Diabets and the incidence of cancer in a Portuguese urban cohort

Susana Raquel Moura Barros Diabetes and the incidence of cancer in a Portuguese urban cohort Porto, 2012 ii Diabetes and the incidence of cancer in a Portuguese urban cohort Dissertação de candidatura ao grau de Mestre em Saúde Pública apresentada à Faculdade de Medicina e ao Instituto de Ciências Biomédicas Abel Salazar da Universidade do Porto. Susana Barros Porto, 2012 iii Este trabalho foi realizado no Departamento de Epidemiologia Clínica, Medicina Preditiva e Saúde Pública da Faculdade de Medicina da Universidade do Porto e no Instituto de Saúde Pública da Universidade do Porto (ISPUP), sob orientação da Doutora Bárbara Peleteiro e co-orientação do Professor Doutor Nuno Lunet. Neste trabalho colaborei na operacionalização das hipóteses em estudo, na análise dos dados e na interpretação dos resultados. iv Agradecimentos À minha família, agradeço pelo apoio, paciência e carinho. Pelos valores incutidos, pelo exemplo e pela confiança em mim depositada. Um sincero e profundo Obrigada! À Doutora Bárbara Peleteiro, orientadora, agradeço o apoio desde o primeiro dia, pela confiança, compreensão e tolerância nos momentos difíceis atravessados. Um suporte fundamental para a realização da dissertação. Ao Professor Doutor Nuno Lunet, co-orientador, agradeço o conhecimento científico e a presença nas alturas certas. Aos Professores, com quem tive o privilégio de aprender ao longo do meu percurso académico, agradeço o estímulo e a transmissão de conhecimentos. Aos colegas do Mestrado em Saúde Publica com quem tive o privilégio de percorrer esta jornada, agradeço a convivência, a boa-disposição e a partilha. A todos os meus amigos, que não precisam de ser enumerados, agradeço a compreensão pela minha ausência, a força por vós inspirada, o carinho e a amizade neste meu percurso. Por último, mas não menos importante, um especial agradecimento à Helena Carreira, companheira e amiga de longa data, pela amizade, pelo alento e disponibilidade nos vários momentos. Mas também pela partilha de conhecimentos e pela ajuda técnica para a elaboração da dissertação. v Contents Resumo ................................................................................................................................ vi Abstract ................................................................................................................................ ix List of figures ........................................................................................................................ xii List of tables ........................................................................................................................ xiv List of abbreviations ............................................................................................................. xvi Background ........................................................................................................................... 0 Diabetes and cancer - definition, incidence and mortality ................................................... 1 Epidemiological findings regarding the association between diabetes and cancer ............. 2 Mechanisms for the link between diabetes and cancer ....................................................... 3 Main objective ...................................................................................................................... 12 Methods ............................................................................................................................... 14 Study participants ............................................................................................................. 15 Data collection .................................................................................................................. 15 Identification of cancer cases ........................................................................................... 17 Data analysis .................................................................................................................... 25 Ethics ............................................................................................................................... 26 Results ................................................................................................................................ 28 Discussion ........................................................................................................................... 34 Conclusion ........................................................................................................................... 42 References .......................................................................................................................... 45 vi Resumo vii Introdução: A diabetes mellitus e o cancro são duas doenças comuns, cuja incidência tem aumentado globalmente. Alguns estudos têm explorado uma possível associação entre ambas, mostrando uma maior frequência de cancro em indivíduos diabéticos do que em indivíduos da população geral. Vários mecanismos biológicos têm sido propostos para explicar esta associação: a diabetes mellitus e o cancro partilham vários factores de risco (modificáveis e não-modificáveis), que podem influenciar o processo neoplásico por vários mecanismos metabólicos, incluindo hiperinsulinemia, hiperglicemia e adiposidade, particularmente a obesidade abdominal, que está associada com um estado de inflamação crónico de baixo grau. Adicionalmente, estudos observacionais sugeriram que alguns fármacos usados para tratar a hiperglicemia estão associados quer com um maior quer com um menor risco de cancro. O principal objectivo deste estudo foi quantificar a associação entre diabetes mellitus tipo 2 e a incidência de cancro, numa coorte de indivíduos adultos de uma zona urbana Portuguesa. Métodos: Este estudo foi realizado no âmbito da coorte EPIPorto, constituída por uma amostra de 2485 habitantes não institucionalizados da cidade do Porto, com idades entre os 18 e os 92 anos, recrutados por aleatorização de dígitos telefónicos. Entrevistadores treinados efectuaram entrevistas presenciais, utilizando um questionário estruturado, para recolher informações sociodemográficas, clínicas e de estilos de vida, além de avaliações físicas, incluindo medições antropométricas. Colheram-se amostras de sangue venoso depois de um jejum de 12 horas durante a noite. Na avaliação inicial, os indivíduos foram classificados em: não diabéticos sem alterações no metabolismo da glicose se o valor da glicemia em jejum fosse abaixo de 1,10 g/L; não diabéticos com alterações no metabolismo da glicose, se o valor da glicemia em jejum se situasse entre 1,10 g/L e 1,25 g/L; ou diabéticos se o valor da glicemia em jejum fosse acima de 1,25 g/L, ou se estivessem a ser tratados com insulina ou anti-diabéticos orais, ou se reportassem um diagnóstico prévio de diabetes tipo 2. Foram elegíveis para este estudo os indivíduos sem um diagnóstico prévio de cancro e excluíram-se os indivíduos sem a informação necessária para a sua caracterização relativamente à principal exposição em estudo, tendo a amostra final sido constituída por 2035 indivíduos. Identificaram-se os indivíduos que posteriormente desenvolveram cancro (com excepção de melanoma da pele) através do Registo Oncológico Regional do Norte (RORENO). O tempo mediano de seguimento da coorte foi de 8,1 anos (percentil 25-75: 6,211,6). Para quantificar a associação entre diabetes e cancro, calcularam-se razões de taxas de incidência (RR) ajustadas para o sexo, idade e escolaridade, e os correspondentes viii intervalos de confiança a 95% (IC95%), utilizando a regressão de Poisson. Também se efectuaram análises estratificadas