Diabetes Mellitus and Colon Carcinogenesis: Expectation for Inhibition of Colon Carcinogenesis by Oral Hypoglycemic Drugs

Abstract

The global deaths due to colorectal cancer and diabetes mellitus have increased by 57% and 90%, respectively. The relationship between various cancers and diabetes mellitus has been shown in multiple epidemiological studies. Hence, better management of diabetes mellitus is expected to reduce the risk of various cancers. This review focuses on colorectal cancer and aims to summarize recent findings on the antitumor effects of various oral hypoglycemic drugs on colorectal cancer and their estimated mechanisms. Of the seven classes of oral hypoglycemic agents, only metformin was found to have suppressive effects on colorectal cancer in both clinical and basic research. Clinical and basic researches on suppressing effects of glinides, dipeptidyl peptidase-4 inhibitors, thiazolidinedione, α-glucosidase inhibitors, and sodium glucose cotransporter-2 inhibitors against colon carcinogenesis have been insufficient and have not arrived at any conclusion. Therefore, further research regarding these agents is warranted. In addition, the suppressive effects of these agents in healthy subjects without diabetes should also be investigated.

1. Introduction

The global deaths due to colorectal cancer (CRC) and diabetes mellitus (DM) have increased by 57% and 90%, respectively. The risk of CRC was estimated to be 27% higher in type 2 DM patients than in non-diabetic patients. Due to imbalances in lifestyles, including diet and exercise, diseases related to obesity and metabolic syndrome have increased. The number of cases of DM, especially type 2 DM, is increasing, and this trend is expected to continue in the future [1].

The relationship between DM and various cancers has been shown in many epidemiological studies [2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23], therefore, adequate DM management is expected to decrease the risk of various cancers. In fact, studies that acknowledge the carcinogenesis-inhibiting effect of several oral hypoglycemic agents currently in clinical use have increased in recent years.

This review focuses on CRC and aims to summarize the recent findings regarding the antitumor effect of various oral glycemic drugs on CRC and their estimated underlying mechanisms.

2. Diabetes Mellitus and Colorectal Cancer: Basic Introduction and Epidemiological Relevance

2.1. Diabetes Mellitus

Diabetes mellitus (DM) is a disease in which hyperglycemia presents chronically because of insulin deficiency or a decrease in the effect of insulin, where glucose tolerance is reduced. In addition to the major complications of retinal diseases, nephropathy and neuropathy, which are microvascular disorders, the development of serious complications without subjective symptoms increases, arteriosclerosis of larger vessels progresses, and the risk of heart disease increases.

DM is classified into two types, type 1 DM and type 2 DM, depending on the necessity of supplementing insulin. Type 1 DM is characterized by deficient insulin production associated with progressive destruction of pancreatic β cells, while type 2 DM is characterized by insulin dysfunction associated with environmental influences such as overeating, lack of exercise and obesity.

The number of people suffering from DM worldwide will be 629 million in 2045, compared with an estimated 426 million in 2017. In addition, the prevalence rate of DM in adult humans was 8.47% in 2017 [24]. The number of people with type 2 DM is increasing in every country and about 1 in 11 adults worldwide now have diabetes mellitus, 90% of whom have type 2 DM [25,26].

Treatment of DM is mainly by increasing insulin availability, improving insulin sensitivity, inhibiting gluconeogenesis, delaying the absorption of carbohydrate in the intestinal tract, and promoting excretion of glucose into the urine [1].

2.2. Colorectal Cancer

Colorectal cancer (CRC) is the third most commonly diagnosed cancers and the second leading cause of cancer-related global deaths [27,28]. The incidence increases with age and they usually occur in persons of 50 years old and above. Its burden is expected to increase by 60% to more than 2.2 million new cases and 1.1 million deaths due to cancer by 2030 [29].

Molecular and pathological studies on carcinogenesis of colorectum have earned much knowledge from early on as a breakthrough in genetic analysis of hereditary CRC. CRC is not caused by a single genetic abnormality, its predominant mechanism has been the multistep carcinogenesis theory, which states that colon cancer develops by the accumulation of multiple gene abnormalities [30]. Increased genomic instability due to the accumulation of genomic abnormalities is involved in carcinogenesis of various cancers including CRC. Accumulation of genomic abnormalities in CRC is classified into two types: chromosomal instability (CIN) which is accompanied by structural change of the chromosome and microsatellite instability (MIN) which is caused due to an abnormal repetitive sequence (microsatellite) of DNA in the genome without any structural change in chromosome. It has been reported that 80 to 85% of colon cancers are of the CIN type and about 15% are of the MIN type [31]. The genetic abnormality that causes MIN type is due to inactivation of the mismatch repair gene. Inactivation of cancer suppressor genes (e.g., APC, TP 53, TGFβ, etc.) and activation of oncogenes (e.g., KRAS) are also involved in colorectal carcinogenesis [1,32,33].

