Continuing Education Activity

Latent autoimmune diabetes of adults (LADA) is autoimmune diabetes that begins in adulthood and does not need insulin for glycemic control at least in the first six months after diagnosis. It shares genetic, immunologic, and metabolic features with both type 1 and type 2 diabetes mellitus (DM). This activity reviews the management of latent autoimmune diabetes and highlights the role of the interprofessional team in evaluating and treating this condition.

Objectives:

• Outline the epidemiology of latent autoimmune diabetes in adults (LADA).
• Describe the advances in the pathophysiology of latent autoimmune diabetes in adults.
• Review the evaluation and management of latent autoimmune diabetes of adults.
• Explain the importance of improving care coordination amongst the interprofessional team members to improve outcomes for adult patients with latent autoimmune diabetes.

Introduction

Diabetes mellitus (DM) is a disease spectrum ranging from classic insulinopenic type 1 diabetes (T1DM) at one end to classic insulin-resistant type 2 diabetes (T2DM) at the other. Latent autoimmune diabetes of adults (LADA) is a form of DM with features of both T1DM and T2DM and has therefore been termed Type 1.5 DM. In Japan, the synonym used is slowly progressive insulin-dependent type 1 diabetes mellitus (SPIDDM). The American Diabetes Association (ADA) lists LADA as T1DM that evolves more slowly than the classic disease and does not recognize it as a specific type of DM. The World Health Organization's term for LADA is 'slowly evolving immune-related diabetes.'

LADA is, by definition, a disease of adults. The Immunology for Diabetes Society (IDS) has specified three criteria for the diagnosis of LADA:

1.  Age greater than 35 years
2.  Positive autoantibodies to islet beta cells
3.  Insulin independence for at least the initial 6 months after initial diagnosis

Although attractive, this set of criteria has been challenged, especially because the choice of insulin as a treatment is highly physician dependent. It is immunologically similar to T1DM as antibodies to islet beta cells are present, albeit at lower titers, and immune destruction progresses at a much slower rate when compared to classic T1DM. The majority of these patients present with hyperglycemia that is not dramatic like T1DM and is misdiagnosed and managed as T2DM. Only later is it realized that they have poor control with many conventional agents, especially sulfonylureas, and eventually require insulin therapy.

LADA itself is a heterogeneous disease where some patients have high antibody titers, a low BMI, and progress to insulin therapy faster, and others who have low antibody titers, features of insulin resistance like higher BMI, and progress more slowly to requiring insulin.

Early recognition of LADA is paramount so that appropriate strategies are employed to delay beta-cell destruction and reduce complications. This article reviews the advances in the pathophysiology of LADA and its implications in the evaluation and treatment.

Etiology

LADA is determined by genetic factors. As in T1DM, the risk for acquiring LADA is highest in carriers of certain HLA haplotypes. The HLA genes code for the major histocompatibility antigens (MHC) which have important immunoregulatory functions, and therefore it is not surprising that LADA is caused by dysregulated immunity. However, the factors that precipitate autoimmunity have not been established.

Unlike T2DM, there is a paucity of studies investigating the role of environmental factors like lifestyle in LADA. The reasons for this are:

• The need for autoantibody measurement in every newly diagnosed adult diabetes patient to identify and classify patients as LADA
• The absence of comparable cohort groups
• Unavailability of lifestyle information antecedent to the diagnosis of LADA 
• The inadequate sample size of LADA patients in the studies enhances the power of the study.[1]

In the few studies in this regard, it is evident that LADA shares several lifestyle risk factors with T2DM, namely, excess body weight, greater waist-hip ratio, low birth weight, intake of two or more sweetened beverages daily, and heavy smoking. These risk factors are of greater significance in the subset of LADA with lower titers of autoantibodies and higher BMI. Although weaker, they hold even for those with higher antibody titers. Increased physical activity, moderate alcohol use, and the intake of fatty fish have a protective effect on the risk of LADA.[2] Two or more cups of coffee daily increase the risk for LADA, unlike the beneficial effect seen in T2DM, but this observation is sourced from a single study and needs to be validated.