por exposição aos principais factores de risco para cancro, nomeadamente, excesso de peso, perímetro da cintura aumentado, tabagismo, etilismo, baixo consumo de vitamina C, fruta, e vegetais, além dos padrões alimentares. Resultados: Durante o período de seguimento da coorte, foram diagnosticados 115 tumores primários, correspondendo a uma taxa de incidência (TI) de 586/100000 pessoas-ano (IC95%: 488703). As topografias tumorais mais frequentes foram mama (16,5%), próstata (13,9%), cólon e recto (13,9%), estômago (7,0%), pulmão (7,0%), fígado (5,2%) e bexiga (5,2%). Comparativamente com os não diabéticos (TI=508/100000, IC95%: 413-626), os indivíduos não diabéticos mas com alterações do metabolismo da glicose medida em jejum apresentaram um risco aumentado de cancro (RR=1,99, IC95%: 1,04-3,83). Contudo, este resultado deixou de ser estatisticamente significativo após ajuste para o sexo, idade e escolaridade. O risco de cancro foi superior entre os indivíduos diabéticos, independentemente do sexo, idade e escolaridade (RR=2,04, IC95%: 1,18-3,51). Quando se considerou os participantes não diabéticos, com ou sem alterações do metabolismo da glicose em jejum, como categoria de referência, o risco de cancro diminuiu ligeiramente (RR=1,95, IC95%: 1,14-3,34). A diabetes associou-se significativamente com um risco aumentado de tumores do tracto gastrointestinal (RR=2,89, IC95%: 1,23-6,82), particularmente cancro colorectal (RR=3,34, IC95%: 1,05-10,66). Os cancros de mama (RR=2,49, IC95%: 0,69-8,99), estômago (RR=3,68, IC95%: 0,72-18,84), pulmão (RR=3,77, IC95%: 0,73-19,41), fígado (RR=2,17, IC95%: 0,24-19,37), tracto urinário (RR=2,57, IC95%: 0,54-12,14) e órgãos genitais femininos (RR=3,83, IC95%: 0,40-37,08) também apresentaram um risco aumentado entre os diabéticos e observou-se uma associação inversa para o cancro da próstata (RR=0,88, IC95%: 0,12-6,76). Não se verificaram diferenças significativas na associação entre diabetes e cancro entre os estratos de exposição aos principais factores de risco para as doenças oncológicas. Conclusão: A incidência de cancro foi superior nos indivíduos com diabetes do que nos não diabéticos, com associações de diferente magnitude de acordo com o tipo de cancro. Os indivíduos com alterações do metabolismo glicídico tiveram também um risco aumentado de cancro. Este estudo sugere que o desenvolvimento de programas de gestão da diabetes poderão passar a ter de incluir também uma especial atenção à prevenção primária e secundária de cancro nestes indivíduos. ix Abstract xvi List of abbreviations xvii DM Diabetes Mellitus T2D Type 2 diabetes T1D Type 1 diabetes IGT Impaired glucose tolerance IFG Impaired fasting glucose FBG Fasting blood glucose IDF International Diabetes Federation IGF Insulin-like growth factor IGF-1 Insulin-like growth factor 1 IGF-2 Insulin-like growth factor 2 IR Insulin receptor IR-A Insulin receptor isoform A IR-B Insulin receptor isoform B IGF-IR Insulin-like growth factor receptor IGFBP Insulin-like growth factor binding proteins GLUT Glucose transporter type BMI Body mass index MMSE Mini Mental State Examination OR Odds ratio RR Relative risk MetS Metabolic syndrome IRR Incidence rate ratio CI Confidence interval HCC Hepatocellular carcinoma Il-6 Interleukin-6 TZD Thiazolidinediones Background 1 1. Diabetes and cancer - definition, incidence and mortality Diabetes mellitus (DM) and cancer are common diseases that have a tremendous impact on health worldwide, and are associated with severe acute and chronic complications [1-3]. DM is recognized as a group of heterogeneous disorders with the common elements of hyperglycemia and glucose intolerance, due to insulin deficiency, impaired effectiveness of insulin action, or both. Diabetes is typically divided into 2 major subtypes, type 1 and type 2, along with less common types like gestational diabetes. Type 1 diabetes is sometimes called insulin-dependent, immune-mediated or juvenile-onset diabetes. It usually accounts for only a minority of the total burden of diabetes in a population, but its incidence is increasing in both developing and developed countries [4]. Type 2 diabetes (T2D) constitutes about 85 to 95% of all diabetes in high-income countries and may account for an even higher percentage in low-and middle-income countries. T2D is characterized by insulin resistance and relative insulin deficiency, either of which may be present at the time that diabetes becomes clinically manifested. The diagnosis of T2D usually occurs after the age of 40 years but can occur earlier, especially in populations with high diabetes prevalence, where the number of reports of children developing T2D is increasing. Impaired glucose tolerance (IGT), along with impaired fasting glucose (IFG), is recognized as being a stage before diabetes when blood glucose levels are higher than normal. Thus, people with IGT are at high risk of developing T2D [5]. Both diabetes and cancer are common diseases whose incidence is increasing globally. According to the International Diabetes Federation (IDF), about 366 million people worldwide, or 8.3% of adults between the ages of 20 and 79, were estimated to have diabetes in 2011. About 80% of these live in lowand middle-income countries [5]. If these trends continue, by 2030, about 552 million people, or one adult in 10, will have diabetes. This equates to approximately 9 million new cases diagnosed every year [1]. This prevalence is more than double the prevalence observed in the year 2000 [6]. The largest increases will take place in the regions dominated by developing economies [5]. In Portugal, the prevalence of self-reported diabetes and mean fasting glucose increased in the last two decades (1987-2009) [7]. In 2010, the PREVADIAB study found a diabetes prevalence of 11.7% with a significant difference between men (14.2%) and women (9.5%). This study evaluated a national sample of Portuguese subjects, and applied objective criteria to define diabetes. While 6.6% of subjects had previously been diagnosed with diabetes, 5.1% were undiagnosed. If ‘pre-diabetes’ is also considered, about one-third (34.9%) of the population aged 20–79 years is affected [8]. 2 Worldwide, the prevalence of cancer has been difficult to establish because many areas are not covered by cancer registries, but in 2008 there were an estimated 12.7 million new cancer cases diagnosed, of these 6.6 million cases were in men and 6.0 million in woman. This number is expected to increase to 21 million by 2030 [9]. The most commonly diagnosed cancers are those of the lung/bronchus, breast, and colorectal, whereas the most common causes of cancer deaths are lung, stomach, and liver cancer [2]. In the North region of Portugal, the most commonly diagnosed cancers were those of the prostate, colon, lung and stomach in men and the breast, colon, stomach and thyroid in women in 2007 [10]. Worldwide, cancer and diabetes are the 3rd and 11th leading cause of death, respectively [11]. Diabetes is a major global cause of premature mortality that is widely underestimated, because only a minority of persons with diabetes dies from a cause uniquely related to the condition [12]. Epidemiologic evidence suggests that people with diabetes are at a significantly higher risk of many forms of cancer [13-16]. 