Type 1 vs. Type 2 Diabetes - Pathophysiology

SECTION 1

Historical Context & Motivation

The recognition that diabetes mellitus encompasses more than a single disease entity has been one of the most consequential advances in modern endocrinology. Ancient physicians in Egypt, India, and Greece described a wasting illness characterized by polyuria and sweet-tasting urine, yet it was not until the twentieth century that clinicians began to appreciate the profound pathophysiological differences between what we now call Type 1 diabetes (T1D) and Type 2 diabetes (T2D). Understanding this historical trajectory illuminates why the distinction matters for prognosis, treatment, and prevention strategies in contemporary clinical practice.

1552 BCE • Ebers Papyrus

One of the earliest known medical texts describes a condition of excessive urination, likely corresponding to diabetes mellitus, though no distinction between subtypes existed.

1921 • Discovery of Insulin

Frederick Banting and Charles Best isolated insulin from canine pancreatic extracts, transforming T1D from a uniformly fatal disease into a manageable chronic condition.

1936 • Two-Type Classification Proposed

Harold Himsworth distinguished "insulin-sensitive" from "insulin-insensitive" diabetes, laying the groundwork for the modern T1D versus T2D classification.

1979 • WHO/NDDG Formal Classification

The National Diabetes Data Group and the World Health Organization formally codified insulin-dependent diabetes mellitus (IDDM) and non-insulin-dependent diabetes mellitus (NIDDM) as distinct entities based on clinical phenotype and treatment requirements.

1997–Present • Etiology-Based Nomenclature

The ADA reclassified diabetes by pathogenic mechanism—autoimmune β-cell destruction (Type 1) versus progressive insulin resistance with relative secretory failure (Type 2)—replacing the older treatment-based terminology.

The shift from treatment-based labels to mechanism-based classification reflects a deeper question that this lesson addresses: how do two diseases that share the hallmark of chronic hyperglycemia arise from fundamentally different immunological, metabolic, and genetic disruptions? Answering this question is essential for healthcare professionals who must distinguish overlapping presentations, select appropriate interventions, and anticipate the distinct complications that each subtype carries.

SECTION 2

Core Principles & Definitions

At its foundation, diabetes mellitus is a syndrome defined by chronic hyperglycemia resulting from defects in insulin secretion, insulin action, or both. To compare the two major subtypes, clinicians rely on several core principles that span immunology, endocrinology, and metabolic physiology. The following foundational concepts underpin every diagnostic and therapeutic decision made in diabetology.

Absolute vs. Relative Insulin Deficiency

In T1D, autoimmune destruction of pancreatic β-cells eliminates endogenous insulin production entirely, constituting an absolute deficiency. In T2D, β-cells initially overproduce insulin to compensate for peripheral resistance, but this compensatory capacity eventually fails, creating a relative deficiency.

Autoimmunity vs. Metabolic Syndrome

T1D is fundamentally an autoimmune disease mediated by autoreactive T lymphocytes against β-cell antigens. T2D is strongly associated with metabolic syndrome—a cluster of visceral adiposity, dyslipidemia, hypertension, and insulin resistance driven by inflammatory and metabolic dysregulation.

Insulin Resistance

Insulin resistance is the diminished biological response of target tissues—skeletal muscle, liver, and adipose tissue—to a given concentration of circulating insulin. Although primarily a hallmark of T2D, some degree of resistance can emerge in T1D patients as well, complicating the classical dichotomy.

Genetic Susceptibility Profiles

T1D is strongly linked to HLA class II alleles (particularly HLA-DR3, HLA-DR4, and HLA-DQ8), which govern antigen presentation. T2D susceptibility involves dozens of polygenic loci (e.g., TCF7L2, PPARG, KCNJ11) that influence β-cell function and insulin signaling, and its heritability is paradoxically higher than T1D.

β-Cell Mass and Function Over Time

In T1D, β-cell mass declines precipitously—often over months to a few years—reaching near-total destruction. In T2D, β-cell function deteriorates gradually over decades, with glucotoxicity and lipotoxicity accelerating apoptosis and impairing secretory granule exocytosis.

