Published online Nov 15, 2026. doi: 10.4251/wjgo.123971
Revised: August 5, 2026
Accepted: September 24, 2026
Published online: November 15, 2026
Processing time: 156 Days and 1.6 Hours
Sodium taurocholate cotransporting polypeptide deficiency (NTCPD) causes persistent hypercholanemia and neonatal cholestasis, but its genotypic and phenotypic spectrum across ethnicities and geographic backgrounds remains poorly characterized.
To characterize the genetic architecture, biochemical trajectories, and genotype-phenotype correlations of NTCPD in a multi-ethnic pediatric cohort.
In this case cohort study, we analyzed 125 patients with genetically confirmed NTCPD from Kunming Children’s Hospital (2020-2025). Longitudinal bioche
The cohort comprised 69 male and 56 female patients (83 homozygous, 25 heterozygous, and 17 compound heterozygous). The c.800C>T variant was present in all compound heterozygotes, together with 11 additional variants (four likely pathogenic variants and seven variants of uncertain significance). The cohort comprised 73 Han and 52 individuals from ethnic minority groups, predominantly Yi and Hui, distributed across 15 prefectures. Biochemical profiles indicated self-limiting hypercholanemia, with total bile acid levels declining from 178.8 ± 60.9 µmol/L (0-2 months) to 141.2 ± 77.9 µmol/L (> 4 months, P = 0.0002). Homozygous and compound heterozygous genotypes were strongly associated with hypercholanemia and vitamin E deficiency, independent of demographics. Sex-stratified analysis revealed genotype- associated elevations in aspartate aminotransferase in males and more pronounced disturbances in vitamin E and iron levels in females.
In Yunnan’s multi-ethnic population, NTCPD exhibited as self-limiting course and genetic heterogeneity across ethnic and geographic subgroups. These findings provide preliminary evidence to inform larger studies and the development of monitoring strategies.
Core Tip: This 5-year retrospective study of 125 pediatric patients with sodium taurocholate cotransporting polypeptide deficiency from Yunnan Province is the largest multi-ethnic cohort to date. The results reveals that the c.800C>T founder variant predominates across 12 ethnic groups, while 11 additional variants (four likely pathogenic, seven variants of un
- Citation: Wang MF, Chen R, Li J, Cun ZK, Li Q, Jiang HC, Li JW. Genotype-phenotype landscape of Sodium taurocholate cotransporting polypeptide deficiency in Yunnan Province, China: A five-year retrospective, multi-ethnic cohort study. World J Gastrointest Oncol 2026; 18(11): 123971
- URL: https://www.wjgnet.com/1948-5204/full/v18/i11/123971.htm
- DOI: https://dx.doi.org/10.4251/wjgo.123971
Neonatal cholestasis is a clinically significant condition in pediatric hepatology, with diverse causes involving complex gene-environment interactions. Within this spectrum, sodium taurocholate cotransporting polypeptide deficiency (NTCPD) is a distinct monogenic disorder of bile acid metabolism; it was first reported in humans by Vaz et al[1] and subsequently detailed in clinical case studies. Pathogenic mutations in the SLC10A1 gene impair hepatic clearance of conjugated bile salts, resulting in persistent hypercholanemia and potentially contributing to the burden of neonatal hyperbilirubinemia[2,3].
The SLC10A1 c.800C>T (p.Ser267Phe) variant is a founder mutation prevalent in East Asian populations[4-6], yet its distribution exhibits substantial geographic and ethnic heterogeneity[7,8]. Although numerous pathogenic SLC10A1 variants have been reported[9-11], the full spectrum of variants in ethnically diverse cohorts remains incomplete. Moreover, the mechanistic basis for clinical variability and the influence of genetic modifiers on disease expression await systematic investigation[12,13].
Yunnan Province in Southwest China provides a unique setting to explore the heterogeneity of NTCPD because of its substantial ethnic diversity and concentrated populations of Han Chinese and several genetically distinct ethnic minority groups[6]. Previous studies have largely overlooked this regional complexity, leaving important gaps in knowledge regarding genotype–phenotype correlations across different demographic[4,14,15]. To address these limitations, we established a retrospective, multi-ethnic pediatric cohort in Yunnan to comprehensively characterize the genetic architecture and natural history of NTCPD[16]. Using longitudinal assessments and genomic analyses, we investigated associations between SLC10A1 variants, biochemical parameters, and clinical outcomes, with particular emphasis on ethnicity and sex as potential modifiers of disease expression[13,17].
In this study, we enrolled 125 children with persistent hypercholanemia who presented to Kunming Children’s Hospital between January 2020 and February 2025. All inpatients routinely underwent serum total bile acid (TBA) quantification via enzymatic methods and standard liver function tests. Persistent hypercholanemia was defined according to the following criteria: (1) TBA levels exceeding five times the age-specific upper limit of normal in at least three consecutive measurements; (2) Key liver enzymes [including alanine aminotransferase (ALT), aspartate aminotransferase (AST), and gamma-glutamyl transferase (GGT)] remaining below twice the upper limit of normal; and (3) Normal serum albumin levels and coagulation function. Patients with cholestasis attributable to infectious, structural, or immunological causes were excluded. The follow-up period ranged from 0.1 months to 62 months, with 84 children completing three or more follow-up visits. Data from a prospectively maintained registry were retrospectively analyzed. The study protocol was approved by the institutional ethics committee and complied with the ethical guidelines of the Declaration of Helsinki. Written informed consent was obtained from the legal guardian of each participant.
Demographic characteristics (age, sex, weight, height), ethnic and geographic background, family history, past medical history, and clinical symptoms (jaundice, stool color, and pruritus) were recorded through interviews with the parents or caregivers. Clinicians assessed hepatosplenomegaly by palpation during physical examination.
Laboratory measurements included TBA, total bilirubin (TBIL), direct bilirubin (DBIL), indirect bilirubin (IBIL), ALT, AST, GGT, alkaline phosphatase (ALP), and alpha-fetoprotein (AFP). The levels of vitamins A, D, and E and the trace elements zinc and iron were also assessed, as was thyroid function.
A 3 mL sample of peripheral venous blood was collected from each family member. Genomic DNA was extracted using a blood genomic DNA extraction kit (Qiagen, Shanghai, China), and a genomic library was constructed using a standard library construction kit (MyGenostics, Beijing, China). All procedures were performed in strict accordance with the reagent instructions. The target region was captured by a GenCap liquid capture kit (MyGenostics, Beijing, China), and then the enriched library was sequenced using an Illumina NextSeq 500 sequencer (Illumina, CA, United States). After adapters and low-quality reads were removed, the clean reads were mapped to the reference human genome (hg19) using Burrows-Wheeler Aligner software 0.7.15. After detecting genetic variations by Genome Analysis Toolkit software 3.7, single nucleotide polymorphisms and deletions or insertions were used for annotations of the identified variations. Candidate variants were validated in family members using bidirectional Sanger sequencing. The target gene was sequenced on an ABI PRISM3730 genetic analyzer (Applied Biosystems; Thermo Fisher Scientific). The sequencing results were compared with the corresponding reference sequences in GeneBank (www.ncbi.nlm.nih.gov) using DNASTAR software (Madison).
Comparisons among the three age groups (0-2 months, 2-4 months, and > 4 months) were performed using one-way analysis of variance (ANOVA). Bartlett’s test was used to assess the homogeneity of variances, and Brown-Forsythe and Tukey post hoc tests were performed as appropriate.
For continuous biochemical outcomes, multiple linear regression was used to evaluate the relationships between SLC10A1 gene variants and multiple biochemical parameters, including liver enzymes (ALT, AST, GGT, ALP), bilirubin fractions (TBIL, DBIL), TBA, and vitamin concentrations. All analyses were conducted in the subgroup of 84 children who completed three or more follow-up visits. The following three regression models were constructed: Model 1 was unadjusted; Model 2 was adjusted for age; and Model 3 was further adjusted for age, ethnic background, and geographic location. In all models, the normal genotype served as the reference group. The results are presented as beta coefficients with corresponding 95% confidence interval (CI). To explore potential effect modification by sex, we performed sex-stratified analyses (male = 1, female = 0) and tested for gene-sex interaction terms.
For the binary outcome of thyroid dysfunction (normal vs abnormal), binary logistic regression was applied using the same three models and covariate adjustment strategy, with results presented as odds ratio (OR) and 95%CI.
