>A (R389H)	c.1810C>G (Q604E)			19 , 27)
31	11m	F	Romanian	1	NA	NA	NA	2.4	1023.0	837.0	c.46C>T (Q16X)	c.1336C>T (R446X)			19 , 28 , 29)
32	8	F	Chinese	1	0	0	1	NA	427.0	346.0	c.987C>A (Y329X)	c.1311C>G (N437K)			19 , 20 , 29 , 30)
33	18m	F	Chinese	1	0	0	0	7.1	705.0	565.0	c.1166G>A (R389H)	c.1336C>T (R446X)			19 , 20 , 29 , 30)
34	3m	F	Chinese	0	0	0	0	9.2	402.0	304.0	c.1166G>A (R389H)	c.1336C>T (R446X)			19 , 20 , 29 , 30)
35	23m	F	Chinese	1	0	0	0	7.1	640.0	519.0	c.1166G>A (R389H)	homo			19 , 20 , 29 , 30)
36	12	F	Chinese	1	0	0	0	6.1	343.0	263.0	c.1166G>A (R389H)	c.850G>A (G269R)			19 , 20 , 29 , 30)
37	31	M	Asian	0	0	0	3	33.1	315.9	209.2	c.1256G>A (R419H)	homo			5 , 31)
38F	32	F	French	1	0	0	2	30.5	470.0	NA			c.1214G>A (R405H)	c.1709_1711del (F570del)	32)
39F	27	F	French	1	0	0	0	43.9	688.0	617.0			c.1214G>A (R405H)	c.1709_1711del (F570del)	32)
40	33	F	French	1	0	0	2	24.9	805.0	NA			c.788G>A (R263Q)	NA (D160T)	32)
41	17	F	French	1	0	0	0	15.1	724.0	630.0			c.490C>T (R164X)	c.1476T>A (Y492X)	32)
42	19	M	French	1	0	0	0	14.0	409.0	372.0			c.1214G>A (R405H)	c.1083G>A (W361X)	32)
43	59	F	French	1	2	0	1	NA	336.0	255.2			c.904C>T (Q302X)	homo	33)
44	15	F	Chinese	1	0	0	0	24.1	290.0	177.9			c.490C>T (R164X)	c.1949T>G (L650R)	20 , 34)
45	15m	F	Korean	1	NA	NA	0	19.4	675.0	540.0	c.904+1G>A (IVS7+1G/A)	c.1336C>T (R446X)			19 , 29 , 35)
46	12	M	Turkish	0	0	NA	3	20.5	NA	NA	c.1464-1G>T (IVS10-1G/T)	homo			5 , 36)
47G	25	F	Chinese	1	1	0	3	27.6	205.7	154.7	c.64C>T (Q22X)	homo			20 , 37)
48G	24	F	Chinese	1	0	0	3	48.5	220.0	150.8	c.64C>T (Q22X)	homo			20 , 37)
49G	23	M	Chinese	1	0	0	3	36.0	134.6	88.9	c.64C>T (Q22X)	homo			20 , 37)
50	34	F	Chinese	1	0	0	3	57.5	220.0	150.8	c.1336C>T (R446X)	homo			20 , 37)
51	43	F	Chinese	1	2	0	3	78.3	338.7	235.5			g.43683-727del (M614-K628del)	g.43866-7indelsT (V642SfsTer662)	20 , 37)
52H	61	M	Chinese	0	0	0	3	48.9	331.8	220.0	c.1324+2A>G (IVS9+2A/G)	c.1336C>T (R446X)			20 , 37)
53H	58	M	Chinese	1	0	0	3	29.6	143.9	65.0	c.1324+2A>G (IVS9+2A/G)	c.1336C>T (R446X)			20 , 37)
54H	57	M	Chinese	1	0	0	3	23.5	343.8	223.9	c.1324+2A>G (IVS9+2A/G)	c.1336C>T (R446X)			20 , 37)
55H	53	F	Chinese	1	0	0	3	61.9	338.7	235.5	c.1324+2A>G (IVS9+2A/G)	c.1336C>T (R446X)			20 , 37)
56	31	M	Chinese	0	0	0	2	68.9	315.9	198.8	c.1256G>A (R419H)	homo			20 , 37)
57	58	M	Chinese	1	0	0	3	67.0	127.6	109.0			g.12283-4insC (L86PfsTer185)	homo	20 , 37)
58	45	F	Chinese	1	0	0	3	90.2	227.8	108.7	c.904+2G>A (IVS7+2G/A)	homo			20 , 37)
59	59	F	Chinese	0	0	0	2	87.2	213.1	128.4			c.788G>A (R263Q)	g.41193-202 ins (E500DfsTer604)	20 , 37)
60	2	F	Spanish	1	0	0	3	38.4	623.0	570.0			c.1720G>A (G574R)	homo	38)
