A phase 2 liver trial where the add-on drug didn't help — and the base drug quietly did
Read the Sunday Brief →

Study wrapper · #136

Personalized Combination of a Ketogenic Diet and Low-Dose Semaglutide for Cardiometabolic Health: A Retrospective Case Series.

Parker G, Morris MD, Heggie JR, et al. Journal of personalized medicine. 2026.
Weak / noneCase reportMentions: Semaglutide

Editor's note

A small, uncontrolled retrospective case series (seven adults, six women) describing a clinician-supervised 6-month program pairing a whole-food ketogenic diet with low-dose semaglutide (<=1.0 mg/week). Researchers reported mean total weight loss of 21.9 kg, of which about 92% was attributed to fat mass by bioimpedance, with skeletal muscle largely preserved, plus falls in fasting insulin and visceral fat and broadly favorable lipid and inflammation markers. The interest here is the lean-mass-preservation question surrounding GLP-1 weight loss. But the design is weak: no control group, only seven patients, bioimpedance (BIA) rather than DXA/imaging for body composition, and an individualized protocol that confounds diet with drug. The authors themselves call these findings hypothesis-generating. Weight it accordingly, as a signal to test in controlled trials, not as evidence that the combination outperforms either component.

Plain-language abstract

This is a small look-back at seven adults (six women, one man) with overweight or obesity who took part in a supervised 6-month program combining a whole-food ketogenic (very low-carb) diet with a low weekly dose of semaglutide (up to 1.0 mg). Doctors measured body composition and blood markers at 1, 3 and 6 months. On average, patients lost 21.9 kg, and about 92% of that loss was body fat as estimated by a bioimpedance scale; muscle mass was mostly kept (average loss just 1.2 kg), and one patient gained muscle. Fasting insulin dropped, belly (visceral) fat fell by about 37%, and most patients showed better triglycerides, HDL cholesterol and inflammation markers. Because there was no comparison group, only seven people, and both the diet and drug dose were tailored to each person, the authors describe these results as preliminary ideas that need testing in proper controlled studies.