For early detection and early treatment, fecal occult blood test (FOBT) and endoscopy are useful. The sensitivity for CRC in the 1-day method of FOBT is 30–56% and the specificity is 96–97%; in 2-day and 3-day methods of FOBT, the sensitivity for CRC is 83–92% and the specificity is 90–96% [34,35,36]. FOBT is a minimally invasive simple test and is useful for screening of CRC. Endoscopy not only assesses the intestine directly but can also perform tissue biopsy and helps to perform endoscopic mucosal resection if there are abnormal lesions such as adenomatous polyps. Early stage CRC is treated with surgery, and locally advanced CRC is treated with adjuvant chemotherapy along with surgery. Neoadjuvant chemo-radiation is one of the standard treatments for rectal cancer with nodal disease [37]. Adjuvant chemotherapy schemes contain 5-fluorouracil and oxaliplatin. Metastatic CRC is treated with irinotecan or oxaliplatin combined with fluoropyrimidine and leucovorin (FOLFIRI or FOLFOX regimens) [38].

2.3. Epidemiological Relevance of Diabetes Mellitus and Colorectal Cancer Risk

In a study conducted in Taiwan with data analyzed from 36,270 DM (type 1:type 2 = 1447:34,823) patients and 145,080 subjects without DM, it was shown that the incidence of cancer at any site was significantly higher in patients with DM than in those without DM, and the risk of carcinogenesis imparted by DM was greatest in gastroenterological malignancies such as CRC, liver and pancreas as well as lung, breast and oral cancer [39,40]. In addition, various prospective cohort studies [3,7,8,9,12,41] and meta-analysis [5,6] have reported that the risk of various cancers increases significantly in DM-affected groups [25].

Focusing on CRC, a cohort study conducted in the United States on 850,000 diabetic patients during the period of 1960–1972 is listed as an early report on the relationship between DM and CRC. The adjusted incidence density ratio of CRC was 1.30 (95% confidence interval (CI) 1.03–1.65) for diabetic males, but no significant incidence was found for females [42]. A more recent US study observed an increased adjusted hazard ratio (HR) for CRC regardless of gender, which could be due to a change in lifestyle from the 1960s to 1990s [11].

Since the early 2000s, several cohort studies have been conducted [2,4,11,13,14,15,16,17,18,19,20,21,22,23], showing a significant rise in CRC risk compared to diabetes-related groups in diabetes-affected groups before the 2000s (

Table 1. Epidemiological relevance of diabetes mellitus and risk of colorectal cancer.