61I	18m	F	Japanese	1	NA	NA	0	10.1	523.0	407.0	c.1166G>A (R389H)	c.1673_1677delCTTTT (P558Qfs＊14)			19 , 29 , 39)
62I	10m	F	Japanese	0	NA	NA	0	3.3	555.0	NA	c.1166G>A (R389H)	c.1673_1677delCTTTT (P558Qfs＊14)			19 , 29 , 39)
63	13m	F	Japanese	1	NA	NA	0	8.0	870.0	796.0	c.1166G>A (R389H)	c.130C>T (S44A)			19 , 29 , 39)
64	2	M	Japanese	1	NA	NA	0	11.5	756.0	589.0	c.1336C>T (R446X)	c.1673_1677delCTTTT (P558Qfs＊14)			19 , 29 , 39)
65	10	F	Mexican	1	NA	NA	3	68.1	295.0	224.0	c.144-1G>A (IVS1-1G/A)	c.1523delC			19 , 29 , 40)
66	16m	F	Japanese	1	NA	NA	0	24.6	866.0	679.0	c.1166G>A (R389H)	c.1256G>A (R419H)			19 , 41)
67	14	F	Turkish	0	NA	NA	0	20.2	300.0	219.0	c.1336C>T (R446X)	homo			19 , 42)
68	13	F	Japanese	1	1	1	0	18.2	681.0	609.0	c.1256G>A (R419H)	c.1763-1G>A (IVS12-1G>A)			29 , 43)
69	46	F	Spanish	1	NA	NA	2	27.7	104.4	NA	c.914C>G (T305R)	c.1890delT (F630Lfs＊8)			5)
70	2	F	Italian/ Chinese	1	0	0	0	7.9	770.0	690.0	c.751C>A (Q251X)	c.1336C>T (R446X)			19 , 29 , 44)
71	7	M	Italian	0	0	0	0	22.8	598.0	528.0			c.320C>G (S107X)	c.1083G>A (W361X)	44)
72J	20s	M	Tunisian	1	NA	NA	0	NA	WNL	WNL			c.965-1G>C (IVS6-1G/C)	homo	45)
73J	20s	F	Tunisian	1	NA	NA	0	NA	WNL	WNL			c.965-1G>C (IVS6-1G/C)	homo	45)
74	5	F	Japanese	1	0	0	0	9.5	868.0	660.0	c.1166G>A (R389H)	homo			19 , 29 , 46)
75	16	M	Japanese	1	1	NA	0	7.2	155.0	90.0	c.1166G>A (R389H)	homo			19 , 47)
76	21	M	Pakistani	1	2	0	3	17.5	152.0	NA	c.1336C>T (R446X)	homo			29 , 48)
77	18m	F	Japanese	1	NA	NA	0	24.6	866.0	679.0	c.1166G>A (R389H)	c.1256G>A (R419H)			19 , 29 , 49)
78	19	M	north Indian	0	0	0	3	NA	NA	NA	c.727C>T (R243X)	homo			50)
79	13	M	American	1	0	0	0	5.4	742.0	679.0			c.1225A>G (N409D)	c.1756+2T>A (IVS11+2T/A)	51)
80	25	F	Japanese	1	2	0	NA	4.6	NA	220.0			c.1256T>A (I419N)	c.694+5G>C (IVS5+5G/C)	52)
81	16m	F	Chinese	1	NA	NA	1	30.9	791.6	568.1			c.490C>T (R164X)	c.1707-1709delTTC	53)
82	6	M	Chinese	1	0	0	0	9.6	559.0	349.0			c.1528_1530delATC (I510del)	c.788G>A (R263Q)	20 , 54)
83	5	F	Chinese	1	0	0	0	7.4	599.0	524.0	c.1311C>G (N437K)	c.751C>T (Q251X)			20 , 54)
84	16m	F	Chinese	1	0	0	0	4.7	613.0	483.0	c.1336C>T (R446X)	c.751C>T (Q251X)			20 , 54)
85	1	F	Japanese	1	NA	NA	NA	1.6	540.0	453.0	c.1166G>A (R389H)		c.1285A>C (M429L)		15 , 29)
86	33	F	German	1	0	0	0	7.4	327.0	221.0			c.1267G>A (E423K)	c.1487insA (I497Dfs＊105)	55)
87	14	M	Chinese	1	2	1	3	21.4	412.0	332.0	c.1336C>T (R446X)	homo			20 , 29 , 56)
88	10	F	Chinese	0	0	0	3	4.2	240.9	NA	c.1915delA (I639X)	c.1336C>T (R446X)			57)
89	14	M	Chinese	0	0	0	3	13.9	123.0	NA	c.1118+3A>C (IVS8+3A/C)	c.904+1G>A (IVS7+1G/A)			57)
90	11	F	Chinese	0	0	0	3	8.8	158.2	NA			c.683T>C (L228P)	c.1724G>A (G575D)	57)
