Table 1

Approaches used to identify monogenic diabetes in pediatric populations

Type of studyCountryAreaInitial cohort (n)Cohort characteristicsTesting strategy (subgroup tested)Genes testedPrevalence in genetically testedMinimal prevalence of monogenic diabetesReference
Systematic studies ordered by number in study
 Multicenter population basedU.S.6 centers: California, Ohio, Hawaii, South Carolina, Washington5,9631) Diagnosed <20 years1) AB-ve (×2), fasting C-peptide ≥0.8 ng/mL (n = 586)1) HNF1A, HNF4A, GCK1) 8.4% (47/586)1.2%1
2) Diagnosed <6 months2) Diagnosed <6 months (n = 7)2) KCNJ11, INS, ABCC82) 71.4% (5/7)0.2% (total 1.4%)19
 Nationwide population basedNorwayNationwide2,756Newly diagnosed, age 0–14 years1) AB-ve (×2) and affected parent (n = 46)1) HNF1A, HNF4A, MIDD1) 13.0% (6/46)1.1%2
2) AB-ve, HbA1c <7.5% (58 mmol/mol) and not receiving insulin (n = 10)2) GCK2) 30.0% (3/10)
3) Diagnosed <12 months (n = 24)3) KCNJ11, ABCC8, INS3) 16.6% (4/24)
 Epidemiological data/ nationwide genetic test resultsPoland3 centers: Lodz, Katowice, Gdansk2,568Age 0–18 years1) AB-ve, affected parent, noninsulin dependent1) HNF1A, HNF4A, HNF1B32.1% (100/311)3.1–4.2%7
2) HbA1c <7.5% (58 mmol/mol)2) GCK
3) Diagnosed <6 months3) KCNJ11, ABCC8, INS
4) Syndromic diabetes4) WFS, Alstrom
 Single pediatric clinic populationU.S.New York939Clinical diagnosis T1DAB-ve (×3) plus eitherGCK,8.6% (5/58)0.5%*4
Age 6 months to 20 yearsHbA1c ≤7% (53 mmol/mol) and ≤0.5 units insulin/kg/day >1 year postdiagnosis C-peptide positive or 3 gen. FH (n = 58)HNF1A
 Pediatric clinics in single cityAustraliaSydney4971) Clinical diagnosis T1DAB-ve (×4 on 2 occasions) (n = 19)1) HNF1A, HNF4A 2) INS, KCNJ115% (1/19)1.2%*20
2) Diagnosed 6 months to 16 years
 Single pediatric clinic populationSpainMadrid2521) Clinical diagnosis T1DAB-ve (×5) (n = 25)1) HNF1A, HNF4A8.0% (2/25)0.8%*8
2) Diagnosed 6 months to 17 years2) KCNJ11, INS
 Pediatric clinic: case historiesNew ZealandSouth Island160Pediatric diabetes <18 yearsAB-ve (×2?) (n = 4)GCK, HNF1B, HNF1A2.5% (4/160)2.5%21
 NationwideJapanCenters throughout JapanN/KAge 6 months to 20 years1) AB-ve (×2), BMI <25 kg/mm2, dominant family history or1) HNF1A, GCK, HNF4A, MIDD47.5% (38/80)34
2) Renal cysts (n = 80)2) HNF1B
 Single pediatric clinic populationU.S.ColoradoN/KDiabetes <25 yearsC-peptide ≥0.1 ng/mL, AB-ve (×3) (n = 97)HNF1A, HNF4A, GCK, PDX1, HNF1B22.7% (22/97)N/K35
Type of studyCountryAreaInitial cohort of patients with diabetes and the population taken from (n)Cohort characteristicHow monogenic diabetes was definedMonogenic diabetes diagnosis, n (% all diabetes)Prevalence per 100,000 populationReference
Nonsystematic studies relying on clinical recognition and clinical testing
 Postal questionnaire surveyU.K.Nationwide15,255 (59 million population)Diabetes <16 years non-T1DConfirmed by genetic test20 (0.13%)0.1723
 Questionnaire and telephone surveyGermanyState of Baden-Württemberg2,640 (2.6 million) population0–20 yearsClinician diagnosis (45% genetically confirmed)58 (2.1%)2.324
 Assessment of Childhood Diabetes RegistryGermanySaxony (34 pediatric clinics)865 new cases Prevalence cases not stated (4.8 million population)Newly diagnosed, age 0–15 yearsConfirmed by genetic test21 (2.4%) prevalence in incident casesCannot be calculated26
 Surveillance questionnaire (physician reporting)CanadaNationalNot stated (35 million population Canada)Newly diagnosed, non-T1D <18 yearsClinical diagnosis genetically confirmed in ∼50%31 (% cannot be calculated)0.3225#
 Observational investigation of databaseAustria/Germany262 pediatric clinics40,567 populationAge <20 years, diagnosis for <18 yearsClinician diagnosis MODY usually confirmed by genetic test (polymorphisms not excluded#)339 all cases (0.8%), 263 (0.65%) genetic positive#Cannot be calculated27
  • AB-ve, autoantibody negative; 3 gen. FH, three-generation family history; N/K, not known; T1D, type 1 diabetes.

  • *Only patients with a clinical diagnosis of type 1 diabetes were included, so the prevalence is likely to be underestimated.

  • #Subsequent study (30) indicated 38% of reported HNF1A patients were polymorphisms not mutations.