Abstract
Context
The applicability of the maturity-onset diabetes of the young (MODY) risk calculator to non-White European populations remains unknown.
Objective
We aimed to test its real-world application in Hispanic youth.
Methods
We conducted a retrospective chart review of Hispanic youth (<23 years) with diabetes (n = 2033) in a large pediatric tertiary care center in the United States. We calculated MODY probability for all subjects, splitting them into 2 cohorts based on the original model: individuals who were started on insulin within 6 months of diabetes diagnosis (Cohort 1) and those who were not (Cohort 2).
Results
Cohort 1 consisted of 1566 individuals (median age [25p, 75p]: 16 [13, 19] years, 49% female), while Cohort 2 comprised 467 youth (median age [25p, 75p]: 17 [15, 20] years, 62% female). The mean MODY probability was 5.9% and 61.9% in Cohort 1 and Cohort 2, respectively. The mean probability for both cohorts combined was 18.8%, suggesting an expected 382 individuals with MODY, which is much higher than previous estimations (1-5%; ie, 20-102 individuals in this cohort). A total of 18 individuals tested positive for MODY among the limited number of individuals tested based on clinical suspicion and genetic testing availability (n = 44 out of 2033 tested, 2.2% of overall cohort).
Conclusion
The MODY risk calculator likely overestimates the probability of MODY in Hispanic youth, largely driven by an overestimation in those not early-insulin treated (predominantly young-onset type 2 diabetes). The calculator needs updating to improve its applicability in this population. In addition, further research is needed to help better identify MODY in Hispanic youth.
Keywords: MODY, maturity-onset diabetes of the young, MODY calculator, MODY probability
Diabetes is one of the most common chronic diseases worldwide. While some forms of diabetes are rare, such as maturity-onset diabetes of the young (MODY), their recognition is important because a precise diagnosis has substantial implications for clinical care including treatment, risk of complications, and family counseling. MODY accounts for 1% to 5% of all pediatric diabetes cases (1-5). However, it is often an underdiagnosed condition. For instance, only 6% of cases with a genetic variant associated with MODY had a prior MODY clinical diagnosis in 586 islet antibody–negative subjects who had a fasting C-peptide level 0.8 ng/mL or greater in the SEARCH study (4).
The challenges in MODY diagnosis stem from several factors, including overlapping clinical manifestations with other types of diabetes, limited availability of genetic testing on a broad scale, high costs of testing where available, and variability in the quality of testing procedures (6). Shields et al developed the MODY risk calculator (MRC) with a goal of helping clinicians in identifying patients who would benefit the most from undergoing MODY genetic testing (7). This tool calculates the risk of having MODY based on the estimated prevalence of the condition by using easily accessible demographic, clinical, and biochemical variables (age, sex, body mass index [BMI], HbA1c, parental history of diabetes, and treatment of diabetes). For the United Kingdom, post-test probabilities exceeding 25% for individuals who were not started on insulin within 6 months of diagnosis or surpassing 10% for those who were commenced insulin therapy earlier are considered appropriate risk levels to prompt the recommendation for MODY genetic testing (7). However, these thresholds may vary for each country based on significant variables, such as economic benefit, and the prevalence of MODY and other types of diabetes. In addition, development and validation of the MRC was conducted exclusively in White Europeans and the original data on which it was developed had no cases with pediatric type 2 diabetes (T2D). The lack of children and adolescents with T2D in the original data raises a question regarding the accuracy of MRC and may limit its applicability in populations with higher risk of young-onset T2D.
To our knowledge, this tool has not been specifically studied previously in Hispanic youth, where risk of young-onset T2D is much higher than in White Europeans. Therefore, our objective is to test the real-world application of this tool for the Hispanic youth treated at Texas Children's Hospital (Houston, TX, USA). We hypothesized that MRC may overestimate the risk of MODY in Hispanic youth, in whom pediatric T2D is prevalent.
Materials and Methods
Participants
We conducted a retrospective chart review to identify all Hispanic youth with diabetes who were treated at Texas Children's Hospital, the largest children's hospital in the United States, between January 2019 and March 2022, using electronic medical record data. Inclusion criteria were (1) diagnosis of diabetes mellitus, (2) age less than 23 years old, and (3) Hispanic ethnicity. Exclusion criteria were (1) missing key information essential for MRC, such as current age, sex, BMI, age of diagnosis, timing of insulin initiation, current treatment regimen, HbA1c, and parental history of diabetes. The flow diagram illustrating the study population is presented in Fig. 1.
