Diabetes and cancer, common threads and missing links

Cancer Letters 374 (2016) 54–61

Contents lists available at ScienceDirect

Cancer Letters j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / c a n l e t

Mini-review

Diabetes and cancer, common threads and missing links Fang Hua, Jiao-Jiao Yu, Zhuo-Wei Hu * Immunology and Cancer Pharmacology Group, State Key Laboratory of Bioactive Substance and Function of Natural Medicines, Institute of Materia Medica, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing, 100050, China

A R T I C L E

I N F O

Article history: Received 30 December 2015 Received in revised form 3 February 2016 Accepted 3 February 2016 Keywords: Hyperinsulinemia Insulin resistance Inflammation Obesity ER-stress Autophagy

A B S T R A C T

Diabetes mellitus is a serious and growing health problem worldwide and is associated with severe acute and chronic complications. Accruing epidemiological and clinical evidence have suggested that an increased cancer incidence is associated with diabetes as well as certain diabetes risk factors and diabetes medications. Several pathophysiological mechanisms for this relationship have been postulated, including insulin resistance and hyperinsulinemia, enhanced inflammation, aberrant metabolic state, endoplasmic reticulum stress, and deregulation of autophagy. In addition to these potential mechanisms, a number of common risk factors, including obesity, may be behind the association between diabetes and cancer. Furthermore, different anti-diabetic medications may modify cancer risk and mortality in patients with diabetes. This Review discusses evidence to support the relationship between diabetes and cancer development as well as the underlying mechanisms. We also discuss the relationship of current diabetes treatments and cancer risk or prognosis. Understanding the mechanisms that connect type 2 diabetes or diabetes treatments to cancer are crucial for establishing the fundamental strategies concerning about primary prevention, early detection and effective therapy against these diseases. © 2016 The Authors. Published by Elsevier Ireland Ltd. This is an open access article under the CC BYNC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Introduction Diabetes is a common chronic disease with tremendous impact on health worldwide. Statistical data from the World Health Organization (WHO) reports that about 347 million people worldwide have diabetes and it is predicted to become the 7th leading cause of death in the world by the year 2030. Epidemiologic study suggests that people with diabetes are at significantly higher risk for many forms of cancer and greater cancer mortality, predominantly type 2 diabetes mellitus (T2DM) [1]. In 2010, the American Cancer Society and the American Diabetes Association released a consensus report that recommended regular cancer screening for diabetics and careful consideration in selecting diabetes medication for patients at very high risk for cancer or recurrent cancer [2]. Different cancer types, as well as T2DM could share a number of major risk factors. Some risk factors are nonmodifiable, such as age and sex. There is no doubt that the incidence and mortality of cancer and diabetes are increasing with age, although the two diseases have trends of attacking young adults. Sex is another risk factor; overall cancer occurs more frequently in men. Similarly, men show slightly higher risk of diabetes than women. Cancer and diabetes also share many modifiable risk factors, including obesity, diet, physical activity, tobacco smoking and alcohol drinking. Except for the

* Corresponding author. Tel.: +86 10 8316 5034; fax: +86 10 8316 5034. E-mail address: [email protected] (Z.-W. Hu).

common risk factors, meta-analyses have revealed T2DM to be an independent risk factor for the development of several different types of cancer [3,4]. The fact that we should pay more attention is that some medications used to treat hyperglycemia are associated with either increased or decreased risk of cancer. Although, the mechanisms that underlie the associations between T2DM and cancer risk remain far from understood, the insulin-insulin like growth factor (IGF) axis, inflammation, autophagy, endoplasmic reticulum stress (ER stress) and other mechanisms have been proposed to be important in this process. In this review, we describe the epidemiological evidence supporting the relationship between cancer and T2DM, and discuss the biologic links and mechanisms associated with T2DM, the metabolic syndrome, and obesity that may promote cancer initiation, growth, and metastases. Epidemiological evidence for the connection of T2DM to cancer Metabolic syndrome is a clustering of at least three of five of the following disorders: abdominal obesity, high blood pressure, elevated fasting plasma glucose, high triglycerides, and low highdensity lipoprotein (HDL) levels. Metabolic syndrome is associated with a greater risk of developing diabetes with insulin resistance as the cornerstone. Researchers have paid much attention to the relationship between diabetes and cancer. A multi-national program named as “The Metabolic Syndrome and Cancer (Me-Can) Project” was initiated in 2006 to investigate factors of the metabolic syn-

http://dx.doi.org/10.1016/j.canlet.2016.02.006 0304-3835/© 2016 The Authors. Published by Elsevier Ireland Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/ by-nc-nd/4.0/).

F. Hua et al./Cancer Letters 374 (2016) 54–61

drome on the association with cancer risk. Me-Can Project and other investigators reported that a number of cancers, including liver, pancreas, endometrium, colorectal, gastric, breast and bladder are associated with higher metabolic syndrome score, although gender difference can be observed in some types of cancer [5–10]. Among these cancers, the greatest increase in risk has been found in hepatocellular carcinoma (HCC). T2DM is associated with an increased incidence of HCC, the relative risks (RR) is 2.01, with the 95% confidence intervals (CI) as 1.61–2.51, in comparison with individuals without T2DM [11]. The increased incidence of HCC is independent of geographic location, alcohol consumption, history of cirrhosis, or infections with hepatitis B virus (HBV) or hepatitis C virus (HCV). The relationships between pancreatic cancer and diabetes are more complex as the two diseases may be of reciprocal causation. Diabetes is a risk factor for pancreatic cancer (RR = 1.97, 95% CI = 1.78–2.18) [12]. Meanwhile, evidence suggest that recentlydeveloped diabetes may be a consequence of pancreatic cancer due to impaired phosphatidylinositol 3-kinase (PI3-K) signaling or islet dysfunction [13]. Moreover, other pathological changes caused by pancreatic cancer may also contribute to the new-onset diabetes. Pancreatic tumor mass of ductal adenocarcinoma most commonly arises from the head of the pancreas, which are the sections where the bile duct joins with the pancreatic duct. Hence, the bile duct obstruction is often happened in pancreatic cancer patients. A recently published work investigated the relationship between blood glucose homeostasis and partial pancreatectomy. The authors revealed that surgically reversible blood glucose dysregulation diagnosed concomitantly with a (peri-) pancreatic tumor appears secondary to compromised liver function due to tumor compression of the common bile duct and the subsequent increase in insulin resistance, which is called as “cholestasis-induced diabetes” [14]. Bile acid is involved in the regulation of hepatic glucose metabolism by the nuclear receptor farnesoid X receptor (FXR) and induce glucagon-like peptide-1 (GLP-1) secretion by the G-proteincoupled membrane receptor TGR5-mediated pathways [15]. In addition, new-onset diabetes in pancreatic cancer patients is likely to be a paraneoplastic phenomenon caused by tumor-secreted adrenomedullin. Levels of adrenomedullin were higher in patients with pancreatic cancer who developed diabetes compared those who did not [16]. Mechanically, pancreatic cancer causes paraneoplastic β-cell dysfunction by shedding adrenomedullin+/ CA19-9+ exosomes into circulation that inhibit insulin secretion through ER stress and failure of the unfolded protein response (UPR) [17]. The RR of cancer at other sites ranging from 1.27 (95% CI 1.16– 1.39) for breast cancer [18] to 2.22 (95% CI 1.8–2.74) for endometrial cancer [19]. In these cancers, diabetes may be an independent risk factor. A prospective study of six European cohorts revealed that abnormal glucose metabolism is associated with an increased risk of cancer and cancer death overall and at several cancer sites independent of body mass idex (BMI). RR (95% CI) per 1 mmol/L increment of glucose for overall incident cancer is 1.05 (1.01–1.10) in men and 1.11 (1.05–1.16) in women, and corresponding RRs for cancer death are 1.15 (1.07–1.22) and 1.21 (1.11–1.33), respectively [5]. Notably, T2MD is significantly inversely associated with risk of developing prostate cancer [20]. However, obese men with prostate cancer have higher cancer mortality rates than those of patients with normal body weight [21]. During the last several years, epidemiologic evidence linking antidiabetic drugs with cancer risk has been considered. Anti-diabetic therapies may either target for lowering of glucose by insulin sensitizers (e.g. metformin and thiazolidinediones) or stimulate the pancreas to secret much more insulin (e.g. sulfonylurea drugs). In addition, once T2DM patients become insulin dependent, they are treated with different analogs of insulin. Recently, a meta-analysis demonstrated that the use of metformin or thiazolidinediones is

