Breakthrough Discovery: Protein Markers in Cord Blood Signal Future Type 1 Diabetes Risk

Recent research from the University of Florida (UF) in collaboration with Linköping University has unveiled a promising avenue for predicting type 1 diabetes (T1D) right at birth through analysis of umbilical cord blood. Type 1 diabetes, an autoimmune condition where the body's immune system attacks insulin-producing beta cells in the pancreas, affects approximately 1.6 million Americans, with incidence rising by about 3-4% annually in the United States. Traditional prediction methods rely on genetic screening for human leukocyte antigen (HLA) risk alleles or detection of autoantibodies, which appear only after the disease process has begun. However, this new study identifies specific protein biomarkers in cord blood— the blood remaining in the umbilical cord and placenta after birth—that correlate strongly with later T1D development, offering a noninvasive, early detection opportunity.

Led by assistant research scientist Angelica Ahrens at UF's Institute of Food and Agricultural Sciences (IFAS), the team utilized machine learning algorithms on proteomics data from cord blood samples. Proteomics involves studying the entire set of proteins (proteome) to understand biological processes, here revealing an inflammatory signature present even in utero. This discovery suggests that the foundational biological pathways leading to T1D may initiate during fetal development, shaped by prenatal factors, though not predetermining the disease.

The study drew from the All Babies in Southeast Sweden (ABIS) cohort, tracking 16,683 children born between 1997 and 1999. Researchers analyzed cord blood from 146 who later developed T1D and 286 matched controls, employing Olink proteomics platforms to measure hundreds of inflammatory and immune proteins. Machine learning models identified a subset predicting T1D with an area under the curve (AUC) of 0.89—indicating high accuracy—independent of genetic HLA risk. This outperforms many current tools and highlights inflammation's role from fetal life.

Decoding Type 1 Diabetes: From Autoimmunity to Lifelong Management

Type 1 diabetes mellitus (T1D) typically manifests in childhood or adolescence, though adult-onset cases occur. The destruction of pancreatic beta cells leads to absolute insulin deficiency, requiring lifelong exogenous insulin therapy via injections or pumps. In the US, it accounts for 5-10% of diabetes cases, with lifetime costs exceeding $1 million per patient due to complications like neuropathy, retinopathy, and cardiovascular disease. Early prediction is crucial because interventions like immunomodulatory drugs (e.g., teplizumab, FDA-approved in 2022 for delaying onset) work best presymptomatically.

Current screening in high-risk families uses HLA genotyping from cord blood or heel-prick samples, identifying 90% of future cases but with low positive predictive value (PPV) of 1-5%. Autoantibody testing (e.g., GAD65, IA-2, insulin, ZnT8) confirms presymptomatic stages but requires repeated blood draws post-infancy. Cord blood biomarkers could revolutionize this by providing a one-time, discarded-sample analysis at birth, avoiding ethical concerns of newborn screening for non-treatable conditions.

The Power of Cord Blood: A Window into Fetal Health

Umbilical cord blood, rich in hematopoietic stem cells, reflects late fetal metabolism, maternal influences, and early immune programming. Typically discarded post-delivery, banking it for research or therapy is routine in studies like ABIS or TEDDY. Step-by-step, collection involves clamping the cord, drawing 20-50 mL into tubes, freezing, and later thawing for assays. Proteomics detects low-abundance proteins via targeted panels (e.g., Olink's 3,000+ protein targets), followed by data normalization and AI-driven pattern recognition.

In the UF study, UF's HiPerGator supercomputer—America's fastest university-owned high-performance computing system—enabled scalable analysis of vast datasets, underscoring universities' role in computational biology.

Spotlight on the UF-Llinköping Collaboration: Methodology and Results

The interdisciplinary team, including UF Microbiology Chair Eric Triplett and Linköping's Johnny Ludvigsson, refined predictive models iteratively. Key proteins involved innate immunity, cytokine signaling, and extracellular matrix remodeling, pointing to fetal inflammation priming beta-cell vulnerability. Notably, maternal exposure to per- and polyfluoroalkyl substances (PFAS), ubiquitous 'forever chemicals' in water and food, correlated with altered profiles, suggesting environmental epigenetics.