References	Study	Country	Sample	Period (Years)	Risk of CRC Among DM Participants (95% CI)
					Males	Females	Overall
Nilsen et al., 2001 [2]	Cohort	Norway	751,922	12	RR 0.66 (0.35–1.34)	RR 1.55 (1.04–2.31)	N.A.
Inoue et al., 2006 [4]	Cohort	Japan	97,771	12	HR 1.36 (1.00–1.85)	HR 0.83 (0.42–1.68)	N.A.
Jarvandi et al., 2013 [11]	Cohort	USA	484,020	12	C: HR 1.24 (1.12–1.38) R: HR 1.34 (1.14–1.57)	C: HR 1.37 (1.16–1.60) R: HR 1.43 (1.08–1.88)	C: HR 1.27 (1.17–1.39) R: HR 1.36 (1.18–1.56)
Oberaigner et al., 2014 [13]	Cohort	Austria	5,709	22	SIR 1.11 (0.81–1.49)	SIR 0.94 (0.62–1.36)	N.A.
Wang et al., 2015 [17]	Cohort	China	327,000	7	C: SIR 1.47 (1.29–1.67) R: SIR 1.25 (1.09–1.43)	C: SIR 1.33 (1.15–1.54) R: SIR 1.29 (1.10–1.51)	C: SIR 1.40 (1.27–1.55) R: SIR 1.26 (1.14–1.40)
Harding et al. 2015 [14]	Cohort	Australia	953,382	12	SIR 1.18 (1.15–1.21)	SIR 1.16 (1.13–1.20)	N.A.
Liu et al., 2015 [15]	Cohort	Sweden	380,196	47	N.A.	N.A.	C: SIR 1.33 (1.28–1.38) R: SIR 1.19 (1.13–1.25)
Dankner et al., 2016 [18]	Cohort	Israel	218,6196	11	SIR 1.45 (1.37–1.55)	SIR 1.48 (1.39–1.57)	N.A.
de Kort et al., 2016 [19]	Cohort	Netherlands	120,852	21	HR 0.95 (0.75–1.20)	HR 1.08 (0.85–1.37)	N.A.
Ballotari et al., 2017 [20]	Cohort	Italy	383,799	4	IRR 1.44 (1.25–1.55)	IRR 1.44 (1.25–2.60)	IRR 1.32 (1.12–1.55)
de Jong et al., 2017 [22]	Cohort	Netherlands	34,038	14	N.A.	N.A.	C: HR 1.40 (1.10–1.70) R: HR 0.88 (0.63–1.20)
Chen et al., 2017 [21]	Cohort	Asia	771,297	23	C: HR 1.57 (1.27–1.93) R: HR 1.26 (0.99–1.61)	C: HR 1.48 (1.20–1.81) R: HR 1.55 (1.12–2.16)	HR 1.41 (1.26–1.57)
Saarela et al., 2018 [23]	Cohort	Finland	428,326	27	C: SIR 1.29 (1.24–1.33) R: SIR 1.16 (1.10–1.21)	C: SIR 1.16 (1.12–1.20) R: SIR 1.13 (1.06–1.19)	C: SIR 1.22 (1.19–1.25) R: SIR 1.14 (1.10–1.18)

Cohort means prospective cohort. DM, diabetes mellitus; CI, confidence interval; C, colon; R, rectum; RR, relative risk; HR, hazard ratio; SIR, standardized incidence rate; IRR, incident rate ratio; N.A., not applicable.

).

2.4. Expectation for Reduction of Carcinogenic Risk by Diabetes Treatment, Especially Oral Hypoglycemic Drugs

Although there is no significant difference in the incidence of malignancy between type 1 DM and type 2 DM [25], type 2 DM accounts for more than 90% of the total number of diabetic patients. Considering the current situation in DM, especially type 2 DM, good control of DM by drug therapy is expected to reduce the risk of developing various malignant tumors. While treatment for type 1 DM is mainly supplementation with insulin, various oral hypoglycemic drugs are currently developed and used for type 2 DM. Recent studies have reported that various oral hypoglycemic agents, such as metformin, the preferred and most widely used pharmacological agent for type 2 DM, has tumor-suppressive properties.

3. Molecular Mechanisms of the Association between Diabetes Mellitus and Colorectal Cancer

DM and CRC share multiple risk factors, such as obesity, Western diet, sedentary lifestyle, and alcohol/tobacco use. For this reason, it is easy to imagine the theory that DM might be a causal factor for CRC development [43]. Several representative pathophysiological mechanisms, such as the insulin-like growth factor (IGF) signaling system and Wnt signaling system, have been proposed for association between DM and CRC, especially for the increased risk of developing CRC in people with DM.

3.1. The Insulin-like Growth Factor Signaling System

IGF signaling system is associated with cell proliferation, diabetes, hyperinsulinemia and obesity. IGF signaling proteins, such as IGF-1 and IGF-2, play important roles not only in cell growth and differentiation, but also in the occurrence and development of tumors. The biological effect of IGF signaling proteins are mediated by IGF-1 receptor (IGF-1R) that is structurally associated with the insulin receptor [44,45]. Ligands for IGF-1R not only include IGF-1 and IGF-2, but also insulin; ligands induce autophosphorylation of IGF-1R. This results in activation of the phosphatidylinositol 3-kinase-AKT(PI3K)/mammalian target of rapamycin (mTOR) signaling pathway and RAS protein/mitogen-activated protein Kinase (MAPK) pathway. This leads to promotion of cell proliferation and suppression of apoptosis, which leads to carcinogenesis [46]. IGF-1 is also involved in the development of angiogenesis and metastases by increasing the expression of vascular endothelial growth factor (VEGF) [47,48]. IGF-1 prevented apoptosis by inhibiting p53 and also caused angiogenesis by inducing hypoxia-inducible factor 1α (HIF1α) involved in the expression of VEGF [49,50].

