The ability to identify obese subjects who will lose weight in response to energy restriction is an important strategy in obesity clinical care.
Fifty obese/overweight subjects were submitted to an energy restricted high protein diet for 6 weeks followed by a 6 weeks-maintenance diet. Based on their trajectories of weight loss during the study, three subjects clusters were identified. Cluster A (n=17) and Cluster B (n=15) lost more weight during the diet period, however during the stabilization phase cluster A continued to loose weight, whereas cluster B remained stable. Cluster C (n=17) lost less and rapidly regained weight during the stabilization. At baseline, subjects in cluster C had the highest plasma insulin (P=0.01), IL-6 (P=0.05) and adipose tissue inflammatory marker (HAM56, P=0.03), and the lowest plasma free fatty acid excursions after a glucose charge (FFA AUC, P=0.03). Gut microbiota was profiled from faecal samples by qPCR in all the subjects. Intriguingly subjects in cluster C had the highest level of Lactobacillus/Leuconostoc/Pediococcus group before diet. During the dietary program subjects in cluster C consumed more starchy foods, less protein and raw vegetables. Spearman correlations revealed positive relationship between weight regain after diet and HOMA-IR (P=0.0002, rs=0.5), inflammatory markers (leucocytes numbers: P=0.05, rs=0.27; Neutrophils: P=0.05, rs=0.28; IL-6: P=0.002, rs=0.43) as well as the number of Lactobacillus/Leuconostoc/Pediococcus group (p=0.005, rs=0.4) at baseline. Bayesian network (BN) was performed for prediction of the 3 clusters by using the data prior to the weight-loss dietary program. According to the learnt structure of BN, the levels of 4 biomarkers (plasma insulin, IL-6, leucocytes numbers and adipose tissue HAM56) at baseline were sufficient to characterize the distribution of the 3 clusters. The prediction of clusters was 75.5% (37 among 49 subjects). We concluded that individual responses to hypocaloric high protein dietary program could be predicted by plasma insulin, IL-6, leucocytes numbers and adipose tissue HAM56 levels prior to diet.