Figure 1.
Flow diagram of the study population selection.
The study protocol was approved by the Institutional Review Board at Baylor College of Medicine. The requirement for informed consent was waived by the board due to the retrospective nature of the study. All procedures were conducted in accordance with relevant guidelines and regulations.
Procedures
We collected comprehensive demographic, clinical, biochemical, and MODY genetic test data for all eligible patients from electronic medical records. These data included age, sex, race/ethnicity (self-reported), age at diabetes diagnosis, type of diabetes, current BMI, BMI at diagnosis, occurrence of diabetic ketoacidosis within the past year, diabetes treatment regimen, presence of pancreatic islet antibodies, parental history of diabetes, non-high–density lipoprotein cholesterol levels at last check, MODY genetic testing results (if conducted), C-peptide, glucose, HbA1c, presence of acanthosis nigricans, and occurrence of diabetic ketoacidosis at diagnosis. Subsequently, we calculated a MODY risk score for each patient using the MRC developed by the University of Exeter group (7). The risk score is based on 2 equations: (a) For people insulin treated within 6 months of diagnosis (where the comparison is primarily type 1 diabetes [T1D] vs MODY) and (b) For people not insulin treated within 6 months of diabetes (where the comparison is primarily T2D vs MODY).
Statistical Analyses
Analyses were split into 2 cohorts based on whether patients were insulin treated within 6 months or not (Cohorts 1 and 2, respectively), to enable assessment of the performance of the 2 equations used by the MODY calculator. Demographic, clinical, and biochemical characteristics were summarized for each cohort (Cohort 1 and Cohort 2) and, within each cohort, by group (MODY genetic testing done vs not) using descriptive statistics. Also, among those who underwent MODY genetic testing, characteristics were also compared between individuals with a genetically confirmed MODY diagnosis vs not. For purpose of the group analysis, only those who had a pathogenic or likely pathogenic variant in a known MODY-related gene were considered as MODY positive. Continuous measures were expressed using mean with standard deviation, median with 25th and 75th percentile (25p, 75p) and frequencies with percentages as appropriate. Data were inspected for normality using the Shapiro–Wilk and D’Agostino-Pearson tests in GraphPad Prism 9.1.1. Wilcoxon rank sum test, t-test, chi-square test, and Fisher's exact test were selected based on appropriateness for analysis. All analyses were performed using GraphPad Prism (GraphPad Prism 9.1.1, GraphPad Software, San Diego, CA, USA). Statistical significances were noted if 2-sided P values were less than .05. MODY risk score data were presented with median with 25th and 75th percentile to demonstrate the distribution of probabilities. We assessed the performance of the MODY calculator by examining the potential calibration of the model (ie, whether the probabilities are appropriate or whether they are overestimated or underestimated in this population). Overall calibration was assessed by calculating the overall mean probability in each cohort, which would provide an estimate of the expected number of MODY cases (eg, if in 100 individuals, the mean probability was 10%, this would mean the expected number of individuals with MODY according to the calculator is 10). We could not directly compare this to the MODY prevalence in this cohort as testing was only carried out on a limited subgroup, but compared this estimate to the expected proportion of MODY based on previous population prevalence estimates (1-5% (1-5)).
Results
We studied 2033 Hispanic youth with diabetes, dividing them into Cohort 1 and Cohort 2 based on the timing of insulin initiation (within 6 months of diagnosis vs not, respectively).
Patient Characteristics in Each Cohort
Baseline characteristics of variables required for the MRC are summarized in Tables 1 and 2.
Table 1.
