Genotype-phenotype correlations of fasting C-peptide and lipids in HNF1A-MODY: a single-center series and multi-center cross-sectional analysis in Chinese population

Background: HNF1A-MODY is one of the most prevalent subtypes of maturity-onset diabetes of the young (MODY). Individuals with HNF1A-MODY display considerable clinical heterogeneity, potentially attributable to specific mutation sites. However, in the Chinese population, the relationship between distinct mutation sites and clinical manifestations remains to be investigated.

Methods: In the initial analysis, 23 HNF1A-MODY patients diagnosed at the Department of Endocrinology, Qilu Hospital were included. These patients were followed up regularly to monitor glycemic control status and the progression of complications. In the subsequent analysis, baseline information of 113 Chinese HNF1A-MODY retrieved from public databases were further enrolled. Analysis of covariance was conducted to investigate the genotype-phenotype associations.

Results: This study included a total of 136 patients. Among the 23 from Qilu Hospital, 22 distinct HNF1A gene variants were identified, including 8 novel ones. After excluding cases classified as “variant of uncertain significance”, the analysis showed that the median age of onset was earliest in patients with DNA-binding domain mutations (15.70 years), compared to the dimerization or transactivation domain mutations (p = 0.044). Fasting C-peptide levels were markedly lower in the dimerization domain and DNA-binding domain group (p = 0.005). Patients with DNA-binding domain mutations demonstrated lower low-density lipoprotein cholesterol (p = 0.049) and total cholesterol (p = 0.016) levels, but higher high-density lipoprotein cholesterol (p = 0.036) levels. Analysis of covariance indicated that mutations in the dimerization domain (mean difference = -0.757, p = 0.001) and DNA-binding domain (mean difference = -0.331, p = 0.041) were independently associated with lower fasting C-peptide, and DNA-binding domain mutations were also associated with low-density lipoprotein cholesterol (mean difference = -0.554, p = 0.015) and higher high-density lipoprotein cholesterol (mean difference = 0.224, p = 0.015) levels, whereas the other domain mutations showed no statistically significant associations.

Conclusion: This study revealed the correlation between HNF1A mutation regions and pancreatic islet function as well as blood lipids in Chinese HNF1A-MODY patients, thereby underscoring the importance of early genetic identification in formulating individualized therapeutic strategies to improve prognosis.

1 Introduction

Maturity-onset diabetes of the young (MODY) is a monogenic form of diabetes characterized by autosomal dominant inheritance. To date, 14 distinct MODY subtypes have been identified, among which HNF1A-MODY, caused by HNF1A gene mutations, represents one of the most prevalent subtypes, accounting for 30-70% of all MODY cases (1). However, recent study indicates that within the Chinese MODY population, the proportion of HNF1A-MODY cases is 16.03% (2). The HNF1A gene encodes hepatocyte nuclear factor 1α (HNF1α), a transcriptional regulator that activates insulin gene expression in pancreatic β-cells. Structurally, HNF1α consists of three functional domains (3): an N-terminal dimerization domain (residues 1-32), a DNA-binding domain (residues 91-281), and a C-terminal transactivation domain (residues 282-631). HNF1A-MODY is characterized by defective insulin secretion and progressive hyperglycemia, with a high prevalence of microvascular complications (4). The clinical phenotypes of HNF1A-MODY patients exhibit considerable heterogeneity and overlap with both type 1 diabetes mellitus (T1DM) and type 2 diabetes mellitus (T2DM) (5), often leading to misdiagnosis. Therefore, early and accurate genetic diagnosis followed by precision therapy is essential for improving clinical management and long-term outcomes.

The marked heterogeneity in the clinical manifestations of HNF1A-MODY may be attributed to the variability in mutation loci (6). Previous studies have indicated that both the type and location of HNF1A mutations influence the age at diagnosis (7, 8), response to sulfonylurea therapy (9), and risk of diabetic complications (10). Additionally, genetic modifiers and exposure to intrauterine hyperglycemia may contribute to the variability in HNF1A-MODY diagnosis age (11, 12). However, the current researches were mainly based on data from Caucasian and other non-Chinese Asian populations, while systematic analyses of Chinese HNF1A-MODY patients remain scarce. Additionally, the impact of domain-specific HNF1A mutations on metabolic indicators such as pancreatic function and lipids has not been thoroughly investigated.

This study analyzed patients with HNF1A-MODY admitted to Qilu Hospital of Shandong University, along with reported Chinese cases from published literature, to investigate the genotype-phenotype correlations associated with domain-specific HNF1A mutations. This study aims to provide a basis for developing personalized management strategies by investigating the association between specific mutation sites and patients’ clinical manifestations, thereby improving the prognosis of HNF1A mutation carriers.

2 Materials and methods

2.1 Subjects

This study included patients from two sources. First, we enrolled 23 HNF1A-MODY patients with HNF1A gene variants identified through gene sequencing, who were treated at the Department of Endocrinology, Qilu Hospital of Shandong University from January 2021 to April 2025. These patients were followed up regularly to monitor their glycemic control status and the progression of complications. Fasting plasma glucose < 7 mmol/L and glycated hemoglobin < 6.5% are defined as good glycemic control (4), whereas failure to meet either criterion indicates poor glycemic control.

Additionally, 113 patients were identified through a review of the literature. Using the keywords and their Chinese equivalents, including “Maturity-Onset Diabetes of the Young, Type 3”, “MODY, Type 3”, “MODY3”, “HNF1A-MODY”, “HNF1-alpha MODY”, “hepatic nuclear factor 1 alpha”, “HNF1A”, and “HNF1-alpha”, a comprehensive literature search was conducted in databases including China National Knowledge Infrastructure (CNKI), Wanfang Data Knowledge Service Platform, China Science and Technology Journal Database (CQVIP), Chinese Medical Journal Network (CMJN), PubMed and Web of Science to identify relevant literature on Chinese patients with HNF1A-MODY. The search period was set from inception of the database to April 30, 2025. The detailed search strategy is provided in Supplementary Appendix S1.

