Rethinking the Relationship between Insulin and Cancer.

In addition to being a major metabolic hormone, insulin is also a growth factor with a mitogenic effect on all cells, more marked in malignant cells that often overexpress the insulin receptor. In patients with metabolic diseases characterized by hyperinsulinemia (obesity, type 2 diabetes, and metabolic syndrome), the incidence of several types of cancer is increased, as is cancer-related mortality. Because of the worldwide growing prevalence of metabolic diseases and the diffuse use of insulin and its analogs for treating diabetes, the relationship between insulin and cancer has become a clinically relevant issue. Clinical studies have not clarified the degree to which hyperinsulinemia can influence cancer occurrence and prognosis. To better understand this issue, an improved scientific approach is required, with more careful consideration of the mechanisms related to hyperinsulinemia and carcinogenesis.

Long-acting insulin analogs and cancer.

AIMS: Hyperinsulinemia is a recognized risk factor for cancer and plays a major role for the increased cancer incidence in diabetic patients. Whether insulin analogs, and particularly long-acting analogs, worsen the pro-cancer effect of excess insulin is still controversial. DATA SYNTHESIS: In this paper we summarize the biological bases for the potential detrimental effect of long-acting analogs on cancer cells and review the in vitro and in vivo evidence on this issue. Because of their different molecular structure relative to native insulin, insulin analogs may activate the insulin receptor (IR) and the post receptor pathways differently. Most, but not all, in vitro evidence indicate that long-acting analogs may have a stronger mitogenic potency than insulin on cancer cells. Notably insulin glargine, the most studied long-acting analog, also has a higher affinity for the insulin-like growth factor (IGF)-1 receptor, a potent growth mediator. In vitro observations, however, may not reflect what occurs in vivo when analogs are metabolized to derivatives with a different mitogenic activity. Clinical studies, mostly retrospective and predominantly concerning glargine, provide contrasting results. The only perspective trial found no cancer increase in patients treated with glargine. All these studies, however, have severe weaknesses because of the insufficient evaluation of important factors such as dose administered, length of exposure, patient follow-up duration and site-specific cancer investigation. Moreover, whether cancer promotion is a long-acting analog class characteristic or a specific effect of a single agent is not clear. CONCLUSIONS: In conclusion the carcinogenic risk of long-acting analogs, and specifically glargine, can be neither confirmed nor excluded. A personalized and shared decision, considering all the individual risk factors (metabolic and non-metabolic), is the suggestion for the clinician.

Lymph node location is a risk factor for papillary thyroid cancer-related death.

PURPOSE: Papillary thyroid cancer (PTC) has good prognosis with a very low chance of mortality. The prognostic role of metastatic lymph node location was judged controversial and more recently (TNM VIII ed.) was considered to have no impact on the prognosis of older patients. The aim of the study was to evaluate the role of metastasized node location on PTC-related mortality. METHODS: PTC-related mortality was analysed in a consecutive retrospective series of 1653 PTC patients followed at our Thyroid Clinic (mean follow-up 5.9 years). RESULTS: Sixteen out of 1653 patients (0.96%) died because of PTC. Average age was 68 years at presentation and 74.7 at death. F/M ratio was 1:1. The death rate increased in relation to the lymph node status: 0.2% in N0, 0.3% in N1a and 3.0% in N1b. CONCLUSIONS: The presence of lymph node metastases in the N1b compartment should be considered as a risk factor for distant metastatic spread and for cancer-related death and included in post-surgery evaluation.

Abnormal 1-hour post-load glycemia during pregnancy impairs post-partum metabolic status: a single-center experience.