Demographic and clinical characteristics of Cohort 1 - Insulin use within <6 months of diagnosis (n = 1566) and their MODY risk scores.
| Comparison between patients tested for MODY vs not tested (n = 1566) | N (nonmissing) | Combined (n = 1566) | MODY genetic testing (n = 30) | No MODY genetic testing (n = 1536) | |
| Age, yearsb | 1566 | 16 (13,19) | 18 (15, 20) | 16 (13, 19) | |
| Age at diagnosis, yearsa | 1566 | 10.8 (7.4, 13.7) | 11.5 (10, 13.4) | 10.8 (7.4, 13.7) | |
| Female, n (%)a | 1566 | 773 (49.4%) | 16 (53.3%) | 757 (49.3%) | |
| Parents with diabetes, n (%)c | 1566 | 513 (32.8%) | 20 (66.7%) | 493 (32.1%) | |
| Current BMI, percentilea | 1566 | 91.6 (72.4, 97.55) | 84.9 (63.5, 97.4) | 91.7 (72.7, 97.6) | |
| Last HbA1c, %c | 1564 | 8.7 (7.4, 10.6) | 11.3 (9.4, 13.1) | 8.7 (7.4, 10.5) | |
| MODY risk score, %a | 1566 | 1.9 (0.7, 4.9) | 3.3 (1.62, 6.57) | 1.9 (0.7, 4.91) | |
| Comparison among patients tested for MODY (n = 30) | N (nonmissing) | Combined (n = 30) | MODY positive (n = 6) | MODY negative (n = 24) | |
| Age, yearsa | 30 | 18 (15, 20) | 18 (13.5, 22.2) | 18 (15.3-19) | |
| Age at diagnosis, yearsb | 30 | 11.5 (10, 13.4) | 10.7 (8.5,11.8) | 12.2 (10.2-13.6) | |
| Female, n (%)a | 30 | 16 (53.3%) | 2 (33.3 %) | 14 (58.3%) | |
| Parents with diabetes, n (%)a | 30 | 20 (66.6%) | 5 (83.3%) | 15 (62.5%) | |
| Current BMI, percentilea | 30 | 84.8 (63.5, 97.4) | 65.7 (45.8, 85.4) | 87.92 (65.5, 97.7) | |
| Last HbA1c, %a | 30 | 9.25 (7.7, 12.2) | 9.6 (8.6,10.9) | 8.9 (7.6,12.3) | |
| MODY risk score, % a | 30 | 3.3 (1.6, 6.6) | 4.45 (2.15, 9.30) | 2.6 (1.01, 5.99) |
Data presented as median (25p, 75p) or n (%).
P > .05.
P = .05-.01.
P < .01.
Table 2.
Demographic and clinical characteristics of Cohort 2 - No insulin within 6 months of diagnosis (n = 467) and their MODY risk scores.
| Comparison between patients tested for MODY vs not tested (n = 467) | N (nonmissing) | Combined (n = 467) | MODY genetic testing (n = 14) | No MODY genetic testing (n = 453) | |
| Age, yearsa | 467 | 17 (15, 20) | 19 (14.8,22.2) | 17 (15,20) | |
| Age at diagnosis, yearsc | 467 | 14 (11.7, 15.9) | 11.8 (8.3,14) | 14.1 (11.9,16.1) | |
| Female, n (%)a | 467 | 288 (61.7%) | 7 (50%) | 281(62%) | |
| Parents with diabetes, n (%)a | 467 | 271 (58%) | 11 (78.5%) | 260 (57.3%) | |
| Current BMI, percentilec | 466 | 98.6 (96.6, 99.4) | 84.5 (26,94.7) | 98.7 (96.8, 99.4) | |
| Last HbA1c, %a | 467 | 6.6 (5.8, 8.2) | 6.2 (5.9,6.7) | 6.6 (5.8,8.2) | |
| MODY risk score, %c | 467 | 75.49 (58, 75.5) | 75.49 (75.49, 75.49) | 75.49 (57.98, 75.49) | |
| Comparison among patients tested for MODY (n = 14) | N (nonmissing) | Combined (n = 14) | MODY positive (n = 12) | MODY negative (n = 2) | |
| Age, yearsa | 14 | 19 (14.8, 22.2) | 19 (14.3, 22.8) | 19.5 (17, 22) | |
| Age at diagnosis, yearsa | 14 | 11.8 (8.3, 13.9) | 12.8 (9.6, 14.1) | 8.2 (8, 8.4) | |
| Female, n (%)a | 14 | 7 (50%) | 5 (41.7%) | 2 (100%) | |
| Parents with diabetes, n (%)a | 14 | 11 (78.6%) | 9 (75%) | 2 (100%) | |
| Current BMI, percentileb | 13 | 84.5 (20.1, 29) | 83.8 (8.3, 92.8) | 97.7 (96.4, 99) | |
| Last HbA1c, % | 14 | 6.3 (5.9, 6.7) | 6.2 (5.8, 6.4) | 8.6 (7.3 ,9.9) | |
| MODY risk score, %a | 14 | 75.49 (75.49, 75.49) | 75.49 (75.49, 75.49) | 75.49 (75.49, 75.49) |
Data presented as median (25p, 75p) or n (%).
a P > .05.
bP = .05-.01.
cP < .01.
