Introduction
Worldwide, the incidence of obesity and type 2 diabetes mellitus (T2DM) has increased significantly, creating a substantial economic and healthcare burden. Novel treatment options have been designed with the aim of reducing the numerous complications associated with these metabolic disorders, as well as reducing morbidity and mortality and improving the quality of life of those who suffer from these disorders. These conditions, often described by the term 'diabesity' to highlight their pathophysiological link, increase the risk of cardiovascular disease, thereby increasing morbidity and mortality. Among the most modern therapeutics targeting 'diabesity' are glucagon-like peptide 1 receptor agonists (GLP-1 RAs). GLP-1 RAs are widely used as a glucose-lowering therapy with weight reduction and cardiovascular benefits in T2DM, having also beneficial effects in non-diabetic obesity as a weight loss adjuvant therapy. One of the most recently approved GLP-1 RAs for T2DM is semaglutide. This narrative review highlights recently published data on the effects and safety of semaglutide in diabetic obesity, also emphasizing its cardiovascular benefits and potential side effects, and also delves into the effects of ketogenic diets on colorectal cancer.
The Rise of Diabesity and the Role of GLP-1 RAs
Obesity is a metabolic disease with increasing prevalence over the past decades, becoming an important economic and health care burden. In 2016, the World Health Organization estimated that worldwide more than 650 million adults were obese. 'Diabesity' describes the pathophysiologic link between obesity and T2DM and was first introduced by Sims et al in 1973. In 2019, the International Diabetes Federation estimated that 463 million individuals worldwide have diabetes, projecting that by 2045, there will be >700 million cases. GLP-1 RAs are a drug class that can achieve both glycemic control and weight loss. Their glucose-lowering effects are mainly attributed to glucose-dependent insulin secretion, glucagon inhibition, and decreased gastric emptying. Given the effects on the central nervous system, GLP-1 RA usage may lead to body weight reduction.
Understanding GLP-1 and Semaglutide
GLP-1 is an incretin hormone secreted in a biphasic pattern by the neuroendocrine L cells in the distal ileum and colon after consumption of nutrients, particularly glucose and other carbohydrates. It has a short elimination half-life (1-2 min) due to proteolysis by dipeptidyl peptidase IV and renal elimination. GLP-1 receptors are expressed in numerous organs, mainly occurring in the pancreas, central nervous system (hypothalamus) and the gastrointestinal tract, but also in the heart and kidneys. GLP-1 stimulates insulin secretion from the β-pancreatic cells in a glucose-dependent manner, also promoting β-pancreatic cell survival and proliferation. Furthermore, GLP-1 reduces glucagon secretion by α-pancreatic cells through complex endocrine mechanisms, which include somatostatin stimulation and insulinotropic effects on the β-pancreatic cells. By slowing down gastric emptying, GLP-1 further reduces blood glucose and appetite. This effect on appetite is not only attributed to the delayed gastric emptying but also to its influence on the hypothalamus as a neurotransmitter, particularly on the lateral hypothalamus, and the paraventricular and arcuate nucleus.
At present, six injectable (subcutaneous) GLP-1 RAs and one oral formulation are available in Europe for T2DM treatment. Based on their pharmacological properties, GLP-1 RAs are classified into short- or long-acting agents. Short-acting GLP-1 RAs include exenatide standard-release (Byetta) and lixisenatide (Lyxumia). Their major mechanism of action is based on slowing gastric emptying and lowering postprandial glucose. Long-acting GLP-1 RAs include exenatide modified-release (Bydureon), liraglutide (Victoza), dulaglutide (Trulicity) and semaglutide (Ozempic). Compared to short-acting compounds, long-acting agents have a mechanism of action that mainly comprises stimulating insulin secretion and inhibiting glucagon release, thus influencing both postprandial and fasting glucose. Liraglutide was the first antidiabetic treatment approved as a weight reduction drug in non-diabetic obesity. Sold under the name of Saxenda, liraglutide at a dose of 3.0 mg once daily was approved by both Food and Drug Administration in 2014 and the European Medicines Agency in 2015 for long-term weight management. Semaglutide (Ozempic) is a long-acting GLP-1 RA and its administration is once-weekly subcutaneously at doses of 0.5 and 1.0 mg, with 0.25 mg/week being the initiation dose for the first 4 weeks. The safety and efficacy of semaglutide was investigated in the Semaglutide Unabated Sustainability in Treatment of Type 2 Diabetes (SUSTAIN) clinical trial program.