2. Epidemiological findings regarding the association between diabetes and cancer The association between diabetes and cancer was described long before the 21st century, as far back as 1885 [17]. Most studies were initiated after introduction of insulin treatment, which prolonged the patient´s life and therefore resulted in over-diagnoses of cancer cases. Some stages can be distinguished in the research of this association. The first substantial reports, based on post-mortem examinations, suggested an increased incidence of different cancers in diabetic patients. The second stage of the research focused on analyzing the association using more objective and thorough epidemiological methods. However, it provided only partial confirmation of the previous reports [17]. The most recent epidemiological research consisted of well-prepared and well-documented studies based on analysis of cancer diagnosis and mortality among diabetic patients and included both casecontrol and cohort studies as well meta-analysis [13, 14, 18-20]. Because of the different physiology between the two subtypes of diabetes, these findings cannot be extended to type 1 diabetes (T1D). There are few studies of T1D and cancer. One cohort study from Sweden showed modest excess risk for stomach, cervix and endometrial cancers in T1D [21]. Overall, 8% to 18% of individuals suffering from cancer also have diabetes. While the literature indicates a strong and consistent increased risk of cancer in people with T2D, the strength of association depends on the specific cancer site [15]. 3 The strongest relations have been demonstrated for cancers of liver, pancreas, endometrium and kidney and lesser for cancers of the colon/rectum, breast and bladder [14, 15, 19]. The evidence for others cancer is inconclusive or have a small number of studies (e. g. non-Hodgkin lymphoma, lung, ovarium) [14, 15]. However, diabetes is associated with a lower risk of prostate cancer [14, 22, 23]. Previous studies reported a difference in magnitude of cancer risk associated with diabetes between men and women, such as liver, colon and stomach cancer [18, 24]. These sex-specific differences in the association between diabetes and cancer require more careful analyses. The impact of T2D on the prognosis of patients with cancer has been studied less well than its impact on the risk of cancer. Results of some, but not all, epidemiological studies suggest that diabetes may significantly increase mortality in patients with cancer [3, 25-27]. For example, in one study, which included up to 50% of Swedish patients with T2D, of whom 16,123 were diagnosed with subsequent cancer, and 999,982 cancer patients without T2D, clearly revealed an impact of T2D on cancer mortality. After adjusting for the confounding factors, patients who had cancer with T2D experienced a significant increase in cause-specific mortality ranging from 1.11 to 1.80 for different types of cancer [26]. 3. Mechanisms for the link between diabetes and cancer Several hypotheses have been suggested to explain the potentially causal relationship between glycemic level and cancer (Figure 1). T2D and cancer share many risk factors, modifiable and non-modifiable. The environmental or lifestyle factors could contribute at the same time to metabolic factors underlying both T2D and cancer, including visceral adiposity and a consequent state of inflammation, hyperglycemia and hyperinsulinemia [3, 14, 20, 28]. Furthermore, insulin, glycemia, adiposity and inflammation can be modified by antidiabetic pharmacotherapy and so, a better understanding of their pathophysiologic links to cancer allow us to more effectively target T2D treatment [29, 30]. 4 Figure 1. The link between T2D and cancer. (a) The shared metabolic factors underlying both T2D and cancer, including visceral adiposity, inflammation, hyperglycemia, and hyperinsulinemia. (b) The metabolic and mitogeneic pathways that are activated through different molecules [16] 3.1 Common risk factors to diabetes and cancer It is known that both T2D and cancer increase with age; those aged ≥ 55 years comprise almost 80% of newly diagnosed cancers and 23.1% of patients aged ≥ 60 years develop T2M compared to 10.7% of younger adults [5]. Men have slightly more cancer and diabetes compared to women after adjusting for other risks [31]. African-Americans appear as a subgroup more prone to cancer-related deaths, T2D, and obesity compared to nonHispanic white counterparts [31, 32]. 5 T2D is strongly related to overweight and obesity, two conditions associated with an increased risk of several cancers [13]. The cancers most consistently associated with overweight and obesity are those of breast (in post menopausal women), colon/rectum, endometrium, pancreas, adenocarcinoma of the esophagus, kidney, gallbladder and liver [33]. Central obesity, a marker of insulin resistance and a key player in both T2D and the metabolic syndrome, has also been linked to breast, colorectal, liver, and endometrial malignancies [34]. A majority of studies, despite employing different study designs and populations, suggest that diets low in red and processed meat and high in vegetables, fruits, and whole grains are associated with a lower risk of many types of cancer [35]. Diets that are low in red and processed meat but high in monounsaturated fatty acids, fruits, vegetables, wholegrain cereals, and dietary fiber may protect against T2D, possibly through improving insulin sensitivity. Low-carbohydrate diets (which often include a greater consumption of red meats and fat) have also been associated with weight loss and improvements in insulin sensitivity and glycemic control [36]. Evidence from observational epidemiologic studies consistently shows that higher levels of physical activity are associated with a lower risk of the colon, postmenopausal breast, and endometrial cancer [35]. Prevention of diabetes with increased physical activity has also been proven by numerous trials [37, 38]. Moreover, tobacco and excess alcohol usage are linked to cancer and can also worsen diabetes complications [14, 35, 36, 39]. 3.2 Biological hypotheses proposed to link diabetes to cancer Diabetes may influence the neoplastic process by metabolic mechanisms, including hyperinsulinemia (either endogenous due to insulin resistance or exogenous due to administered insulin or insulin secretogogues), hyperglycemia and adiposity, particularly central obesity, creating a low-grade chronic inflammation state (Figure 1). Hyperinsulinemia The role of insulin in promoting cancer growth was first recognized by studies in experimental animals as early as 1972 [40]. In the genesis of T2D, reduced insulin sensitivity plays a key role, inducing compensatory hyperinsulinaemia with an increased level of circulating insulin-like growth 6 factors (IGF), well-know to stimulate cell proliferation and inhibit apoptosis in many organs including the liver, pancreas, colon, ovary, breast, which are the most frequent sites with an increased risk of cancer in T2D patients [14, 20]. Insulin stimulates cell growth and differentiation acting within a complex system of interrelated hormones, receptors and binding proteins, known as the insulin/IGF axis. This system includes insulin, insulin-like growth factors (IGF-I, IGF-II), insulin receptor in two isoforms (IR-A, IR-B), growth factors receptors (IGF-IR, IGF-IIR), hybrid receptors and insulin-like growth factor binding proteins (IGFBP-1 to 6) [41]. Both insulin and IGF-1 have affinity to both the insulin receptor (IR) and IGF-1 receptor (IGF-1R) due of similar structural homology [42]. These receptors, expressed in majority of cancer cells tend to have stronger mitogenic and antiapoptotic effects, and the hyperinsulinemia that occurs in insulin-resistant individuals may enhance this effect [14, 28]. Insulin may also indirectly promote cancer development via IGF-1. Insulin reduces the hepatic production of IGF-binding protein-1, and possibility IGF-binding protein-2, with resultant increases in the levels of circulating free, bioactive form of the growth factor, IGF-1 [43]. In addition, malignant cells predominantly express the A isoform of the IR (IR-A) or hybrid IR-A/IGF-1R forms, which