Statistical analyses were performed using SPSS (version 26.0, IBM Corp., Armonk, NY, United States). Graphical representations were generated with GraphPad Prism (version 10.0, GraphPad Software, San Diego, CA, United States). A two-sided P value < 0.05 was considered statistically significant. Given the exploratory nature of this study and the limited subgroup sample sizes, no multiple testing, such as the Bonferroni correction, was applied to the sex-stratified analyses. The results should therefore be interpreted as descriptive effect estimates rather than confirmatory evidence and require independent validation.
A total of 125 children with NTCPD were enrolled, including 69 males (55.2%) and 56 females (44.8%). The age of onset was predominantly early in life: 54 patients (43.2%) presented at 0-2 months, 35 (28.0%) at 2-4 months, 10 (8.0%) at 4-6 months, and 16 (12.8%) at 6-12 months. Later onset was observed in a smaller number of patients, including 3 (2.4%) at 12-24 months, 5 (4.0%) at 24-36 months, and 2 (1.6%) beyond 36 months. The mean body weight was 6.44 ± 2.46 kg (range: 1.67-14.00 kg). Thyroid function screening showed that 104 children (83.2%) were euthyroid, while 21 (16.8%) had subclinical or overt thyroid dysfunction. Genetic analysis identified homozygous SLC10A1 variants in 83 patients (66.4%), heterozygous variants in 25 (20.0%), and compound heterozygous variants in 17 (13.6%). All patients exhibited normal physical development. No cases of pruritus or steatorrhea were reported. Physical examinations revealed no significant abnormalities (Table 1 and Figure 1A-D).
| Variable | Patient cohorts (n = 125) |
| Gender | |
| Female | 56 |
| Male | 69 |
| Age of onset (months) | |
| 0-2 | 54 |
| 2-4 | 35 |
| 4-6 | 10 |
| 6-12 | 16 |
| 12-24 | 3 |
| 24-36 | 5 |
| > 36 | 2 |
| Weight (kg) | 6.44 ± 2.46 (1.67-14.00) |
| Thyroid function | |
| Normal | 104 |
| Abnormal | 21 |
| Gene | |
| HOM | 83 |
| HET | 25 |
| Compound HET | 17 |
| Ethnic group | |
| Han Chinese | 73 |
| China’s ethnic minorities | 52 |
| Region (Yunnan Province) | |
| Kunming city | 49 |
| Other regions | 76 |
The study population consisted of 125 participants from 12 ethnic groups. The Han ethnic group was the majority (n = 73, 58.4%), and 52 participants (41.6%) belonged to ethnic minorities. The distribution was as follows: Yi (n = 17, 13.6%), Hui (n = 14, 11.2%), Dai (n = 6, 4.8%), Bai (n = 5, 4.0%), Wa (n = 2, 1.6%), Pumi (n = 2, 1.6%), and one participant each (0.8%) from the Tujia, Lahu, Blang, Mongolian, and Zhuang groups. This distribution reflects the regional ethnic diversity of Yunnan Province and the demographic predominance of the Han population (Table 1 and Figure 1E). Due to the limited sample size within each minority subgroup, these data are presented as descriptive summaries and were not subjected to formal interaction testing; they serve as a foundation for future studies designed to evaluate ethnic-specific genetic effects.
The study enrolled 125 participants from 15 prefectures in Yunnan Province. Kunming contributed the largest number of participants (n = 49, 39.2%), followed by Honghe (n = 13, 10.4%) and Zhaotong (n = 12, 9.6%). Other regions with more than five participants included Qujing (n = 10, 8.0%), Dali (n = 8, 6.4%), Yuxi (n = 6, 4.8%), Xishuangbanna (n = 6, 4.8%), and Baoshan (n = 5, 4.0%). The remaining seven prefectures had fewer participants: Wenshan and Lijiang (each n = 4, 3.2%), Pu’er (n = 3, 2.4%), Dehong (n = 2, 1.6%), and Nujiang, Chuxiong, and Lincang (each n = 1, 0.8%) (Table 1 and Figure 1F).
Laboratory parameters were analyzed across three age strata (0-2 months, 2-4 months, and > 4 months) relative to established reference intervals (Table 2; Figures 2 and 3). TBA levels were markedly elevated, measuring 8- to 12-fold above the upper reference limit in all age groups. The highest concentrations were observed in the 0-2-month cohort (178.8 ± 60.9 µmol/L), which were significantly higher than those in the 2-month to 4-month cohort (128.5 ± 81.3 µmol/L; P = 0.0002). The levels in the > 4-month cohort were 141.2 ± 77.9 µmol/L.
| Indices | Reference ranges | Paediatric patients | P value | ||
| (0-2 months), n = 54 | (2-4 months), n = 35 | (> 4 months), n = 36 | |||
| TBA | 0-15 µmol/L | 178.8 ± 60.87a | 128.54 ± 81.26b | 141.15 ± 77.92 | a vs b, P = 0.0002 |
| TBIL | 0-26 µmol/L | 84.08 ± 67.86a | 43.41 ± 46.8b | 9.28 ± 4.22c | a vs b, P = 0.0030; b vs c, P < 0.0001; a vs c, P < 0.0001 |
| IBIL | 0-14 µmol/L | 63.49 ± 57.21a | 33.87 ± 44.37b | 5.71 ± 3.3c | a vs b, P = 0.0122; b vs c, P < 0.0001; a vs c, P = 0.0004 |
| GGT | 5-19 U/L | 128.53 ± 112.17a | 87.2 ± 127.38b | 22.57 ± 30.45c | a vs c, P < 0.0001; b vs c, P = 0.0054 |
| ALP | 146-500 U/L | 425.48 ± 157.45a | 343.93 ± 133.28b | 261.16 ± 93.14c | a vs b, P = 0.0148; b vs c, P = 0.0041; a vs c, P < 0.0001 |
| DBIL | 0-10 µmol/L | 20.74 ± 38.75 | 9.54 ± 11.28 | 4.14 ± 3.35 | NS |
| ALT | 7-30 U/L | 37.50 ± 29.95 | 42.44 ± 48.95 | 40.69 ± 39.15 | NS |
| AST | 14-44 U/L | 64.24 ± 54.81 | 64.44 ± 54.09 | 56.17 ± 22.54 | NS |
| AFP | < 200 µg/L | 5472.47 ± 12515.0 | 1515.87 ± 2992.81 | 72.72 ± 183.27 | NS |
| Vit A | 1.05-2.8 µmol/L | 1.467 ± 0.55 | 1.79 ± 0.73 | 2.00 ± 0.58 | NS |
| Vit D | 50-147 nmol/L | 79.60 ± 34.71 | 97.75 ± 37.55 | 122.52 ± 35.44 | NS |
| Vit E | 16.25-33.5 µmol/L | 26.25 ± 8.19 | 29.0 ± 7.01 | 29.23 ± 7.14 | NS |
| Zinc | 3.7-7.3 mg/L | 3.91 ± 0.59 | 3.47 ± 0.9 | 4.33 ± 0.49 | NS |
| Fe | 373.5-557.2 mg/L | 399.35 ± 30.26 | 631.93 ± 890.43 | 409.24 ± 32.72 | NS |
Both conjugated and unconjugated hyperbilirubinemia demonstrated a similar pattern of decline with age. TBIL decreased from 84.1 ± 67.9 µmol/L in the 0-month to 2-month group to 9.3 ± 4.2 µmol/L in the > 4-month group (P < 0.0001), while IBIL decreased from 63.5 ± 57.2 µmol/L to 5.7 ± 3.3 µmol/L (P < 0.0001). DBIL also exhibited a downward trend, though this did not reach statistical significance.
Cholestatic indices followed a comparable pattern, with GGT activity peaking during early infancy (128.5 ± 112.2 U/L) and declining significantly in the > 4-month group (22.6 ± 30.5 U/L; P = 0.005). ALP levels decreased from 425.5 ± 157.5 U/L to 261.2 ± 93.1 U/L (P < 0.0001).
In contrast, markers of hepatocellular injury remained mild and age-stable. ALT and AST were only modestly elevated and showed no significant intergroup differences. AFP was substantially elevated in the youngest cohort (5472 ± 12515 µg/L) but declined to near-normal levels thereafter (72.7 ± 183 µg/L).
The nutritional parameters generally improved with age. Vitamin D concentrations increased from 79.6 ± 34.7 nmol/L to 122.5 ± 35.4 nmol/L, while vitamins A and E remained within reference ranges. Zinc and iron levels were largely normal across groups, with only minor, non-significant variations observed.
In summary, NTCPD presents as a self-limiting, age-modulated cholestatic disorder characterized by prominent early hypercholanemia, elevated GGT and AFP levels, but minimal evidence of hepatocellular injury and generally preserved nutritional status.