Epidemiology

LADA is the most frequent form of adult-onset autoimmune DM. It is also the most prevalent form of autoimmune diabetes as a whole.[3] There are geographic and ethnic differences in its incidence. In the multicentric 'Action LADA' study from Europe, almost 10 percent of 6000 adults with adult-onset DM had islet cell autoantibodies. In the United Kingdom Prospective Diabetes Study (UKPDS), the antibody positivity among those with a presumptive diagnosis of T2DM in adults was 12 percent. Similarly, studies from Norway showed a 10 percent incidence, whereas studies from the middle-east, Korea, and China showed between 4 and 6 percent. Even within individual studies involving multiple centers, there were variations in the incidence between northern and southern latitudes both in Europe and Asia.

Most patients with LADA are positive for a single islet autoantibody, and glutamic acid decarboxylase antibody (GADA) is the most predominant. Some population groups have a varying prevalence of different autoantibodies, and measuring just one may underestimate the prevalence of LADA. Autoantibodies have been observed to appear and disappear during longitudinal follow-up. In these situations, the role of assay interference from anti-idiotype antibodies should be considered.[4]

A form of latent autoimmune diabetes in the young (LADY) has been described.[5][6]

Pathophysiology

Islet beta cell autoimmunity antedates the onset of LADA by several years. This was observed in nearly two-thirds of patients with LADA in a prospective study and substantiated autoimmunity as the first insult. This is followed by insulin resistance that causes overt hyperglycemia and the diagnosis of autoimmune insulin-independent DM. An assessment of insulin resistance (IR) using HOMA-IR has shown patients with LADA have IR similar to T2DM even after correction for BMI. Thus, the pathophysiology of LADA involves both autoimmunity and the metabolic derangements of IR.[7]

Most patients with T2DM and some with LADA and T1DM present with features of metabolic syndrome (MetS). Irrespective of glucose as a criterion, MetS was more frequent in T2DM than in LADA. When glucose levels were included, MetS was higher in LADA than in T1DM.[8]

Pancreatic pathology:

Pancreatic tissue from humans with LADA and from a rat model mimicking LADA were analyzed using immunohistochemistry and PCR. Predominant macrophage (CD68) infiltration was shown in the islet cells as opposed to T1DM, which showed CD8T lymphocytes. There were islet groups with and without infiltration. The cell type shift translated to greater expression of Interleukin 1 beta cytokine secretion and decreased expression of TNF alfa in T lymphocytes. Also, the proliferation marker of nuclear antigen (PCNA) and the anti-inflammatory cytokine interleukin 10 (IL10) were elevated while the apoptotic promoters caspase3 and TUNEL were diminished. The result was increased beta-cell gene transcription, greater C-peptide levels, slower progression of beta-cell destruction, and slower onset of LADA compared to T1DM.[9]

The presence of more than one diabetes-associated autoantibody (DAA) predicts a faster progression of beta-cell failure. In patients with LADA, the sensitivity of GADA in predicting insulin requirement diminishes in older patients.

In a Chinese study, it was shown that GADA alone was insufficient to identify all cases of LADA, although it was the most dominant. This highlights the geographic and ethnic variability in the distribution of DAAs in LADA. 

Anti-gliadin antibodies (both IgA and IgG) are increased in LADA compared to T2DM and compromised intestinal mucosal barrier may allow environmental antigens to gain access via the oral route and initiate immunological events to induce the disease. Furthermore, similar to T1DM, anti-TPO antibodies are increased in LADA when compared to T2DM.[10][11]

LADA subtypes differ with GADA titers that are inversely related to BMI.[12]

Regulatory T lymphocytes produce proteins like transcription factor forkhead box protein 3 (FOXP3), which suppresses autoimmunity. Hypermethylation of DNA and decreased expression of FOXP3 diminish protection against immune destruction. It has been shown that FOXP3 and other proteins are lower in high GADA subtype compared to low GADA, indicating higher beta-cell destruction in the former.[13][14]