Ahrens noted, "Biology is being shaped during a period when systems are still highly adaptable," emphasizing prevention windows. Triplett highlighted cord blood's non-invasive appeal over genetic or autoantibody tests.

TEDDY Study: US-Led Quest for Environmental Triggers in T1D

The Environmental Determinants of Diabetes in the Young (TEDDY) study exemplifies US higher education's leadership. Funded by NIH, it screened over 400,000 newborns across six US sites—University of Florida, University of Pittsburgh, University of Colorado, University of Washington, Kaiser Permanente Georgia, and Washington—for HLA risk via dried blood spots (proxy for cord blood). Of 8,500 high-risk infants enrolled since 2004, TEDDY tracks diet, infections, microbiome, and metabolome to pinpoint T1D triggers.

Recent TEDDY findings include gene expression biomarkers stratifying progression rates and calculators combining genetics with family history for PPV up to 15%. While not proteomics-focused, TEDDY's biorepository supports biomarker validation, with USF receiving $69.9M in 2021 to continue.

Building on Prior Research: Lipids, Metabolites, and Epigenetics

Earlier Finnish DIPP and Norwegian MoBa studies found decreased phospholipids (e.g., phosphatidylcholines) in cord blood of future T1D cases, predicting autoimmunity with moderate accuracy. Metabolomics showed no strong prediction beyond genetics (AUC ~0.5), but 2025 methylome research linked maternal T1D to offspring islet autoimmunity risk via DNA methylation at susceptibility loci.

• Lipidome alterations: Lower choline phospholipids at birth precede beta-cell autoimmunity by years.
• Metabolites: Bile acids, amino acids neutral in large cohorts.
• Epigenetics: Cord blood methylation predicts metabolic dysfunction up to 18 years later.

UF's proteomics advances these, integrating multi-omics for robust models.

Toward Clinical Translation: Screening Tests and Prevention Strategies

Imagine routine cord blood proteomics at birth, flagging high-risk infants for monitoring or trials. Benefits include:

• High AUC prediction without genetics/privacy issues.
• Targeted interventions: Probiotics, antigen-specific immunotherapy, or PFAS avoidance.
• Cost savings: Early teplizumab delays onset by 2-3 years, reducing complications.

Challenges: Validation in diverse US populations, assay standardization, ethical newborn screening. Ongoing trials like cord blood infusions (NCT00305344) explore regenerative potential, though results mixed.

Career Opportunities in T1D Biomarker Research at US Universities

US universities drive T1D innovation, offering roles for immunologists, bioinformaticians, and clinicians. UF's Diabetes Institute seeks associate/full professors; Human Islet Research Network (HIRN) posts research associates at City of Hope. Postdocs analyze TEDDY data, leveraging HiPerGator-like resources.

Professionals can find research jobs, clinical research positions, and postdoc opportunities advancing biomarker discovery. Craft a standout CV with tips from our career advice. Rate faculty via Rate My Professor or browse university jobs.

Stakeholder Perspectives: From Researchers to Families

Ludvigsson emphasizes multifactorial prevention: "Different early changes of lifestyle and environment may gradually make type 1 diabetes less common." Families in TEDDY report empowerment through risk knowledge, though anxiety risks exist. Pediatric endocrinologists advocate balanced counseling, focusing on modifiable factors like diet and pollutants.

Future Outlook: Multi-Omics and Personalized Prevention

Integrating proteomics with TEDDY's genomics/microbiome promises AUC >0.95 predictions. AI advancements and biorepositories will accelerate trials. By 2030, cord blood panels could be standard, slashing US T1D burden ($14.4B annually). US higher ed remains pivotal, training next-gen scientists.

For higher ed career guidance, visit higher ed career advice or search higher ed jobs.

Photo by Alan Johnson on Unsplash