Genetic profiling of 125 pediatric patients with NTCPD revealed three distinct genotypic patterns: 83 cases (66.4%) carried homozygous SLC10A1 variants; 25 (20.0%) were heterozygous, and 17 (13.6%) were compound heterozygous. Comprehensive analysis of the 17 compound heterozygous probands and their families identified 11 distinct SLC10A1 variants. The c.800C>T (p.Ser267Phe) variant was present in all compound heterozygous cases as one allele.
Parental genotyping demonstrated that 15 of the 17 cases followed a classical autosomal recessive inheritance pattern, where the c.800C>T variant was inherited from one parent and the secondary variant was inherited from the other parent. However, two cases (probands 1 and 10) showed atypical inheritance patterns, with both maternal variants residing in cis. In proband 1, the mother carried both the c.800C>T and c.185C>T variants on the same allele, while in proband 10, the mother carried both the c.800C>T and c.402G>A variants similarly in cis.
According to ACMG/AMP guidelines, four variants were classified as likely pathogenic (LP): C.682_683del, c.263T>C (identified in three families); c.54delT (found in three families); and c.54del. The remaining seven variants were categorized as variants of uncertain significance (VUS) and included the following: C.185C>T, c.944-32G>C (observed in three families); c.799T>C, c.402G>A, c.422dupA (detected in two families); c.997G>A (identified in two families); and c.413T>C. This genetic landscape highlights both the predominance of the c.800C>T founder variant and the substantial allelic heterogeneity underlying NTCPD in this cohort (Tables 1 and 3; Figures 1D and 4).
| Father | Mother | ACMG classification | |
| Proband (n = 17) | Compound heterozygous | ||
| c.800C>T, c.185C>T | c.800C>T | c.800C>T, c.185C>T | c.185C>T: VUS |
| c.800C>T, c.944-32G>C | c.944-32G>C | c.800C>T | c.944-32G>C: VUS |
| c.800C>T, c.944-32G>C | c.800C>T | c.944-32G>C | c.944-32G>C: VUS |
| c.800C>T, c.263T>C | c.263T>C | c.800C>T | c.944-32G>C: VUS |
| c.800C>T, c.682_683del | c.682_683del | c.800C>T | c.682_683del: LP |
| c.799T>C, c.800C>T | c.800C>T | c.799T>C | c.799T>C: VUS |
| c.800C>T, c.263T>C | c.800C>T | c.263T>C | c.263T>C: P |
| c.800C>T, c.54delT | c.800C>T | c.54delT | c.54delT: LP |
| c.800C>T, c.54delT | c.800C>T | c.54delT | c.54delT: LP |
| c.800C>T, c.402G>A | c.800C>T | c.800C>T, c.402G>A | c.402G>A: VUS |
| c.422dupA, c.800C>T | c.800C>T | c.422dupA | c.422dupA: VUS |
| c.422dupA, c.800C>T | c.800C>T | c.422dupA | c.422dupA: VUS |
| c.997G>A, c.800C>T | c.997G>A, c.800C>T | c.800C>T | c.997G>A: VUS |
| c.997G>A, c.800C>T | c.997G>A, c.800C>T | c.800C>T | c.997G>A: VUS |
| c.800C>T, c.54del | c.54del | c.800C>T | c.54del: LP |
| c.800C>T, c.263T>C | c.800C>T | c.263T>C | c.263T>C: P |
| c.800C>T, c.413T>C | c.800C>T, c.413T>C | c.800C>T | c.413T>C: VUS |
| Proband (n = 25) | Heterozygous | ||
| c.800C>T | c.800C>T | c.800C>T: P | |
| c.800C>T | c.800C>T | ||
| c.263T>C | c.263T>C | c.263T>C: P | |
| c.800C>T | c.800C>T | ||
| c.800C>T | c.800C>T | ||
| c.800C>T | c.800C>T | ||
| c.745C>T | c.745C>T | c.745C>T: VUS | |
| c.800C>T | c.800C>T | ||
| c.800C>T | c.800C>T | ||
| c.800C>T | c.800C>T | ||
| c.800C>T | c.800C>T | ||
| c.800C>T | c.800C>T | ||
| c.800C>T | c.800C>T | ||
| c.800C>T | c.800C>T | ||
| c.800C>T | c.800C>T | ||
| c.800C>T | c.800C>T | ||
| c.800C>T | c.800C>T | ||
| c.800C>T | c.800C>T | ||
| c.800C>T | c.800C>T | ||
| c.800C>T | c.800C>T | ||
| c.800C>T | c.800C>T | ||
| c.800C>T | c.800C>T | ||
| c.800C>T | c.800C>T | ||
| c.800C>T | c.800C>T | ||
| c.800C>T | c.800C>T | ||
| Proband (n = 83) | homozygous | ||
| c.800C>T | c.800C>T | c.800C>T | c.800C>T: P |
Based on the ACMG/AMP variant classification and the observed genotype–phenotype associations in this cohort, we propose the following clinical attribution framework for SLC10A1-related phenotypes: (1) Classical NTCPD (homozygous or compound heterozygous). Patients with biallelic SLC10A1 variants (homozygous or compound heterozygous) were assigned to the classical NTCPD group. This category included 83 homozygotes, all carrying c.800C>T in the homozygous state, and 17 compound heterozygotes. In our regression models, this group corresponds to Gene 1 (HOM) and Gene 3 (Compound HET); and (2) NTCPD associated hypercholanemia (heterozygous or monoallelic). The 25 heterozygous patients (all c.800C>T carriers) were classified as having NTCPD associated hypercholanemia or a monoallelic phenotype. This term highlights their clinical presentation and persistent hypercholanemia according to our enrollment criteria, rather than adherence to a recessive Mendelian framework.
Several factors support including these heterozygotes. First, all met the phenotype-driven entry criteria. Second, during infancy, when the bile acid load is high and hepatic transporter reserve may be limited, a single defective allele could still impair bile acid clearance. Third, our targeted sequencing cannot rule out a second variant in noncoding regions. Fourth, the cis-acting variants observed in two probands suggest that allelic configuration may influence expressivity in some families. In the regression analyses, this group is designated as Gene 2 (HET) and was evaluated separately to reflect the broader clinical spectrum of SLC10A1 related hypercholanemia.
Multivariable-adjusted analyses of the relationships between gene variants and serum biochemical indices in the 84 pediatric participants are presented in Table 4. Three sequential regression models were constructed: A crude model (Model 1), a model adjusted for age (Model 2), and a fully adjusted model that additionally accounted for age, ethnicity, and geographic location (Model 3).
| Model 1 | Model 2 | Model 3 | |||||||
| Characteristic | Beta | 95%CI | P value | Beta | 95%CI | P value | Beta | 95%CI | P value |
| Gene | TBA | ||||||||
| 1 | 82 | 31-132 | 0.002 | 83 | 31-134 | 0.002 | 86 | 26-147 | 0.007 |
| 2 | -2.0 | -58 to 54 | > 0.9 | -2.1 | -59 to 55 | > 0.9 | -6.1 | -72 to 60 | 0.9 |
| 3 | 95 | 36-155 | 0.002 | 95 | 35-155 | 0.003 | 95 | 23-166 | 0.011 |
| GGT | |||||||||
| 1 | -182 | -284 to -80 | < 0.001 | -176 | -279 to -73 | 0.001 | -253 | -380 to -125 | < 0.001 |
| 2 | -189 | -302 to -76 | 0.002 | -189 | -303 to -76 | 0.002 | -255 | -394 to -116 | < 0.001 |
| 3 | -149 | -269 to -29 | 0.017 | -149 | -270 to -29 | 0.017 | -244 | -393 to -95 | 0.002 |
| AFP | |||||||||
| 1 | 800 | -4142 to 5741 | 0.8 | 1294 | -3717 to 6305 | 0.6 | 490 | -7846 to 8826 | > 0.9 |
| 2 | -441 | -5886 to 5004 | 0.9 | -447 | -5883 to 4988 | 0.9 | -1149 | -9866 to 7568 | 0.8 |
| 3 | 6,416 | 207-12624 | 0.048 | 6349 | 150-12547 | 0.050 | 5334 | -350 to 14619 | 0.3 |
| Vit D | |||||||||
| 1 | -24 | -56 to 7.6 | 0.14 | -27 | -59 to 5.1 | 0.10 | -18 | -55 to 19 | 0.4 |
| 2 | -4.1 | -40 to 32 | 0.8 | -4.0 | -39 to 32 | 0.8 | -0.91 | -41 to 40 | > 0.9 |
| 3 | -43 | -81 to -3.7 | 0.035 | -42 | -81 to -3.8 | 0.035 | -35 | -80 to 10 | 0.13 |
| Vit E | |||||||||
| 1 | -13 | -19 to -7.5 | < 0.001 | -13 | -19 to -7.4 | < 0.001 | -15 | -22 to -8.1 | < 0.001 |
| 2 | -7.8 | -14 to -1.5 | 0.018 | -7.8 | -14 to -1.4 | 0.019 | -9.3 | -17 to -1.9 | 0.017 |
| 3 | -14 | -23 to -6.1 | 0.001 | -14 | -23 to -6.0 | 0.001 | -16 | -25 to -6.6 | 0.002 |
TBA: Gene variants 1 and 3 were consistently associated with elevated TBA levels across all models. In Model 3, variant 1 was associated with an increase of 86 units (95%CI: 26-147; P = 0.007); variant 3 with an increase of 95 units (95%CI: 23-166; P = 0.011); and variant 2 was not significantly associated with TBA in any model (Model 3: Β = -6.1, 95%CI: -72 to 60; P = 0.9).