The IgG4 class of GADA is more prevalent in LADA than T1DM, and this yields a T helper 2 lymphocyte (TH2) immune response, which may be one explanation for delayed onset of diabetes in LADA compared to T1DM. Another reason may be the binding site on the GAD molecule for the GADA. In LADA, there is more amino-terminal binding, whereas, in T1DM, it is in the carboxy-terminal. In the UKPDS, some patients who were GADA positive even after 5 years did not progress to insulin requirement, and the epitope specificity may explain this. Also, a high-affinity subtype of GADA to GAD65 antigen predicts the rate of beta-cell failure.[15]

A variant of the IA2A antibody (256-760) is more frequently found in LADA and may prove useful in its detection.[16][17] 

In the UKPDS and subsequent studies, a subset of T2DM negative for islet autoantibodies had T lymphocytes that were immunoreactive to islet cell antigens. The autoimmunity in these patients was more severe than in those with autoantibodies alone. They demonstrated a lower stimulated C-peptide response and subsequently went on to require insulin earlier than the group of T2DM without both antibodies and reactive T cells, thus creating a separate category of T-LADA.[18][19]

The inflammatory biomarkers which are associated with obesity vary when sera from various types of DM are analyzed. If these results are reproduced in larger populations, their utility to diagnose and differentiate LADA from T1DM and T2 DM in conjunction with antibody testing may be enhanced.[20]

LADA shares immunological features with both T1DM and T2DM. In a recent study analyzing fresh-blood-derived peripheral blood mononuclear cells, it was shown that LADA was similar to T2DM in antigen-presenting cell characteristics and the number of regulatory B lymphocytes, whereas it mimicked T1DM in the number of Natural killer cells.[21]

In patients with LADA, there is a positive correlation between BMI and interleukin 17, a pro-inflammatory cytokine from peripheral B lymphocytes implicated in T2DM and obesity.[22]

LADA has more genetic similarity to T1DM than T2DM. Studies have shown certain HLA types increase while others decrease the risk for LADA.[23] The number of 'risk alleles' is lesser in LADA, explaining the later onset of the disease. Genetic loci that are common in T1DM are associated with LADA. These include the major histocompatibility complex (MHC) region, PTPN22, SH2B3, and INS. In a recently published study, the key T2D risk allele TCF7L2 had a lower occurrence in LADA cases; it appears that this locus may not play a role in the etiology of LADA.[24]

The only genetic similarity with T2DM at the HNF1A locus was not reproduced in a subsequent study by the same investigators who instead found a strong signal at the PFKFB3 locus during a landmark Genome-wide association study (GWAS).[25]

Recent data do not support the independent effects of class I MHC in LADA, unlike in T1DM.[26]

Modulating immunoreactivity of T cells may be a viable therapeutic target in LADA, just like T1DM.

Serious life events did not seem to precipitate LADA, unlike childhood T1DM.[27]

History and Physical

Patients with LADA may present with symptoms of polyuria, polydipsia, nocturia, fatigue, visual changes, tingling in the feet, and weight loss or may be asymptomatic.

A personal or family history of autoimmune disease should raise suspicion that a patient with hyperglycemia has LADA or T1DM.[28]

A history of low-birth-weight is a strong risk factor for both LADA and T2DM.[29] 

A history of smoking, alcohol consumption, the number of sweetened beverages are risk factors.

The amount and type of physical activity and exercise should be sought and recorded to quantitate the risk of LADA.[30]

The blood pressure must be accurately measured.

Several diagnostic tools have been proposed to identify patients with LADA.

A diagnostic screening tool with three criteria was used to identify LADA in diabetic patients older than 50 years of age:

1. A low or normal BMI.
2. A fasting blood glucose 270 mg/dl or higher, HbA1C 10% or greater despite good compliance.
3. Loss of weight despite a diet constant in calorie content.

This tool is said to detect three-fourths of patients with LADA.[31] 

Another investigator developed a clinical risk score:

• Age less than 50 years.
• Symptomatic hyperglycemia.
• BMI of less than 25 kg/m2.
• A personal or family history of autoimmune disease

Two or more criteria when positive yielded a sensitivity of 90 percent and specificity of 71 percent while less than 2 criteria virtually excluded LADA.[32] 

Although the above algorithms have been described and validated in some studies, it must be mentioned that the phenotypes of LADA may resemble T1DM or T2DM and therefore exhibit a normal or high BMI, respectively, with the latter having higher blood pressure and triglycerides albeit midway between T1DM and T2DM.[33]

Some patients with LADA may have features of the MetS. Two phenotypes have been described based on GADA titers. LADA1 with ICA and high GADA titers and a T1DM phenotype and LADA2 with lower GADA titers, single antibody, and a T2DM phenotype.[34]

Patients with LADA are usually ketosis resistant at first diagnosis.