Semaglutide: Efficacy in Glycemic Control and Weight Loss
When a new diagnosis of T2DM is made, it is crucial to educate the patient regarding the importance of a healthy lifestyle, which includes avoiding excess calories (particularly high-glycemic-index carbohydrates) and increasing physical activity to prevent cardiovascular and metabolic complications. The American Diabetes Association recommends metformin as the first-line drug for T2DM therapy, if not contraindicated. GLP-1 RAs are known to lower blood glucose by stimulating insulin production in the pancreatic β-cells and inhibiting glucagon release by the pancreatic α-cells, combined with slowing gastric emptying and reducing appetite and food consumption. Therefore, given their beneficial effects on glucose metabolism and weight loss potential, GLP-1 RAs are currently recommended as a second-line therapy in T2DM.
Read also: Weight Loss Guide Andalusia, AL
The SUSTAIN clinical trial program has provided comprehensive data on semaglutide's efficacy. Key findings include:
- SUSTAIN-1: Compared the safety and efficacy of semaglutide (0.5 and 1.0 mg) over the course of 30 weeks vs. placebo in patients with T2DM that had no prior drug therapy, only diet and exercise interventions. Compared to the placebo, both doses of semaglutide produced a significant reduction in glycated hemoglobin (HbA1c) [-1.45% with semaglutide 0.5 mg vs. -1.55% with semaglutide 1.0 mg vs. -0.2% with placebo; the estimated treatment difference (ETD) for semaglutide 0.5 mg vs. placebo was -1.43% and the ETD for semaglutide 1.0 mg vs. placebo was -1.53%; P for both doses of semaglutide vs.
- SUSTAIN-2: Compared semaglutide 0.5 and 1.0 mg with sitagliptin 100 mg over the course of 56 weeks in patients with T2DM inadequately controlled with metformin, thiazolidinediones or both. HbA1c reduction was greater with both doses of semaglutide vs. sitagliptin (-1.3% with 0.5 mg semaglutide vs. -1.6% with 1.0 mg semaglutide vs. -0.5% with 100 mg of sitagliptin; ETD -0.77% with semaglutide 0.5 mg and -1.06% with semaglutide 1.0 mg; P for both doses of semaglutide vs. sitagliptin <0.0001 for non-inferiority and superiority).
- SUSTAIN-3: Semaglutide 1.0 mg was compared with once-weekly exenatide in the SUSTAIN-3 clinical trial, which was performed over the course of 56 weeks on 813 adults with T2DM on previous oral antidiabetic agents. A reduction in HbA1c of 1.5% with semaglutide and 0.9% with exenatide was noted (ETD, 0.62%; P<0.0001 for noninferiority and superiority for semaglutide vs.
- SUSTAIN-4: Assessed the safety and efficacy of semaglutide compared to insulin glargine in patients with T2DM inadequately controlled with metformin (with or without sulfonylureas). At week 30, semaglutide at 0.5 and 1.0 mg achieved greater HbA1c reductions than insulin glargine (1.21 vs. 1.64 vs. 0.83%; ETD, -0.38% with semaglutide 0.5 mg and -0.81% with semaglutide 1.0 mg with ETD; P<0.0001 for ETD for both doses of semaglutide vs. insulin glargine).