have more mitogenic effects that the B isoform IR. The A receptor isoform can stimulate insulin-mediated mitogenesis, even in cells deficient in IGF-1 [43, 44]. Thus, in the hyperinsulinemic individual, insulin´s mitogenic properties may further oncogenic proliferation. Multiple signaling pathways are activated after insulin receptors or IGF-I receptors interact with their ligands. By phosphorylating adaptor proteins, most notably the insulin receptor substrate family, the initial kinase event is linked to downstream signaling pathways. Once activated, these signaling pathways may stimulate multiple cancer phenotypes including proliferation, protection from apoptotic stimuli, invasion, and metastasis, potentially enhancing the promotion and progression of many types of cancer cells. It is also clear that insulin/IGF may stimulate normal cells that are involved in cancer progression. All these effects might contribute to carcinogenesis [16, 45]. Hyperinsulinemia may ultimately upregulate the insulin mitogenic pathway compared to the insulin metabolic pathway. Hyperglycemia Hyperglycemia, one of the main characteristics of diabetes, is considered one possible reason for increased risk of cancer in diabetic patients [46]. Hyperglycemia may not 7 only promote tumorigenesis indirectly by stimulating insulin production, but also may have a direct effect, since cancer cells depend on glycolysis for energy. In order for cancer cells to maintain their intensive and uncontrolled proliferation, it requires a lot of energy and substrates. For these purposes, various substances can be used, e.g. glucose, whose input is ensured in tumor cells by an increase in the expression of glucose membrane transporters, especially forms of GLUT1, GLUT3 and GLUT12 [14] . Heavy consumption of glucose is characteristic of neoplastic cells but glucose metabolism proceeds differently than in the unmodified cells. In healthy cells, glucose is converted to pyruvate, and later ATP is formed. This is done with the participation of oxygen in the Krebs cycle, and taking place in the mitochondria. Cancer cells convert the pyruvates into lactic acid and use glucose for the synthesis of DNA, RNA, proteins, lipids, namely for the cell construction. This different metabolism of cancer cells leads to an increased production of lactate, despite the full availability of oxygen [14, 41]. This phenomenon, which was described in 1927 is called the Warburg effect [47]. The excess of lactic acid production lowers the pH of the extra cellular matrix, thus contributing to the death of normal tissue cells and to the increased activity of collagenases, which facilitate tumor cells migration [41]. Nevertheless, increased intake of sugar and refined carbohydrates has been positively correlated with the risk of cancer. One mechanism by which glucose could induce cancer progression is by the induction of oxidative stress. Inducing hyperglycemia in the human breast cancer cell line MDA MB-231 leads to increased expression of the oxidative stress-responsive gene, thioredoxin interacting protein, and subsequent increased levels of reactive oxygen species. Thus, hyperglycemia in obesity and T2D could accelerate tumor growth and progression. HbA1c is a test that measures the amount of glycated hemoglobin in blood, and gives a stable estimate of blood glucose control over the last 1 to 3 months. Inconsistent results were reported from previous observational studies on the relationship between HbA1c levels and cancer incidence or mortality. For example, Miao Jonasson et al performed a nationwide population-based observational study based on Swedish patient registers and they did not observed associations between HbA1c and incidence of all cancers or cancers of specific types in patients with type 2 diabetes [48]. Their results are not consistent with the Hong Kong study of T2D which was based on 973 new insulin users and 971 matched non-users of insulin. This study found that HbA1c per percentage was associated with a 1.24-fold increase in cancer risk [49]. Hyperglycemia, which is an independent risk factor for cancer, is usually associated with further metabolic changes - obesity, insulin resistance and hyperinsulinemia, which also contribute to the development of cancer. 14 Methods 15 Study participants The present investigation was based on the EPIPorto cohort. It comprises a representative sample of 2485 Portuguese adults (61.8% women), aged 18 to 92 years, residing in Porto, an urban center in northwest Portugal, recruited between 1999 and 2003. As previously described [74], the participants were recruited by random digit dialing using households as the sampling unit. Once a household was selected, all residents were identified by age and sex, and one resident (aged more than 17 years) was randomly selected as the respondent, without allowing a replacement if there was a refusal. A participation rate of 70 % was achieved [74]. A visit to the Department of Clinical Epidemiology, Predictive Medicine and Public Health was scheduled by telephone according to the participant’s convenience. A personal interview, using a structured questionnaire, and a physical examination was performed by trained interviewers. A venous blood sample was drawn after a 12-hour overnight fast. All the samples were analyzed at the central laboratory of the university hospital. In individuals aged above 64 years, a rapid evaluation of cognitive function was done using the Mini-Mental State Examination test (MMSE) [75], and subjects were excluded when scoring below than 24 points [76]. At baseline, subjects were classified as non-diabetic without impaired fasting blood glucose (FBG) if FBG was below 1.10 g/L, non-diabetic with impaired FBG if FBG was between 1.10 g/L and 1.25 g/L, or diabetic if FBG was above 1.25 g/L or if being treated with insulin or oral anti-diabetic drugs or if reported a previous diagnosis of T2D. Data collection A structured standard questionnaire was applied by trained interviewers to obtain data on socio-demographic characteristics, personal and family medical history, and behavioral features. Subjects were classified as daily smokers (current consumers of at least one cigarette per day, on average), occasional smokers (current consumers of less than one cigarette per day, on average), ex-smokers (not smoking for more than six months) and never-smokers [77]. Daily and occasional smokers were grouped into current smokers for data analysis. Regarding the consumption of alcoholic beverages, participants were classified as daily drinkers (current consumers of at least one alcoholic drink per day, on average), occasional drinkers (current consumers of less than an alcoholic drink per day, on average), ex-drinkers 16 (not drinking for more than six months) and never-drinkers. For analysis, daily and occasional drinkers were grouped into current drinkers. Daily ethanol intake during the previous year was estimated, and cut-offs for men and women were defined according to the proposed by the American Heart Association [78]. Dietary intake during the previous year was estimated using a validated semiquantitative food-frequency questionnaire [79, 80], comprising 82 food item/group and beverage categories. For each item, participants were asked to indicate the average frequency of consumption (nine possible responses, ranging from never to six or more times per day), as well as the amount consumed (using a photograph manual with three portion sizes, small, medium and large), and the number of months during which the foods had been consumed in the previous year. Any foods that were not specified in the food-frequency questionnaire but eaten regularly (once a week or more frequently) were listed in an open