Summary: In children aged 2-6 years with type 1 diabetes requiring at least 6 units of insulin per day, automated insulin delivery with the MiniMed 780G system in auto mode was compared with the same pump in manual mode with suspend-before-low. In this open-label crossover trial, time in range was 68.3% versus 58.3% (adjusted difference 9.9 percentage points, 95% CI 8.0 to 11.7) and HbA1c was 7.00% versus 7.61% (difference -0.61 percentage points, 95% CI -0.76 to -0.46); auto mode met non-inferiority and was superior on both endpoints.
PICO Summary
| Element | Detail |
|---|---|
| Population | 98 children aged 2-6 years with type 1 diabetes requiring at least 6 units of insulin/day (mean age 4.7 years); open-label, multicentre, randomised crossover trial across 12 hospitals in Finland, Italy, Slovenia, and the UK (NCT05574062). |
| Intervention | MiniMed 780G system in auto mode (automated insulin delivery) for a 12-week period within the crossover (sequence AM-MM n=50; total n=98 analysed by intention to treat). |
| Comparison | The same MiniMed 780G system in manual mode with suspend-before-low enabled for a 12-week period (sequence MM-AM n=48), separated by a 2-week washout. Not standard care or a separate device. |
| Outcome | Time in range (70-180 mg/dL): 68.3% (auto) vs 58.3% (manual+SBL); adjusted difference 9.9 percentage points (95% CI 8.0 to 11.7), non-inferiority met (margin 7.5 pp) with superiority. HbA1c: 7.00% vs 7.61%; difference -0.61 percentage points (95% CI -0.76 to -0.46), non-inferiority met (margin 0.4 pp) with superiority. Safety: 9 serious adverse events (5 in auto mode, 2 in manual+SBL, 1 run-in, 1 washout), all deemed unrelated to the device, including 1 diabetic ketoacidosis event during auto mode; no severe hypoglycaemia reported. |
MiniMed 780G auto mode in young children (LENNY)
RCT crossover · type 1 diabetes · 12 weeks
In children aged 2-6 years, MiniMed 780G auto mode raised time in range by about 10 percentage points and lowered HbA1c by 0.6 points versus the same pump in manual mode with suspend-before-low, meeting non-inferiority with superiority and no severe hypoglycaemia.
Expert Commentary
This crossover trial provides reassuring evidence that the MiniMed 780G in auto mode is a sound option for very young children with type 1 diabetes. The prespecified non-inferiority endpoint was met, and the secondary superiority tests were also positive, with time in range improved by roughly 10 percentage points and HbA1c lowered by 0.6 percentage points relative to the same pump in manual mode with suspend-before-low. The crossover design, in which each child served as their own control, strengthens these within-participant comparisons, and the absence of any severe hypoglycaemia is notable in an age group where hypoglycaemia is feared most. Several caveats temper the enthusiasm. The trial was open-label, so families and clinicians knew which mode was active, and behaviour around glucose monitoring may have differed between arms. The study was funded by Medtronic, the device manufacturer, and several authors are company employees, which warrants the usual caution when effect sizes are large. One diabetic ketoacidosis event occurred during auto mode, a reminder that automation does not remove the need for vigilance. The comparator was the same advanced pump in manual mode rather than usual care, so the gain reflects automation specifically, not the move from injections. Can I use this with my patients? Yes, for children aged 2-6 years with type 1 diabetes who already use or can access this system, auto mode is a reasonable and well-supported choice. Clinicians should pair it with continued education on ketone monitoring and sick-day rules.
References
Battelino T, Kuusela S, Shetty A, Rabbone I, Cherubini V, Campbell F, et al. Efficacy and safety of automated insulin delivery in children aged 2-6 years (LENNY): an open-label, multicentre, randomised, crossover trial. Lancet Diabetes Endocrinol. 2025;13(8):662-673. doi:10.1016/S2213-8587(25)00091-9