Sitosterolemia presenting as unexplained cytopenias: hematologic manifestations and genetic spectrum

Abstract

Background

Sitosterolemia is a rare autosomal recessive disorder caused by pathogenic variants in the ATP-binding cassette subfamily G member 5 (ABCG5) or ATP-binding cassette subfamily G member 8 (ABCG8) gene, which increase the intestinal absorption of plant sterols and reduce their biliary excretion. It is classically linked to xanthomas and premature cardiovascular disease; however, many patients first present with haematologic abnormalities, often leading to diagnostic delay.

Methods

We retrospectively reviewed six consecutive patients with clinical features suggestive of sitosterolemia and molecular variants in the ABCG5 or ABCG8 gene who were treated at a tertiary referral center between January 2024 and June 2026. We analyzed their clinical features, hematologic findings, genetic variants, treatment, and follow-up outcomes.

Results

Six patients (three men and three women; median age, 40 years [range, 22–55 years]) presented with anemia, thrombocytopenia, and splenomegaly but were initially misdiagnosed with more common disorders, such as immune thrombocytopenia, iron-deficiency anemia, megaloblastic anemia, and Evans syndrome. All patients had stomatocytes and giant platelets on peripheral blood smears; three also had tendon xanthomas, and one had premature coronary artery disease. Genetic testing predominantly identified ABCG8 variants, as well as two ABCG5 variants, including a homozygous ABCG5 c.1724 T > G (p.Leu575Arg) variant that has not been previously reported and is currently classified as a variant of uncertain significance. All patients were advised to follow a phytosterol-restricted diet, and five received ezetimibe. Four patients with available follow-up data showed improvements in hemoglobin, platelet count, and spleen size during 4–12 months of follow-up. One patient was lost to follow-up, and one patient had insufficient follow-up for outcome assessment.

Conclusion

Sitosterolemia should therefore be considered in patients with unexplained cytopenias, stomatocytosis, macrothrombocytopenia, and splenomegaly. Early recognition and genetic confirmation of this condition allows timely treatment with dietary modification and ezetimibe, thereby improving hematologic outcomes and preventing unnecessary investigations.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s44313-026-00176-9.

Introduction

Sitosterolemia is an uncommon inherited disorder of sterol metabolism caused by pathogenic variants in the ATP-binding cassette subfamily G member 5 (ABCG5) or ATP-binding cassette subfamily G member 8 (ABCG8) gene. Defective sterol transport leads to excessive accumulation of plant sterols and a variable clinical phenotype. Xanthomas and premature atherosclerotic cardiovascular disease are well-recognized manifestations; however, many patients first present with unexplained cytopenias, hemolysis, splenomegaly, and macrothrombocytopenia. Stomatocytes and giant platelets on peripheral blood smears are important diagnostic indicators, and early diagnosis is essential because dietary phytosterol restriction and ezetimibe can reverse hematologic abnormalities.

Methods

This retrospective case series included six consecutive patients who had clinical features suggestive of sitosterolemia and molecular variants in the ABCG5 or ABCG8 gene and were treated at a tertiary referral center between January 2024 and June 2026. Patients without genetic testing were excluded. Eligible patients were identified from the institutional hematology database and medical records, and their clinical, laboratory, treatment, and follow-up data were reviewed. Genetic testing was performed on peripheral blood samples using next-generation sequencing–based clinical exome sequencing. Plasma sitosterol and campesterol levels were not measured because these assays were not available at our center. The institutional ethics committee waived the requirement for informed consent because of the retrospective nature of the study.

Results

All six patients presented with unexplained cytopenias and splenomegaly and had initially been misdiagnosed with more common conditions, such as immune thrombocytopenia, iron-deficiency anemia, megaloblastic anemia, or hereditary spherocytosis. Before diagnosis, one patient had received intermittent blood transfusions, one had been treated with eltrombopag, and three had received corticosteroids. All patients had previously been treated for nutritional anemia. Peripheral blood smears consistently showed stomatocytes and giant platelets. Five patients carried pathogenic or likely pathogenic variants in ABCG5 or ABCG8, and one had a variant of uncertain significance (Table 1). Dietary phytosterol restriction and ezetimibe led to progressive improvement in hematologic parameters and spleen size in four patients over 4–12 months of follow-up; one patient was lost to follow-up, and one had insufficient follow-up for outcome assessment (Table 2).

Table 1.

