Prediabetes is a common condition affecting hundreds of millions of people worldwide and, if untreated, it often leads to type 2 diabetes and related problems such as heart and kidney disease. Now, researchers have found that gut microbiota composition can predict whether a person with prediabetes will benefit from dietary changes such as an increased fiber intake.

The findings, published in Nature Communications, suggest that gut microbiota profiles can help support personalized dietary interventions.

Lifestyle changes such as diet and exercise can help people with prediabetes, but individual responses vary due to differences in genetics and gut bacteria. The gut microbiota plays a key role in breaking down dietary fiber to improve insulin activity. However, whether microbial features can predict individual response remains unclear. 

Researchers led by Delei Song at Shanghai Jiao Tong University in China enrolled more than 800 people with prediabetes in a clinical study to investigate whether, and how, dietary fiber supplements can help improve blood sugar control.

Prediabetes subgroups

Study participants were randomly assigned to receive standard care with or without fiber supplements. Researchers grouped participants using multiple health measures, including age, weight, and blood sugar, rather than just blood sugar, and analyzed their gut microbiota. 

This approach identified four subgroups of people with prediabetes, each with distinct health profiles, such as differences in insulin production, heart and liver health, and family history of diabetes. These subgroups also showed differences in gut bacteria and blood metabolites, with some having less diverse gut microbiotas.

Only two of these subgroups benefited from dietary fiber, and this improvement depended on whether their gut bacteria could respond to fiber, the researchers found. 

Personalized medicine 

To predict who would benefit from dietary fiber, they researchers created a score based on changes in three key blood sugar measures. Then, they used machine learning to link these outcomes to specific gut bacteria. A set of specific gut bacteria could predict with good accuracy whether a person would respond well to fiber, the team found. 

Next, the researchers tested this approach in two independent groups of people with type 2 diabetes, showing it could predict both short-term and long-term benefits.

“Our study suggests that the gut microbiota response influences the effectiveness of dietary fiber intervention and provides a clinically applicable model to guide microbiome-targeted personalized medicine for prediabetes,” the authors say.