Abstract
Oral semaglutide has potent anti-hyperglycemic efficacy in phase III trials. However, the complicated dosing instructions hamper to use this drug; therefore, we evaluated the efficacy and safety of oral semaglutide in subjects with type 2 diabetes in a real-world clinical setting. In this multi-center retrospective observational study, we analyzed subjects with type 2 diabetes newly treated with an oral semaglutide for >6 months at four medical centers located in Sapporo, Japan. The changes in glycated hemoglobin, body weight, and other metabolic parameters were evaluated and any adverse event leading to semaglutide discontinuation were recorded from February 2021 to December 2022. This study was registered with the University Hospital Medical Information Network Center (UMIN000050583). Of 543 subjects who met the inclusion criteria, data for 434 subjects (age 55.5 ± 12.6 years; body mass index 29.6 ± 6.0 kg/m2) were analyzed. After a 6 months of observation period, semaglutide 3 mg, 7 mg, or 14 mg was used by 55 (12.7%), 241 (55.5%), and 138 (31.8%) of subjects, respectively. Both glycated hemoglobin and body weight significantly improved: 7.65 ± 1.11% to 6.88 ± 0.91% (p < 0.001) and 80.2 ± 19.2 kg to 77.6 ± 19.2 kg (p < 0.001), respectively. Efficacy was also confirmed in the subgroup switched from other anti-hyperglycemic agents, including dipeptidyl peptidase-4 inhibitors. In total, 154 subjects had symptomatic gastrointestinal symptoms and 39 (7.2%) were discontinued semaglutide due to the adverse events. None of the participants experienced severe hypoglycemic events. Oral semaglutide in subjects with type 2 diabetes improved glycemic control and body weight in a real-world clinical setting.
SUBJECTS with type 2 diabetes (T2D) are at a high risk of microvascular and macrovascular complications, leading to an impaired prognosis and quality of life [1, 2]. Comprehensive care involving not only glycemic control but also improvements in body weight, blood pressure, and lipid metabolism are required to suppress the onset and progression of such complications [3, 4]. Among the various anti-hyperglycemic agents, regimens that have been shown to reduce all-cause mortality and inhibit the development of cardiovascular events and microvascular complications in high-risk subjects have been recommended in the recent guidelines [5, 6].
Glucagon-like peptide-1 receptor agonists (GLP-1RAs) potently improve glycemic control in a glucose-concentration-dependent manner [7]. The superiority of GLP-1RAs over other anti-diabetic medications for glucose lowering effects has been demonstrated in previous meta-analyses [8]. Another distinct feature of GLP-1RAs is body weight loss through appetite suppression and inhibition of gastric peristalsis [9, 10]. In addition, GLP-1RAs are recommended in subjects with T2D at a high risk of cardiovascular diseases based on the findings of cardiovascular outcome trials [5, 6, 11-13]. Despite these positive features, major factors for GLP-1RA non-use are related with the injectable formulation.
The only available oral GLP-1RA, oral semaglutide, has similar pharmacological actions to subcutaneous GLP-1RAs. Phase III trials of oral semaglutide showed that its glucose lowering and weight loss effects are more potent than those of subcutaneous GLP-1RAs agents [14, 15]. However, oral semaglutide is complicated to administer because it needs to be taken in a fasting state with no more than 120 mL water. And its efficacy is not fully achieved unless it is taken appropriately. In real-world clinical practice, unlike in phase III trials, patients are often treated with multiple anti-diabetic agents; therefore, it is unclear whether the efficacy of oral semaglutide can be achieved in such situations. Taken together, it is important to verify the efficacy of semaglutide in clinical practice.
In the present study, we aimed to evaluate the efficacy and safety of oral semaglutide in Japanese subjects with T2D in a real-world clinical setting and determine whether the efficacy varied with concomitant treatments.
Materials and Methods
Study subjects
Participants who were treated at four medical centers and clinics specializing in diabetes located in Sapporo, Japan were recruited in this cohort (the Sapporo-Oral SEMA study). Eligible criteria were as follows: 1) a diagnosis as T2D; 2) 20 years of age or more; and 3) newly treated with oral semaglutide from February 2021 to December 2022. Patients who had an allergy to semaglutide, experienced diabetic ketoacidosis and/or severe infections, or were incompatible with the trial for other reasons (as determined by the physicians) were excluded from the analysis. There was no restriction for the use of concomitant and/or previous medications for diabetes including subcutaneous GLP-1RAs.
Study protocol
This was a multicenter retrospective observational cohort study. Clinical data for 9–12 months of observation were obtained from the participants’ medical records. The baseline was defined as the date when oral semaglutide was initiated. Regarding glycated hemoglobin (HbA1c) and body weight, the values 6 months prior to semaglutide administration were collected as a reference. When such data were unavailable, data from 3 months before the administration were used. Participants who stopped oral semaglutide within 6 months after initiation were also recorded to investigate the adverse events (AEs) related to oral semaglutide treatment. Data were collected for each visit as follows: 3–6 months before the initiation of oral semaglutide; at baseline; and 1, 3, and 6 months after starting semaglutide treatment. Oral semaglutide was initiated at 3 mg once daily, followed by the escalation to 7 mg after at least 4 weeks and then scaled up to 14 mg, if necessary, based on the judgment of the physicians in charge and the agreement of the subject. Oral semaglutide was given as an addition to current anti-hyperglycemic agents (add-on) or as a switch from dipeptidyl peptidase-4 (DPP-4) inhibitors, other GLP-1RAs, or other agents. Participants were instructed to take oral semaglutide in the morning in a fasted state with 120 mL of water, at least 30 min before breakfast and any other oral medication. The treatments were supervised through each medical care center.
There were two primary endpoints in this study: 1) changes in glycated hemoglobin (HbA1c) from baseline; and 2) safety of oral semaglutide assessed by any AEs leading to drug discontinuation. The AEs of special interest included gastrointestinal symptoms and hypoglycemia. Severe hypoglycemia was defined as symptoms requiring external assistance. The secondary endpoints comprised changes in 1) body weight, 2) blood pressure, 3) pulse rate, 4) liver enzyme activity, 5) serum creatinine concentration and estimated glomerular filtration rate, 6) urinary albumin/creatinine ratio, 7) factors associated with the changes in HbA1c, and 8) the incidences of any AEs. Sub-analyses were also conducted to assess the efficacy of semaglutide on these parameters in participants who switched from any anti-hyperglycemic agent to semaglutide versus those who received semaglutide as add-on to their current therapies.