Literature screening followed the PRISMA guidelines. Initially, duplicates were removed using EndNote software, followed by a manual review of authors and affiliations to exclude duplicate reports. A two-stage screening was then performed: firstly, titles and abstracts were reviewed to exclude obviously irrelevant studies, non-human research, and non-original case reports (e.g., reviews, commentaries). Secondly, full texts of the initially screened articles were reviewed, and studies were selected for inclusion according to predetermined inclusion and exclusion criteria. Finally, we examined the participants’ dates of birth reported by the included studies. Cases were screened based on date of birth and sex to identify and exclude duplicate subjects. Inclusion Criteria: (1) Chinese HNF1A-MODY patient was confirmed by gene sequencing; (2) explicitly reported the location and type of HNF1A mutation; (3) reported at least three of the following parameters: age of onset, fasting plasma glucose (FPG), 2-hour postprandial glucose (2hPG), fasting C-peptide (FCP), 2-hour postprandial C-peptide (2hPCP), glycated hemoglobin (HbA_1c), lipid profile, antidiabetic therapies, or diabetic complications. Exclusion Criteria: (1) unavailable or overlapping data; (2) lack of mutation information or key clinical characteristics. The screening process is presented in a PRISMA flowchart (Figure 1).

Figure 1

Figure 1. Literature flow diagram.

2.2 Clinical data collection

Clinical data, including the gender, age of onset, age of diagnosis, duration of diabetes, presence of classic diabetes symptoms (including polydipsia, polyuria, polyphagia, and unexplained weight loss — the presence of any one of these symptoms is considered indicative of classic hyperglycemic symptoms (13).), family history, history of smoking and drinking, body mass index (BMI), FPG, 2hPG, fasting insulin (FINS), 2-hour postprandial insulin (2hPINS), FCP (measured by chemiluminescence immunoassay after an 8-hour fasting period), 2hPCP, HbA_1c, islet autoantibody (including IAA, ICA, IA-2A, and GAD), triglyceride (TG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), uric acid (UA), hypersensitive C-reactive protein (hs-CRP), glucosuria, diabetic complications — diabetic retinopathy is screened through fundus examination, diabetic peripheral neuropathy is screened via nerve conduction velocity and vibration perception threshold testing and macrovascular complications are assessed using carotid ultrasound, history of underlying diseases (hypertension, coronary heart disease, etc.), antidiabetic therapies, history of other medications (including lipid-lowering therapy, etc.), and the location and type of HNF1A mutation were systematically collected. To ensure comparability across data sources, variables with inconsistent measurement units were standardized prior to pooled analysis.

2.3 Statistical analysis

Statistical analysis was performed using SPSS 27.0 software. Normally distributed numerical data were expressed as mean ± standard deviation and were compared using independent sample t-test or one-way ANOVA, followed by Bonferroni-adjusted post-hoc test. Non-normally distributed data were presented as M (Q1, Q3), and compared using the Mann-Whitney U test or Kruskal-Wallis H test, with post-hoc pairwise comparisons conducted using Dunn’s test. Categorical data were expressed as n (%), and group comparisons were performed using the chi-square test or Fisher’s exact test. To investigate whether different mutation domains are independently associated with specific clinical manifestations after adjusting for potential confounding factors, analysis of covariance (ANCOVA) was employed. To control for variation across different data sources, we incorporated data source as a random effect and performed analyses using linear mixed model. A p-value < 0.05 was considered statistically significant.

3 Results

3.1 Baseline characteristics of HNF1A-MODY patients in Qilu cohort

This study enrolled 23 patients diagnosed with HNF1A-MODY at Qilu Hospital (Tables 1, 2). Whole-exome sequencing identified 22 distinct HNF1A variants, including 8 novel ones, which were classified as pathogenic, likely pathogenic, or variant of uncertain significance according to the standards and guidelines recommended by the American College of Medical Genetics (ACMG) (Table 1). In terms of domain distribution, 2 variants were located in the dimerization domain, 9 in the DNA-binding domain, and 12 in the transactivation domain. Mutations were most frequently located in exons 2 and 4, accounting for 43.48% (10/23). The main variant types were missense (52.17%, 12/23) and frameshift mutations (26.09%, 6/23). Clinically (Table 2), all patients presented with early-onset diabetes characteristic of HNF1A-MODY, with an age of onset ranging from 9 to 29 years, all < 35 years. Most patients (81.82%, 18/22) had a BMI < 24 kg/m². A positive family history of diabetes was reported in 82.61% (19/23) of cases. At disease onset, 43.48% (10/23) of the patients manifested classic symptoms of polydipsia, polyuria, polyphagia, weight loss. 2 patients tested positive for islet autoantibodies. Regarding complications, 40.91% (9/22) had developed at least one diabetic complication, with diabetic peripheral neuropathy (36.36%, 8/22) and diabetic retinopathy (22.72%, 5/22) being the most common. In terms of glucose-lowering regimens, 8.70% (2/23) of patients were managed with lifestyle intervention alone; 56.52% (13/23) received oral antidiabetic drug monotherapy; and the remaining 34.78% (8/23) required insulin, with 8.70% (2/23) on insulin monotherapy and 26.08% (6/23) on a combination of oral agents and insulin. Given that the pathogenicity of novel variants had not yet been functionally validated, we excluded cases classified as “variant of uncertain significance” from subsequent analyses. A comparison of the clinical characteristics among different mutation domains was provided in Supplementary Table 2.

Table 1

Table 1. The genotypes of patients with HNF1A-MODY in Qilu Hospital.

Table 2

Table 2. Clinical and laboratory data of patients with HNF1A-MODY in Qilu hospital.