- SUSTAIN-5: Semaglutide was investigated as an add-on vs. placebo in patients with T2DM on basal insulin, with or without metformin. At week 30, HbA1c exhibited a significant reduction of 1.4 and 1.8% with semaglutide 0.5 and 1.0 mg, respectively, vs. 0.1% with placebo (P for both doses of semaglutide vs. placebo <0.0001).
- SUSTAIN-7: Proved the superiority of 0.5 and 1.0 mg semaglutide in improving the mean HbA1c when compared to dulaglutide 0.75 and 1.5 mg. Semaglutide 0.5 mg reduced the mean HbA1c by 1.5 vs. 1.1% with dulaglutide 0.75 mg. Furthermore, 1.0 mg of semaglutide produced a reduction of 1.8% in HbA1c vs.
- SUSTAIN-8: Compared once-weekly semaglutide 1.0 mg with once-daily canagliflozin 300 mg in patients with T2DM inadequately controlled with metformin. Semaglutide was superior to canagliflozin in reducing HbA1c (ETD, -0.49%; P<0.0001).
- SUSTAIN-9: The efficacy and safety of semaglutide were assessed when added to a sodium glucose cotransporter-2 (SGLT-2) inhibitor in patients with T2DM with poor glycemic control. Semaglutide in addition to a SGLT-2 inhibitor significantly reduced HbA1c (ETD, -1.42%; P<0.0001) compared with placebo.
- SUSTAIN-10: Semaglutide (1.0 mg/week) was compared with liraglutide (1.2 mg/day) in subjects with T2DM treated with 1-3 oral antidiabetic drugs. A total of 577 subjects were randomized to receive either semaglutide or liraglutide. Patients receiving semaglutide had a superior reduction in HbA1c (ETD, -0.69%; P<0.0001). Both treatments had similar safety profiles, with semaglutide having a higher frequency of gastrointestinal reactions compared to liraglutide (43.9 vs.
In SUSTAIN-1, a marked body weight loss was observed with both doses of semaglutide when compared to placebo. Specifically, with semaglutide at 0.5 and 1.0 mg, a weight reduction of 3.73 and 4.53 kg kg was achieved, respectively, while the placebo had an insignificant loss of 0.98 kg (ETD vs. placebo, -2.75 and -3.56 kg with semaglutide 0.5 and 1.0 mg, respectively; P for both doses of semaglutide vs. placebo <0.0001). In SUSTAIN-2, at week 56, a weight loss of 4.3 kg with semaglutide 0.5 mg and 6.1 kg with semaglutide 1.0 mg, and 1.9 kg with sitagliptin 100 mg was achieved (ETD, -2.35 kg with semaglutide 0.5 mg and -4.20 kg with semaglutide 1.0 mg vs. sitagliptin; P for both doses of semaglutide vs. The SUSTAIN-3 trial indicated that semaglutide-treated subjects achieved a greater weight reduction when compared to exenatide-treated subjects (-5.6 vs. -1.9 kg; ETD -3.78 kg; P<0.0001). SUSTAIN-4 compared semaglutide vs. insulin glargine. Body weight loss was observed in semaglutide-treated subjects and at week 30, a loss of 3.5 kg with semaglutide 0.5 mg and 5.2 kg with semaglutide 1.0 mg, compared to a weight gain of 1.15 kg with insulin glargine was observed. This result came with no surprise given the appetite-reducing effects of GLP-1 and the anabolic effects of insulin. When added to basal insulin, in patients with T2DM with or without metformin treatment, semaglutide produced a significant body weight reduction vs. placebo according to the results of SUSTAIN-5 (-3.7 kg with semaglutide 0.5 mg vs. -6.4 kg with semaglutide 1.0 mg vs. -1.4 kg with placebo; P for both doses of semaglutide…
Semaglutide and Cardiovascular Benefits
Semaglutide arm had a lower primary outcome (first occurrence of cardiovascular death, nonfatal myocardial infarction or nonfatal stroke) occurrence vs. placebo (6.6 vs. 8.9%); semaglutide arm had lower rates of new or worsening nephropathy vs. placebo (3.8 vs. 6.1%); semaglutide arm had a higher incidence of retinopathy complications vs. placebo (3.0 vs.