An opt-out consent procedure was used in this study. This study was registered with the University Hospital Medical Information Network Center (UMIN000050583) and the study protocol was approved by the Institutional Review Board of the Japan Clinicians Diabetes Association and Hokkaido University (approval number 022-0200). The study was carried out in accordance with the principles of the Declaration of Helsinki, as revised in 2013.
Statistical analysis
Results are expressed as mean ± standard deviation for normally distributed data or median (interquartile range) for non-normally distributed data. Changes from baseline values are expressed as mean/median and 95% confidence interval (95% CI). Categorical variables are expressed as numbers and percentages. Differences between two groups were evaluated using Student’s t-tests. The results within the groups were compared by a paired t-test or Wilcoxon signed-rank test. Changes in variables from baseline are expressed as mean or median (95% CI), and differences among the groups were analyzed using the Kruskal-Wallis test followed by the Dunn post-hoc test. p < 0.05 indicated statistical significance. All statistical analyses were performed using GraphPad Prism (GraphPad Software, San Diego, CA, USA) or JMP Pro 16.0.0 (SAS, Cary, NC, USA).
Results
Study population and semaglutide doses
A total of 543 subjects with T2D met the inclusion criteria during the study period. Of them, 109 discontinued oral semaglutide treatment within 6 months. Ninety-two subjects (21.2%) who were newly treated with oral semaglutide and 342 subjects (78.8%) who switched their treatment regimens from other anti-hyperglycemic agents to oral semaglutide were analyzed (Fig. 1). The demographic and clinical characteristics of the subjects at baseline are summarized in Table 1. The mean age, body mass index, and HbA1c values were 55.5 ± 12.6 years, 29.6 ± 6.0 kg/m2, and 7.65 ± 1.11%, respectively. Notably, 68% of participants were treated with DPP-4 inhibitors and 6.9% with other GLP-1RAs, both of which were switched to oral semaglutide (Table 1 and Supplementary Table 1). The add-on group, in which oral semaglutide was simply added, was younger, more obese, and had less medication use for diabetes compared with the switch group (Supplementary Table 2). Dosages of oral semaglutide were adjusted by the attending physicians during treatment, resulting in 55 subjects treated with 3 mg/day, 241 with 7 mg/day, and 138 with 14 mg/day at 6 months (Supplementary Fig. 1). The highest dose of oral semaglutide was selected in subjects with higher body weight and/or those treated with other GLP-1RAs (Supplementary Table 3).
Table 1
Demographics and characteristics at baseline
| Variables |
Subjects (n = 434) |
| Sex (male/female) |
245/189 |
| Age (years) |
55.5 ± 12.6 |
| HbA1c (%) |
7.65 ± 1.11 |
| Body weight (kg) |
80.2 ± 19.2 |
| Body mass index (kg/m2) |
29.6 ± 6.0 |
| Duration of diabetes, n (%) |
| <5 years |
136 (31.3) |
| 5 to 10 years |
80 (18.4) |
| 10 to 15 years |
77 (17.7) |
| 15≤ years |
141 (32.5) |
| Antihyperglycemic agents, n (%) |
| SGLT2 inhibitors |
300 (69.1) |
| Metformin |
297 (68.4) |
| DPP-4 inhibitors |
295 (68.0) |
| Sulfonylureas |
74 (17.1) |
| Glinides |
41 (9.4) |
| Alfa-glucosidase inhibitors |
32 (7.4) |
| GLP-1 receptor agonists |
30 (6.9) |
| Thiazolidinediones |
21 (4.8) |
| Insulin injection |
19 (4.4) |
Data are mean ± standard deviation or number (%). HbA1c, glycated hemoglobin; SGLT2, sodium-glucose cotransporter 2; DPP-4, dipeptidyl peptidase-4; GLP-1, glucagon-like peptide-1.
Before the induction of oral semaglutide, glycemic control assessed by HbA1c deteriorated from 7.43 ± 1.14% to 7.65 ± 1.11% (p < 0.001). Semaglutide treatment gradually improved HbA1c over time, eventually reaching 6.88 ± 0.91% at 6 months (p < 0.001 vs. baseline; Fig. 2A). In the subanalysis by the method of semaglutide induction, both add-on and switching from current medications significantly reduced HbA1c, but the reduction was greater in the add-on group (Supplementary Fig. 2). The changes in HbA1c in this add-on group were significantly correlated with baseline HbA1c levels, but not with body mass index (Fig. 3). By dosage of oral semaglutide, all dosages were effective in lowering HbA1c both in the add-on and switching from other oral medications including DPP-4 inhibitors, although the efficacy was partially limited in switching from injectable GLP-1RAs (Table 2 and Supplementary Fig. 3).
Table 2
Changes in HbA1c over 6 months according to administration method and dose of oral semaglutide
| Variable |
Baseline
(3 mg; n = 55) |
Mean change at
6 months |
Baseline
(7 mg; n = 241) |
Mean change at
6 months |
Baseline
(14 mg; n = 138) |
Mean change at
6 months |
p-value for changes in HbA1c among the groups |
| Total cohort |
7.47 ± 1.17 |
–0.72***
(–0.97 to –0.47) |
7.64 ± 0.96 |
–0.75***
(–0.86 to –0.65) |
7.73 ± 1.30 |
–0.81***
(–0.99 to –0.62) |
0.235 |
| Add-on subgroup |
7.43 ± 1.25 |
–0.90**
(–1.44 to –0.36) |
7.29 ± 0.97 |
–0.97***
(–1.25 to –0.69) |
7.76 ± 1.87 |
–1.29***, ††
(–1.85 to –0.73) |
0.759 |
| Switch from DPP-4 inhibitors subgroup |
7.53 ± 1.14 |
–0.74***
(–1.08 to –0.40) |
7.72 ± 0.94 |
–0.70***
(–0.82 to –0.59) |
7.89 ± 1.02 |
–0.70***
(–0.85 to –0.54) |
0.404 |
| Switch from injectable GLP-1RAs subgroup |
7.67 ± 1.92 |
–0.07
(–2.21 to 2.08) |
7.92 ± 0.58 |
–0.64*
(–1.16 to –0.12) |
7.05 ± 0.99 |
–0.37
(–0.87 to 0.13) |
0.353 |
Values are presented as mean ± standard deviation or mean change (95% confidence interval). * p < 0.05, ** p < 0.01, *** p < 0.001 vs. baseline (Student’s t-test). †† p < 0.05 vs. switch from injectable GLP-1RAs subgroup (Kruskal-Wallis test followed by Dunn’s multiple comparison test). P-values were calculated using Kruskal-Wallis test. DPP-4, dipeptidyl peptidase-4; GLP-1RAs, glucagon-like peptide-1 receptor agonists; HbA1c, glycated hemoglobin.