55

associated with a lower risk of overall cancer incidence (RR = 0.86, 95% CI 0.83–0.90; RR = 0.93, 95% CI 0.91–0.96 respectively). In contrast, insulin, sulfonylureas and alpha glucosidase inhibitor use is associated with an increased risk of overall cancer incidence (RR = 1.21, 95% CI 1.08–1.36; RR = 1.20, 95% CI 1.13–1.27; RR = 1.10, 95% CI 1.05–1.15, respectively) [22]. Such a fact not only reminds the clinicians to choose the anti-diabetic medications deliberately, but also facilitates the exploration of biological links and mechanisms to decipher the linking between the two diseases. Biological mechanisms linking diabetes and cancer Several mechanisms have been proposed to explain links between diabetes and increased cancer risk (Fig. 1), including deregulation of insulin and IGF signaling, obesity and inflammation, metabolic symbiosis; moreover, ER stress and autophagy have also emerged as important cellular mechanisms linking diabetes to cancer. Insulin-IGF axis Insulin resistance is common in individuals with obesity or T2DM, in which insulin action is impaired in peripheral target tissues and then circulating insulin levels are frequently increased. Hyperinsulinemia also results in reduced levels of IGF binding protein-1 (IGFBP-1) and IGFBP-2 (which normally bind to and inhibit the action of IGF-1), thus increasing the levels of free and bioactive IGF-1. Both insulin and IGF-1 activate the receptor tyrosine kinase pathway, insulin receptor (IR) and IGF-1 receptor (IGF-1R) respectively, which are expressed at higher levels in malignant cells. Activation of these receptors results in activation of insulin response substrate-1 (IRS-1) and downstream mitogen-activated protein kinase (MAPK) pathway, phosphoinositol-3 kinase/Akt (PI3KAkt) pathway, as well the Janus kinase/signal transducer and activator of transcription (JAK/STAT) pathway. The results of activating these pathways are promoting protein synthesis, increasing cellular proliferation, protection from apoptotic stimuli and participating in the initiation and maintenance of cancer stem cells [23]. There is a high degree of homology between IGF1R and IR, they can form hybrid receptors. IGF1 and IGF2 have high affinity for the hybrid receptors compared with insulin and increased levels of hybrid receptors have been observed in many cancer tissues [24]. Experimental evidence suggested that decreasing circulating insulin levels in mice or down-regulating the IR in cancer cells and xenografts reduced tumor growth, angiogenesis, lymphangiogenesis, and metastasis [25,26]. Moreover, increased circulating insulin can induce a reduction of the sex hormone-binding globulin, leading to increases in bioavailable sex hormones [27]. Elevated endogenous sex steroid levels are associated with a higher risk of postmenopausal breast and endometrial cancers. Because the pivotal roles of IGFs in tumorigenesis and growth, more than 10 of IGF/IGF-1 inhibitors (monoclonal antibodies against IGF-1R or its ligands, and IGF-1R tyrosine kinase inhibitors) have entered clinical trials but demonstrating unsatisfactory results [28], suggesting that other critical mechanisms exist and participate in the mediation of cancerpromoting roles of insulin/IGF signaling. Obesity and the inflammation Diabetes and obesity are closely linked; approximately 80%– 90% of patients diagnosed with T2DM are also obese. Diabetes and obesity display mutual promotion effects. Insulin resistance creates increased levels of insulin and glucose in the blood stream, which is a major underlying cause of excess weight and obesity, whereas inflammation factors releasing by the infiltrated macrophages in pancreatic tissue destroy insulin-producing β cells and facilitate the progression of T2DM [29]. Obesity performs the cancer promo-

56

F. Hua et al./Cancer Letters 374 (2016) 54–61

Fig. 1. Biological mechanisms connecting diabetes and cancer. Schematic representation of the five main mechanisms that are hypothesized to link diabetes and cancer risk. Abbreviations: T2DM, type 2 diabetes mellitus; TAMs, tumor-associated macrophages; CAFs, cancer-associated fibroblasts; CAAs, cancer-associated adipocytes; FFA, free fatty acid; CSCs, cancer stem cells; LPS, lipopolysaccharides; IGFBP, Insulin-like growth factor-binding protein; IGF, Insulin-like growth factor; MAPK, mitogenactivated protein kinase; PI3K, Phosphatidylinositol-3 kinases; JAK, Janus kinase ; STAT, signal transducers and activators of transcription; IR, insulin receptor; IGF-1R, Insulinlike growth factor receptor; UDP-GlcNAc, uridine diphosphate N-acetylglucosamine; SREBPs, sterol regulatory element-binding proteins; SCAP, SREBP cleavage activating protein; GRP78, glucose-regulated protein 78; ER stress, endoplasmic reticulum stress; FOXO, forehead box O; ULK1, unc-51 like autophagy activating kinase 1; mTOR, mammalian target of rapamycin.

tion effect through many aspects. First, obesity promotes the establishment of tumor microenvironment. The survival of cancer cells is critically dependent on their interactions with neighboring nonmalignant cells in the tumor stroma. Cancer-associated fibroblasts (CAFs), tumor-associated macrophages (TAMs) and cancerassociated adipocytes (CAAs) are well considered as tumor-promoting cell types. Adipose tissue within the tumor microenvironment actively contributes to tumor growth and metastasis by functioning as an endocrine organ, through secretion of leptin, adiponectin, free fatty acid (FFA), proinflammatory cytokines, proangiogenic factors, and extracellular matrix constituents [30]. Furthermore, adipose tissue can act as an energy reservoir for embedded cancer cells [31]. The obesity related tumor microenvironment also helps the initiation and maintenance of cancer stem cells. Clinically, individuals with obesity display more resistance to chemotherapy or radiotherapy than do lean ones, partly due to the increased number of CSCs present in the adipose tissue. During obesity, adipose tissue expansion is associated with insufficient vascularization which results in hypoxia and infiltration of not only macrophages, but also T cells and natural killer (NK) cells. These cells generate large amounts of pro-inflammatory factors, including tumor necrosis factor α (TNFα), interleukin-6 (IL-6), IL-8, IL-18 and et al. Each of these

