Abstract
Background
Dipeptidyl peptidase-4 inhibitors (DPP-4is) are the most widely used oral hypoglycemic drugs in Japan. However, once-daily oral semaglutide has been reported to reduce glycated hemoglobin (HbA1c) and body weight (BW) without causing significant hypoglycemia. Here, we aimed to evaluate the efficacy and safety of switching from a DPP-4i to oral semaglutide in Japanese patients with type 2 diabetes (T2D).
Methods
We performed a single-center retrospective study of the changes in HbA1c and BW in 68 patients with T2D who were switched from a DPP-4i and took oral semaglutide for ≥ 6 months, without changes in any other oral hypoglycemic agent.
Results
Mean HbA1c decreased from 7.8 to 7.0% (p < 0.001) and BW decreased from 74.2 to 71.2 kg (p < 0.001) over 6 months. The decrease in HbA1c was more pronounced in participants with high baseline HbA1c (r = − 0.542, p < 0.001). There was also a trend (r = 0.236, p = 0.052) toward a decrease in BW in individuals with shorter disease duration. There were reductions in either HbA1c or BW in 65 participants (95.6%). In addition, the larger the decrease in HbA1c was, the greater was the decrease in BW (r = 0.480, p < 0.001). Eighteen participants (20.1%) discontinued the drug within 6 months, of whom 10 (11.6% of the total) did so because of suspected adverse effects and the discontinuation rate was the highest in older, non-obese patients.
Conclusions
Switching from a DPP-4i to oral semaglutide may be useful for Japanese patients with T2D who have inadequate glycemic or BW control. However, its utility may be limited by gastrointestinal adverse effects in certain patients.
Keywords: Type 2 diabetes, Glucagon-like peptide-1 receptor agonist, Dipeptidyl peptidase-4 inhibitor, Oral semaglutide
Introduction
Although the drug treatment options for type 2 diabetes (T2D) are expanding, dipeptidyl peptidase-4 inhibitors (DPP-4i) remain the most widely used antidiabetic drugs in Japan [1, 2], and their use has been reported to be associated with a high level of compliance with long-term use [1]. This may be explained by their particularly potent hypoglycemic effects in Asian patients [3, 4], their oral administration, and their few side effects. In particular, DPP-4is are the most commonly prescribed first-line treatment for T2D in Japan, with a particularly large number of prescriptions being written for older patients, suggesting that they are safe in this group [2]. However, the effects of antidiabetic drugs on body weight (BW) are important from the perspective of reducing obesity and preventing geriatric syndromes, for instance, cognitive decline, activities of daily living (ADL) decline, falls, sarcopenia, depression, urinary and low nutrition. Oral semaglutide administration is associated with significant weight loss (2.0–3.0 kg) at high doses [5, 6], which is in sharp contrast to DPP-4is, which have almost no effect on BW. The current Japanese algorithm for the pharmacotherapy of T2D [2] prioritizes the use of DPP-4is for patients who do not have obesity because of their safety and efficacy. Meanwhile, biguanides and sodium-glucose cotransporter-2 inhibitors (SGLT2is) are prioritized for patients with obesity, followed by GLP-1 receptor agonists (GLP-1RAs) and DPP-4is. In Japanese patients with T2D who are taking a DPP-4i, switching to oral semaglutide may be considered when their HbA1c or BW control is inadequate. However, there have been no reports of the efficacy or safety of switching from a DPP-4i to oral semaglutide in Japanese patients. Although the extension of the PIONEER 7 trial [7] was conducted in patients who switched from sitagliptin 100 mg to oral semaglutide, Japanese patients were not included. Therefore, in the present study, we investigated the efficacy and safety of switching from a DPP-4i to oral semaglutide using clinical data collected at our T2D clinic in Japan.
Materials and methods
Study design
We conducted a single-center retrospective study of Japanese patients with T2D who attended Misaki Internal Medicine Clinic between February 2021 and September 2022.
The inclusion criteria were as follows: Japanese patients with T2D who were switched from a DPP-4i to oral semaglutide in an outpatient setting and they took for at least 6 months. The exclusion criteria included patients with cancer, those who did not visit the clinic regularly as instructed, those who had poor adherence to the medication regimen, those who had a lot of leftover medication, those who transferred to a new hospital or were hospitalized during the observation period, and those who exhibited a change in other hypoglycemic agents during the period. In all patients, once daily administration of 3 mg oral semaglutide was initiated. The dose titration criteria were fundamentally as described in the package insert, and the dose was increased to 7 mg if there were no significant side effects after 4 or more weeks. When there were no side effects and the drug was judged to be insufficiently effective, the dose was increased to 14 mg after another 4 or more weeks. However, if the patient requested a dose reduction due to severe side effects or increased medical costs associated with increasing the dose of oral semaglutide, the dose was reduced to 7 mg and the medication was continued at the physician’s discretion after hearing the patient’s symptoms and financial status.