In hyperinsulinemia caused by insulin resistance associated with type 2 diabetes, high insulin levels in the portal circulation upregulate the growth hormone receptor, enhance the growth hormone receptor signaling, and increase IGF-1 production [51,52].

3.2. The Wnt Signaling System

The Wnt signaling system is the fundamental mechanism that directs cells for proliferation, as a result, mutations in the Wnt signaling pathway are often associated with cancers and other diseases.

In the absence of Wnt, cytoplasmic β-catenin protein is constantly degraded by the effect of Axin complex, which is composed of Axin, the tumor suppressor adenomatous polyposis coli (APC) gene product, casein kinase 1 (CK1), and glycogen synthase kinase 3 (GSK3). CK1 and GSK3 sequentially phosphorylate the amino terminal region of β-catenin, resulting in β-catenin recognition by β-Trcp and subsequent β-catenin ubiquitination and proteasomal degradation. As a result, the transfer of β-catenin into the nucleus is inhibited, and Wnt target genes are suppressed by the DNA-bound T cell factor/lymphoid enhancer factor (TCF/LEF) family of proteins.

In the presence of Wnt, a Wnt ligand binds to a frizzled (Fz) receptor and its co-receptor, low-density lipoprotein receptor related protein 6 (LRP6) or its close relative LRP5. The Wnt-Fz-LRP5/6 complex together with the recruitment of the scaffolding dishevelled (Dvl) protein results in the recruitment of the axin complex to the receptors. These events lead to inhibition of β-catenin phosphorylation and to the stabilization of β-catenin, which accumulates and transfers to the nucleus to form complexes with TCF/LEF and activates Wnt target gene expression [53].

Also, LRP5 appears to serve as a co-receptor in insulin signaling. The insulin receptor (IR)/LRP5 interaction could be a mode of action in the Wnt effect on Akt, ERK1/2, and GSK3 phosphorylation. The IR/LRP5 interaction acts could play a role in the pathogenesis of obesity and insulin resistance. The inducibility of this interaction is specific for the insulin receptor and is not observed with the IGF-1 receptor. Insulin and Wnt signaling interactions could also play a role in the association of these pathways in conditions like obesity, type 2 DM and cancer [54].

3.3. Glucagon-like Peptide-1

Another possible molecular mechanism linking DM and CRC includes the involvement of the hormone glucagon-like peptide-1 (GLP-1), secreted by the intestinal endocrine L cells, on the Wnt signaling pathway and the oncogenes. Because of insulin resistance, GLP-1 secretion is reduced in type 2 DM patients. Reduction of GLP-1 secretion with increased expression of proto-oncogenes, such as c-Myc, causes compensatory activation of the Wnt signaling pathway leading to intestinal cell proliferation and CRC development [55].

3.4. Gut Microbiota

There are at least 100 trillion bacteria that live in our gut system, which are known as the gut microbiome. Approximately 90% of the microbial system in healthy adults are Firmicutes and Bacteroidetes [56]. The gut microbiota can work with the host to promote health but can sometimes initiate or promote disease [57,58,59,60].

The abundance of Gram-positive bacteria that produce butyrate was reduced in patients with metabolic syndrome who were treated with vancomycin. This was correlated with impaired insulin sensitivity; results showed that reduced levels of butyrate produced in gut microbiota might lead to disease pathogenesis of type 2 diabetes [61].

Among short chain fatty acids (SCFA) produced by gut microbiota, acetic acid and butyric acid increase mucus production from intestinal mucosal goblet cells and protect intestinal tract. When the action of goblet cells is suppressed by the decrease of SCFA, this leads to a decrease in the function of the intestinal barrier, and lipopolysaccharides (LPS) produced by Gram negative bacilli, mostly Proteobacterias, transfer from the intestinal side to the lumen, where it comes in contact with blood. An increase of LPS levels in blood causes insulin resistance in insulin-sensitive organs such as skeletal muscle and liver [62]. Increased insulin resistance results in hyperinsulinemia and as a result may activate the IGF signaling system and Wnt signaling system, leading to colon carcinogenesis.

