One in five insulin-treated MODY referrals may have type 1 diabetes

Design

Retrospective diagnostic study of a clinical referral cohort

Population

1,129 insulin-treated individuals referred for MODY gene panel testing

Primary outcome

Estimated type 1 diabetes prevalence and triage performance of a genetic risk score

Effect

Estimated 20% (95% CI 14.9–25.2%) type 1 diabetes; 16.1% of unsolved cases flagged; AUC 0.83 in children <10

A Diabetes Care study examined 1,129 insulin-treated people referred for MODY gene panel testing, with a type 1 diabetes genetic risk score calculated from ten extra variants on the same panel. Comparing score distributions with reference populations suggested about 20% of referrals had type 1 diabetes.

Among those with no monogenic cause found, the enrichment was greater, about 26%, and reached 76% in children. Age-specific thresholds flagged 16% of genetically unsolved cases as probable type 1 diabetes. Used before testing, the score discriminated poorly overall (AUC 0.60) but well in children under 10 (AUC 0.83), where more than half of tests could have been avoided without missing a MODY diagnosis.

The score is a probability, not a diagnosis, and this was a single UK referral laboratory with a mostly European-ancestry population; type 1 genetic risk scores perform less well in South Asian populations. Autoantibodies and C-peptide remain the first tests. The value is in what happens after a negative gene panel: rather than leaving the label as 'possible MODY', consider whether this is type 1 diabetes.

• Check islet autoantibodies and C-peptide before requesting MODY genetic testing in an insulin-treated patient.
• Ask whether the laboratory can report a type 1 genetic risk score alongside the MODY panel.
• After a negative MODY panel, consider type 1 diabetes rather than leaving the diagnosis unresolved.
• Interpret type 1 genetic risk scores cautiously in South Asian patients, in whom they were not validated here.

Why it matters

An unsolved MODY referral often leaves a patient without a diagnosis, and this gives that label a likely answer.

Don't overread it

The score was derived and tested in a mostly European-ancestry UK referral population.

The statistics, in plain English

An AUC of 0.5 is a coin toss and 1.0 is perfect. At 0.60 overall the score barely separates groups, but at 0.83 in children under 10 it is useful for deciding who needs the gene panel.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.