We collected the sex, age, height, BW, body mass index (BMI), duration of diabetes, family history of diabetes, diabetic neuropathy, diabetic retinopathy, diabetic nephropathy, glycated hemoglobin (HbA1c), C-peptide immunoreactivity/plasma glucose concentration 2 h after a meal (CPR/PG), aspartate aminotransferase (AST), alanine aminotransferase (ALT), γ-glutamyltranspeptidase (γ-GTP), low-density lipoprotein-cholesterol (LDL-C), high-density lipoprotein-cholesterol (HDL-C), triglyceride (TG), estimated glomerular filtration rate (eGFR), treatment prior to switching to oral semaglutide, and diabetic complications from the medical records of the participants. The data collected within the 30 days preceding the start of oral semaglutide administration were regarded as the baseline data, and data were also collected 6 months after the switch (180 ± 15 days following the baseline date). The presence or absence of gastrointestinal symptoms, hypoglycemia (defined as severe hypoglycemia, symptomatic hypoglycemia, or glucose concentration < 70 mg/dL, based on the self-monitoring of blood glucose concentration), and worsening retinopathy during the 6 months following the switch to oral semaglutide were also extracted from the participants’ medical records. The efficacy of the semaglutide was assessed by comparing the changes in clinical factors between baseline and 6 months after the switch, and the clinical factors that affected the changes in HbA1c and BW were also identified. The safety assessments performed included recording the presence or absence of gastrointestinal symptoms, the incidence of hypoglycemia (the proportions of the participants taking each dose who experienced hypoglycemia), and whether or not their retinopathy worsened.
Statistical analysis
We express continuous datasets as means ± standard deviations and categorical datasets as numbers and percentages. Mean values for clinical parameters at baseline and 6 months after switching were compared using the paired Student’s t-test, and the relationships of the change in HbA1c with other continuous parameters were analyzed using Pearson’s correlation. In addition, multiple regression analysis, using the change in HbA1c as the dependent variable, was performed using a stepwise method. SPSS ver. 28 (IBM Inc., Armonk, NY, USA) was used for statistical analysis. p < 0.05 was considered to represent statistical significance and p < 0.1 was considered to represent a trend.
Results
The baseline characteristics of the 68 participants and the changes that occurred in their metabolic parameters after switching from a DPP-4i to oral semaglutide are shown in Table 1. In all the participants, oral semaglutide was initiated at a dose of 3 mg per day, and after 6 months, the doses being administered were 7 mg in 52 participants and 14 mg in the other 16. The ages, durations of diabetes, BMIs, and HbA1cs of these two groups were 61.9 ± 14.0 years, 13.1 ± 8.5 years, 28.3 kg/m2, and 7.8% ± 0.8%, respectively. Forty-five of the participants (66.2%) were also administering an SGLT2i and 41 (60.3%) were also administering metformin. The DPP-4is that were being administered before the switch were as follows: vildagliptin (n = 30), sitagliptin (n = 15), teneligliptin (n = 8), alogliptin (n = 7), anagliptin (n = 3), linagliptin (n = 2), omarigliptin (n = 1), trelagliptin (n = 1), and saxagliptin (n = 1). There were reductions in HbA1c (from 7.8 ± 0.8 to 7.0 ± 0.8%, p < 0.001) and BMI (from 28.3 ± 4.8 to 27.1 ± 4.7 kg/m2, p < 0.001). In the participants who were taking 7 mg semaglutide, their HbA1c decreased from 7.8 ± 0.8 to 7.0 ± 0.8% (p < 0.001) and their BMI decreased from 27.9 ± 5.0 to 26.7 ± 4.9 kg/m2 (p < 0.001). In the 14 mg group, their HbA1c decreased from 7.7 ± 0.6 to 7.1 ± 0.6% (p = 0.003) and their BMI decreased from 29.4 ± 3.8 to 28.3 ± 3.6 kg/m2 (p = 0.003). There were reductions in either HbA1c or BW in 65 of the participants (95.6%), but both the HbA1c and BW of 1 participant (1.5%) increased. There were also significant reductions in their serum LDL-cholesterol concentration from 107.8 ± 25.1 to 94.2 ± 21.2 mg/dL (p < 0.001).
Table 1.