Evaluation

When an adult patient presents with hyperglycemia with or without symptoms and is controlled without the need for insulin in the first six months, LADA should be considered. A positive antibody to one of the islet antigens is the hallmark of LADA. Worldwide, the most prevalent islet autoantibody utilized is GADA. Others include IA-2A, which is associated increasingly with the LADA2 phenotype, insulin antibodies, and zinc transporter isoform 8 (ZnT8) antibody that occurs with varying frequencies. LADA has more of the other DAAs apart from GADA and IA2A than T1DM. 

Patients with LADA have residual C-peptide levels typically between those with T1DM and T2DM. In T1DM, C-peptide is absent at first clinical presentation, and in T2DM, it is often increased. The levels of C-peptide correlate inversely with GADA titers. A stimulated C-peptide has a greater predictive value than a fasting level. Among the methods studied, the glucagon stimulation test (GST) and Mixed Meal Tolerance Test (MMTT) have been validated and found to be useful. The former is shorter in duration but associated with transient nausea while the latter takes a long time but is free of side effects. Measurement of C-peptide is preferable to the measurement of insulin as it has a longer half-life, not subjected to the first-pass hepatic metabolism, and a steady-state clearance from the circulation. Insulin undergoes first-pass hepatic metabolism, has variable clearance, a much shorter half-life, and exogenously administered insulin can confound results. C-peptide is cost-effective when used as the initial test to distinguish LADA from T2DM but has to be later confirmed with antibody testing. In LADA, the MMTT has also been utilized for choosing the treatment modality and predicting the time to transition to managing with insulin.[35] 

HLA typing is not routinely utilized in the evaluation of LADA.

All other routine investigations for other DM patients should be employed in LADA at recommended intervals and as dictated by the clinical situation.

Tests:

• Fasting glucose
• Glycosylated hemoglobin (HbA1C)
• Self-monitoring of blood glucose (SMBG)
• Measures of glycemic variability [best done by continuous glucose monitoring(CGM)]
• Lipid profile
• Estimated glomerular filtration rate (eGFR)
• Serum creatinine
• Urinalysis for albumin excretion (spot and 24-hour specimen with simultaneous creatinine)
• Tests for peripheral neuropathy (Semmes Weinstein monofilament test)
• Retinopathy screening by an ophthalmologist.

If diabetes-related complications set in, other tests may be indicated.

Treatment / Management

Once the diagnosis of LADA is made, non-pharmacological therapies including a diet with carbohydrate and calorie counting, exercise, and precautions to prevent complications similar to those employed in patients with T1DM and T2DM. Since LADA is a heterogeneous condition, pharmacological treatment has to be personalized to gain the maximum therapeutic advantage.

The two goals of pharmacological treatment are (1) to obtain good glycemic control and (2) to prevent or delay complications. Therapies that will preserve beta-cell function are a priority. Insulin has been the treatment of choice for LADA. Studies have shown preserved beta-cell function as evidenced by a maintained stimulated C-peptide response, normal HbA1C levels, and a decrease in autoantibody concentrations.[36]

Sulfonylureas are a poor choice for LADA. They deplete beta-cells of insulin as depicted by falling C-peptide levels, the persistence of antibodies, and the earlier progression to insulin.

Although metformin may help initially in glycemic control in patients with LADA with higher BMI, it alone cannot achieve the second and more important goal of preserving beta-cell function or delaying its destruction. 

Thiazolidinediones have antiinflammatory effects on beta-cells and can prolong their survival and can be useful if used in the earlier stage in LADA. They can also be combined with insulin. Rosiglitazone is the only drug studied in LADA.[37] Unfortunately, the cardiovascular risks associated with this drug have extremely limited its use in the past decade. More long-term studies are warranted with other compounds in this class, like pioglitazone.

