Relative Effectiveness and Safety of the GLP-1 (Glucagon-Like Peptide 1) Receptor Agonists, Semaglutide and Liraglutide in the Treatment of Obese Type 2 Diabetics: A Prospective Observational Cohort Study in Poland

Abstract

Purpose

To compare the efficacy and safety of semaglutide and liraglutide over one year in obese Polish patients with type 2 diabetes.

Patients and Methods

In this prospective, observational cohort study conducted in Poland in 2024, 460 patients aged 18–80 were enrolled: 333 received semaglutide (Group 1), and 133 received liraglutide (Group 2).

Results

After 12 months, HbA1c levels significantly decreased in both groups: Group 1: from 6.09 ± 1.14% to 5.42 ± 0.82% (mean decrease: 0.67 ± 0.37%, p<0.0001). Group 2: from 5.78 ± 0.75% to 5.17 ± 0.54% (mean decrease: 0.61 ± 0.28%, p<0.001). BMI decreased by 5.36 ± 3.45 kg/m² in Group 1 and 4.41 ± 4.63 kg/m² in Group 2 (p<0.0001), with greater reduction in Group 1 (p=0.017). Gastrointestinal adverse effects were most common, including nausea, vomiting, constipation, gastritis, and diarrhea, with higher incidence in Group 1. HbA1c reduction correlated with baseline HbA1c, age, and gender; BMI reduction correlated with baseline HbA1c and BMI.

Conclusion

Semaglutide was more effective than liraglutide in reducing HbA1c and BMI. Gastrointestinal symptoms were the most frequent side effects in both groups.

Introduction

In 2025, two terms were introduced into clinical practice: “preclinical obesity” and “clinical obesity”, aimed at early diagnosis and treatment of patients with excess adipose tissue and its health consequences.¹ However, for over half a century, another term, “diabesity”, has been used to demonstrate the close link between excess body weight and type 2 diabetes mellitus (T2DM).² Excess visceral fat leads to insulin resistance and is ultimately a milestone in the development of diabetes.²

The sedentary lifestyle behavior, together with the high-caloric diet, results in increasing number of people with obesity, the incidence of which has more than doubled in the years 1990–2022.² In 2022, 16% of the population aged 18 years and over was obese, and there was about 890 million obese adults worldwide.³ It is estimated that the number of obese adults will rise to 1.53 billion in 2035, an increase of 72%.⁴ According to data from 2021, 10.5% of the population aged 20–79 years had diabetes, with almost half unaware of this condition. It is projected that by 2045, one in eight adults, approximately 783 million, will have type 2 diabetes, an increase of 46%.⁵

USD 4.32 trillion—nearly 3% of global gross domestic product (GDP) —annually by 2035. This is comparable with the impact of covid-19 in 2020. By 2060, the global annual costs of obesity and overweight (OAO) would reach USD 18 trillion.⁶ In other words, if current trends are allowed to continue, economic impacts of OAO are estimated to rise to an average of 3.29% of GDP globally on average in 2060.^6,7

The health consequences of obesity are well defined, and include, among others, T2DM, dyslipidemia, arterial hypertension, metabolic-associated fatty liver disease (MAFLD), obstructive sleep apnea (OSA), heart failure, ischemic heart disease, cancer and osteoarthritis.¹ All of them are responsible for higher general and cardiovascular morbidity and mortality among obese patients. In 2019, 5 million deaths from non-communicable diseases were associated with a higher than optimal body mass index (BMI).⁸

The pharmaceutical industry has attempted to synthesize a drug that could have a positive effect on all the metabolic effects of obesity. This ideal drug effectively reduces body weight, normalizes HbA1c levels, normalizes elevated non-HDL cholesterol levels and blood pressure values, while maintaining a good safety profile. These attempts led to the synthesis of glucagon-like peptide-1 (GLP-1) analogues. GLP-1 is an incretin hormone that is responsible for glucose homeostasis and the post-prandial insulinotropic incretin effect.^9–11 It is the 30 amino acid peptide, derives from L cells in the distal ileum and colon.⁹ According to current guidelines, GLP-1 receptor agonists (GLP-1RA) are the first-line therapy in T2DM, especially in patients with atherosclerotic cardiovascular disease or obesity, and some of them are recommended for obese people without T2DM.^12,13 Among them, semaglutide and liraglutide are widely used. Despite their widespread use, few studies have directly compared the long-term efficacy and safety of semaglutide versus liraglutide in routine clinical settings. Furthermore, no published data to date have evaluated this comparison in obese Polish patients with T2DM—a group characterized by unique epidemiological and healthcare system factors. Such population-specific evidence is vital to inform national treatment strategies and optimize therapeutic outcomes.

The aim of the present study was to compare efficacy and safety of two GLP-1 analogues, semaglutide and liraglutide, in the one-year-long treatment of T2DM in obese Polish patients.