Patient demographics, clinical presentation, and genetic findings

Characteristic	Case 1	Case 2	Case 3	Case 4	Case 5	Case 6
Age/Sex	41/Female	39/Female	55/Female	22/Male	37/Male	40/Male
Year of presentation	2015	2024	2024	2026	2023	2026
Presenting features	Anemia, thrombocytopenia, splenomegaly, tendon xanthomas	Pancytopenia, splenomegaly	Anemia, thrombocytopenia, splenomegaly	Anemia, thrombocytopenia, splenomegaly	Premature coronary artery disease, Anemia, thrombocytopenia, splenomegaly	Anemia, thrombocytopenia, splenomegaly, dyslipidemia
Prior diagnosis / treatment	Presumed ITP; steroids, iron, vitamin B12, blood transfusion	Nutritional supplementation, steroids	Iron, vitamin B12	Iron, vitamin B12	Steroids, TPO agonist, iron, vitamin B12	Iron, vitamin B12
Peripheral smear	Stomatocytes, giant platelets	Stomatocytes, giant platelets	Stomatocytes, giant platelets	Stomatocytes, giant platelets	Stomatocytes, giant platelets	Stomatocytes, giant platelets
Year of genetic diagnosis	2024	2024	2024	2026	2025	2026
Gene	ABCG8	ABCG5	ABCG8	ABCG8	ABCG8	ABCG5
Location	Exon 4	Exon 12	Exon 4	-Exon 4 and Exon 1	Exon 4	Exon 6
Variant type	Frameshift	Missense	Nonsense	Frameshift deletion and Start-loss variant	Frameshift	Nonsense substitution
Depth analyzed	245 ×	120 ×	141 ×	128 ×	267 ×	123 ×
HGVS (cDNA)	ENST00000272286.4 c.408del	NM_022436 c.1724 T > G	ENST00000272286.4 c.361C > T	NM_022437.2c.408del NM_022437.2 c.2 T > C	NM_022437.3 c.408del	NM_022436.3 c.727C > T
HGVS (protein)	p.Gln137SerfsTer8	p.Leu575Arg	p.Arg121Ter	p.Gln137SerfsTer8 + p.Met1?	p.Gln137SerfsTer8	p.Arg243Ter
Zygosity	Heterozygous	Homozygous	Heterozygous	Compound heterozygous	Homozygous	Homozygous
ACMG classification	Likely pathogenic	VUS	Pathogenic	Pathogenic + VUS	Pathogenic	Pathogenic
ACMG evidence	PVS1, PM2	PM2, PP3	PVS1, PM2, PP5	PVS1, PM2	PVS1, PM2, PP5	PVS1, PM2, PP1
Allele frequency	Not detected in the 1000 Genomes Project or gnomAD v3.1/v2.1	gnomAD: 0.0015%;	gnomAD v3.1: 0.0013%; gnomAD v2.1: 0.0023%;	c.408del: gnomAD < 0.001%, c.2T > C: gnomAD < 0.001%	gnomAD: 0.001%; ExAC: 0.0016%	dbSNP: reported rare, gnomAD: rare (< 0.001%)
Conservation	Conserved across species	Not reported	Conserved across species	Not reported	Not reported	Not reported
Previous reports	Trehan et al. 2025 [1]	Not reported	Lu et al. 2001 [2] Parekh et al. 2025 [3]	c.408del Trehan et al. 2025 [1] c.2T > C, not reported	Trehan et al. 2025 [1]	Lu et al. 2001 [2], Trehan et al. 2025 [1], Desai et al. 2021 [4], Warang et al. 2026 [5]
Clinvar data base	Clinvar: pathogenic/likely pathogenic (Variation ID 2150526)	Clinvar: VUS	Clinvar: pathogenic (Variation ID 2733772)	Clinvar: pathogenic/likely pathogenic (Variation ID 2150526) and Start-loss variant ClinVar: (Variation ID 1677567)	Clinvar: pathogenic/likely pathogenic (Variation ID 2150526)	Clinvar: Pathogenic (Variation ID 4977)

Abbreviations: ACMG American College of Medical Genetics and Genomics, dbSNP Single Nucleotide Polymorphism Database, ExAC Exome Aggregation Consortium, gnomAD Genome Aggregation Database, HGVS Human Genome Variation Society nomenclature, ITP immune thrombocytopenic purpura, PM2, PP1, PP3, PP5, PVS1 ACMG evidence codes (moderate, supporting, and very strong pathogenic evidence criteria, respectively), TPO thrombopoietin, VUS variant of uncertain significance, VUSD variant of uncertain significance (deleterious)

Table 2.