DPP4 inhibitors have shown promise alone or when combined with insulin in preserving beta-cell function in LADA. They affect metabolic control by prolonging levels of GLP1 and other peptides. Their primary action is to increase levels of GLP1, thereby suppressing glucagon and increasing insulin secretion after a glucose load by activating DPP4 receptors in the GI tract and brain. DPP4 receptors have also been identified on the surface of T lymphocytes, where they may affect immune regulation. This latter action may be of importance in slowing the beta-cell immune destruction in LADA. Thus far, these studies with DPP4 inhibitors have been either underpowered or not long-term. While this drug category has potential value, additional studies are needed before routinely recommending them for patients with LADA.[38][39][40][41][42]

A study using the GLP1 receptor agonist dulaglutide has shown reductions in HbA1C levels and improvement in beta-cell function in patients with LADA with results comparable to T2DM.[43]

The benefit from the addition of vitamin D to insulin or DPP4 inhibitors in improved glycemic control and preserving beta-cell function has been demonstrated in separate studies.[44][45]

Similarly, a Chinese herbal decoction combined with insulin has yielded positive results.[41]

SGLT2 inhibitors have not been well-studied in LADA. However, some case reports of euglycemic ketoacidosis have appeared. Therefore this category is not recommended.[46]

There are no studies with metformin alone in LADA, and with understandings of both the pathogenesis of LADA and the pharmacological profile of metformin, the latter has no role as a single agent for this condition. 

The novel immunomodulating treatment with alum formulated recombinant human GAD65 by administering 3 injections of 4 mcg each into a lymph node in the groin along with daily oral vitamin D is in phase II of the GADinLADA trial. The study will evaluate the effects of the drug on the parameters of the beta-cell function for one year. This is based on the encouraging results from the DIAGNODE2 T1DM trial and the Diamyd LADA trial, which involved the injection of the GAD 65 formulation subcutaneously 20 mcg per dose for 2 doses one month apart. Several beneficial changes were noted in the immunological system. These included a shift of the GADA isotypes to IgG3/igG4, increased beneficial cytokines, increased FOXP3, and TGF beta, all favoring a Th2 lymphocyte response, more regulatory T lymphocytes, and downplay of activated T lymphocytes thereby promoting a state of immune tolerance. These beneficial effects were persistent even after four years.[47]

In some patients with obesity and DM, bariatric surgery (BS) is undertaken for weight loss and glycemic control. Some cases with presumed T2DM and obesity, in reality, have LADA. In such patients, BS causes successful weight loss, but inadequate glycemic control post-surgery and the remission of diabetes often seen in T2DM are absent in LADA confirming a progressive beta-cell failure. Patients with LADA have an increased post-operative risk of ketoacidosis and should be diligently watched. Some experts have even suggested screening for beta-cell function before BS to avoid disappointment with glycemic control in all obese patients with DM.[48]

Differential Diagnosis

The main challenge is to distinguish patients with LADA from those with T2DM, with which it is often confused. By definition, T2DM has absent autoantibodies to islet cell components, a high normal or often elevated fasting and stimulated C-peptide, and does not need insulin for an extended period. If the risk factors of T2DM are reduced, there is a good chance of control with non-pharmacological measures with or without minimal medications. Physicians should consider screening for LADA in patients with T2DM who do not achieve adequate glycemic control within a reasonable period despite compliance to therapy. This is particularly true if they are not obese, lack the features of the MetS, or they, or their first degree relatives, have other autoimmune disorders, including Hashimoto thyroiditis, Graves disease, celiac disease, rheumatoid arthritis, or pernicious anemia.

T1DM presents dramatically, often with ketoacidosis, needs insulin on detection, and has more than one autoantibody and is easily differentiated from LADA.

Sometimes, a young adult with maturity-onset diabetes of the young (MODY) is mistakenly diagnosed as T1DM, T2DM, or LADA. MODY is rare, has a strong family history, has residual C-peptide, and absent humoral and cellular immunity to islet cell antigens and can thereby be distinguished from LADA.