4. Suppressive Effect of Colonic Carcinogenesis by Oral Hypoglycemic Drugs

4.1. Types of Oral Hypoglycemic Drugs for Type 2 Diabetes Mellitus

There are seven classes of oral hypoglycemic drugs, which can be classified according to their respective main functions; insulin secretion promoting system for decreased insulin secretion, insulin resistance improving system for increased insulin resistance, and glucose absorption/excretion regulation system for the state of hyperglycemia.

Drugs whose main function is based upon an insulin secretion-promoting system include sulfonylureas, immediate-release insulin secretagogues (glinides), and dipeptidyl peptidase (DPP)-4 inhibitors, whereas those with main functions based upon an insulin resistance-improving system include biguanides and thiazolidinedione. Drugs whose main function is based upon a glucose absorption/excretion regulation system include alpha-glucosidase inhibitors (AGIs) and sodium-glucose cotransporter 2 (SGLT2) inhibitors.

4.2. Basic Pharmacological Action of Oral Hypoglycemic Drugs and Reports on Suppressive Effect of Colonic Carcinogenesis Risk

4.2.1. Sulfonylureas

The main effect of sulfonylureas is to increase plasma insulin concentrations. They are effective only when functional pancreatic β-cells are present. The rise in plasma insulin concentrations occurs for two reasons. Firstly, they stimulate insulin secretion by pancreatic β-cells. Secondly, they decrease hepatic clearance of insulin. Sulfonylureas have been traditionally classified into first-generation agents (tolbutamide, chlorpropamide, tolazamide, and acetohexamide) and second-generation agents (glibenclamide, glipizide, gliclazide, glimepiride, and gliquidone) [63]. Currently, the second generation is used predominantly.

We investigated cohort studies that examined various carcinogenic risks including CRC using sulfonylureas reported between 2000 and 2019. There was no report showing the effect of sulfonylureas to reduce the risk of CRC regardless of the presence or absence of significant differences [64,65,66,67,68,69]. Conversely, several studies showed that the risk of CRC slightly increases by using sulfonylureas, but to a lesser extent than insulin [70,71,72].

Based on these results, unfortunately, sulfonylureas seems to have no clear suppressive effect on colonic carcinogenesis.

4.2.2. Immediate-release Insulin Secretagogues (Glinides)

The main effect of glinides (repaglinide and nateglinide) is to increase the plasma insulin concentrations. The rise in plasma insulin concentrations occurs for inhibiting ATP-sensitive potassium channels in the pancreatic β-cell membrane, thus providing improved control of postprandial glucose concentrations [73].

We investigated cohort studies that examined various carcinogenic risks including CRC using glinides reported between 2000 and 2019. There are few comprehensive studies verifying the risk of cancer, including CRC, under glinides treatment [71,74]. Also, among those reports there is no clear positive effect on the risk of cancer. However, a recent in-vitro work attested that repaglinide has anti-cancer properties [75].

Due to inadequate studies, no further information is available concerning the effects of these drugs on the risk of CRC.

4.2.3. DPP-4 Inhibitors

Incretin which is an intestinal hormone, such as GLP-1, is secreted from intestinal epithelial cells with elevation of postprandial blood glucose level. GLP-1 stimulates insulin secretion and inhibits glucagon secretion, in turn increasing glucose utilization and diminishing hepatic glucose production [76]. GLP-1 in peripheral plasma is degraded by DPP-4; therefore, DPP-4 inhibitors (sitagliptin, vildagliptin, saxagliptin, alogliptin, and linagliptin) increases GLP-1 in the peripheral circulation. As a result, there is a decrease in postprandial blood glucose and fasting blood glucose and HbA1c [76].

We investigated cohort studies that examined various carcinogenic risks including CRC using DPP-4 inhibitors reported between 2000 and 2019. However, there are few reports on the relationship between DPP-4 inhibitors and CRC. We found only one clinical study conducted in 2018 on the risk of colon carcinogenesis; the study reported that there was no improvement in the risk of developing colon carcinogenesis by DPP-4 inhibitors [77,78]. In in-vitro studies conducted between 2000 and 2019, there were three reports that confirmed the tumor suppressing by DPP-4 inhibitors [79,80,81], on the contrary, one study reported that the tumor suppressing effect could not be confirmed.

The number of reported cases is small; thus, it cannot be said whether the effect of suppressing CRC is present or absent. The DPP-4 inhibitor itself is still a novel drug and more data from both clinical and basic researches is required.