factors might play an etiologic role in regulating malignant transformation or cancer progression. Second, obesity changes of host– microbiota interactions. Obesity enhances the intestinal permeability leading to lipopolysaccharides (LPS) leaking out into the body from the gram-negative part of the intestinal microbiota, which contributes to trigger the low-grade inflammation and carcinogenesis [32,33]. Obesity can also provoke senescence-like features in hepatic stellate cells (HSCs) and promote tumorigenesis in hepatocytes adjacent to these HSCs. Gut Gram-positive bacteria was strikingly increased in high fat diet feeding mice, thereby increasing the levels of a gut bacterial metabolite-deoxycholic acid (DCA) to cause DNA damage. The enterohepatic circulation of DCA provokes senescenceassociated secretory phenotype (SASP) in HSCs, which in turn secretes various inflammatory and tumor-promoting factors in the liver, thus facilitating hepatocellular carcinoma development [34]. Excess adiposity especially shows increased risk of postmenopausal breast, endometrial and ovarian cancers due to the higher rates of conversion of androgenic precursors to estradiol through increased aromatase enzyme activity in peripheral adipose tissue. Abundant experimental evidence have demonstrate that estrogens can exert mitogenic and mutagenic effect to induce direct or indirect DNA damage, genetic instability and mutations in cells in

F. Hua et al./Cancer Letters 374 (2016) 54–61

normal and neoplastic mammary tissues [35]. The Endogenous Hormones and Breast Cancer Collaborative Group (EHBCCG) reported that postmenopausal breast cancer risk is increased among women with higher concentrations of circulating sex hormones and the association of BMI with postmenopausal breast cancer risk was almost entirely explained by the increase in estradiol levels with higher BMI [36]. For endometrial cancer, increased estradiol levels not only increase endometrial cell proliferation and inhibit apoptosis but also stimulate the local synthesis of IGF1 in endometrial tissue [37]. Metabolic symbiosis Establishing the relationship between cancer and diabetes/ obesity has kindled a wide interest in the metabolism of cancer cells and particularly about the Warburg effect. Cancer cells, like bacteria, cannot burn fats. So cancer cells are dependent on glucose for energy. Cancer cells have a high glycolysis rate even in the presence of oxygen. This feature of cancer cells is well known as the Warburg effect. Compared with oxidative phosphorylation, aerobic glycolysis is an inefficient way to generate adenosine 5′-triphosphate (ATP), so the cancer cells have to increase the glucose uptake to provide nourishment for their rapid growth and replication requirement. That means, cancer cells are sugarholics! The hyperglycemia state in diabetic patients just forms a suitable soil for cancer cells’ growth and development. Some researchers suggested that aerobic glycolysis occurs in stromal cells rather than cancer cells, and provides lactate and pyruvate to cancer cells through a paracrine exchange of these nutrients. This phenomenon is referred to as the reverse Warburg effect [38]. Similarly, the lipolytic activity was increased in stromal adipocytes to provide FFA as an energy source for cancer cells [39]. In cancer and metabolic diseases, elevated lipogenesis is another common pathophysiological characteristic. Sterol regulatory element-binding proteins (SREBPs) are critical transcription factors in this process, which control the expression of genes important for the uptake and synthesis of cholesterol, fatty acids, and phospholipids. Elevated blood glucose activates SREBP by stabilizing SREBP cleavage activating protein (SCAP), a central regulator of the SREBP pathway; stimulates pancreatic insulin secretion, which activates SREBP-dependent lipogenic gene expression; and generates acetyl-CoA, the substrate for lipogenesis [40]. Furthermore, glucose metabolism can connect with epigenetic changes in oncogenic pathways. Recently, a study demonstrated that phenotypes in basal-like breast cancer are promoted by a metabolic switch to glucose metabolism through the promoter methylation of fructose-1,6-bisphosphatase (FBP1), which is a ratelimiting enzyme in gluconeogenesis [41]. On the other hand, chronic hyperglycemic conditions may cause epigenetic changes in oncogenic pathways in cancer cells. Transient hyperglycemia induces the recruitment of the transcription factor Set7 to the nuclear factorκB (NF-κB) p65 promoter, resulting in increased NF-κB activation and increased inflammation [42]. ER stress The endoplasmic reticulum is a cellular calcium store and is responsible for post-translational modification, folding, and assembly of newly synthesized secretory and membrane-bound proteins. Protein flux through the ER must be carefully monitored to prevent dysregulation of ER homeostasis and stress. ER stress elicits the UPR which influences both cellular life and death decisions. In obesity, ER stress is induced by an augmented demand for protein synthesis under nutrient excess and by elevated levels of saturated FFA. During the progression of T2DM, the demand of insulin production is increased to compensate for ongoing insulin resistance. Maturation of proinsulin into insulin requires its processing in the ER, it is believed that this increased demand, together with in-

57

creased circulating FFA and hyperglycemia, triggers ER stress in islet β cells [43,44]. Chronic ER stress eventually leads to β cell death, which further exacerbates hyperglycemia. The activation of ER stress in cancer cells is associated with facilitating cancer cell survival and tumor growth. The low pH, low oxygen tension, and low nutrient supply in solid tumors result in the accumulation of unfolded, misfolded aggregated proteins and reactive oxygen species (ROSs), which could signal cell death. However, cancer cells have developed a capacity to survive under these extreme conditions through modulation of the UPR response. Glucose-regulated protein 78 (GRP78) is an ER protein chaperone, which plays a major role in the adaptive response to ER stress. Highly expressed GPR78 is commonly found in many cancers and correlates with the tumor recurrence, therapeutic resistance and cancer stemness phenotype [45–49]. High glucose and leptin can induce high expression of GPR78 [50,51], providing some clues of the connection between ER stress and diabetes-related cancer. Except for GPR78, the components of the UPR pathway, such as inositol-requiring enzyme-1α (IRE1α), X-box-binding protein-1 (XBP1) and protein kinase RNA-like endoplasmic reticulum kinase (PERK) have also been implicated in cancer [52] and glucose homeostasis [53,54]. However, much work needs to be done to elucidate the exact connection and mechanism of ER with diabetes-related cancers. Autophagy Cell homeostasis is maintained by a precisely regulated balance between synthesis and degradation of cellular components. Eukaryotic cells have two major degradation systems, the lysosome and the proteasome. The proteasome selectively recognizes only ubiquitinated substrates, which are primarily short-lived proteins. In contrast, long-lived proteins and redundant or damaged intracellular organelles can be delivered to the lysosome by autophagy. Accumulating evidence have suggested that these two systems participate in the pathogenesis of many diseases including diabetes and cancers. The emerging role of autophagy in cancer is one of a doubleedged sword. Autophagy confers tumor cells with superior stress (nutrient, oxygen deprivation together with inefficient ATP production) tolerance, which limits damage, maintains viability, sustains dormancy, and facilitates recovery. Although autophagy is a survival pathway used by both normal and tumor cells to survive starvation and stress, paradoxically, autophagy defects are found in many human tumors. Allelic deletion of the essential autophagy gene beclin1 is frequently observed in human breast, ovarian, and prostate cancers [55]. Heterozygous disruption of the beclin 1, deficiency of atg5 or atg7 render mice tumor prone [56,57]. Under starvation conditions, the activity of the phosphoinositide 3-kinase (PI3K) pathway and mammalian target of rapamycin (mTOR) is suppressed resulting in the activation of autophagy. However, in cancer cells PI3K pathway is constitutively activated by mutations, which causes unrelenting proliferation of tumor cells than the survival disadvantage conferred by autophagy suppression. Cancer is a kind of proteinopathy with accumulation of mutated oncoproteins. Defects in autophagy increase the protein quality control burden of cancer cells. Furthermore, autophagy inhibition can induce cell death, inflammation, and genome damage, which further enhance tumor progression. Then what’s the role of autophagy in diabetes-related cancer? Autophagy is inhibited by the insulin-amino acid-mTOR signaling pathway. Insulin inhibits autophagy in many ways: first by activating mTOR in synergy with amino acids [58], which results in the phosphorylation and inhibition of unc-51 like autophagy activating kinase 1 (ULK1) [59]; and second by Akt-mediated phosphorylation and inhibition of the transcription factor FoxO3,

