Introduction

Insulin resistance syndrome, formerly known as insulin receptor abnormalities, is characterized by pronounced attenuation of insulin action due to functional impairment of insulin receptor signaling [1]. This syndrome has conventionally been classified into type A and type B conditions caused by abnormalities of the insulin receptor gene and by autoantibodies to the insulin receptor, respectively [2, 3]. The concept of insulin resistance syndrome was first proposed by Kahn et al. in 1976 [1]. The subsequent accumulation of basic and clinical knowledge in Japan by 1990 led to the development of diagnostic criteria for insulin receptor abnormalities type A and type B by the hormone receptor abnormality study group of the Ministry of Health and Welfare, with these criteria being published as a comprehensive study report in 1996 [4]. In this study report, Donohue syndrome and Rabson-Mendenhall syndrome, which are also caused by defects of the insulin receptor gene, were classified as distinct from insulin receptor abnormalities [4].

Although more than 25 years have passed since the publication of this study group report, no revisions have been made to the proposed diagnostic criteria. As a result, certain descriptions, including those of diagnostic tests (such as examination of the number and function of insulin receptors with patient-derived cells), are not consistent with the current clinical situation. Revision of the disease classification and diagnostic criteria has also been needed because of the recent discovery of insulin resistance syndrome triggered by mutations in genes related to signaling downstream of the insulin receptor [5,6,7,8,9]. Moreover, it has become necessary to clarify the classification of individuals with insulin receptor autoantibodies who manifest hypoglycemia but not insulin resistance, possibly as a result of the agonistic nature of the antibodies [10].

The report of a nationwide survey of insulin resistance syndrome conducted by the hormone receptor abnormality study group of the Science Research Grants division (research program on rare and intractable diseases) of the Ministry of Health, Labour, and Welfare (MHLW) was published in 2020 [11]. The accumulation of this clinical information was seen as an opportunity to establish a working group at the Japan Diabetes Society to develop, in cooperation with the MHLW study group, disease classification, and diagnostic criteria for insulin resistance syndrome. The resulting report presented here proposes and explains a new classification and diagnostic criteria for this condition.

Classification of insulin resistance syndrome (Table 1)

Table 1 Classification of insulin resistance syndrome
Full size table

Insulin resistance syndrome has conventionally been classified into type A, caused by abnormalities of the insulin receptor gene (INSR), and type B, caused by autoantibodies to the insulin receptor [2, 3]. Although INSR abnormalities also give rise to Donohue syndrome and Rabson-Mendenhall syndrome, these conditions were classified as being different from insulin resistance syndrome [4] as a result of the severity of the defects in receptor function and distinctive physical characteristics of affected individuals [12]. However, from a genetic and pathological point of view, it seems appropriate to consider these two syndromes as severe forms of type A insulin resistance syndrome. In general, individuals with type A insulin resistance syndrome harbor a heterozygous abnormality of INSR, whereas those with Donohue syndrome or Rabson-Mendenhall syndrome often possess homozygous or compound heterozygous abnormalities [13,14,15].

Cases of severe insulin resistance triggered by abnormalities of genes related to insulin receptor signal transduction have also been identified. An abnormality of AKT2 was first identified in such a case [5], with a defect of TBC1D4, which encodes a substrate of the protein kinase AKT, subsequently being similarly reported [6]. Given that reports of such cases have been few in number, insulin resistance syndrome due to mutations in such genes appears to be rare.

In 2013, mutations of PIK3R1 were identified as the cause of SHORT syndrome [7,8,9]. PIK3R1 encodes a regulatory subunit (p85α) of phosphatidylinositol 3-kinase, which plays a key role in regulation of the metabolic actions of insulin [16]. SHORT syndrome is so named as a result of the physical features of affected individuals: short stature (S); hyperreflexia/hernia (H); ocular depression (O); Rieger anomaly (R), a malformation of the cornea and iris; and teething delay (T). About half of individuals with SHORT syndrome also develop diabetes mellitus [17]. PIK3R1 abnormalities have recently been identified in individuals diagnosed with juvenile-onset diabetes with insulin resistance in Japan [18]. It is thus appropriate to classify PIK3R1 abnormalities as a cause of insulin resistance syndrome. In addition, mutations of the protein kinase C isoform ε gene (PRKCE) have been identified in individuals with SHORT syndrome-like physical characteristics as well as insulin resistance [19].

With regard to type B insulin resistance syndrome, instances of amelioration of the syndrome by treatment of comorbid autoimmune disease have been described [20], whereas complete remission as a result of eradication of Helicobacter pylori in a patient with comorbid autoimmune thrombocytopenia was reported [21]. In addition, individuals with autoantibodies to the insulin receptor that manifest hypoglycemia but not hyperinsulinemia have been identified [10]. Such a condition is likely triggered by autoantibodies that activate insulin receptor function.

