Definition

Diabetes mellitus is a group of disorders of carbohydrate metabolism in which glucose is both underutilized (due to reduced glycogenolysis) and overproduced (due to increased gluconeogenesis). This imbalance causes elevated blood sugar levels.

Diagnosis can be made through elevated plasma glucose or increased HbA1c. In 1998, diabetes mellitus was classified into four types:

  1. Type 1 Diabetes Mellitus – An autoimmune disease leading to destruction of pancreatic beta cells, resulting in absolute insulin deficiency.
  2. Type 2 Diabetes Mellitus – A combination of insulin resistance and relative insulin deficiency with secretion defects.
    • Type 2A: without obesity
    • Type 2B: with obesity
  3. Type 3 Diabetes Mellitus – Other specific types:
    • Genetic defects of beta cells (e.g., MODY diabetes)
    • Genetic defects in insulin action
    • Diseases of the pancreas and islets of Langerhans (pancreoprivic diabetes)
    • Drug- or chemical-induced diabetes
    • Infection-induced diabetes
    • Unusual forms of immune-mediated diabetes
  4. Type 4 Diabetes Mellitus – Gestational diabetes

Diabetes Type 1, Type 2, Type 3 … Is that enough?

Type 2 Diabetes: A very heterogeneous group

In addition to very specific forms of diabetes (Type 3 diabetes), newly diagnosed diabetes mellitus is usually classified as either Type 1 or Type 2.
People with Type 1 diabetes are typically treated with insulin right away (since Type 1 diabetes is defined by an absolute lack of insulin).

In people with Type 2 diabetes, however, the initial therapy and response to treatment vary widely — often leading to a process of “trial and error.”
The chosen therapy depends heavily on the treating physician, which can result in under- or overtreatment. The group of people with Type 2 diabetes is therefore very heterogeneous.

Experienced clinicians consider factors such as age, abdominal fat, and the dynamics of disease progression; however, a systematic and comparable classification has not yet been established.

Five subgroups in newly diagnosed Type 2 diabetes: A fairly stable pattern

A Swedish research group proposed a new classification for Type 2 diabetes in 2018 [1]. This proposal was based on a study of nearly 9,000 individuals with newly diagnosed Type 2 diabetes. They identified six relevant variables on which to base their classification:

  • GAD antibodies, which are typical for Type 1 diabetes
  • Age at diagnosis
  • Body mass index (BMI)
  • HbA1c (long-term blood sugar value)
  • HOMA, a calculation used to assess:
    a) Insulin resistance/sensitivity
    b) Insulin secretion/endogenous insulin production

Using these six variables, the researchers identified five subgroups among the 9,000 participants:

  1. Group 1 (7%): Severe autoimmune diabetes – most closely resembling classic Type 1 diabetes. The affected individuals were young, relatively lean, difficult to manage, and had GAD antibodies in lab tests.
  2. Group 2 (18%): Severe insulin-deficient diabetes. These individuals were similar to Group 1 — young, relatively lean, difficult to manage — but without GAD antibodies and with low endogenous insulin production.
  3. Group 3 (15%): Severe insulin-resistant diabetes. These individuals were relatively overweight and showed marked insulin resistance in lab tests.
  4. Group 4 (22%): Mild obesity-related diabetes. The affected individuals were overweight but not insulin-resistant, with only mild metabolic disturbances.
  5. Group 5 (39%): Mild age-related diabetes. These individuals had mild metabolic disturbances similar to Group 4 but were older.

Interestingly, the percentage distribution among other populations with newly diagnosed diabetes mellitus was very similar. This suggests that the classification may be a useful model for structuring the highly heterogeneous group of people with Type 2 diabetes.

Classification into five “clusters”: Practical implications?

Long-term follow-up data show that these groups differ in required therapy as well as in risk and distribution of complications.
Thus, these “clusters” may help support the goal of personalized therapy (precision medicine) in modern diabetes care.

For example:

  • Group 2 (severe insulin-deficient diabetes) clearly benefits from early insulin therapy.
  • Groups 1 and 2 are most likely to develop retinopathy,
  • Group 3 is prone to nephropathy and fatty liver,
  • Groups 4 and 5 generally show mild disease progression with fewer complications — where overtreatment should be avoided and therapy may be de-escalated if appropriate.

Conclusion

  • The reported subgroups (“clusters”) reveal a clear pattern within the heterogeneous population of people newly diagnosed with Type 2 diabetes.
  • The subgroups show distinct characteristics regarding treatment needs and the development of complications.
  • This may lead to a new classification of Type 2 diabetes that could support individualized therapy planning (precision medicine).
    Further studies and validation in other patient populations will be needed to confirm these models.

Reference:
[1] Ahlqvist E, Storm P, Käräjämäki A, et al. Novel subgroups of adult-onset diabetes and their association with outcomes: a data-driven cluster analysis of six variables. Lancet Diabetes Endocrinol. 2018 May;6(5):361–

  1. Swedish 5-Cluster Classification

Cluster

Key Features

Autoantibodies

BMI

C-peptide

Insulin Resistance

Therapy

SAID (Severe Autoimmune Diabetes)

Type 1-like, young, early insulin need

+

Low

Low

Low

Insulin

SIDD (Severe Insulin-Deficient Diabetes)

Young, no antibodies, severe insulin deficiency

–

Low

Low

Low

Insulin ± GLP-1/SGLT2

SIRD (Severe Insulin-Resistant Diabetes)

High BMI, preserved β-cells, high risk kidney/NAFLD

–

High

Normal/high

High

Metformin, GLP-1, SGLT2

MOD (Mild Obesity-related Diabetes)

Young, overweight, mild progression

–

High

Normal

Moderate

Lifestyle ± oral meds

MARD (Mild Age-Related Diabetes)

Older age, mild course

–

Normal

Normal

Low

Lifestyle ± Metformin

LADA (Latent Autoimmune Diabetes in Adults)

Belongs to SAID group. Autoantibody-positive, often misdiagnosed as type 2. Insulin required later than in type 1.