58

F. Hua et al./Cancer Letters 374 (2016) 54–61

which controls the transcription of autophagy-related genes, including LC3 and Bnip3 [60]; third by Inhibiting expression of autophagy-related genes, such as VPS34 and Atg12 in a FoxO1dependent manner [61]. The existing evidence indicates that hyperinsulinemia-induced autophagy inhibition in diabetes may be the cause of tumorigenesis and tumor progression. Recently, our work provides circumstantial evidence to support the notion that TRB3-induced autophagy suppression is a critical link between diabetes and cancer [62]. TRB3 is a stress-induced protein. Many stress factors in diabetic conditions, such as hyperglycemia, hyperinsulinemia, high IGF-1, hypoxia, ER stress, oxidative stress and even inflammatory factors, upregulate the expression of TRB3 in lung, liver and other organs. Simultaneous higher expression of TRB3 and phosphorylated insulin receptor substrate1 (IRS1) in several human cancer tissues along with poor prognosis indicates the clinical relevance of these signal molecules. TRB3 physically interacts with the autophagic receptor P62. This interaction hinders the binding of P62 to LC3 and to ubiquitinated substrates, leading to suppression of autophagic flux and accumulation of P62. Interestingly, autophagy defect-induced accumulated P62 not only binds to ubiquitinated proteins but also impedes the delivery of ubiquitinated substrates to the proteasome. Thus, metabolic stressenhanced TRB3 interferes with both autophagic and proteasomal substrate clearance, resulting in the accumulation of P62 and many tumor-promoting factors in tissue to promote tumorigenesis and progression. These observations emphasize the importance of effective autophagic flux and homeostasis of protein in the inhibition of cancer. Diabetes treatments and cancer Improved glucose control is one of the central goals of effective diabetes management, which strives to minimize morbidity and mortality by reducing the risk of diabetes-associated complications. When selecting pharmacologic therapies against diabetes, several factors are considered by clinicians and patients, including the type of diabetes; the glucose-lowering potential of a given agent, known acute and chronic side effects of treatment, treatment costs, and the multiple chronic comorbidities in patient. During the last several years, a few studies have focused on cancer risk in relation to different types of anti-diabetic treatments. Metformin The biguanide metformin is the most commonly used therapeutic medication in patients with T2MD, often prescribed as initial or combination therapy. A comprehensive meta-analysis has confirmed that taking metformin reduces cancer incidence by 30–

50%, most significantly in pancreatic cancer, hepatocellular carcinoma, and colon cancer [63]. Systemically, metformin reduces circulating insulin and IGF-1 levels, thereby decreasing receptor tyrosine kinase signaling to downstream targets. The anti-cancer effects of metformin may be caused by adenosine 5′-monophosphate (AMP)-activated protein kinase (AMPK)-dependent or -independent mechanism [63]. Metformin inhibits the mitochondrial complex I (NADH ubiquinone oxidoreductase) to interrupt mitochondrial respiration and induce a decrease of ATP production. ATP decreasing causes a reduction in the AMP/ATP ratio, which stimulates AMPK activation. The activated AMPK phosphorylates tuberous sclerosis complex protein 2 (TSC2) and inhibits mammalian target of rapamycin complex 1 (mTORC1). The suppression of the mTORC1 pathway decreases protein synthesis and cell growth. The activation of AMPK by metformin also induces a p53-dependent apoptosis and cell cycle inhibition through decreasing the cyclin D1 expression. Under specific conditions, metformin can also suppress mTORC1 signaling through AMPK-independent mechanisms through either inhibition of the Rag GTPases or induction of REDD expression [64,65]. Interestingly, in vivo studies show that metformin has a stronger antineoplastic activity in mice on a high-energy diet than in mice on a control diet [66]. This suggests that the insulinlowering action of metformin may contribute to its anti-neoplastic activity, and that it may have less impact on cancers in less hyperinsulinemic patients. The interplay between apoptosis and autophagy induced by metformin is another aspect for concerning its antitumor efficacy. Metformin has been reported to inhibit melanoma and lymphoma development through induction of autophagy and apoptosis. Here, apoptosis is considered as a consequence of autophagy [67]. In another study, metformin was found to exert antineoplastic effects on esophageal squamous cell carcinoma cells in vitro and invivo through Stat3 inactivation and Bcl-2 repression, facilitating crosstalk between apoptosis and autophagy [68]. Recent evidence indicates that a small proportion of cells in human cancers are cancer stem cells (CSCs), which confer tumor metastasis and therapy resistance. Metformin have been reported to markedly improved the response to conventional chemotherapy drugs by eradicating CSCs in many cancer types (Box 1). Regarding the accumulating basic and epidemiologic evidence about the antineoplastic activity of metformin, more than 250 ongoing clinical trials are registered on www.clinicaltrials.gov, assessing the potential of metformin as an adjuvant or neoadjuvant chemotherapy agent or as an enhancer of classic chemotherapy against almost all types of cancer. However, a recently published randomized controlled clinical trial shows that addition of a conventional anti-diabetic dose of metformin does not improve outcome in patients with advanced pancreatic cancer treated with gemcitabine and erlotinib [77]. In this study, the conventional anti-diabetic doses of metformin

Box 1. Metformin and cancer stem cells. Hirsch et al. first reported metformin’s specific action against breast cancer stem cells [69]. The authors demonstrated that combination of metformin and doxorubicin reduces tumor mass and prevents relapse much more effectively than either drug alone in a xenograft mouse model. Metformin can also synergistically interact with trastuzumab, an anti-HER2 monoclonal antibody to suppress selfrenewal and proliferation of cancer stem/progenitor cells in HER2-positive carcinomas [70]. In pancreatic cancer cells, metformin significantly decreases cell survival, clonogenicity, wound-healing capacity, sphere-forming capacity (pancreatospheres), and increases disintegration of pancreatospheres in both gemcitabine-sensitive and gemcitabine-resistant pancreatic cancer cells [71]. Other studies provide evidence of metformin in targeting of ovarian, lung and prostate cancer stem cells and enhancing the chemosensitivity and radiosensitivity of these cancer cells [72–74]. The cancer stem cell targeting effect of metformin can be summarized as: (1) activation of AMPK leading to inactivation of mTOR and suppression of its downstream effectors such as p70 ribosomal S6 kinase1 (p70S6K1) and Eukaryotic translation initiation factor 4E-binding protein 1 (4EBP1) [74]; (2) inhibit the progression of TGF-β-induced epithelialto-mesenchymal transition (EMT) by retaining the expression of E-cadherin and preventing concurrent appearance of vimentin expression [75]; (3) inducing the expression of miR-26a and other specific miRNAs [71]; (4) inhibiting the inflammatory pathway necessary for transformation and CSC formation [76]. These studies suggest that metformin can be useful for overcoming therapeutic resistance of many cancers.