PURPOSE: Recent evidence indicates that people with normal glucose tolerance (NGT) but 1-h post-load plasma glucose (1-h OGTT) ≥ 155 mg/dl have an increased risk for developing Type 2 diabetes mellitus (T2DM), determining a new risk category with deeper metabolic impairment. The aim of this study was to identify, among women with gestational diabetes (GDM), which alterations at OGTT during pregnancy are more frequently associated with 1-h OGTT ≥ 155 mg/dl at post-partum examination. METHODS: Among 297 women affected by GDM, we retrospectively evaluated 244 resulted NGT after delivery. Based on post-partum glucose levels at 1-h OGTT, these people were divided into 188 cases (77.0%) with 1-h OGTT < 155 mg/dl (L-NGT) and 56 (23.0%) with 1-h OGTT ≥ 155 mg/dl (H-NGT). RESULTS: Abnormal glucose levels at 1-h OGTT during pregnancy (≥ 180 mg/dl) were more frequent in H-NGT than in L-NGT (39.3 vs. 24.6%, odds ratio 3.7 [95% CI 1.4-9.6]; p = 0.016). Moreover, H-NGT showed more frequently the simultaneous alteration of all three OGTT plasma glucose values during pregnancy (10.7 vs. 2.1%, odds ratio 4.5 [95% CI 1.5-20.3]; p = 0.038) and less frequently the alteration of fasting plasma glucose alone (14.3 vs. 30.8%, odds ratio 0.4 [95% CI 0.1-0.7]; p = 0.028). CONCLUSIONS: Abnormal 1-h OGTT during pregnancy predicts an increased risk for post-partum 1-h OGTT ≥ 155 mg/dl in women with previous GDM. Even if NGT after delivery, these women may require a closer long-term post-partum follow-up, being at higher risk to develop future glucose intolerance.

Latero-cervical lymph node metastases (N1b) represent an additional risk factor for papillary thyroid cancer outcome.

PURPOSE: Papillary thyroid carcinoma (PTC), the most common thyroid cancer histotype, has a good prognosis even when spread to the neck lymph node (LN). The prognostic role of LN metastases' location is still controversial. The aim of the present study was to evaluate the clinical relevance of the number and location of LN metastases at presentation in PTCs. METHODS: This retrospective study included a consecutive series of 1653 PTC patients followed for a mean period of 5.9 years in a referral thyroid cancer clinic. All patients have undergone thyroidectomy with the dissection of at least six LNs. According to the LN status, patients were subdivided into 569 N0 (34.4%), 644 N1a (39.0%, central compartment) and 440 N1b (26.6%, latero-cervical compartment). RESULTS: Age at diagnosis was significantly lower in N1b (39.8, IQR 30.7-51.6) and N1a (40.1, IQR 31.3-50.1) than in N0 (44.7, IQR 36.6-55.0 yrs). The male gender was more prevalent in N1b than in N1a and N0 (F/M = 1.9/1, 4.0/1 and 5.5/1, respectively). Persistent/recurrent disease at last control was significantly more frequent in N1b (29.8%) than in N1a (14.3%), and in N1a than in N0 (4.2%) (p < 0.01 for all). Also, distant metastases were significantly (p < 0.001) more frequent in N1b (14.1%) than in N1a (4.3%) and N0 (1.6%). LN metastases' number (>5) was a significant risk factor for persistent/recurrent disease only for N1a patients. CONCLUSIONS: These data indicate that persistent/recurrent disease and distant metastases are significantly more frequent in patients with latero-cervical LN (N1b) metastases and that the LN location should be used for a better postsurgical risk stratification.

Tall cell and diffuse sclerosing variants of papillary thyroid cancer: outcome and predicting value of risk stratification methods.

PURPOSE: Tall cell (TCV) and diffuse sclerosing (DSV) variants are aggressive variants of papillary thyroid cancer (PTC). We compared the risk of recurrent/persistent disease in patients with TCV, DSV and classical PTC (cPTC) and evaluated the prognostic accuracy of initial vs. ongoing risk stratification. METHODS: A consecutive series of DSV (n = 54), TCV (n = 72) and cPTC (n = 184) patients was retrospectively analyzed. TCV and DSV patients were first risk stratified for recurrent/persistent disease without considering the histotype as a risk factor and subsequently, 6-24 months after initial treatment, re-classified on the basis of the response to therapy (ongoing risk stratification). RESULTS: Extrathyroidal extension was more frequent in DSV than in TCV and cPTC patients (p < 0.05); moreover, only DSV tumors had a higher rate of recurrent/persistent disease when compared to cPTC treated with the same protocol (total thyroidectomy followed by 131I treatment) (p < 0.01). After initial treatment, 54.2% of TCV and 20.4% of DSV patients were classified at low risk, while at ongoing risk stratification, the excellent response (low risk) was higher for both TCV (77.8%) and DSV (50.0%) patients relative to initial stratification (both p < 0.01). Using ongoing risk classification, positive predictive value (PPV) for persistent/recurrent disease was higher relative to initial risk stratification for both TCV (PPV = 93.8 vs. 39.4%) and DSV (PPV = 63.0 vs. 34.9%), p < 0.05 for both. CONCLUSIONS: In our series DSV, but not TCV patients, had poorer outcome than cPTC treated with the same protocol. Moreover, the ongoing risk stratification predicted outcome better than the initial classification in both TCV and DSV patients.