Materials and Methods

Study Design

A prospective, observational cohort study was conducted in Poland from 1 January 2024 to 30 December 2024. Based on data from major clinical trials (eg, SUSTAIN-10 and LEAD-6), the expected difference in HbA1c reduction between semaglutide and liraglutide was approximately 0.7%, with a standard deviation (SD) of 1.1%. For BMI reduction, the expected difference was 2.0 kg/m², with an SD of 3.5 kg/m².^14,15 Using the formula for comparing two means, with a two-sided α = 0.05 and power = 80%: the minimum required sample size to detect a difference in HbA1c was 39 patients per group (78 total) and the minimum required sample size to detect a difference in BMI was 48 patients per group (96 total). Based on experiences with less cooperative patients, finally 460 consecutive outpatients and inpatients aged between 18 and 80 were recruited to the study. During pre-treatment period, they were presented with therapeutic options in accordance with the current guidelines of the Polish Diabetes Association.¹³ Therapeutic options were fully discussed with the patient during several medical consultations with his attending physician. All patients were free to choose the study drug. Because subcutaneous injections of semaglutide are given once a week, this option was more likely to be chosen by the patients. During the study period, each subject successively increased the dose of semaglutide and liraglutide: in case of semaglutide from 0.25 mg/once- weekly up to 1.0 mg/once-weekly; in case of liraglutide – from 0.6 mg/day up to 3.0 mg/day. Patients who experienced adverse events associated with GLP-1 analogues and who did not tolerate maintenance doses of both drugs were not included in the study group.

Eligibility Criteria

The criteria marked for inclusion are shown below. (1) Participants: adult patients (aged 18–80 years) with the similar duration of obesity and with a newly diagnosed T2DM, pharmacologically naive, with a similar state of metabolic balance (similar HbA1c level), not achieving weight loss during previous (at least one year) non-pharmacological treatment for obesity (2) ability to give informed consent, understand and comply with the study procedures. (3) No use of any weight loss or weight gain drugs other than used in the study.

The criteria marked for exclusion are shown below. (1) ketoacidosis, acute myocardial infarction (AMI), congestive heart failure (CHF), renal and liver failure, being pregnant and breastfeeding (2) no ability to give informed consent, understand and comply with the study procedures. (3) using other anti-obesity drugs during the study period.

The decision to include a patient in the study was made by their attending physician. All the recruited patients were fully informed about study goals and conditions. Before the survey, they gave written informed consent to participate in the project. The Bioethics Committee of the Poznan University of Medical Sciences approved the study protocol [149/24]. This study was conducted in accordance with the Declaration of Helsinki.

Study Technique

Patients Were Divided into Two Groups

Group 1

333 patients treated with semaglutide (at a maintenance dose of 1 mg once-weekly, administered subcutaneously). During the study period, this was the highest dose of subcutaneous semaglutide registered on the Polish pharmacy market.

Group 2

133 patients treated with liraglutide (at a maintenance dose of 3 mg once-daily, administered subcutaneously).

At baseline and after follow-up of 12 months active treatment, fasting blood samples were drawn, a physical examination was carried out and updated patient’s medical records were analysed. Glycated haemoglobin (HbA1c) and BMI were estimated. HbA1c was measured by high-performance liquid chromatography and by an immunoturbidimetric method.¹⁶ Weight, height, waist and hip circumferences were measured and BMI was calculated by a nurse specialized in diabetes care. Patients were weighed wearing underclothes. The assessment of the safety profile of the analyzed GLP-1 analogues, semaglutide vs liraglutide, was conducted for each patient enrolled in the study after three months of active treatment. The safety profile was evaluated by the attending physician during a follow-up visit based on an interview with the patient. The results were then recorded in the patient’s medical history.

The patient’s demographic data (age, gender), information on the progress of the disease (presence of comorbid diseases) and control were also collected at baseline from an analysis of the patient’s medical records.

Statistical Analysis

The quantitative parameters were presented using mean value, median, and standard deviation. Categorical data were presented as counts and percentages. The results between analysed groups were compared using the Student’s t-test. Where data did not follow the normal distribution (Shapiro–Wilks test), the comparison was performed using the Mann–Whitney test. The Chi-square test for independence was used to analyse categorical data. The relationship between the decline HbA1c and BMI level as a result of applied GLP-1 analogues treatment and study parameters was analysed using multivariable regression analysis. The analysis was performed with the use of the TIBCO Software Inc (2017) statistical package. Statistica (data analysis software system), version 13. http://statistica.io. All tests were considered significant at p<0.05.

Results

Study Group Characteristics

The study groups numbered 460 patients: 333 patients treated with semaglutide (Group 1) and 133 patients treated with liraglutide (Group 2). Both groups did not differ significantly in terms of gender. Female predominated in both study groups (58.26% in the group of patients who received semaglutide and 60.90% in the group of patients who received liraglutide). At baseline, both groups differed statistically significant in terms of presence of hypertension (39.64%- group 1 vs 28.79%-group 2, p=0.028). There were no statistically significant differences between both groups in terms of other analysed parameters (Table 1).

Table 1.

Baseline Demographic and Clinical Characteristics of Patients with Diabetes Type 2 and Obesity Treated by Semaglutide vs Liraglutide, n=466

	Semaglutide	Liraglutide	P value
Group size n	333	133	0.648
Female n (%)	194 (58.26)	81 (60.90)
Male n (%)	139 (41.74)	52 (39.10)
Age n (%)			0.227
18–25	17 (5.11)	4 (3.01)
26-35	82 (24.62)	42 (31.58)
36–50	180 (54.05)	73 (54.89)
51-70	1 (0.30)	1 (0.75)
>70	53 (15.92)	13 (9.77)
Duration of the obesity (years) (mean ± SD)	5.48 ± 5.48	5.33 ±5.12	0.743
Body mass index (kg/m2) (mean ± SD)	35.34 ± 5.89	36.09 ± 6.15	0.231
HbA1c (%) (mean ± SD)	6.09 ± 1.14	5.78± 0.75	0.123
Hypertension n (%)	132 (39.64)	38 (28.79)	0.028
Hypothyroidism n (%)	65 (19.52)	37 (27.82)	0.053
Hypercholesterolemia n (%)	80 (24.10)	32 (24.06)	0.993
Obstructive sleep apnea syndrome n (%)	25 (7.51)	12 (9.02)	0.585

Abbreviation: SD, standard deviation.