section. The consumption of fruit and vegetables intake was estimated by adding up the frequencies of consumption of medium servings, corrected for seasonality, of the items referring to different fruits (apple or pear, orange or mandarin, banana, kiwi, strawberry, cherry, peach or plum, melon or watermelon, persimmon, fig or loquat or apricot, and grapes) and vegetables (lettuce, watercress, tomato, cucumber, green and white cabbages, broccoli, cauliflower or Brussels sprout, spinach or spring greens or turnip greens, spinach, bean pod, carrot, turnip, green pepper, and onion), as previously described in detail [81, 82]. For analysis, the cut-off of five servings per day recommended by the World Cancer Research Fund on their 2007 report on “Food, Nutrition, Physical activity and the Prevention of Cancer” was used to classify participants [83]. Energy and nutrient intake were obtained from consumption frequency and portion size data using the software Food Processor Plus R (ESHA Research), based on values from the US Department of Agriculture, further adapted for typical Portuguese foods using the Portuguese tables of food composition [84] and data from other studies that analysed the composition of Portuguese foods [79]. Dietary patterns were defined by principal components and cluster analyses, as previously described in detail [85]. The four different dietary patterns were identified in each sex. In women, the pattern with the highest consumption of vegetables, including vegetable soup, fruits, and dairy products, and with the lowest consumption of red meat, fast foods and soft drinks was named “healthy”; the pattern characterized by the lowest intake of all food group in particular fruit and vegetables, was called “low fruit and vegetables”; the pattern with the highest intake of red meat and alcohol, but also with the lowest consumption of dairy products and vegetable soup, was designated as “red meat and alcohol”; lastly, the pattern 17 with the highest intake of white meat, sweets, and fast foods and the second highest intakes of red meat, vegetables (including vegetable soup), and dairy products was named “in transition to fast food.” In men, the pattern with the highest consumption of vegetable soup, fruits, dairy products, and cereals, and the lowest of red meat, fast foods, and alcoholic beverages, was designated as “healthy”; the highest intake of fish and vegetables characterized the “fish” pattern; the pattern with the highest consumption of red meat, alcohol, and fast foods, and the lowest of fruits, vegetable soup, dairy products, and cereals was named “red meat and alcohol”; lastly, a pattern with an intermediate consumption of most food groups but of white meat was designated as “intermediate intake”. For analysis, the “healthy” dietary pattern in both men and women was compared with the others. The concentrations of total cholesterol were also assessed in the overnight fasting serum samples. If total cholesterol concentration was above 200 mg/dl or if subjects were on specific treatment for this lipid abnormality, ‘dyslipidemia’ was considered. Trained interviewers measured blood pressure on a single occasion using the recommendations of the American Heart Association [86]. A standard mercury sphygmomanometer with the cuff on the right upper arm was used. Blood pressure levels were calculated as the mean of three readings. If systolic blood pressure was 140 mg/Hg and/or diastolic blood pressure was 90 mg/Hg or if subjects were on anti-hypertensive therapy, ‘hypertension’ was considered. Anthropometric measurements were obtained after a 12-hour overnight fast, with the participant wearing light clothing and no footwear. Body weight was measured to the nearest 0.1kg using a digital scale, and height was measured to the nearest centimetre in the standing position using a wall stadiometer. Body mass index (BMI) was calculated as weight (kg) divided by squared height (m), and further divided into three categories: normal and underweight (< 24.9 kg/m2), overweight (25.0-29.9 kg/m2) or obese (≥ 30.0 kg/m2) [87, 88]. Central obesity was defined by waist circumference, which was measured midway between the lower limit of the rib cage and the iliac crest, to the nearest centimeter, with the subject standing, using a flexible and non-distensible tape and avoiding exertion of pressure on the tissues. Subjects were classified as having central obesity, according to the World Health Organization criteria [88]. Identification of cancer cases The identification of participants with a cancer diagnosis (except skin non-melanoma) was accomplished using the North Region Cancer Registry (RORENO) and death 18 certificates information. This population-based cancer registry was set up in 1988 and covers the whole northern region of Portugal (approximately 3.3 million inhabitants). The main procedure for gathering data is passive notification of cases, both by public or private hospitals/clinics and private pathology laboratories.The identification of cohort participants in the cancer registry database was performed using the name and date of birth of each participant, up to December 2009. Data analysis Subjects without a previous cancer diagnosis were eligible for this study (n=2,432), comprising a median follow-up time of 8.1 years (percentiles 25-75: 6.2-11.5). After excluding participants with a MMSE score lower than 24 points (n=74), those without information on diabetes (n=303) and those previously diagnosed with T1D (n=20), our sample comprised 2,035 participants, of whom 115 were diagnosed with cancer after the initial cohort evaluation. During follow-up, 35 participants with a tumour and 45 without a tumour had died. Figure 2. Flow-chart of the participants included in this study. 19 Age-, sexand education-adjusted incidence rate ratios (IRR) and corresponding 95% confidence intervals (95%CI) were computed using Poisson regression to quantify the association between diabetes and cancer. We also conducted these analyses among subgroups of subjects exposed to the main cancer risk factors like BMI, waist circumference, smoking, alcohol, fruits and vegetables consumption, total fiber and vitamin C intake, and dietary patterns. Ethics The local ethics committee (Centro Hospitalar de São João) approved the study protocol. All participants gave written informed consent to participate; the study was carried out in accordance with the Helsinki Declaration II. 20 Results 21 Table 1 presents the characteristics of the cohort participants. In our sample, approximately 60% were females, with a median age of 52 years and a median 9 schooling years. More than half of the subjects were overweight (40.3%) or obese (22.4%), and about 30% had central obesity. At baseline, approximately one-quarter were current smokers and three-quarters were current drinkers, with more than half having moderate alcohol consumption. More than half of the subjects had a fruits and vegetables intake below the five a day recommendation (59.8%), but vitamin C intake was according the recommendations for most participants (79.9%). Total fiber intake was also below the sexand age-specific recommendations for most subjects (77.4%). About 20% of the men and women had a healthy dietary pattern, but most women had a fast food dietary pattern (48.2%) while most men had a fish and vegetables dietary pattern (44.0%). Table 1. Baseline characteristics of the cohort participants (N=2035). N (%) a Median (percentile 25-percentile 75) Sex Women Men 1247 (61.3) 788 (38.7) Age (years) 52 (42-63) Education (years) 9 (4-13) BMI (Kg/m2) < 25.0 25.0-29.9 ≥ 30.0 759 (37.3) 821 (40.3) 455 (22.4) Waist circumference (cm) ≤ 88.0 in women or ≤ 102.0 in men > 88.0 in women or > 102.0 in men 1405 (69.8) 609 (30.2) Smoking status Never smokers Former smokers Current smokers 1103 (54.3) 426 (21.0) 503 (24.7) Alcohol consumption status Never drinkers Former drinkers Current drinkers 373 (18.4) 147 (7.2) 1512 (74.4) Alcohol consumption (g ethanol/day) 0 b ≤ 15 in women or ≤ 30 in men > 15 in women or > 30 in men 520 (25.7) 1007 (49.7) 500 (24.7) Fruits and vegetables intake (servings/day) c 4.42 (3.09-6.02) Vitamin C intake (mg/day) c 119.41 (88.40-162.32) Total fiber intake (g/day) c 18.57 (13.78-24.44) Dietary patterns Women ‘Lower fruit and vegetables’ ‘Red meat and alcohol’ ‘Healthy’ ‘In transition to fast food’ Men ‘Healthy’ ‘Fish and vegetables’ ‘Red meat and alcoho’l ‘Intermediate intake’ 102 (8.3) 239 (19.5) 293 (23.9) 590 (48.2) 157 (20.6) 335 (44.0) 168 (22.0) 102 (13.4) BMI – body mass index. a The sum of the number of participants may be lower than the total due to missing data, and percentages may add up more than 100% due to rounding; b includes never and former drinkers; c data available only for 2024 subjects who fulfilled the food frequency questionnaire. 