Heavy metals in the volcanic environment and thyroid cancer.

In the last two decades thyroid cancer incidence has increased worldwide more than any other cancer. Overdiagnosis of subclinical microcarcinomas has certainly contributed to this increase but many evidences indicate that a true increase, possibly due to environmental factors, has also occurred. Thyroid cancer incidence is markedly increased in volcanic areas. Thus, the volcanic environment is a good model to investigate the possible factors favoring thyroid cancer. In the volcanic area of Mt. Etna in Sicily, as well as in other volcanic areas, a non-anthropogenic pollution with heavy metals has been documented, a consequence of gas, ash and lava emission. Soil, water and atmosphere contamination, via the food chain, biocontaminate the residents as documented by high levels in the urines and the scalp hair compared to individuals living in adjacent non-volcanic areas. Trace amounts of metals are essential nutrients but, at higher concentrations, can be toxic for living cells. Metals can behave both as endocrine disruptors, perturbing the hormonal system, and as carcinogens, promoting malignant transformation. Similarly to other carcinogens, the transforming effect of heavy metals is higher in developing organisms as the fetus (contaminated via the mother) and individuals in early childhood. In the last decades environment metal pollution has greatly increased in industrialized countries. Although still within the "normal" limits for each single metal the hormesis effect (heavy metal activity at very low concentration because of biphasic, non linear cell response) and the possible potentiation effect resulting from the mixture of different metals acting synergistically can explain cell damage at very low concentrations. The effect of metals on the human thyroid is poorly studied: for some heavy metals no data are available. The scarce studies that have been performed mainly focus on metal effect as thyroid endocrine disruptors. The metal concentration in tissues has been rarely measured in the thyroid. Heavy metal accumulation and metabolism in the thyroid or the carcinogenic activity of different doses and different speciation of metals has not been investigated. These studies are now warranted to better understand thyroid biology and heavy metal role in human thyroid carcinogenesis.

Insulin, insulin receptors, and cancer.

Insulin is a major regulator of cell metabolism but, in addition, is also a growth factor. Insulin effects in target cells are mediated by the insulin receptor (IR), a transmembrane protein with enzymatic (tyrosine kinase) activity. The insulin receptor, however, is represented by a heterogeneous family of proteins, including two different IR isoforms and also hybrid receptors resulting from the IR hemireceptor combination with a hemireceptor of the cognate IGF-1 receptor. These different receptors may bind insulin and its analogs with different affinity and produce different biologic effects. Since many years, it is known that many cancer cells require insulin for optimal in vitro growth. Recent data indicate that: (1) insulin stimulates growth mainly via its own receptor and not the IGF-1 receptor; (2) in many cancer cells, the IR is overexpressed and the A isoform, which has a predominant mitogenic effect, is more represented than the B isoform. These characteristics provide a selective growth advantage to malignant cells when exposed to insulin. For this reason, all conditions of hyperinsulinemia, both endogenous (prediabetes, metabolic syndrome, obesity, type 2 diabetes before pancreas exhaustion and polycystic ovary syndrome) and exogenous (type 1 diabetes) will increase the risk of cancer. Cancer-related mortality is also increased in patients exposed to hyperinsulinemia but other factors, related to the different diseases, may also contribute. The complexity of the diseases associated with hyperinsulinemia and their therapies does not allow a precise evaluation of the cancer-promoting effect of hyperinsulinemia, but its detrimental effect on cancer incidence and mortality is well documented.

Influence of early-life and parental factors on childhood overweight and obesity.

PURPOSE: We recently reported that a high BMI and high waist circumference prevalence is present in Sicilian children and that the male gender is associated with a significant risk of obesity. Early-life and parent-related risk factors were investigated 1521 Sicilian children (752 females and 769 males, aged 9.0-14.0 years) to identify biological and environmental factors that can contribute to obesity onset. METHODS: Anthropometric measurements of children, their urban vs rural area provenience, birth weight and neonatal feeding were collected. In addition, the BMI and educational level of their parents and the perception of their child weight status were investigated. RESULTS: In the study cohort, the prevalence of overweight and obesity was 27.2 and 14.1 %, respectively, significantly (p < 0.05) higher in males than in females. Breastfeeding emerged as a protective factor (OR 0.64; p < 0.0005), while risk factors for developing childhood obesity were a birth weight ≥4.0 kg (OR 1.83; p < 0.05), an overweight or obese mother (OR 2.33; p < 0.0001) or father (OR 1.68; p < 0.0001) and a mother with a low/medium education level (OR 1.72; p < 0.005). CONCLUSION: Understanding risk factors for pediatric obesity is a prerequisite to identify children at highly risk of being obese and to predispose early intervention strategies.