Comparison of the Therapeutic Effect of Semaglutide and Liraglutide

Glycaemic Control

The performed statistical analysis of data revealed that after 12 months, HbA1c level significantly decreased from 6.09 ± 1.14% to 5.42 ± 0.82 (mean decrease- 0.67 ± 0.37%, p<0.0001) in the semaglutide group. Similar decrease was observed in patients treated with liraglutide 5.78 ± 0.75% to 5.17 ± 0.54% (mean decrease- 0.61 ± 0.28, p<0.001). There was no statistically significant difference in HbA1c reduction between both groups (p=0.711, Table 2) (Figure 1).

Table 2.

Comparison Between the Therapeutic Effect of Semaglutide versus Liraglutide in Patients with Diabetes Type 2 and Obesity, n=466

Outcome Parameters	Baseline		Follow-up		Mean Change from Baseline ± SD		P-Value
	Semaglutide	Liraglutide	Semaglutide	Liraglutide	Semaglutide	Liraglutide
HbA1c (%) (mean ± SD)	6.09 ± 1.14	5.78 ± 0.75 0.123*	5.42 ± 0.82	5.17 ± 0.54 0.009**	0.67 ± 0.37 p<0.0001***	0.61 ± 0.28 p<0.0001***	0.711#
Body mass index (kg/m2) (mean ± SD)	35.34 ± 5.89	36.09 ± 6.15 0.203*	29.98 ± 5.81	31.42 ± 6.03 0.017**	5.36 ± 3.45 p<0.0001***	4.41 ± 4.63 p<0.0001***	0.017#
Waist Circumference (cm) (mean ± SD)	112.3 ± 10.4	111.5 ± 10.7 0.428*	101.6 ± 9.8	104.3 ± 10.2 0.007**	10.7 ± 5.3 <0.0001***	7.2 ± 5.9 <0.0001***	0.002#

Notes: *p-value for between-group comparison (Semaglutide vs Liraglutide) at baseline. **p-value for between-group comparison at follow-up. ***p-value for within-group comparison (baseline vs follow-up). #p-value for comparison of changes between groups (Semaglutide vs Liraglutide).

Figure 1.

HbA1c Change After 12 Months of Treatment. This figure illustrates the change in glycated hemoglobin (HbA1c) levels in patients treated with Semaglutide and Liraglutide from baseline to 12 months. Bars represent mean HbA1c levels expressed as a percentage. Semaglutide and Liraglutide both led to a reduction in HbA1c after 12 months of treatment. Baseline values are shown in light blue, and follow-up values after 12 months are shown in dark blue.

Body Mass Index

After 12 months, a significant decrease in BMI was observed in both groups, by 5.36 ± 3.45 kg/m2 in the semaglutide group and by 4.41 ± 4.63 kg/m2 in the liraglutide group (p<0.0001) (Table 2). Semaglutide compared to liraglutide resulted in significantly greater reduction in BMI in the analyzed study period (p=0.017) (Table 2) (Figure 2).

Figure 2.

BMI Change After 12 Months of Treatment. This figure compares the body mass index (BMI) of patients treated with Semaglutide and Liraglutide at baseline and after 12 months. The bars represent mean BMI values in kg/m². Semaglutide and Liraglutide both show a decrease in BMI over the 12-month treatment period. Baseline values are shown in light green, and follow-up values after 12 months are shown in dark green.

Waist Circumference

After 12 months, a significant decrease in waist circumference (WC) was observed in both groups, by 10.7 ± 5.3 cm in the semaglutide group and by 7.2 ± 5.9 cm in the liraglutide group (p<0.0001) (Table 2). Semaglutide compared to liraglutide resulted in significantly greater reduction in waist circumference in the analyzed study period (p=0.002) (Table 2) (Figure 3). We did not have prior data on the expected difference in waist circumference between groups. However, the observed effect size was 3.5 cm with a standard deviation of 5.6 cm. Based on these values, the minimum required sample size to achieve a statistical power of 80% would have been 41 participants per group. Given our actual sample sizes, the calculated power of the test was approximately 99.9%.

Figure 3.

Waist Circumference Change After 12 Months of Treatment. This figure illustrates the changes in waist circumference for patients treated with Semaglutide and Liraglutide over a 12-month period. Bars represent mean waist circumference values in centimeters. A reduction in waist circumference is observed in both treatment groups after 12 months. Baseline measurements are represented in blue, while follow-up measurements are shown in green. Numerical values are annotated above each bar.

Comparison of the Safety Profile of Semaglutide and Liraglutide

Both analyzed GLP-1 analogues primarily caused gastrointestinal adverse effects (GI AEs), including nausea (48.64% - 1 vs 41.35% - 2; p = 0.154), vomiting (47.74% - 1 vs 36.09% - 2; p = 0.022), constipation (45.64% - 1 vs 41.35% - 2; p = 0.400), gastritis (26.72% - 1 vs 21.81% - 2; p = 0.270), and diarrhea (15.91% - 1 vs 23.31% - 2; p = 0.061). GI AEs were observed more frequently in semaglutide group than in liraglutide group. Semaglutide therapy was statistically significantly associated with a higher incidence of vomiting compared with liraglutide therapy (p = 0.022) (Table 3).