Regarding effects on body weight, oral semaglutide effectively reduced body weight from 80.2 ± 19.2 kg to 77.6 ± 19.2 kg (p < 0.001; Table 3 and Fig. 2B). This effect occurred regardless of the oral semaglutide dosage, although the extent of body weight loss was less in subjects who switched from subcutaneous GLP-1RA to oral semaglutide (∆ body weights were: –2.5 kg in the add-on subgroup, –2.9 kg in the switch from DPP-4 inhibitors subgroup, and –1.8 kg in the switch from GLP-1Ras subgroup; p < 0.05 between the switch from DPP-4 inhibitors and switch from GLP-1Ras subgroups) (Supplementary Table 4). During the 6-month treatment period, systolic and diastolic blood pressure, levels of hepatic liver enzymes, lipid profiles assessed by triglyceride, and low-density lipoprotein cholesterol significantly improved. Although semaglutide resulted in a small but significant deterioration in renal function, urinary albumin excretion was reduced after oral semaglutide induction (Table 3).
Table 3
Comparisons of the variables between baseline and 6 months
|
Baseline |
6 months |
p-value |
| Body weight (kg) |
80.2 ± 19.2 |
77.6 ± 19.2 |
<0.001 |
| Body mass index (kg/m2) |
29.6 ± 6.0 |
28.5 ± 6.6 |
<0.001 |
| Systolic blood pressure (mmHg) |
132.8 ± 15.4 |
127.8 ± 15.6 |
<0.001 |
| Diastolic blood pressure (mmHg) |
80.3 ± 11.8 |
78.9 ± 10.9 |
0.008 |
| Pulse rate (/min) |
85.8 ± 13.4 |
86.7 ± 13.2 |
0.064 |
| AST (U/L) |
29.1 ± 17.3 |
26.4 ± 12.3 |
<0.001 |
| ALT (U/L) |
37.2 ± 27.4 |
34.1 ± 23.9 |
0.003 |
| γ-GTP (U/L) |
34 (22, 57) |
30 (19, 51) |
<0.001 |
| eGFR (mL/min/1.73 m2) |
78.9 ± 23.4 |
76.3 ± 22.5 |
<0.001 |
| Creatinine (mg/dL) |
0.77 ± 0.26 |
0.79 ± 0.26 |
<0.001 |
| UACR (mg/g.Cre)† |
19.7 (9.0, 54.7) |
14.0 (6.7, 36.3) |
0.002 |
| HDL-C (mg/dL) |
57.2 ± 14.8 |
57.1 ± 14.1 |
0.995 |
| Triglyceride (mg/dL) |
135 (96, 198) |
124 (90, 184) |
0.001 |
| LDL-C (mg/dL) |
94.0 ± 28.0 |
87.8 ± 25.6 |
<0.001 |
Data are mean ± standard deviation or median (interquartile). Data were analyzed using Student’s t-test or the Wilcoxon signed-rank test. † Data were obtained from 346 subjects. AST, aspartate aminotransferase; ALT, alanine aminotransferase; γ-GTP, gamma-glutamyl transpeptidase; eGFR, estimated glomerular filtration rate; UACR, urinary albumin/creatinine ratio; HDL-C, high-density cholesterol; LDL-C, low-density cholesterol.
Then, patient background characteristics were compared with and without improvement of HbA1c and BMI for 6 months. Subjects whose HbA1c improved over 6 months had a lower baseline BMI and a higher HbA1c (Supplementary Table 5). Although the patient population with a lower BMI over 6 months tended to display a lower blood pressure and significantly lower γ-GTP and eGFR, no differences in baseline glycemic management or body weight were observed (Supplementary Table 6).
Safety of oral semaglutide induction
The breakdown of adverse events is summarized in Supplementary Tables 7 and 8. Of the 543 subjects who met eligibility criteria, 154 (28.4%) had symptomatic gastrointestinal symptoms. Most were able to continue oral semaglutide, but 39 discontinued because of intolerable gastrointestinal symptoms (Supplementary Fig. 4 and Supplementary Table 9). Of these 39 patients, 5 were in the addition group, 30 switched from DPP-4 inhibitors, and 2 switched from other GLP-1RAs. The breakdown of severe gastrointestinal symptoms was as follows: nausea (n = 26), gastric pain (n = 5), diarrhea (n = 5) and other symptoms (n = 3). Other symptomatic AEs leading to drug discontinuation were appetite loss (n = 2), general fatigue (n = 2), leg cramps (n = 1), and hypoglycemia (n = 1). Notably, severe hypoglycemia was not identified during the study period.
Discussion
The present cohort study evaluated the efficacy and safety of oral semaglutide for T2D in a real-world clinical setting. HbA1c was significantly improved, and body mass was reduced by 2.5 kg (3.2% reduction from baseline). Of note, semaglutide showed efficacy as both an add-on therapy and in subjects who switched from other anti-hyperglycemic agents including DPP-4 inhibitors. Although gastrointestinal symptoms as AEs of oral semaglutide were frequent, most of the subjects continued semaglutide and the discontinuation rate was similar to that of phase III trials [15, 16].