Cohort 1 (those who were insulin treated within 6 months of diagnosis) consisted of 1566 Hispanic individuals (median age [25p, 75p]: 16 [13, 19] years, 49% female). This cohort had a clinical diagnosis of T1D in 69%, T2D in 30%, clinical MODY diagnosis without molecular confirmation in 0.6%, and genetically confirmed MODY in 0.4% of participants. The median age (25p, 75p) at diagnosis was 10.8 (7.4, 3.7) years. At diagnosis, median HbA1c (25p, 75p) was 11.3% (9.9, 12.9), median BMI percentile (25p, 75p) 90.9 (44.2, 98.5), and random median C-peptide (25p, 75p) at diagnosis 0.78 (0.35, 2.1) ng/mL. Thirty-three percent of subjects had parental history of diabetes. In this cohort, only 1.9% (n = 30) of youth underwent MODY genetic testing, and MODY diagnosis was genetically confirmed in 6 individuals (20%, 6/30). Youth who underwent MODY genetic testing exhibited older age, higher C-peptide (median [25p, 75p] 1.7 [1.2, 2.9] vs 0.75 [0.3, 2.1] ng/mL), higher last HbA1c, and higher percentage of parental history of diabetes than those who did not undergo MODY genetic testing. Among individuals with available MODY genetic testing results, those with genetically confirmed MODY had younger age at diabetes onset than youth with negative MODY test results (Table 1).
Cohort 2 (those who were not insulin treated within 6 months of diagnosis) comprised 467 youth (median age [25p, 75p]: 17 [15, 20] years, 62% female). This cohort had a clinical diagnosis of T1D in 5%, T2D in 91%, clinical MODY diagnosis without molecular confirmation in 1.3%, and genetically confirmed MODY in 2.6% of patients. The median age (25p, 75p) at diagnosis was 14 (11.7, 15.9) years. This group of participants had overall characteristics consistent with T2D, with median BMI on the 99th percentile, 79% of participants having acanthosis nigricans, and random median C-peptide (25p, 75p) at diagnosis of 4.1 (2.7, 6.5) ng/mL. Fifty-eight percent had parental history of diabetes. In this cohort, only 3% (n = 14) of youth underwent MODY genetic testing, and MODY diagnosis was genetically confirmed in 12 individuals (85.7%, 12/14). Among these patients, the affected genes were identified as follows: GCK in 75% (n = 9), HNF4A in 8.3% (n = 1), and HNF1A in 16% (n = 2). When comparing clinical data between those with positive MODY testing results and those with either a negative test (n = 2) or no MODY test (n = 467), we found that the median C-peptide levels, BMI percentile at diagnosis, and age at diagnosis (25p, 75p) for the MODY group were 0.9 (0.6, 0.9) ng/mL, 74.4 (18.8, 84.9), and 12.8 (9.6, 14.1) years compared with 4.2 (2.7, 6.5) ng/mL, 99.1 (98.3, 99.5), and 14 (11.8, 16) years in the negative/nontested group. A family history of diabetes was reported in both cohorts, with 75% in the MODY group and 57% in the negative/nontested group. Acanthosis nigricans was observed in 66% of the negative/nontested group, whereas none of the MODY group exhibit acanthosis; however, there were missing data in both groups.
Performance of the MODY Calculator in Hispanic Youth
Among 1566 Hispanic youth in Cohort 1 who were treated with insulin within 6 months of diagnosis, the mean MODY probability was 5.9%, indicating an expected 92 individuals (5.9%) with MODY. For the 467 individuals in Cohort 2 who were treated with insulin later or not treated with insulin at all, the mean probability was 61.9%, suggesting an expected 289 individuals (61.9%) with MODY. The overall mean MODY probability for both cohorts was 18.8%, indicating an expected 382 individuals with MODY (18.8% of 2033 individuals). This prevalence estimate surpasses previous estimates (4, 8-10) of MODY prevalence (1-5%), largely because of very high MODY probabilities in Cohort 2 (Fig. 2), suggesting the model is substantially overestimating probabilities in this population.
