Signs of "Future Diseases" in the Blood of 8-Year-Olds: 5,000 Proteins Indicate a New Gateway to Preventive Medicine

Signs of "Future Diseases" in the Blood of 8-Year-Olds: 5,000 Proteins Indicate a New Gateway to Preventive Medicine

Heart disease, type 2 diabetes, and chronic kidney disease do not suddenly begin in adulthood. Small changes accumulate in the body for many years, sometimes decades, before symptoms appear. So, is it possible to detect these changes in children and alter their trajectory before they develop into diseases?

Research teams from Vanderbilt University Medical Center, the University of Texas Health Science Center at Houston, and the University of North Carolina at Chapel Hill have focused on numerous proteins circulating in the blood. The study, published in the journal "Nature Metabolism" on September 11, 2026, suggests that even children as young as eight may exhibit molecular patterns that overlap with adult cardiovascular, kidney, and metabolic diseases.

This does not mean that "a blood test can definitively predict future diseases." However, it might open a path to detect risks earlier and more precisely, which traditional indicators like weight and blood sugar levels might miss. The most significant message from the research is the reinterpretation of signs found in children's bodies not as a "predetermined fate" but as indicators of a period where intervention is still possible.


"CKMD" - Viewing the Heart, Kidney, and Metabolism as a Chain

At the core of the research is the concept of CKMD, which combines cardiovascular, kidney, and metabolic diseases. Obesity, insulin resistance, hypertension, lipid abnormalities, fatty liver, and kidney function decline do not necessarily progress independently. Dysregulation of blood sugar starting from visceral fat and chronic inflammation can increase the burden on blood vessels and kidneys, potentially leading to myocardial infarction or heart failure.

This chain is well-known in adult medical care. However, since children have not yet developed clear diseases like myocardial infarction, it is challenging to determine when the risk begins. Applying adult standards directly could overlook children whose numbers fall within the "normal range" but are already undergoing undesirable changes for their age.

According to the research team, tests commonly used in adults, like fasting blood sugar, may not capture a significant portion of blood sugar abnormalities or type 2 diabetes risk in children. Therefore, they targeted a collection of proteins that increase or decrease according to the body's state, known as the "proteome."


Extracting Six "Signatures" from About 5,000 Proteins

The subjects were 273 Hispanic/Latino children and young people participating in a long-term health study in Cameron County, Texas, located on the U.S.-Mexico border. The average age was 13.1 years, with 53% being female. More than one-third of the participants exhibited physical characteristics related to CKMD, such as obesity, elevated blood pressure, insulin resistance, and unfavorable blood lipids.

Researchers analyzed the relationship between approximately 5,000 proteins in the blood and 25 health indicators spanning the liver, adipose tissue, blood vessels, and blood sugar regulation. They then statistically summarized 16 indicators that could be measured in both children and adults and organized them into the following six potential patterns.

  • Fat Accumulation with Inflammation

  • Liver Fat and Liver Fibrosis

  • Cholesterol

  • Blood Pressure

  • Kidney Function

  • Insulin Resistance

Using LASSO regression, a type of machine learning, they created combinations of multiple proteins representing each condition. Instead of searching for a single "magic biomarker," they adopted an approach to read patterns played by numerous proteins. This is similar to how weather forecasts combine temperature, pressure, humidity, and wind direction.

The explanatory power of the model varied significantly by field. While the pattern for liver fat and fibrosis was relatively high, the models for blood pressure and insulin resistance were low. This difference is an important consideration when thinking about future clinical tests.


Children's Patterns Overlapped with Adult Diseases

Next, the research team verified whether the relationships found in children also applied to adults. Among 685 adults from the same region, they compared the relationships between proteins and health indicators, finding that 1,877 out of 2,916 significant combinations, or 64%, were also significant and in the same direction in adults. The strongest matches were in areas related to fat accumulation and the liver. For 494 of the adults, they directly applied the six protein scores derived from children to confirm their association with corresponding CKMD indicators.

Furthermore, they verified this with 28,256 participants from the UK Biobank, who have different geographical and ethnic backgrounds. The median follow-up period for mortality analysis was 13.7 years. The protein scores for "Fat Accumulation with Inflammation," "Liver Fat and Fibrosis," and "Insulin Resistance" created from children's data were associated with high risks of all-cause mortality, type 2 diabetes, multiple cardiovascular diseases, sleep apnea, and fatty liver in adults.

Notably, the protein pattern related to the liver was particularly strongly linked to adult outcomes across multiple organs. Fatty liver may not be just a liver issue but could serve as an early warning of metabolic abnormalities.

However, what was confirmed here is a "correlation." This study did not track the 273 children for decades to see which diseases they actually developed. Instead, molecular scores created from a group of children were applied to a different adult group, and they were found to be associated with adult diseases and mortality. It is not at a stage where it can predict on an individual level, "You will have a myocardial infarction in a few years based on this test value."


The Correct Interpretation of "Reversal" with GLP-1 Drugs

What is likely to attract the most attention in this study is the relationship with GLP-1 receptor agonists. Drugs like semaglutide are rapidly gaining popularity for the treatment of type 2 diabetes and obesity. The research team cross-referenced 1,032 proteins associated with some indicators of pediatric CKMD with the proteome data from the STEP 1 trial of semaglutide in adults.