Table 3.

Comparison Between the Safety Profile of Semaglutide versus Liraglutide in Patients with Diabetes Type 2 and Obesity, n=466

	Semaglutide n=333	Liraglutide n=133	P value
Outcome parameters
Nausea n (%)	162 (48.64)	55 (41.35)	0.154
Vomiting n (%)	159 (47.74)	48 (36.09)	0.022
Constipation n (%)	152 (45.64)	55 (41.35)	0.400
Gastritis n (%)	89 (26.72)	29 (21.81)	0.270
Diarrhoea n (%)	53 (15.91)	31 (23.31)	0.061
Rash n (%)	8 (2.40)	3 (2.25)	0.925
Abdominal pain n (%)	6 (1.80)	1 (0.75)	0.400
Reflux n (%)	6 (1.80)	0 (0.00)	0.119
Vision disorders n (%)	25 (7.51)	5 (3.75)	0.137
Headaches n (%)	30 (9.01)	8 (6.01)	0.286

Multivariable Regression Analysis of Predictive Factors

The multivariate regression analysis for confounding factors influencing the reduction of HbA1c in both groups showed that this parameter was significantly associated with the baseline HbA1c value, the type of GLP-1 analogue, gender and age. In turn, the final reduction in BMI was significantly associated with the baseline HbA1c and BMI value (Table 4 and Table 5). BMI reduction was greater in the semaglutide-treated group (Table 5).

Table 4.

Multivariable Regression Analysis of Decline HbA1c Level as a Result of Applied GLP-1 Analogues Treatment in Patients with Diabetes Type 2 and Obesity, n=466

Variable	Change in HbA1c
	Coefficient (95% CI)	P value
Gender (male)	0.08 (0.01–0.14)	0.022
Age
26-35	0.091 (0.02–0.16)	0.010
36-50	0.095 (0.03–0.16)	0.006
Above 50	0.096 (0.02–0.17)	0.012
Baseline value of HbA1c (%)	0.28 (0.27–0.30)	<0.0001

Table 5.

Multivariable Regression Analysis of Decline BMI Level as a Result of Applied GLP-1 Analogues Treatment in Patients with Diabetes Type 2 and Obesity, n=466

Variable	Change in BMI
	Coefficient (95% CI)	P value
Baseline value of HbA1c (%)	−0.35 (−0.59–0.11)	0.004
Baseline value of BMI (kg/m2)	0.17 (0.13–0.22)	< 0.0001
GLP-1 type (semaglutide)	0.96 (0.19–1.73)	0.015

Discussion

This study aims to compare the safety and effectiveness of semaglutide and liraglutide among Polish obese patients with T2DM. According to the presented results, semaglutide had a significant effect on weight loss and caused a more pronounced dose-dependent reduction in HbA1c compared to the liraglutide group. The present analysis extends the findings of our primary study by incorporating WC as a measure of visceral adiposity, an important factor in the metabolic risk profile of patients with T2DM. While BMI is traditionally used to assess obesity-related risk, it does not distinguish between lean and fat mass, nor does it reflect the distribution of adiposity. In contrast, WC—and by extension, the waist-to-height ratio (WHtR)—serves as a more accurate surrogate marker for central obesity and cardiometabolic risk, especially in individuals with T2DM and insulin resistance.^17,18

Our results indicate that treatment with semaglutide resulted in a significantly greater reduction in WC compared to liraglutide after 12 months of therapy (−10.7 ± 5.3 cm vs −7.2 ± 5.9 cm; p = 0.002). These findings are clinically relevant as central adiposity, particularly visceral fat, is strongly linked to the pathogenesis of insulin resistance, chronic inflammation, and cardiovascular disease.¹⁹ The observed reduction in WC aligns with previous studies that have demonstrated the superior efficacy of semaglutide in reducing total and abdominal fat mass compared to other GLP-1 receptor agonists.^20,21

Moreover, a growing body of evidence supports WHtR as a more consistent predictor of T2DM and cardiovascular risk than BMI.²² Ashwell et al proposed a WHtR cut-off of 0.5 as a simple and universal boundary to define increased risk across different ethnic groups and ages.¹⁷ Given that WC is a critical determinant of WHtR, the pronounced decrease in WC among semaglutide-treated patients suggests a likely improvement in WHtR and, by implication, cardiometabolic health.

The presented findings are consistent with prior real-world and clinical trial data showing that semaglutide leads to greater reductions in total body weight and central fat. For instance, the STEP trials revealed that semaglutide 2.4 mg weekly produced substantial reductions in visceral fat compared to placebo, accompanied by improvements in metabolic parameters.²³ Similar results were confirmed in meta-analyses showing that semaglutide leads to more favorable changes in body composition, including fat mass, lean mass, and WC, than liraglutide.^19,24

Importantly, the reduction in abdominal adiposity may partially explain semaglutide’s favorable impact on cardiovascular outcomes observed in major trials such as SUSTAIN-6.²⁵ This reinforces the clinical utility of WC as a monitoring parameter, especially in patients where BMI changes may not fully capture improvements in metabolic health.

