Circulating plasma microRNAs miR-150 and miR-375 levels are associated with age-related endotypes of newly diagnosed Type 1 Diabetes

This study identifies circulating plasma miR-150-5p and miR-375-3p as robust, non-invasive biomarkers that distinguish aggressive, early-onset Type 1 diabetes endotypes from later-onset forms, offering a novel tool for endotype-centered patient stratification and clinical trial enrichment.

The Gist

The Big Picture: One Disease, Many Faces

Imagine Type 1 Diabetes (T1D) isn't just one single disease, but rather a "shape-shifter." For a long time, doctors treated everyone with T1D the same way: "You have high blood sugar, so take insulin."

However, this study suggests that T1D actually comes in two very different "flavors" or endotypes (think of them as different species of the same animal):

1. The "Turbo" Version (T1DE1): This happens mostly in young children (under 7). It's aggressive, fast, and destructive. The body's immune system attacks the insulin-producing cells (beta cells) with a heavy, rapid fire.
2. The "Slow Burn" Version (T1DE2): This happens mostly in older kids and adults (13+). It's slower, more gradual, and the body holds onto some insulin production for longer.

The Problem: Right now, the only way to tell which "flavor" a patient has is to look at their age. But age isn't a perfect biological test. A 12-year-old could be either type, and we need a better way to know so we can treat them correctly.

The Solution: The "Smoke Signals" (MicroRNAs)

The researchers wanted to find a blood test that could act like a "smoke signal" to tell them which version of the disease a patient has, without needing to look at their age or do a biopsy of their pancreas.

They looked for tiny molecules called microRNAs (miRNAs).

• The Analogy: Imagine your cells are factories. When a factory is under attack or breaking down, it doesn't just stay silent; it sends out tiny "emergency flares" or "smoke signals" into the bloodstream. These flares are the microRNAs.
• The Goal: Catch these flares in a blood sample to see what kind of trouble is happening inside the body.

The Discovery: Two Specific Flares

The team analyzed blood samples from over 260 newly diagnosed patients. They found two specific "flares" that were much brighter in the "Turbo" group (T1DE1) than in the "Slow Burn" group (T1DE2):

1. miR-150 (The Immune Alarm): This flare comes from the immune system's soldiers (B-cells and T-cells).
   • What it means: In the "Turbo" group, the immune system is screaming "ATTACK!" very loudly.
2. miR-375 (The Beta-Cell Distress Call): This flare comes specifically from the insulin-producing cells (beta cells).
   • What it means: In the "Turbo" group, the insulin factories are being destroyed so fast that they are sending out massive distress signals.

The Key Finding: Even though these two groups are separated by age, the biology is what matters. The researchers proved that these flares weren't just there because the kids were young; they were there because the disease was aggressive.

The Twist: The "Middle Ground" Problem

There was a tricky group of patients: kids aged 7 to 12. They didn't fit neatly into "Young/Turbo" or "Old/Slow." They were the "Middle Ground" (T1DE-Int). Doctors didn't know if these kids were heading toward the aggressive "Turbo" path or the slower "Slow Burn" path.

The New Strategy:
Instead of asking, "How old are you?", the researchers asked, "What do your smoke signals say?"

They created a Traffic Light System based on the two flares:

• Red Light (miRC1): Both the Immune Alarm (miR-150) AND the Distress Call (miR-375) are HIGH.
  • Verdict: This patient is biologically similar to the aggressive "Turbo" group, even if they are 12 years old. They need aggressive treatment.
• Green Light (miRC2): The signals are lower.
  • Verdict: This patient is biologically similar to the "Slow Burn" group.

Why This Matters (The "So What?")

This is a game-changer for Precision Medicine.

1. Better Treatment: If a 12-year-old has the "Red Light" profile, doctors know immediately that their disease is aggressive. They might need stronger immunotherapy (medicine to calm the immune system) right away, rather than waiting.
2. Clinical Trials: When testing new drugs, researchers can now group patients by their biology (Red Light vs. Green Light) rather than just their age. This makes trials more accurate and helps find cures faster.
3. Non-Invasive: It's just a blood draw. No need for painful biopsies or guessing based on age.