A recently published meta-analysis of randomized controlled trials of oral semaglutide showed that 7 mg and 14 mg reduced HbA1c by 1.06% and 1.10%, respectively, compared with placebo [17]. This finding is comparable to that of the current analysis. By contrast, HbA1c reduction with 7 mg and 14 mg of semaglutide compared with other anti-diabetic agents was only 0.26% and 0.38%, respectively, in the same meta-analysis [17]. Moreover, a phase III trial of the efficacy of oral semaglutide (3 mg/day) vs. sitagliptin (100 mg/day) resulted in less HbA1c reduction with semaglutide than sitagliptin (–0.6% vs. –0.8%, respectively) [16]. Unlike the previous report, our trial resulted in significant HbA1c reduction by 0.74% in subjects switched from DPP-4 inhibitors to 3 mg semaglutide (Table 2). The reasons for such discrepancy might be explained by the differences in the titration method of semaglutide administration. The dosage of semaglutide was fixed in each treatment arm in the phase III trials. By contrast, in real-world clinical settings (as in our study), low-dose semaglutide would be sufficient in certain individuals but dosage would be increased up to 7 mg or 14 mg if clinical response was inadequate. As shown in an analysis of exposure-response pharmacodynamics, oral semaglutide blood concentrations showed great variability compared with the injectable formulation even in strictly conducted phase III trials [18]. Considering that the reduction of HbA1c was not different for the different dosages in our trial, the appropriate dose to exert HbA1c-lowering effects may vary widely from individual to individual. Thus, a switching strategy from conventional anti-hyperglycemic agents including DPP-4 inhibitors might be a treatment option to achieve preferable glycemic management.
One reason to initiate oral semaglutide treatment is the beneficial effect on excessive body weight and metabolic disorders. In previous phase III trials, high-dose oral semaglutide resulted in greater body weight reduction than conventional GLP-1RAs [14, 15]. As anticipated, high-dose semaglutide was selected in the subjects with high body weight (Supplementary Tables 3 and 4). Baseline BMI was not an indicator of HbA1c reduction after semaglutide induction in our study. One explanation is that our trial included subjects switching from injectable GLP-1RAs. Another possible reason is the individual differences in the optimal dosage as described above [18]. The high frequency of subjects switched from conventional GLP-1RAs to 14 mg semaglutide despite acceptable glycemic control might be explained by its efficacy on body weight loss (Table 2). Oral semaglutide administration also significantly improved albuminuria but slightly reduced estimated glomerular filtration rate. Numerous reports have argued for the renoprotective effects of GLP-1RAs both via direct and indirect mechanism. The direct mechanisms are renal vasodilation [19], inhibition of oxidative stress in the glomerulus [20], natriuresis [21], and the indirect mechanisms are improvement of metabolic parameters including glycemic control, blood pressure, and excessive body weight [22]. GLP-1RAs mainly reduce body fat rather than muscle affecting serum creatinine value [23, 24]. A decrease in estimated glomerular filtration rate during the observation period in the current trial might be explained by optimization of glomerular hyperfiltration via improved metabolic abnormalities, including hyperglycemia and excessive body weight [25].
The present study had several limitations. First, the study design of a single-arm, retrospective open-labeled study can yield selection bias; i.e., semaglutide may have been administered to subjects who were expected to receive better effects. To reduce the bias, we enrolled all subjects with consent who were treated with oral semaglutide during a certain period. Second, the unique method of administration required for oral semaglutide might have affected patient attitudes toward diabetes and medication adherence. Our retrospective study design could not assess the individual medication adherence. Third, changes in medications for concomitant comorbidities were not restricted according to the study design. However, only a small percentage of cases changed concomitant medications after semaglutide administration (29 out of 434 analyzed cases), and most of these cases were reductions in the medications for diabetes. Finally, semaglutide could not be used in combination with incretin-related drugs, thus treatment had to be switched in subjects being treated with DPP-4 inhibitors or existing GLP-RAs. Because of the diversity of the baseline treatment regimens, it was difficult to compare the efficacy of oral semaglutide with certain pre-treatment drugs conclusively. Although categorical analyses were performed, prospective switching studies from incretin-related drugs to semaglutide are desirable.
In conclusion, induction of oral semaglutide in subjects with T2D could yield additional benefits on metabolic abnormalities including hyperglycemia, overweight, dyslipidemia and hepatorenal dysfunction without causing severe hypoglycemia in real-world clinical settings. Although certain precautions to prevent gastrointestinal symptoms and select appropriate patients should be undertaken, oral semaglutide may contribute to improve the quality of life with the better management of the comorbidities in the subjects with T2D.
Acknowledgments
The authors would like to express their gratitude to all the patients and staff who participated in this study. The authors would particularly like to thank Dr. Hideaki Miyoshi for his contribution in conceiving, designing, and supervising the original study. The manuscript was edited by Carol Wilson, PhD from Edanz (https://jp.edanz.com/ac).
Disclosure of Ethical Statements
Approval of the research protocol: institutional review board of Hokkaido University Hospital Clinical Research and Medical Innovation Center (022-0200).
Informed Consent: An opt-out consent procedure was used.
Approval date of Registry and the Registration No. of the study/trial: This study was registered in the University Hospital Medical Information Network Clinical Trials Registry (UMIN000050583).
Animal Studies: N/A.
Disclosure
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
The authors declare the following financial interests/personal relationships that may be considered as potential competing interests: Nomoto H, Nakamura A, and Atsumi T have received honoraria for lectures and research funding from the organizations listed below. Nomoto H has received honoraria for lectures from Novo Nordisk Pharma Ltd. and Sumitomo Pharma Co., Ltd. and has obtained research support from Kowa Pharmaceutical Co. Nakamura A has received honoraria for lectures from Novo Nordisk Pharma Ltd. and has obtained research support from Mitsubishi Tanabe Pharma, Nippon Boehringer Ingelheim Co., Kowa Co., Ltd., and Taisho Pharmaceutical Co., Ltd. Atsumi T has received honoraria for lectures from GlaxoSmithKline plc., Eli Lilly Japan K.K., Nippon Boehringer Ingelheim Co., Ltd. Eisai Co. Ltd., AbbVie Inc., UCB Japan Co. Ltd., Novartis Pharma K.K., Pfizer Inc. Gilead Sciences, Inc., Daiichi Sankyo Co., Ltd., AstraZeneca plc., Asahi Kasei Pharma Co., Astellas Pharma Inc., Chugai Pharmaceutical Co., Ltd.; and research grants from Eli Lilly Japan K.K., AbbVie Inc., Bristol-Myers Squibb Co., Daiichi Sankyo Co., Ltd., Asahi Kasei Pharma Co., Eisai Co. Ltd., Chugai Pharmaceutical Co., Ltd., AstraZeneca plc., Novartis Pharma K.K., Pfizer Inc. and UCB Japan Co., Ltd.; and consultancy fees from GlaxoSmithKline plc., Eli Lilly Japan K.K., Nippon Boehringer Ingelheim Co., Ltd., Boehringer Ingelheim, Bristol Myers Squibb Co., Amgen K.K., Novartis Pharma K.K., Otsuka Pharmaceutical Co., Ltd., Ono Pharmaceutical Co., Ltd., Gilead Sciences, Inc. and Daiichi Sankyo Co., Ltd. The other authors declare no conflicts of interest.