3.2 Follow-up data of HNF1A-MODY patients at Qilu hospital

During a median follow-up of 2 years, 2 patients were lost to follow-up. Among the remaining 21 patients, 71.43% (15/21) achieved good glycemic control, defined as FPG < 7 mmol/L and HbA_1c < 6.5% (4). Within this limited observation window, two new cases of diabetic peripheral neuropathy and one case of diabetic retinopathy were documented. After excluding cases classified as “variant of uncertain significance”, the analysis of hypoglycemic regimens (Figure 2) showed that sulfonylureas were the most common medication among patients with good glycemic control (40.00%), whereas insulin usage was highest in the poor control group. Comparative analysis based on glycemic control status (Table 3) demonstrated that the poor glycemic control group had higher baseline BMI (p = 0.010) and LDL-C (p = 0.042) than the good control group. Regarding hypoglycemic therapy, the good glycemic control group was mainly managed with oral hypoglycemic monotherapy (p = 0.025), while the poor control group more frequently required insulin-based combination therapy (p = 0.039). No significant intergroup differences were observed in age of onset, mutation domain distribution, blood glucose, HbA_1c, or C-peptide levels at baseline.

Figure 2

Figure 2. Analysis of hypoglycemic drugs for HNF1A-MODY patients. (A) all types of hypoglycemic drugs for patients with HNF1A-MODY, (B) the proportion of hypoglycemic drugs in HNF1A-MODY patients with good glycemic control, (C) the proportion of hypoglycemic drugs in HNF1A-MODY patients with poor glycemic control. Abbreviations: DPP-4i, dipeptidyl peptidase 4 inhibitor; AGI, alpha glucosidase inhibitor; SGLT-2i, sodium-glucose transporter 2 inhibitor.

Table 3

Table 3. Baseline clinical characteristics of patients with HNF1A-MODY by glycemic control status.

3.3 Clinical manifestations of Chinese HNF1A-MODY patients

Following the description of clinical manifestations from the Qilu Hospital cohort, we performed a detailed genotype-phenotype analysis in Chinese patients with HNF1A-MODY. Given that the pathogenicity of novel variants had not yet been functionally validated, we excluded cases classified as “variant of uncertain significance” from subsequent analyses. After a comprehensive literature review and exclusion of cases with incomplete clinical data or duplicate reports, baseline information of 113 HNF1A-MODY patients from published literature were included. Combined with the baseline data of the Qilu Hospital cohort (Table 4), the median age of onset was 19.00 years in the dimerization domain group, 15.70 years in the DNA-binding domain group, and 22.00 years in the transactivation domain group, indicating an earlier disease onset in patients with DNA-binding domain mutations (p = 0.044). Pancreatic islet function varied by domain. FCP levels were markedly lower in in the dimerization domain (1.00 ± 0.41 ng/ml) and DNA-binding domain group (1.08 ± 0.57 ng/ml) than the transactivation domain group (1.51 ± 0.85 ng/ml, p = 0.005). 2hPCP levels in the DNA-binding domain group [2.54 (1.56, 3.13) ng/ml] were also significantly lower than in the transactivation domain group [3.61 (1.83, 5.72) ng/ml, p = 0.041]. Lipid profiles also showed significant differences. HDL-C levels were higher in the DNA-binding domain group (1.45 ± 0.46 mmol/L, p = 0.036) than in the dimerization domain (1.27 ± 0.47 mmol/L) and transactivation domain group (1.20 ± 0.30 mmol/L). Conversely, the levels of LDL-C (DNA-binding domain group 2.47 ± 0.79 mmol/L vs. dimerization domain group 2.96 ± 0.46 mmol/L vs. transactivation domain group 2.98 ± 0.96 mmol/L, p = 0.049) and TC (DNA-binding domain group 4.15 ± 0.98 mmol/L vs. dimerization domain group 4.99 ± 0.75 mmol/L vs. transactivation domain group 4.93 ± 1.37 mmol/L, p = 0.016) showed different patterns. Regarding hypoglycemic therapy, patients with dimerization and transactivation domain variants were more frequently managed with diet and exercise alone (p = 0.008), while oral hypoglycemic agents were most common in the DNA-binding domain group (61.3%, p = 0.019). No significant differences were observed among groups in classic diabetes symptoms, BMI, FPG, 2hPG, HbA_1c, TG, UA, insulin resistance indices, islet autoantibody, family history, or diabetic chronic complications.

Table 4

Table 4. Clinical characteristics of Chinese HNF1A-MODY patients by different domains.

3.4 DNA-binding domain mutations independently associated with lower FCP, LDL-C and higher HDL-C levels

Our previous findings demonstrated that FCP levels were significantly lower in the dimerization domain and DNA-binding domain groups compared to the transactivation domain group. LDL-C and TC levels in the DNA-binding domain group were also significantly lower than those in the other two groups, while HDL-C was the opposite. To investigate whether the effects of mutation domains were independent of potential confounding factors, we enrolled an additional control group of 50 adolescents with type 2 diabetes who were randomly selected and conducted ANCOVA. The results showed that after adjusting for age of onset, gender, family history, duration of diabetes, BMI, HbA_1c, TyG index, and prior antidiabetic therapies, the mutation domain remained an independent influencing factor for FCP levels. Mutations in the dimerization domain (mean difference = -0.757, p = 0.001) and DNA-binding domain (mean difference = -0.331, p = 0.041) were independently associated with lower FCP levels, while the effect of the transactivation domain did not reach statistical significance (Table 5A). In addition, after adjusting for age of onset, gender, history of smoking and drinking, BMI, HbA_1c, TG, and lipid-lowering therapy, DNA-binding domain mutations were associated with lower LDL-C levels (mean difference = -0.554, p = 0.015) (Supplementary Table 3). DNA-binding domain mutations were also associated with higher HDL-C levels (mean difference = 0.224, p = 0.015) after adjusting for age of onset, gender, history of smoking and drinking, HbA_1c, TG, TC, and lipid-lowering therapy. (Table 5B). In contrast, DNA-binding domain mutations were not independent influencing factors for lower TC levels (Supplementary Table 3).