91	15	F	Chinese	0	0	0	3	23.2	269.1	NA	c.751G>A (Q251X)	homo			57)
92	12	F	Chinese	1	0	0	0	21.0	543.0	395.0	c.904+1G>A (IVS7+1G/A)	c.1528C<A (H510N)			58)
93	9	M	Chinese	1	0	0	0	NA	557.0	430.0	c.1166G>A (R389H)	homo			19 , 20)
94	7	M	Chinese	1	0	0	0	NA	478.0	359.0	c.751C>A (Q251X)	homo			19 , 20)
95K	11	F	Bosnian	1	0	1	0	23.0	467.9	394.4	c.1336C>T (R446X)	homo			59)
96K	7	F	Bosnian	0	0	0	0	13.8	315.9	239.8	c.1336C>T (R446X)	homo			59)
97	67	F	Chinese	1	2	1	2	28.4	183.0	NA	c.1166G>A (R389H)	G90E			20 , 60)
98	45	M	Chinese	1	0	0	0	17.5	153.0	NA	c.904+2G>A (IVS7+2G/A)	homo			20 , 60)
99	44	F	Chinese	1	0	0	0	28.5	213.0	NA	c.1336C>T (R446X)	homo			20 , 60)
100	39	F	Chinese	1	0	0	0	NA	NA	NA	c.1166G>A (R389H)	homo			20 , 60)
101	8	M	Chinese	0	0	0	3	1.3	238.6	145.8			c.965-1G>A (IVS6-1G/A)	c.1418C>G (S473X)	20 , 61)
102	60	M	Japanese	1	2	NA	2	13.2	354.0	270.0	c.1108_1118+ 2del	c.1673_1677delCTTTT (P558Qfs＊14)			16)
103	18	F	Korean	0	0	NA	0	19.4	340.7	241.3	c.1292C>T (P431L)	c.1336C>T (R446X)			62)
104	7	F	Korean	0	0	NA	0	17.6	288.5	209.2			c.1405_1406del (S469Qfs＊12)	homo	62)
105	6	F	Japanese	1	1	1	NA	22.4	423.0	370.0			c.225G>A (W75X)	c.1256T>A (I419N)	63)
106	8	F	Australian	1	0	0	0	19.7	363.5	286.2			c.1444del (L482Wfs＊40)	c.1640T>C (L547P)	64)
107	34	F	Thai	0	0	NA	2	NA	NA	185.0	c.1217G>A (R406Q)	c.751C>A (Q251X)			65)
108	15	F	Indian	0	NA	NA	3	NA	WNL	WNL	c.727C>T (R243X)	homo			66)
109	23	M	Indian	0	NA	NA	3	NA	WNL	WNL	c.1222C>T (R408X)	c.1255C<T (R419C)			66)
110	29	F	Indian	0	NA	NA	3	NA	WNL	WNL	c.332G<A (G111E)	homo			66)
111	7	M	Indian	0	NA	NA	3	NA	WNL	WNL	c.1222C>T (R408X)	homo			66)
112	6	M	Belgian	0	NA	NA	3	NA	NA	NA	c.1222C>T (R408X)	homo			67)
113	21	F	Danish	1	0	NA	2	27.0	379.0	294.0	c.1336C>T (R446X)	homo			68)
114	17	M	Turkish	NA	NA	NA	3	5.7	181.0	119.0	c.1464-1G>T (IVS10-1G/T)	homo			69)
115	35	M	Turkish	NA	NA	NA	2	4.0	180.0	115.0			c.1762G>C (A588P)	homo	69)
116	12	M	Turkish	NA	NA	NA	2	4.7	162.0	95.0			c.1762G>C (A588P)	homo	69)
117	45	M	Turkish	NA	NA	NA	2	4.0	177.0	113.0	c.1336C>T (R446X)	homo			69)
118	10	M	Turkish	NA	NA	NA	2	8.2	204.0	139.0			c.1895T>C (V632A)	homo	69)
119	16	M	Turkish	NA	NA	NA	3	21.4	258.0	190.0	c.161G>A (W54X)	homo			69)
120	20	M	Turkish	NA	NA	NA	2	10.0	188.0	122.0	c.1375C>T (Q459X)	homo			69)
121	26	F	Chinese	1	1	NA	3	NA	323.3	271.5	c.1336C>T (R446X)	c.1325-3_c.1325- 2delCA (IVS9-3_IVS9-2/delCA)			70)
122	15	M	Japanese	1	2	0	0	5.3	NA	180.0	c.1166G> A (R389H)	homo			71)
123	17	M	Japanese	1	2	1	2	4.0	440.0	379.0	c.1256G>A (R419H)	c.904+1G>A (IVS7+1G/A)			29)
124	6	F	Chinese	0	1	NA	0	3.3	638.0	527.0	c.1337G>A (R446Q)	c.1118+3G>T			72)
