FormalPara Key Summary Points

Why carry out this study?

Although the efficacy and safety of oral semaglutide have been established in patients with type 2 diabetes in global phase 3a trials, real-world evidence regarding the effectiveness of dose escalation of oral semaglutide remains lacking

Clarifying the effects of escalating oral semaglutide to the maximum dose would aid physician decision-making

This study aimed to elucidate the therapeutic advantages of escalating the dose of oral semaglutide from 7 to 14 mg through clinical data analysis

What was learned from the study?

Our study demonstrated that escalating the dose of oral semaglutide from 7 to 14 mg in clinical practice resulted in an additional improvement of − 0.5% in HbA1c levels and a reduction of − 2.0 kg in body weight without serious adverse events

Escalating oral semaglutide to 14 mg could be an effective strategy for intensifying glycemic therapy and managing body weight in patients with type 2 diabetes

Introduction

Type 2 diabetes mellitus (T2D) is a chronic metabolic disorder characterized by hyperglycemia resulting from insulin resistance and β-cell dysfunction. The management of T2D involves a multifaceted approach, including lifestyle modifications, oral antidiabetic agents, and, if necessary, parenteral therapies. Among the therapeutic options, glucagon-like peptide 1 (GLP-1) receptor agonists have emerged as a significant class of antidiabetic agents. They are notable for their ability to improve glycemic control, promote weight loss, and reduce cardiovascular risk without increasing the risk of hypoglycemia [1,2,3]. However, GLP-1 receptor agonists have typically been available only in subcutaneous formulations, which can be a barrier to their prescription. Nonetheless, to broaden the range of therapeutic choices available to patients, a novel oral formulation of semaglutide was developed in conjunction with an absorption enhancer, sodium N-(8-[2-hydroxylbenzoyl]amino)caprylate (SNAC) [4, 5]. This is anticipated to result in the broader and earlier use of GLP-1 receptor agonists in the treatment of T2D, especially in patients who are reluctant to use injectable formulations.

Compelling evidence has been presented by the PIONEER clinical trial program, which included eight international and two Japanese trials, regarding the safety and effectiveness of oral semaglutide for the treatment of T2D at various dosages, including 3, 7, and 14 mg [6,7,8,9,10,11,12,13,14,15]. Compared to lower doses and the competitors, 14 mg of oral semaglutide demonstrated superior and clinically meaningful benefits in glycemic control and body weight loss [6, 8, 12,13,14,15]. However, the real-world implications of escalating the dose of oral semaglutide from 7 to 14 mg remain unclear. Real-world data can offer insights into the practical applications of clinical trial findings, including the effectiveness and tolerability of oral semaglutide in a broader patient population. For instance, the IGNITE study has provided early evidence for the use of oral semaglutide in routine clinical practice, highlighting potential treatment gaps and the need for dose optimization to achieve better glycemic control [16]. In that study, 37.0% of patients received only the initial 3 mg dose prescription despite the recommended dose escalation to 7 mg after 30 days; only 13.6% were escalated to the 14 mg dose.

This study aimed to examine the efficacy and safety of escalating the dose of oral semaglutide from 7 to 14 mg using real-world clinical data. By analyzing the outcomes of patients who have transitioned to the maximum dose in clinical practice, we can better understand the potential benefits and challenges associated with this dose adjustment. The findings of this study will be crucial for healthcare providers to make informed decisions regarding dose escalation to maximize the therapeutic benefits of oral semaglutide in patients with T2DM.

Methods

Study Design and Subjects

This retrospective observational cohort study, utilizing medical records, was conducted at a single center in Japan. The study period commenced on February 5, 2021 (the launch date of oral semaglutide in Japan), with the database locked on September 21, 2023. The inclusion criteria were as follows: all adult patients (aged ≥ 18 years) diagnosed with T2D who have been prescribed 14 mg oral semaglutide at least once before April 20, 2023, at the Toho University Omori Medical Center, Tokyo. The exclusion criteria included patients with type 1 diabetes, individuals with excessive alcohol consumption (more than 60 g/day for men and 30 g/day for women), those with dementia, patients hospitalized for ≥ 1 week for blood glucose management in the 12 weeks before or after the first prescription of 14 mg oral semaglutide, pregnant individuals during the data collection period, those deemed non-adherent by the primary physician, patients who initiated oral semaglutide at another medical facility, and patients deemed unsuitable for participation by the attending physician. Non-adherence was defined as patients who either did not fill their prescriptions or did not take the medication as prescribed, as verified through medical records.