Typical Features:

  • Age at diagnosis >30 years
  • Positive autoantibodies (especially GAD)
  • Initially no insulin therapy needed, later required
  • Slower progression than classic type 1
  • C-peptide initially normal, later declines

Feature

LADA

Age at diagnosis

>30 years

Autoantibodies

GAD+ (main marker)

Insulin need

Delayed (months–years)

Progression

Slower than type 1

C-peptide

Initially normal → declines

Therapy

Initially oral meds (Metformin, DPP-4i), insulin later; avoid sulfonylureas

Diabetes Mellitus Historical Notes

Diabetes mellitus has been known as a disease for over 3,500 years. In the second century AD, a Greek physician used the term Diabetes (Greek for “to flow through”) to describe the symptom of frequent urination associated with the disease. In India, the first mention appeared around 600 BC in Ayurvedic texts. The condition was called “honey-sweet urine” or “honey urine” because it was observed that the urine had a sweet taste and attracted ants. In China, diabetes was mentioned around 200 BC, where it was called “wasting thirst disease” or “consuming thirst,” with the main symptoms—intense thirst, frequent urination, and weight loss—clearly described.

In the 17th century, Thomas Willis recognized that the urine of diabetics tasted sweet, and in the 19th century it was understood that diabetes is related to a disturbance in sugar metabolism.

Historical Notes

Time / Region

Observation

~600 BC (India)

First mention in Ayurvedic texts; called honey-sweet urine (sweet taste, attracted ants).

~200 BC (China)

“Wasting thirst disease” – intense thirst, frequent urination, weight loss described.

2nd century AD (Greece)

Term Diabetes (“to flow through”) coined for frequent urination.

17th century (Thomas Willis)

Urine of diabetics noted to taste sweet.

19th century

Recognized as disorder of sugar metabolism.

  1. Epidemiology

Region

Diabetes Cases (millions)

% of Type 1

Worldwide

828

~10%

USA

38.4

~5.7%

Europe (EU)

31.6

~5–10%

India

212

~5–10%

China

140

<5%

Latin America

35

~2.3%

  • Global prevalence: ~14% of adults
  • Type 1 proportion worldwide: ~80 million adults (~10% of cases)
  1. Autoantibodies in Diabetes

Antibody

Target

Relevance

Typical Use

GAD-AK

Glutamate decarboxylase (GAD65)

Most frequent in LADA; sensitive in adults

LADA, Type 1

IA-2-AK

Tyrosine phosphatase IA-2

More common in children; faster β-cell loss

Type 1

ZnT8-AK

Zinc transporter 8

Improves sensitivity; esp. in GAD-/IA-2–negative

Type 1

IAA

Insulin

Only useful pre-insulin therapy; young children

Early type 1

ICA

Islet cell antibodies

Historic, less specific

Rarely used

  1. Associated Autoimmune Diseases

Disease

Notes / Screening

Autoimmune thyroiditis (Hashimoto, Graves)

Most common (15–30%); TSH, fT4, TPO-AK

Celiac disease

4–8%; tTG-IgA, total IgA

Addison’s disease

Rare; cortisol, ACTH, 21-OH-AK

Vitiligo

Depigmentation

Pernicious anemia

B12 deficiency, intrinsic factor AK

Autoimmune hepatitis

Liver enzymes, ANA, SMA

Rheumatoid arthritis / SLE

Less common

Myasthenia gravis

Very rare

  1. Polyendocrine Autoimmune Syndromes (APS)

Type

Name / Common Features

Genetics

APS-1

Addison + hypoparathyroidism + candidiasis

Childhood; AIRE mutation

APS-2

Addison + thyroiditis ± type 1 DM

Adolescence/adults; HLA-linked

APS-3

Thyroiditis + other autoimmune (no adrenal)

Common; no adrenal involvement

  1. Carpenter Syndrome (Not autoimmune)
  • Genetic (RAB23 mutation)
  • Features: craniosynostosis, polydactyly, intellectual disability, possible heart/genital defects
  • No autoimmune/endocrine link
  1. Diagnostic Criteria (non-pregnant)

Test

Diabetes

Prediabetes

HbA1c

≥6.5% (≥48 mmol/mol)

5.7–6.4% (39–47 mmol/mol)

Fasting plasma glucose

≥126 mg/dL (≥7.0 mmol/L)

100–125 mg/dL (5.6–6.9 mmol/L)

2h OGTT

≥200 mg/dL (≥11.1 mmol/L)

140–199 mg/dL (7.8–11.0 mmol/L)

Random glucose + symptoms

≥200 mg/dL (≥11.1 mmol/L)

–

Diagnosis should be confirmed by a second method unless classic symptoms are present.