Appendix
List of research collaboration facilities and investigators for the Sapporo-Oral SEMA study.
Department of Rheumatology, Endocrinology and Nephrology, Faculty of Medicine and Graduate School of Medicine, Hokkaido University: Sho Furusawa, Hiroshi Nomoto, Aika Miya, Hiraku Kameda, Akinobu Nakamura, Tatsuya Atsumi.
Sapporo Diabetes and Thyroid Clinic: Chiho Oba-Yamamoto, Jun Takeuchi.
Kurihara Clinic: Kumiko Yamashita, Miki Ito, Hiroyoshi Kurihara.
Aoki Clinic: Shin Aoki.
Supplementary Table 1
GLP-1 receptor agonists used at baseline
| GLP-1 receptor agonists (n = 30) |
|
| Semaglutide (0.25 mg/week) |
5 (16.7) |
| Semaglutide (0.5 mg/week) |
6 (20.0) |
| Semaglutide (1.0 mg/week) |
4 (13.3) |
| Dulaglutide (0.75 mg/week) |
13 (43.3) |
| Liraglutide (0.9 mg/day) |
1 (3.3) |
| Exenatide (10 mg/day) |
1 (3.3) |
Values are presented as numbers (%). GLP-1, glucagon-like peptide-1.
Supplementary Table 2
Comparison of patients’ characteristics between the treatment strategies
| Variables |
Add-on group (n = 92) |
Switch group (n = 342) |
p-value |
| Sex (male/female) |
45/47 |
200/142 |
0.100 |
| Age (years) |
51.7 ± 11.3 |
56.6 ± 12.7 |
0.001 |
| Body weight (kg) |
86.7 ± 21.0 |
78.4 ± 18.4 |
<0.001 |
| Body mass index (kg/m2) |
32.2 ± 7.1 |
28.7 ± 6.3 |
<0.001 |
| Duration of diabetes, n (%) |
|
|
<0.001 |
| <5 years |
54 (58.7) |
82 (24.0) |
|
| 5 to 10 years |
14 (15.2) |
66 (19.3) |
|
| 10 to 15 years |
15 (16.3) |
62 (18.1) |
|
| 15≤ years |
9 (9.8) |
132 (38.6) |
|
| HbA1c (%) |
7.49 ± 1.41 |
7.69 ± 1.01 |
0.113 |
| AST (U/L) |
31.9 ± 23.4 |
28.3 ± 15.2 |
0.081 |
| ALT (U/L) |
40.3 ± 32.1 |
36.4 ± 26.0 |
0.226 |
| eGFR (mL/min/1.73 m2) |
82.1 ± 21.0 |
78.0 ± 23.9 |
0.137 |
| Antihyperglycemic agents, n (%) |
| SGLT2 inhibitors |
58 (63.0) |
242 (70.8) |
0.155 |
| Metformin |
47 (48.9) |
250 (73.1) |
<0.001 |
| DPP-4 inhibitors |
0 (0.0) |
295 (86.3) |
<0.001 |
| Sulfonylureas |
8 (8.7) |
66 (19.3) |
0.016 |
| Glinides |
4 (4.4) |
37 (10.8) |
0.060 |
| Alfa-glucosidase inhibitors |
3 (3.3) |
29 (8.5) |
0.089 |
| GLP-1 receptor agonists |
0 (0.0) |
30 (8.8) |
0.003 |
| Thiazolidinediones |
0 (0.0) |
21 (6.1) |
0.015 |
| Insulin injection |
4 (4.4) |
15 (4.4) |
0.987 |
Data are mean ± standard deviation or number (%). HbA1c, glycated hemoglobin; AST, aspartate aminotransferase; ALT, alanine aminotransferase; eGFR, estimated glomerular filtration rate; SGLT2, sodium-glucose cotransporter 2; DPP-4, dipeptidyl peptidase-4; GLP-1, glucagon-like peptide-1.
Supplementary Table 3
Comparison of the baseline characteristics among the doses of oral semaglutide
| Variable |
3 mg (n = 55) |
7 mg (n = 241) |
14 mg (n = 138) |
p-value
among the groups |
| Sex (male/female) |
25/30 |
131/110 |
89/49 |
0.034 |
| Age (years) |
53.4 ± 12.0 |
56.8 ± 13.0 |
54.2 ± 11.8 |
0.059 |
| HbA1c (%) |
7.47 ± 1.17 |
7.64 ± 0.96 |
7.73 ± 1.30 |
0.183 |
| Body weight (kg) |
77.1 ± 17.6 |
77.2 ± 17.2 |
86.6 ± 21.5** ††† |
<0.001 |
| AST (U/L) |
23 (17, 33) |
24 (19, 33) |
24 (18, 32) |
0.720 |
| ALT (U/L) |
26 (19, 47) |
28 (20, 47) |
30 (20, 48) |
0.517 |
| γ-GTP (U/L) |
31 (19, 56) |
32 (21, 56) |
36 (23, 65) |
0.377 |
| eGFR (mL/min/1.73 m2) |
77.1 ± 22.7 |
77.5 ± 22.9 |
82.1 ± 24.3 |
0.151 |
| Creatinine (mg/dL) |
0.80 ± 0.33 |
0.78 ± 0.25 |
0.76 ± 0.25 |
0.691 |
| Dosing methods for semaglutide |
| Add-on (n, %) |
14 (25.0) |
44 (18.3) |
34 (24.6) |
0.261 |
| Switch from DPP-4 inhibitors (n, %) |
34 (61.8) |
176 (73.3) |
84 (60.9) |
0.024 |
| Switch form GLP-1RAs (n, %) |
3 (5.4) |
10 (4.2) |
17 (12.3) |
0.009 |
| Switch from other agents (n, %) |
4 (7.1) |
10 (4.2) |
3 (2.2) |
0.259 |
Values are presented as mean ± standard deviation, median (interquartile), or number (%). ** p < 0.01 vs. 3 mg group, ††† p < 0.001 vs. 7 mg group. P-values were calculated using ANOVA, Kruskal-Wallis test followed by post-hoc analysis, or chi-square test. HbA1c, glycated hemoglobin; AST, aspartate aminotransferase; ALT, alanine aminotransferase; γ-GTP, gamma-glutamyl transpeptidase; eGFR, estimated glomerular filtration rate; DPP-4, dipeptidyl peptidase-4; GLP-1RAs, glucagon-like peptide-1 receptor agonists.