Prognosis

Patients with LADA have mortality as high as T2DM despite having more favorable metabolic parameters. In the HUNT study, which is to date the largest population-based data on mortality in autoimmune diabetes (LADA), it was shown that hyperglycemia was the only significant influencing factor, and not the components of the metabolic syndrome, in determining mortality that was chiefly due to cardiovascular disease.[49]

It, therefore, goes without saying that strict glycemic control is the key to improving the prognosis in LADA.

Complications

There is evidence that small-fiber neuropathy (SFN) occurs early and with increased frequency in LADA when compared to T2DM, which is related to higher HbA1C and poor glycemic control. Patients with LADA have severe SFN more often than those with age and duration-matched T2DM patients. However, the affliction of large nerve fibers is not different from T2DM. It is recommended that during the evaluation of a patient with LADA, tests to detect SFN should be included. Small nerve fibers carry pain and temperature sensations, mediate sweating and regulate vascular tone, thereby controlling blood flow. Tests for SFN include cold sensation threshold (CST), warm sensation threshold (WST), intraepidermal nerve fiber density (IENFD), and corneal confocal microscopy. The sensitivity of nerve conduction studies (NCS) to diagnose SFN is low and not recommended.[50]

There is a window of opportunity if detected early as the treatment of hyperglycemia may reverse SFN and decrease morbidity.[51][52]

Long-term follow-up of patients with LADA reveals a lower risk in the first nine years but a higher risk in later years for microvascular complications when compared to T2DM even after adjustment for several factors.[53]

patients with LADA have as much carotid artery atherosclerosis as T1DM and T2DM despite a better vascular risk profile.[54]

Three studies, the Botnia study, the Freemantle diabetes study, and the HUNT study all concurred on the increased cardiovascular disease and mortality in LADA, similar to T2DM.

It is clear, therefore, that LADA is associated with both microvascular and macrovascular complications like T1DM and T2DM.

Bone resorption is reduced in LADA, but the mechanisms are different from those in T2DM.[55]

Consultations

• Primary care (family medicine, internal medicine)
• Endocrinology
• Laboratory medicine
• Clinical genetics
• Ophthalmology
• Podiatry
• Bariatric surgery

Deterrence and Patient Education

Patients with LADA need insight into the nature of their disease and the importance of strict glycemic control to prevent microvascular and macrovascular complications.

Pearls and Other Issues

• LADA is a form of DM that has features common to both T1DM and T2DM.
• Early diagnosis is paramount to initiating appropriate treatment and preventing complications.
• New insights into the pathophysiology of LADA explain the slow progression of beta-cell destruction.
• Genetic similarities with both T1DM and T2DM have been revealed in GWAS.
• A C-peptide test, basal or after a mixed-meal, may be used as an initial, cost-effective test to screen patients with LADA to identify which patients need confirmatory testing for islet autoantibodies.
• Sulfonylureas are a poor choice for LADA as it results in beta-cell failure and faster progression to insulin.
• Insulin, DPP4 inhibitors alone and in combination with insulin, thiazolidinediones, and GLP1 receptor agonists have shown promise in achieving glycemic control and preserving beta-cell function.
• Novel immunomodulatory therapies like GAD65 injected into lymph nodes are undergoing trials and appear promising.

Enhancing Healthcare Team Outcomes

According to the World Health Organization (WHO), there were 422 million people with DM globally. As the prevalence of LADA in a population of T2 DM is between 4 and 12 percent depending on the population, between 17 to 50 million will have LADA. This number is likely to grow exponentially in the coming years.

The primary care practitioner (PCP) is very likely to encounter patients with LADA frequently and should be equipped with the knowledge and understanding to recognize and manage this condition promptly.

The endocrinologist sees difficult and complex patients and coordinates care with the PCP, ophthalmologist, podiatrist, and geneticist.

Laboratory medicine advises on the appropriateness of tests, performs biochemical and serological tests, and communicates them to the treating physician on time.

Bariatric surgeons must have a high index of suspicion for LADA in the obese with diabetes and counsel them regarding the less than optimal glycemic control post-surgery.

A coordinated effort by all the above health care providers is needed to achieve good glycemic control, prevent or delay complications, and substantially reduce morbidity, mortality, and health care costs. [Level 3]