Table 5A

Table 5A. Analysis of covariance of FCP levels in HNF1A-MODY patients.

Table 5B

Table 5B. Analysis of covariance of HDL-C levels in HNF1A-MODY patients.

Given the high proportion of insulin-treated patients across the included studies, to verify model stability, we further excluded clinically atypical cases with “early or sustained insulin dependence” (insulin initiation within 10 years of disease duration) and repeated the ANCOVA. The results showed that dimerization domain and DNA-binding domain mutations remained associated with lower FCP, and DNA-binding domain mutations continued to be linked to higher HDL-C levels (Supplementary Table 4).

Furthermore, we constructed linear mixed models with FCP, LDL-C, and HDL-C as outcome variables. The mutation domains were included as fixed effects, while data source was incorporated as a random intercept to account for variability across sources. After controlling for variation among different data sources, the mutation domain retained a statistically significant association with FCP, LDL-C, and HDL-C (Supplementary Table 5).

4 Discussion

This study analyzed the clinical characteristics of Chinese patients with HNF1A-MODY and revealed associations between mutations in different functional domains and clinical phenotypes. This multicenter cross-sectional analysis of Chinese population revealed that patients with DNA-binding domain variants exhibited a significantly younger age of onset, as well as lower FCP, LDL-C, TC levels, and higher HDL-C levels compared to patients with dimerization or transactivation domain mutations. Short-term follow-up revealed that HNF1A-MODY patients with poor glycemic control might had higher baseline BMI and LDL-C levels. ANCOVA indicated that DNA-binding domain mutations were independently associated with lower FCP, LDL-C and higher HDL-C levels.

HNF1A-MODY is a monogenic form of diabetes caused by pathogenic variants in the HNF1A gene. The HNF1A gene, located on chromosome 12q24 (14), encodes the hepatocyte nuclear factor 1 alpha (HNF1α) protein, which is predominantly expressed in the liver, pancreas, kidney, and gastrointestinal tract (15). HNF1α plays a critical role in regulating genes involved in insulin secretion, glucose metabolism, and lipid homeostasis. To date, approximately 500 HNF1A variants have been reported in association with MODY phenotypes (16). The HNF1A gene exhibits high polymorphism, with the highest mutation rate in exons 2 and 4 (17, 18). A retrospective study of patients with HNF1A-MODY reported 492 different mutations (10), and missense mutations (227, 46.1%) and frameshift mutations (198, 40.2%) were the most common mutations. Whereas in the Qilu Hospital cohort, 22 distinct HNF1A gene variants were identified, eight of which were novel, mainly located in exons 2 and 4, with missense and frameshift mutations being the main types, consistent with the previous findings.

The HNF1α protein comprises three functional domains: the dimerization domain mediates homodimerization or heterodimerization with homologous proteins such as HNF1β (19), which is fundamental for DNA-binding and transcriptional regulation, and aberrant assembly of the dimeric complex leads to impaired target gene binding and reduced transactivation activity. The DNA-binding domain consists of a POU-specific domain (POUS) and a POU homeodomain (POUH). POUS maintains protein stability, and POUH is critical for initiating protein-DNA interactions at promoter regions of target genes (20). The nuclear localization of HNF1α depends on specific nuclear localization signals, with residues 158-171, 197-205, and 271–282 are considered to be nuclear localization regions (21), mutations affecting these regions may lead to loss of transcriptional regulation. The transactivation domain recruits coactivators such as p300/CBP to activate the transcription of glucose metabolism-related genes (22). In our study, the number of cases in the dimerization domain group was relatively small, which may be attributed to the overall low frequency of mutations in this domain across populations. A retrospective study on HNF1A-MODY patients reported that (10) in non-Asian populations, mutation rates were approximately 7.8% in the dimerization domain, 29.2% in the DNA-binding domain, and 47.5% in the transactivation domain. Among Asian populations, the corresponding proportions were 3.8%, 48.8%, and 47.5%, with no significant difference in domain-specific mutation distribution between the two groups. Of the 432 patients with coding−region mutations for whom mutation sites were reported, only 15 occurred in the dimerization domain. In our Qilu Hospital cohort, mutations were observed in 8.7% of cases within the dimerization domain, 39.1% within the DNA-binding domain, and 52.2% within the transactivation domain, consistent with previous reports.

Previous researches demonstrated that the DNA-binding domain of HNF1α plays a more critical role in regulating blood glucose, and the dimerization domain plays the second most important role (17), indicating a possible genotype-phenotype correlation in HNF1A-MODY. To explore the association between genotype and clinical phenotypes in Chinese HNF1A-MODY patients, our study enrolled a total of 136 patients with from our cohort, together with previously reported cases from China identified through a review of published literature. Combined analysis with literature-sourced HNF1A-MODY cases revealed that patients with the dimerization and DNA-binding domain mutations presented a younger age of onset and lower FCP levels. Besides, patients with DNA-binding domain mutation exhibited significantly lower 2hPCP levels than those with transactivation domain mutation, suggesting that DNA-binding domain mutation cause more severe insulin secretion defects. This clinical difference indicates that mutations in the transactivation domain may lead to a milder diabetic phenotype. The impact of mutations varies by domain: mutations in the dimerization or DNA-binding domain impair DNA-binding capacity (23) and may cause more severe insulin secretion defects, resulting in a younger age of onset and lower C-peptide levels. The transactivation domain demonstrates greater tolerance to mutations that result in small changes in HNF1α protein structure than the dimerization or DNA-binding domain (24, 25). ANCOVA further confirmed that dimerization domain and DNA-binding domain mutation were independently associated with lower FCP levels. A comparison of the activity of the mutant and wild-type proteins revealed that mutations in the transactivation domain disrupted transcriptional activity less than mutations in the DNA-binding domain, as it primarily affected coactivator recruitment rather than DNA-binding affinity (26). This study is the first to reveal in a Chinese population that dimerization domain and DNA-binding domain mutations in HNF1A-MODY are associated with poorer pancreatic islet function. This finding provides a basis for assessing patient prognosis and formulating treatment strategies.