Supplementary Table 4
Changes in body weight over 6 months according to administration method and dose of oral semaglutide
| Variable |
Baseline
(3 mg; n = 55) |
Mean change
at 6 months |
Baseline
(7 mg; n = 241) |
Mean change
at 6 months |
Baseline
(14 mg; n = 138) |
Mean change
at 6 months |
p-value for changes in body weight among the groups |
| Total cohort |
77.1 ± 17.6 |
–2.7***
(–3.7 to –1.8) |
77.2 ± 17.2 |
–2.8***
(–3.2 to –2.4) |
86.6 ± 21.5 |
–2.5***
(–3.1 to –1.9) |
0.494 |
| Add-on subgroup |
88.5 ± 21.6 |
–2.4
(–4.9 to 0.0) |
81.5 ± 17.5 |
–2.7***
(–4.0 to –1.4) |
92.9 ± 23.6 |
–2.4**
(–4.0 to –0.7) |
0.899 |
| Switch from DPP-4 inhibitors subgroup |
73.3 ± 15.0 |
–3.3***
(–4.4 to –2.1) |
76.3 ± 17.1 |
–3.0***
(–3.4 to –2.6) |
85.2 ± 20.7 |
–2.6***
(–3.3 to –2.0) |
0.603 |
| Switch from injectable GLP-1RAs subgroup |
64.9 ± 11.5 |
0.5
(–12.5 to 11.6) |
78.3 ± 21.0 |
–1.4
(–3.5 to 0.8) |
82.8 ± 20.8 |
–2.2*
(–4.1 to –0.2) |
0.411 |
Values are presented as mean ± standard deviation or mean change (95% confidence interval). * p < 0.05, ** p < 0.01, *** p < 0.001 vs. baseline. P-values were calculated using Kruskal-Wallis test. DPP-4, dipeptidyl peptidase-4; GLP-1RAs, glucagon-like peptide-1 receptor agonists; HbA1c, glycated hemoglobin.
Supplementary Table 5
Differences in baseline characteristics between the improvement and non-improvement groups based on changes in HbA1c during the 6-month study period
| Variables |
Improvement
group (n = 368) |
Non-improvement
group (n = 65) |
p-value between the groups |
| Systolic blood pressure (mmHg) |
132.8 ± 15.3 |
132.8 ± 15.8 |
0.965 |
| Diastolic blood pressure (mmHg) |
80.6 ± 11.9 |
78.9 ± 11.2 |
0.309 |
| Pulse rate (bpm) |
85.2 ± 12.9 |
89.2 ± 15.7 |
0.033 |
| BMI (kg/m2) |
29.3 ± 5.7 |
31.2 ± 6.8 |
0.020 |
| HbA1c (%) |
7.73 ± 1.10 |
7.21 ± 1.06 |
<0.001 |
| AST (U/L) |
24.0 (18.3, 33.0) |
22.0 (18.0, 33.0) |
0.607 |
| ALT (U/L) |
29.0 (20.0, 46.0) |
27.0 (19.0, 51.0) |
0.745 |
| γ-GTP (U/L) |
34.0 (22.0, 58.0) |
35.0 (22.0, 54.5) |
0.930 |
| eGFR (mL/min/1.73 m2) |
78.4 ± 23.0 |
82.0 ± 25.4 |
0.246 |
| Creatinine (mg/dL) |
0.73 (0.59, 0.88) |
0.75 (0.57, 0.92) |
0.563 |
| UACR (mg/g.Cre) |
19.9 (8.9, 56.0) |
17.3 (10.0, 35.5) |
0.569 |
| HDL-C (mg/dL) |
55.0 (47.0, 66.0) |
52 (44.0, 62.0) |
0.120 |
| Triglyceride (mg/dL) |
136.5 (97.3, 201.8) |
126.0 (83.0, 179.5) |
0.177 |
| LDL-C (mg/dL) |
92.0 (75.0, 112.0) |
88.0 (71.5, 107.0) |
0.273 |
| Treated with GLP-1RAs (n, %) |
19 (5.2) |
11 (16.9) |
0.002 |
Participants were divided into improvement and non-improvement groups according to whether their HbA1c had improved or not. Values are presented as mean ± standard deviation, median (interquartile range), or number (%). Differences between the groups were compared using unpaired t-test, Wilcoxon rank sum test, or chi-square test. BMI, body mass index; HbA1c, glycated hemoglobin; AST, aspartate aminotransferase; ALT, alanine aminotransferase; γ-GTP, γ-glutamyl transpeptidase; eGFR, estimated glomerular filtration rate; UACR, urinary albumin-creatinine ratio; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; GLP-1RAs, glucagon-like peptide-1 receptor agonists.
Supplementary Table 6
Differences in baseline characteristics between the weight loss group and the non-weight loss group based on changes in BMI during the 6-month study period
| Variables |
Weight loss
group (n = 318) |
Non-weight loss
group (n = 109) |
p-value between the groups |
| Systolic blood pressure (mmHg) |
132.0 ± 14.3 |
135.2 ± 18.1 |
0.062 |
| Diastolic blood pressure (mmHg) |
79.5 ± 11.8 |
82.7 ± 11.9 |
0.018 |
| Pulse rate (bpm) |
85.6 ± 12.6 |
86.5 ± 15.7 |
0.588 |
| BMI (kg/m2) |
29.7 ± 6.0 |
29.2 ± 5.8 |
0.435 |
| HbA1c (%) |
7.70 ± 1.10 |
7.50 ± 1.13 |
0.110 |
| AST (U/L) |
23.0 (19.0, 33.0) |
24.0 (18.0, 33.5) |
0.842 |
| ALT (U/L) |
27.0 (19.0, 45.0) |
30.0 (20.0, 50.5) |
0.354 |
| γ-GTP (U/L) |
31.5 (20.8, 54.0) |
38.0 (24.5, 67.0) |
0.048 |
| eGFR (mL/min/1.73 m2) |
76.9 ± 23.0 |
85.3 ± 23.3 |
0.001 |
| Creatinine (mg/dL) |
0.74 (059, 0.90) |
0.71 (0.58, 0.84) |
0.207 |
| UACR (mg/g.Cre) |
20.0 (9.0, 54.8) |
17.1 (9.0, 53.4) |
0.752 |
| HDL-C (mg/dL) |
55.0 (46.0, 66.0) |
52.0 (47.0, 62.0) |
0.177 |
| Triglyceride (mg/dL) |
134.0 (97.8, 191.0) |
143.0 (93.5, 226.0) |
0.296 |
| LDL-C (mg/dL) |
92.0 (74.0, 110.0) |
93.0 (75.0, 110.0) |
0.599 |
Participants were divided into weight loss and non-weight loss groups according to whether their BMI had improved or not. Values are presented as mean ± standard deviation or median (interquartile range). Differences between the groups were compared using unpaired t-test or Wilcoxon rank sum test. BMI, body mass index; HbA1c, glycated hemoglobin; AST, aspartate aminotransferase; ALT, alanine aminotransferase; γ-GTP, γ-glutamyl transpeptidase; eGFR, estimated glomerular filtration rate; UACR, urinary albumin-creatinine ratio; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol.