125	26	F	Chinese	1	0	0	3	12.5	382.8	309.4	c.1336C>T (R446X)	c.1325-3_c.1325- 2delCA (IVS9-3_IVS9-2/delCA)			73)
126	9	F	Chinese	1	0	0	0	21.0	543.0	395.0	c.904+1G>A (IVS7+1G/A)	c.1528C<A (H510N)			20)
127	4	F	Chinese	0	0	0	0	6.5	183.0	146.0	c.130T>C (S44P)	c.1166G>A (R389H)			20)
128	8	F	Chinese	0	0	0	0	7.8	398.0	315.0	c.64C>T (Q22X)	c.1528C<A (H510N)			20)
129	2	M	Chinese	1	0	0	0	5.0	329.0	296.0	c.751C>A (Q251X)	c.1166G>A (R389H)			20)
130	41	F	Russian	1	2	NA	2	2.3	391.3	340.3			c.1083G>A (W361X)	homo	74)
131	10	M	Turkish	0	NA	NA	3	18.8	130.0	NA	c.904+5G>C (IVS7+5G/C)	homo			75)
132	7	M	Chinese	1	0	NA	0	11.6	588.6	513.5	c.296T>G (M99R)	c.76C<T			76)
133	25	F	Chinese	0	2	NA	0	13.7	375.1	309.4	c.985T>C (Y329H)	c.1621G>A (G541R)			77)
134L	9	M	Chinese	1	1	0	0	NA	585.1	542.9	c.904+1G>A (IVS7+1G/A)	c.1324+1delG (IVS9+1delG)			78)
135L	7	F	Chinese	1	1	0	0	NA	271.1	152.4	c.904+1G>A (IVS7+1G/A)	c.1324+1delG (IVS9+1delG)			78)
136	10	F	Spanish	1	0	0	2	115.9	411.0	252.0			c.1083G>A (W361X)	c.1608G>A (W536X)	79)
137	11	NA	Chinese	1	NA	NA	1	NA	511.6	285.8	c.335dupA (V113fs)	c.751C>A (Q251X)			80)
138	4m	NA	Chinese	1	NA	NA	1	NA	363.5	266.4	c.1337G>A (R446Q)	c.335dupA (V113fs)			80)
139	8	NA	Chinese	1	NA	NA	0	NA	371.2	259.1	c.1336C>T (R446X)	c.1166G>A (R389H)			80)
140M	8	M	Chinese	1	1	NA	1	NA	560.3	500.0	c.904+1G>A (IVS7+1G/A)	c.1324+1delG (IVS9+1delG)			80 , 81)
141M	6	F	Chinese	1	NA	NA	0	NA	301.6	211.1	c.904+1G>A (IVS7+1G/A)	c.1324+1delG (IVS9+1delG)			80)
142	11	NA	Chinese	1	NA	NA	1	NA	460.2	332.6	c.751C>A (Q251X)	c.1528C>G (H510D)			80)
143	15m	NA	Chinese	1	NA	NA	1	NA	580.8	473.3	c1336C>T (R446X)	c.1166G>A (R389H)			80)
144	3	NA	Chinese	1	NA	NA	0	NA	556.5	492.7			c.1720G>A (G574R)	c.1877G>T (G626V)	80)
145	7	NA	Chinese	1	NA	NA	0	NA	580.1	400.2	c.904+1G>A (IVS7+1G/A)	homo			80)
146	10	NA	Chinese	1	NA	NA	0	NA	437.0	357.7	c.751C>A (Q251X)	c.335dupA (V113fs)			80)
147	16	F	Indian	0	0	NA	3	NA	NA	NA	c.727C>T (R243X)	homo			82)
148	30	M	Saudi Arabian	1	0	0	3	NA	267.2	194.1			c.1715T>C (L572P)	homo	83)
149	32	F	Chinese	1	1	1	3	19.7	326.0	203.0	c.1769C>A (S590X)	homo			84)
150N	8	F	Chinese	1	0	NA	2	NA	358.5	290.8	c.904+1G>A (IVS7+1G/A)	c.1166G>A (R389H)			85)
151N	8	F	Chinese	1	0	NA	2	NA	392.5	305.9	c.904+1G>A (IVS7+1G/A)	c.1166G>A (R389H)			85)
152	8	F	Russian	1	0	NA	0	0.6	425.4	371.2			c.1715T>C (L572P)	homo	86)
153	6	F	Russian	0	0	NA	0	1.2	487.2	371.2			c.1715T>C (L572P)	c.1534G>A (G512R)	86)
154	14	M	Russian	1	1	1	2	2.9	464.0	348.0			c.1715T>C (L572P)	c.1083G>A (W361X)	86)
155	48	F	Brazilian	1	NA	NA	3	NA	265.9	210.8			c.1476T>A (Y492X)	homo	87)