Clinical characteristics of the participants at baseline and 6 months after switching to oral semaglutide
| Parameter | Full cohort | (n = 68) | 7 mg | (n = 52) | 14 mg | (n = 16) | |||
|---|---|---|---|---|---|---|---|---|---|
| Baseline | 6 months | p value | Baseline | 6 months | p value | Baseline | 6 months | p value | |
| N (M/F) | 68 (35/33) | 52 (25/27) | 16 (10/6) | ||||||
| Age, years | 61.9 ± 14.0 | 63.5 ± 14.3 | 56.5 ± 11.8 | ||||||
| Duration of diabetes, years | 13.1 ± 8.5 | 13.5 ± 8.7 | 11.6 ± 7.7 | ||||||
| Family history of diabetes, n (%) | 45 (66.1%) | 35 (67.3%) | 10 (62.5%) | ||||||
| CPR/PG × 100, ng/mL/mg | 2.83 ± 1.40 | 2.86 ± 1.53 | 2.74 ± 0.82 | ||||||
| Neuropathy, n (%) | 29/68 (42.6) | 23/52 (44.2) | 6/16 (37.5) | ||||||
| Retinopathy, n (%) | 20/63 (31.7) | 15 /48 (31.2) | 5/15 (33.3) | ||||||
| Nephropathy, n (%) | 25/68 (36.8) | 16/52 (30.8) | 9/16 (56.2) | ||||||
| eGFR, mL/min/1.73m2 | 80.2 ± 21.7 | 80.5 ± 21.5 | 79.2 ± 22.1 | ||||||
| Height, cm | 161.2 ± 1.7 | 159.6 ± 10.0 | 166.6 ± 11.3 | ||||||
| Body weight, kg | 74.2 ± 18.5 | 71.2 ± 18.6 | < 0.001 | 71.7 ± 18.2 | 68.7 ± 18.4 | < 0.001 | 82.3 ± 17.0 | 79.3 ± 16.8 | 0.002 |
| BMI, kg/m2 | 28.3 ± 4.8 | 27.1 ± 4.7 | < 0.001 | 27.9 ± 5.0 | 26.7 ± 4.9 | < 0.001 | 29.4 ± 3.8 | 28.3 ± 3.6 | 0.003 |
| Waist circumference, cm | 95.6 ± 12.0 | 94.9 ± 12.7 | 98.1 ± 9.3 | ||||||
| HbA1c, % | 7.8 ± 0.8 | 7.0 ± 0.8 | < 0.001 | 7.8 ± 0.8 | 7.0 ± 0.8 | < 0.001 | 7.7 ± 0.6 | 7.1 ± 0.6 | 0.003 |
| AST, U/L | 26.4 ± 10.6 | 24.8 ± 8.6 | 0.116 | 26.3 ± 8.5 | 24.5 ± 7.3 | 0.261 | 26.6 ± 8.5 | 24.5 ± 7.3 | 0.116 |
| ALT, U/L | 29.7 ± 21.3 | 27.7 ± 16.1 | 0.178 | 29.8 ± 23.5 | 28.0 ± 17.5 | 0.308 | 29.3 ± 11.2 | 26.9 ± 10.3 | 0.270 |
| γ-GTP, U/L | 35.1 ± 21.6 | 34.4 ± 22.9 | 0.647 | 36.2 ± 22.1 | 35.5 ± 23.4 | 0.748 | 31.6 ± 20.0 | 30.5 ± 20.7 | 0.645 |
| LDL-cholesterol, mg/dL | 107.8 ± 25.1 | 94.2 ± 21.2 | < 0.001 | 105.5 ± 25.6 | 92.3 ± 20.1 | < 0.001 | 115.1 ± 21.9 | 100.3 ± 23.4 | 0.030 |
| HDL-cholesterol, mg/dL | 56.4 ± 16.0 | 54.8 ± 15.4 | 0.154 | 56.9 ± 15.5 | 55.6 ± 15.9 | 0.331 | 54.8 ± 17.3 | 52.2 ± 13.4 | 0.188 |
| Triglycerides, mg/dL | 170.9 ± 113.3 | 149.5 ± 83.4 | 0.081 | 172.5 ± 122 | 143.2 ± 77.5 | 0.048 | 165.9 ± 78.1 | 169.8 ± 97.2 | 0.847 |
| Insulin, n (%) | 0 (0%) | 0 (0%) | 0 (0%) | ||||||
| GLP-1RA, n (%) | 0 (0%) | 0 (0%) | 0 (0%) | ||||||
| DPP-4i, n (%) | 68 (100%) | 52 (100%) | 16 (100%) | ||||||
| SGLT2 inhibitor, n (%) | 45 (66.2%) | 32 (61.5%) | 13 (81.3%) | ||||||
| Biguanide, n (%) | 41 (60.3%) | 32 (61.5%) | 9 (56.3%) | ||||||
| Sulfonylurea, n (%) | 11 (16.2%) | 8 (15.4%) | 3 (18.8%) | ||||||
| Glinide, n (%) | 8 (11.8%) | 7 (13.5%) | 1 (6.3%) | ||||||
| Thiazolidine, n (%) | 13 (19.1%) | 11 (21.2%) | 2 (12.5%) | ||||||
| α-GI, n (%) | 8 (11.8%) | 7 (13.5%) | 1 (6.3%) |
p-values were calculated using the paired Student’s t-test. Values are presented as mean ± standard deviation, unless otherwise indicated. CPR C-peptide immunoreactivity, PG plasma glucose, eGFR estimated glomerular filtration rate, BMI body mass index, HbA1c glycated hemoglobin, AST aspartate aminotransferase, ALT alanine aminotransferase, γGTP γ-Glutamyltranspeptidase, LDL low-density lipoprotein, HDL high-density lipoprotein, GLP-1RA glucagon-like peptide-1 receptor agonist, DPP-4i dipeptidyl peptidase-4 inhibitor, SGLT2i sodium-glucose cotransporter-2 inhibitor, α-GI α-glucosidase inhibitor
Table 2 shows the results of the correlation analysis of the relationships of the changes in HbA1c and BW with clinical parameters. For the entire cohort of 68 participants, there was a significant negative correlation (r = − 0.542, p < 0.01) between the change in HbA1c and the baseline HbA1c, implying that the higher the baseline HbA1c was, the greater was the decrease in HbA1c. There was also a trend towards a positive correlation between the duration of diabetes and BW (r = 0.236, p = 0.052), implying that the shorter the duration of disease was, the greater was the decrease in BW.
Table 2.