Figure 2.
(A) Estimated percentage of MODY diagnosis based on the MODY risk calculator (MRC) vs previous prevalence estimates of 3% (1-5%). (B) Estimated absolute number of MODY cases (n) based on the MRC vs previous prevalence estimates of 3% (1-5%).
While the probability of MODY was statistically lower in those not tested for MODY in Cohort 2 (P = .003) this group had overall extremely high MODY probability, with 60% having the maximum probability for the model (75.49%). All 14 individuals tested for MODY had the maximum model probability (Table 2 and Fig. 3).
Figure 3.
Dot plot comparing MODY risk scores between different subgroups. Data shown as median with 25th and 75th percentiles. (A) (n = 30) Comparison between MODY positive vs MODY negative in those who were started on insulin within 6 months of diagnosis (ie, Cohort 1) and had MODY genetic testing done. (B) (n = 1566) Comparison between those who had MODY genetic testing (MGT) vs those who did not in Cohort 1. (C) (n = 14) Comparison between MODY positive vs MODY negative in those who were not started on insulin within 6 months of diagnosis (ie, Cohort 2) and had MODY genetic testing done. (D) (n = 467) Comparison between those who had MODY genetic testing (MGT) vs those who did not in Cohort 2.
Discussion
We found that MRC likely overestimates MODY probability in Hispanic youth with diabetes, particularly in those who have a T2D phenotype (Cohort 2). Hispanic youth with diabetes had a high probability of MODY (18.8%), far in excess of previously reported MODY prevalence estimates in pediatric diabetes populations (1-5%) (4, 8-10), and this was largely driven by the substantial probability overestimation (61.9%) in the nonearly insulin–treated cohort (Cohort 2), which predominantly consisted of youth with T2D. This study demonstrates that mean probabilities for MODY in Hispanic youth far exceeds the 5% to 20% thresholds recommended by the American Diabetes Association Guidelines and the National Health Service in the UK (7, 11, 12), raising concerns about potential overtesting, particularly among nonearly insulin–treated patients with diabetes.
The higher probabilities seen in Cohort 2 are likely due to the higher prevalence of T2D in Hispanic youth in the United States compared with the UK population where the MRC was validated, and the lack of individuals with T2D below 18 years of age in the UK dataset used for the development of MRC (7). The estimated prevalence of T2D in youth in the UK was 0.21/100 000 in the early 2000s (13). Most recent data also confirm that youth-onset T2D is a relatively rare condition in the UK (14). In contrast, youth-onset of T2D in the United States is not infrequent; in fact, Cohort 2 had 91% of cases with a clinical diagnosis of T2D. Not surprisingly, recent data suggest a T2D prevalence of 0.67 per 1000 youth among those aged 10 to 19 years in the United States, with ethnic minorities carrying the largest burden. Specifically, in Hispanic youth, the prevalence of T2D is 1.03 for every 1000 youth (15). In addition, key variables discriminating MODY from non-MODY in the UK data (7) were age at diagnosis (heavily skewed to the upper age limit of 35 for the non-MODY patients) and BMI, whereas the differences in these variables in the Hispanic youth were less pronounced.
Overestimation of the MODY probability in Hispanic youth warrants further studies to validate its effective use in this population. This validation may require revising the relative contribution of each variable to the risk score (eg, BMI), incorporating the prevalence of youth-onset T2D for different geographical location into the MRC, adding newer variables to the model, and/or providing cautious cutoff values that prompt MODY genetic testing. A revised MRC is particularly of importance in Hispanic youth who did not start insulin within 6 months of diagnosis and/or have never been treated with insulin. The absence of younger individuals with T2D in the model could lead to higher MODY risk score calculations in those younger than 18 years of age who were not treated with insulin within 6 months of diagnosis. This may cause the model to consider them likely to have MODY in an algorithm comparing MODY vs T2D. These concerns were raised already by Misra et al in the South Asian population in the UK, where they had a lower pick-up rate upon genetic testing than Western Europeans (12% vs 29%), likely due to higher prevalence of youth-onset T2D in this population (16). Regarding BMI in the model for Cohort 2, a concern arises as the risk score becomes significantly lower (<25%) only when BMI reaches >45 kg/m2, which could be an acceptable threshold for adults but not children or youth. The complexity of BMI for Hispanic youth lies not only in the high prevalence of obesity, but also in the fact that Hispanics appear to have a higher likelihood of developing T2D at lower BMI levels (17). Obesity prevalence in Mexican American boys and girls has reached 31% and 25%, respectively, in youth younger than 19 years old (18). The prevalence of MODY in Hispanic youth is unknown; however, the SEARCH study identified that out of all cases identified with MODY, 31% were from Hispanic origin (4). Lastly, the MRC was validated to be used in people younger than 35 years old (7), and our study group only includes patients younger than 23 years old, which could have played a role in the overestimation of the MRC in this particular cohort.