In the Case No column, the letters A to N denote the same pedigree.

In the Xanthoma column, 0 denotes “without xanthoma” and 1 denotes “with xanthoma”.

In the Atherosclerosis and Ischemic Heart Disease columns, 0 denotes “without lesion”, 1 denotes “with atherosclerotic change”, and 2 denotes “with ischemic heart disease”.

In the Valvular heart disease column, 0 denotes “without Valvular heart disease” and 1 denotes “with Valvular heart disease”.

In the Hematologic abnormalities column, 0 denotes “without hematologic abnormalities”, 1 denotes “with anemia”, 2 denotes “with thrombocytopenia”, and 3 denotes “with anemia and thrombocytopenia”

Discussion

Measurement of sitosterol concentration is useful for diagnosing sitosterolemia. However, it requires GC-MS or LC-MS, and it is not straightforward. Therefore, genetic testing for sitosterolemia plays a vital role in enhancing diagnostic precision and guiding effective therapeutic strategies. Identifying pathogenic variants in ABCG5 and ABCG8 also allows for a definitive diagnosis of sitosterolemia, setting it apart from phenotypically similar disorders such as FH ¹⁶⁾ and cerebrotendinous xanthomatosis (CTX) ¹⁷⁾ . Moreover, genetic testing enables early detection of pathogenic variants, allowing timely interventions like a low plant sterol diet and ezetimibe or bile acid sequestrants to prevent complications such as premature atherosclerosis. Furthermore, genetic confirmation supports cascade screening within families, allowing the identification and early treatment of affected relatives, which can reduce morbidity and mortality. These aspects emphasize the indispensable role of genetic testing in improving the diagnosis and management of sitosterolemia.