Supplementary Table 7
Adverse events during the study period of the oral semaglutide introduction method
| Adverse events |
Oral semaglutide introduction method |
Total (n = 543) |
| Add-on (n = 118) |
Switch from DPP-4 inhibitors (n = 360) |
Switch from injectable GLP-1RAs (n = 45) |
Switch from other antihyperglycemic agents (n = 20) |
| Nausea |
27 [5] |
96 [19] |
4 [2] |
2 [0] |
129 [26] |
| Diarrhea |
2 [0] |
13 [5] |
0 [0] |
1 [0] |
16 [5] |
| Gastric pain |
1 [1] |
3 [4] |
0 [0] |
1 [0] |
5 [5] |
| Others |
1 [1] |
2 [2] |
2 [0] |
0 [0] |
4 [3] |
| Appetite loss |
0 [0] |
2 [2] |
0 [0] |
0 [0] |
2 [2] |
| General fatigue |
0 [0] |
1 [1] |
1 [1] |
0 [0] |
2 [2] |
| Hypoglycemia |
0 [0] |
1 [1] |
0 [0] |
0 [0] |
1 [1] |
| Leg cramps |
0 [0] |
1 [1] |
0 [0] |
0 [0] |
1 [1] |
| Stomatitis |
0 [0] |
0 [0] |
1 [1] |
0 [0] |
1 [1] |
Values are presented as the number of cases. Values in brackets are the number of cases leading to semaglutide discontinuation. DPP-4, dipeptidyl peptidase-4; GLP-1RAs, glucagon-like peptide-1 receptor agonists.
Supplementary Table 8
Adverse events during the study period by oral semaglutide dose
| Adverse events |
Semaglutide dose |
Total (n = 543) |
| 3 mg/day (n = 140) |
7 mg/day (n = 257) |
14 mg/day (n = 146) |
| Nausea |
50 [23] |
57 [3] |
22 [0] |
129 [26] |
| Diarrhea |
8 [3] |
7 [2] |
1 [0] |
16 [5] |
| Gastric pain |
1 [4] |
2 [1] |
2 [0] |
5 [5] |
| Others |
3 [2] |
1 [1] |
0 [0] |
4 [3] |
| Appetite loss |
1 [1] |
1 [1] |
0 [0] |
2 [2] |
| General fatigue |
1 [1] |
1 [1] |
0 [0] |
2 [2] |
| Hypoglycemia |
1 [1] |
0 [0] |
0 [0] |
1 [1] |
| Leg cramps |
1 [1] |
0 [0] |
0 [0] |
1 [1] |
| Stomatitis |
1 [1] |
0 [0] |
0 [0] |
1 [1] |
Values are presented as the number of cases. Within brackets are the number of cases leading to semaglutide discontinuation.
Supplementary Table 9
Characteristics of patients exhibiting symptomatic gastrointestinal disorders
| Variables |
Patients showing symptomatic gastrointestinal disorders (n = 154) |
Discontinuation of oral semaglutide due to gastrointestinal disorders (n = 39) |
| Sex (male/female) |
76/78 |
18/21 |
| Age (years) |
55.8 ± 13.5 |
60.7 ± 15.8 |
| Body weight (kg) |
76.9 ± 17.1 |
69.9 ± 13.3 |
| Body mass index (kg/m2) |
28.9 ± 5.9 |
26.8 ± 5.6 |
| Duration of diabetes, n (%) |
| <5 years |
44 (28.6) |
8 (20.5) |
| 5 to 10 years |
25 (16.2) |
4 (10.3) |
| 10 to 15 years |
34 (22.1) |
9 (23.1) |
| 15≤ years |
51 (33.1) |
18 (46.2) |
| HbA1c (%) |
7.69 ± 1.05 |
7.90 ± 1.27 |
| AST (U/L) |
26.9 ± 13.2 |
24.1 ± 9.9 |
| ALT (U/L) |
35.5 ± 26.0 |
29.8 ± 26.3 |
| eGFR (mL/min/1.73 m2) |
80.5 ± 23.6 |
74.3 ± 24.2 |
| Antihyperglycemic agents, n (%) |
| SGLT2 inhibitors |
104 (67.5) |
25 (64.1) |
| Metformin |
106 (68.8) |
32 (82.1) |
| DPP-4 inhibitors |
115 (75.7) |
30 (78.9) |
| Sulfonylureas |
40 (26.0) |
17 (43.6) |
| Glinides |
9 (5.8) |
2 (5.1) |
| Alfa-glucosidase inhibitors |
10 (6.5) |
2 (5.1) |
| GLP-1 receptor agonists |
4 (2.6) |
2 (5.3) |
| Thiazolidinediones |
4 (2.6) |
1 (2.6) |
| Insulin injection |
6 (3.9) |
2 (5.1) |
Data are mean ± standard deviation or number (%). HbA1c, glycated hemoglobin; AST, aspartate aminotransferase; ALT, alanine aminotransferase; eGFR, estimated glomerular filtration rate; SGLT2, sodium-glucose cotransporter 2; DPP-4, dipeptidyl peptidase-4; GLP-1, glucagon-like peptide-1.
