Get access

Genetic risk profiles for a childhood with severely overweight



What is already known about this subject

  • Genome-wide association studies have identified susceptibility loci, conferring only small effects to obesity.
  • We selected 12 from the most accepted susceptibility genes to overweight, which have been corroborated in different populations and series of thousands of adult individuals.
  • Single-nucleotide polymorphisms variants can have a cumulative effect on common forms of obesity, to create a genetic risk score.

What this study adds

  • We concentrated on 109 single-nucleotide polymorphisms (SNPs) located in 12 loci previously reported in association to body mass index at genome-wide significant level, for individuals of European ancestry.
  • To identify the frequency and effect size of common obesity SNPs in a discovery subsample of a group of children with high familial predisposition to morbid obesity.
  • To select those genetic variants that best associate individuals with high-susceptibility genetic profiles to overweight, and to create and validate a genetic risk score (GRS).
  • We compared the predictive power of each approach in an independent birth cohort, not aimed at studying obesity, and which can be considered as a suitable proxy to measure the clinical relevance of our GRS.


The objective of this study was the description of a valid genetic risk score (GRS) to predict individuals with high susceptibility to childhood overweight by their genetic profiles.

Design and methods

Case-control study including a group of children with high-risk familial predisposition to morbid obesity. Birth cohort from general population constituted the validation sample. For the discovery sample, 218 children with non-syndromic obesity and 190 control individuals were included. The validation sample was 653 children from two birth cohorts belonging to the INMA (Infancia y Medio Ambiente [Environment and Childhood] )project. 109 SNPs located in the genes FTO, SEC16B, BDNF, ETV5, SH2B1, GNPDA2, LYPLAL1, MSRA, TFAP2, KCTD15, MTCH2 and NEGR1, previously reported in association to body mass index (BMI) were analysed. For the validation sample, association between genome-wide data and BMI measurements between 3.5 and 5 years of age, were evaluated.


The GRS includes six SNPs in the genes FTO, TFAP2B, SEC16B, ETV5 and SH2B1. The score distribution differs among cases and controls (P = 9.2 × 10−14) showing a significant linear association with obesity (odds ratio [OR] per allele = 1.69; confidence interval [CI] 95% = 1.46–1.97; P = 4.3 × 10−1 and area under the receiver operating characteristic curve [AUC] = 0.727; CI 95% = 0.676–0.778). The results were validated by the INMA cohort (OR per allele = 1.23 CI 95% = 1.03–1.48 and AUC = 0.601 CI 95% = 0.522–0.680).


The use of our proposed genetic score provides useful information to determine those children who are susceptible to obesity. To improve the efficiency of clinical prevention and treatment of obesity, it is essential to design individualized based protocols in advance knowledge of the molecular basis of inherited susceptibility.