In conclusion, the greater reduction in WC observed with semaglutide treatment underscores its superior efficacy not only in general weight loss but specifically in reducing visceral adiposity—a key target in the management of T2DM. This further supports semaglutide’s utility in patients with central obesity and highlights the importance of integrating WHtR or WC as secondary outcome measures in future studies and clinical practice.

It is worth noting that the improvement in glycemic control occurred while maintaining a good safety profile. The most common AEs among study participants were GI side effects. Compared with liraglutide, all semaglutide dosing regimens resulted in more GI AEs. The risk of GI side effects increased with increasing semaglutide dose.

Subcutaneous Semaglutide and Liraglutide in Meta-Analyses and Clinical Trials

Our findings are in line with conclusions from several recent meta-analyses. A meta-analysis conducted by Wen. et al revealed that semaglutide had a greater impact on mean body weight reduction compared to liraglutide. Semaglutide and liraglutide were associated with minimal to moderate severity side effects, most of which were GI.²⁶ Another meta-analysis also by Wen et al proved that semaglutide had a greater mean weight loss compared to liraglutide, and all of analyzed GLP-1 analogues were responsible for high rates of minimal to moderate-severity AEs.²⁷

Ma et al included sixty-one RCTs in their meta-analysis. Based on intraclass comparison, semaglutide 2.4 mg was among the most effective interventions in decreasing body weight, HbA1c level and fasting plasma glucose, reducing blood pressure values.²⁴ Similarly, in a meta-analysis by Hu et al semaglutide was superior to other GLP-1 analogues in decreasing body mass and improving glycemic control. What is more, semaglutide proved to be associated with superior outcomes for heart failure and cardiovascular death in obese patients without T2DM, whereas liraglutide showed negative impact on heart failure outcomes in a diabetic population with a reduced ejection fraction. Both GLP-1 analogues were significantly associated with less new or persistent macroalbuminuria.²⁰ It is worth emphasizing, that a recently published meta-analysis confirmed that a reduction in body mass, HbA1c level and fasting plasma glucose was more pronounced in the case of therapy with once-weekly semaglutide in comparison to once-daily liraglutide. However, the percentage of severe AEs and GI side effects was similar during both therapies.²¹

Body mass reduction is slightly increased in patients suffering from GI AEs than those who do not. A mediation analysis of the SUSTAIN 1–5 trials showed minor contribution of nausea/vomiting to the greater weight reduction during the therapy with once-weekly semaglutide versus its comparators.²⁸ Lingvay et al analyzed the data from SUSTAIN 3, 7 and 10, and concluded that nausea/vomiting contributed minimally (<0.1 kg) to the superior weight reduction with semaglutide versus GLP-1RA comparators at the end of treatment.²⁹ However, a recent study that showed a significant difference in the profiles of reporting risk and time-to-onset during the treatment with semaglutide or liraglutide, underlines that individual treatment choices should be considered when choosing an agent among GLP-1 analogues.³⁰

Murvelashvili et al showed that semaglutide 1.0 mg weekly lead to superior weight loss compared with liraglutide 3.0 mg daily for treating post-metabolic and bariatric surgery (MBS) weight recurrence, regardless of procedure type or the magnitude of weight recurrence.³¹

In a real-world study conducted by Gameil et al it was shown, that semaglutide and dulaglutide were associated with the lower risk of clinically relevant cholelithiasis than liraglutide in patients with T2DM.³²

Several studies have directly compared the efficacy and safety of semaglutide and liraglutide, contributing to the growing body of evidence supporting the superior weight loss outcomes associated with semaglutide. In a pivotal Phase 2 randomized controlled trial by O’Neil et al, semaglutide demonstrated significantly greater reductions in body weight compared to liraglutide (−13.8% to −11.2% vs −7.8%) over 52 weeks, with up to 65% of patients achieving ≥10% weight loss.³³ Similarly, Deng et al reported, through a systematic review of 18 studies involving non-diabetic overweight/obese adults, that 86.6% of semaglutide-treated participants achieved ≥5% weight loss versus 65.3% with liraglutide, reinforcing semaglutide’s higher efficacy.³⁴

Head-to-head switching studies further support semaglutide’s clinical advantage. Iijima et al observed that patients with T2DM switching from liraglutide to semaglutide achieved significantly improved glycemic control and greater weight loss compared to those switched to dulaglutide.³⁵ In another comprehensive systematic review and network meta-analysis, Xie et al confirmed semaglutide 2.4 mg’s superior weight reduction effect (−12.47 kg) compared with liraglutide 3.0 mg (−5.24 kg), albeit with a higher incidence of adverse events.³⁶

Real-world evidence also aligns with clinical trial findings. Rana et al examined a veteran population and found no statistically significant difference in HbA1c reduction between semaglutide and liraglutide; however, semaglutide was associated with greater numerical reductions in both insulin requirements and body weight.³⁷ Lastly, Alsugair et al conducted a network meta-analysis and demonstrated semaglutide’s superiority over liraglutide in reducing both HbA1c and body weight, particularly at the 1 mg once-weekly dose.³⁸

Collectively, these findings highlight semaglutide’s greater efficacy in weight loss and glycemic control compared to liraglutide across diverse populations and study designs, although tolerability and individual patient characteristics should guide therapeutic choices.