GGT: All three gene variants showed significant inverse associations with GGT levels; these associations remained consistent and were strengthened in Model 3. Specifically, in Model 3, variant 1 was associated with a reduction of 253 units (95%CI: -380 to -125; P < 0.001), variant 2 with a reduction of 255 units (95%CI: -394 to -116; P < 0.001), and variant 3 with a reduction of 244 units (95%CI: -393 to -95; P = 0.002).
AFP: A significant positive association was observed between variant 3 and AFP levels in Model 1 (β = 6,416, 95%CI: 207-12624; P = 0.048) and Model 2 (β = 6349, 95%CI: 150-12547; P = 0.050). However, this association was attenuated and lost statistical significance in Model 3 (β = 5334, 95%CI: –3950 to 14619; P = 0.3). No significant associations were found between variants 1 or 2 and AFP in any model.
Vitamin D: Variant 3 was significantly associated with lower vitamin D levels in Model 1 (β = -43, 95%CI: -81 to -3.7; P = 0.035) and Model 2 (β = –42, 95%CI: -81 to -3.8; P = 0.035). However, this association was not sustained in Model 3 (β = -35, 95%CI: -80 to 10; P = 0.13). Variants 1 and 2 did not show any significant association with vitamin D levels.
Vitamin E: All three gene variants were consistently associated with lower vitamin E levels across all models. In Model 3, variant 1 was associated with a decrease of 15 units (95%CI: -22 to -8.1; P < 0.001), variant 2 with a decrease of 9.3 units (95%CI: -17 to -1.9; P = 0.017), and variant 3 with a decrease of 16 units (95%CI: -25 to -6.6; P = 0.002).
Logistic regression analyses were performed to examine the relationships between gene variants and thyroid dysfunction in the pediatric cohort, with the results detailed in Table 5. Three models were constructed: An unadjusted model (Model 1), a model adjusted for sex (Model 2), and a fully adjusted model that included sex, age, ethnicity, and location (Model 3). In Model 1, gene variant 1 demonstrated a significant inverse association with thyroid dysfunction (OR = 0.07, 95%CI: 0.01-0.42, P = 0.006). This association remained significant after adjusting for sex in Model 2 (OR = 0.07, 95%CI: 0.01-0.45, P = 0.007). However, following full adjustment for covariates in Model 3, the association was attenuated and no longer statistically significant (OR = 0.14, 95%CI: 0.01-1.58, P = 0.12). Gene variant 2 was not significantly associated with thyroid dysfunction in any of the models. In Model 1, the OR was 0.17 (95%CI: 0.02-1.18, P = 0.085); in Model 2, the OR was 0.17 (95%CI: 0.02-1.33, P = 0.10); and in Model 3, the OR was 0.42 (95%CI: 0.03-5.62, P = 0.5). A similar pattern was observed for gene variant 3, which showed a non-significant trend toward a reduced risk of thyroid dysfunction in both Model 1 (OR = 0.11, 95%CI: 0.01-0.96, P = 0.060) and Model 2 (OR = 0.11, 95%CI: 0.01-0.96, P = 0.060). This association also lost significance in Model 3 (OR = 0.40, 95%CI: 0.02-6.60, P = 0.5).
| Model 1 | Model 2 | Model 3 | |||||||
| Characteristic | OR | 95%CI | P value | OR | 95%CI | P value | OR | 95%CI | P value |
| Gene | |||||||||
| 1 | 0.07 | 0.01-0.42 | 0.006 | 0.07 | 0.01-0.45 | 0.007 | 0.14 | 0.01-1.58 | 0.12 |
| 2 | 0.17 | 0.02-1.18 | 0.085 | 0.17 | 0.02-1.33 | 0.10 | 0.42 | 0.03-5.62 | 0.5 |
| 3 | 0.11 | 0.01-0.96 | 0.060 | 0.11 | 0.01-0.96 | 0.060 | 0.40 | 0.02-6.60 | 0.5 |
Sex-stratified multivariable linear regression analysis was performed to examine associations between genotypes and serum biochemical indices in male pediatric participants (n = 41), with results presented in Table 6. Three models were constructed: An unadjusted model (Model 1), a model adjusted for age (Model 2), and a fully adjusted model that included age, ethnicity, and location (Model 3).
| Model 1 | Model 2 | Model 3 | |||||||
| Characteristic | Beta | 95%CI | P value | Beta | 95%CI | P value | Beta | 95%CI | P value |
| Gene | TBA | ||||||||
| 1 | 108 | 3.1-214 | 0.051 | 109 | 2.2-216 | 0.053 | 206 | 69-342 | 0.007 |
| 2 | 30 | -78 to 138 | 0.6 | 29 | -80 to 139 | 0.6 | 105 | -37 to 246 | 0.2 |
| 3 | 92 | -28 to 211 | 0.14 | 91 | -30 to 212 | 0.15 | 176 | 26-326 | 0.031 |
| AST | |||||||||
| 1 | 18 | -6.0 to 42 | 0.2 | 18 | -6.1 to 43 | 0.2 | 55 | 28-82 | < 0.001 |
| 2 | 17 | -7.7 to 42 | 0.2 | 17 | -8.2 to 42 | 0.2 | 51 | 22-79 | 0.002 |
| 3 | 14 | -14 to 41 | 0.3 | 13 | -14 to 41 | 0.3 | 49 | 19-79 | 0.004 |
| ALP | |||||||||
| 1 | -317 | -484 to -150 | < 0.001 | -305 | -463 to -148 | < 0.001 | -349 | -611 to -87 | 0.016 |
| 2 | -262 | -433 to -90 | 0.005 | -271 | -433 to -109 | 0.002 | -293 | -565 to -21 | 0.046 |
| 3 | -354 | -544 to -165 | < 0.001 | -359 | -538 to -181 | < 0.001 | -375 | -662 to -87 | 0.018 |
| Vit A | |||||||||
| 1 | 0.40 | -0.59 to 1.4 | 0.4 | 0.34 | -0.61 to 1.3 | 0.5 | 0.12 | -1.2 to 1.5 | 0.9 |
| 2 | 0.28 | -0.73 to 1.3 | 0.6 | 0.32 | -0.65 to 1.3 | 0.5 | 0.40 | -1.0 to 1.8 | 0.6 |
| 3 | 1.1 | 0.00-2.2 | 0.060 | 1.1 | 0.07-2.2 | 0.046 | 0.94 | -0.55 to 2.4 | 0.2 |
TBA: In Model 3, genotype 1 was significantly associated with increased TBA levels (β = 206, 95%CI: 69-342, P = 0.007). Genotype 3 also showed a significant positive association with TBA in Model 3 (β = 176, 95%CI: 26-326, P = 0.031). No significant associations were observed between genotype 2 and TBA levels across any models.
AST: All three genotypes demonstrated significant positive associations with AST levels in the fully adjusted model. Specifically, genotype 1 was associated with an increase in AST (β = 55, 95%CI: 28-82, P < 0.001), as were genotype 2 (β = 51, 95%CI: 22-79, P = 0.002) and genotype 3 (β = 49, 95%CI: 19-79, P = 0.004). No significant associations were observed in the less adjusted models.
ALP: All genotypes showed consistent inverse associations with ALP levels across all models. In Model 3, genotype 1 was associated with a reduction in ALP (β = 349, 95%CI: 611-87, P = 0.016), along with genotype 2 (β = 293, 95%CI: 565-21, P = 0.046) and genotype 3 (β = 375, 95%CI: 662-87, P = 0.018).
Vitamin A: A marginally significant positive association was observed between genotype 3 and vitamin A levels in Model 2 (β = 1.1, 95%CI: 0.07-2.2, P = 0.046). However, this association was not maintained in Model 3. No significant associations were found between other genotypes and vitamin A levels in any model.