Ketogenic Diet and Colorectal Cancer
Worldwide, colorectal cancer (CRC) is the third most prevalent cancer, with over 1.9 million new cases recorded in 2020. CRC incidence and mortality rates have been associated with various lifestyle-related risk factors, in particular dietary habits. Recent years have seen a rise in CRC incidence in persons aged 50 years and under, further underlining the importance of studying these factors. The ketogenic diet (KD) has been shown to possess anti-cancer properties in the context of CRC, and several studies have focused on the importance of the restriction of energy intake to slow the growth of rapidly proliferating cells, as well as on anti-cancer signaling through ketone bodies. One recently published study investigated the effects of the KD on CRC development and described a mechanism through which KD exerts its anti-cancer effects, namely through the ketone body β-hydroxybutyrate acting on the surface receptor Hcar2, inducing the transcriptional regulator Hopx, thus altering gene expression and inhibiting cell proliferation. These studies often focus on the effects of KD on the cancer cells themselves, and rarely on other cells of the tumor microenvironment, such as immune cells and cancer-associated fibroblasts, or on external factors, such as the gut microbiome. Specifically in CRC, cancer progression is accompanied by a state of dysbiosis of the gut microbiome. The study of the role of the different bacteria in CRC is of particular interest, considering the affected tissue type is in contact with the gut microbiome. Until recently, few studies had investigated the effects of the KD on the gut microbiome in health and disease, but this has now changed, with a number of new studies published over the last three years. These studies remain largely descriptive regarding dietary-induced gut microbiome changes. However, the role of the gut microbiome in maintaining the reported anti-cancer effects of the KD in the context of CRC has not yet been investigated. Therefore, we set out to investigate the importance of the gut microbiome in the context of KD consumption in CRC.
Experimental Design
We first generated a humanized gut microbiome mouse model of CRC, through the transplantation of stool samples from five healthy human donors. We then combined this model with a therapeutically administered KD and were able to confirm the cancer-suppressing properties of the KD. Most importantly, we could demonstrate, for the very first time, the causal role of the gut microbiome in maintaining this effect, through transplantation of the microbial community. We report long-lasting alterations in gut microbiome function in the absence of maintained selective pressure by the KD. In our experiments, fecal free stearic acid levels were increased not only upon KD consumption, but also upon cecal transplantation, suggesting that the shift in the microbial community contributes to the changes in the fatty acid pool. Importantly, stearate-producing members were enriched in ketogenic conditions, whereas consumers were depleted and supplementation of stearic acid reduced tumor burden in vivo.
Read also: Beef jerky: A high-protein option for shedding pounds?
Results: KD Reduces Tumor Burden and Modulates Immune Response
To study the relevance of the gut microbiome in CRC development in a model relevant to human disease, we first established a representative healthy human microbiome by mixing fecal samples from five healthy donors (5HD; Supplementary Fig. 1a for patient metadata). The achieved 5HD mix recapitulated the compositions of the individual donor samples, as evidenced by family, genus and species level analysis of whole genome shotgun sequencing (WGS, Supplementary Fig. 1b-d, respectively), while fecal metabolite levels varied across samples (Supplementary Fig. 1e). We performed fecal microbial transplantation (FMT) of the 5HD mix into germ-free (GF) and antibiotic-treated specific-pathogen-free (SPF) mice in an AOM/DSS CRC model (Fig. 1a, b, details can be found in the Methods section) followed by a switch from standard to ketogenic diet after the last cycle of DSS. We utilized both GF and SPF mice to leverage the fully functional immune system of SPF mice to study immune responses in CRC and the unique advantage of GF mice allowing engraftment of a human microbiome into mice in the absence of competition with a resident mouse microbiome. Antibiotic-treated SPF mice were subjected to bi-weekly FMT during the induction of CRC to maintain enrichment of the donor fecal microbiome in relation to the residual host mouse microbiome. FMT was ceased at diet change to allow for a natural evolution of the gut microbiome in response to the diet and CRC progression. No significant difference in body weight between the standard diet (SD) and ketogenic diet (KD) groups was observed throughout the experiments, although dips in mouse body weight were observed following the antibiotic treatment and following each cycle of DSS (Supplementary Fig. 2a (GF) and 2b (SPF)). There was no difference in colon length between groups (Supplementary Fig. 2c, GF). KD-fed mice exhibited a lower colonic tumor burden (Fig. 1c (GF) and Fig. 1d (SPF)) and a smaller tumor size (Fig. 1e, f and g (SPF)) than SD-fed mice.