F. Hua et al./Cancer Letters 374 (2016) 54–61

59

might fail to accumulate to a sufficient concentration to cause energetic stress. In the future research, the dose of metformin; other biguanides, such as phenformin which have pharmacokinetic advantages over metformin; and patients selection criteria should be carefully taken into consideration [78].

Adverse Event Reporting System (AERS) database also suggest the increased risk of acute pancreatitis and pancreatic cancer with GLP-1 receptor agonist and DPP-4 inhibitor use. However, these findings are not supported by population-based observational studies, suggesting long term follow-up, are needed [88,89].

Thiazolidinediones

Insulin and insulin analogs

Thiazolidinediones (TZDs) act as insulin sensitizers by virtue of their effect on peroxisome proliferator activated receptor-γ (PPARγ), which do not increase insulin secretion directly or cause hypoglycemia when used alone. Two drugs in this class, pioglitazone and rosiglitazone, are currently used in clinical cases. In vitro studies suggest some anti-cancer properties of these agents, such as inhibiting cell growth, inducing apoptosis and cell differentiation. However, some data suggest PPAR agonists may associate with increased cancer risk and potentiate tumorigenesis. A nationwide, population-based, case–control study using the Taiwan National Health Insurance Research Database reported that the use of TZDs may be associated with a decreased risk of colorectal cancer in patients with diabetes [79]. A population-based case–control study suggested that TZD derivatives used in T2DM patients reduces gastric cancer occurrence [80]. However, meta-analysis, populationbased cohort study and nested case–control study all revealed that Long-term exposures to pioglitazone and rosiglitazone were associated with higher odds of bladder cancer [81–83]. Despite the distinct effect of TZDs in bladder cancer and other cancer types, its anti-tumor effect in digestive organs was summarized as: (1) TZDs induce inhibition of the ubiquitin-proteasome system and/or MEK– ERK signaling, which result in the cell growth arrest through increased level of p27Kip1; (2) TZDs induce apoptosis through increased levels of apoptotic molecules, such as p53 and phosphatase and tensin homolog (PTEN) and/or decreased level of anti-apoptotic molecules, such as Bcl-2 and survivin; (3) TZDs induce inhibition of cancer cell invasion through inhibiting of MEK–ERK signalingmediated up-regulation of E-cadherin and claudin-4, and/or decreased expression of matrix metalloproteinases (MMPs) such as MMP-2 and MMP-9 [84].

Insulin is always necessary for patients with type I diabetes. It is also used for many patients with T2MD to achieve the ideal glycemic control targets. Several epidemiologic studies suggest a possible association of insulin therapy with increased risk of cancer [22]. Subcutaneous injection of insulin results in high levels of circulating insulin thereby increasing the risk of cancer associated with hyperinsulinemia. Insulin glargine is widely used long-acting recombinant human insulin analog. In 2009, four papers published simultaneously in Diabetologia with conflicting results [85,90–92] made people focus attention on the potential relationship of insulin glargine and the increased incidence of cancer. Recently a systematic review and meta-analysis of published cohort and case– control studies examined the risk of many type of cancer associated with use of exogenous human insulin or insulin analogs in diabetic patients. This study revealed that exogenous insulin exposure is associated with an increased risk of cancer in pancreas, liver, kidney, stomach and respiratory system and decreased risk of prostate cancer, when compared to no insulin use. Furthermore, a decreased risk of colon cancer as well as a marginally significant increased risk of breast cancer was observed in users of glargine insulin compared to users of non-glargine insulin [93]. Potential mechanisms by which administration of insulin or insulin analogs might affect neoplastic disease include several aspects: (1) insulin glargine has much higher affinity to the IGF-1 receptor, and higher mitogenic potency, than human insulin or other analogs; (2) pharmacokinetics is another issue that should be considered. Longacting analogs have a slower dissociation rate than that of native insulin (about 1.5–3 times longer); (3) the growth-promoting effect of insulin analogs on malignant cells due to its mitogenic effects [66]. Conclusions and further research

Insulin secretagogues and incretin-based therapies Secretagogues, such as sulfonylureas and the rapid-acting glinides, stimulate β-cells to release insulin. Sulfonylureas have been used to treat T2DM for more than 50 years. Clinical studies and epidemiological investigations have associated sulfonylureas with an increased risk of cancer. A retrospective cohort study of 62 809 people with diabetes revealed that sulfonylureas significantly increased the risk of developing solid tumor (HR: 1.36; 95% CI: 1.19– 1.54; p < 0.001), especially pancreatic cancer (HR: 4.95; 95% CI: 2.74– 8.96; p < 0.001) [85]. Although the association between sulfonylureas and cancer risk is genuine, the mechanisms still need further investigations. GLP-1 is an incretin hormone released after meals by L cells in the ileum. It increases the secretion of insulin from the pancreas in a glucose-dependent manner, suppresses the secretion of glucagon and delay gastric emptying. There are two GLP-1mimetic drugs currently approved for clinical use to treat type-2 diabetes, that is, exenatide and liraglutide. Inhibitors of dipeptidyl peptidase 4, also DPP-4 inhibitors are a class of oral hypoglycemics that block DPP-4. They can be used to treat diabetes mellitus type 2. DPP4 inhibitors work by blocking dipeptidyl peptidase IV (DPP4), an enzyme that breaks down gut peptides, especially GLP-1. DPP-4 inhibitors work indirectly to raise GLP-1. Blocking DPP-4 makes GLP-1 levels rise and increases insulin release after meals and when glucose levels are high. Current studies suggested a probable role of GLP1R activation on the development of pancreatic cancer and thyroid cancer [86,87]. Data from the Food and Drug Administration (FDA)

Numerous epidemiologic studies report a direct connection of cancers in the liver, pancreas, colorectum, endometrium, breast (in postmenopausal women), kidney, and biliary tract with type 2 diabetes mellitus. From a biological perspective, the increase in the risk of cancer is likely related to the interplay between obesity, inflammation, hyperinsulinemia and hyperglycemia. Increases in circulating levels of glucose and/or insulin could affect malignant transformation or tumor growth through effects on cellular energy metabolism, ER stress, dysfunction of autophagy/ubiquitinproteasomal system and by increased levels of bioactive IGF-1 induced by insulin. Different anti-diabetic medications may modify cancer risk and mortality in patients with diabetes. Generally speaking, insulin sensitizing drugs, such as metformin and TZDs are associated with a lower risk of cancer incidence. On the other hand, treatments that will result increase of endogenous level of insulin (insulin secretagogues and GLP-1-based therapies) or exogenous insulin or insulin analogs, are associated with an increase in risk of cancer. Despite the obtained evidence and the critical role of insulin-IGF axis in carcinogenesis and tumor progression, many large clinical trials involving patients with adult tumors, including nonsmall cell lung cancer, breast cancer, and pancreatic cancer, failed to show clinical benefit in the overall patient population as expected [28]. These facts indicate the complexity of the mechanisms in the connection of diabetes to cancer. Our recently published data document that antagonism of insulin/IGF signaling producing an unsatisfactory efficacy against cancer may be attributed to the high