Type 2 diabetic patients with Graves' disease have more frequent and severe Graves' orbitopathy.

BACKGROUND AND AIMS: Due to the worldwide increasing prevalence of diabetes (DM), patients with both diabetes and Graves' disease (GD) have become more frequent. Sporadic reports indicate that Graves' orbitopathy (GO), a GD complication that affects orbital soft tissues, can be severe in DM patients. The relationship between these diseases is not well understood. This study aims at evaluating the association of GD and GO with autoimmune and non-autoimmune diabetes (DM) and to assess diabetic features that influence GD and GO prevalence and severity. METHODS AND RESULTS: This retrospective study evaluated GD, GO and DM association in 1211 consecutive GD patients (447 with GO and 77 with DM). A case-control study was carried out to evaluate DM relationship with GO severity by comparing at 1:2 ratio GO patients with or without DM. A strong association was found between GD and T1DM (p = 0.01) but not T2DM. Instead, the presence of GO was strongly associated with T2DM (p = 0.01). Moreover, GO was more frequently severe in GD patients with T2DM (11/30 or 36.6%) than in those without T2DM (1/60 or 1.7%, p = 0.05). T2DM was the strongest risk factor for severe GO (OR = 34.1 vs. 4.4 p < 0.049 in cigarette smokers). DM duration, obesity and vascular complications, but not metabolic control were significant determinants of GO severity. CONCLUSIONS: GD is associated with T1DM but not with T2DM, probably because of the common autoimmune background. GO, in contrast, is more frequent and severe in T2DM, significantly associated with obesity, diabetes duration and diabetic vasculopathy but not metabolic control.

Rare diseases in clinical endocrinology: a taxonomic classification system.

PURPOSE: Rare endocrine-metabolic diseases (REMD) represent an important area in the field of medicine and pharmacology. The rare diseases of interest to endocrinologists involve all fields of endocrinology, including rare diseases of the pituitary, thyroid and adrenal glands, paraganglia, ovary and testis, disorders of bone and mineral metabolism, energy and lipid metabolism, water metabolism, and syndromes with possible involvement of multiple endocrine glands, and neuroendocrine tumors. Taking advantage of the constitution of a study group on REMD within the Italian Society of Endocrinology, consisting of basic and clinical scientists, a document on the taxonomy of REMD has been produced. METHODS AND RESULTS: This document has been designed to include mainly REMD manifesting or persisting into adulthood. The taxonomy of REMD of the adult comprises a total of 166 main disorders, 338 including all variants and subtypes, described into 11 tables. CONCLUSIONS: This report provides a complete taxonomy to classify REMD of the adult. In the future, the creation of registries of rare endocrine diseases to collect data on cohorts of patients and the development of common and standardized diagnostic and therapeutic pathways for each rare endocrine disease is advisable. This will help planning and performing intervention studies in larger groups of patients to prove the efficacy, effectiveness, and safety of a specific treatment.

Clinical and molecular mechanisms favoring cancer initiation and progression in diabetic patients.

Cancer incidence and mortality are higher among diabetic patients. This review examines the mechanisms, both general and site-specific, for this increase. Hyperglycemia and hyperinsulinemia, which are the major abnormalities that characterize diabetes, can promote cancer via both independent and synergic mechanisms. Insulin is both a metabolic hormone and a growth factor that promotes cell proliferation. When insulin levels are increased due to either insulin resistance or insulin treatment, their mitogenic effect is more marked in malignant cells that frequently overexpress the insulin receptor and, more specifically, its A isoform that has predominant mitogenic activity. Hyperglycemia provides energy for malignant cell proliferation and, via the peculiar energy utilization of cancer cells, favors cancer growth and neoangiogenesis. Additionally, diabetes-associated obesity has cancer-promoting effects due to mechanisms that are specific to excess fat cells (such as increased peripheral estrogens, increased pro-mitogen cytokines and growth factors). Also fat-associated chronic inflammation can favor cancer via the cell damage caused by reactive oxygen species (ROS) and via the production of inflammatory cytokines and transcription factors that stimulate cancer growth and invasiveness. Finally, the multiple drugs involved in the treatment of diabetes can also play a role. Diabetes-associated comorbidities, tissue-specific inflammation, and organ-specific dysfunctions can explain why the risk of cancer can differ by tissue type among diabetic patients. The increased risk of cancer-related mortality is moderate among individual patients with diabetes (RR = 1.25), but the pandemic nature of the disease means that a considerable number of lives could be spared through a better understanding of the factors associating diabetes and cancer.