Relationships of the changes in HbA1c and BW with clinical parameters, evaluated using Pearson’s correlation
| Change in HbA1c | Change in BW | |||
|---|---|---|---|---|
| r | p | r | p | |
| Sex | − 0.130 | 0.291 | − 0.156 | 0.203 |
| Age (years) | − 0.115 | 0.348 | 0.045 | 0.718 |
| Baseline BMI (kg/m2) | 0.164 | 0.183 | − 0.159 | 0.194 |
| Diabetes duration (years) | − 0.159 | 0.195 | 0.236 | 0.052 |
| Baseline HbA1c (%) | − 0.542 | < 0.001 | − 0.022 | 0.858 |
| CPR/ PG × 100 (ng/mL/mg) | 0.059 | 0.634 | − 0.11 | 0.376 |
BMI body mass index, CPR C-peptide immunoreactivity, PG plasma glucose, BW body weight, HbA1c glycated hemoglobin
There were significant positive correlations between the changes in HbA1c and BW (r = 0.480, p < 0.001) in the 68 participants (Fig. 1). HbA1c decreased in 60 participants (88.2%) and BW also decreased in 60 (88.2%). Furthermore, both HbA1c and BW decreased in 55 participants (80.9%), and one of the two decreased in 65 (95.6%). We next performed a stepwise multiple regression analysis using the change in HbA1c as the dependent variable, and sex, age, duration of disease, baseline HbA1c, baseline BMI, and CPR/PG × 100 as independent variables (Table 3). We found that baseline HbA1c was the only significant explanatory variable (β = − 0.485, p < 0.001; adjusted R2 = 0.223, ANOVA p < 0.01). We then performed the same analysis with the change in BW as the dependent variable, and found no significant explanatory variable.
Fig. 1.
Relationship between the change in HbA1c and the change in BW of the participants after they were switched from a DPP-4i to oral semaglutide. There was a significantly positive correlation between the changes in HbA1c and BW (r=0.480, p<0.001) in the 68 participants. HbA1c decreased in 60 participants (88.2%) and BW also decreased in 60 (88.2%). Both HbA1c and BW decreased in 55 participants (80.9%), and one of HbA1c or BW decreased in 65 (95.6%). DPP-4i, dipeptidyl peptidase-4 inhibitor; HbA1c, glycated hemoglobin; BW, body weight.
Table 3.
Multiple regression analysis of changes in HbA1c
| Variables | β | t value | p value |
|---|---|---|---|
| Sex* | − 0.173 | − 1.609 | 0.112 |
| Age | − 0.127 | − 1.167 | 0.248 |
| Duration of disease | − 0.024 | − 0.205 | 0.838 |
| Baseline HbA1c | − 0.485 | − 4.469 | < 0.001 |
| Baseline BMI | 0.078 | 0.688 | 0.494 |
| CPR/PG × 100 | − 0.120 | − 1.044 | 0.300 |
HbA1c glycated hemoglobin, BMI body mass index, CPR C-peptide immunoreactivity, PG plasma glucose Adjusted R2 = 0.223, ANOVA p < 0.01
*Dummy variables of 1 for males and 2 for females were used
There were no instances of symptomatic hypoglycemia, severe hypoglycemia, or the worsening of retinopathy within the 6 months following the switch to oral semaglutide. The gastrointestinal symptoms experienced by the participants are shown in Table 4. In all, 33 participants (48.5%) experienced heartburn or nausea, 8 (11.8%) experienced constipation, 5 (7.4%) experienced diarrhea, 5 (7.4%) experienced abdominal distention, 36 (52.9%) experienced a loss of appetite, and 5 (7.4%) experienced no gastrointestinal symptoms. Furthermore, we have examined the relationship between the presence or absence of diabetic complications and gastrointestinal symptoms of oral semaglutide. Of the 27 participants without diabetic complications (neuropathy, retinopathy, and nephropathy), 14 (51.8%) had gastrointestinal symptoms. Of the 29 participants with neuropathy, 14 (48.2%) had gastrointestinal symptoms. Of the 20 participants with retinopathy, 8 (40.0%) had gastrointestinal symptoms, and of the 25 participants with nephropathy, 8 (32.0%) had gastrointestinal symptoms. In addition, of the 86 patients who were prescribed oral semaglutide during the period covered in this study, 18 (20.1%) discontinued the medication, and 10 of these (11.6% of the total) did so because of suspected adverse effects. We divided the 86 patients who were prescribed oral semaglutide during the period into four groups according to their age and BMI, and compared the incidence of discontinuation owing to adverse effects among the groups (Fig. 2). The frequency of discontinuation was highest in those participants who were ≥ 65 years of age and had BMIs < 25 kg/m2. However, none of the participants aged < 65 years and with BMIs < 25 kg/m2 discontinued the medication. On the other hand, when comparing the group who were aged ≥ 65 years and BMIs ≥ 25 kg/m2 with the group who were aged < 65 years and BMIs ≥ 25 kg/m2, discontinuation rate was almost the same (8.0% vs 7.7%). The mean duration of diabetes in the 18 participants who discontinued the medication was 18.2 ± 10.3 years, while the mean duration in the 68 participants who continued their medication was 13.1 ± 8.5 years.
Table 4.