The MRC has been studied in other populations across the world in recent years with some limitations. The performance of the MRC was evaluated and validated in a subset of Chinese individuals who were between 15 and 35 years old who were diagnosed clinically with T2D (19). In this population, the optimal cutoff for MODY probability was 40.7%. This cutoff value is higher than the 25% that was used in the UK population for those with a T2D phenotype (7). In Brazil, the optimal cutoff for MODY probability was reported as 60% for individuals diagnosed with diabetes before 35 years of age (20). However, the study population was heavily enriched with patients with genetically confirmed MODY accounting for almost 30% of the cohort, so significantly different than the overall diabetes population. In a retrospective study conducted in 72 selected Turkish patients with a clinically diagnosed T2D who had elevated C-peptide (mean 2.35 ng/mL), the mean MODY probability was 11.2%, but no genetic data were reported for the cohort (21). In other studies examining the MODY calculator in their geographical region, assessment of its discriminatory performance was limited to highly selected groups of individuals referred for MODY testing on the basis of their clinical phenotype (22-26). Hence, these previous investigations do not demonstrate its discriminatory power and calibration on an unselected population with diabetes reflective of routine clinical practice. Broadening these various efforts to evaluate the performance of the MRC in different ethnicities and populations is needed, yet it is essential to ensure that the evaluation is conducted at the diabetes population level rather than the laboratory level so that recommendations can be given to clinicians to guide MODY genetic testing decisions.
Our study further illustrates the significant gap in performing MODY genetic testing in Hispanic youth as only 2.2% of the population had MODY genetic testing completed. This undertesting is particularly noticeable in Cohort 2, in which the MODY positivity rate was over 85% in those who underwent a MODY genetic test. Multiple factors might have led to this undertesting, including but not limited to the level of clinical suspicion of individual providers, affordability, and insurance coverage. The barriers to MODY genetic testing and potential solutions are beyond the scope of this manuscript and warrant further investigation.
The limitations of the study include lack of systematic MODY genetic testing for the entire cohort and relatively lower number of individuals with confirmed MODY diagnosis, which meant that formal testing of discrimination and more detailed assessment of calibration could not be carried out. However, even in the absence of widespread genetic testing, our study offers valuable insights into the real-world applicability of MRC in Hispanic youth with diabetes by comparing their MODY probabilities to known MODY prevalence estimates. Additionally, our findings underscore the need for a revised MRC for Hispanic youth with diabetes. While our study identified remarkably high MODY probabilities in Hispanic youth with T2D phenotype, it is important to note that these findings may not be directly applicable to non-Hispanic youth with T2D. However, one could argue that the lack of younger individuals with T2D in the original cohort may similarly affect the accuracy of MODY probabilities in youth with T2D from other populations.
In conclusion, the MRC likely overestimates the MODY probabilities in Hispanic youth with diabetes, especially among those with T2D phenotype. These findings suggest exercising caution when utilizing the MRC to inform decisions regarding genetic testing in this population, particularly in those with features of T2D. Moreover, our findings highlight the need for further research with systematic genetic testing to evaluate the performance of MODY risk assessment tools in diverse populations and to potentially refine these tools to improve their accuracy across different demographic groups. Such efforts are crucial for ensuring optimal management and treatment strategies for individuals at risk of MODY.
Abbreviations
- BMI
body mass index
- MODY
maturity-onset diabetes of the young
- MRC
MODY risk calculator
- T1D
type 1 diabetes
- T2D
type 2 diabetes
Contributor Information
Guido Alarcon, Department of Pediatrics, The Division of Diabetes and Endocrinology, Baylor College of Medicine, Texas Children’s Hospital, Houston, TX 77030, USA.