Treatment, hematologic response, and follow-up outcomes

	Case 1	Case 2	Case 3	Case 4	Case 5	Case 6
Baseline hemoglobin (g/dL)	8.2	7.6	8.6	9.8	8.7	11.0
Post-treatment hemoglobin (g/dL)	10.5	NA	11.0	11.0	13.5	NA
Baseline leukocyte count (×109/L)	4.7	2.3	4.77	4.1	4.2	7.7
Post-treatment leukocyte count (×109/L)	6.8	NA	8.3	6.35	8.3	NA
Baseline platelet count (×109/L)	39	63	85	106	62	100
Post-treatment Platelet Count (×109/L)	92	NA	189	200	189	NA
Pre-treatment epleen size (cm)	18.6	14.6	15.5	14.2	13.9	14.3
Post treatment spleen size (cm)	13.2	NA	12.5	11.0	10.5	NA
Duration of ezetimibe Therapy	12 months	Not treated	8 months	4 months	7 months	1 month
Dietary intervention restricting vegetable oils, nuts, seeds	Yes	Yes	Yes	Yes	Yes	Yes
Clinical outcome	Hematologic improvement	Lost to follow-up	Hematologic improvement	Hematologic improvement	Hematologic improvement	Recently diagnosed; follow-up not available

Reference ranges: Hemoglobin, 13.0–16.5 g/dL in adult males and 12–16 g/dL in adult females; total leukocyte count, 4.0–11.0 × 10⁹/L; platelet count, 150–400 × 10⁹/L; spleen size, not palpable/impalpable on clinical examination (longitudinal span on imaging generally ≤ 12 cm)

Abbreviations: g/dL grams per deciliter, NA not available (recently diagnosed; follow-up pending)

Discussion

Sitosterolemia is caused by pathogenic ABCG5 or ABCG8 variants that increase intestinal absorption and reduce biliary excretion of plant sterols. It has traditionally been linked to hypercholesterolemia, xanthomas, and premature cardiovascular disease. However, hematologic manifestations, such as hemolytic anemia, macrothrombocytopenia, pancytopenia, and splenomegaly, are common presenting features, and these atypical presentations often lead to delayed or incorrect diagnosis [6, 7].

The hematologic manifestations of sitosterolemia result from the accumulation of phytosterols within erythrocytes and platelet membranes. Incorporation of sitosterol and other plant sterols alters membrane lipid composition, reduces membrane deformability, and induces stomatocytosis and macrothrombocytopenia. These membrane abnormalities increase osmotic fragility, impair erythrocyte hydration and calcium homeostasis, and promote accelerated splenic clearance, resulting in hemolytic anemia, thrombocytopenia, and splenomegaly [3, 5, 8]. Warang et al. showed that erythrocyte injury in sitosterolemia is not solely due to an altered membrane lipid composition. They found increased intracellular calcium levels, enhanced calcium influx during osmotic stress, and increased reactive oxygen species in erythrocytes, suggesting that calcium dysregulation and oxidative stress contribute to membrane instability and hemolysis. They also observed the persistence of overhydrated stomatocytic erythrocytes despite a reduction in plasma phytosterol levels after ezetimibe therapy, indicating that membrane abnormalities may persist despite treatment. Continued phytosterol incorporation into erythrocyte membranes may alter membrane fluidity, promote microvesicle shedding, and reduce the membrane surface area. Because mature erythrocytes cannot repair membrane damage, these structural changes may persist and contribute to ongoing hemolysis in some patients [5]. This combination of stomatocytic hemolytic anemia, macrothrombocytopenia, and splenomegaly overlaps with more common hematologic disorders; therefore, patients are often misdiagnosed with hereditary sperocytosis, immune thrombocytopenia, Evans syndrome, or bone marrow failure syndromes [4, 7, 9]. The presence of stomatocytes and giant platelets on the peripheral blood smear, together with pathogenic ABCG5 or ABCG8 variants, strongly supports a diagnosis of sitosterolemia and helps avoid unnecessary investigations and inappropriate treatment.

In our cohort, stomatocytes and giant platelets were observed on the peripheral blood smear in all six patients, and a representative Wright–Giemsa-stained peripheral blood smear from Case 5 is shown in Fig. 1. This is consistent with earlier reports that hematologic abnormalities can precede overt dyslipidemia or cardiovascular disease and may be the earliest indication of sitosterolemia [1, 3, 4, 10]. Similar to the findings of Warang et al., patients in our cohort had normal routine lipid profiles despite having sitosterolemia. This is because standard enzymatic assays do not specifically measure phytosterols, whereas sterol-specific chromatographic techniques are required for biochemical confirmation [3, 4]. However, the required assays were not available at our center; therefore, the diagnosis in our cohort relied on characteristic peripheral blood smear findings together with genetic testing. As in previous reports, all six patients first presented to hematology services with unexplained cytopenias rather than cardiovascular disease; five had anemia with thrombocytopenia, and one had pancytopenia [1, 3]. Only one patient had established coronary artery disease, and another had dyslipidemia, reflecting the variable clinical spectrum of sitosterolemia [3, 4]. One patient remained undiagnosed for nine years after symptom onset, underscoring the need to consider sitosterolemia in patients with otherwise unexplained cytopenias (Table 1) [1, 3].

Fig. 1.