✦

KEY TAKEAWAY

Think of the pancreatic β-cell as a factory and insulin as its product. In Type 1 diabetes, the factory is bombed by the immune system—production goes to zero, and you must import insulin from outside. In Type 2 diabetes, the factory is intact but its customers (muscle, liver, fat) have changed their locks (receptor dysfunction); the factory works overtime to produce more keys (insulin), but eventually its machinery wears out and output drops. Both scenarios result in insufficient insulin action, but the root cause—and therefore the repair strategy—is entirely different.

SECTION 3

Visual Explanation — Pathogenic Pathways

The following diagram contrasts the pathogenic cascades of Type 1 and Type 2 diabetes side by side, beginning with the precipitating factors and progressing through the cellular mechanisms that ultimately converge on chronic hyperglycemia. Note how the two pathways diverge at the level of the β-cell: one is destroyed extrinsically by immune attack, while the other fails intrinsically under metabolic stress.

The left pathway (blue) depicts the autoimmune cascade of T1D, from HLA-linked genetic susceptibility through T-cell–mediated insulitis to absolute insulin deficiency. The right pathway (violet) shows the metabolic progression of T2D, from polygenic risk amplified by lifestyle factors through insulin resistance to compensatory β-cell failure. Both converge on chronic hyperglycemia (center, red).

Several features of the diagram merit emphasis. First, notice that the environmental triggers differ categorically: T1D requires an immunological inciting event—often a viral infection such as coxsackievirus or enterovirus that may share molecular mimicry with β-cell antigens—whereas T2D is precipitated by metabolic overload from caloric excess and physical inactivity. Second, the temporal course is strikingly different; T1D frequently presents acutely (over weeks to months), whereas T2D develops insidiously over years, often remaining undiagnosed during a protracted period of impaired glucose tolerance or impaired fasting glucose collectively termed prediabetes.

SECTION 4

Mechanistic Deep Dive — Cellular and Molecular Pathways

Type 1 Diabetes: Autoimmune β-Cell Destruction

The immunopathogenesis of T1D begins with a breach of central and peripheral tolerance. In genetically susceptible individuals, autoreactive CD4⁺ and CD8⁺ T lymphocytes escape thymic deletion and recognize β-cell–specific antigens including glutamic acid decarboxylase 65 (GAD65), insulinoma-associated antigen 2 (IA-2), insulin, and zinc transporter 8 (ZnT8). The infiltration of islets by mononuclear cells—termed insulitis—leads to β-cell apoptosis via perforin-granzyme and Fas-FasL pathways, as well as cytokine-mediated killing (IL-1β, TNF-α, IFN-γ). Symptoms typically manifest when approximately 80–90% of functional β-cell mass is lost.

Type 2 Diabetes: Insulin Resistance and Progressive Secretory Failure

In T2D, the initiating defect is insulin resistance in skeletal muscle, hepatocytes, and adipocytes. At the molecular level, excess free fatty acids and pro-inflammatory cytokines (e.g., TNF-α, IL-6) secreted by hypertrophied visceral adipocytes activate serine/threonine kinases—including JNK and IKKβ—that phosphorylate insulin receptor substrate-1 (IRS-1) on inhibitory serine residues rather than activating tyrosine residues. This impairs downstream PI3K/Akt signaling, reducing GLUT4 translocation to the plasma membrane in muscle and diminishing glycogen synthesis. Hepatic insulin resistance promotes unrestrained gluconeogenesis, contributing to fasting hyperglycemia.

Concurrently, β-cells mount a compensatory response by increasing insulin output. Over time, sustained hyperglycemia (glucotoxicity) and elevated circulating lipids (lipotoxicity) generate reactive oxygen species, endoplasmic reticulum stress, and islet amyloid polypeptide (IAPP) aggregation within the islets. These insults collectively erode β-cell function and mass, transforming compensated insulin resistance into frank diabetes. The concept of a declining disposition index—the product of insulin sensitivity and β-cell function—captures this transition quantitatively.

DISPOSITION INDEX

DI = Insulin Sensitivity × β-Cell Function

The disposition index (DI) represents the hyperbolic relationship between insulin secretion and sensitivity. A normal glucose-tolerant individual maintains DI on the hyperbolic curve; individuals who fall off this curve—due to inadequate compensatory secretion—progress to T2D.

HOMA-IR (INSULIN RESISTANCE INDEX)

HOMA-IR = (Fasting Insulin [μU/mL] × Fasting Glucose [mmol/L]) ÷ 22.5

The Homeostatic Model Assessment of Insulin Resistance (HOMA-IR) is a widely used clinical surrogate for insulin resistance. Values > 2.5 are generally considered indicative of insulin resistance, though cut-offs vary by population and assay.

