What's new?

This study is the first to systematically compare the effectiveness of oral and injectable semaglutide in lowering blood sugar level in patients with type 2 diabetes. While both forms control glucose levels effectively, the injectable one offers additional benefits in weight and triglyceride level reduction. Our findings suggest that patients and clinicians can consider the oral option without compromising glycemic control, potentially improving treatment adherence in individuals reluctant to use injections. The subcutaneous form may be favored in patients seeking additional benefits in weight management or triglyceride level reduction. Ultimately, formulation choice should be tailored to individual patient preferences, clinical goals, and comorbidities.

Introduction

Glucagon‑like peptide‑1 receptor agonists (GLP‑1RAs) have become a cornerstone of type 2 diabetes (T2D) and obesity management thanks to their pleiotropic metabolic effects extending beyond glycemic control and lowering the level of glycated hemoglobin (HbA_1c). These effects include direct body weight regulation and cardiovascular protection.¹ Semaglutide, a long‑acting GLP‑1RA, is unique in being available in both subcutaneous and oral formulations, offering flexibility in treatment options.^2,3

The subcutaneous formulation of semaglutide, given once weekly, has significantly reduced HbA_1c level and body weight in various clinical settings, as evidenced by the SUSTAIN (Semaglutide Unabated Sustainability in Treatment of Type 2 Diabetes) clinical trial program. Research trials demonstrated that subcutaneous semaglutide is superior to several other antidiabetic agents, including sitagliptin, liraglutide, and insulin glargine, in maintaining glycemic control and supporting weight reduction.^2,4

Oral semaglutide, taken once daily, offers an option for patients who prefer to avoid injections. The PIONEER (Peptide Innovation for the Early Diabetes Treatment) clinical trial program has established the efficacy of oral semaglutide in reducing HbA_1c level and body weight as comparable to its subcutaneous counterpart and superior to other oral antidiabetic agents, such as sitagliptin and empagliflozin.^2,3,5

However, no clear evidence exists to determine whether subcutaneous or oral semaglutide is superior for managing T2D or obesity. To address the frequent patient concern regarding a possibility to avoid injections, we conducted a meta‑analysis of randomized controlled trials (RCTs), followed by a formal noninferiority assessment of oral vs subcutaneous semaglutide in individuals with T2D.

The preference for oral medications over injectables, observed in many patients, holds the potential for better compliance and improved clinical outcomes.^3,6 That is why, when introducing a long‑lasting treatment, such as for T2D, a physician should consider patient preferences, particularly their aversion to injections, which can significantly impact adherence and overall treatment satisfaction. The current body of evidence is not sufficient to make a definitive recommendation on the superior route of administration, as both formulations have shown comparable efficacy and safety profiles. Our study aimed to guide clinical decision making, serving as a roadmap for personalized therapy, a critical goal in modern diabetes care.

Patients and methods

Results

Primary outcomes

Given that semaglutide is approved as an antidiabetic medication, it was expected that both oral and subcutaneous formulations would achieve clinical and statistical significance in reducing HbA_1c levels from baseline. Both formulations significantly reduced HbA_1c level from baseline, oral by 1.16 percentage points (95% CI, –1.39 to –0.92), and subcutaneous by 1.4 percentage points (95% CI, –1.66 to –1.33; Table 2; Supplementary material, Figure S1). Subsequent noninferiority testing confirmed that oral semaglutide was not inferior to the injectable form using a 50% margin of the subcutaneous effect, although the results were inconclusive at higher margins (Figure 1; Supplementary material, Figure S2).