Improving the prediction of malignancy in cytologically suspicious thyroid nodules.

BACKGROUND: Fine needle cytology aspirates (FNA) classified as THY4 are a heterogeneous group suspicious for malignancy [papillary thyroid cancer (PTC)], which is confirmed in 50-80% of cases after surgery. AIM: To better stratify THY4 FNA specimens for the relative risk of malignancy. METHODS: We retrospectively analyzed 78 thyroid nodules classified as THY4 because of the presence of atypical cells, hypercellular trabeculae and/or intranuclear inclusions (ICI), in the absence of papillae. Two subgroups were identified: group 1 (38 nodules), showing ICI with (no.=17) or without (no.=21) trabeculae and cellular atypia, and group 2 (40 nodules), showing trabeculae and atypia but without ICI. RESULTS: PTC was detected at histology in 56/78 of the patients (71.8%). Malignancy occurred in 36/38 (94.7%) of the patients in group 1 and in 20/40 (50.0%) of the patients in group 2. Therefore, the positive predictive value (PPV) for PTC was 97.3% in the ICI+ specimens (group 1), with a sensitivity of 64.3% and specificity of 95.2%. When only ICI was present, without atypia and trabeculae, the PPV and specificity were similar (95.0 and 95.2%, respectively), but the sensitivity was decreased (48.7%). In specimens without ICI (group 2), the PPV was only 50.0%; however, combined with young age (<40 yr) and male gender, it reached a value similar to that of group1. CONCLUSIONS: In ICI+ specimens compared to ICI-, the risk of PTC is nearly doubled, since PPV increases from 50.0% to 97.3%. This observation suggests that surgery should be considered mandatory in all lesions classified THY4 at FNA, although the relevant difference in terms of cancer risk between ICI- vs ICI+ nodules might be an useful information for both the clinician and the patient.

Adiponectin increases glucose-induced insulin secretion through the activation of lipid oxidation.

The expression of adiponectin receptors has been demonstrated in human and rat pancreatic beta cells, where globular (g) adiponectin rescues rat beta cells from cytokine and fatty acid-induced apoptosis. The aim of our study was to evaluate whether adiponectin has a direct effect on insulin secretion and the metabolic pathways involved. Purified human pancreatic islets and rat beta cells (INS-1E) were exposed (1 h) to g-adiponectin, and glucose-induced insulin secretion was measured. A significant increase in glucose-induced insulin secretion was observed in the presence of g-adiponectin (1 nmol/l) with respect to control cells in both human pancreatic islets (n = 5, p < 0.05) and INS-1E cells (n = 5, p < 0.001). The effect of globular adiponectin on insulin secretion was independent of AMP-dependent protein kinase (AMPK) activation or glucose oxidation. In contrast, g-adiponectin significantly increased oleate oxidation (n = 5, p < 0.05), and the effect of g-adiponectin (p < 0.001) on insulin secretion by INS-1E was significantly reduced in the presence of etomoxir (1 µmol/l), an inhibitor of fatty acid beta oxidation. g-Adiponectin potentiates glucose-induced insulin secretion in both human pancreatic islets and rat beta cells via an AMPK independent pathway. Increased fatty acid oxidation rather than augmented glucose oxidation is the mechanism responsible. Overall, our data indicate that, in addition to its anti-apoptotic action, g-adiponectin has another direct effect on beta cells by potentiating insulin secretion. Adiponectin, therefore, in addition to its well-known effect on insulin sensitivity, has important effects at the pancreatic level.

The tall cell variant of papillary thyroid carcinoma: clinical and pathological features and outcomes.