Regarding treatment, sulfonylureas are widely acknowledged as the first-line therapy for HNF1A-MODY (23); however, a notable proportion of patients in our study were treated with insulin. Real-world data and reviews indicate that insulin use is indeed common among genetically confirmed HNF1A-MODY patients. The real-world study (27) revealed that the insulin usage rate within one year after diabetes diagnosis was 52% (35% insulin monotherapy, 17% insulin combined with oral hypoglycemic agents). Even after genetic diagnosis of HNF1A-MODY, insulin usage remained at 42% (25% insulin alone, 17% insulin combined with oral agents). A review on the clinical characteristics of HNF1A-MODY patients (10) reported that insulin usage was 16.7% in those with dimerization domain mutations, 33.8% in DNA-binding domain, and 33.2% in transactivation domain. In our study, after excluding patients with variants of uncertain significance, the corresponding insulin usage rates for these domains were 33.3%, 35.3%, and 45.28%, respectively. The widespread use of insulin in HNF1A-MODY is likely attributable to diagnostic delay or treatment patterns, rather than true insulin dependence. Therefore, inferring domain-specific treatment responses based on current insulin use should be approached with caution. Instead, prospective therapeutic trials with sulfonylurea should be systematically conducted following genetic diagnosis.

In addition, our data showed that patients with DNA-binding domain mutations had lower LDL-C and TC levels, but higher HDL-C levels. HNF1α is involved in bile acid and plasma cholesterol metabolism (28). Prior studies have shown that liver-specific knockdown of HNF1α reduces LDL-C levels in normolipidemic mice (29), though similar effects have not been observed in humans. Our results suggest a potential domain-specific correlation with lipid regulation, but the precise mechanisms require further investigation.

This study also characterized the clinical profile and short-term treatment patterns associated with different glycemic control statuses. After a 2-year follow-up of the patients from Qilu Hospital, the majority achieved target glycemic levels; however, some developed new-onset diabetic peripheral neuropathy or diabetic retinopathy within this observation period. Notably, poor glycemic control was associated with higher baseline levels of BMI and blood lipids. These findings appear to support the association between chronic hyperglycemia and microvascular complications in this cohort, and may also underscore the potential role of weight and lipid management in short-term metabolic outcomes. Regarding pharmacotherapy, patients with good glycemic control were more likely to use sulfonylurea monotherapy, whereas those with poor control required insulin-based combination therapy. This observation aligns with the characteristic hypersensitivity to sulfonylureas in HNF1A-MODY patients, for whom these agents are considered first-line therapy (23). Although recent case reports identified novel HNF1A variants exhibiting resistance to sulfonylureas and other glucose-lowering drugs (9, 30), such variants were not identified in our cohort. Given that HNF1A-MODY patients exhibit progressive β-cell dysfunction and remain at high risk for both microvascular and macrovascular complications (31–33), early diagnosis coupled with stringent glycemic control is crucial for reducing the prevalence of late-diabetic complications (34, 35). However, given the relatively short follow-up period, longer-term studies will be needed to clarify prognosis and the durability of treatment effects.

This study has several limitations. First, the relatively small sample size, particularly the low number of cases in the dimerization domain group may affect statistical power and limit the generalizability of the findings. Second, the relatively short follow-up period is insufficient for comprehensively assessing long-term complication risks and does not provide strong support for long-term prognosis. To evaluate the impact on long-term outcomes, we are conducting ongoing follow-up of the patient cohort. Third, literature-derived cases exhibited variability in diagnostic criteria and laboratory methods, along with incomplete clinical variables, which may introduce selection and measurement biases and increase clinical data heterogeneity. Fourth, while we performed a sensitivity analysis excluding patients with early insulin dependence, the overall high prevalence of insulin therapy in our combined cohort may still reflect regional diagnostic delays or historical treatment patterns rather than intrinsic sulfonylurea unresponsiveness. This could introduce residual confounding when assessing metabolic phenotypes. Moreover, a subset of patients in the adolescent type 2 diabetes control group did not undergo genetic testing, which may to some extent affect the specificity of intergroup comparisons. Furthermore, the absence of functional studies on the identified variants restricts mechanistic interpretation of domain−specific mutation effects. Future multicenter studies with larger cohorts and longer follow-up durations, combined with functional characterization of mutant proteins, are needed to better elucidate the predictive value of different variant types for drug response and complication risks.

5 Conclusion

In conclusion, this study demonstrated domain-specific effects of HNF1A mutations. Pancreatic islet function was significantly associated with the location of the mutation. Mutations in the dimerization domain and DNA-binding domains were associated with an earlier age of onset and more severe insulin secretion defects. Regarding lipid metabolism, our analyses indicated an observational association between DNA-binding domain mutations and a distinct lipid profile characterized by lower LDL-C and higher HDL-C levels. This study summarized the association between clinical phenotypes and genotypes in patients with HNF1A-MODY, revealed the correlation between HNF1A mutation regions and pancreatic islet function as well as blood lipid levels, thereby providing a theoretical foundation for predicting pancreatic islet function prognosis through gene sequencing and guiding individualized treatment options.

Data availability statement

The original contributions presented in the study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding authors.

Ethics statement

The studies involving humans were approved by Ethics Committee of Qilu Hospital of Shandong University. The studies were conducted in accordance with the local legislation and institutional requirements. Written informed consent for participation in this study was provided by the participants’ legal guardians/next of kin.

