NEW ORLEANS – Early weight loss on liraglutide – specifically, dropping at least 4% of body weight at 16 weeks – is a strong and clinically useful identifier of patients with a high likelihood of significant weight loss 13 months into treatment, with an accompanying improvement in cardiometabolic risk factors, according to Ken Fujioka, MD.
Conversely, patients who aren’t early responders to subcutaneous liraglutide at 3 mg/day are unlikely to achieve at least 5% weight loss after a full year on the drug, the regulatory benchmark for clinically meaningful weight loss, added Dr. Fujioka, an internist and director of the center for weight management at the Scripps Clinic in La Jolla, Calif.
Dr. Ken Fujioka
“Using this early response criterion at week 16 to predict long-term weight loss is to me a very valuable tool. Obesity is an odd disease because it has so many different causes. Finding the right drug is tough, and how long to keep trying with a particular medication is something we haven’t known. So I think the biggest change in obesity medicine is the creation of stopping rules that allow you to say, ‘OK, maybe this isn’t going to work. There’s some other reason you’re gaining weight, so let’s move on to something else,’” Dr. Fujioka said at the meeting presented by the Obesity Society of America and the American Society for Metabolic and Bariatric Surgery.
“When you stop the medication, you improve the risk-benefit ratio by removing all risk. That’s a win-win to me, and I applaud the FDA for getting on the pharmaceutical companies to make sure they put stopping rules in their medication labels,” he added.
He presented a post hoc pooled analysis of two previously published large, double-blind phase III clinical trials of subcutaneous liraglutide at 3 mg/day (Saxenda) or placebo in combination with a diet and exercise intervention for weight loss: the 3,731-patient SCALE Obesity and Prediabetes trial and the 846-patient SCALE Diabetes trial. In both trials liraglutide was started at a dose of 0.6 mg and titrated to 3.0 mg by week 4. The lifestyle intervention entailed a 500-kcal/day deficit diet and a minimum of 150 minutes of physical activity per week.
The purpose of the pooled analysis was to identify the best early predictor of response status at 56 weeks by examining the impact of 3%, 4%, and 5% weight loss after 8, 12, and 16 weeks of treatment as cut points. This post hoc analysis was prespecified at the request of the Food and Drug Administration before the trials were completed.
The bottom line: The best predictor of long-term outcome on liraglutide, a glucagonlike peptide–1 analog, proved to be a weight loss of 4% or greater at 16 weeks. It had an 81% positive predictive value and a 76% negative predictive value for at least a 5% weight loss at 56 weeks. It correctly predicted weight outcomes at 56 weeks in 80.1% of patients, the highest success rate of all the combinations studied. This finding was the impetus for the current product labeling, which contains the stopping rule. Dr. Fujioka shared study data not included in the labeling; namely, the marked contrast in how early responders and early nonresponders fared at 56 weeks.
The mean weight loss at 56 weeks in nondiabetic early responders to liraglutide was 10.8%, compared with only 3% in early nondiabetic nonresponders. Diabetic early responders averaged an 8.5% weight loss at 56 weeks, while early nonresponders had a mean 3.1% weight loss.
In the SCALE Obesity and Prediabetes trial, 50% of early responders to liraglutide ended up with a greater than 10% weight loss at 56 weeks, and 21% had more than 15% weight loss, compared with rates of 6% and 2%, respectively, in early nonresponders.
In the SCALE Diabetes study, 38% of early responders had greater than 10% weight loss long term, a rate nearly fourfold higher than in early nonresponders. Moreover, 10% of early responder diabetic patients had greater than 15% weight loss, versus a mere 2% of early nonresponders, the internist continued.
The ratio of early responders to early nonresponders in the nondiabetic population was 77%:23%. In diabetic patients, it was 63%:37%.
Turning to cardiometabolic endpoints, Dr. Fujioka noted that early responders in the SCALE Obesity and Prediabetes trial went on to show a mean reduction in systolic blood pressure of 5.1 mm Hg at 56 weeks, compared with a 2–mm Hg decrease in early nonresponders. Early responders also averaged a 10.5-cm shrinkage in waist circumference from a baseline of 115 cm, which was more than twice that observed at 56 weeks in early nonresponders. HDL-cholesterol level rose by 3.9% in early responders but remained unchanged over time in early nonresponders.
Diabetic patients who were early responders to liraglutide 3.0 mg/day had a mean 44.2-mg/dL reduction in fasting plasma glucose at 56 weeks from a baseline of 158 mg/dL, compared with a 30.1-mg/dL decrease in early nonresponders.
“The drop in fasting blood glucose is very quick – within a matter of weeks – so if you already have diabetic patients on drugs that are going to bring their blood sugar down, you may have to back titrate those other drugs really quickly. You don’t want to make your patients hypoglycemic,” the physician said.
Mean hemoglobin A1c values in early responder diabetic patients fell by 1.6% from a baseline of 7.9%, a full 0.5% greater reduction than in early nonresponders.
By far the most frequent adverse events in the two SCALE trials were gastrointestinal, with nausea leading the way. Rates were modestly higher in the early responders.