India is often called the diabetes capital of the world. With an estimated 101 million adults currently living with the condition and another 136 million at risk with prediabetes, Type 2 diabetes has become a major public health concern in the country. The condition arises when blood sugar levels remain too high and cannot be regulated properly. Genetics, obesity, sedentary lifestyles, unhealthy diets, family history, high blood pressure, abnormal cholesterol, gestational diabetes, and stress all contribute to driving the disease. Because many cases are diagnosed late, complications affecting the eyes, kidneys, nerves, heart, and brain are common. Nearly one-third of patients also develop chronic kidney disease.
Standard clinical tests for diabetes typically measure fasting blood glucose, HbA1c, and creatinine levels for assessing kidney health. However, these tests capture only a fraction of the complex biochemical disruptions underlying the disease and cannot always predict who faces the greatest risk.
In a new study, researchers led by Prof. Pramod Wangikar from the Indian Institute of Technology Bombay (IIT Bombay) and Dr. Rakesh Kumar Sahay and Dr. Manisha Sahay from Osmania Medical College, along with researchers from Clarity Bio Systems India Pvt. Ltd., Pune, used metabolomics—the study of small molecules in the blood—to find biochemical patterns that could help identify patients at risk of kidney complications. Their findings, published in the Journal of Proteome Research in July 2025, may help doctors diagnose problems earlier and create more personalised treatments.
“Type 2 diabetes is not just about high blood sugar. It disrupts amino acids, fats, and other pathways in the body. Standard tests often miss this hidden activity, which may often begin years before the onset of clinical symptoms,” explains Ms. Sneha Rana, a Ph.D. scholar in Prof. Wangikar’s lab and the first author of this study.
To gain a more comprehensive view of the disease, researchers performed metabolic profiling of hundreds of metabolites at once. Metabolites are tiny molecules present in the body that reflect the ongoing activity in cells. By analysing them, hidden shifts in body chemistry that precede clinical symptoms can be detected.
Previous metabolomics studies have highlighted associations between diabetes and molecules such as branched-chain amino acids (BCAAs), acylcarnitines, and certain lipids. But most of that research has been done in European or East Asian populations. Since genetics and lifestyles vary across populations, markers found in one region may not be applicable elsewhere.
“Indians often develop diabetes at a younger age and lower body mass index than Western populations, and they face higher risks of complications like kidney disease. It was important to check if Indian patients show different metabolic patterns,” says Dr. Rakesh Sahay.
For the study, the team collected whole blood samples from 52 volunteers at Osmania General Hospital in Hyderabad between June 2021 and July 2022. The participants included 15 healthy controls, 23 patients with type 2 diabetes, and 14 patients with diabetic kidney disease (DKD). Using two complementary techniques — liquid chromatography-mass spectrometry (LC-MS) and gas chromatography-mass spectrometry (GC-MS) — the team scanned for nearly 300 metabolites.
They found 26 metabolites that differed between diabetic patients and healthy controls. Some were expected, such as glucose, cholesterol, and 1,5-anhydroglucitol (a short-term marker of blood sugar). But others, like valerobetaine, ribothymidine, and fructosyl-pyroglutamate, had not been linked to diabetes before.
“This suggests that diabetes is a much broader metabolic disorder beyond just glucose dysregulation,” notes Prof. Wangikar.
The researchers also discovered two distinct subgroups among diabetic patients. One group appeared metabolically closer to healthy individuals, while the other showed major changes related to stress, inflammation, and energy production.
“These markers could one day be used by doctors, just like cholesterol tests are used to assess heart disease risk. This means some patients may need more aggressive treatment, while others might benefit more from lifestyle changes,” notes Dr. Rakesh Sahay.
When comparing patients with kidney disease to the other groups, the team identified seven metabolites that steadily increased from healthy to diabetic to DKD patients. These included sugar alcohols like arabitol and myo-inositol, as well as ribothymidine and a toxin-like compound called 2PY, which accumulates when the kidneys are damaged.
“By monitoring these molecules, we could predict kidney complications much earlier,” says Ms. Rana. “These markers could replace or add on to the existing markers like creatinine, eGFR estimation and measuring albuminuria by identifying those diabetics at risk of kidney disease even before these markers indicate the onset of kidney dysfunction, facilitating the early use of therapies which prevent progression of kidney disease,” says Dr. Manisha Sahay.
Unlike most earlier studies, which looked only at the liquid part of the blood (plasma or serum), this research analysed whole blood.
“The work has potential for clinical translation as a test could be developed based on dried blood spots made from finger pricks, and this is an ongoing work in our lab,” says Prof. Wangikar.
Whole blood captures metabolites from red blood cells in addition to plasma, giving a more comprehensive but different metabolic snapshot. This advantage of whole-blood testing may also explain why some well-known markers from Western studies, like BCAAs, did not show up strongly here. Their abundance and distribution can vary between whole blood and plasma, and across populations.
This study currently has a small sample size. The researchers plan to extend it to obtain data from a larger number of subjects with diabetes and various complications, with the objective of developing simple clinical tests that can not only help in the early diagnosis of diabetes but also identify those at higher risk of complications. This will pave the way for personalised care in the long run.
“In India, we often rely on a one-size-fits-all approach to diabetes treatment. With these new markers, we can start tailoring care to each patient’s unique profile,” concludes Ms. Rana.
Funding information: This study was funded by Koita Center for Digital Health, IIT Bombay and Department of Biotechnology.