💡

Clinical Pearl

The presence of C-peptide in serum can help distinguish T1D from T2D in ambiguous cases. C-peptide is co-secreted with insulin in equimolar amounts and reflects endogenous insulin production. A very low or undetectable C-peptide level suggests T1D (β-cell destruction), while a normal or elevated level in the setting of hyperglycemia suggests T2D (insulin resistance with residual secretion).

SECTION 5

Detailed Comparison — Clinical, Immunological, and Metabolic Features

While the pathogenic mechanisms differ fundamentally, Type 1 and Type 2 diabetes can sometimes present with overlapping features—particularly in adults with latent autoimmune diabetes of adults (LADA) or in obese adolescents who may harbor both insulin resistance and autoimmunity. A systematic comparison of key features is therefore indispensable for accurate classification.

Feature	Type 1 Diabetes	Type 2 Diabetes
Typical age of onset	Childhood/adolescence (can occur at any age)	Adulthood (increasingly seen in adolescents)
Body habitus	Often lean or normal weight	Typically overweight or obese (BMI ≥ 25)
Onset tempo	Acute (days to weeks); may present with DKA	Insidious (months to years); often incidental finding
Autoantibodies	Present: anti-GAD65, anti-IA-2, anti-ZnT8, IAA	Absent (unless LADA)
C-peptide level	Low or undetectable	Normal or elevated early; declines late
Insulin requirement	Mandatory from diagnosis (exogenous insulin)	Not initially required; may need insulin later
Primary metabolic defect	Absolute insulin deficiency	Insulin resistance + relative deficiency
Ketoacidosis risk	High (common presenting feature)	Low (may occur with severe illness)
Acanthosis nigricans	Uncommon	Common (marker of hyperinsulinemia)
Associated conditions	Hashimoto thyroiditis, celiac disease, Addison disease	Hypertension, dyslipidemia, NAFLD, PCOS

Comparative features of Type 1 and Type 2 Diabetes Mellitus

This graph illustrates the temporal trajectory of β-cell function in each diabetes subtype. The blue curve (T1D) shows a precipitous decline, with clinical symptoms appearing once function drops below approximately 10–20%. The violet curve (T2D) illustrates a gradual decline over one to two decades, with diagnosis often occurring when function has already fallen to roughly 50%.

SECTION 6

Worked Example — Clinical Case Differentiation

The following clinical vignette demonstrates how to integrate history, physical examination findings, and laboratory data to distinguish between Type 1 and Type 2 diabetes in a patient presenting with new-onset hyperglycemia.

Case: New-Onset Hyperglycemia in a 14-Year-Old Male

1
Step 1 — Gather Clinical Information

A 14-year-old male presents to the emergency department with a 3-week history of polyuria, polydipsia, and unintentional weight loss of 5 kg. He has no significant past medical history and no family history of T2D, though his maternal grandmother has autoimmune thyroiditis. BMI is 19.2 kg/m² (50th percentile). He appears dehydrated with Kussmaul respirations.

Acute onset, lean habitus, weight loss, signs of acidosis
2
Step 2 — Review Laboratory Data

Blood glucose: 420 mg/dL. HbA1c: 12.3%. Arterial blood gas: pH 7.18, pCO₂ 18 mmHg, HCO₃⁻ 8 mEq/L. Serum ketones: strongly positive. Urinalysis: glucosuria and ketonuria. Fasting C-peptide: 0.15 ng/mL (reference: 0.8–3.1 ng/mL). Autoantibodies: anti-GAD65 positive, anti-IA-2 positive.

DKA confirmed (pH < 7.30, elevated ketones, low bicarb); C-peptide near zero; autoantibodies positive
3
Step 3 — Calculate HOMA-IR (for comparison)

Fasting insulin: 2.1 μU/mL. Fasting glucose: 23.3 mmol/L (420 mg/dL ÷ 18). HOMA-IR = (2.1 × 23.3) ÷ 22.5 = 48.93 ÷ 22.5 ≈ 2.17. Despite extreme hyperglycemia, the HOMA-IR is not markedly elevated because the numerator (insulin) is profoundly low—this pattern is consistent with absolute insulin deficiency rather than insulin resistance.