As a result, proteins associated with unfavorable conditions in children tended to decrease after adults underwent 68 weeks of semaglutide treatment, while proteins associated with favorable conditions tended to increase. Leptin, FABP4, CES1, ACY1, and others decreased, while SHBG and IGFBP2 increased. Overall, the direction of change was consistent with health improvement.

There is something crucial that must not be confused here. This study is not a clinical trial that tested whether administering semaglutide to at-risk children can prevent future heart disease. The protein changes confirmed with the drug were in adults, and it is not clear whether these changes are due to the drug's direct action or effects mediated by weight loss. Therefore, it does not conclude that "children should be given GLP-1 drugs if they test positive."

Nevertheless, this result is valuable because it shows that protein patterns are not fixed from birth and can change with treatment, at least in adults. The "modifiable risk states" emphasized by researchers do not mean that treatment methods are already established but rather that there is a reasonable basis to explore the possibility of intervention.


Why Precision Prevention for Children is Gaining Attention Now

In the U.S., approximately one in five children and young people are classified as obese, with estimates from 2021 to 2023 indicating that 21.1% of those aged 2 to 19 are obese, and 7% are severely obese. This background is not only related to diet and exercise but also involves complex interactions with income, living environment, access to food, sleep, stress, and access to healthcare.

The use of GLP-1 drugs is also rapidly increasing. Prescriptions for those aged 12 to 25 reportedly increased by about 600% from 2020 to 2023. However, these drugs are expensive, have side effects like nausea, and there are still unanswered questions about the effects of long-term use from a young age and weight regain after discontinuation. Without tools to determine who would benefit the most, necessary children might not receive the treatment, while the treatment scope could expand to children with minimal benefit.

If proteome testing becomes practical in the future, it might allow for prioritizing monitoring, lifestyle support, specialized care, and pharmacotherapy based on molecular states of the liver, inflammation, and blood sugar regulation, rather than using drugs uniformly based on a certain weight. This is the image of precision prevention described by the research.

However, being precise does not equate to being fair. If expensive tests and drugs are inaccessible due to insurance or income, the technology could widen, rather than narrow, health disparities. Additionally, there is a need to protect data with caution akin to genetic information to prevent inappropriate handling of children's future risks by schools, insurance companies, employers, etc.


Reactions on SNS—Quiet Sharing from the Academic Community Before Major Debates

The paper and news release were just published on September 11, 2026, and as of the 12th, reactions on SNS were still limited. A representative example that could be confirmed was the sharing of the paper title and link by the official Nature Metabolism X account. At this point, a large number of comments from independent researchers, pediatricians, or parents, or major debates with divided opinions were not found in searches.

 

Outside of SNS, several scientific and medical news outlets reported on the potential to identify risks from as young as eight years old and the early detection of signs of adult diseases. However, the article comment sections had zero comments at the time of confirmation. A lack of reactions does not necessarily mean a lack of interest. Given the short time since the announcement and the highly specialized content, it is reasonable to consider that there is a time lag before the research results spread to the general public.

If discussions expand on SNS in the future, there are likely three main focal points. The first is the expectation of being able to know the future risks of asymptomatic children early. The second is the concern about labeling children based on weight, potentially increasing anxiety and prejudice. The third is the danger of simplistically perceiving GLP-1 drugs as a "panacea for early prevention." Especially if only headlines spread, there is a risk of misunderstanding as "weight-loss drugs from age eight." The study did not show an immediate recommendation for treatment but provided a molecular map for future verification of which interventions are effective for whom.


How to View the Limitations of This Study

The researchers themselves have pointed out several limitations. First, the core group of children is Hispanic/Latino living in a region of South Texas, with high obesity risk and many residents facing economic challenges. Although some external validation was achieved with the UK Biobank, it is unknown whether the same model applies to children with different ancestral backgrounds, income levels, lifestyles, and environmental exposures.

Next, there are influences unique to the growth period. During puberty, hormones and body composition change significantly, and blood proteins fluctuate. 72.5% of participants were concentrated in the pubertal stage, and the study could not sufficiently investigate differences from earlier developmental stages. There is also a lack of data on long-term repeated measurements of protein changes during childhood.

Furthermore, different protein measurement platforms were used for each group, which may have affected the results due to differences in detection range and accuracy. Although recalibration was performed in the analysis, to use it in actual clinical practice, it is necessary to confirm reproducibility in a large pediatric population measured using the same method, and clarify reference values, cost-effectiveness, and the rates of false positives and false negatives.


The Goal Should Be "Early Support" Rather Than "Early Use of Drugs"

The future envisioned by this study is not a world where a disease name is pronounced from a drop of a child's blood. Rather, it is a world where families and healthcare providers can start support early, using changes that were difficult to see with traditional checkups as clues. The foundation should be nutrition, exercise, sleep, psychological support, and improvement of family and community environments, with specialized tests and medications combined only when necessary. That order must not be forgotten.

Proteins in the blood are small messages from the heart, kidneys, liver, adipose tissue, pancreas, and other organs. This study showed that when these countless messages are bundled, the outline of adult diseases might be faintly visible from childhood.

However, having a map is different from establishing a safe path. What is needed next is research that tracks diverse children over the long term and trials that confirm whether interventions in children improve not only protein patterns but also actual health and quality of life. What should be expected is not a "test that predicts the future" but "medicine that increases options before disease develops."


Sources

Related Articles