4.2.4. Biguanides

The main effect of biguanides (mainly metformin) is to inhibit hepatic gluconeogenesis and to improve peripheral utilization of glucose. However, metformin has been reported to reduce insulin sensitivity in peripheral tissues [82]. The hypoglycemic mechanism by metformin is the suppression of gluconeogenesis and the reduction of hepatic glucose production. In addition, although the contribution is small, it also reduces the absorption of glucose from the intestinal tract and the improvement of glucose uptake and utilization in skeletal muscle and adipose tissue [82]. Other recent studies have shown that metformin suppresses hepatic glucose production by antagonizing glucagon action [83], and metformin improves insulin action by adenosine monophosphate-activated protein kinase (AMPK), which suppresses acetyl-CoA carboxylase phosphorylation [84]. Metformin may also improve glucose homeostasis by interacting with the incretin axis through the action of GLP-1 [85,86].

To date, no epidemiological study was found that examined the role of metformin in the risk of colon carcinogenesis before 2000. During the period 2000 to 2019, 34 epidemiological reports were identified. Among them, 25 reports concluded that metformin was effective in reducing the risk of CRC [65,67,69,72,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109], 8 reports concluded that there is no evidence for the involvement of metformin in CRC risk compared to the control group and other oral hypoglycemic agents [66,68,110,111,112,113,114,115], and 1 report concluded that metformin increased the risk of CRC [64].

Activation of AMPK is a major factor in the mechanism of oncogenesis by metformin (Figure 1). AMPK suppresses hepatic gluconeogenesis, promotes glucose uptake in muscle and adipose tissue, sensitizes cells to insulin, and reduces insulin levels, thereby reducing IGF-1 levels. As described above, enhancement of the IGF signaling pathway is involved in the development of various carcinomas including colon cancer. Downstream responses are downregulated by reduced signal input of the IGF-1 signaling pathway. Furthermore, AMPK also has mTOR-type suppressive action and can reduce cell proliferation signals.

Recently, in addition to simply suppressing the effect of primary carcinogenesis, metformin has been examined for suppression of postoperative recurrence [116], reduction of drug resistance, and enhancement of the effect by combination with existing chemoradiotherapy in CRC patients. Furthermore, there are also increasing number of studies [117,118,119,120] on the suppressive effect of metformin on colon carcinogenesis in non-DM patients [121,122].

4.2.5. Thiazolidinedione

The main effect of thiazolidinedione (rosiglitazone and pioglitazone) is to improve insulin sensitivity at the sites of insulin action, however, there is no action on promoting insulin secretion.

The intracellular target molecule of thiazolidinedione is a receptor-type transcription factor called peroxisome proliferator-activated receptor gamma (PPARγ), which forms heterodimers with retinoid X receptor (RXR) and is a specific DNA recognition sequence. When a thiazolidine drug binds to PPARγ/RXR heterodimer, transcription activity on DNA is increased, and various metabolic actions are exerted through the induction of target gene expression. Thiazolidine promotes cell death of obese large fat cells and increases differentiation into non-obese small fat cells through the adipocyte differentiation. As a result, insulin sensitivity is improved by reducing the factors, such as free fatty acid and TNF-α, which had lowered insulin sensitivity.

We investigated cohort studies that examined various carcinogenic risks including CRC using thiazolidine reported between 2000 and 2019. As far as we searched, no epidemiological study was found that examined thiazolidine and the risk of colon carcinogenesis. There are very few reports on basic research. One of those reports reported that the administration of a thiazolidine derivative in human cancer cell line showed an increase in apoptosis by inhibiting the DNA topoisomerase I activity [123] and a cell proliferation inhibitory effect via inactivation of NFkβ by suppressing GSK3 activity [124]. Clinical research and anti-tumor effects in further basic research is required.

4.2.6. Alpha-glucosidase Inhibitors

The main effect of alpha-glucosidase inhibitors (acarbose, miglitol and voglibose) is to retard carbohydrate digestion and reduce the rate of postprandial glucose absorption [125].

Intestinal brush border membrane-bound intestinal alpha glucosidase has the action of hydrolyzing oligosaccharides, trisaccharides and disaccharides to glucose and other monosaccharides in the small intestine. AGIs reduce the digestive rate of carbohydrates by competitively and reversibly inhibiting this enzyme, and the carbohydrates are less likely to be broken down into glucose molecules, thereby reducing postprandial hyperglycemia and hyperinsulinemia. Acarbose also inhibits pancreatic alpha-amylase, an enzyme that has the function of hydrolyzing starch to oligosaccharides in the lumen of the small intestine [125].