The primary endpoint of this study was the change in HbA1c levels 24 weeks after the first prescription of 14 mg oral semaglutide (index date). Secondary endpoints included alterations over the 24-week period in physical assessments, such as body weight, body mass index (BMI), and blood pressure, as well as metabolic parameters, including liver enzymes, renal function, urine albumin excretion ratio (UACR), lipid profile, and uric acid levels. Additionally, secondary endpoints included HbA1c (other than the primary endpoint), the proportion of patients achieving HbA1c < 7.0%, changes in the number of glucose-lowering medications and total daily dose of insulin, and the occurrence of any adverse events (AEs). The percentage of patients who experienced a weight change of − 3% or less and − 5% or less from the index date was also calculated. In addition, the primary and secondary endpoints were assessed in two subgroups: those with shorter (early dose-titration group) and longer (late dose-titration group) durations from the initial prescription of 3 mg oral semaglutide to the first prescription of 14 mg oral semaglutide, based on the median duration. Moreover, we sought to identify potential factors associated with HbA1c improvement.

Statistical Analyses

The dataset was constructed by extracting data from three non-overlapping time periods relative to the index date: 12 weeks pre-index and index date (baseline), 12 ± 6 weeks, and 24 ± 6 weeks (Fig. 1). The measurements closest to the target day (index date, 12 weeks, and 24 weeks) were used. Continuous variables are expressed as means (± standard deviation, SD), and categorical variables are expressed as numbers or percentages. Data subsequent to the reduction or discontinuation of oral semaglutide during the observation period were not included in outcome measurements. Missing values were not imputed in the analysis. Comparison between the baseline and each observation point for primary and secondary endpoints was performed using McNemar’s test for categorical variables and the t test for continuous variables. Simple regression analysis was performed to identify potential independent predictors of outcomes associated with HbA1c levels, although multivariate regression analysis was not conducted because of the limited sample size. All statistical analyses were outsourced to EviPro Co., Ltd. (Tokyo, Japan), using SAS version 9.4 (SAS Institute, Cary, NC, USA). Statistical significance was set at a p value < 0.05.

Fig. 1
figure 1

Study design. The index date indicates the initial prescription date for 14 mg of oral semaglutide. The baseline period was defined as 12 weeks before the index date. For each endpoint, 12 and 24 weeks after the index date and the baseline were compared

Full size image

Ethical Approval

All procedures performed in this study were in accordance with the Declaration of Helsinki as revised in 2013. The Ethics Committee of Toho University Omori Medical Center, Tokyo, Japan, approved all procedures performed in this study (No. M23185 23052). This study used retrospective non-identifiable data obtained by the treating physicians; therefore, on the basis of the decision of the local Ethics Committee of Toho University Omori Medical Center, informed consent was not required. Patients had the opportunity to object using their data for retrospective scientific research; however, none objected.

Results

Patient Characteristics

Figure 2 depicts the flowchart of patient enrollment, indicating that of the 77 initially screened subjects, 11 were deemed ineligible (Fig. 2). Consequently, data analysis was performed for the remaining 66 patients who met the inclusion criteria. Table 1 summarizes the baseline clinical characteristics of the participants. Most patients were middle-aged (with an average age of approximately 56 years) and overweight or with obesity (with an average BMI of approximately 32 kg/m2). The mean duration of diabetes at baseline was approximately 11 years. At baseline, the mean HbA1c level was 7.4%, and patients were typically managed with two glucose-lowering agents in addition to oral semaglutide. Commonly prescribed glucose-lowering agents included biguanides (64%), sodium-glucose cotransporter 2 (SGLT2) inhibitors (62%), and insulin (24%). The mean duration from the initiation of 3 mg oral semaglutide to the commencement of 14 mg dosing was 136 days. Although the mean estimated glomerular filtration rate (eGFR) remained above 60 mL/min/1.73 m2 (with an average eGFR of approximately 73 mL/min/1.73 m2), alanine transaminase (ALT) levels were mildly elevated, surpassing 30 IU/L (with an average ALT of approximately 33 IU/L) (Table 2).