Characterization of full-length lamina propria (LP), mesenteric lymph node (MLN) and spleen immune cell profiles (Supplementary Fig. 2d (GF) and 2e (SPF)) showed a reduction in LP CD4+ T cell populations in KD-fed mice. We particularly observed a decrease in IL17-producing T cells (CD4+ IL17+) (Supplementary Fig. 2f), a pathogenic subpopulation of CD4+ T cells known to play a critical role in CRC development as evidenced by reduced expression of IL-6, STAT3, IFNγ, and TNFα, leading to fewer and smaller tumors in IL-17A-deficient mice12. Only LP natural killer T cells and type 3 innate lymphoid cells (GF LP) were increased - both have been associated to gut homeostasis and health13. The expression of colonic S100A8, a subunit of calprotectin (a biomarker of inflammation used in inflammatory bowel disease14), as well as fecal lipocalin-2 were however unchanged in KD-fed mice, suggesting an overall similar inflammatory milieu of the gut lumen (Supplementary Fig. 2g (GF), h and i (SPF)). Thus, it remains unclear whether the observed reduction in immune cell populations is a result mediated by the KD or if it is a direct consequence of decreased tumor development. Conducting time-course analyzes at various stages of the AOM/DSS-induced carcinogenic process would be essential to elucidate the temporal sequence of events leading to the observed reduction in colonic tumor burden.
To further assess the importance of KD in modulating inflammation in CRC, we employed a low-grade inflammatory model, in which mice received four doses of AOM and were randomized into two subgroups - GF and 5HD mix FMT, each with a KD and SD branch (Supplementary Fig. 3a, b). We did not observe any significant differences in colonic tumor burden between groups (Supplementary Fig. 3c), nor in immune cell profiles (Supplementary Fig. 3d), suggesting that the KD specifically exerted its anti-cancer effect in an inflammatory setting. The differences in immune populations between the KD and SD in the AOM/DSS model, which are absent in this low-grade inflammatory model, may be due to the lack of damage-associated molecular patterns (DAMPs) from DSS-induced epithelial damage. This suggests that the KD attenuates chronic inflammation activated by such damage, potentially leading to a reduced tumor burden in this inflammation-driven CRC model. Of note, long-term KD feeding induced fatty liver and fibrosis in all sub-groups (Supplementary Fig. 3e), as previously reported for long-term KD consumption15.
Metabolic Changes Induced by KD
Ordination analyzes based on the underlying β-diversity among host plasma samples collected from animals undergoing ketogenic dietary intervention in the AOM/DSS inflammatory model of CRC, revealed clustering based on dietary condition (Supplementary Fig. 4a (GF) and Supplementary Fig. 4b (SPF)). β-hydroxybutyrate plasma levels were elevated in KD-fed mice (Supplementary Fig. 4c (GF) and 4d (SPF)), confirming ketosis. KD-fed mice presented altered amino acid, fatty acid, sugar, carbohydrate, vitamin, and bile acid metabolism (Supp. Data File 1 and Supplementary Fig. 4e (GF), Supp. Data File 2 and Supp Fig. 4f (SPF)). Furthermore, mannose was increased in KD-fed mice (Supp. Data Files 1 (GF) and 2 (SPF)). Interestingly, mannose has been reported to possess antitumor properties16, impairing tumor cell growth in vitro and in vivo17, and targeting tumor-associated macrophages18. While treatment with β-hydroxybutyrate showed no effect, treatment with the ketone bodies acetone and sodium acetoacetate tended to reduce the differentiation of naive T cells into Th17 cells (Supplementary Fig. 4g and h), which play an important role in the modulation of colorectal tumorigenesis19 and are affected by KD in a dose-dependent manner5. Treatment with a commercially available lipid mixture mimicking KD (containing arachidonic, linoleic, linolenic, myristic, oleic, palmitic and stearic acid) led to a dose-dependent reduction in Th17 differentiation (Supplementary Fig. 4i), suggesting that Th17 cells are susceptible to fatty acid modulation. Collectively, these data indicate that KD induces systemic functional metabolic changes in the host, which are not limited to the induction of ketosis.