We investigated cohort studies that examined various carcinogenic risks including CRC using AGIs reported between 2000 and 2019. However, there are few reports on the relationship between AGIs and CRC. As far as we could search for clinical studies on the risk of colon carcinogenesis by cohort studies, only two studies were reported in 2015, and both studies reported that AGIs reduces the risk of CRC [39,126].

Although the mechanisms of anti-tumor effects, other than improvement of hyperinsulinemia, are still poorly discussed, several possibilities have been suggested. There are reports that intestinal transit time of feces is shortened in diabetic patients during AGIs administration [127], and reports show that long-term bile acid exposure of intestinal epithelia due to delayed feces induces colorectal tumorigenesis [128]. Based on the above data, it is suggested that shortening the intestinal transit time of stool by AGIs administration may reduce the risk of colorectal tumorigenesis. Further, AGIs have been reported to increase the growth inhibition of transformed cells in the colorectal mucosa and the associated butyrate levels [129]. In addition, the antineoplastic effects of AGIs have been reported, including the prevention of angiogenesis and the inhibition of tumor growth [130]. These findings suggest that AGIs can suppress colorectal tumor formation directly and indirectly [131].

Further studies in clinical research and anti-tumor effects in basic research is required.

4.2.7. SGLT2 Inhibitors

The main effect of SGLT2 inhibitors (dapagliflozin, canagliflozin and empagliflozin) is to prevent reabsorption of glucose filtered through the glomeruli and increase its urinary excretion [132,133].

Blood glucose is filtered in the glomerulus, and the filtered glucose is reabsorbed in the proximal convoluted tubule, mediated by two classes of carrier proteins: the SGLTs and the glucose transporters (GLUTs) [134]. The SGLTs are located on the luminal surface of the proximal tubule epithelium and transport glucose into the cells against the concentration gradient by transporting glucose with sodium. The SGLT1 and SGLT2 are involved in the renal reabsorption of glucose. By the actions of SGLT2 that are selectively expressed in the kidney, about 90% of the glucose filtered from blood is reabsorbed in the proximal tubule of the kidney. About 10% of glucose filtered through the action of low volumes of SGLT1 is reabsorbed in the distal tubule [135]. In patients with type 2 diabetes, renal glucose transport thresholds are increased as compared to normal individuals as a result of SGLT2 upregulation. The intrinsic risk of hypoglycemia with SGLT2 inhibitors is expected to be low, because this effect is insulin independent, and the amount of urinary glucose excretion is determined in part by blood glucose concentrations [136,137,138,139].

We investigated cohort studies that examined various carcinogenic risks including CRC using SGLT2 inhibitors reported between 2000 and 2019. As far as we could search, no clinical studies could be found that examined the relationship between SGLT2 inhibitors and CRC risk. Even in basic research, although there have been reports of suppressive effects on prostate cancer and pancreatic cancer, enough studies on CRC could not be found.

Because of the lack of adequate studies, no further information is available concerning the effect of this drug on the risk of CRC.

5. Conclusions

We examined clinical research and basic research on the colon carcinogenesis suppression effect for the past 20 years by seven groups of oral hypoglycemic drugs. Sulfonylureas did not show an effect on colon carcinogenesis in both clinical and basic research. Biguanides (mainly metformin) have been shown to suppress colon carcinogenesis in many clinical studies and basic studies, because of a long clinical application period. Further, the usefulness of the combination with existing chemotherapy and radiation therapy, post-operative relapse suppression effect has also been studied, and good results are being obtained. Therefore, it seems that the study of metformin’s inhibitory effect on colon cancer is not new. There are not many clinical studies and basic studies with glinides, DDP-4 inhibitors and thiazolidinedione, and their presence or absence of colon carcinogenesis suppressive effect is unknown. Although AGIs and SGLT-2 inhibitors can be expected to suppress colon carcinogenesis, there is still insufficient information on the degree of risk reduction and its mechanism.

The improvement of clinical research and basic research on the colon carcinogenesis suppression effects by glinides, DDP-4 inhibitor, thiazolidinedione, AGIs, and SGLT-2 inhibitors are remaining issues for future studies. Although metformin is being studied, examinations on the CRC-suppressive effect by metformin in a healthy person is also a remaining issue.