Associations between genotypes and serum biochemical parameters in female participants (n = 43) were examined using sex-stratified multivariable linear regression analysis, with the results summarized in Table 7. Three models were constructed: An unadjusted model (Model 1), a model adjusted for age (Model 2), and a fully adjusted model that included age, ethnicity, and location (Model 3).
| Model 1 | Model 2 | Model 3 | |||||||
| Characteristic | Beta | 95%CI | P value | Beta | 95%CI | P value | Beta | 95%CI | P value |
| Gene | TBA | ||||||||
| 1 | 89 | 28-150 | 0.007 | 89 | 27-152 | 0.008 | 76 | -8.3 to 160 | 0.089 |
| 2 | 13 | -65 to 92 | 0.7 | 13 | -67 to 93 | 0.7 | -1.9 | -96 to 92 | > 0.9 |
| 3 | 104 | 33-175 | 0.007 | 104 | 32-176 | 0.007 | 97 | 0.44-194 | 0.059 |
| GGT | |||||||||
| 1 | -215 | -373 to -56 | 0.012 | -210 | -372 to -48 | 0.015 | -346 | -565 to -128 | 0.004 |
| 2 | -223 | -428 to -19 | 0.039 | -223 | -431 to -16 | 0.041 | -307 | -551 to -64 | 0.020 |
| 3 | -172 | -357 to 13 | 0.076 | -172 | -359 to 15 | 0.079 | -327 | -578 to -76 | 0.016 |
| AFP | |||||||||
| 1 | 1589 | -6335 to 9513 | 0.7 | 2004 | -6132 to 10140 | 0.6 | -1748 | -16919 to 13422 | 0.8 |
| 2 | -273 | -10648 to 10102 | > 0.9 | -253 | -10761 to 10255 | > 0.9 | -1385 | -17234 to 14463 | 0.9 |
| 3 | 11291 | 916-21666 | 0.043 | 11147 | 629-21665 | 0.049 | 9749 | -6415 to 25914 | 0.3 |
| Vit D | |||||||||
| 1 | -34 | -70 to 2.3 | 0.075 | -33 | -70 to 4.3 | 0.092 | -5.9 | -50 to 38 | 0.8 |
| 2 | -9.3 | -56 to 37 | 0.7 | -9.3 | -56 to 38 | 0.7 | -0.01 | -49 to 49 | > 0.9 |
| 3 | -66 | -110 to -21 | 0.007 | -66 | -111 to -21 | 0.007 | -34 | -88 to 19 | 0.2 |
| Vit E | |||||||||
| 1 | -15 | -23 to -7.5 | < 0.001 | -15 | -24 to -7.2 | 0.001 | -22 | -34 to -10 | 0.003 |
| 2 | -8.8 | -19 to 1.2 | 0.10 | -8.8 | -19 to 1.4 | 0.10 | -11 | -22 to -0.48 | 0.060 |
| 3 | -18 | -35 to -0.72 | 0.052 | -18 | -36 to -0.37 | 0.056 | -23 | -42 to -4.8 | 0.027 |
| Fe | |||||||||
| 1 | -34 | -58 to -8.9 | 0.018 | -36 | -62 to -11 | 0.015 | -58 | -100 to -15 | 0.032 |
| 2 | -36 | -68 to -3.8 | 0.045 | -36 | -69 to -3.7 | 0.047 | -44 | -84 to -4.0 | 0.068 |
| 3 | -25 | -72 to 22 | 0.3 | -25 | -73 to 22 | 0.3 | -48 | -106 to 9.4 | 0.15 |
| Thyroid dysfunction | |||||||||
| 1 | 0.06 | 0.00-0.55 | 0.017 | 0.08 | 0.01-0.71 | 0.030 | 0.23 | 0.01-5.14 | 0.3 |
| 2 | 1.00 | 0.07-13.8 | > 0.9 | 1.00 | 0.07-14.0 | > 0.9 | 1.57 | 0.08-41.5 | 0.8 |
| 3 | 0.10 | 0.00-1.20 | 0.094 | 0.17 | 0.01-2.43 | 0.2 | 0.45 | 0.01-16.0 | 0.7 |
TBA: Genotype 1 was positively associated with TBA in Models 1 (β = 89, 95%CI: 28-150, P = 0.007) and 2 (β = 89, 95%CI: 27-152, P = 0.008), but not in Model 3 (β = 76, 95%CI: -8.3 to 160, P = 0.089). Genotype 3 was significantly associated with TBA in Models 1 and 2 (β = 104, P = 0.007 for both), with a borderline association in Model 3 (β = 97, 95%CI: 0.44-194, P = 0.059). Genotype 2 was not associated with TBA in any model.
GGT: All genotypes were inversely associated with GGT levels in Model 3: Genotype 1 (β = -346, 95%CI: –565 to -128, P = 0.004), genotype 2 (β = –307, 95%CI: -551 to -64, P = 0.020), and genotype 3 (β = -327, 95%CI: -578 to -76, P = 0.016).
AFP: Genotype 3 was positively associated with AFP in Models 1 (β = 11291, 95%CI: 916-21666, P = 0.043) and 2 (β = 11147, 95%CI: 629-21665, P = 0.049), but not in Model 3 (β = 9749, 95%CI: -6415 to 25914, P = 0.3). Genotypes 1 and 2 showed no significant associations.
Vitamin D: Genotype 3 was inversely associated with vitamin D in Models 1 and 2 (β = –66, P = 0.007 for both), but not in Model 3 (β = -34, 95%CI: -88 to 19, P = 0.2). Genotype 1 showed a trend in Models 1 and 2, which was attenuated in Model 3. Genotype 2 was not associated with vitamin D in any model.
Vitamin E: Genotypes 1 and 3 were significantly inversely associated with vitamin E in Model 3 (β = -22, 95%CI: -34 to -10, P = 0.003 and β = -23, 95%CI: -42 to -4.8, P = 0.027, respectively). Genotype 2 showed a marginal association in Model 3 (β = -11, 95%CI: -22 to -0.48, P = 0.060).
Iron: Genotype 1 was inversely associated with iron in all models, including Model 3 (β = -58, 95%CI: -100 to -15, P = 0.032). Genotype 2 showed a similar trend in Model 3 (β = -44, 95%CI: -84 to -4.0, P = 0.068). Genotype 3 was not associated with iron levels.
Within the female subgroup, genotype 1 was inversely associated with thyroid dysfunction in Model 1 (OR = 0.06, 95%CI: 0.00-0.55, P = 0.017) and 2 (OR = 0.08, 95%CI: 0.01-0.71, P = 0.030), but not in Model 3 (OR = 0.23, 95%CI: 0.01-5.14, P = 0.3). Genotype 3 showed a marginal association only in Model 1 (OR = 0.10, 95%CI: 0.00-1.20, P = 0.094).
Based on the comprehensive findings from this 5-year retrospective multi-ethnic cohort study, we delineate the complex genotype-phenotype landscape of NTCPD in Yunnan Province, China. Our analysis reveals several critical dimensions of this disorder. Given the rarity of NTCPD, this study should be considered exploratory rather than confirmatory in nature. Our objective was not to provide definitive clinical guidance, but to map the preliminary landscape of this disorder in a genetically diverse population, thereby generating testable hypotheses for subsequent investigations.
The cohort demonstrated notable demographic variations, with a male predominance (55.2%) and early infancy presentation (43.2% within 0-2 months). Thyroid dysfunction was observed in 16.8% of participants, while genetic analysis identified three distinct patterns: Homozygous (66.4%), heterozygous (20.0%), and compound heterozygous (13.6%) variants. The ethnic composition reflected Yunnan’s exceptional diversity, with 41.6% belonging to minority groups, predominantly Yi (13.6%), Hui (11.2%), Dai (5.0%), and Bai (4.1%). This ethnic distribution closely corresponded with geographical patterns, where cases from minority-concentrated prefectures aligned with traditional autonomous regions[8,15]. The widespread distribution across multiple prefectures with high ethnic minority densities suggests a correlation between the distribution of NTCPD cases and regional ethnic concentrations, likely reflecting underlying population-specific genetic susceptibilities[18-20].
Analysis across age strata revealed a dynamic biochemical progression that delineates the natural course of NTCPD. Cholestatic parameters demonstrated a particularly prominent age-dependent resolution. TBA levels peaked dramatically in the 0-2 months cohort before declining significantly. Parallel changes were observed in bilirubin metabolism, with both total and IBIL decreasing precipitously from markedly elevated neonatal levels to near-normal ranges beyond 4 months. GGT activity followed a similar trajectory, showing substantial elevation during early infancy but normalizing significantly after 4 months of age.