Oral Semaglutide in Clinical Trials and Meta-Analyses

In the present study, participants were treated with semaglutide administered subcutaneously. Numerous studies have compared the efficacy and safety of oral semaglutide with other GLP-1 analogues. In PIONEER-4 study, oral semaglutide was non-inferior to subcutaneous liraglutide and superior to placebo in reducing HbA1c level, and superior in reducing body mass compared with both liraglutide and placebo at week 26. Safety and tolerability of both medications were similar. The authors assumed that use of oral semaglutide could potentially lead to earlier initiation of GLP-1 receptor agonist therapy in diabetic patients.³⁹ Dungan et al analyzed data from PIONEER trials and found out that the proportion of patients achieving clinically relevant reductions in both HbA1c and body mass were significantly greater in the group treated with orally administered semaglutide versus comparators (liraglutide, empagliflozin, sitagliptin).⁴⁰

Chubb et al conducted a meta-analysis, which proved that once-daily oral semaglutide 14 mg, as an add-on to basal insulin, was an effective pharmacotherapy for achieving glycemic control, HbA1c and weight reduction at 26 ± 4 weeks. The authors pointed out that once-daily oral semaglutide 14 mg was as good or even better drug in the context of efficacy and safety versus most injectable GLP-1 analogues.⁴¹

In turn, Wright et al analyzed the data from PIONEER 4 and PIONEER 8. Oral semaglutide and subcutaneous liraglutide were similarly effective in decreasing HbA1c at 26 weeks, but semaglutide presented greater reduction at 52 weeks. In terms of body mass reduction, semaglutide was more effective at 26 and 52 weeks. The tolerability profile of oral semaglutide was similar to that seen for injectable GLP-1 analogues. GI AEs were detected most frequently, were transient and usually occurred with dose escalation.⁴²

Li et al performed a meta-analysis, which showed that compared with active comparators, oral semaglutide significantly decreased HbA1c level, body mass, and increased the number of patients who achieved HbA1c < 7.0%. Compared with placebo or active comparators, oral semaglutide did not increase the incidence of all AEs, but did increase the incidence of GI AEs (nausea, diarrhea and vomiting).⁴³

Interesting findings come from a meta-analysis conducted by Zhang et al who showed that compared with placebo or control drugs (empagliflozin, sitagliptin, liraglutide, and dulaglutide), oral dose of semaglutide significantly decreased HbA1c level. This GLP-1 analogue also proved to be more effective than placebo or positive control drugs in reducing body weight, HbA1c level, fasting plasma glucose level, and BMI with overall dose-dependent efficacy. The incidence of GI AEs caused by oral semaglutide was higher than that of the placebo or positive control drugs. However, there was no case of hypoglycemia during this treatment.⁴⁴

Semaglutide in Obese or Overweight Patients without Type 2 Diabetes

Xie et al performed a systematic review of studies conducted in populations with OAO, but without T2DM, and indicated that semaglutide 2.4 mg had a significant advantage in decreasing body mass and HbA1c level, but the incidence of total AEs was also the highest and could cause hypoglycemia.³⁶ Moiz et al also conducted a meta-analysis of 26 RCTs comprising over 15 thousands patients with the excess of body mass and without T2DM. They concluded that semaglutide was more efficacious for weight loss than liraglutide, with reported safety concerns predominantly GI.⁴⁵

Factors Influencing HbA1c and BMI Reduction

Randomized controlled trials (RCTs) do not always provide sufficient data to clarify the issue of potential predictors of ultimate weight loss or HbA1c levels during GLP-1RA therapy.⁴⁶ This study indicates several confounding factors influencing HbA1c reduction: baseline HbA1c, gender, and age, and also reports that ultimate BMI reduction was significantly associated with baseline HbA1c, baseline BMI and type of GLP-1 analogue.

Factors Influencing HbA1c Reduction

Baseline HbA1c

A study by Mazzei et al analyzing patient-specific factors during GLP-1 agonist treatment revealed that higher baseline HbA1c levels were associated with greater reductions in HbA1c, indicating that patients with poorer initial glycemic control may experience more significant improvements. Additionally, female participants demonstrated a more pronounced HbA1c reduction compared to male participants (3.23% vs 1.35%, respectively; P = 0.001), suggesting a potential gender-related difference in treatment response.⁴⁷

Interestingly, in the meta-analysis by Esposito et al comparing liraglutide and exenatide, the best predictor of achieving the target HbA1c level was the baseline HbA1c.⁴⁸

Gender

As far as the gender specificity of GLP-1 RA action is concerned, the data collected so far do not provide a clear answer. Regarding hypoglycemic efficacy, most studies did not find sex differences.^49–55 One study reported a male superiority.⁵⁶

Age

An audit conducted by the Association of British Clinical Diabetologists assessed the impact of age on response to injectable semaglutide. The study revealed that HbA1c reductions were consistent across all age groups, indicating that age may not significantly influence the glycemic response to semaglutide treatment.⁵⁷

In summary, results presented so far highlight the complex interplay of factors influencing HbA1c reduction during GLP-1 receptor agonist therapy. Further research is warranted to elucidate these relationships and optimize individualized treatment strategies.