Anh Nguyen, Department of Pediatrics, The Division of Diabetes and Endocrinology, Baylor College of Medicine, Texas Children’s Hospital, Houston, TX 77030, USA.
Angus Jones, Exeter Centre of Excellence in Diabetes (EXCEED), University of Exeter Medical School, Exeter EX2 5DW, UK.
Beverley Shields, Exeter Centre of Excellence in Diabetes (EXCEED), University of Exeter Medical School, Exeter EX2 5DW, UK.
Maria J Redondo, Department of Pediatrics, The Division of Diabetes and Endocrinology, Baylor College of Medicine, Texas Children’s Hospital, Houston, TX 77030, USA.
Mustafa Tosur, Department of Pediatrics, The Division of Diabetes and Endocrinology, Baylor College of Medicine, Texas Children’s Hospital, Houston, TX 77030, USA; Children's Nutrition Research Center, USDA/ARS, Houston, TX 77030, USA.
Funding
The work in this manuscript is supported by K23-DK129821 (M.T.), R01-DK124395 (M.J.R., M.T.) from the National Institute of Health-National Institute of Diabetes and Digestive and Kidney disease (NIH-NIDDK) and by Diabetes UK 21/0006328 (B.S.).
Author Contributions
G.A. and M.T. designed the study. A.N. collected data from the electronic medical record. G.A. wrote the initial draft and edited the manuscript. M.T. reviewed and critically revised the manuscript. M.J.R., A.N., B.S., and A.J. reviewed and edited the manuscript. G.A. and M.T. are the guarantors of this work and had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. All authors approved the final version of the manuscript.
Disclosures
The authors have no relevant conflict of interest to disclose.
Data Availability
Some or all datasets generated during and/or analyzed during the current study are not publicly available but are available from the corresponding author on reasonable request.
Prior Presentation
Parts of the content of this manuscript was presented at the American Diabetes Association 83rd Scientific Meeting in San Diego, CA, in June 2023.
References
- 1. Ellard S, Weedon M, Colclough K, et al. Absence of islet autoantibodies and modestly raised glucose values at diabetes diagnosis should lead to testing for MODY: lessons from a 5-year pediatric Swedish national cohort study. Diabetes Care 2020;43(1):82‐89. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Delvecchio M, Mozzillo E, Salzano G, et al. Monogenic diabetes accounts for 6.3% of cases referred to 15 Italian pediatric diabetes centers during 2007 to 2012. J Clin Endocrinol Metab. 2017;102(6):1826‐1834. [DOI] [PubMed] [Google Scholar]
- 3. Shields BM, Hicks S, Shepherd MH, Colclough K, Hattersley AT, Ellard S. Maturity-onset diabetes of the young (MODY): how many cases are we missing? Diabetologia. 2010;53(12):2504‐2508. [DOI] [PubMed] [Google Scholar]
- 4. Pihoker C, Gilliam LK, Ellard S, et al. Prevalence, characteristics and clinical diagnosis of maturity onset diabetes of the young due to from the SEARCH for diabetes in youth. J Clin Endocrinol Metab. 2013;98(10):4055‐4062. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Tosur M, Philipson LH. Precision diabetes: lessons learned from maturity-onset diabetes of the young (MODY). J Diabetes Investig. 2022;13(9):1465‐1471. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Riddle MC, Philipson LH, Rich SS, et al. Monogenic diabetes: from genetic insights to population-based precision in care. Reflections from a diabetes care editors’ expert forum. Diabetes Care. 2020;43(12):3117‐3128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Shields BM, Mcdonald TJ, Ellard S, Campbell MJ, Hyde C, Hattersley AT. The development and validation of a clinical prediction model to determine the probability of MODY in patients with young-onset diabetes. Diabetologia 2012;55(5):1265‐1272. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Johnson SR, Ellis JJ, Leo P, et al. Comprehensive genetic screening: the prevalence of maturity-onset diabetes of the young gene variants in a population-based childhood diabetes cohort. Pediatr Diabetes. 2019;20(1):57‐64. [DOI] [PubMed] [Google Scholar]
- 9. Fendler W, Borowiec M. Prevalence of monogenic diabetes amongst Polish children after a nationwide genetic screening campaign. Diabetologia. 2012;55(10):2631‐2635. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Shepherd M, Shields B, Hammersley S, et al. Europe PMC funders group systematic population screening, using biomarkers and genetic testing, identifies 2.5% of the UK pediatric diabetes population with monogenic diabetes. Diabetes Care 2016;39(11):1879‐1888. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. NHS . National Genomic Test Directory. Testing criteria for rare and inherited disease. https://www.england.nhs.uk/wp-content/uploads/2018/08/rare-and-inherited-disease-eligibility-criteria-v2.pdf. Published 2021.