In contrast to these evolving cholestatic markers, indicators of hepatocellular injury remained essentially stable. Both ALT and AST maintained only modest elevations without significant intergroup differences, reinforcing the non-progressive nature of the condition[21]. The age-dependent decline in AFP paralleled normal physiological liver maturation, and was therefore not interpreted as a disease-specific phenomenon[22,23].
The nutritional parameters generally demonstrated favorable profiles. The vitamin D concentrations increased significantly with advancing age (from 79.6 nmol/L ± 34.7 nmol/L to 122.5 nmol/L ± 35.4 nmol/L), while vitamins A and E, along with zinc and iron levels, remained largely within reference ranges throughout infancy. This pattern suggests that most nutritional biomarkers remained within age-appropriate reference ranges, suggesting that clinically significant malabsorption was not evident in this cohort despite the underlying transporter defect[24,25]. It is important to note that despite the significant associations observed for vitamin E and iron, the absolute levels for most nutritional parameters remained within or near reference ranges. Thus, these findings should not be interpreted as evidence of generalized mala
Comprehensive genotypic analysis revealed a distribution pattern where homozygous variants predominated (66.4%), consistent with characteristic autosomal recessive inheritance[1,6]. However, the substantial proportion of heterozygous cases (20.0%) presents intriguing clinical implications, as these patients may harbor undetected secondary variants or demonstrate modifying influences from other genetic or environmental factors[27,28]. The compound heterozygous subgroup (13.6%) proved particularly informative for understanding allelic heterogeneity[11]. The c.800C>T variant was invariably present in all compound heterozygous cases, confirming its established role as the primary founder mutation in East Asian populations[5,6]. Partnered with this common variant, we identified 11 distinct secondary mutations, demonstrating remarkable mutational diversity[9,11]. According to ACMG classification standards[29,30], four variants were categorized as LP, providing clear guidance for genetic counseling and prenatal diagnosis[31]. The remaining seven VUS pose a clinical challenge yet offer potential insights into novel disease mechanisms[28]. The seven VUS variants identified in this study (c.185C T, c.944-32G>C, c.799T>C, c.402G>A, c.422dupA, c.997G>A, c.413T>C) currently lack functional validation. However, their segregation patterns within affected families provide important supportive evidence: (1) c.944-32G>C was observed in trans with c.800C>T in three independent families, with all affected members exhibiting hypercholanemia; (2) c.422dupA was recurrent in two families, consistently co-segregating with the pheno
To delineate genotype-phenotype relationships while accounting for potential confounders, we employed sequential multivariable regression models with increasing stringency. This analytical approach revealed distinct patterns of association across biochemical parameters. Notably, homozygous and compound heterozygous variants demonstrated robust positive associations with TBA levels that persisted across all models. In the fully adjusted model, the effect estimates remained stable or even strengthened indicating that these genotype effects are independent of major demographic confounders. In contrast, heterozygous variants showed no significant association with TBA in any model, supporting the classic autosomal recessive inheritance pattern of NTCPD[1,21]. Moreover, a paradoxical inverse relationship emerged for GGT, with all genotypes showing consistent negative associations that intensified following full adjustment for demographic variables. This pattern suggests the presence of competing factors that may partially mask the genuine genetic effect on GGT regulation in NTCPD[16,18]. Vitamin E metabolism exhibited another robust association, with all genotypic groups demonstrating significantly lower levels across all models. The persistence of this relationship after comprehensive adjustment indicates an association between SLC10A1 genotypes and lower serum vitamin E levels, extending even to heterozygous carriers, although the clinical significance of this remains uncertain[1,24,25]. However, several initially observed associations proved susceptible to demographic confounding. Apparent relationships between compound heterozygous variants and both AFP and vitamin D levels attenuated to non-significance after full adjustment. Similarly, the initially observed inverse association with thyroid dysfunction disappeared upon controlling for demographic factors, indicating these were likely spurious associations driven by ethnic or geographic clustering rather than direct genetic effects.
These findings highlight several key insights. The core biochemical hallmarks of NTCPD, consisting of elevated TBA and reduced vitamin E levels, maintain genuine genetic associations that are not influenced by demographic factors[1,10,12]. The relationship observed with GGT reveals a complex regulatory mechanism that may be unique to NTCPD among cholestatic disorders[18,20]. Furthermore, our results emphasize the need for comprehensive adjustment of demographic variables in genetic association studies, particularly in ethnically diverse populations[8,12,13].
First, regarding the TBA-genotype link, the NTCP protein, encoded by SLC10A1, mediates sodium-dependent uptake of conjugated bile acids into hepatocytes. The c.800C>T (p.Ser267Phe) alteration lies within the transmembrane region and has been shown to reduce transport activity by 40%-70% in functional assays[32,33]. In homozygotes, this severe functional loss impairs the hepatic clearance of bile acids returning via the enterohepatic circulation, leading to hypercholanemia. The graded effect observed across genotype groups (homozygotes showing the highest TBA levels, followed by compound heterozygotes, then heterozygotes) is consistent with this dose dependent model.
Second, in terms of the possible explanations for sex related differences, several mechanisms could account for the sex stratified patterns we observed, although all remain speculative. These include estrogen mediated regulation of bile acid transporters, as animal studies have shown that estrogen can downregulate NTCP and OATP expression, although our prepubertal cohort limits direct extrapolation. X chromosome linked genes involved in bile acid metabolism may also contribute to baseline expression differences between males and females. Additionally, hepatic developmental trajectories, including the maturation of drug metabolizing enzymes, are known to follow sex dependent patterns in infancy, and similar processes may influence bile acid transport systems; however, functional studies are needed to clarify these possibilities.
Third, regarding the potential basis for the vitamin E findings, vitamin E is a lipid soluble nutrient whose absorption relies on intact bile acid mediated micelle formation. Altered bile acid composition resulting from NTCP deficiency could interfere with this process and reduce vitamin E uptake efficiency. The iron association is less straightforward but may involve shared regulatory pathways between bile acid metabolism and iron homeostasis, such as the hepcidin-ferroportin axis. However, this connection remains poorly understood and requires further investigation.
Our sex-stratified analysis uncovered fundamental differences in how genetic variants manifest biochemically between male and female pediatric patients. Sex emerged as a potential effect modifier in our regression models; however, this observation requires independent validation in NTCPD, revealing distinct pathophysiological pathways.
In male patients, the most significant associations emerged specifically in the fully adjusted model, suggesting that demographic factors substantially confound genotype-phenotype relationships in this group. The association between homozygous and compound heterozygous variants with TBA levels became highly significant in the fully adjusted model with substantially larger effect sizes. This pattern indicates that the genetic contribution to TBA elevation in males becomes fully apparent only after accounting for demographic confounders. More importantly, all three genotypes demonstrated highly significant positive associations with AST levels exclusively in the fully adjusted model. This repre
Female patients exhibited a distinct pattern: While associations were often apparent in less stringent models, they showed particular vulnerability in nutritional parameters. Notably, the TBA associations observed in initial models attenuated after full adjustment. However, female patients demonstrated more highly significant differences in serum vitamin E and iron levels. Both homozygous and compound heterozygous variants showed significant inverse associations with vitamin E levels in the fully adjusted model, with effect sizes exceeding those in males. Furthermore, homozygous variants exhibited consistent inverse associations with iron levels across all models; the inverse genotype-iron association reached statistical significance only in female patients, a finding that merits further mechanistic investigation.
These sex-stratified findings have important clinical implications. For male patients, our findings support the hypothe
While providing valuable insights into NTCPD, our study has several limitations that should be considered when interpreting the results and designing future research. The present cohort, while substantial for a rare disorder, presents limitations in subgroup analyses, particularly when examining compound heterozygous cases across multiple strata. This restriction reduces statistical power for detecting genuine differences and limits deeper genotype-phenotype correlation analyses. Furthermore, the exclusive focus on Yunnan Province may limit generalizability to other populations with different demographic characteristics. Although the ethnic diversity of our cohort is a unique strength, the small sizes of individual minority groups preclude robust statistical inference regarding ethnicity as an independent modifier of genotype-phenotype relationships. Moreover, the observed geographic clustering of cases in minority-concentrated prefectures is intriguing but remains hypothesis-generating. Larger, multicenter collaborative studies are needed to disentangle genetic from environmental contributions to phenotypic variation in NTCPD. Furthermore, the observational nature of our study design inherently restricts causal inference. Although we demonstrate strong associations between genotypes and biochemical phenotypes, definitive proof of causality requires functional validation through experimental models. Additionally, while we adjusted for major known confounders, residual confounding from unmeasured factors cannot be excluded. Our genetic analysis has specific limitations. The functional consequences of the identified variants remain largely uncharacterized. Although we classified variants according to the ACMG guidelines, none have undergone functional validation through bile acid transport assays. Moreover, our targeted sequencing approach may have missed deep intronic mutations, regulatory variants, or larger structural variations that could explain phenotypes in some heterozygous cases. Finally, the clinical follow-up duration represents another limitation. While NTCPD is generally considered benign, its true long-term outcomes remain incompletely defined. Our medium-term follow-up data cannot address important questions about potential long-term sequelae, which will require dedicated longitudinal studies extending over decades.