22 During follow-up, 115 primary tumours were diagnosed, corresponding to an incidence rate of 586/100000 person-years (95%CI: 488-703). The most frequent tumour topographies were breast (n=19), prostate (n=16), colon (n=11), lung (n=8), stomach (n=8), bladder (n=6), liver (n=6), kidney (n=5), rectum (n=5) and cervix (n=4). For 4 tumours the topography was unknown. Non-diabetic subjects with impaired FBG were at an increased risk of developing cancer (RR=1.99, 95%CI: 1.04-3.83) (Table 2). However, after adjustment for sex, age and education, the association was not statistically significant. The risk of cancer was higher for diabetic subjects, and this association remained statistically significant after adjustment (RR=2.04, 95%CI: 1.18-3.51). When considering all non-diabetic participants, without and with impaired FBG, as the reference category, the risk was slightly decreased (RR=1.95, 95%CI: 1.14-3.34). Diabetes was significantly associated with an increased risk of digestive tumours (RR=2.89, 95%CI: 1.23-6.82), particularly colorectal neoplasms (RR=3.34, 95%CI: 1.0510.66) (Table 2). Breast, stomach, lung, liver, urinary tract and female genital organs cancer risk were also higher among the diabetic. Although not statistically significant, an inverse association was found for prostate cancer (RR=0.88, 95%CI: 0.12-6.76). 23 Table 2. Cancer risk associated with type 2 diabetes mellitus according to different criteria to define type 2 diabetes mellitus and according to cancer type. N (%) Cases during follow-up N (%) Crude RR (95%CI) Adjusted a RR (95%CI) According to different criteria to define type 2 diabetes mellitus According to FBG level and self-reported information b Non-diabetic without impaired FBG with impaired FBG Diabetic 1698 (88.4) 105 (5.5) 117 (6.1) 89 (5.0) 10 (8.7) 16 (12.0) 1 1.99 (1.04-3.83) 2.77 (1.63-4.72) 1 1.63 (0.85-3.15) 2.04 (1.18-3.51) According to cancer type c All cancers Non-diabetic Diabetic 1902 (93.5) 133 (6.5) 99 (5.2) 16 (12.0) 1 2.63 (1.55-4.46) 1 1.95 (1.14-3.34) Breast cancer (C50*) Non-diabetic Diabetic 1171 (93.9) 76 (6.1) 16 (1.4) 3 (4.0) 1 2.96 (0.86-10.15) 1 2.49 (0.69-8.99) d Prostate cancer (C61*) Non-diabetic Diabetic 731 (92.8) 57 (7.2) 15 (2.0) 1 (1.8) 1 1.14 (0.15-8.67) 1 0.88 (0.12-6.76) d Colorectal cancer (C18-20*) Non-diabetic Diabetic 1902 (93.5) 133 (6.5) 12 (0.6) 4 (3.0) 1 5.44 (1.75-16.86) 1 3.34 (1.05-10.66) Stomach cancer (C16*) Non-diabetic Diabetic 1902 (93.5) 133 (6.5) 6 (0.3) 2 (1.5) 1 5.40 (1.09-26.77) 1 3.68 (0.72-18.84) Digestive cancer (C15-C16,C18-C20*) Non-diabetic Diabetic 1902 (93.5) 133 (6.5) 25(1.3) 7 (5.3) 1 4.51 (1.95-10.42) 1 2.89 (1.23-6.82) Lung cancer (C34*) Non-diabetic Diabetic 1902 (93.5) 133 (6.5) 6 (0.3) 2 (1.5) 1 5.45 (1.10-26.99) 1 3.77 (0.73-19.41) Liver cancer (C22*) Non-diabetic Diabetic 1902 (93.5) 133 (6.5) 5 (0.3) 1 (0.8) 1 3.34 (0.38-27.7) 1 2.17 (0.24-19.37) Urinary tract (C64*-C67*) Non-diabetic 1902 (93.46) 9 (0.47) 1 1 Diabetic 133 (6.54) 2 (1.50) 3.64 (0.79-16.85) 2.57 (0.54-12.14) Female genital organs (C51*, C53-55*) Non-diabetic 1171 (93.91) 6 (0.51) 1 1 Diabetic 76 (6.09) 1 (1.32) 2.63 (0.32-21.83) 3.83 (0.40-37.08) d FBG – fasting blood glucose; * cancer was classified according the International Statistical Classification of Diseases and Related Health Problems 10th Revision (ICD-10), 2010 [89]; a Adjusted for sex, age and education, except if otherwise specified; b subjects were classified as non-diabetic without impaired FBG if FBG ≤ 1.10 g/L, non-diabetic with impaired FBG if 1.10 g/L < FBG ≤ 1.25 g/L, or diabetic if FBG > 1.25 g/L or if being treated with insulin or oral anti-diabetic drugs or if reported a previous diagnosis of type 2 diabetes mellitus; c subjects were considered diabetic if FBG > 1.25 g/L or if being treated with insulin or oral anti-diabetic drugs or if reported a previous diagnosis of type 2 diabetes mellitus; d adjusted for age and education. No significant differences in cancer risk associated with diabetes were observed between strata of exposure to the main lifestyle risk factors for oncological diseases. However, the risk was lower in those having healthier lifestyles, such as never smokers (RR=1.71, 95%CI: 0.82-3.54), in those having moderate alcohol consumption (RR=1.75, 30 In contrast, we observed a non significant inverse association between T2D and prostate cancer, which is consistent with most previous epidemiologic studies [23, 120]. A meta-analysis of 19 studies found an inverse association between diabetes mellitus and prostate cancer (RR=0.84; 95%CI: 0.76-0.90) [120]. The reduced risk for prostate cancer in diabetic men could be due to the lower level of plasma androgens, as higher androgens levels are known to be associated with an increased risk of prostate cancer [20]. Furthermore elevated testosterone levels have been associated with an increased risk of prostate cancer [120]. Men with T2D and obesity often have low testosterone levels and this may help explain the decreased rates of prostate cancer in these men. In fact, studies have suggested that as many as 43% of men with T2D have reduced total testosterone levels and as many as 57% have decreased free testosterone levels [121]. Thus, as prostate cancer is positively associated with testosterone levels and men with T2D and obesity have low testosterone levels, it is reasonable to conclude that men with T2D and obesity have a lower risk of prostate cancer than men without these conditions. Stocks et al reported on a prospective study and found that men with increased insulin resistance and worsening glycemic control had a decreased risk of prostate cancer [122]. The association observed in our study could alternatively be explained by an increased use of health care services by diabetic subjects. Diabetes is a chronic condition, with increased medical attention, and may contribute to earlier detection of any present but previously undiagnosed cancer. In this study, diabetic subjects had a doubled number of medical appointments in the year before the interview, compared with non-diabetic (median: 4 appointments vs. 2 appointments, p<0.001). Notwithstanding, the median number of appointments was similar for those who had developed a tumour and those who did not regardless of diabetes status (median of two appointments in the last year, p=0.611). Furthermore, it has been described that individuals more recently diagnosed with diabetes have an increased risk of cancer due to the same indication bias [102]. However, in our study we found no difference in the risk of cancer according to the duration of diabetes (RR=1.19, 95%CI: 0.39-3.68 for patients diagnosed with