- 12. American Diabetes Association Professional Practice Committee . 2. Diagnosis and classification of diabetes: standards of care in diabetes-2024. Diabetes Care. 2024;47(Supplement_1):S20‐S42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Ehtisham S, Hattersley AT, Dunger DB, Barrett TG; British Society for Paediatric Endocrinology and Diabetes Clinical Trials Group . First UK survey of paediatric type 2 diabetes and MODY. Arch Dis Child. 2004;89(6):526‐529. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Candler TP, Mahmoud O, Lynn RM, Majbar AA, Barrett TG, Shield JPH. Continuing rise of type 2 diabetes incidence in children and young people in the UK. Diabet Med 2018;35(6):737‐744. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Lawrence JM, Divers J, Isom S, et al. Trends in prevalence of type 1 and type 2 diabetes in children and adolescents in the US, 2001-2017. JAMA. 2021;326(8):717‐727. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Misra S, Shields B, Colclough K, et al. South Asian individuals with diabetes who are referred for MODY testing in the UK have a lower mutation pick-up rate than white European people. Diabetologia. 2016;59(10):2262‐2265. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Zhu Y, Sidell MA, Arterburn D, et al. Racial/ethnic disparities in the prevalence of diabetes and prediabetes by BMI: patient outcomes research to advance learning (PORTAL) multisite cohort of adults in the U. S. Diabetes Care. 2019;42(12):2211‐2219. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Wang Y, Beydoun MA, Min J, Xue H, Kaminsky LA, Cheskin LJ. Has the prevalence of overweight, obesity and central obesity levelled off in the United States? Trends, patterns, disparities, and future projections for the obesity epidemic. Int J Epidemiol. 2020;49(3):810‐823. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Zhao J, Chen Y, Ma F, et al. MODY probability calculator is suitable for MODY screening in China: a population-based study. J Endocr Soc. 2024;8(5):bvae047. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Santomauro AC Jr, Magalhães ÁLF, Motta FT, et al. The performance of the MODY calculator in a non-Caucasian, mixed-race population diagnosed with diabetes mellitus before 35 years of age. Diabetol Metab Syndr. 2023;15(1):15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Keskinler MV, Erbakan AN, Oguz A. MODY probability ratios in patients diagnosed with type 2 diabetes Mellitus at a young age. Medeni Med J. 2020;35(4):290‐294. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Harrington F, Greenslade M, Colclough K, Paul R, Jefferies C, Murphy R. Monogenic diabetes in New Zealand—an audit based revision of the monogenic diabetes genetic testing pathway in New Zealand. Front Endocrinol (Lausanne). 2023;14:1116880. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. da Silva Santos T, Fonseca L, Santos Monteiro S, et al. MODY probability calculator utility in individuals’ selection for genetic testing: its accuracy and performance. Endocrinol Diabetes Metab. 2022;5(5):e00332. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. De Sousa SMC, Wu KHC, Colclough K, et al. Identification of monogenic diabetes in an Australian cohort using the Exeter maturity-onset diabetes of the young (MODY) probability calculator and next-generation sequencing gene panel testing. Acta Diabetol. 2024;61(2):181‐188. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Tarantino RM, de Medeiros Abreu G, de Fonseca ACP, et al. MODY probability calculator for GCK and HNF1A screening in a multiethnic background population. Arch Endocrinol Metab. 2020;64(1):17‐23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Hohendorff J, Zapala B, Ludwig-Slomczynska AH, et al. The utility of MODY probability calculator in probands of families with early-onset autosomal dominant diabetes from Poland. Minerva Med. 2019;110(6):499‐506. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Some or all datasets generated during and/or analyzed during the current study are not publicly available but are available from the corresponding author on reasonable request.