Author contributions

MW: Formal Analysis, Methodology, Writing – original draft, Data curation, Investigation. HH: Investigation, Writing – original draft, Data curation, Formal Analysis. HL: Writing – original draft, Investigation, Formal Analysis, Data curation. HT: Resources, Writing – original draft, Data curation. XH: Writing – review & editing. LC: Funding acquisition, Writing – review & editing, Resources. MT: Writing – review & editing. LW: Writing – review & editing, Resources, Formal Analysis, Methodology.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This work was supported by the Major basic research project of Shandong Natural Science Foundation (Grant Number ZR2022ZD15) and the National Key Research and Development Program of China (Grant Number 2022YFA1004800).

Acknowledgments

We gratefully acknowledge Professor Han Wu from Qilu Hospital of Shandong University for his expert statistical support.

Conflict of interest

Author HT was employed by Jinan Aixin Zhuoer Medical Testing Co., Ltd.

The remaining authors declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fendo.2026.1735596/full#supplementary-material

References

1. Peixoto-Barbosa R, Reis AF, and Giuffrida FMA. Update on clinical screening of maturity-onset diabetes of the young (MODY). Diabetol Metab Syndr. (2020) 12:50. doi: 10.1186/s13098-020-00557-9

PubMed Abstract | Crossref Full Text | Google Scholar

2. Zhongyun Z, Juan Z, Danjie L, Xuyang C, Lei W, Rulai H, et al. Distribution characteristics of special types of diabetes mellitus in Chinese population: A literature-based analysis from 2011 to 2021. Chin J Endocrinol Metab. (2023) 39:336–44. doi: 10.3760/cma.j.cn311282-20220723-00444

Crossref Full Text | Google Scholar

3. Miyachi Y, Miyazawa T, and Ogawa Y. HNF1A mutations and beta cell dysfunction in diabetes. Int J Mol Sci. (2022) 23:3222. doi: 10.3390/ijms23063222

PubMed Abstract | Crossref Full Text | Google Scholar

4. Yong X, Cheng H, Tao Y, Lixin S, Ming L, Tianpei H, et al. Expert consensus on screening and treatment of maturity-onset diabetes of the young. Chin J Diabetes Mellitus. (2022) 14:423–32. doi: 10.3760/cma.j.cn115791-20220211-00073

Crossref Full Text | Google Scholar

5. Zhang H, Colclough K, Gloyn AL, and Pollin TI. Monogenic diabetes: a gateway to precision medicine in diabetes. J Clin Invest. (2021) 131:e142244. doi: 10.1172/jci142244

PubMed Abstract | Crossref Full Text | Google Scholar

6. Bonner C and Saponaro C. Where to for precision treatment of HNF1A-MODY? Diabetologia. (2022) 65:1825–9. doi: 10.1007/s00125-022-05696-4

PubMed Abstract | Crossref Full Text | Google Scholar

7. Bellanné-Chantelot C, Carette C, Riveline JP, Valéro R, Gautier JF, Larger E, et al. The type and the position of HNF1A mutation modulate age at diagnosis of diabetes in patients with maturity-onset diabetes of the young (MODY)-3. Diabetes. (2008) 57:503–8. doi: 10.2337/db07-0859

PubMed Abstract | Crossref Full Text | Google Scholar

8. Ludwig-Słomczyńska AH, Seweryn MT, Radkowski P, Kapusta P, Machlowska J, Pruhova S, et al. Variants influencing age at diagnosis of HNF1A-MODY. Mol Med. (2022) 28:113. doi: 10.1186/s10020-022-00542-0

PubMed Abstract | Crossref Full Text | Google Scholar

9. Tan CSH, Ang SF, and Lim SC. Response to multiple glucose-lowering agents in a sib-pair with a novel HNF1α (MODY3) variant. Eur J Hum Genet. (2020) 28:518–20. doi: 10.1038/s41431-019-0561-8

PubMed Abstract | Crossref Full Text | Google Scholar

10. Zhao Q, Ding L, Yang Y, Sun J, Wang M, Li X, et al. Clinical characteristics of patients with HNF1-alpha MODY: A literature review and retrospective chart review. Front Endocrinol (Lausanne). (2022) 13:900489. doi: 10.3389/fendo.2022.900489

PubMed Abstract | Crossref Full Text | Google Scholar

11. Kim SH, Ma X, Klupa T, Powers C, Pezzolesi M, Warram JH, et al. Genetic modifiers of the age at diagnosis of diabetes (MODY3) in carriers of hepatocyte nuclear factor-1alpha mutations map to chromosomes 5p15, 9q22, and 14q24. Diabetes. (2003) 52:2182–6. doi: 10.2337/diabetes.52.8.2182

PubMed Abstract | Crossref Full Text | Google Scholar

12. Stride A, Shepherd M, Frayling TM, Bulman MP, Ellard S, and Hattersley AT. Intrauterine hyperglycemia is associated with an earlier diagnosis of diabetes in HNF-1alpha gene mutation carriers. Diabetes Care. (2002) 25:2287–91. doi: 10.2337/diacare.25.12.2287

PubMed Abstract | Crossref Full Text | Google Scholar

13. American Diabetes Association Professional Practice Committee for Diabetes. 2. Diagnosis and classification of diabetes: standards of care in diabetes-2026. Diabetes Care. (2026) 49:S27–s49. doi: 10.2337/dc26-S002

PubMed Abstract | Crossref Full Text | Google Scholar

14. Yamagata K, Oda N, Kaisaki PJ, Menzel S, Furuta H, Vaxillaire M, et al. Mutations in the hepatocyte nuclear factor-1alpha gene in maturity-onset diabetes of the young (MODY3). Nature. (1996) 384:455–8. doi: 10.1038/384455a0