In this study, we compiled the most comprehensive catalog of clinical features as well as pathogenic variants in ABCG5 and ABCG8 to date, leveraging an extensive search for case reports, database analyses of each variant, and linking each variant to serum sitosterol levels. A total of 53 variants in ABCG5 and 52 variants in ABCG8 were investigated. Among these, 50 variants in ABCG5 and 51 variants in ABCG8 were classified as pathogenic, while 3 and 1 missense variants in ABCG5 and ABCG8, respectively, were designated as VUS. The classification of these variants as VUS was primarily due to the lack of serum sitosterol levels data, highlighting the necessity for further data collection to clarify their clinical significance. On the other hand, the variants identified as pathogenic in this study constitute an invaluable catalog for advancing the genetic diagnosis of sitosterolemia. This comprehensive resource will significantly enhance future diagnostic accuracy and provide crucial support for genetic diagnosis of sitosterolemia.

Geographical and ethnic variations in variant distribution were also observed, shedding light on potential founder effects. The three most frequent pathogenic variants in ABCG5 (R446X, R389H, and IVS7+1G/A) were predominantly seen in East Asian populations. Conversely, the most common pathogenic variants in ABCG8 (W361X, S107X, and L572P) were primarily observed in individuals of European descent. This geographical clustering of variants aligns with the concept of founder effects, where certain variants are more prevalent in specific populations due to historical genetic bottlenecks or localized propagation of the variant. Our catalog captures the unique distribution and characteristics of variants specific to various regions worldwide, making it a valuable resource for region-specific genetic diagnosis of sitosterolemia.

There are several limitations. First, this study is a retrospective analysis of case reports, and the number of variant carriers presented in this study does not necessarily reflect the real-world distribution of cases. However, it successfully captures regional biases in the distribution of variants, providing insights into the presence of founder effects. Second, functional assays and familial segregation analyses to validate the pathogenicity of variants were not performed in this study. Instead, we relied on serum sitosterol levels as well as the ACMG classification framework to determine pathogenicity. This is because sitosterolemia follows an autosomal recessive inheritance pattern, and an increase in serum sitosterol levels is theoretically observed only when both alleles carry loss-of-function variants. Third, serum sitosterol levels data were not available for all identified variants, leading to the classification of some variants as VUS. Further studies with expanded clinical data are needed to accurately evaluate the pathogenicity of these variants. Fourth, the cut-off value for serum sitosterol level was set at 1.0 mg/dL or higher, which has been used as the current diagnostic criteria for sitosterolemia in Japan. However, further study may be required to define another diagnostic cutoff using data on serum sterol levels in healthy individuals, heterozygous and homozygous cases of ABCG5 and ABCG8 variants. At that time, we will reconsider the pathogenicity for some variants of the ABCG5 and ABCG8.

In conclusion, we provided the largest catalog of pathogenic variants of ABCG5 and ABCG8 in the world. This could aid in determining the pathogenicity of variants in ABCG5 and ABCG8 and serve as an essential catalog for the genetic diagnosis of sitosterolemia.

Acknowledgements and Notice of Grant Support

This study was supported by a grant from the Labor and Welfare Sciences Research Grant for Research on Rare and Intractable Diseases (24FC1012) and OT and KM are supported by JSPS 22K08670.

Conflicts of Interest

Takeshi Okada has nothing to disclose. Hayato Tada has nothing to disclose. Kazuhisa Tsukamoto has received honoraria from Bayer Yakuhin, Ltd., Kowa Company, Ltd., and Sanofi K.K. KT also received clinical research funding from MSD. K.K. Masako Togo has nothing to disclose. Masa-aki Kawashiri has nothing to disclose. Kimitoshi Nakamura has nothing to disclose. Kazushige Dobashi has nothing to disclose. Katsunori Ikewaki has nothing to disclose. Hiroshi Yoshida has received honoraria from Kowa Company, Ltd., Novartis Pharma K.K., Novo Nordisk Pharma, Ltd., and Otsuka Pharmaceutical Company, Ltd. HY also received fees for promotional materials from Kowa Company, Ltd., and Denka Company, Ltd. Shizuya Yamashita has received honoraria from Kowa Company, Ltd., Novartis Pharma K.K., Otsuka Pharmaceutical Company, Ltd., and Immuno-Biological Laboratories Company, Ltd. Mariko Harada-Shiba received stock holdings or options from Liid Pharma; honoraria from Amgen Inc., Medpace Japan K.K., Novartis Pharma K.K., Protocera Inc., BML Inc., Ultragenyx Japan Company, Ltd., and Recordati Rare Diseases Japan.