diabetes for more than 5 years compared with those diagnosed in the last 5 years). These results strengthen the hypothesis of a causal relation between diabetes and cancer. When investigating the impact of diabetic treatment on cancer occurrence within the cohort, we noted that metformin, either as a monotherapy or in combination was associated 31 with a decreased (RR=0.48, 95%CI: 0.05-4.56) and sulforynureas with an increase risk of cancer (RR=3.54, 95%CI: 0.41-30.5). This finding is in the line with a study conducted by Hsieh et al. They showed that diabetic patients treated with insulin or sulfonylureas had significantly higher risk of all cancers, compared to those treated with metformin [odds ratio (OR)=1.58; 95%CI: 1.39-1.80, and OR=1.78; 95%CI: 1.41-2.26, respectively] [123]. Associations between other oral anti-diabetic drugs or insulin analogues, either alone or combination have also been reported [30, 63, 67]. Unfortunately, the number of T2D patients taking these medications in our cohort was low and no risk estimates could be computed. Furthermore, the distinction between long and fast acting analogues was not possible, nor did we have data on drug dosages. Although we had data on mortality for our cohort participants, we opted for not conducting a survival analysis for those diagnosed with cancer during follow-up, aiming to disclosure the impact of type 2 diabetes on cancer survival. This methodological option was due to the non-exhaustiveness of the mortality data and to the probably low quality of these data in our cohort, since the available information for death was collected from different sources, such as the death certificates and response by relatives. To best of our knowledge, this is the first study conducted in Portugal addressing this association, using a prospective approach. Our study confirms previous reports on the association of T2D with cancer, however, the impact varied across different cancer types suggesting the need to evaluate specific cancers individually. Our results highlight an additional attention that health care professionals need to take into account when dealing with diabetic patients. They point out how important is to monitor these subjects properly, as cancer prevention and early detection should be important components of diabetes management in light of the exponentially increasing prevalence of this chronic condition. 32 Conclusion 33 Our prospective study confirms previous reports on the association of T2D with cancer, which is independent of other risk factors for oncological diseases. However, the impact of diabetes varied across different cancer types suggesting the need to evaluate specific cancers individually. We also found that subjects with impaired fasting have an increased risk of cancer, though diabetes was not established at that time. Furthermore specific antihyperglycemic medications are associated with increase or decrease cancer risk. This study may support the development of primary and secondary prevention programs aimed to increased awareness of diabetes patients and healthcare personnel to the importance of cancer prevention efforts targeted to this specific population. 34 References 35 1. Shaw, J.E., R.A. Sicree, and P.Z. Zimmet, Global estimates of the prevalence of diabetes for 2010 and 2030. Diabetes Res Clin Pract, 2010. 87(1): p. 4-14. 2. Ferlay, J., et al. GLOBOCAN 2008, Cancer Incidence and Mortality Worldwide. IARC CancerBase No. 10 2010 [Available from: http://globocan.iarc.fr. 3. Suh, S. and K.W. Kim, Diabetes and cancer: is diabetes causally related to cancer? Diabetes Metab J, 2011. 35(3): p. 193-8. 4. Danaei, G., et al., National, regional, and global trends in fasting plasma glucose and diabetes prevalence since 1980: systematic analysis of health examination surveys and epidemiological studies with 370 country-years and 2.7 million participants. Lancet, 2011. 378(9785): p. 31-40. 5. International Diabetes Federation, IDF Diabetes Atlas, 5th edn. 2011: Brussels, Belgium. 6. Wild, S., et al., Global prevalence of diabetes: estimates for the year 2000 and projections for 2030. Diabetes Care, 2004. 27(5): p. 1047-53. 7. Pereira M, et al., Trends in prevalence of diabetes mellitus and mean fasting glucose levels in Portugal (1987-2009): a systematic review. in Epidemiology for a fair and healthy society. Abstracts of the IEA-EEF (European Epidemiology Federation of the International Epidemiological Association) European Congress of Epidemiology 2012. September 2012. Porto, Portugal. Eur J Epidemiol, 2012. 27 Suppl 1: p. S14. 8. Gardete-Correia, L., et al., First diabetes prevalence study in Portugal: PREVADIAB study. Diabet Med, 2010. 27(8): p. 879-81. 9. World Cancer Research Fund International. Cancer statistics: worldwide. 2012; Available from: http://www.wcrf.org/cancer_statistics/world_cancer_statistics.php. 10. RORENO, Registo Oncologico Regional do Norte 2007, Instituto Português de Oncologia do Porto, Editor. 2011: Porto. 11. World Health Organization, Global Burden of Disease: 2004 Update. 2008. 12. van Dieren, S., et al., The global burden of diabetes and its complications: an emerging pandemic. Eur J Cardiovasc Prev Rehabil, 2010. 17 Suppl 1: p. S3-8. 13. Noto, H., et al., Substantially increased risk of cancer in patients with diabetes mellitus: a systematic review and meta-analysis of epidemiologic evidence in Japan. J Diabetes Complications, 2010. 24(5): p. 345-53. 14. Giovannucci, E., et al., Diabetes and cancer: a consensus report. Diabetes Care, 2010. 33(7): p. 1674-85. 15. Nicolucci, A., Epidemiological aspects of neoplasms in diabetes. Acta Diabetol, 2010. 47(2): p. 87-95. 36 16. Sun, G. and S.R. Kashyap, Cancer risk in type 2 diabetes mellitus: metabolic links and therapeutic considerations. J Nutr Metab, 2011. 2011: p. 708183. 17. Czyzyk, A. and Z. Szczepanik, Diabetes mellitus and cancer. Eur J Intern Med, 2000. 11(5): p. 245-252. 18. Chodick, G., et al., Diabetes and risk of incident cancer: a large population-based cohort study in Israel. Cancer Causes Control, 2010. 21(6): p. 879-87. 19. Hense, H.W., et al., Cancer incidence in type 2 diabetes patients - first results from a feasibility study of the D2C cohort. Diabetol Metab Syndr, 2011. 3(1): p. 15. 20. Simon, D. and B. Balkau, Diabetes mellitus, hyperglycaemia and cancer. Diabetes Metab, 2010. 36(3): p. 182-91. 21. Zendehdel, K., et al., Cancer incidence in patients with type 1 diabetes mellitus: a population-based cohort study in Sweden. J Natl Cancer Inst, 2003. 95(23): p. 1797800. 22. Pierce, B.L., Why are diabetics at reduced risk for prostate cancer? A review of the epidemiologic evidence. Urol Oncol, 2012. 30(5): p. 735-43. 23. Rodriguez, C., et al., Diabetes and risk of prostate cancer in a prospective cohort of US men. Am J Epidemiol, 2005. 161(2): p. 147-52. 24. Zhang, P.H., et al., Increased risk of cancer in patients with type 2 diabetes mellitus: A retrospective cohort study in China. BMC Public Health, 2012. 12(1): p. 567. 25. Barone, B.B., et al., Long-term all-cause mortality in cancer patients with preexisting diabetes mellitus: a systematic review and meta-analysis. JAMA, 2008. 300(23): p. 2754-64. 26. Liu, X., et al., The impact of type 2 diabetes mellitus on cancer-specific survival: a follow-up study in Sweden. Cancer, 2012. 118(5): p. 1353-61. 27. Peairs, K.S., et al., Diabetes mellitus and breast cancer outcomes: a systematic review and meta-analysis. J Clin Oncol, 2011. 29(1): p. 40-6. 28. Vigneri, P., et al., Diabetes and cancer. Endocr Relat Cancer, 2009. 16(4): p. 110323. 29. Carstensen, B., D.R. Witte, and S. Friis, Cancer occurrence in Danish diabetic patients: duration and insulin effects. Diabetologia, 2012. 55(4): p. 948-58. 