Searchable abstracts of presentations at key conferences in endocrinology
Endocrine Abstracts (2019) 63 GP139 | DOI: 10.1530/endoabs.63.GP139

1University of Milan, Milan, Italy; 2University of Copenhagen, Frederiksberg C, Denmark; 3Health & Care, sro, Prague, Czech Republic; 4General Hospital in Prague, Prague, Czech Republic; 5Sapienza University of Rome, Rome, Italy; 6University of Navarra, Pamplona, Spain; 7CIBERobn and IMDEA Food Institute, Madrid, Spain; 8Pennington Biomedical Research Center, Baton Rouge, USA; 9Weill Cornell Medicine, New York, USA; 10Boston University, Boston, USA; 11Massachusetts General Hospital, Boston, USA; 12University of Alabama at Birmingham, Birmingham, USA; 13Scripps Clinic, San Diego, USA; 14Geisinger Obesity Institute, Danville, USA; 15Gelesis, Inc., Boston, USA.


Introduction: Identification of early responders is widely used in pharmacotherapy for weight loss to minimize unnecessary exposure to risks and unnecessary expenses. Gelesis Loss Of Weight (GLOW; NCT02307279), a multicenter, double-blind, placebo-controlled pivotal study, demonstrated that Gelesis100 offers a compelling new potential approach in the management of overweight and obesity.

Objective: A post-hoc area under the curve (AUC) for the receiver operating characteristic (ROC) analysis was conducted on data from the GLOW study to identify the optimal timepoint for predicting body weight responders (≥5% body weight loss at Week 24).

Material and methods: An AUC of the ROC was completed as a post-hoc analysis to model the predictive power of early weight loss for responder status in 322 subjects who were treatment completers. The AUC ROC analysis optimized the balance between the positive predictive value (PPV) or percent of subjects with early weight loss response who were body weight responders and the negative predictive value (NPV) or percent of subjects with early weight loss response who were not body weight responders. The cutoff for PPV and NPV was set at 80%.

Results: AUC ROC analysis showed that an early response to Gelesis100 treatment (≥ 3% weight loss from baseline at Week 8) successfully predicted clinically meaningful weight loss (≥ 5%) at Week 24. More than 85% of subjects who achieved ≥ 5% weight loss at Week 24 had lost ≥ 3% of body weight at Week 8. Notably, early responders (n=93) achieved a mean weight loss of 9.9% vs. 2.1% in non-early responders (n=77) at Week 24. The placebo arm did not reach the required threshold for sensitivity and specificity until near the end of treatment. Safety and tolerability of Gelesis100 in the early responders were similar to the non-early responders, except for the incidence of abdominal pain that was higher in the early responders (P=0.0415) but of mild intensity in the majority of cases, and demonstrated no increased risk compared to placebo. No serious adverse events were observed.

Conclusion: The results of AUC ROC analysis suggest that a weight loss of ≥ 3% as early as 8 weeks after treatment with Gelesis100 is predictive of clinically meaningful weight loss (≥ 5%) after 24 weeks. Although there are no overall increased safety risks with Gelesis100 treatment, predicting successful treatment early allows efficient use of resources and provides a key milestone to motivate subjects with overweight or obesity for personalized care.

Volume 63

21st European Congress of Endocrinology

Lyon, France
18 May 2019 - 21 May 2019

European Society of Endocrinology 

Browse other volumes

Article tools

My recent searches

No recent searches.