Gastrointestinal symptoms reported by the participants taking semaglutide
| Total (n = 68) | 7 mg (n = 52) | 14 mg (n = 16) | |
|---|---|---|---|
| Heartburn/nausea, n (%) | 33 (48.5%) | 25 (48.0%) | 8 (50.0%) |
| Constipation, n (%) | 8 (11.8%) | 6 (11.5%) | 2 (12.5%) |
| Diarrhea, n (%) | 5 (7.4%) | 4 (7.7%) | 1 (6.3%) |
| Abdominal distention, n (%) | 5 (7.4%) | 3 (5.8%) | 2 (12.5%) |
| Appetite loss, n (%) | 36 (52.9%) | 30 (57.7%) | 6 (37.5%) |
| None, n (%) | 5 (7.4%) | 4 (7.7%) | 1 (6.3%) |
Fig. 2.
Discontinuation of oral semaglutide owing to suspected adverse effects. We divided the 86 patients who were prescribed oral semaglutide during the study into four groups according to their age and body mass index (BMI), and compared the incidence of discontinuation owing to adverse effects among these groups. Eighteen participants (20.1%) discontinued the medication, of whom 10 (11.6% of the total) discontinued the medication because of suspected adverse effects. The frequency of discontinuation was the highest in patients aged ≥ 65 years and BMI < 25 kg/m2.
Discussion
This was the first study to evaluate the efficacy and safety of the use of oral semaglutide in Japanese patients with T2D who had their prescription switched from a DPP-4i, the most widely used type of hypoglycemic agent in Japan. Switching from a DPP-4i to oral semaglutide significantly reduced either HbA1c or BW in the participants, and it was particularly effective in those with high baseline HbA1c.
Fujiwara et al. [8] compared Japanese patients with T2D who initiated oral semaglutide treatment with those who were being injected with a GLP-1RA or a DPP-4i using data extracted from the Medical Data Vision database, and found that those taking oral semaglutide were younger, had a higher BMI, and reported using a larger number of antidiabetic medications. Although oral semaglutide should be commenced relatively early in the course of the disease, with the intention of preserving pancreatic endocrine function and extrapancreatic action, in real-world settings, a large proportion of the participants in this study (44.1%) were prescribed semaglutide after having taken three or more drugs. The maximum doses used by each participant during the treatment period were recorded, with 7 mg being the most common (48.7%), followed by 3 mg (39.6%) and 14 mg (11.7%). However, in the present study, all the participants administered either 7 mg or 14 mg per day. We believe that this may be because the physicians and patients recognized the benefits of switching from a DPP-4i to oral semaglutide for their HbA1c and BW.
Because there have been no previous studies on the efficacy of switching from a DPP-4i to oral semaglutide in a real-world clinical setting in Japan, we discuss the results of the present study in conjunction with the results of previous clinical trials. In the extension of the PIONEER 7 trial [7], more than twice as many patients who were switched to oral semaglutide (the dose could be freely adjusted) achieved an HbA1c < 7% and lost approximately 2 kg more than those who continued on sitagliptin 100 mg, suggesting that the switch was effective, but this study did not include Japanese patients. However, a subgroup analysis of PIONEER 3 data [9], in which the use of oral semaglutide and sitagliptin 100 mg were compared in Japanese patients, showed that after 26 weeks, 7 mg of oral semaglutide was as effective as 100 mg of sitagliptin at reducing HbA1c (− 1.0% vs. − 0.9%), but was more effective at inducing weight loss (− 2.7 kg vs. − 0.3 kg). In addition, the 14 mg dose of semaglutide was superior to sitagliptin in reducing HbA1c (− 1.4% vs. − 0.9%) and BW (− 3.7 kg vs. − 0.3 kg). The prevalence of gastrointestinal symptoms was as high as 48.1% in the 7 mg semaglutide group and 64.7% in the 14 mg group, but 38.5% in the 100 mg sitagliptin group. In real-world clinical practice in Japan, treatment is often initiated with a DPP-4i in patients with T2D who have not used incretin-based drugs previously, owing to concerns regarding safety and side effects. Nevertheless, because DPP-4is can only increase the GLP-1 concentration to within the physiological range, it may be that some patients need to be treated with a GLP-1RA to increase this to a higher concentration. Therefore, we suggest that switching to oral semaglutide may be effective for patients with T2D who are already taking a DPP-4i but have inadequate HbA1c control, as in the present study, or for those who want to reduce their BW and have difficulty with the administration of injectable semaglutide.
The injectable formulation of semaglutide was the first form to be launched, followed by the oral formulation. It has been reported that there are no differences in the reductions in HbA1c and BW that can be achieved using semaglutide when the plasma concentrations achieved are similar, regardless of whether it is administered orally or subcutaneously [10]. Therefore, oral semaglutide may be an effective treatment for patients who are reluctant to self-inject or have difficulty with the injection technique. Takahara et al. [11] performed a continuous glucose monitoring (CGM) study of the effects of switching from subcutaneous semaglutide 0.5 mg to oral semaglutide 7 mg in Japanese patients with T2D, and found that the mean glucose concentrations of the participants increased by 9 mg/dL, and that this was accompanied by greater inter-individual variability. As for the satisfaction of the participants, 48% reported that they preferred the oral formulation, 35% reported that they preferred the injectable formulation, and 17% expressed no preference.