Summary Analogy

Imagine two houses on fire.

• House A is a small shed with a massive, raging inferno (The "Turbo" T1D).
• House B is a large mansion with a slow, smoldering fire (The "Slow Burn" T1D).

Previously, firefighters (doctors) only looked at the address (Age) to decide how many trucks to send. If the address was "123 Kids Lane," they sent a huge team. If it was "456 Adult Ave," they sent a smaller team.

But sometimes, a 12-year-old lives in a mansion that is burning like a shed, or a 7-year-old lives in a shed that is just smoldering.

This study invented a Smoke Detector (the blood test). Now, instead of looking at the address, the firefighters look at the smoke. If the smoke is thick and black (High miR-150 and miR-375), they send the heavy trucks immediately, regardless of the address. If the smoke is light, they send a standard crew.

This ensures the right resources go to the right house, saving more "homes" (beta cells) and lives.

Technical Summary

1. Problem Statement

Type 1 Diabetes (T1D) is a heterogeneous disease with distinct pathogenic pathways. Recent histopathological studies have identified two primary immunopathological endotypes based on age at onset:

• T1DE1: Early onset (<7 years), characterized by aggressive lymphocytic insulitis (high CD20+ B cells and CD8+ T cells), interferon signatures, and rapid, extensive $\beta$-cell destruction.
• T1DE2: Later onset (≥13 years), characterized by milder insulitis, slower $\beta$-cell decline, and preserved C-peptide levels.

The Gap: While age at onset correlates with these endotypes, it is an imperfect proxy for the underlying tissue immunopathology. There is a critical lack of scalable, non-invasive blood-based biomarkers that can distinguish these endotypes at diagnosis, clarify the underlying immune-islet biology, and allow for the re-stratification of patients (including those in the "intermediate" age group, 7–12 years) for precision medicine and clinical trials.

2. Methodology

The study utilized a multi-omics approach across two independent INNODIA cohorts of newly diagnosed Stage 3 T1D patients (Discovery: $n=115$; Replication: $n=147$).

• Cohort Stratification: Patients were stratified by age at diagnosis into T1DE1 (<7y), T1DE2 (≥13y), and an intermediate group (T1DE-Int, 7–12y).
• miRNA Profiling:
  • Discovery: Small RNA-sequencing (RNA-seq) was performed on plasma samples collected within 6 weeks of diagnosis.
  • Validation: Candidate miRNAs were validated using Digital Droplet PCR (ddPCR) for absolute quantification (copies/µL).
  • Controls: To rule out age as a confounding factor, the study analyzed public datasets of non-T1D pediatric cohorts (healthy controls, celiac disease, and childhood asthma) and applied age-adjusted statistical models.
• Multi-Omic Integration:
  • Immunomics: High-parameter flow cytometry (Multi-FACS) was used on a subset of patients to assess peripheral blood B- and T-cell frequencies.
  • Transcriptomics: Whole-blood RNA-seq was analyzed to correlate circulating miRNA levels with gene expression, specifically looking for inverse correlations with predicted miRNA targets.
• Statistical Analysis: Differential expression analysis (DESeq2), ROC curve analysis for biomarker performance, and a rule-based "AND" classifier were used to re-stratify the entire population.