Table 2. Meta‑analysis of oral and subcutaneous semaglutide

Outcome (mean change from baseline)	Oral semaglutide	Subcutaneous semaglutide	Mean difference subcutaneous vs oral semaglutide	P value
Values represent mean changes from baseline with 95% CIs. Lipid outcomes represent a change in lipid profiles as ratio to baseline. a Clinical significance;   b Statistical significance;   c Nonsignificant Abbreviations: DBP, diastolic blood pressure; HDL‑C, high‑density lipoprotein cholesterol; LDL‑C, low‑density lipoprotein cholesterol; SBP, systolic blood pressure; others, see Table 1
HbA1c, %	–1.16 (–1.39 to –0.92)a	–1.40 (–1.66 to –1.13)a	–0.24 (–0.59 to 0.11)	0.18
SBP, mm Hg	–2.44 (–3.47 to –1.4)b	–3.38 (–4.78 to –1.98)b	–0.94 (–2.68 to 0.8)	0.29
DBP, mm Hg	0.64 (0–1.27)c	–0.59 (–1.45 to 0.28)c	–1.23 (–2.3 to –0.16)b	0.024b
Total cholesterol, ratio	–0.03 (–0.06 to –0.01)b	–0.06 (–0.16 to 0.04)c	–0.03 (–0.13 to 0.07)	0.57
HDL‑C, ratio	0 (–0.02 to 0.03)c	0.01 (–0.04 to 0.06)c	0.01 (–0.03 to 0.05)	0.58
LDL‑C, ratio	–0.05 (–0.09 to –0.01)b	–0.03 (–0.21 to 0.08)c	–0.01 (–0.16 to 0.14)	0.9
Triglyceride, ratio	–0.06 (–0.08 to –0.03)b	–0.15 (–0.19 to –0.1)a	–0.09 (–0.14 to –0.04)b	<⁠0.001b
Weight loss, kg	–2.88 (–3.47 to –2.3)b	–5.07 (–6.32 to –3.83)b	–2.19 (–3.57 to –0.81)b	0.002b
Weight loss ≥5%, odds ratio	5.05 (3.62–7.07)b	7.56 (3.6–15.87)b	–	–
BMI, kg/m2	–0.97 (–1.2 to –0.74)b	–1.85 (–2.49 to –1.22)a	–0.88 (–1.55 to –0.21)b	0.01b
Waist circumference, cm	–2.42 (–3.19 to –1.65)b	–4.03 (–5.72 to –2.34)b	–1.16 (–3.47 to –0.09)	0.09

Safety monitoring

In Figure 2, we present details on AEs and all‑cause mortality. Due to inconsistencies in AE reporting across the analyzed RCTs, safety monitoring was categorized into 3 groups: serious AEs (SAEs), all AEs, and all‑cause mortality. The PIONEER 6 study raised concerns, as it classified all AEs as SAEs, even relatively mild ones, such as iron deficiency anemia, abdominal pain, or nausea, with no nonserious AEs reported.¹¹

Considering these discrepancies, the meta‑analysis of oral semaglutide RCTs showed no effect of the treatment on total AEs (OR, 1.5; 95% CI, 0.89–2.52) but a significant reduction in SAEs (OR, 0.81; 95% CI, 0.68–0.95) and all‑cause mortality (OR, 0.55; 95% CI, 0.34–0.9) in comparison with placebo. On the other hand, subcutaneous administration showed no difference from placebo in the odds of all‑cause mortality (OR, 1; 95% CI, 0.06–16) and SAEs (OR, 1.45; 95% CI, 0.53–3.92) but significantly increased the OR of AEs (OR, 3.22; 95% CI, 1.55–6.72; Figure 2).