In the present study, a high baseline HbA1c was only associated with a larger reduction in HbA1c when switching from a DPP-4i to oral semaglutide. Yamada et al. [12] also showed the effectiveness of oral semaglutide in Japanese clinical settings. In this study, HbA1c decreased considerably from baseline, regardless of age, sex, BMI, the presence or absence of chronic kidney disease, and the duration of diabetes. In addition, a multivariate analysis showed that the improvement in HbA1c over 6 months was larger when the baseline HbA1c was higher, which is a similar finding to that made in the present study. In addition, a subgroup analysis of PIONEER 9 and PIONEER 10 data regarding Japanese patients with T2D showed that high baseline HbA1c is associated with a greater reduction in HbA1c, irrespective of the BMI of the patients and the identity of the other medications used [13]. Of note, the PIONEER 8 trial [14] was conducted in patients who were already being treated with basal insulin and had a long mean duration of disease (approximately 15 years), in whom the addition of oral semaglutide was effective in reducing HbA1c. Although residual pancreatic beta cell function is useful for predicting the effects of long-acting GLP-1RA [15, 16], there have been no reports of the effects of oral semaglutide on insulin secretory capacity or HbA1c. In the present study, we did not identify a relationship between the change in HbA1c and CPR/PG × 100. Therefore, further studies on the use of oral semaglutide will be required to clarify the relationship between insulin secretion capacity and its HbA1c-lowering effect.
We monitored the incidence of gastrointestinal adverse effects as a safety assessment. In the present study, nausea and heartburn were more frequent than in clinical trials, such as the extension of the PIONEER 7 trial [7]. Heartburn was less frequently reported by the participants than nausea, but many reported mild heartburn when asked, and because the two symptoms were placed in a single category, as in the present study, approximately half of the participants experienced them. However, symptoms are often mitigated by continued use and dose adjustment, and it was possible to continue the treatment for > 6 months with appropriate discussion with the participants. Gastrointestinal symptoms are common when oral semaglutide treatment is started or when the dose is increased, and the results of the present study suggest the importance of listening to patients and setting expectations in clinical practice. Of note, in this study, there was no clear association between the presence of diabetic complications and gastrointestinal symptoms with oral semaglutide. In addition, the frequency of discontinuation of the medication was higher in the participants who were aged ≥ 65 years and had BMIs < 25 kg/m2; and none of the participants who were < 65 years of age and had a BMI < 25 kg/m2 discontinued the drug. In contrast, when comparing the group aged ≥ 65 years and BMI ≥ 25 kg/m2 with the group aged < 65 years and BMI ≥ 25 kg/m2, the discontinuation rate was almost the same. We suggest that age, especially in individuals with BMIs < 25 kg/m2, may have a significant effect. This is not inconsistent with the findings of the post hoc analysis of the PIONEER 9 and 10 trials regarding discontinuation of the medication [17].
In these studies, in the participants as a whole, adverse events generally occurred more frequently in patients aged ≥ 65 vs. < 65 years, and there was a higher incidence of premature discontinuation of the medication because of adverse events in the older age group. However, the present findings may not be applicable to all patients with T2D in Japan because the participants had a mean age of 61.9 years and a mean baseline HbA1c of 7.8%. The baseline glycemic control of the cohort was satisfactory, given the mean age of the participants, which may have made them more inclined to discontinue the medication when side effects developed.
Reducing the incidence of cardiovascular events is important in the pharmacotherapy of T2D, and oral semaglutide has been shown to affect cardiovascular risk factors, such as blood pressure, circulating lipid concentrations, albuminuria, and insulin resistance. In a retrospective cohort study of 47 Japanese patients with T2D, Yanai et al. [18] found that not only the HbA1c and BW, but also the blood pressure, lipid concentrations (LDL-C and non-HDL-C), and albuminuria of the participants had improved 3 and 6 months after the initiation of oral semaglutide. After 6 months of treatment, HbA1c improved from 8.4 ± 1.5 to 7.8 ± 1.5% (p < 0.05) in 26 participants, and BW significantly decreased from 79.9 ± 19.3 to 77.3 ± 19.5 kg (p < 0.05) in 25 participants. Although 30 participants (63.8% of the total) switched from a DPP-4i to oral semaglutide in this study, the results for these individuals were not described. Arai et al. [19] reported the efficacy and safety of oral semaglutide in 16 patients with non-alcoholic fatty liver disease (NAFLD) complicated by T2D. In this study, the liver-related biochemistry, plasma glucose, HbA1c, homeostasis model assessment-insulin resistance, controlled attenuation parameter (CAP), fibrosis-4 index, ferritin, type IV collagen 7 s, and TG of the participants had improved after 24 weeks of oral semaglutide treatment.
There have been no reports of the cost-effectiveness of the use of oral semaglutide in Japan. A study conducted in Denmark that compared the cost-effectiveness of the use of oral semaglutide and subcutaneous semaglutide to that of empagliflozin, canagliflozin, and sitagliptin, showed that the daily oral or weekly subcutaneous administration of semaglutide are likely to be associated with both higher cost and greater health benefits, but are likely to be commonly considered cost-effectiveness [20]. Conversely, Jin et al. suggested that as first-line agents, SGLT2is and GLP-1RAs improve T2D outcomes, but their cost would need to fall for them to be cost-effective [21]. However, drug prices and insurance systems differ from country to country, and drug efficacy and the risks of complications are affected by ethnicity.
