New Tool Makes Use of People’s Genetic Blueprint to Predict Obesity Risk

Ana Pena, PhD avatar

by Ana Pena, PhD |

Share this article:

Share article via email
genes, athletic endurance

Researchers have created a tool — a genome-wide polygenic score (GPS) — to predict a person’s genetic susceptibility to excess weight and obesity from early life into adulthood.

This predictive tool could offer new opportunities to improve clinical prevention and provide new insights on the underlying causes of genetic obesity.

The study, “Polygenic Prediction of Weight and Obesity Trajectories from Birth to Adulthood, was published in the journal Cell.

Even though obesity is many times thought of as a result of unhealthy lifestyles or environmental factors — which often leads to stigmatization of obese people — a great part of obesity can stem from genetic causes.

Most inherited cases, however, are not caused by mutations at a single gene (e.g. inactivation of MC4R gene), but instead are thought to derive from the cumulative effect of many genetic variants in one person, an effect referred to as a “polygenic.” Genetic variants are slightly different forms of the same gene that determine specific traits, such as eye or hair color. 

Alone, each of those variants would have a rather modest effect, but together their effects can add up to stronger consequences, such as obesity.

A recent genome-wide association study (GWAS) — a genome based approach to find genetic variations linked to a particular disease — measured the association between 2.1 million common genetic variants and the BMI of more than 300,000 people. At the time, however, no striking effects were attributed to any of those variations alone.

Now, a team led by researchers at Massachusetts General Hospital and Harvard Medical School went on to reanalyze those data in a search for polygenic combinations that could predict increased genetic risk of obesity.

They built six candidate genome-wide polygenic scores (GPS), which also integrated information from 503 people of European ancestry from the 1000 Genomes Study.

To identify the best score, researchers tested which correlated the most with BMI using a dataset of 119,951 adults, ages 40 to 69, enrolled in the UK Biobank.

The best score “considerably outperformed earlier scores” and its predictive power in respect to BMI, weight, and severe obesity was validated in a dataset of 306,135 people, ranging from middle age to time of birth, enrolled in four independent studies.

The results showed the score each person was attributed matched well the differences in weight and severe obesity within middle-aged adults.

Compared with the remaining individuals, those with a high score were associated with a 4.2-, 6.6-, and 14.4-fold increased risk of a BMI of 40, 50, and 60 kg/m2 or more, respectively.

Extreme scores were more commonly associated with people with severe obesity, indicating an association between both.

In fact, an extreme score was as predictive of a greater BMI as an MC4R mutation well-known to cause a form of genetic obesity, strengthening the commonly held belief that a combination of gene variants can equal genetic obesity disorders caused by a single gene.

Beyond weight, the score also associated with risk of diseases commonly associated with obesity. A higher score correlated with a 28% greater risk of coronary artery disease, a 72% increased risk for diabetes, a 38% increased risk for high blood pressure, a 34% increased risk for congestive heart failure, a 23% increased risk for ischemic stroke, and a 41% increased risk for venous thromboembolism.

In agreement with the prior observation, a higher scored correlated with a 19% greater risk of death.

One important question researchers wanted to see answered was whether people with a genetic susceptibility for obesity already have quantifiable differences in weight at birth.

They explored this question by looking at 7,861 subjects enrolled in the Avon Longitudinal Study of Parents and Children (ALSPAC) which followed individuals from the time of their birth to 18 years of age.

The results showed that birthweight had only a minimal correlation with the polygenic score. However, this correlation got stronger during early childhood, with even larger differences as individuals entered adulthood.

Newborns within the top polygenic scores were born with a mean weight of 3.47 kg, whereas those in the lowest scores were born with a little less than that (60 grams). By age 8, however, this difference increased to 3.5 kg. and by 18 years, it reached 12.3 kg.

The risk of developing severe obesity as young adults aged also varied according to the polygenic score. About 15.6% of young adults in the top scores ended up developing severe obesity, compared with 5.6% of those in lower scores and only 1.3% of those in the lowest score.

“The ability to predict disease via genome interpretation will raise both important opportunities and potential challenges for clinical medicine,” senior author Sekar Kathiresan, director of the Center of Genomic Medicine at the Massachusetts General Hospital and the Cardiovascular Disease Initiative at the Broad Institute, said in a press release.

“We are in the early days for figuring out how and when best to disclose genetic information and how we can best empower patients to overcome any genetic risks identified, but we are incredibly excited about the potential,” he said.

By identifying who is more likely to become obese, the tool might also shed light on the molecular processes behind the disorder and help uncover novel risk factors.

“Finally, a clear understanding of the genetic predisposition to obesity may help to de-stigmatize obesity among patients, their health care providers, and the general public,” the team stated.

Despite the strength of these associations, “polygenic susceptibility to obesity is not deterministic,” researchers warned. This is to say that a high score does not necessarily mean a person will have excess weight or become obese. Environmental influences, other genetic variations, or other factors may well enter this equation and contribute to increase or attenuate such risk.

Recent Posts