Discussion

Combining the results of available RCTs, our meta‑analysis supports the use of both semaglutide formulations for effective glycemic control in the patients with T2D. Even though the noninferiority was confirmed at the relatively low margin of effect (50%), we could not prove any significant difference between the 2 formulations. The efficacy of once‑weekly subcutaneous semaglutide and once‑daily oral semaglutide has been investigated mostly in the SUSTAIN and PIONEER phase 3 clinical trial programs.^6,11-31 Notably, most of these studies compared semaglutide with other antidiabetic agents, often in combination with additional therapies or in populations with varying degrees of the disease severity. Given that our primary objective was to determine whether one administration route is superior to another, and to assess whether possible differences in clinical effectiveness should influence the choice of the formulation, we aimed to ensure maximal comparability across the included studies. Therefore, we included only the trials conducted in similar conditions—specifically, the studies involving patients without severe comorbidities, comparing semaglutide with placebo, and excluding patients requiring insulin therapy. Overall, both formulations significantly reduced HbA_1c levels, with subcutaneous semaglutide showing a reduction of 1.4 percentage points and oral semaglutide of 1.16 percentage points from baseline. Our findings suggest that both forms are similarly effective at improving glycemic control in the patients with T2D, offering flexibility without compromising outcomes. Although our study focused on comparing different formulations of semaglutide, Zhong et al⁴ demonstrated the superiority of semaglutide over other oral antidiabetic drugs in terms of glycemic control and weight loss. Their analysis did not statistically assess the difference between oral and subcutaneous semaglutide, nevertheless, they showed a reduction in HbA_1c level similar to that observed in our analysis (1.37 percentage points for subcutaneous vs 1.02 percentage points for oral semaglutide). Notably, their meta‑analysis did not include more recent and large‑scale trials, such as PIONEER 6,²⁴ PIONEER 11,²⁶ or STEP 2 (Research Study Investigating How Well Semaglutide Works in People With Type 2 Diabetes Suffering From Overweight or Obesity),²⁹ but included trials with active comparators and analyzed a more heterogeneous patient population. In contrast, a recent meta‑analysis by Karedath et al³² reported a numerically greater HbA_1c level reduction with subcutaneous semaglutide (standardized MD, 0.21; 95% CI, 0.04–0.38), consistent with our finding that the injectable formulation had an insignificantly stronger effect on HbA_1c level reduction (1.4% vs 1.16% for oral formulation). However, their analysis was limited by vague inclusion criteria, reliance on only 4 studies, most of which were retrospective, and the absence of a formal noninferiority assessment.

Since T2D is associated with significant morbidity and mortality, primarily due to its complications, including cardiovascular diseases or diabetic kidney disease, applying a treatment that also affects comorbidities seems to be especially important in managing diabetic patients.^33,34 From this perspective, GLP‑1RAs and sodium‑glucose cotransporter‑2 (SGLT‑2) inhibitors emerged as recommended treatment options by the American Diabetes Association (ADA) Standards of Care and the ADA / European Association for the Study of Diabetes.^35,36

The DAPA‑CKD (Dapagliflozin and Prevention of Adverse Outcomes in Chronic Kidney Disease) trial³⁷ demonstrated that dapagliflozin significantly reduced the risk of kidney failure, death from cardiovascular causes, and hospitalization for heart failure in patients with chronic kidney disease (CKD), both with and without diabetes.³⁷ A pooled analysis of SUSTAIN 6 and LEADER (Liraglutide Effect and Action in Diabetes: Evaluation of Cardiovascular Outcome Results) trials indicated that semaglutide and liraglutide significantly reduced albuminuria and slowed the decline in eGFR in patients with T2D, suggesting nephroprotective effects.³⁸ Similar results were observed in a recent FLOW study (A Research Study to See How Semaglutide Works Compared to Placebo in People with Type 2 Diabetes and Chronic Kidney Disease), where semaglutide had renoprotective effects in people without diabetes but suffering from CKD, showing a 24% lower hazard ratio for major kidney disease events, as compared with placebo.³⁹

While both SGLT‑2 and GLP‑1RAs provide nephroprotection, the latter also lead to significant weight loss, which further reduces cardiovascular risk. In terms of weight management, subcutaneous semaglutide was superior, achieving greater weight loss and BMI reduction than the oral form. However, the ORs for achieving at least 5% of initial body weight loss did not differ between the formulations. Subcutaneous semaglutide reduced body weight by on average 5.07 kg, while oral semaglutide showed a weight reduction of 2.88 kg. This observation is further supported by the findings of Shi et al,⁴⁰ who reported greater weight loss with injectable semaglutide in a comprehensive network meta‑analysis of contemporary T2D treatment. However, for other outcomes, their analysis treated all GLP‑1 RAs as a single group, limiting conclusions specific to semaglutide formulations. Additionally, Formichi et al⁴¹ found superior outcomes in weight and BMI reduction, as well as waist circumference, with subcutaneous semaglutide in comparison with the oral form in a retrospective cohort. Yet, significant baseline differences between the groups, such as higher initial body weight and shorter diabetes duration in the subcutaneous group, may have influenced these results. It is worth noting, however, that our study included only trials on the patients with T2D. The analyzed trials rarely reported initial body weight, but we may assume that the majority of patients had normal weight or overweight. In contrast, most trials of the STEP program involved obese or overweight patients without diabetes. Most of the published RCTs with subcutaneous semaglutide were done for weight management purposes. In such a population, the effect of semaglutide was far more pronounced in comparison with a diabetic population.

