What "in range" feels like, and who automated insulin delivery actually helps

The edition in brief

Three findings shape this edition. In 161 adults with type 1 diabetes wearing blinded continuous glucose monitoring for 14 days, time in a personal range (mean glucose plus or minus 50 mg/dL) tracked pain, fatigue, activity demands and perceptual speed more consistently than guideline time in range did, while guideline range tracked diabetes distress and self-care better. The two metrics measure different things, and a patient who feels worse at target may be describing habituation rather than resisting advice. Second, among 833 autoantibody-positive people with overweight or obesity in the TrialNet Pathway to Prevention study, BMI normalised in 26.8% and that was associated with about a 48% lower risk of progressing to stage 3 type 1 diabetes over a median 4.1 years. The association sat in youth and was not seen in adults, and the study was observational. A pooled glycomics analysis links agalactosylated IgG glycan patterns to incident diabetic kidney disease but not, after correction, to neuropathy. On the regulatory side, a Class II FDA recall of compounded semaglutide multi-dose vials from one US firm is ongoing for particulate contamination: a manufacturing-quality action, not a signal about the drug. The practice-changer: in 713 youth followed 18 months after starting automated insulin delivery, every trajectory group improved fast for three months and then plateaued or slipped, only 12.7% reached time in range of 70% or more, and the worst trajectory was predicted by baseline behaviours the algorithm did not override.