30. Chang, C.H., et al., Cancer risk associated with insulin glargine among adult type 2 diabetes patients--a nationwide cohort study. PLoS One, 2011. 6(6): p. e21368. 31. Centers for Disease Control and Prevention. National Diabetes. General Information and National Estimates on Diabetes in the United States, 2007. 2008; Available from: http://www.cdc.gov/diabetes/pubs/pdf/ndfs_2007.pdf. 32. Jemal, A., et al., Cancer statistics, 2009. CA Cancer J Clin, 2009. 59(4): p. 225-49. 37 33. Pischon, T., U. Nothlings, and H. Boeing, Obesity and cancer. Proc Nutr Soc, 2008. 67(2): p. 128-45. 34. Elks, C.M. and J. Francis, Central adiposity, systemic inflammation, and the metabolic syndrome. Curr Hypertens Rep, 2010. 12(2): p. 99-104. 35. Joint FAO/WHO Expert Consultation, WHO Technical Report Series 916: Diet, Nutrition and the Prevention of Chronic Diseases, World Health Organization. 2003: Geneva, Switzerland. 36. Kastorini, C.M. and D.B. Panagiotakos, Dietary patterns and prevention of type 2 diabetes: from research to clinical practice; a systematic review. Curr Diabetes Rev, 2009. 5(4): p. 221-7. 37. Hamman, R.F., et al., Effect of weight loss with lifestyle intervention on risk of diabetes. Diabetes Care, 2006. 29(9): p. 2102-7. 38. Physical Activity Guidelines Advisory Committee report, 2008. To the Secretary of Health and Human Services. Part A: executive summary. Nutr Rev, 2009. 67(2): p. 114-20. 39. Haire-Joshu, D., R.E. Glasgow, and T.L. Tibbs, Smoking and diabetes. Diabetes Care, 2004. 27 Suppl 1: p. S74-5. 40. Heuson, J.C., N. Legros, and R. Heimann, Influence of insulin administration on growth of the 7,12-dimethylbenz(a)anthracene-induced mammary carcinoma in intact, oophorectomized, and hypophysectomized rats. Cancer Res, 1972. 32(2): p. 233-8. 41. Piatkiewicz, P. and A. Czech, Glucose metabolism disorders and the risk of cancer. Arch Immunol Ther Exp (Warsz), 2011. 59(3): p. 215-30. 42. Ciaraldi, T.P. and T. Sasaoka, Review on the in vitro interaction of insulin glargine with the insulin/insulin-like growth factor system: potential implications for metabolic and mitogenic activities. Horm Metab Res, 2011. 43(1): p. 1-10. 43. Gao, J., et al., Targeting the insulin-like growth factor axis for the development of novel therapeutics in oncology. Cancer Res, 2012. 72(1): p. 3-12. 44. Belfiore, A., et al., Insulin receptor isoforms and insulin receptor/insulin-like growth factor receptor hybrids in physiology and disease. Endocr Rev, 2009. 30(6): p. 586623. 45. Denley, A., et al., Differential activation of insulin receptor substrates 1 and 2 by insulin-like growth factor-activated insulin receptors. Mol Cell Biol, 2007. 27(10): p. 3569-77. 46. Giovannucci, E. and D. Michaud, The role of obesity and related metabolic disturbances in cancers of the colon, prostate, and pancreas. Gastroenterology, 2007. 132(6): p. 2208-25. 38 47. Warburg, O., F. Wind, and E. Negelein, The Metabolism of Tumors in the Body. J Gen Physiol, 1927. 8(6): p. 519-30. 48. Miao Jonasson, J., et al., HbA1C and cancer risk in patients with type 2 diabetes--a nationwide population-based prospective cohort study in Sweden. PLoS One, 2012. 7(6): p. e38784. 49. Yang, X., et al., Associations of hyperglycemia and insulin usage with the risk of cancer in type 2 diabetes: the Hong Kong diabetes registry. Diabetes, 2010. 59(5): p. 1254-60. 50. Calle, E.E., et al., Overweight, obesity, and mortality from cancer in a prospectively studied cohort of U.S. adults. N Engl J Med, 2003. 348(17): p. 1625-38. 51. Malnick, S.D. and H. Knobler, The medical complications of obesity. QJM, 2006. 99(9): p. 565-79. 52. Arcidiacono, B., et al., Insulin resistance and cancer risk: an overview of the pathogenetic mechanisms. Exp Diabetes Res, 2012. 2012: p. 789174. 53. van Kruijsdijk, R.C., E. van der Wall, and F.L. Visseren, Obesity and cancer: the role of dysfunctional adipose tissue. Cancer Epidemiol Biomarkers Prev, 2009. 18(10): p. 2569-78. 54. Cusi, K., The role of adipose tissue and lipotoxicity in the pathogenesis of type 2 diabetes. Curr Diab Rep, 2010. 10(4): p. 306-15. 55. Stofkova, A., Resistin and visfatin: regulators of insulin sensitivity, inflammation and immunity. Endocr Regul, 2010. 44(1): p. 25-36. 56. Ahima, R.S., Metabolic actions of adipocyte hormones: focus on adiponectin. Obesity (Silver Spring), 2006. 14 Suppl 1: p. 9S-15S. 57. Jackson, M.B. and R.S. Ahima, Neuroendocrine and metabolic effects of adipocytederived hormones. Clin Sci (Lond), 2006. 110(2): p. 143-52. 58. Nathan, D.M., et al., Medical management of hyperglycemia in type 2 diabetes: a consensus algorithm for the initiation and adjustment of therapy: a consensus statement of the American Diabetes Association and the European Association for the Study of Diabetes. Diabetes Care, 2009. 32(1): p. 193-203. 59. Decensi, A., et al., Metformin and cancer risk in diabetic patients: a systematic review and meta-analysis. Cancer Prev Res (Phila), 2010. 3(11): p. 1451-61. 60. Evans, J.M., et al., Metformin and reduced risk of cancer in diabetic patients. BMJ, 2005. 330(7503): p. 1304-5. 61. Libby, G., et al., New users of metformin are at low risk of incident cancer: a cohort study among people with type 2 diabetes. Diabetes Care, 2009. 32(9): p. 1620-5. 39 62. Yang, X., et al., Low HDL cholesterol, metformin use, and cancer risk in type 2 diabetes: the Hong Kong Diabetes Registry. Diabetes Care, 2011. 34(2): p. 375-80. 63. Gallagher, E.J. and D. LeRoith, Diabetes, cancer, and metformin: connections of metabolism and cell proliferation. Ann N Y Acad Sci, 2011. 1243: p. 54-68. 64. Noto, H., et al., Cancer risk in diabetic patients treated with metformin: a systematic review and meta-analysis. PLoS One, 2012. 7(3): p. e33411. 65. Colmers, I.N., et al., Use of thiazolidinediones and the risk of bladder cancer among people with type 2 diabetes: a meta-analysis. CMAJ, 2012. 184(12): p. E675-83. 66. Koro, C., S. Barrett, and N. Qizilbash, Cancer risks in thiazolidinedione users compared to other anti-diabetic agents. Pharmacoepidemiol Drug Saf, 2007. 16(5): p. 485-92. 67. Yang, X., et al., Use of sulphonylurea and cancer in type 2 diabetes-The Hong Kong Diabetes Registry. Diabetes Res Clin Pract, 2010. 90(3): p. 343-51. 68. Ruiter, R., et al., Lower risk of cancer in patients on metformin in comparison with those on sulfonylurea derivatives: results from a large population-based follow-up study. Diabetes Care, 2012. 35(1): p. 119-24. 69. Monami, M., et al., Sulphonylureas and cancer: a case-control study. Acta Diabetol, 2009. 46(4): p. 279-84. 70. Currie, C.J., C.D. Poole, and E.A. Gale, The influence of glucose-lowering therapies on cancer risk in type 2 diabetes. Diabetologia, 2009. 52(9): p. 1766-77. 71. Dejgaard, A., et al., No evidence of increased risk of malignancies in patients with diabetes treated with insulin detemir: a meta-analysis. Diabetologia, 2009. 52(12): p. 2507-12. 72. Weinstein, D., et al., Insulin analogues display IGF-I-like mitogenic and anti-apoptotic activities in cultured cancer cells. Diabetes Metab Res Rev, 2009. 25(1): p. 41-9. 73. Jonasson, J.M., et al., Insulin glargine use and short-term incidence of malignanciesa population-based follow-up study in Sweden. Diabetologia, 2009. 52(9): p. 1745-54. 74. Ramos, E., C. Lopes, and H. Barros, Investigating the effect of nonparticipation using a population-based case-control study on myocardial infarction. Ann Epidemiol, 2004. 14(6): p. 437-41. 75. Folstein, M.F., S.E. Folstein, and P.R. McHugh, "Mini-mental state". A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res, 1975. 12(3): p. 189-98. 76. Anthony, J.C., et al., Limits of the 'Mini-Mental State' as a screening test for dementia and delirium among hospital patients. Psychol Med, 1982. 12(2): p. 397-408.