3. Key Contributions

1. Identification of a Specific miRNA Signature: The study identified miR-150-5p (immune-related) and miR-375-3p ($\beta$-cell related) as a reproducible signature distinguishing T1DE1 from T1DE2.
2. Decoupling Age from Biology: The authors demonstrated that while these miRNAs correlate with age in T1D, they do not correlate with age in non-T1D pediatric cohorts. Furthermore, miR-150-5p remained significantly associated with the T1DE1 endotype even after rigorous age adjustment, suggesting it reflects specific disease biology rather than just chronological age.
3. Mechanistic Insight: The study linked circulating miR-150-5p to immune activation (specifically B- and T-lymphocytes) and miR-375-3p to $\beta$-cell stress/destruction, providing a molecular readout of the "immune-$\beta$ cell axis."
4. Novel Stratification Tool: The authors developed a miRNA-based classifier (miRC1/miRC2) that re-stratifies T1D patients, including the intermediate age group, into two clinically distinct groups independent of their age at onset.

4. Key Results

Clinical Phenotypes

• T1DE1 patients exhibited significantly lower fasting C-peptide, higher first-year C-peptide decline (~54% vs ~17% in T1DE2), and distinct autoantibody patterns (higher IAA, lower GADA/ZnT8) compared to T1DE2.
• T1DE2 patients showed slower disease progression and better preserved $\beta$-cell function.

miRNA Findings

• Differential Expression: Both miR-150-5p and miR-375-3p were significantly upregulated in the plasma of T1DE1 patients compared to T1DE2 across both cohorts.
• Validation: ddPCR confirmed higher absolute copy numbers of both miRNAs in T1DE1.
• Age Independence: In non-T1D cohorts (healthy, celiac, asthma), neither miRNA correlated with age. In age-adjusted models for T1D, miR-150-5p retained strong significance ($P < 0.05$), while miR-375-3p's significance was attenuated in one cohort but remained robust in the other.
• Cellular Origin:
  • miR-375-3p: Confirmed as highly enriched in pancreatic $\beta$-cells, reflecting $\beta$-cell death/stress.
  • miR-150-5p: Highly enriched in B- and T-lymphocytes. Crucially, flow cytometry showed no difference in the absolute frequency of circulating B or T cells between endotypes. This suggests the elevated miR-150-5p is due to qualitative differences (e.g., activation state, increased secretion per cell) rather than cell abundance.
• Transcriptomic Correlation: Circulating miR-150-5p levels were inversely correlated with whole-blood expression of MPPE1 and RABGAP1L (predicted targets enriched in neutrophils, NK cells, and monocytes), suggesting a broader immune network reconfiguration.

Re-stratification (miRC1 vs. miRC2)

Using a logical "AND" rule (requiring both miRNAs to be above specific Youden-derived cutoffs), the authors created two new groups:

• miRC1 (High miRNA): Characterized by high miR-150-5p and miR-375-3p. These patients resembled the T1DE1 phenotype: lower C-peptide, faster decline, and higher autoantibody titers (GAD65, IA-2, ZnT8).
• miRC2 (Low miRNA): Characterized by lower miRNA levels, resembling the T1DE2 phenotype with better preserved $\beta$-cell function.
• Intermediate Group Resolution: Approximately 47% of the intermediate age group (7–12y) were re-assigned to miRC1, and 44% to miRC2, effectively resolving the "gray area" of T1D diagnosis based on biology rather than age.

5. Significance and Implications

• Precision Medicine: This study provides the first non-invasive, blood-based biomarkers capable of distinguishing T1D endotypes at diagnosis, moving beyond the limitations of age-based stratification.
• Clinical Trial Enrichment: The ability to identify patients with "aggressive" biology (miRC1/T1DE1) regardless of age allows for better enrichment of clinical trials for immunotherapies targeting specific immune pathways (e.g., B-cell depletion).
• Biological Insight: The findings support a model where T1DE1 is driven by a highly active, B-cell-rich immune response causing rapid $\beta$-cell destruction, detectable via the release of specific miRNAs.
• Future Directions: The authors propose including assays for miR-150-5p and miR-375-3p in future immunotherapy trials to personalize treatment strategies and monitor disease progression.

Limitations Noted: The study is cross-sectional (diagnosis only), limiting longitudinal predictions. T1DE1 is under-represented due to epidemiology, and functional validation of the miRNA-target interactions in purified cell subsets is required.