References
- 1 Davis TM, Chubb SA, Bruce DG, Davis WA (2016) Metabolic memory and all-cause death in community-based patients with type 2 diabetes: the fremantle diabetes study. Diabetes Obes Metab 18: 598–606.
- 2 Rubin RR, Peyrot M (1999) Quality of life and diabetes. Diabetes Metab Res Rev 15: 205–218.
- 3 Ueki K, Sasako T, Okazaki Y, Kato M, Okahata S, et al. (2017) Effect of an intensified multifactorial intervention on cardiovascular outcomes and mortality in type 2 diabetes (J-DOIT3): an open-label, randomised controlled trial. Lancet Diabetes Endocrinol 5: 951–964.
- 4 Gaede P, Lund-Andersen H, Parving HH, Pedersen O (2008) Effect of a multifactorial intervention on mortality in type 2 diabetes. N Engl J Med 358: 580–591.
- 5 Davies MJ, Aroda VR, Collins BS, Gabbay RA, Green J, et al. (2022) Management of hyperglycemia in type 2 diabetes, 2022. A consensus report by the American Diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD). Diabetes Care 45: 2753–2786.
- 6 Bouchi R, Kondo T, Ohta Y, Goto A, Tanaka D, et al. (2022) A consensus statement from the Japan Diabetes Society (JDS): a proposed algorithm for pharmacotherapy in people with type 2 diabetes. Diabetol Int 14: 1–14.
- 7 Seino Y, Yabe D (2013) Glucose-dependent insulinotropic polypeptide and glucagon-like peptide-1: incretin actions beyond the pancreas. J Diabetes Investig 4: 108–130.
- 8 Tsapas A, Avgerinos I, Karagiannis T, Malandris K, Manolopoulos A, et al. (2020) Comparative effectiveness of glucose-lowering drugs for type 2 diabetes: a systematic review and network meta-analysis. Ann Intern Med 173: 278–286.
- 9 Madsbad S (2016) Review of head-to-head comparisons of glucagon-like peptide-1 receptor agonists. Diabetes Obes Metab 18: 317–332.
- 10 Nauck MA, Kemmeries G, Holst JJ, Meier JJ (2011) Rapid tachyphylaxis of the glucagon-like peptide 1-induced deceleration of gastric emptying in humans. Diabetes 60: 1561–1565.
- 11 Marso SP, Daniels GH, Brown-Frandsen K, Kristensen P, Mann JF, et al. (2016) Liraglutide and cardiovascular outcomes in type 2 diabetes. N Engl J Med 375: 311–322.
- 12 Marso SP, Bain SC, Consoli A, Eliaschewitz FG, Jódar E, et al. (2016) Semaglutide and cardiovascular outcomes in patients with type 2 diabetes. N Engl J Med 375: 1834–1844.
- 13 Gerstein HC, Colhoun HM, Dagenais GR, Diaz R, Lakshmanan M, et al. (2019) Dulaglutide and cardiovascular outcomes in type 2 diabetes (REWIND): a double-blind, randomised placebo-controlled trial. Lancet 394: 121–130.
- 14 Pratley R, Amod A, Hoff ST, Kadowaki T, Lingvay I, et al. (2019) Oral semaglutide versus subcutaneous liraglutide and placebo in type 2 diabetes (PIONEER 4): a randomised, double-blind, phase 3a trial. Lancet 394: 39–50.
- 15 Yabe D, Nakamura J, Kaneto H, Deenadayalan S, Navarria A, et al. (2020) Safety and efficacy of oral semaglutide versus dulaglutide in Japanese patients with type 2 diabetes (PIONEER 10): an open-label, randomised, active-controlled, phase 3a trial. Lancet Diabetes Endocrinol 8: 392–406.
- 16 Rosenstock J, Allison D, Birkenfeld AL, Blicher TM, Deenadayalan S, et al. (2019) Effect of additional oral semaglutide vs. sitagliptin on glycated hemoglobin in adults with type 2 diabetes uncontrolled with metformin alone or with sulfonylurea: the PIONEER 3 randomized clinical trial. JAMA 321: 1466–1480.
- 17 Li A, Su X, Hu S, Wang Y (2023) Efficacy and safety of oral semaglutide in type 2 diabetes mellitus: a systematic review and meta-analysis. Diabetes Res Clin Pract 198: 110605.
- 18 Overgaard RV, Hertz CL, Ingwersen SH, Navarria A, Drucker DJ (2021) Levels of circulating semaglutide determine reductions in HbA1c and body weight in people with type 2 diabetes. Cell Rep Med 2: 100387.
- 19 Ronn J, Jensen EP, Wewer Albrechtsen NJ, Holst JJ, Sorensen CM (2017) Glucagon-like peptide-1 acutely affects renal blood flow and urinary flow rate in spontaneously hypertensive rats despite significantly reduced renal expression of GLP-1 receptors. Physiol Rep 5: e13503.
- 20 Fujita H, Morii T, Fujishima H, Sato T, Shimizu T, et al. (2014) The protective roles of GLP-1R signaling in diabetic nephropathy: possible mechanism and therapeutic potential. Kidney Int 85: 579–589.
- 21 Skov J, Pedersen M, Holst JJ, Madsen B, Goetze JP, et al. (2016) Short-term effects of liraglutide on kidney function and vasoactive hormones in type 2 diabetes: a randomized clinical trial. Diabetes Obes Metab 18: 581–589.
- 22 Thomas MC (2017) The potential and pitfalls of GLP-1 receptor agonists for renal protection in type 2 diabetes. Diabetes Metab 43 Suppl 1: 2S20–2S27.
- 23 Bunck MC, Diamant M, Eliasson B, Cornér A, Shaginian RM, et al. (2010) Exenatide affects circulating cardiovascular risk biomarkers independently of changes in body composition. Diabetes Care 33: 1734–1737.
- 24 Astrup A, Carraro R, Finer N, Harper A, Kunesova M, et al. (2012) Safety, tolerability and sustained weight loss over 2 years with the once-daily human GLP-1 analog, liraglutide. Int J Obes (Lond) 36: 843–854.
- 25 Tonneijck L, Muskiet MH, Smits MM, van Bommel EJ, Heerspink HJ, et al. (2017) Glomerular hyperfiltration in diabetes: mechanisms, clinical significance, and treatment. J Am Soc Nephrol 28: 1023–1039.