Factors Influencing BMI Reduction

Baseline HbA1c

Research indicates that baseline HbA1c levels may not significantly predict weight loss outcomes with GLP-1RA. A study conducted by Mazzei et al who analyzed patient-specific factors in GLP-1RA therapy, concluded that there was no significant correlation between baseline HbA1c and weight reduction, suggesting that glycemic control at treatment initiation does not substantially influence BMI reduction.⁴⁷

Baseline BMI

Higher baseline BMI has been associated with greater weight loss during GLP-1 analogue treatment. A study conducted by Gasoyan et al showed that individuals with higher starting BMIs experienced more significant weight reduction when treated with semaglutide or liraglutide.⁵⁸

Type of GLP-1 Analogue

The specific GLP-1 analogue used also influences weight loss outcomes. A study comparing semaglutide and liraglutide found that semaglutide led to greater average weight loss (5.1% of body weight) compared to liraglutide (2.2%). Higher dosages and longer treatment durations further enhanced weight loss results.⁵⁹

Gender

When considering weight loss, most studies reported a female superiority.^{51,54,56,59,60} Interestingly, one study with liraglutide concluded that in addition to being a woman, a higher baseline BMI was associated with greater body mass reduction.⁶¹ In turn, Milani et al reported that male participants treated with liraglutide achieved greater weight loss and BMI reduction.⁴⁹

Other Factors

It is important to note that the clinical effectiveness of semaglutide, particularly in terms of sustained glycemic control and weight reduction, generally requires long-term, continuous use. As demonstrated in previous trials, including SUSTAIN 6 and PIONEER 6, consistent adherence over extended periods is essential for optimal therapeutic outcomes.⁶² As it was underlined, GI AEs, although often transient and mild to moderate in severity, may affect treatment persistence.⁶³ Finally, pharmacological therapy alone is insufficient; lifestyle modifications, including diet and physical activity, remain crucial components for maintaining long-term metabolic improvements. These factors should be considered when interpreting our results and when making individualized treatment decisions in clinical practice.

Mechanisms Underlying Superior Weight and Fat Reduction with Semaglutide

As emphasized, in the presented study semaglutide demonstrated superior efficacy in reducing BMI and WC—both key markers of metabolic health and cardiovascular risk. These findings warrant a deeper exploration of the underlying mechanisms and broader implications of GLP-1RA therapy beyond the numerical results. The greater weight and WC reduction observed with semaglutide may be attributable to its distinct pharmacokinetic and pharmacodynamic properties. Semaglutide, a once-weekly GLP-1 analogue with high albumin binding affinity, offers prolonged receptor activation and sustained incretin effect compared to liraglutide, which is administered once daily. This extended half-life results in more consistent appetite suppression, delayed gastric emptying, and enhanced satiety signaling through hypothalamic pathways, particularly in the arcuate nucleus and area postrema—regions implicated in appetite and nausea regulation.^64,65 Furthermore, evidence suggests that semaglutide has a stronger influence on reward-related eating behavior by modulating dopaminergic pathways in the brain, thereby reducing cravings for high-fat, high-sugar foods.⁶⁶ These central effects on food intake, coupled with peripheral actions on gastrointestinal motility and insulin sensitivity, likely explain its superior efficacy in promoting weight loss and reducing visceral adiposity.

Clinical Implications: Central Obesity, Insulin Resistance, and Cardiovascular Risk

The addition of WC analysis in our study offers valuable clinical insight. Central obesity is a more robust predictor of cardiometabolic risk than BMI alone, as it closely reflects visceral fat burden—a driver of systemic inflammation, insulin resistance, and atherogenesis.¹⁹ The significant reduction in WC with semaglutide compared to liraglutide suggests greater efficacy in mitigating these pathophysiological processes. This is particularly relevant in the context of T2DM, where normal-weight individuals may still exhibit high visceral fat levels and metabolic dysfunction (Metabolically Obese, Normal-Weight, the “MONW” phenotype).

While we did not measure insulin levels or resistance markers (eg, HOMA-IR), previous trials have demonstrated that semaglutide improves insulin sensitivity independently of weight loss, possibly through reductions in hepatic steatosis and low-grade inflammation.⁶⁷ Therefore, the greater impact of semaglutide on waist circumference may also reflect its ability to target hepatic and visceral fat depots more effectively than liraglutide.

Cardiovascular Protection: Beyond Glycemia and Weight

Both semaglutide and liraglutide have shown cardiovascular benefits in large-scale trials; however, emerging evidence suggests that these benefits may not solely derive from glucose lowering or weight reduction. GLP-1RAs exert pleiotropic effects, including anti-inflammatory, anti-atherosclerotic, and endothelial-protective properties. For instance, semaglutide has been shown to reduce pro-inflammatory cytokines and improve vascular function through nitric oxide–dependent pathways.⁶⁸ These effects may be particularly beneficial in obese patients with T2DM who are at elevated cardiovascular risk.

In the presented study, the absence of direct cardiovascular biomarkers (eg, C-reactive protein, CRP, N-terminal pro-B-type natriuretic peptide, NT-pro-BNP) is a limitation. However, the significant anthropometric improvements observed—especially in central fat—suggest that semaglutide may confer a more favorable cardiometabolic profile. This is consistent with findings from the SUSTAIN-6 and PIONEER trials, which reported reductions in major adverse cardiovascular events (MACE) with semaglutide use.²⁵

Safety and Tolerability

As observed in previous studies, GI AEs were the most common side effects in both groups, with a higher incidence in the semaglutide arm. While the frequency of vomiting was significantly greater with semaglutide, most events were transient and occurred during dose escalation. Interestingly, previous analyses have shown that GI side effects may only minimally contribute to weight loss, suggesting that the observed benefits are driven more by pharmacologic rather than AEs.²⁹

The exact mechanisms behind GI AEs caused by GLP-1 RA remain unclear. While delayed gastric emptying may contribute to nausea, it also occurs during fasting and is not consistently linked to gastric emptying rates.^69,70 Recent evidence suggests a central nervous system component, as a modified exenatide with reduced brain penetration caused less vomiting in animal models while preserving glucose-lowering effects. This supports the theory that central pathways, rather than just GI effects, may play a key role in mediating these adverse symptoms associated with GLP-1 analogue therapy.⁷¹

Study Limitations and Strengths

Patients’ lifestyle factors, including physical activity and dietary habits, may have influenced the outcomes and were not systematically controlled.