HOMA-IR ≈ 2.17 — misleadingly near normal due to negligible insulin secretion
4
Step 4 — Integrate Findings and Classify

The clinical picture is unambiguous: acute onset in a lean adolescent with DKA, undetectable C-peptide, and positive islet autoantibodies. The family history of autoimmune thyroiditis further supports an autoimmune diathesis. This constellation establishes the diagnosis of Type 1 diabetes mellitus.

Diagnosis: Type 1 Diabetes Mellitus — initiate insulin therapy immediately
5
Step 5 — Contrast with Hypothetical T2D Presentation

Had this patient been a 45-year-old obese male with a gradual onset of fatigue and blurred vision, an HbA1c of 8.5%, a fasting glucose of 210 mg/dL, a C-peptide of 4.8 ng/mL (elevated), negative autoantibodies, and acanthosis nigricans on physical exam, the diagnosis would instead be Type 2 diabetes. Initial management would focus on lifestyle modification and metformin rather than immediate insulin.

T2D contrast: older, obese, insidious onset, elevated C-peptide, no autoantibodies

SECTION 7

Complications — Shared and Divergent Sequelae

Both T1D and T2D share the common pathway of chronic hyperglycemia, which means they share microvascular and macrovascular complications. However, the relative incidence, timing, and co-morbidity profiles differ significantly between the two subtypes, reflecting their distinct metabolic milieus.

Complication Category	Type 1 Diabetes	Type 2 Diabetes
Diabetic Ketoacidosis (DKA)	Common; may be the presenting feature. Results from absolute insulin deficiency and unopposed lipolysis → ketogenesis.	Rare; may occur during severe physiological stress. More prone to hyperosmolar hyperglycemic state (HHS).
Retinopathy	Develops after 5–10 years of disease; nearly universal after 20 years.	May be present at diagnosis due to years of unrecognized hyperglycemia.
Nephropathy	Microalbuminuria typically appears 5–15 years post-diagnosis.	Leading cause of end-stage renal disease; often co-exists with hypertensive nephrosclerosis.
Cardiovascular Disease	Elevated risk (2–4×); accelerated atherosclerosis even in young adults.	Markedly elevated risk; CVD is the leading cause of death. Often co-occurs with dyslipidemia, hypertension.
Neuropathy	Primarily distal symmetric polyneuropathy; autonomic neuropathy also common.	Similar distribution; may be present at diagnosis. Contributes to diabetic foot ulcers.
Other Autoimmune Conditions	High co-occurrence: thyroiditis, celiac disease, pernicious anemia, Addison disease.	Not typically associated with autoimmune conditions; associated instead with NAFLD, OSA, PCOS.

Complications comparison between T1D and T2D

✦

KEY TAKEAWAY

Consider two rivers that both flood a town: one bursts its dam suddenly (T1D — acute insulin loss), while the other rises slowly because its drainage channels are clogged (T2D — insulin resistance). The flood damage (hyperglycemia-driven complications) is similar in both cases—water-logged buildings, eroded foundations—but the upstream causes demand different engineering solutions. Rebuilding the dam requires a completely different strategy from unclogging the drains, just as exogenous insulin replacement differs from addressing insulin resistance through lifestyle and pharmacotherapy.

SECTION 8

Connection to Advanced Theory — Emerging Subtypes and Precision Diabetology

The binary T1D/T2D classification, while clinically useful, is increasingly recognized as an oversimplification. Emerging research in precision diabetology seeks to reclassify diabetes into more granular subtypes based on molecular, immunological, and phenotypic data, with the goal of tailoring prevention and treatment strategies to individual patients.

Concept	Classical Framework	Advanced / Emerging Framework
Classification system	Binary: T1D (autoimmune) vs. T2D (metabolic)	Ahlqvist cluster model (2018): 5 subtypes—SAID, SIDD, SIRD, MOD, MARD—based on autoantibodies, age, BMI, HbA1c, HOMA-B, HOMA-IR
LADA recognition	Often misclassified as T2D in adults > 30 years	Recognized as autoimmune (GAD+) but slowly progressive; may represent a continuum with T1D. Classified as SAID in cluster models.
Monogenic diabetes	Often unrecognized; lumped into T1D or T2D	MODY (HNF1A, GCK, HNF4A mutations) and neonatal diabetes now identified via genetic testing, with targeted therapy (e.g., sulfonylureas for KCNJ11 mutations)
Immunotherapy for T1D	No disease-modifying therapy; insulin replacement only	Teplizumab (anti-CD3 monoclonal antibody) FDA-approved 2022 to delay Stage 3 T1D onset by ≈ 2 years in high-risk individuals
β-Cell regeneration	Not achievable	Stem cell–derived islet transplantation (e.g., VX-880, VX-264) and CRISPR-based approaches are in clinical trials for T1D

