1. ¹Department of Clinical Pharmacology, Merck Research Laboratories, Rahway, New Jersey, USA
2. ²Office of the Director, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, Maryland, USA
3. ³Biostatistics, Quintiles, Austin, Texas, USA
4. ⁴Diabetes/Endocrine Medical, Eli Lilly and Company, Indianapolis, Indiana, USA
5. ⁵Metabolic Diseases, Roche, Nutley, New Jersey, USA
6. ⁶Metabolic Pathways Center of Excellence for Drug Discovery, GlaxoSmithKline, Research Triangle Park, North Carolina, USA
7. ⁷Office of Cellular, Tissue, and Gene Therapies, Center for Biologics Evaluation and Research, US Food and Drug Administration, Rockville, Maryland, USA
8. ⁸Global Statistical Sciences, Lilly Corporate Center, Eli Lilly and Company, Indianapolis, Indiana, USA
9. ⁹Late Development Statistics, Merck Research Laboratories, Rahway, New Jersey, USA
10. ¹⁰P5 Medicines, Cary, North Carolina, USA
11. ¹¹The Biomarkers Consortium, Foundation for the National Institutes of Health, Bethesda, Maryland, USA

Correspondence: MT Vassileva, (mvassileva@fnih.org)

Received 1 April 2009; Accepted 7 April 2009; Published online 24 June 2009.

Abstract

This study, conducted under the Metabolic Disorders Steering Committee of the Biomarkers Consortium (a public–private partnership managed by the Foundation for the National Institutes of Health (FNIH)), analyzed blinded data on 2,688 type 2 diabetes (T2D) patients from randomized clinical trials conducted by four pharmaceutical companies. An increase in the levels of adiponectin was observed after peroxisome proliferator–activated receptor (PPAR)-agonist treatment (P < 0.0001), but not after treatment with non-PPAR drugs. This increase correlated with decreases in levels of glucose, hemoglobin A_1c (Hb_A1c), hematocrit, and triglycerides, and increases in levels of blood urea nitrogen, creatinine, and high-density lipoprotein cholesterol (HDL-C). Early (6–8 weeks) increases in levels of adiponectin after treatment with PPAR agonists showed a negative correlation (r = -0.21, P < 0.0001) with subsequent changes in levels of Hb_A1c. Changes in adiponectin level did not appear to be associated with baseline level of Hb_A1c. Logistic regression demonstrated that an increase in the level of adiponectin predicts a decrease in the level of Hb_A1c. These analyses confirm previously demonstrated relationships between adiponectin levels and metabolic parameters and support the robust predictive utility of adiponectin across the spectrum of glucose tolerance. Cross-company precompetitive collaboration is a feasible and powerful approach to biomarker qualification.