On the basis of these various findings, the working group now defines insulin resistance syndrome as a condition characterized by severe attenuation of insulin action as a result of functional impairment of the insulin receptor or of downstream signaling molecules. We also propose that the syndrome be classified into two types: genetic insulin resistance syndrome, including type A insulin resistance syndrome, and type B insulin resistance syndrome.

Diagnostic criteria for genetic insulin resistance syndrome (Table 2)

Table 2 Diagnostic criteria for genetic insulin resistance syndrome
Full size table

Genetic insulin resistance syndrome is often suspected on the basis of low birth weight or other physical characteristics in infancy, whereas it is often diagnosed after the detection of glucose intolerance, such as by a school urine examination, during childhood or adolescence [11]. The key finding for the diagnosis of this condition is the presence of insulin resistance without an apparent cause such as obesity, endocrine disease, chronic inflammatory disease, malignant disease, the use of glucocorticoids or other drugs including antimalignancy agents, and the presence of antibodies to insulin or the insulin receptor. The presence of insulin resistance can be determined on the basis of endogenous hyperinsulinemia, with a fasting serum insulin concentration of > 30 μU/mL being apparent in most Japanese cases of genetic insulin resistance syndrome [11]. Measurement of fasting insulin levels in 89 adult Japanese patients with type 2 diabetes and a body mass index of < 25 kg/m2 revealed no cases with a fasting insulin level of > 30 μU/mL [11]. However, depending on the course of treatment, some cases of genetic insulin resistance syndrome manifest fasting insulin concentrations of < 30 μU/mL [22, 23].

Physical characteristics associated with genetic insulin resistance syndrome include acanthosis nigricans, hypertrichosis, and polycystic ovary, all of which are thought to be triggered by chronic insulin resistance and hyperinsulinemia [2]. Whereas Donohue syndrome and Rabson-Mendenhall syndrome are associated with severe intrauterine growth retardation, type A insulin resistance syndrome and insulin resistance syndrome due to PIK3R1 abnormalities are often associated with low birth weight. Donohue syndrome and Rabson-Mendenhall syndrome are discriminated from other forms of genetic insulin resistance syndrome on the basis of their associated physical characteristics and comorbid abnormalities such as distinctive facial features, tooth abnormalities, and pineal body abnormalities. However, given that the symptoms of these two syndromes overlap, it is often difficult to differentiate between them. Cases that result in death of newborns or in early childhood mortality are therefore often diagnosed as Donohue syndrome, whereas those that do not result in such early death are often classified as Rabson-Mendenhall syndrome. Donohue syndrome was formerly known as leprechaunism [24], a term that is now considered ethically inappropriate and is no longer used.

About 40 families with PIK3R1 abnormalities as causes of SHORT syndrome or insulin-resistant diabetes have been reported worldwide [11, 18, 25]. Such genetic abnormalities thus appear to be the second most frequent cause of genetic insulin resistance syndrome, after those affecting INSR. In addition to short stature in adulthood, facial features such as a small, protruding jaw, occipital protrusions, and large, low-set auricles are associated with almost all cases of genetic insulin resistance syndrome caused by PIK3R1 abnormalities, whereas other classical body features of SHORT syndrome are not necessarily observed in these patients [18]. It is important that Silver-Russell syndrome be distinguished from SHORT syndrome, given that it is also characterized by similar facial features, including an inverted triangle–shaped face, a relatively large head, intrauterine growth retardation, and short stature. Mutations in PIK3R1 also cause immunological defects, with a patient manifesting both the characteristics of SHORT syndrome and an immune disorder having been reported [25].

It is sometimes necessary to differentiate genetic insulin resistance syndrome from lipoatrophic diabetes in the clinical setting, given that the former condition is often also accompanied by a decrease in subcutaneous fat mass, likely as a result of impairment of both insulin-induced triglyceride synthesis and proliferation and differentiation of adipocytes. It has been proposed that genetic insulin resistance syndrome associated with PIK3R1 abnormalities, in particular, be categorized as a subtype of lipoatrophy [8].