Gut Microbiome's Role in KD's Anti-Cancer Effects
The gut microbiome may affect CRC development either by directly interacting with host cells (attachment, invasion and translocation), or indirectly, through bacterial metabolism (genotoxins, SCFAs and others)20. We first excluded bacterial translocation as a primary factor in the KD-mediated phenotype, as mucus layer thickness (Supplementary Fig. 5a, b (GF)), LPS levels in plasma (Supplementary Fig. 5c (GF) and 5 d (SPF)), and occludin expression (Supplementary Fig. 5e, f (SPF)) remained unaltered upon KD consumption, in line with Ang et al.5. We next performed cecal microbial transplantation (CMT) of samples collected from the SPF diet mice (Fig. 1d) into a next generation of SD-fed GF mice undergoing the same cancer model (Fig. 2a, Supplementary Fig. 6a). We did not observe a statistically significant difference in colonic occludin expression in ketogenic cecal recipient mice (KC, Supplementary Fig. 6b and c) when compared to standard cecal recipients (SC), however we did detect a decrease in S1008A levels (Supplementary Fig. 6d). Strikingly, KC mice exhibited a lower colonic tumor burden than SC mice (Fig. 2b). Average colonic tumor size per mouse did not differ between groups (Fig. 2c), however, KC mice exhibited a smaller proportion of large tumors (Fig. 2d, e). Furthermore, colonic mRNA IL17 levels were decreased in KC mice compared to SC mice (Supplementary Fig. 6e), in line with the decreased number of IL17-producing T cells in the lamina propria of KD-fed mice (Supplementary Fig. 2f). Altogether, our results demonstrate the causal role of the gut microbiome in mediating the previously observed KD-driven anticancer effect.
Read also: Inspiring Health Transformation
KD's Impact on Gut Microbiome Composition
Ordination analyzes based on the underlying β-diversity among fecal samples collected from KD- and SD-fed AOM/DSS mice revealed clustering of 5HD mix used for FMT with the individual healthy donor stool samples and with the murine stool samples collected three days after FMT (TG, Fig. 3a (GF) and 3b (SPF)), confirming the validity of the 5HD mix used for transplant and its engraftment. Furthermore, we observed increasing differences between KD and SD samples over time (Fig. 3c (GF) and 3d (SPF)). Early time points cluster together (T0, collected at diet change), while later time points diverge (T1-T4). Differential analysis revealed Phocea spp., Faecalitalea spp. (Fig. 3e (GF), Supp. Data Files 3 (GF) and 4 (SPF)) and Akkermansia spp., Intestimonas spp., Lachnoclostridium spp., Bilophila spp. and others (Fig. 3f (SPF)), to be more abundant in KD, while Barnesiella spp., Eisenbergiella spp., Ruminococcus 1 and 2 spp. (GF), and Bifidobacterium spp., Turicibacter spp. (SPF), and others were depleted. Furthermore, the differential abundance of various members of Barnesiellaceae spp., Erysipelotrichaceae spp., Lachnospiraceae spp. and Ruminococcaceae spp. suggested that dietary composition affects bacterial members of the same families in different manners.