60

F. Hua et al./Cancer Letters 374 (2016) 54–61

expression level of TRB3 induced by a diversity of metabolic stresses surrounding cancer cells and tumor tissue. TRB3 links insulinIGF1 to cancer development and progression through interacting with P62 via suppressing autophagic and proteasomal degradation. Indeed, interrupting the TRB3-P62 interaction with an α-helical peptide derived from P62 attenuates tumor growth and metastasis through activating autophagic flux, the therapeutic effect is more significant in the diabetic mice than in the nondiabetic ones. This study remind us that: (1) interrupting the TRB3-P62 interaction may provide a potential strategy against cancers in patients with diabetes; (2) maintaining an unrestricted and effective autophagic flux, together with moderately activated autophagic signaling are essential for tumor inhibition and minimize the cancer risk of some diabetes treatments. Funding This work was supported by grants from National Natural Science Foundation of China (81273529 to ZWH; 81472717 to FH). FH was also supported by grants from the Basic Research Program of the Peking Union Medical College (33320140070) and from the Basic Research Program of Institute of Materia Medica (2014ZD01). Conflict of interest The authors declare that they have no conflict of interests. References [1] M. Jalving, J.A. Gietema, J.D. Lefrandt, et al., Metformin: taking away the candy for cancer?, Eur. J. Cancer 46 (13) (2010) 2369–2380. [2] E. Giovannucci, D.M. Harlan, M.C. Archer, et al., Diabetes and cancer: a consensus report, CA Cancer J. Clin. 60 (4) (2010) 207–221. [3] H. Yuhara, C. Steinmaus, S.E. Cohen, et al., Is diabetes mellitus an independent risk factor for colon cancer and rectal cancer?, Am. J. Gastroenterol. 106 (11) (2011) 1911–1921. [4] V.A. Grote, S. Becker, R. Kaaks, Diabetes mellitus type 2 – an independent risk factor for cancer?, Exp. Clin. Endocrinol. Diabetes 118 (1) (2010) 4–8. [5] T. Stocks, K. Rapp, T. Bjorge, et al., Blood glucose and risk of incident and fatal cancer in the metabolic syndrome and cancer project (me-can): analysis of six prospective cohorts, PLoS Med. 6 (12) (2009) e1000201. [6] G. Nagel, T. Stocks, D. Spath, A. Hjartåker, et al., Metabolic factors and blood cancers among 578,000 adults in the metabolic syndrome and cancer project (Me-Can), Ann. Hematol. 91 (10) (2012) 1519–1531. [7] B. Lindkvist, M. Almquist, T. Bjorge, et al., Prospective cohort study of metabolic risk factors and gastric adenocarcinoma risk in the Metabolic Syndrome and Cancer Project (Me-Can), Cancer Causes Control 24 (1) (2013) 107–116. [8] C. Haggstrom, K. Rapp, T. Stocks, et al., Metabolic factors associated with risk of renal cell carcinoma, PLoS ONE 8 (2) (2013) e57475. [9] W. Borena, S. Strohmaier, A. Lukanova, et al., Metabolic risk factors and primary liver cancer in a prospective study of 578,700 adults, Int. J. Cancer 131 (1) (2012) 193–200. [10] D. Johansen, T. Stocks, H. Jonsson, et al., Metabolic factors and the risk of pancreatic cancer: a prospective analysis of almost 580,000 men and women in the Metabolic Syndrome and Cancer Project, Cancer Epidemiol. Biomarkers Prev. 19 (9) (2010) 2307–2317. [11] C. Wang, X. Wang, G. Gong, Q. Ben, W. Qiu, Y. Chen, et al., Increased risk of hepatocellular carcinoma in patients with diabetes mellitus: a systematic review and meta-analysis of cohort studies, Int. J. Cancer 130 (7) (2012) 1639–1648. [12] P. Batabyal, S. Vander Hoorn, C. Christophi, et al., Association of diabetes mellitus and pancreatic adenocarcinoma: a meta-analysis of 88 studies, Ann. Surg. Oncol. 21 (7) (2014) 2453–2462. [13] T. Salvatore, R. Marfella, M.R. Rizzo, et al., Pancreatic cancer and diabetes: a two-way relationship in the perspective of diabetologist, Int. J. Surg. 21 (Suppl. 1) (2015) S72–S77. [14] F. Ehehalt, D. Sturm, M. Rosler, et al., Blood glucose homeostasis in the course of partial pancreatectomy – evidence for surgically reversible diabetes induced by cholestasis, PLoS ONE 10 (8) (2015) e0134140. [15] J. Prawitt, S. Caron, B. Staels, Bile acid metabolism and the pathogenesis of type 2 diabetes, Current Diab. Rep. 11 (3) (2011) 160–166. [16] G. Aggarwal, V. Ramachandran, N. Javeed, et al., Adrenomedullin is up-regulated in patients with pancreatic cancer and causes insulin resistance in beta cells and mice, Gastroenterology 143 (6) (2012) 1510–1517, e1. [17] N. Javeed, G. Sagar, S.K. Dutta, et al., Pancreatic cancer-derived exosomes cause paraneoplastic beta-cell dysfunction, Clin. Cancer Res. 21 (7) (2015) 1722–1733. [18] P. Boyle, M. Boniol, A. Koechlin, et al., Diabetes and breast cancer risk: a meta-analysis, Br. J. Cancer 107 (9) (2012) 1608–1617.