The OASIS 1 trial⁴² (Research Study to Investigate How Well Semaglutide Tablets Taken Once Daily Work in People Who Are Overweight or Living With Obesity) evaluated the effects of oral semaglutide 50 mg daily in adults with overweight or obesity without T2D. The study found that semaglutide led to a significant mean body weight reduction of 15.1%, as compared with 2.4% on placebo over 68 weeks. Additionally, 85% of the participants in the semaglutide group and 26% in the placebo group achieved at least 5% weight loss. The STEP 3 trial⁴³ assessed the impact of once‑weekly subcutaneous semaglutide at 2.4 mg combined with intensive behavioral therapy in adults with overweight or obesity. The results showed a mean weight reduction of 16% on semaglutide vs 5.7% on placebo over 68 weeks. Furthermore, 86.6% of the participants on semaglutide achieved at least 5% weight loss, as compared with 47.6% in the placebo group.⁴³ A recent systematic review and meta‑analysis of RCTs, including the STEP trials, confirmed that semaglutide at 2.4 mg once weekly resulted in a mean weight loss of 12.1% relative to placebo, with 33.4% of the participants achieving at least 20% weight loss.⁴⁴ The SELECT trial⁴⁵ (Semaglutide Effects on Heart Disease and Stroke in Patients With Overweight or Obesity) also demonstrated long‑term weight loss benefits with semaglutide, showing a mean weight reduction of 10.2% sustained over 208 weeks in patients with obesity and preexisting cardiovascular disease. Higher effectiveness in those trials, however, may arise from higher initial body weight in the studies in the obese / overweight population without T2D, which is a known confounding factor for weight loss.

Discussing the cardiovascular risk, the effect of semaglutide on BP and lipid profile needs to be mentioned. In our analysis, both oral and subcutaneous semaglutide significantly reduced SBP, but neither formulation achieved a clinically significant reduction. In terms of lipid profiles, oral semaglutide significantly lowered total cholesterol, LDL‑C, and triglyceride levels, but these changes were not clinically meaningful. On the other hand, subcutaneous semaglutide significantly reduced triglyceride levels, reaching both clinical and statistical significance, while showing no impact on other lipid parameters. Our results seem to be in line with STEP 1 and STEP 4 trials on obese / overweight patients without diabetes.⁴¹ As shown in the meta‑analysis of their results, subcutaneous semaglutide at 2.4 mg once weekly led to reductions in SBP and DBP, as compared with placebo. Specifically, a mean reduction was approximately 4.78 mm Hg for SBP and 2.56 mm Hg for DBP. These reductions varied in different weight loss categories, and were maintained with continued treatment but deteriorated upon discontinuation. Similarly, semaglutide improved the lipid profile, but the results were strongly related to the percentage of weight lost, reaching clinical significance only in the subgroups with the greatest effect (≥15% total weight loss).⁴⁶

Regarding safety, oral semaglutide seems to be a better‑tolerated formulation. It showed no significant influence on total AE rate but significantly reduced serious AEs and all‑cause mortality in comparison with placebo. Subcutaneous semaglutide, however, increased the number of total AEs but showed no difference in all‑cause mortality and SAEs in comparison with placebo. Nevertheless, it has to be pointed out that for both formulations the total number of death cases was small, and all‑cause mortality reduction on oral semaglutide results from a single study, PIONEER 6.²⁴

Considering the persistent nature of the disease, patient adherence to the therapy is especially important in managing T2D. Thus, patient preferences should also be considered in treatment selection. Clinically, both formulations of semaglutide are comparably effective in glycemic control, with subcutaneous semaglutide offering additional benefits in weight management. Oral semaglutide may be preferred by patients who are averse to injections, potentially improving adherence. However, for those needing stronger weight loss effects, the subcutaneous form may be preferable.^41,47 That is why physicians should consider the specific clinical needs of each patient when selecting the appropriate formulation of semaglutide.