PubMed Abstract | Crossref Full Text | Google Scholar

15. Lau HH, Ng NHJ, Loo LSW, Jasmen JB, and Teo AKK. The molecular functions of hepatocyte nuclear factors - In and beyond the liver. J Hepatol. (2018) 68:1033–48. doi: 10.1016/j.jhep.2017.11.026

PubMed Abstract | Crossref Full Text | Google Scholar

16. Gaál Z, Szűcs Z, Kántor I, Luczay A, Tóth-Heyn P, Benn O, et al. A comprehensive analysis of hungarian MODY patients-part I: gene panel sequencing reveals pathogenic mutations in HNF1A, HNF1B, HNF4A, ABCC8 and INS genes. Life (Basel). (2021) 11:755. doi: 10.3390/life11080755

PubMed Abstract | Crossref Full Text | Google Scholar

17. Li LM, Jiang BG, and Sun LL. HNF1A:From monogenic diabetes to type 2 diabetes and gestational diabetes mellitus. Front Endocrinol (Lausanne). (2022) 13:829565. doi: 10.3389/fendo.2022.829565

PubMed Abstract | Crossref Full Text | Google Scholar

18. Donath X, Saint-Martin C, Dubois-Laforgue D, Rajasingham R, Mifsud F, Ciangura C, et al. Next-generation sequencing identifies monogenic diabetes in 16% of patients with late adolescence/adult-onset diabetes selected on a clinical basis: a cross-sectional analysis. BMC Med. (2019) 17:132. doi: 10.1186/s12916-019-1363-0

PubMed Abstract | Crossref Full Text | Google Scholar

19. Kind L, Raasakka A, Molnes J, Aukrust I, Bjørkhaug L, Njølstad PR, et al. Structural and biophysical characterization of transcription factor HNF-1A as a tool to study MODY3 diabetes variants. J Biol Chem. (2022) 298:101803. doi: 10.1016/j.jbc.2022.101803

PubMed Abstract | Crossref Full Text | Google Scholar

20. Sneha P, Thirumal Kumar D, George Priya Doss C, Siva R, and Zayed H. Determining the role of missense mutations in the POU domain of HNF1A that reduce the DNA-binding affinity: A computational approach. PloS One. (2017) 12:e0174953. doi: 10.1371/journal.pone.0174953

PubMed Abstract | Crossref Full Text | Google Scholar

21. Bjørkhaug L, Bratland A, Njølstad PR, and Molven A. Functional dissection of the HNF-1alpha transcription factor: a study on nuclear localization and transcriptional activation. DNA Cell Biol. (2005) 24:661–9. doi: 10.1089/dna.2005.24.661

PubMed Abstract | Crossref Full Text | Google Scholar

22. Kind L, Driver M, Raasakka A, Onck PR, Njølstad PR, Arnesen T, et al. Structural properties of the HNF-1A transactivation domain. Front Mol Biosci. (2023) 10:1249939. doi: 10.3389/fmolb.2023.1249939

PubMed Abstract | Crossref Full Text | Google Scholar

23. Valkovicova T, Skopkova M, Stanik J, and Gasperikova D. Novel insights into genetics and clinics of the HNF1A-MODY. Endocr Regul. (2019) 53:110–34. doi: 10.2478/enr-2019-0013

PubMed Abstract | Crossref Full Text | Google Scholar

24. Ellard S and Colclough K. Mutations in the genes encoding the transcription factors hepatocyte nuclear factor 1 alpha (HNF1A) and 4 alpha (HNF4A) in maturity-onset diabetes of the young. Hum Mutat. (2006) 27:854–69. doi: 10.1002/humu.20357

PubMed Abstract | Crossref Full Text | Google Scholar

25. Svalastoga P, Kaci A, Molnes J, Solheim MH, Johansson BB, Krogvold L, et al. Characterisation of HNF1A variants in paediatric diabetes in Norway using functional and clinical investigations to unmask phenotype and monogenic diabetes. Diabetologia. (2023) 66:2226–37. doi: 10.1007/s00125-023-06012-4

PubMed Abstract | Crossref Full Text | Google Scholar

26. Çubuk H and Yalçın Çapan Ö. A review of functional characterization of single amino acid change mutations in HNF transcription factors in MODY pathogenesis. Protein J. (2021) 40:348–60. doi: 10.1007/s10930-021-09991-8

PubMed Abstract | Crossref Full Text | Google Scholar

27. Lanzinger S, Laubner K, Warncke K, Mader JK, Kummer S, Boettcher C, et al. Clinical characteristics, treatment, and treatment switch after molecular-genetic classification in individuals with maturity-onset diabetes of the young: Insights from the multicenter real-world DPV registry. J Diabetes. (2024) 16:e70028. doi: 10.1111/1753-0407.70028

PubMed Abstract | Crossref Full Text | Google Scholar

28. Shih DQ, Bussen M, Sehayek E, Ananthanarayanan M, Shneider BL, Suchy FJ, et al. Hepatocyte nuclear factor-1alpha is an essential regulator of bile acid and plasma cholesterol metabolism. Nat Genet. (2001) 27:375–82. doi: 10.1038/86871

PubMed Abstract | Crossref Full Text | Google Scholar

29. Shende VR, Wu M, Singh AB, Dong B, Kan CF, and Liu J. Reduction of circulating PCSK9 and LDL-C levels by liver-specific knockdown of HNF1α in normolipidemic mice. J Lipid Res. (2015) 56:801–9. doi: 10.1194/jlr.M052969

PubMed Abstract | Crossref Full Text | Google Scholar

30. Demol S, Lebenthal Y, Bar-Meisels M, Phillip M, Gat-Yablonski G, and Gozlan Y. A family with a novel termination mutation in hepatic nuclear factor 1α in maturity-onset diabetes of the young type 3 which is unresponsive to sulphonylurea therapy. Horm Res Paediatr. (2014) 81:280–4. doi: 10.1159/000356925

