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

Abstract

Objective: Identifying maturity-onset diabetes of the young (MODY) in pediatric populations close to diabetes diagnosis is difficult. Misdiagnosis and unnecessary insulin treatment are common. We aimed to identify the discriminatory clinical features at diabetes diagnosis of patients with glucokinase (GCK), hepatocyte nuclear factor-1A (HNF1A), and HNF4A MODY in the pediatric population.

Research design and methods: Swedish patients (n = 3,933) aged 1-18 years, diagnosed with diabetes May 2005 to December 2010, were recruited from the national consecutive prospective cohort Better Diabetes Diagnosis. Clinical data, islet autoantibodies (GAD insulinoma antigen-2, zinc transporter 8, and insulin autoantibodies), HLA type, and C-peptide were collected at diagnosis. MODY was identified by sequencing GCK, HNF1A, and HNF4A, through either routine clinical or research testing.

Results: The minimal prevalence of MODY was 1.2%. Discriminatory factors for MODY at diagnosis included four islet autoantibody negativity (100% vs. 11% not-known MODY; P = 2 × 10^-44), HbA_1c (7.0% vs. 10.7% [53 vs. 93 mmol/mol]; P = 1 × 10^-20), plasma glucose (11.7 vs. 26.7 mmol/L; P = 3 × 10^-19), parental diabetes (63% vs. 12%; P = 1 × 10^-15), and diabetic ketoacidosis (0% vs. 15%; P = 0.001). Testing 303 autoantibody-negative patients identified 46 patients with MODY (detection rate 15%). Limiting testing to the 73 islet autoantibody-negative patients with HbA_1c <7.5% (58 mmol/mol) at diagnosis identified 36 out of 46 (78%) patients with MODY (detection rate 49%). On follow-up, the 46 patients with MODY had excellent glycemic control, with an HbA_1c of 6.4% (47 mmol/mol), with 42 out of 46 (91%) patients not on insulin treatment.

Conclusions: At diagnosis of pediatric diabetes, absence of all islet autoantibodies and modest hyperglycemia (HbA_1c <7.5% [58 mmol/mol]) should result in testing for GCK, HNF1A, and HNF4A MODY. Testing all 12% patients negative for four islet autoantibodies is an effective strategy for not missing MODY but will result in a lower detection rate. Identifying MODY results in excellent long-term glycemic control without insulin.

Figure 1

Study flow diagram. MODY was only identified in islet autoantibody-negative patients. All subjects were tested for GADA, IA-2A, ZnT8A, and IAA islet autoantibodies. Antibody positive was defined as at least one of these four autoantibodies being positive. The autoantibody results were fed back to the clinicians soon after diagnosis. The genetic tests are shown in the shaded boxes. *Genetic tests initiated by the clinician for MODY using routine diagnostic services are shown as “clinical test.” **Genetic tests performed as part of the study to identify cases missed by clinical testing are defined as “research test.”

Figure 2

Patients with MODY had lower HbA_1c at diagnosis than those not known to have MODY. HbA_1c in patients with known MODY (both GCK and HNF1A/HNF4A) and those without known MODY (autoantibody negative [Ab -ve] and autoantibody positive [Ab +ve]). Data shown as a box and whisker plot: the top and bottom ends of the box are the upper and lower quartiles, respectively, and the median is marked as the horizontal line inside the box. The vertical lines indicate the maximum and minimum values excluding extreme outliers, shown as dots. HbA_1c of 7.5% (58 mmol/mol) cutoff for GCK MODY (30) shown as a dashed line.