Fig. 2
figure 2

Flowchart of patient enrollment

Full size image
Table 1 Demographic and clinical characteristics at baseline
Full size table
Table 2 Changes in parameters over 24 weeks
Full size table

Primary Endpoint

Finally, the primary endpoint analysis was conducted on 57 patients (Fig. 2). Figure 3 shows HbA1c trajectories over 24 weeks (Fig. 3). The HbA1c level at 24 weeks following the initiation of 14 mg oral semaglutide was significantly lower than that at baseline [from 7.4 ± 1.0% at baseline to 7.0 ± 0.9% at 24 weeks (p < 0.01)]. Therefore, escalating the oral semaglutide dose from the standard 7 mg to the maximum 14 mg resulted in a notable improvement in blood glucose levels (change difference − 0.5 ± 0.8%, p < 0.01).

Fig. 3
figure 3

HbA1c trajectory over 24 weeks. Values are presented as mean ± SD. The numbers of patients were 66, 65, and 57 at 0, 12, and 24 weeks, respectively. *P values indicate differences from baseline

Full size image

Secondary Endpoint

At 12 weeks after the initiation of 14 mg oral semaglutide, the HbA1c levels showed a significant reduction compared with the baseline levels [decreasing from 7.4 ± 1.0% at baseline to 7.0 ± 0.8% at 12 weeks (p < 0.01)] (Table 2). Furthermore, the proportion of patients achieving HbA1c levels of < 7% significantly increased over 24 weeks [from 33.3% at baseline to 49.2% at 12 weeks (p < 0.01) and 54.4% at 24 weeks (p < 0.01)]. Although no notable change was observed in the number of glucose-lowering medications, a significant reduction in the total daily insulin dose was noted [decreasing from 30.9 ± 18.1 units/day to 26.3 ± 14.9 units/day at 12 weeks (p < 0.01) and 23.8 ± 17.3 units/day at 24 weeks (p < 0.01)]. Body weight also exhibited a significant decrease over the 24-week period [from 90.0 ± 20.5 kg to 89.2 ± 20.8 kg at 12 weeks (p < 0.01) and 88.2 ± 21.4 kg at 24 weeks (p < 0.01)]. During the study period, 13 patients (21.3%) at 12 weeks and 21 patients (41.2%) at 24 weeks achieved a body weight loss of ≥ 3% relative to the baseline. Additionally, 3 patients (4.9%) at 12 weeks and 8 patients (15.7%) at 24 weeks achieved a body weight loss of ≥ 5%. In contrast, no significant alterations were observed in systolic or diastolic blood pressure. Although laboratory analyses indicated potential improvements in liver enzymes, uric acid, and lipid profiles with 14 mg semaglutide, none demonstrated consistent significance over the 24-week period (Table 2). Renal function showed a slight decline at 12 weeks, with no significant difference observed between baseline and 24 weeks. No statistically significant differences were observed in UACR.

The results of the simple regression analysis investigating the association between changes in HbA1c levels at 24 weeks and patient characteristics are presented in Table S1. The BMI and HbA1c levels at baseline were positively associated with improved HbA1c levels at 24 weeks (Supplementary Material, Table S1).

Safety Outcomes

Throughout the study period, AEs were reported in 18 of 66 patients (27.3%) (Table 3). The predominant AEs were gastrointestinal disorders (10.6%), with nausea being the most commonly reported symptom (7.6%). However, most of these events were mild or moderate in severity. Notably, only one patient (1.5%) required a dose reduction of oral semaglutide due to gastrointestinal disorders. Other reasons for discontinuation or dose reduction included perceived treatment ineffectiveness by the attending physician, financial constraints, and patient requests. Additionally, a single instance (1.5%) of blood glucose-confirmed hypoglycemia (54 mg/dL) was observed, with no serious AEs.

Table 3 Summary of adverse events
Full size table

Subgroup Analysis

The median duration of the escalation of the oral semaglutide dose from 3 to 14 mg was 114 days. In the subgroup analysis, significant improvements in HbA1c levels were evident in both the early (< 114 days) and late dose-titration groups (≥ 114 days) (Supplementary Material, Table S2). However, notable weight loss was observed in the early dose-titration group. Conversely, gastrointestinal tract disorders were more prevalent in the early dose-titration group (Supplementary Material, Table S3).