Classical vs. emerging frameworks in diabetes classification and therapy

As you advance in your clinical training, you will encounter patients who defy the clean T1D/T2D dichotomy—an obese teenager with autoantibodies, a lean adult with severe insulin resistance due to a lipodystrophy, or a child with monogenic diabetes misdiagnosed as T1D. The ability to think beyond the binary framework, leveraging autoantibody panels, C-peptide kinetics, and eventually genomic data, will be essential for delivering precision care. The T1D/T2D model remains the indispensable foundation, but it is a starting point, not the final answer.

SECTION 9

Practice Problems

Question 1: CONCEPTUAL

Explain why Type 1 diabetes patients are prone to diabetic ketoacidosis (DKA) while Type 2 diabetes patients more commonly develop hyperosmolar hyperglycemic state (HHS). In your answer, relate the pathophysiology to the presence or absence of endogenous insulin.

Show answer

In T1D, the absolute absence of insulin removes the hormonal brake on lipolysis. Unopposed glucagon and catecholamines drive massive lipolysis, releasing free fatty acids that undergo hepatic β-oxidation to produce ketone bodies (acetoacetate, β-hydroxybutyrate), causing metabolic acidosis. In T2D, residual insulin secretion—even if insufficient to prevent hyperglycemia—is typically enough to suppress lipolysis and ketogenesis. Consequently, T2D patients who decompensate tend to develop severe hyperglycemia with profound dehydration and hyperosmolarity (HHS) rather than ketoacidosis. The key distinction is that even small amounts of circulating insulin can suppress the lipolytic cascade, which is why DKA is the hallmark emergency of T1D but rare in T2D.

Question 2: BASIC CALCULATION

A 52-year-old woman with suspected T2D has a fasting insulin level of 18 μU/mL and a fasting glucose of 7.2 mmol/L. Calculate her HOMA-IR using the formula: HOMA-IR = (Fasting Insulin × Fasting Glucose) ÷ 22.5. Interpret the result.

Show answer

HOMA-IR = (18 × 7.2) ÷ 22.5 = 129.6 ÷ 22.5 = 5.76. A HOMA-IR of 5.76 is well above the commonly cited threshold of 2.5, indicating significant insulin resistance. This is consistent with Type 2 diabetes, in which peripheral tissues require supraphysiological insulin concentrations to maintain glucose homeostasis. Note that the patient's fasting insulin is elevated (not low), which argues against T1D and supports the presence of compensatory hyperinsulinemia characteristic of T2D.

Question 3: INTERMEDIATE

A 28-year-old woman presents with a random glucose of 310 mg/dL and an HbA1c of 10.1%. She has a BMI of 32 kg/m², family history of T2D in both parents, and acanthosis nigricans. Her physician initially diagnoses T2D and starts metformin. Six months later, she develops DKA. Autoantibody testing reveals positive anti-GAD65. What is the most likely diagnosis, and how would you reconcile her initial presentation with the subsequent course?

Show answer

The most likely diagnosis is Latent Autoimmune Diabetes of Adults (LADA), which is often classified under the T1D spectrum despite its adult onset and initially indolent course. The patient's obesity and acanthosis nigricans initially mimicked T2D, but the development of DKA and positive anti-GAD65 autoantibodies reveal an underlying autoimmune process. LADA represents a slowly progressive form of autoimmune β-cell destruction; patients may retain enough endogenous insulin to avoid ketosis for months to years before β-cell reserve becomes critically depleted. This case underscores the importance of measuring autoantibodies (especially anti-GAD65) in any patient with features atypical for T2D—such as younger age, poor response to oral agents, or eventual DKA—because the correct diagnosis mandates insulin therapy and surveillance for other autoimmune conditions.