Wright–Giemsa-stained peripheral blood smear showing stomatocytes and giant platelets (arrow), which are characteristic morphological findings in sitosterolemia. Original magnification, × 40

Genetic analysis showed marked allelic heterogeneity across ABCG5 and ABCG8, with a predominance of ABCG8 variants, consistent with previous reports [2, 7, 11]. A similar pattern was observed by Warang et al., who also reported marked genetic heterogeneity with both novel and previously reported pathogenic variants in ABCG5 and ABCG8. In our cohort, patients had only a single heterozygous ABCG8 pathogenic variant detected. Since copy-number analysis using multiplex ligation-dependent probe amplification (MLPA), deep intronic sequencing, and segregation studies were not available, a second, undetected pathogenic allele could not be ruled out. Nevertheless, these patients had a phenotype highly suggestive of sitosterolemia, including characteristic peripheral blood smear findings and a favorable response to ezetimibe.

Three of the five distinct variants identified in our cohort have been previously reported. The ABCG8 c.361C > T (p.Arg121Ter) and ABCG5 c.727C > T (p.Arg243Ter) variants were first described by Lu et al. and are now listed as pathogenic in ClinVar [2, 11]. The ABCG8 c.408del (p.Gln137SerfsTer8) variant observed in three of our patients was recently reported by Trehan et al. and is classified as pathogenic/likely pathogenic in ClinVar [1, 11]. In contrast, the ABCG8 c.2 T > C (p.Met1Thr) start-loss variant has limited and conflicting evidence in ClinVar and is not well characterized in the literature [11]. The ABCG5 c.1724 T > G (p.Leu575Arg) variant is currently classified as a variant of uncertain significance in ClinVar and, to our knowledge, has not been previously reported in patients with sitosterolemia [11]. Although our patient was homozygous for this variant and had clinical and hematologic findings suggestive of sitosterolemia, the pathogenicity of this variant remains uncertain. Additional evidence from functional studies and segregation analyses is needed to clarify its significance. These findings underline the marked allelic heterogeneity of sitosterolemia and the need to interpret genetic results alongside the clinical and hematologic picture, rather than relying on genotype alone [1–4, 11, 12].

These findings suggest a practical diagnostic approach for patients with unexplained cytopenias. Patients with stomatocytes, giant platelets, splenomegaly, or tendon xanthomas should be promptly evaluated for sitosterolemia once more common causes have been excluded. The initial workup should include a reticulocyte count, direct antiglobulin (Coombs) test, liver function tests, lactate dehydrogenase level, lipid profile, abdominal ultrasonography, osmotic fragility testing, paroxysmal nocturnal hemoglobinuria (PNH) flow cytometry, and plasma phytosterol measurement; genetic testing for ABCG5 and ABCG8 can follow where indicated (Fig. 2).

Fig. 2.

Proposed diagnostic algorithm for the evaluation and diagnosis of sitosterolemia in patients presenting with unexplained cytopenias

All patients were advised to follow a phytosterol-restricted diet, and five received ezetimibe. The four patients with available follow-up data showed improvements in hemoglobin, platelet count, and spleen size, supporting earlier reports that early diagnosis and targeted therapy can reverse hematologic abnormalities and avoid unnecessary immunosuppression or splenectomy [13] (Table 2). In contrast, Warang et al. reported persistent hemolysis and overhydrated erythrocytes despite a reduction in plasma phytosterol levels after ezetimibe therapy, suggesting that established membrane damage may not be completely reversible. They also suggested that therapies targeting calcium homeostasis or oxidative stress may improve treatment response in patients with persistent hemolysis, although further clinical studies are needed [5]. This study has several limitations, including its retrospective, single-center design, small sample size, lack of plasma phytosterol measurements, limited availability of advanced genetic testing, incomplete follow-up, and absence of functional erythrocyte studies.

Conclusion

Sitosterolemia should be considered in patients with unexplained cytopenias, particularly when stomatocytes and giant platelets are observed on a peripheral blood smear in the presence of splenomegaly. Early recognition, genetic confirmation, and prompt treatment with dietary phytosterol restriction and ezetimibe may substantially improve hematologic outcomes while avoiding unnecessary investigations and inappropriate therapy.

Authors’ contributions

N.G. conceptualized the study and critical revision of the manuscript, N.M. collected the data, and reviewed the literature and written manuscript, N.B. reviewed the literature, M.V. contributed in collecting data and revision, J.K. conceptualized the study and critical revision.

Funding

This study did not receive any funding.

Data availability

The data supporting the findings of this study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

The authors confirm that all procedures were performed in accordance with the relevant institutional ethical standards and the principles of the Declaration of Helsinki.

Consent for publication

A waiver of consent for publication was obtained from the Sir Ganga Ram Hospital Committee for this retrospective study.

Competing interests

The authors declare no competing interests.