[19] E. Friberg, N. Orsini, C.S. Mantzoros, et al., Diabetes mellitus and risk of endometrial cancer: a meta-analysis, Diabetologia 50 (7) (2007) 1365– 1374. [20] D. Bansal, A. Bhansali, G. Kapil, et al., Type 2 diabetes and risk of prostate cancer: a meta-analysis of observational studies, Prostate Cancer Prostatic Dis. 16 (2) (2013) 151–158, S151. [21] J. Ma, H. Li, E. Giovannucci, et al., Prediagnostic body-mass index, plasma C-peptide concentration, and prostate cancer-specific mortality in men with prostate cancer: a long-term survival analysis, Lancet Oncol. 9 (11) (2008) 1039–1047. [22] L. Wu, J. Zhu, L.J. Prokop, et al., Pharmacologic therapy of diabetes and overall cancer risk and mortality: a meta-analysis of 265 studies, Sci. Rep. 5 (2015) 10147. [23] W.W. Chang, R.J. Lin, J. Yu, et al., The expression and significance of insulin-like growth factor-1 receptor and its pathway on breast cancer stem/progenitors, Breast Cancer Res. 15 (3) (2013) R39. [24] R. Slaaby, L. Schaffer, I. Lautrup-Larsen, et al., Hybrid receptors formed by insulin receptor (IR) and insulin-like growth factor I receptor (IGF-IR) have low insulin and high IGF-1 affinity irrespective of the IR splice variant, J. Biol. Chem. 281 (36) (2006) 25869–25874. [25] Y. Fierz, R. Novosyadlyy, A. Vijayakumar, et al., Insulin-sensitizing therapy attenuates type 2 diabetes-mediated mammary tumor progression, Diabetes 59 (3) (2010) 686–693. [26] H. Zhang, D.H. Fagan, X. Zeng, et al., Inhibition of cancer cell proliferation and metastasis by insulin receptor downregulation, Oncogene 29 (17) (2010) 2517–2527. [27] B. Arcidiacono, S. Iiritano, A. Nocera, et al., Insulin resistance and cancer risk: an overview of the pathogenetic mechanisms, Exp. Diabetes Res. 2012 (2012) 789174. [28] P. Singh, J.M. Alex, F. Bast, Insulin receptor (IR) and insulin-like growth factor receptor 1 (IGF-1R) signaling systems: novel treatment strategies for cancer, Med. Oncol. 31 (1) (2014) 805. [29] H. Cucak, L.G. Grunnet, A. Rosendahl, Accumulation of M1-like macrophages in type 2 diabetic islets is followed by a systemic shift in macrophage polarization, J. Leukoc. Biol. 95 (1) (2014) 149–160. [30] K.M. Nieman, I.L. Romero, B. Van Houten, et al., Adipose tissue and adipocytes support tumorigenesis and metastasis, Biochim. Biophys. Acta 1831 (10) (2013) 1533–1541. [31] Y. Imai, A.D. Dobrian, M.A. Morris, et al., Islet inflammation: a unifying target for diabetes treatment?, Trends Endocrinol. Metab. 24 (7) (2013) 351–360. [32] A. Hakansson, G. Molin, Gut microbiota and inflammation, Nutrients 3 (6) (2011) 637–682. [33] D.H. Dapito, A. Mencin, G.Y. Gwak, et al., Promotion of hepatocellular carcinoma by the intestinal microbiota and TLR4, Cancer Cell 21 (4) (2012) 504–516. [34] S. Yoshimoto, T.M. Loo, K. Atarashi, et al., Obesity-induced gut microbial metabolite promotes liver cancer through senescence secretome, Nature 499 (7456) (2013) 97–101. [35] R.C. Travis, T.J. Key, Oestrogen exposure and breast cancer risk, Breast Cancer Res. 5 (5) (2003) 239–247. [36] T.J. Key, P.N. Appleby, G.K. Reeves, et al., Endogenous Hormones Breast Cancer Collaborative, body mass index, serum sex hormones, and breast cancer risk in postmenopausal women, J. Natl Cancer Inst. 95 (16) (2003) 1218–1226. [37] E. Weiderpass, K. Brismar, R. Bellocco, et al., Serum levels of insulin-like growth factor-I, IGF-binding protein 1 and 3, and insulin and endometrial cancer risk, Br. J. Cancer 89 (9) (2003) 1697–1704. [38] S. Pavlides, D. Whitaker-Menezes, R. Castello-Cros, et al., The reverse Warburg effect: aerobic glycolysis in cancer associated fibroblasts and the tumor stroma, Cell Cycle 8 (23) (2009) 3984–4001. [39] K.M. Nieman, H.A. Kenny, C.V. Penicka, et al., Adipocytes promote ovarian cancer metastasis and provide energy for rapid tumor growth, Nat. Med. 17 (11) (2011) 1498–1503. [40] C. Cheng, P. Ru, F. Geng, et al., Glucose-mediated N-glycosylation of SCAP is essential for SREBP-1 activation and tumor growth, Cancer Cell 28 (5) (2015) 569–581. [41] C. Dong, T. Yuan, Y. Wu, et al., Loss of FBP1 by Snail-mediated repression provides metabolic advantages in basal-like breast cancer, Cancer Cell 23 (3) (2013) 316–331. [42] D. Brasacchio, J. Okabe, C. Tikellis, et al., Hyperglycemia induces a dynamic cooperativity of histone methylase and demethylase enzymes associated with gene-activating epigenetic marks that coexist on the lysine tail, Diabetes 58 (5) (2009) 1229–1236. [43] S.G. Fonseca, J. Gromada, F. Urano, Endoplasmic reticulum stress and pancreatic beta-cell death, Trends Endocrinol. Metab. 22 (7) (2011) 266–274. [44] N. Kawasaki, R. Asada, A. Saito, et al., Obesity-induced endoplasmic reticulum stress causes chronic inflammation in adipose tissue, Sci. Rep. 2 (2012) 799. [45] X. Xing, Y. Li, H. Liu, et al., Glucose regulated protein 78 (GRP78) is overexpressed in colorectal carcinoma and regulates colorectal carcinoma cell growth and apoptosis, Acta Histochem. 113 (8) (2011) 777–782. [46] H.C. Zheng, H. Takahashi, X.H. Li, et al., Overexpression of GRP78 and GRP94 are markers for aggressive behavior and poor prognosis in gastric carcinomas, Hum. Pathol. 39 (7) (2008) 1042–1049. [47] K.L. Cook, R. Clarke, Role of GRP78 in promoting therapeutic-resistant breast cancer, Future Med. Chem. 7 (12) (2015) 1529–1534. [48] C. Roller, D. Maddalo, The molecular chaperone GRP78/BiP in the development of chemoresistance: mechanism and possible treatment, Front. Pharmacol. 4 (2013) 10.

F. Hua et al./Cancer Letters 374 (2016) 54–61

[49] C.C. Chiu, L.Y. Lee, Y.C. Li, et al., Grp78 as a therapeutic target for refractory head-neck cancer with CD24(−)CD44(+) stemness phenotype, Cancer Gene Ther. 20 (11) (2013) 606–615. [50] M. Thon, T. Hosoi, M. Yoshii, et al., Leptin induced GRP78 expression through the PI3K-mTOR pathway in neuronal cells, Sci. Rep. 4 (2014) 7096. [51] P.L. Mote, J.B. Tillman, S.R. Spindler, Glucose regulation of GRP78 gene expression, Mech. Ageing Dev. 104 (2) (1998) 149–158. [52] E. Chevet, C. Hetz, A. Samali, Endoplasmic reticulum stress-activated cell reprogramming in oncogenesis, Cancer Discov. 5 (6) (2015) 586–597. [53] Y. Gao, D.J. Sartori, C. Li, et al., PERK is required in the adult pancreas and is essential for maintenance of glucose homeostasis, Mol. Cell. Biol. 32 (24) (2012) 5129–5139. [54] J.R. Hassler, D.L. Scheuner, S. Wang, et al., The IRE1alpha/XBP1s pathway is essential for the glucose response and protection of beta cells, PLoS Biol. 13 (10) (2015) e1002277. [55] X.H. Liang, S. Jackson, M. Seaman, et al., Induction of autophagy and inhibition of tumorigenesis by beclin 1, Nature 402 (6762) (1999) 672–676. [56] A. Takamura, M. Komatsu, T. Hara, et al., Autophagy-deficient mice develop multiple liver tumors, Genes Dev. 25 (8) (2011) 795–800. [57] X. Qu, J. Yu, G. Bhagat, et al., Promotion of tumorigenesis by heterozygous disruption of the beclin 1 autophagy gene, J. Clin. Invest. 112 (12) (2003) 1809–1820. [58] P. Codogno, A.J. Meijer, Autophagy and signaling: their role in cell survival and cell death, Cell Death Differ. 12 (Suppl. 2) (2005) 1509–1518. [59] J. Kim, M. Kundu, B. Viollet, et al., AMPK and mTOR regulate autophagy through direct phosphorylation of Ulk1, Nat. Cell Biol. 13 (2) (2011) 132–141. [60] C. Mammucari, G. Milan, V. Romanello, et al., FoxO3 controls autophagy in skeletal muscle in vivo, Cell Metab. 6 (6) (2007) 458–471. [61] H.Y. Liu, J. Han, S.Y. Cao, et al., Hepatic autophagy is suppressed in the presence of insulin resistance and hyperinsulinemia: inhibition of FoxO1-dependent expression of key autophagy genes by insulin, J. Biol. Chem. 284 (45) (2009) 31484–31492. [62] F. Hua, K. Li, J.J. Yu, et al., TRB3 links insulin/IGF to tumour promotion by interacting with p62 and impeding autophagic/proteasomal degradations, Nat. Commun. 6 (2015) 7951. [63] B.J. Quinn, H. Kitagawa, R.M. Memmott, et al., Repositioning metformin for cancer prevention and treatment, Trends Endocrinol. Metab. 24 (9) (2013) 469–480. [64] I. Ben Sahra, C. Regazzetti, G. Robert, et al., Metformin, independent of AMPK, induces mTOR inhibition and cell-cycle arrest through REDD1, Cancer Res. 71 (13) (2011) 4366–4372. [65] A. Kalender, A. Selvaraj, S.Y. Kim, et al., Metformin, independent of AMPK, inhibits mTORC1 in a rag GTPase-dependent manner, Cell Metab. 11 (5) (2010) 390–401. [66] L. Sciacca, R. Le Moli, R. Vigneri, Insulin analogs and cancer, Front. Endocrinol. (Lausanne) 3 (2012) 21. [67] T. Tomic, T. Botton, M. Cerezo, et al., Metformin inhibits melanoma development through autophagy and apoptosis mechanisms, Cell Death Dis. 2 (2011) e199. [68] Y. Feng, C. Ke, Q. Tang, et al., Metformin promotes autophagy and apoptosis in esophageal squamous cell carcinoma by downregulating Stat3 signaling, Cell Death Dis. 5 (2014) e1088. [69] H.A. Hirsch, D. Iliopoulos, P.N. Tsichlis, et al., Metformin selectively targets cancer stem cells, and acts together with chemotherapy to block tumor growth and prolong remission, Cancer Res. 69 (19) (2009) 7507–7511. [70] A. Vazquez-Martin, C. Oliveras-Ferraros, S. Del Barco, et al., The anti-diabetic drug metformin suppresses self-renewal and proliferation of trastuzumabresistant tumor-initiating breast cancer stem cells, Breast Cancer Res. Treat. 126 (2) (2011) 355–364. [71] B. Bao, Z. Wang, S. Ali, et al., Metformin inhibits cell proliferation, migration and invasion by attenuating CSC function mediated by deregulating miRNAs in pancreatic cancer cells, Cancer Prev. Res. (Phila) 5 (3) (2012) 355–364.