In the present study, 68 patients with T2D, a mean HbA1c of 7.8%, and a duration of diabetes of > 10 years who were taking a DPP-4i were switched to oral semaglutide for 6 months and showed significant improvements in HbA1c and BW. This is the first study to demonstrate the efficacy and safety of oral semaglutide in Japanese patients with T2D, and suggests that switching from a DPP-4i to oral semaglutide may represent an effective change of T2D treatment.
The present study has several limitations. First, it was a short-term, single-arm, retrospective cohort study with a small number of participants. Second, the time-in-range and mean amplitude of glycemic excursions were not assessed. Third, the effects of diet and exercise on the outcomes could not be excluded. Fourth, quality of life indices, such as patient satisfaction, were not evaluated. Fifth, the titration criteria could not be perfectly standardized due to gastrointestinal side effect or medical cost, since they were basically determined in accordance with the package insert. Sixth, participants were taking various other oral hypoglycemic agents, and the interaction between these agents and oral semaglutide could have affected the results. Therefore, further studies with patients treated with DPP-4i monotherapy should be needed.
Conclusions
Switching from a DPP-4i to oral semaglutide significantly improved the HbA1c and BM of 68 patients with T2D whose HbA1c was 7.8% ± 0.8% and had had diabetes for > 10 years. Although we did not compare oral semaglutide with injectable semaglutide, the results suggest that oral semaglutide may also be a useful therapeutic option in Japanese patients with T2D who have difficulty with the administration of injectable semaglutide. However, it should be noted that the discontinuation of this medication owing to gastrointestinal symptoms is frequent in patients > 65 years of age and with a BMI < 25 kg/m2.
Acknowledgements
We would like to thank Naotake Hashimoto, Preventive Medicine Research Center of Asahi General Hospital for useful comments. We also thank Mark Cleasby, PhD from Edanz (https://jp.edanz.com/ac) for editing a draft of this manuscript.
Data availability
The data that support the findings of this study are not openly available due to reasons of sensitivity and are available from the corresponding author upon reasonable request.
Declarations
Conflict of interest
Nobuichi Kuribayashi received a lecture fee from Novo Nordisk Pharma. Chihiro Yoneda and Junji Kobayashi declare that they have no conflicts of interest associated with this study.
Research involving human participants and/or animals.
All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and/or with the Helsinki Declaration of 1964 and later versions. The study was also conducted in compliance with the Ethical Guidelines for Medical Research Involving Human Subjects (Ministry of Health, Labour and Welfare and Ministry of Education, Culture, Sports, Science and Technology). Prior to the implementation of this study, the research protocol was reviewed and approved by the Misaki Internal Medicine Clinic Ethics Review Committee.(approval date:29 July 2023; approval number: 23–004).
Informed consent
Informed consent or a substitute thereof was obtained from all patients for inclusion in the study.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Nishimura R, Kato H, Kisanuki K, Oh A, Hiroi S, Onishi Y, Guelfucci F, Shimasaki Y. Treatment patterns, persistence and adherence rates in patients with type 2 diabetes mellitus in Japan: a claims-based cohort study. BMJ Open. 2019;9: e025806. 10.1136/bmjopen-2018-025806 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Bouchi R, Kondo T, Ohta Y, Goto A, Tanaka D, Sato H, Yabe D, Nichimura R, Harada N, Kamiya H, Suzuki R, Yamauchi T. A consensus statement from the Japanese diabetes society (JDS) a proposed algorithm for pharmacotherapy in people with type 2 diabetes. Diabetol Int. 2023;14:1–14. 10.1007/s13340-022-00605-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Kim YG, Hahn S, Oh TJ, Kwak SH, Park KS, Cho YM. Differences in the glucose-lowering efficacy of dipeptidyl peptidase-4 inhibitors between Asians and non-Asians: a systematic review and meta-analysis. Diabetologia. 2013;56:696–708. 10.1007/s00125-012-2827-3 [DOI] [PubMed] [Google Scholar]
- 4.Seino Y, Kuwata H, Yabe D. Incretin-based drugs for type 2 diabetes: focus on East Asian perspectives. J Diabetes Investig. 2016;7(Suppl 1):102–9. 10.1111/jdi.12490 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Yamada Y, Katagiri H, Hamamoto Y, Deenadayalan S, Navarria A, Nishijima K, Seino Y. PIONEER 9 investigators dose-response, efficacy, and safety of oral semaglutide monotherapy in Japanese patients with type diabetes(PIONEER 9): a 52-week,phase2/3a, randomised, controlled trial. Lancet Diabetes Endocrinol. 2020;8:377–91. 10.1016/S2213-8587(20)30075-9 [DOI] [PubMed] [Google Scholar]
- 6.Yabe D, Nakamura J, Kaneto H, Deenadayalan S, Navarria A, Gislum M, Inagaki N. PIONEER Investigators 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. 2020;8:392–406. 10.1016/S2213-8587(20)30074-7 [DOI] [PubMed] [Google Scholar]