The study population consisted exclusively of Caucasian individuals, limiting the generalizability of the results to more diverse ethnic groups.

The observational period was relatively short and longer-term data are necessary to fully assess the durability of the observed benefits and long-term safety of both GLP-1 analogues.

An important limitation also concerns potential selection bias. Therapeutic decisions were made collaboratively between patients and their attending physicians, and patients were free to choose the treatment after detailed discussions. The once-weekly administration of subcutaneous semaglutide may have been a more attractive option for many patients compared to daily liraglutide injections, potentially influencing group allocation and introducing a self-selection effect that could affect comparative outcomes. Future research—including discrete choice experiments or structured patient-reported outcome measures—should more explicitly evaluate how individual preferences and expectations influence treatment selection, adherence, and long-term outcomes.

Moreover, insulin sensitivity indices, frequency of hypoglycemia, and formal quality of life assessments (eg, validated PRO instruments) were not incorporated into the study protocol due to resource and logistic constraints. Future prospective studies should adopt a more comprehensive methodological approach, including measures of insulin resistance (eg, HOMA-IR), hypoglycemia monitoring, and validated quality-of-life questionnaires, to better reflect the full impact of GLP-1 analogue therapy on patient outcomes.

The authors are also aware that including direct quantitation of insulin levels would have further strengthened the mechanistic interpretation of the findings, particularly in evaluating the differential insulinotropic effects of semaglutide and liraglutide. However, as the primary aim of the study was to assess real-world clinical effectiveness and safety in a routine healthcare setting, insulin measurements were not included in the study protocol to maintain feasibility and patient compliance across multiple outpatient sites. Future prospective studies should include insulin and C-peptide measurements to better elucidate the physiological differences between these GLP-1 RAs and their role in glycemic regulation.

The presented study included adult patients aged 18 to 80 y., but the age group of 18–25 y. represented only a small proportion of the cohort. For this reason, the authors did not incorporate specific biomarker assessments commonly used in adolescent-focused trials, such as lipid profiles, CRP, or other cardiometabolic indicators beyond HbA1c and BMI. The authors are aware that studies, especially those including younger adults or adolescents, should incorporate standardized cardiometabolic biomarker profiling to align with emerging evidence and age-specific response patterns.

Semaglutide can modulate both adipose and visceral fat. Adiponectin and omentin −1 are important biomarkers reflecting changes in adipose and visceral fat distribution, and that their inclusion could have provided a more comprehensive understanding of the mechanisms underlying semaglutide’s metabolic effects. However, due to logistical and resource constraints, the authors did not include specialized adipokine measurements such as adiponectin or omentin-1.

The strength of this study lies in its provision of real-world, comprehensive data on the efficacy and tolerability of semaglutide and liraglutide in a homogeneous Polish population with T2DM and obesity. Unlike controlled clinical trials, our findings reflect routine clinical practice, including patient preferences and adherence patterns, offering valuable insights for everyday treatment decision-making.

The obtained results confirm the findings of RCTs, which showed a better effect of semaglutide on weight loss and HbA1c level in real-world conditions. Our findings support the use of semaglutide as a first-line GLP-1RA for patients with obesity and T2DM, particularly when reduction in visceral fat and cardiovascular risk is prioritized.

Expectations

Based on the studies to date, it is reasonable to assume, that new projects should be devoted to demonstrating long-term efficacy and safety of GLP-1 analogues. New potential indications for GLP-1 agonist should be also investigated, such as insulin resistance, prediabetes and sarcopenia.^72,73 Future studies should aim to incorporate insulin resistance indices, detailed body composition analyses, eg, Dual-Energy X-ray Absorptiometry (DXA) or Magnetic Resonance Imaging (MRI), and cardiovascular biomarkers to further delineate the mechanistic pathways involved. Additionally, evaluating patient-reported outcomes and quality of life would provide a more holistic assessment of therapeutic impact.

Currently available results of subgroup analyses of single GLP-1 analogues versus dual GLP-1/GIP agonists (tirzepatide, mazdutide, and CagriSema) or triple agonists (retatrutide) suggest that the latter two medications more effectively reduce body mass, HbA1c level and glycemia.⁷⁴ New, multi-directional therapies are the future of effective and safe treatment of T2DM and obesity.

Conclusion

Our study demonstrated that in patients with T2DM and obesity, semaglutide had a significant effect on weight loss and resulted in a more pronounced reduction in HbA1c compared with liraglutide. The most common AEs were GI, although they were generally mild and self-limiting. These results support the efficacy of GLP-1 RAs in a real-world setting, emphasizing the need for individualized treatment choices based on efficacy, tolerability, and patient preferences. Increased awareness among physicians and patients of achievable outcomes and safety issues is essential to optimize the use of GLP-1 analogues.

Disclosure

The authors report no conflicts of interest in this work.