Discussion

This study investigated the efficacy and safety of escalating the dose of oral semaglutide from 7 to 14 mg using real-world clinical data. Our findings indicated that escalating the dose of oral semaglutide from 7 to 14 mg in routine clinical practice resulted in notable improvements. Specifically, we observed an additional reduction of − 0.5% in HbA1c levels and a decrease of − 2.0 kg in body weight, all achieved without severe AEs. Moreover, a significant increase was observed in the proportion of patients achieving HbA1c levels < 7%, rising from 33.3% at baseline to 54.4% at 24 weeks (p < 0.01). Additionally, approximately 40% of patients achieved at least 3% weight loss compared to baseline, which is associated with beneficial effects on obesity-related risk factors and conditions according to previous findings [17]. These results align with the dose-dependent effects of oral semaglutide observed in the PIONEER clinical trials [6, 8, 12,13,14,15]. Furthermore, consistent with the results of PIONEER 8 [13], we observed a significant reduction in the total daily insulin dose from baseline. Our analysis suggests that escalating the dose of oral semaglutide to 14 mg could serve as an effective strategy for intensifying glycemic therapy and managing body weight in patients with T2D.

In the PIONEER program, the most common AEs were gastrointestinal disorders, occurring in a dose-dependent manner with oral semaglutide, requiring discontinuation in 2–12% of patients receiving 14 mg oral semaglutide [6,7,8,9,10,11,12,13,14,15]. These AEs typically manifested earlier in the study during the dose initiation and escalation phases. In the current study, gastrointestinal disorders were also the predominant AEs, occurring in 10.6% of patients. Notably, these events were more frequent in the early dose-titration group (18.2%) than in the late group (3.0%). However, these gastrointestinal disorders were mild to moderate in intensity and transient, requiring dose reduction in only one patient (1.5%), and none required discontinuation of oral semaglutide. Throughout the study period, only one case (1.5%) of blood glucose-confirmed hypoglycemia (54 mg/dL) was observed. GLP-1RAs are regarded as possessing a minimal risk of hypoglycemia owing to their mechanism of action, which is contingent on glucose levels [18]. In alignment with other GLP-1RAs, severe or blood glucose-confirmed symptomatic hypoglycemia was infrequent in the PIONEER programs, although it seemed to occur more frequently in trials combining oral semaglutide with sulfonylureas (SU) or insulin (PIONEER 3, 5, 7, and 8) [8, 10, 12, 13]. As expected, the background therapy for patients experiencing hypoglycemia in our study was SU. No cases of acute pancreatitis were reported during the study period. Overall, the AEs observed in our study were mild or moderate in severity and did not lead to drug discontinuation. These findings provide valuable reassurance for physicians hesitant to increase the dosage of oral semaglutide to the highest level because of concerns about side effects. This information can help them make informed decisions about dose escalation to achieve therapeutic benefits.

Most of the PIONEER trials indicated that oral semaglutide led to a 1–6 mmHg decrease in systolic blood pressure, and these reductions were statistically significant when comparing the effects of 14 mg oral semaglutide with those of a placebo [6,7,8,9, 12,13,14,15]. However, the present study did not show any significant changes in systolic or diastolic blood pressure. Several factors may have contributed to this inconsistency, including the limited number of patients with confirmed blood pressure changes and unspecified blood pressure measurement conditions (e.g., office vs. home measurements). Furthermore, this study investigated the effect of escalating the dose of oral semaglutide from 7 to 14 mg. Differentiating the interpretation of the effect on blood pressure from the findings of the PIONEER trial, which aimed to discern the differential effects of oral semaglutide and placebo, is important.

Non-alcoholic fatty liver disease (NAFLD) is a significant global health concern, particularly among individuals with obesity and T2D. Previous studies have highlighted the direct and indirect beneficial effects of GLP-1RAs on NAFLD, with reduced ALT and aspartate transaminase (AST) levels following semaglutide administration [19,20,21,22]. Although our study showed a tendency towards improved liver function with an escalation in the dose of oral semaglutide from 7 to 14 mg, consistent significant differences over 24 weeks were not identified.

In the PIONEER program, oral semaglutide was associated with either stable or reduced fasting lipid levels compared with other treatments, with sustained reductions over time [6,7,8,9,10,11,12,13,14,15]. Similarly, the present study indicated improvements in lipid profiles with increasing doses of oral semaglutide. However, triglyceride levels may have been influenced by dietary factors due to the lack of standardization in blood collection conditions. Further studies with larger sample sizes and controlled dietary conditions are required to elucidate the effects of oral semaglutide on lipid metabolism.