- 7.Buse JB, Bode BW, Mertens A, Cho YM, Christiansen E, Hertz CL, Nielsen MA, Pieber TR. PIONEER 7 investigators long-term efficacy and safety of oral semaglutide and the effect of switching from sitagliptin to oral semaglutide in patients with type 2 diabetes: a 52-week, randomized, open-label extension of the PIONEER 7 trial. BMJ Open Diabetes Res Care. 2020;8:e001649. 10.1136/bmjdrc-2020-001649 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Fujiwara K, Brogaard NJ, Lophaven S, Narita Y, Tsuboi S. Profile of people initiating oral semaglutide for the treatment of type diabetes in Japan: results of a cross-sectional study using a large nationwide database. Shinryo to Shinyaku. 2023;60:273–84. [Google Scholar]
- 9.Araki E, Terauchi Y, Watada H, Deenadayalan S, Christiansen E, Horio H, Kadowaki T. Efficacy and safety of oral semaglutide in Japanese patients with type 2 diabetes: a post hoc subgroup analysis of the PIONEER 1, 3, 4 and 8 trials. Diabetes Obes Metab. 2021;23:2785–94. 10.1111/dom.14536 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Overgaard RV, Hertz CL, Ingwersen SH, Navarria A, Drucker DJ. Levels of circulating semaglutide determine reductions in HbA1c and body weight in people with type 2 diabetes. Cell Rep Med. 2021;2: 100387. 10.1016/j.xcrm.2021.100387 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Takahara M, Shiraiwa T, Katakami N, Maeno Y, Yamamoto K, Shiraiwa Y, Yoshida Y, Kozawa J, Shimomura I. Daily glucose profiles after switching from injectable to oral semaglutide in patients with type 2 diabetes mellitus. Intern Med. 2023. 10.2169/internalmedicine.1441-22. 10.2169/internalmedicine.1441-22 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Yamada H, Yoshida M, Funazaki S, Morimoto J, Tonezawa S, Takahashi A, Nagashima S, Kimura M, Otsuka K, Hara K. Retrospective analysis of the effectiveness of oral semaglutide in type 2 diabetes mellitus and its effect on cardiometabolic parameters in Japanese clinical settings. J Cardiovasc Dev Dis. 2023;10:176. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Yabe D, Deenadayalan S, Horio H, Kaneto H, Jensen TB, Terauchi Y, Yamada Y, Inagaki N. Efficacy and safety of oral semaglutide in Japanese patients with type 2 diabetes: a subgroup analysis by baseline variables in the PIONEER9 and PIONEER10 trials. J Diabetes Investig. 2022;13:975–85. 10.1111/jdi.13764 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Zinman B, Aroda VR, Buse JB, Cariou B, Harris SB, Hoff ST, Pedersen KB, Tarp-Johansen MJ, Araki E. PIONEER 8 investigators efficacy, safety, and tolerability of oral semaglutide versus placebo added to insulin with or without metformin in patients with type 2 diabetes: the PIONEER 8 trial. Diabetes Care. 2019;42:2262–71. 10.2337/dc19-0898 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Usui R, Sakuramachi Y, Seino Y, Murotani K, Kuwata H, Tatsuoka H, Hamamoto Y, Kurose T, Seino Y, Yabe D. Retrospective analysis of liraglutide and basal insulin combination therapy in Japanese type 2 diabetes patients: the association between remaining β-cell function and the achievement of the glycated hemoglobin target 1 year after initiation. J Diabetes Investig. 2018;9:822–30. 10.1111/jdi.12773 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Usui R, Yabe D, Kuwata H, Murotani K, Kurose T, Seino Y. Retrospective analysis of safety and efficacy of liraglutide monotherapy and sulfonylurea-combination therapy in Japanese type 2 diabetes: association of remaining β-cell function and achievement of HbA1c target one year after initiation. J Diabetes Complicat. 2015;29:1203–10. 10.1016/j.jdiacomp.2015.07.020 [DOI] [PubMed] [Google Scholar]
- 17.Yamada Y, Yabe D, Herz CL, Horio H, Nakayama J, Nielsen AM, Seino Y. Efficacy and safety of oral semaglutide by baseline age in Japanese patients with type 2 diabetes: a subgroup analysis of the PIONEER 9 and 10 Japan trials. Diabetes Obes Metab. 2022;24:321–6. 10.1111/dom.14571 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Yanai H, Hakoshima M, Adachi H, Katsuyama H. A significant effect of oral semaglutide on cardiovascular risk factors in patients with type 2 diabetes. Cardiol Res. 2022;13:303–8. 10.14740/cr1441 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Arai T, Atsukawa M, Tsubota A, Ono H, Kawano T, Yoshida Y, Okubo T, Hayama K, Nakagawa-Iwashita A, Itokawa N, Kondo C, Nagao M, Iwakiri K. Efficacy and safety of oral semaglutide in patients with non-alcoholic fatty liver disease complicated by type diabetes mellitus: a pilot study JGH. Open. 2022;6(7):503–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Pulleyblank R, Larsen NB. Cost‑effectiveness of semaglutide empaglifozin, canaglifozin, and sitagliptin for treatment of patients with type diabetes in Denmark: a decision‑analytic modelling study. Pharmaco Econ Open. 2023;7(4):579–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Choi JG, Winn AN, Skandari MR, Franco MI, Staab EM, Alexander J, Wan W, Zhu M, Huang ES, Philipson L, Laiteerapong N. First-line therapy for type 2 diabetes with sodium–glucose cotransporter-2 inhibitors and glucagon-like peptide-1 receptor agonists a cost-effectiveness study. Ann Intern Med. 2022;175:1392–400. 10.7326/M21-2941 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data that support the findings of this study are not openly available due to reasons of sensitivity and are available from the corresponding author upon reasonable request.