This 5-year retrospective cohort study characterizes the dynamic genotype–phenotype relationship of NTCPD in a multi-ethnic pediatric population. The disorder manifested with marked early hypercholanemia and followed a self-limiting biochemical course that resolved with age. Genetically, the c.800C>T variant was predominant, although notable allelic heterogeneity was observed, including the identification of several previously unreported variants. Importantly, robust associations were demonstrated between specific genotypes and both TBA and vitamin E levels, which remained significant after full covariate adjustment. The results also revealed clear sex-specific phenotypic differences, namely hepatocellular stress patterns in males and nutritional metabolic imbalances in females. These results emphasize the roles of demographics and biological sex in influencing disease presentation and support the need for comprehensive genetic screening, together with a hypothesis-generating framework that may inform future study design, rather than a validated clinical strategy in NTCPD.
| 1. | Vaz FM, Paulusma CC, Huidekoper H, de Ru M, Lim C, Koster J, Ho-Mok K, Bootsma AH, Groen AK, Schaap FG, Oude Elferink RP, Waterham HR, Wanders RJ. Sodium taurocholate cotransporting polypeptide (SLC10A1) deficiency: conjugated hypercholanemia without a clear clinical phenotype. Hepatology. 2015;61:260-267. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 193] [Cited by in RCA: 175] [Article Influence: 15.9] [Reference Citation Analysis (0)] |
| 2. | Qiu JW, Deng M, Cheng Y, Atif RM, Lin WX, Guo L, Li H, Song YZ. Sodium taurocholate cotransporting polypeptide (NTCP) deficiency: Identification of a novel SLC10A1 mutation in two unrelated infants presenting with neonatal indirect hyperbilirubinemia and remarkable hypercholanemia. Oncotarget. 2017;8:106598-106607. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 33] [Cited by in RCA: 35] [Article Influence: 3.9] [Reference Citation Analysis (0)] |
| 3. | Deng M, Mao M, Guo L, Chen FP, Wen WR, Song YZ. Clinical and molecular study of a pediatric patient with sodium taurocholate cotransporting polypeptide deficiency. Exp Ther Med. 2016;12:3294-3300. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 46] [Cited by in RCA: 41] [Article Influence: 4.1] [Reference Citation Analysis (1)] |
| 4. | You JY, Xiong LY, Wu MF, Fan JS, Fu QH, Qiu MH. Genetic variation features of neonatal hyperbilirubinemia caused by inherited diseases. World J Clin Pediatr. 2024;13:98462. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in RCA: 2] [Reference Citation Analysis (0)] |
| 5. | Liu R, Chen C, Xia X, Liao Q, Wang Q, Newcombe PJ, Xu S, Chen M, Ding Y, Li X, Liao Z, Li F, Du M, Huang H, Dong R, Deng W, Wang Y, Zeng B, Pan Q, Jiang D, Zeng H, Sham P, Cao Y, Maxwell PH, Gao ZL, Peng L, Wang Y. Homozygous p.Ser267Phe in SLC10A1 is associated with a new type of hypercholanemia and implications for personalized medicine. Sci Rep. 2017;7:9214. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 45] [Cited by in RCA: 45] [Article Influence: 5.0] [Reference Citation Analysis (0)] |
| 6. | Xu JX, Lin F, Wu YH, Chen ZK, Ma YB, Yang LY. Etiology analysis for term newborns with severe hyperbilirubinemia in eastern Guangdong of China. World J Clin Cases. 2023;11:2443-2451. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in CrossRef: 3] [Cited by in RCA: 9] [Article Influence: 3.0] [Reference Citation Analysis (0)] |
| 7. | Ho RH, Leake BF, Roberts RL, Lee W, Kim RB. Ethnicity-dependent polymorphism in Na+-taurocholate cotransporting polypeptide (SLC10A1) reveals a domain critical for bile acid substrate recognition. J Biol Chem. 2004;279:7213-7222. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 162] [Cited by in RCA: 143] [Article Influence: 6.5] [Reference Citation Analysis (1)] |
| 8. | Wang M, Chen T, Chen R, Bi Z, Peng J, Shao Q, Li J. Neonatal jaundice caused by compound mutations of SLC10A1 and a novel UGT1A1 gene. Clin Res Hepatol Gastroenterol. 2024;48:102340. [RCA] [PubMed] [DOI] [Full Text] [Cited by in RCA: 2] [Reference Citation Analysis (0)] |
| 9. | Li H, Deng M, Guo L, Qiu JW, Lin GZ, Long XL, Xiao XM, Song YZ. Clinical and molecular characterization of four patients with NTCP deficiency from two unrelated families harboring the novel SLC10A1 variant c.595A>C (p.Ser199Arg). Mol Med Rep. 2019;20:4915-4924. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 3] [Cited by in RCA: 8] [Article Influence: 1.1] [Reference Citation Analysis (0)] |
| 10. | Zhou J, Yang C, Zhu W, Chen S, Zeng Y, Wang J, Zhao H, Chen Y, Lin F. Identification of Genetic Risk Factors for Neonatal Hyperbilirubinemia in Fujian Province, Southeastern China: A Case-Control Study. Biomed Res Int. 2018;2018:7803175. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 20] [Cited by in RCA: 18] [Article Influence: 2.3] [Reference Citation Analysis (0)] |
| 11. | Li H, Chen R, Lin GZ, Lin WX, Yaqub MR, Song YZ. Molecular Epidemiology of Na(+)-Taurocholate Cotransporting Polypeptide Deficiency in Guangdong Province, China: A Pilot Study by Screening for Four Prevalent Variants of the Causative Gene SLC10A1. Front Genet. 2022;13:874379. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in RCA: 2] [Reference Citation Analysis (0)] |
| 12. | Deng LJ, Ouyang WX, Liu R, Deng M, Qiu JW, Yaqub MR, Raza MA, Lin WX, Guo L, Li H, Chen FP, Ouyang Y, Huang YG, Huang YJ, Long XL, Huang XL, Li SJ, Song YZ. Clinical characterization of NTCP deficiency in paediatric patients : A case-control study based on SLC10A1 genotyping analysis. Liver Int. 2021;41:2720-2728. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 4] [Cited by in RCA: 15] [Article Influence: 3.0] [Reference Citation Analysis (1)] |
| 13. | Dawson PA. Hepatic bile acid uptake in humans and mice: Multiple pathways and expanding potential role for gut-liver signaling. Hepatology. 2017;66:1384-1386. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 12] [Cited by in RCA: 16] [Article Influence: 1.8] [Reference Citation Analysis (0)] |
| 14. | Zong XN, Li H. Construction of a new growth references for China based on urban Chinese children: comparison with the WHO growth standards. PLoS One. 2013;8:e59569. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 89] [Cited by in RCA: 140] [Article Influence: 10.8] [Reference Citation Analysis (0)] |
| 15. | Sun WJ, Yu F. [Clinical analysis of two pediatric patients with sodium taurocholate cotransporting polypeptide deficiency]. Zhongguo Yiyao Daobao. 2020;17:185-188. |
| 16. | Sargiacomo C, El-Kehdy H, Pourcher G, Stieger B, Najimi M, Sokal E. Age-dependent glycosylation of the sodium taurocholate cotransporter polypeptide: From fetal to adult human livers. Hepatol Commun. 2018;2:693-702. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 8] [Cited by in RCA: 12] [Article Influence: 1.5] [Reference Citation Analysis (0)] |
| 17. | Song YZ. [Recent advances in the diagnosis and treatment of pediatric genetic cholestasis]. Zhonghua Shiyong Erke Linchuang Zazhi. 2017;32:561-565. |
| 18. | Yan YY, Wang MX, Gong JY, Liu LL, Setchell KDR, Xie XB, Wang NL, Li W, Wang JS. Abnormal Bilirubin Metabolism in Patients With Sodium Taurocholate Cotransporting Polypeptide Deficiency. J Pediatr Gastroenterol Nutr. 2020;71:e138-e141. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 11] [Cited by in RCA: 14] [Article Influence: 2.3] [Reference Citation Analysis (0)] |