A definite diagnosis of genetic insulin resistance syndrome (diagnostic category: definite) is determined by detection of mutations that likely affect the function of INSR or of genes related to insulin receptor signal transduction (including PIK3R1, AKT2, TBC1D4, and PRKCE) [26] together with the presence of characteristic clinical features. However, in the nationwide survey of insulin resistance syndrome, cases were identified in which the syndrome was clinically detected but for which mutations in known causative genes were not detected [11]. It is appropriate to tentatively categorize such cases as genetic insulin resistance syndrome with unidentified causative genes. In addition, it is often difficult to perform tests for many candidate genes in the clinical setting. Thus, even if genetic testing is not performed or does not identify a causative gene, individuals suspected of having the syndrome on the basis of endogenous hyperinsulinemia and other clinical manifestations should be categorized as probable genetic insulin resistance syndrome (diagnostic category: Probable).

The clinical manifestations of type A insulin resistance syndrome vary from normal glucose tolerance with hyperinsulinemia to treatment-refractory diabetes even with a large amount of exogenously administered insulin. In addition, reactive hypoglycemia has been found to be caused by type A insulin resistance syndrome [27, 28]. Patients with type A insulin resistance syndrome or with genetic insulin resistance syndrome due to PIK3R1 abnormalities often develop glucose intolerance after their late teens [1].

Given that metformin and sodium-glucose cotransporter-2 (SGLT-2) inhibitors exert hypoglycemic actions by mechanisms independent of insulin receptor signaling, these drugs are effective in individuals with genetic insulin resistance syndrome. Case reports have thus shown that both metformin and SGLT-2 inhibitors improve glycemic control and allow insulin dose reduction in individuals with this syndrome [29, 30]. In addition to insulin, insulin-like growth factor-1 (IGF-1) is administered to treat Donohue syndrome and Rabson-Mendenhall syndrome [31], although maintenance of good glycemic control in patients with these conditions remains a challenge. Patients with these syndromes often die in infancy or early childhood as a result of infections and other complications of their severe metabolic abnormalities [11, 12].

Diagnostic criteria for type B insulin resistance syndrome (Table 3)

Table 3 Diagnostic criteria for type B insulin resistance syndrome
Full size table

Type B insulin resistance syndrome is characterized by severe attenuation of insulin action due to autoantibodies to the insulin receptor. The number of individuals newly found to be positive for such autoantibodies in Japan is ~ 22 per year [11], indicating that this syndrome is relatively rare. The presence of insulin resistance can be determined on the basis of endogenous hyperinsulinemia. The fasting insulin concentration is also > 30 μU/mL in many cases of type B insulin resistance syndrome (major sign), with the average level having been found to be 1122.1 ± 3292.5 µU/mL [11]. However, depending on the course of treatment, some individuals with this syndrome manifest a fasting insulin value of < 30 μU/mL. Acanthosis nigricans was observed in 17% of cases in Japan [11] (reference finding). On the other hand, individuals with autoantibodies to the insulin receptor who do not manifest hyperinsulinemia but who develop hypoglycemia as a primary symptom (corresponding to a few percent of people positive for such autoantibodies) do not meet the definition of insulin resistance syndrome and are therefore not categorized as having type B insulin resistance syndrome. A definite diagnosis is made on the basis of the detection of insulin receptor autoantibodies in the presence of hyperinsulinemia. In the nationwide survey of insulin resistance syndrome [11], the ratio of men to women with type B insulin resistance syndrome was 19:11, indicating that it is not necessarily found more often in women, as was stated in the previous diagnostic criteria [4]. The disease was detected most often in people in their 60s, with the average age of onset being 59.6 ± 16.5 years [11]. There were no distinct regional differences in prevalence of the disease in Japan.

About 76% of cases of type B insulin resistance syndrome are complicated with hypoglycemia [11], which is thus an important clinical feature together with hyperglycemia. Moreover, among examined individuals with this syndrome, 67% had other autoimmune diseases or abnormalities of immunological test values. The most common associated autoimmune disorder was systemic lupus erythematosus (20%), with others including mixed connective tissue disease, Sjogren’s syndrome, autoimmune thyroid disease, autoimmune thrombocytopenia, systemic scleroderma, and rheumatoid arthritis. Given that treatment of comorbid autoimmune disease can result in remission or cure of type B insulin resistance syndrome, examination of patients for such accompanying autoimmune disease is recommended.

Conclusion

We have formulated disease classification and diagnostic criteria for insulin resistance syndrome on the basis of current knowledge including findings of the recent nationwide survey of insulin resistance syndrome in Japan [11]. The disease classification and diagnostic criteria proposed here should lead to a better understanding of the concept of insulin resistance syndrome as well as facilitate its accurate diagnosis and the selection of appropriate treatments. The Japan Diabetes Society plans to establish a registry of insulin resistance syndrome in collaboration with the MHLW study group. Further revision of the classification and diagnostic criteria for this disease will be necessary in the future on the basis of the additional accumulation of disease information, such as that obtained via the disease registry.