61

[72] J.J. Shank, K. Yang, J. Ghannam, et al., Metformin targets ovarian cancer stem cells in vitro and in vivo, Gynecol. Oncol. 127 (2) (2012) 390–397. [73] D. Iliopoulos, H.A. Hirsch, K. Struhl, Metformin decreases the dose of chemotherapy for prolonging tumor remission in mouse xenografts involving multiple cancer cell types, Cancer Res. 71 (9) (2011) 3196–3201. [74] C.W. Song, H. Lee, R.P. Dings, et al., Metformin kills and radiosensitizes cancer cells and preferentially kills cancer stem cells, Sci. Rep. 2 (2012) 362. [75] S. Cufi, A. Vazquez-Martin, C. Oliveras-Ferraros, et al., Metformin against TGFbeta-induced epithelial-to-mesenchymal transition (EMT): from cancer stem cells to aging-associated fibrosis, Cell Cycle 9 (22) (2010) 4461–4468. [76] H.A. Hirsch, D. Iliopoulos, K. Struhl, Metformin inhibits the inflammatory response associated with cellular transformation and cancer stem cell growth, Proc. Natl. Acad. Sci. U.S.A. 110 (3) (2013) 972–977. [77] S. Kordes, M.N. Pollak, A.H. Zwinderman, et al., Metformin in patients with advanced pancreatic cancer: a double-blind, randomised, placebo-controlled phase 2 trial, Lancet Oncol. 16 (7) (2015) 839–847. [78] M. Pollak, Metformin and pancreatic cancer: a clue requiring investigation, Clin. Cancer Res. 18 (10) (2012) 2723–2725. [79] S.W. Chen, Y.T. Tsan, J.D. Chen, et al., Health data analysis in Taiwan research, use of thiazolidinediones and the risk of colorectal cancer in patients with diabetes: a nationwide, population-based, case-control study, Diabetes Care 36 (2) (2013) 369–375. [80] S.S. Chang, H.Y. Hu, Association of thiazolidinediones with gastric cancer in type 2 diabetes mellitus: a population-based case-control study, BMC Cancer 13 (2013) 420. [81] F.Y. Hsiao, P.H. Hsieh, W.F. Huang, et al., Risk of bladder cancer in diabetic patients treated with rosiglitazone or pioglitazone: a nested case-control study, Drug Saf. 36 (8) (2013) 643–649. [82] A. Neumann, A. Weill, P. Ricordeau, et al., Pioglitazone and risk of bladder cancer among diabetic patients in France: a population-based cohort study, Diabetologia 55 (7) (2012) 1953–1962. [83] S. He, Y.H. Tang, G. Zhao, et al., Pioglitazone prescription increases risk of bladder cancer in patients with type 2 diabetes: an updated meta-analysis, Tumour Biol. 35 (3) (2014) 2095–2102. [84] T. Okumura, Mechanisms by which thiazolidinediones induce anti-cancer effects in cancers in digestive organs, J. Gastroenterol. 45 (11) (2010) 1097–1102. [85] C.J. Currie, C.D. Poole, E.A. Gale, The influence of glucose-lowering therapies on cancer risk in type 2 diabetes, Diabetologia 52 (9) (2009) 1766–1777. [86] M. Elashoff, A.V. Matveyenko, B. Gier, et al., Pancreatitis, pancreatic, and thyroid cancer with glucagon-like peptide-1-based therapies, Gastroenterology 141 (1) (2011) 150–156. [87] R. Rouse, L. Xu, S. Stewart, et al., High fat diet and GLP-1 drugs induce pancreatic injury in mice, Toxicol. Appl. Pharmacol. 276 (2) (2014) 104–114. [88] E.A. Suarez, C.E. Koro, J.B. Christian, et al., Incretin-mimetic therapies and pancreatic disease: a review of observational data, Curr. Med. Res. Opin. 30 (12) (2014) 2471–2481. [89] M. Gokhale, J.B. Buse, C.L. Gray, et al., Dipeptidyl-peptidase-4 inhibitors and pancreatic cancer: a cohort study, Diabetes Obes. Metab. 16 (12) (2014) 1247–1256. [90] L.G. Hemkens, U. Grouven, R. Bender, et al., Risk of malignancies in patients with diabetes treated with human insulin or insulin analogues: a cohort study, Diabetologia 52 (9) (2009) 1732–1744. [91] J.M. Jonasson, R. Ljung, M. Talback, et al., Insulin glargine use and short-term incidence of malignancies-a population-based follow-up study in Sweden, Diabetologia 52 (9) (2009) 1745–1754. [92] H.M. Colhoun, S.E. Group, Use of insulin glargine and cancer incidence in Scotland: a study from the Scottish Diabetes Research Network Epidemiology Group, Diabetologia 52 (9) (2009) 1755–1765. [93] O. Karlstad, J. Starup-Linde, P. Vestergaard, et al., Use of insulin and insulin analogs and risk of cancer – systematic review and meta-analysis of observational studies, Curr. Drug Saf. 8 (5) (2013) 333–348.