BACKGROUND: The tall cell variant (TCV) is a relatively rare variant of papillary thyroid cancer. Since a controversy exists whether or not the TCV has a worse outcome, the aim of our study was to retrospectively compare the clinicopathological features and outcomes in a group of TCV patients and a larger group of patients with classical papillary thyroid carcinoma (cPTC). SUBJECTS AND METHODS: Data from 30 TCV and 293 cPTC patients were analyzed. Among the 293 cPTC, we also selected a "high-risk" cPTC group (no.=103) that was treated with the same protocol used for the TCV patients. All data were managed by Cox analysis. RESULTS: Compared to all cPTC patients, TCV subjects displayed only a significantly higher rate of extrathyroid extension. At multivariate analysis, TCV was not an independent variable for the prediction of a high risk of persistent/recurrent disease. At the last follow-up observation, there was no difference in the disease status between the TCV and all cPTC patients. Moreover, "high-risk" cPTC patients had a significant increase in persistent/recurrent disease. CONCLUSIONS: In our study, although the TCV histotype is associated with a higher prevalence of extrathyroid extension, it is characterized by an outcome that is not significantly different from that of all cPTC patients and is more favorable than that of "high-risk" cPTC patients. Only those TCV patients classified as "high risk" based on specific pathological and clinical features, according to current guidelines, should be treated aggressively, such as with a total thyroidectomy, neck lymph node dissection or ablative radioiodine treatment.

Insulin and its analogs: actions via insulin and IGF receptors.

Insulin analogs are artificially modified insulin molecules that allow better metabolic controls of diabetes through either more rapid or more prolonged activity. The interaction of insulin analogs with the insulin receptor isoforms (IR-A and IR-B) and with the IGF-I receptor (IGF-IR) is similar but not identical to that of insulin, and therefore, their biological effects do not always reproduce insulin actions in terms of quantity, quality and timing. Studies on in vitro models indicate that short-acting analogs elicit molecular and biological effects that are similar, but not identical, to those of insulin via IR-A, IR-B and IGF-IR. In contrast, long-acting analogs behave in a more different way relative to insulin. Although data are not homogeneous and observations on the more recently introduced detemir are scarce, both glargine and detemir often show a decreased binding to IR and increased binding to IGF-IR. Also, intracellular signaling is different with respect to insulin, with a prevalent activation of the ERK rather than the AKT pathway. Finally, an increased mitogenic response has often been observed with these analogs in a variety of cell models. Of course, in vitro studies do not necessarily reflect what occurs in patients, due to the different metabolism of analogs in vivo and their interaction with components of the extracellular environment. After many years of analog's use, observations in patients indicate that insulin analogs are both effective and safe. Prospective clinical studies, however, may add further useful information on the issue of the insulin analogs' possible differences with respect to native insulin.

Insulin analogues differently activate insulin receptor isoforms and post-receptor signalling.

AIMS/HYPOTHESIS: Five insulin analogues, with modified insulin-like molecular structures, are currently approved for treating diabetic patients. They activate cell signalling and biological responses via insulin receptor isoforms (IR-A and IR-B), each having specific characteristics for eliciting cell responses. The molecular and biological effects of these analogues on receptor isoforms in comparison to native insulin are not well defined, and their effects on the IGF1 receptor (IGF1R) are controversial. The characterisation of these effects was the aim of the present study. METHODS: Short-acting (insulin lispro [B28Lys,B29Pro human insulin], insulin aspart [B28Asp human insulin], insulin glulisine [B3Lys,B29Glu human insulin]) and long-acting (insulin glargine [A21Gly,B31Arg,B32Arg human insulin], insulin detemir [B29Lys(epsilon-tetradecanoyl),desB30 human insulin]) insulin analogues were studied in three engineered cell models (R(-), IGF1R-deprived mouse fibroblasts transfected with either only human IR-A or IR-B or IGF1R). Receptor binding and phosphorylation, AKT and extracellular signal-regulated kinase (ERK) activation, cell proliferation and colony formation were evaluated after exposing the cells to each analogue and were compared with insulin, IGF1 and the carcinogenic analogue B10Asp. RESULTS: All short-acting insulin analogues produced molecular and biological effects similar but not identical to those of insulin. Relative to insulin, long-acting analogues more strongly activated the ERK pathway via both IR-A and IGF1R as well as increased cell proliferation. At the concentration tested, no analogue (except B10Asp via IR-A) had increased transforming activity. CONCLUSIONS/INTERPRETATION: Cell models that permit comparisons of the activity of insulin to that of insulin analogues via each receptor individually indicate that only minor differences exist between insulin and short-acting analogues. By contrast, long-acting analogues activate the mitogenic signalling pathway more effectively than insulin and cause increased cell proliferation.