| 19. | Du L, Zhou M, Yang J, Du M, Jiang M, Xiong L. Clinical and genetic characteristics of children with sodium taurocholate cotransporting poly-peptide deficiency. Front Pediatr. 2026;14:1784165. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in RCA: 1] [Reference Citation Analysis (1)] |
| 20. | Dong C, Zhang BP, Wang H, Xu H, Zhang C, Cai ZS, Wang DW, Shu SN, Huang ZH, Luo XP. Clinical and histopathologic features of sodium taurocholate cotransporting polypeptide deficiency in pediatric patients. Medicine (Baltimore). 2019;98:e17305. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 25] [Cited by in RCA: 26] [Article Influence: 3.7] [Reference Citation Analysis (3)] |
| 21. | Van Herpe F, Waterham HR, Adams CJ, Mannens M, Bikker H, Vaz FM, Cassiman D. NTCP deficiency and persistently raised bile salts: an adult case. J Inherit Metab Dis. 2017;40:313-315. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 25] [Cited by in RCA: 27] [Article Influence: 3.0] [Reference Citation Analysis (0)] |
| 22. | Li H, Qiu JW, Lin GZ, Deng M, Lin WX, Cheng Y, Song YZ. [Clinical and genetic analysis of a pediatric patient with sodium taurocholate cotransporting polypeptide deficiency]. Zhongguo Dang Dai Er Ke Za Zhi. 2018;20:279-284. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in RCA: 8] [Reference Citation Analysis (1)] |
| 23. | Wang NL, Lu YL, Zhang P, Zhang MH, Gong JY, Lu Y, Xie XB, Qiu YL, Yan YY, Wu BB, Wang JS. A Specially Designed Multi-Gene Panel Facilitates Genetic Diagnosis in Children with Intrahepatic Cholestasis: Simultaneous Test of Known Large Insertions/Deletions. PLoS One. 2016;11:e0164058. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 26] [Cited by in RCA: 35] [Article Influence: 3.5] [Reference Citation Analysis (0)] |
| 24. | Zhang Q, He Z, Liu Z, Gong L. Integrated plasma and liver gas chromatography mass spectrometry and liquid chromatography mass spectrometry metabolomics to reveal physiological functions of sodium taurocholate cotransporting polypeptide (NTCP) with an Ntcp knockout mouse model. J Chromatogr B Analyt Technol Biomed Life Sci. 2021;1165:122531. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 13] [Cited by in RCA: 12] [Article Influence: 2.4] [Reference Citation Analysis (0)] |
| 25. | Mao F, Wang MX, Hou X, Zhou Z, Yan YY, Fang LJ, Tan Z, Fang WY, Liu T, He W, Li C, Xie XB, Lu SQ, Sui J, Wang F, Han J, Wang JS, Li W. NTCP Deficiency Causes Gallbladder Abnormalities in Mice and Human Beings. Cell Mol Gastroenterol Hepatol. 2021;11:831-839. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 11] [Cited by in RCA: 12] [Article Influence: 2.0] [Reference Citation Analysis (0)] |
| 26. | Mao F, Liu T, Hou X, Zhao H, He W, Li C, Jing Z, Sui J, Wang F, Liu X, Han J, Borchers CH, Wang JS, Li W. Increased sulfation of bile acids in mice and human subjects with sodium taurocholate cotransporting polypeptide deficiency. J Biol Chem. 2019;294:11853-11862. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 17] [Cited by in RCA: 37] [Article Influence: 5.3] [Reference Citation Analysis (1)] |
| 27. | Lin H, Qiu JW, Rauf YM, Lin GZ, Liu R, Deng LJ, Deng M, Song YZ. Sodium Taurocholate Cotransporting Polypeptide (NTCP) Deficiency Hidden Behind Citrin Deficiency in Early Infancy: A Report of Three Cases. Front Genet. 2019;10:1108. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 3] [Cited by in RCA: 11] [Article Influence: 1.6] [Reference Citation Analysis (0)] |
| 28. | Liu HY, Li M, Li Q. De novo mutation loci and clinical analysis in a child with sodium taurocholate cotransport polypeptide deficiency: A case report. World J Clin Cases. 2021;9:11487-11494. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in RCA: 3] [Reference Citation Analysis (0)] |
| 29. | Slijepcevic D, Roscam Abbing RLP, Katafuchi T, Blank A, Donkers JM, van Hoppe S, de Waart DR, Tolenaars D, van der Meer JHM, Wildenberg M, Beuers U, Oude Elferink RPJ, Schinkel AH, van de Graaf SFJ. Hepatic uptake of conjugated bile acids is mediated by both sodium taurocholate cotransporting polypeptide and organic anion transporting polypeptides and modulated by intestinal sensing of plasma bile acid levels in mice. Hepatology. 2017;66:1631-1643. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 90] [Cited by in RCA: 125] [Article Influence: 13.9] [Reference Citation Analysis (1)] |
| 30. | Chinese Society of Hepatology; Chinese Medical Association. [Guidelines on the management of cholestasis liver diseases (2021)]. Zhonghua Gan Zang Bing Za Zhi. 2022;30:253-263. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in RCA: 6] [Reference Citation Analysis (0)] |
| 31. | Song YZ, Deng M. [Sodium taurocholate cotransporting polypeptide deficiency manifesting as cholestatic jaundice in early infancy: a complicated case study]. Zhongguo Dang Dai Er Ke Za Zhi. 2017;19:350-354. [RCA] [PubMed] [DOI] [Full Text] [Cited by in RCA: 8] [Reference Citation Analysis (0)] |
| 32. | Hagenbuch B, Dawson P. The sodium bile salt cotransport family SLC10. Pflugers Arch. 2004;447:566-570. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 216] [Cited by in RCA: 194] [Article Influence: 8.8] [Reference Citation Analysis (1)] |
| 33. | Pan W, Song IS, Shin HJ, Kim MH, Choi YL, Lim SJ, Kim WY, Lee SS, Shin JG. Genetic polymorphisms in Na+-taurocholate co-transporting polypeptide (NTCP) and ileal apical sodium-dependent bile acid transporter (ASBT) and ethnic comparisons of functional variants of NTCP among Asian populations. Xenobiotica. 2011;41:501-510. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 78] [Cited by in RCA: 86] [Article Influence: 5.7] [Reference Citation Analysis (1)] |
| 34. | Song F, Chen Y, Chen L, Li H, Cheng X, Wu W. Association of Elevated Maternal Serum Total Bile Acids With Low Birth Weight and Intrauterine Fetal Growth Restriction. JAMA Netw Open. 2021;4:e2117409. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 9] [Cited by in RCA: 31] [Article Influence: 6.2] [Reference Citation Analysis (0)] |
| 35. | Chen R, Deng M, Rauf YM, Lin GZ, Qiu JW, Zhu SY, Xiao XM, Song YZ. Intrahepatic Cholestasis of Pregnancy as a Clinical Manifestation of Sodium-Taurocholate Cotransporting Polypeptide Deficiency. Tohoku J Exp Med. 2019;248:57-61. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 12] [Cited by in RCA: 12] [Article Influence: 1.7] [Reference Citation Analysis (0)] |
| 36. | Thompson RJ, Arnell H, Artan R, Baumann U, Calvo PL, Czubkowski P, Dalgic B, D'Antiga L, Durmaz Ö, Fischler B, Gonzalès E, Grammatikopoulos T, Gupte G, Hardikar W, Houwen RHJ, Kamath BM, Karpen SJ, Kjems L, Lacaille F, Lachaux A, Lainka E, Mack CL, Mattsson JP, McKiernan P, Özen H, Rajwal SR, Roquelaure B, Shagrani M, Shteyer E, Soufi N, Sturm E, Tessier ME, Verkade HJ, Horn P. Odevixibat treatment in progressive familial intrahepatic cholestasis: a randomised, placebo-controlled, phase 3 trial. Lancet Gastroenterol Hepatol. 2022;7:830-842. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 95] [Cited by in RCA: 146] [Article Influence: 36.5] [Reference Citation Analysis (0)] |
| 37. | Richards S, Aziz N, Bale S, Bick D, Das S, Gastier-Foster J, Grody WW, Hegde M, Lyon E, Spector E, Voelkerding K, Rehm HL; ACMG Laboratory Quality Assurance Committee. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Med. 2015;17:405-424. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 27361] [Cited by in RCA: 26724] [Article Influence: 2429.5] [Reference Citation Analysis (32)] |