Renal function has been reported to remain unaffected during oral semaglutide treatment in the PIONEER program [6,7,8,9,10,11,12,13,14,15]. Conversely, post hoc analysis of the SUSTAIN 1–7 trials revealed initial reductions in the eGFR with semaglutide that plateaued over time, along with marked reductions in the UACR [23]. Similarly, in the present study, renal function at 12 weeks showed a slight decrease relative to baseline but did not continue to decline until 24 weeks, remaining stable at a plateau. However, no significant reduction in UACR was observed, possibly because of the small number of patients who underwent serial urinary albumin quantification, which limited adequate comparisons.

Simple regression analysis revealed a positive association between HbA1c levels at baseline and improvement in HbA1c levels at 24 weeks. This finding is consistent with previous reports indicating a greater reduction in HbA1c levels with higher baseline HbA1c levels [24, 25]. No significant interaction between the reduction in HbA1c levels and baseline BMI was observed in the subgroup analysis of the PIONEER program [24]. However, BMI at baseline was positively associated with improvements in HbA1c levels at 24 weeks in the present study. Nonetheless, this result may have been influenced by other background factors because multivariate analysis could not be performed as a result of the small number of cases. Further research is warranted to ascertain whether a higher baseline BMI is positively associated with reductions in HbA1c levels.

Avoiding clinical inertia and early intensification of diabetes therapy have been emphasized as crucial for achieving HbA1c goals [26]. In the present study, both the early and late dose-titration groups showed significant glycemic improvement, whereas only the early dose-titration group showed significant improvement in weight loss. However, the relatively short median time required for dose titration (114 days) suggests that the early dose-titration group may have been more likely to benefit from the weight loss associated with oral semaglutide initiation.

This study had some limitations. First, being a retrospective observational study conducted at a single center using historical data, the study is susceptible to confounding factors and biases. These factors may impact the results, posing challenges in ascertaining the true effects of the intervention. Additionally, it is a single-arm study without a comparator group, which limits its ability to draw direct comparisons. Although the patients were not stratified on the basis of concomitant medications, prior investigations have indicated that oral semaglutide is effective in reducing HbA1c levels and body weight, regardless of concurrent glucose-lowering agents [27]. Additionally, the constant number of glucose-lowering medications throughout the study period and the predominant reduction in the total daily insulin dose suggested a minimal likelihood of overestimating the effect of escalating the dose of oral semaglutide, validating the obtained results. However, given the potentially heightened effects of 14 mg oral semaglutide in Asian populations [24], further verification is necessary to ascertain the applicability of these results to other racial groups. AEs may have been underestimated owing to inaccurate or selective records; however, it is noteworthy that no serious AEs leading to dose reduction or discontinuation were missed owing to accurate prescription records. Second, the relatively small number of enrolled patients, although all patients prescribed 14 mg of oral semaglutide were included at the time of study planning, is a limitation. In particular, blood pressure and UACR assessments were challenging because they were often omitted or not measured. Further validation of the effects on blood pressure and UACR requires the inclusion of more cases. Third, two patients were lost to follow-up at the end of the 12-week visit as a result of hospital transfer and did not complete the full 24-week duration. Additionally, we excluded non-adherent patients, and data subsequent to the reduction or discontinuation of oral semaglutide during the observation period were not included in outcome measurements, which could introduce bias. Exclusion of non-adherent patients could lead to underreporting of adverse effects. However, only one patient was excluded because of non-adherence in this study. The patient initially switched from another GLP-1 receptor agonist to oral semaglutide but often took it after meals instead of on an empty stomach or forgot to take it altogether because of a busy morning schedule. They subsequently switched to injectable semaglutide and tolerated the maximum dosage without adverse effects, suggesting that proper oral administration would unlikely have led to adverse effects.

Future prospective studies with larger sample sizes, a comparator group continuing on 7 mg oral semaglutide, and longer follow-up durations may provide more evidence regarding the efficacy and safety of escalating the dose of oral semaglutide.

Conclusion

This retrospective observational study, utilizing real-world clinical data, revealed the efficacy and safety of escalating the dose of oral semaglutide from 7 to 14 mg. Our findings demonstrated that this dose escalation resulted in an additional improvement of − 0.5% in HbA1c levels and a reduction of − 2.0 kg in body weight, all achieved without encountering serious AEs. Escalating the dose of oral semaglutide to 14 mg may serve as an effective strategy for intensifying glycemic therapy and managing body weight in patients with T2D.