PubMed Abstract | Crossref Full Text | Google Scholar

31. Sun HY and Lin XY. Genetic perspectives on childhood monogenic diabetes: Diagnosis, management, and future directions. World J Diabetes. (2023) 14:1738–53. doi: 10.4239/wjd.v14.i12.1738

PubMed Abstract | Crossref Full Text | Google Scholar

32. Cobry EC and Steck AK. Review of monogenic diabetes: clinical features and precision medicine in genetic forms of diabetes. Diabetes Technol Ther. (2025) 27:675–86. doi: 10.1089/dia.2024.0602

PubMed Abstract | Crossref Full Text | Google Scholar

33. Steele AM, Shields BM, Shepherd M, Ellard S, Hattersley AT, and Pearson ER. Increased all-cause and cardiovascular mortality in monogenic diabetes as a result of mutations in the HNF1A gene. Diabetes Med. (2010) 27:157–61. doi: 10.1111/j.1464-5491.2009.02913.x

PubMed Abstract | Crossref Full Text | Google Scholar

34. Ang SF, Tan CSH, Chan LWT, Goh LX, Kon WYC, Lian JX, et al. Clinical experience from a regional monogenic diabetes referral centre in Singapore. Diabetes Res Clin Pract. (2020) 168:108390. doi: 10.1016/j.diabres.2020.108390

PubMed Abstract | Crossref Full Text | Google Scholar

35. Sagen JV, Njølstad PR, and Søvik O. Reduced prevalence of late-diabetic complications in MODY3 with early diagnosis. Diabetes Med. (2002) 19:697–8. doi: 10.1046/j.1464-5491.2002.00688_1.x

PubMed Abstract | Crossref Full Text | Google Scholar

36. de Vries AG, Bakker-van Waarde WM, Dassel AC, Losekoot M, Duiker EW, Gouw AS, et al. A novel phenotype of a hepatocyte nuclear factor homeobox A (HNF1A) gene mutation, presenting with neonatal cholestasis. J Hepatol. (2015) 63:1295–7. doi: 10.1016/j.jhep.2015.08.005

PubMed Abstract | Crossref Full Text | Google Scholar

37. Liang H, Zhang Y, Li M, Yan J, Yang D, Luo S, et al. Recognition of maturity-onset diabetes of the young in China. J Diabetes Investig. (2021) 12:501–9. doi: 10.1111/jdi.13378

PubMed Abstract | Crossref Full Text | Google Scholar

38. Lopez AP, Foscaldi SA, Perez MS, Rodriguez M, Traversa M, Puchulu FM, et al. HNF1 alpha gene coding regions mutations screening, in a Caucasian population clinically characterized as MODY from Argentina. Diabetes Res Clin Pract. (2011) 91:208–12. doi: 10.1016/j.diabres.2010.11.024

PubMed Abstract | Crossref Full Text | Google Scholar

39. Kaisaki PJ, Menzel S, Lindner T, Oda N, Rjasanowski I, Sahm J, et al. Mutations in the hepatocyte nuclear factor-1alpha gene in MODY and early-onset NIDDM: evidence for a mutational hotspot in exon 4. Diabetes. (1997) 46:528–35. doi: 10.2337/diab.46.3.528

PubMed Abstract | Crossref Full Text | Google Scholar

40. Bjørkhaug L, Sagen JV, Thorsby P, Søvik O, Molven A, and Njølstad PR. Hepatocyte nuclear factor-1 alpha gene mutations and diabetes in Norway. J Clin Endocrinol Metab. (2003) 88:920–31. doi: 10.1210/jc.2002-020945

PubMed Abstract | Crossref Full Text | Google Scholar

41. Colclough K, Ellard S, Hattersley A, and Patel K. Syndromic monogenic diabetes genes should be tested in patients with a clinical suspicion of maturity-onset diabetes of the young. Diabetes. (2022) 71:530–7. doi: 10.2337/db21-0517

PubMed Abstract | Crossref Full Text | Google Scholar

42. Haring MPD, Vriesendorp TM, Klein Wassink-Ruiter JS, de Haas RJ, Gouw ASH, and de Meijer VE. Diagnosis of hepatocellular adenoma in men before onset of diabetes in HNF1A-MODY: Watch out for winkers. Liver Int. (2019) 39:2042–5. doi: 10.1111/liv.14235

PubMed Abstract | Crossref Full Text | Google Scholar

43. Yan Z, Meijun L, Min L, Qiao Z, and Lixin S. A case report of middle-aged diagnosed maturity onset diabetes of the young type 3 caused by mutation of hepatocyte nuclear factor 1 homeobox A. Chin J Diabetes Mellitus. (2022) 14:1465–8. doi: 10.3760/cma.j.cn115791-20220922-00486

Crossref Full Text | Google Scholar

44. Xu JY, Dan QH, Chan V, Wat NM, Tam S, Tiu SC, et al. Genetic and clinical characteristics of maturity-onset diabetes of the young in Chinese patients. Eur J Hum Genet. (2005) 13:422–7. doi: 10.1038/sj.ejhg.5201347

PubMed Abstract | Crossref Full Text | Google Scholar

45. Xu JY, Chan V, Zhang WY, Wat NM, and Lam KS. Mutations in the hepatocyte nuclear factor-1alpha gene in Chinese MODY families: prevalence and functional analysis. Diabetologia. (2002) 45:744–6. doi: 10.1007/s00125-002-0814-9

PubMed Abstract | Crossref Full Text | Google Scholar

46. Dron JS, Wang J, McIntyre AD, Iacocca MA, Robinson JF, Ban MR, et al. Six years’ experience with LipidSeq: clinical and research learnings from a hybrid, targeted sequencing panel for dyslipidemias. BMC Med Genomics. (2020) 13:23. doi: 10.1186/s12920-020-0669-2

PubMed Abstract | Crossref Full Text | Google Scholar