Question 4: APPLIED

You are designing a community screening program for diabetes in a population with high T2D prevalence. You have the option of using fasting plasma glucose (FPG), oral glucose tolerance test (OGTT), or HbA1c. Discuss the advantages and limitations of each screening modality in the context of early detection of T2D, and explain why these screening tools are less useful for predicting T1D.

Show answer

FPG is inexpensive and convenient but misses individuals with isolated postprandial hyperglycemia (impaired glucose tolerance without impaired fasting glucose). The OGTT captures both fasting and post-challenge dysglycemia and is the gold standard for identifying impaired glucose tolerance—a prediabetic state—but requires a 2-hour clinic visit and has poor reproducibility. HbA1c reflects average glycemia over 2–3 months, does not require fasting, and is highly standardized, but can be inaccurate in patients with hemoglobinopathies, anemia, or recent transfusions, and its diagnostic thresholds vary across ethnic groups. For T2D community screening, HbA1c is often preferred for its convenience, with FPG or OGTT used for confirmation. These tools are less useful for predicting T1D because T1D typically presents acutely after a rapid decline in β-cell function—patients may have normal glucose tolerance until weeks before diagnosis. Instead, T1D prediction relies on autoantibody panels (GAD65, IA-2, ZnT8, IAA) and, increasingly, on staging models (Stages 1–3) that detect autoimmunity and dysglycemia before clinical onset.

Question 5: CRITICAL THINKING

The Ahlqvist (2018) cluster analysis proposed reclassifying adult-onset diabetes into five subtypes: Severe Autoimmune Diabetes (SAID), Severe Insulin-Deficient Diabetes (SIDD), Severe Insulin-Resistant Diabetes (SIRD), Mild Obesity-Related Diabetes (MOD), and Mild Age-Related Diabetes (MARD). Critically evaluate how this model might improve clinical outcomes compared to the traditional T1D/T2D binary, and identify at least two potential barriers to implementing this classification in routine clinical practice.

Show answer

The Ahlqvist cluster model refines the heterogeneous T2D category into subgroups with distinct complication trajectories: SIRD patients carry the highest risk of nephropathy and fatty liver disease, SIDD patients have the greatest risk of retinopathy, and MOD and MARD patients have more benign courses. By identifying these risk profiles early, clinicians could prioritize nephroprotection (e.g., SGLT2 inhibitors or GLP-1 agonists) in SIRD and intensify glycemic control in SIDD, thereby personalizing prevention strategies beyond one-size-fits-all T2D guidelines. However, implementation faces significant barriers. First, the model requires six input variables including HOMA-B and HOMA-IR, which necessitate fasting insulin assays that are not routinely performed in primary care and lack international standardization—assay variability could lead to misclassification. Second, clusters were derived from Scandinavian cohorts, and their generalizability to ethnically diverse populations is unproven; studies in South Asian and African-origin populations have yielded inconsistent cluster distributions. Third, cluster assignment is based on values at diagnosis and may shift over time as the disease progresses, raising questions about dynamic reclassification. Despite these challenges, the model represents a meaningful step toward precision diabetology and will likely evolve as multi-omics data and longitudinal cohort studies refine the cluster boundaries.

SUMMARY

Lesson Summary

Summary

Type 1 diabetes is an autoimmune disease driven by HLA-linked genetic susceptibility and environmental triggers, culminating in T-cell–mediated β-cell destruction and absolute insulin deficiency. It presents acutely—often with diabetic ketoacidosis—typically in children and young adults, and requires lifelong exogenous insulin. Diagnosis is supported by positive islet autoantibodies (GAD65, IA-2, ZnT8) and low or undetectable C-peptide. Associated autoimmune conditions include thyroiditis and celiac disease.

Type 2 diabetes arises from peripheral insulin resistance—particularly in muscle, liver, and adipose tissue—compounded by progressive β-cell secretory failure driven by glucotoxicity and lipotoxicity. It develops insidiously over years, is strongly associated with metabolic syndrome and obesity, and is managed initially with lifestyle modification and oral agents. The disposition index and HOMA-IR quantify the interplay between sensitivity and secretion. Both types converge on chronic hyperglycemia and share microvascular and macrovascular complications, but their divergent pathogeneses demand fundamentally different therapeutic strategies. Emerging frameworks such as the Ahlqvist cluster model and immunotherapies like teplizumab herald a new era of precision diabetology.