Cardiometabolic Effects of Semaglutide in Individuals with Type 2 Diabetes in the United Arab Emirates: Real-World 12-Month Outcomes and Predictors of Response

Abstract

Introduction

There is little real-world evidence on semaglutide effectiveness in the United Arab Emirates (UAE), despite the high local burden of type 2 diabetes (T2DM) and obesity. The aim of this study was to evaluate real-world cardiometabolic outcomes and predictors of response following initiation of semaglutide in a tertiary endocrine clinic.

Methods

This was a retrospective, observational cohort study of adults with T2DM initiating once-weekly subcutaneous semaglutide (June 2022 to January 2025). Primary outcomes were changes in glycated hemoglobin (HbA1c), body weight, and systolic blood pressure (SBP) at 6 and 12 months. Responder and composite endpoints were assessed at 12 months. Predictors of glycemic (HbA1c reduction ≥ 1%) and composite response (HbA1c reduction ≥ 1% and ≥ 5% weight loss) were analyzed by multivariable logistic regression.

Results

In total, 278 patients were included. Adjusted HbA1c decreased by −0.89% (95% confidence interval (CI) −1.07 to −0.70) at 6 months and −0.71% (95% CI −0.91 to −0.50) at 12 months (both p < 0.001). Weight decreased by −3.09 kg (−4.19 to −2.00) and −4.77 kg (−6.00 to −3.53) at 6 and 12 months, respectively (p < 0.001). SBP decreased by −3.56 mmHg (−6.17 to −0.96; p = 0.007) at 6 months. Overall, 39% achieved an HbA1c reduction ≥ 1%, 43% achieved ≥ 5% weight loss, and 23% achieved the composite endpoint at 12 months. Higher baseline HbA1c predicted HbA1c response (odds ratio [OR] 1.85; 95% CI 1.26 to 2.88).

Conclusion

Semaglutide improved glycemia and weight and resulted in modest reductions in SBP in real-world practice. These findings support semaglutide as an effective component of cardiometabolic risk reduction strategies in adults with T2DM in the region, especially in those with at-risk glycemic profiles.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13300-026-01888-y.

Key Summary Points

Why carry out this study?

Type 2 diabetes and obesity affect are highly prevalent in the United Arab Emirates (UAE) population, but there is almost no real-world data on semaglutide from this region.

UAE patients differ from Western trial populations in comorbidity burden, disease severity, and anthropometric profile, limiting the generalizability of existing evidence.

This study evaluated 12-month cardiometabolic outcomes and predictors of response to semaglutide in routine clinical practice in the UAE.

What was learned from the study?

Semaglutide significantly reduced glycated hemoglobin (HbA1c; −0.71% at 12 months), body weight (−4.77 kg), and systolic blood pressure (−3.56 mmHg at 6 months); higher baseline HbA1c was the only consistent predictor of response.

These findings confirm semaglutide’s cardiometabolic effectiveness in a Middle East and North Africa (MENA) real-world setting and support prioritizing treatment in patients with poorer baseline glycemic control.

Introduction

Type 2 diabetes mellitus (T2DM) and overweight and obesity (OAO) are major and persistent global public health challenges [1, 2]. This is especially true in the Middle East and North Africa (MENA) region [3], with obesity (body mass index [BMI] ≥ 30 kg/m²) and overweight (BMI ≥ 25 kg/m²) affecting one in four and a third of adults, respectively [3], as well as T2DM affecting 15% of the population [4]. Both diseases are especially prevalent in the United Arab Emirates (UAE); OAO affects about two-thirds [5] and T2DM about 31% (95% confidence interval (CI) 25% to 37%) of the population [4]. The two diseases are highly interrelated, both in terms of their epidemiology and pathophysiology, as OAO is the primary risk factor for T2DM, and their synergy drives insulin resistance that makes glycemic control more difficult in individuals with comorbid T2DM and obesity [6]. The high prevalence of OAO and T2DM incurs massive personal, healthcare, and societal costs, as both diseases are well-established risk factors for many chronic physical and mental health disorders, especially cardiovascular disease [7, 8]. Adverse outcomes from OAO and T2DM can be mitigated through effective multidisciplinary treatment using lifestyle interventions [9–11], medications [11], surgical options [12, 13], and, more recently, pharmacological interventions such as glucagon-like peptide-1 receptor agonists (GLP-1 RAs) [14–16].

GLP-1 RAs are an important therapy that address both glycemic control and weight, consistent with current guidelines [11]. GLP-1 RAs are not only effective anti-glycemics but they also lead to weight reduction and are increasingly recognized to reduce cardiovascular and renal complications in high-risk populations [17, 18]. Semaglutide, a long-acting once-weekly GLP-1 RA, reduces HbA1c, causes weight loss, and reduces major cardiovascular events (MACE) in patients with T2DM at high cardiovascular risk (by 14% in meta-analyses of randomized controlled trials [RCTs] [15]). These agents are also showing impressive efficacy for individuals with other obesity-related complications, such as established cardiovascular disease (CVD) and/or chronic kidney disease [19–21]. The drug is generally well tolerated (the most common side-effects are nausea, vomiting, and constipation), and the once-weekly administration by injection improves adherence [22].

Although individual metabolic outcomes (such as HbA1c) are useful metrics for assessing treatment efficacy, composite endpoints are increasingly recognized as desirable for capturing overall combined cardiometabolic benefit and are emphasized in American Diabetes Association (ADA)/European Association for the Study of Diabetes (EASD) consensus guidelines [11]. Composite endpoints are particularly relevant in the UAE, where comorbidity is common even in early adulthood and where therapies that benefit multiple systems might provide disproportionate long-term value.

Even though semaglutide is safe and effective, responses are variable, and the generalizability of trial results to specific populations remains uncertain. Furthermore, randomized trial populations differ from real-world populations in the Middle East, with their unique anthropometric and genetic characteristics, more severe and earlier cardiometabolic disease, higher background prevalence of insulin use, and more heterogeneous adherence patterns [23]. Therefore, real-world effectiveness studies are essential for understanding how semaglutide performs in clinical practice, particularly in regions with unique epidemiological and clinical characteristics and that are often underrepresented in trials. European and North American studies suggest that semaglutide is effective outside trial settings, though typically with smaller effect sizes owing to real-world complexity [24, 25]. However, regional data from the Middle East are extremely limited, and no study from the UAE has comprehensively assessed longitudinal anthropometric, glycemic, and cardiometabolic outcomes in patients taking semaglutide.

To address this gap, here we evaluate the cardiometabolic efficacy of semaglutide in a cohort of patients attending a tertiary clinic in the UAE. Specifically, we assessed (i) temporal changes in HbA1c, body weight, systolic blood pressure (SBP), and lipid profiles over 12 months; (ii) proportions achieving key response and composite endpoints; and (iii) predictors of glycemic and composite response to inform clinical decision-making and guideline implementation in the UAE and wider region.

Methods

Study Design

This was a retrospective, observational, cohort study conducted at the Endocrine and Diabetes Clinic, Al Qassimi Hospital, Sharjah, UAE, a tertiary referral center within the Emirates Health Services network. The study adhered to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement. The Emirates Health Services Research Ethics Committee approved the protocol (REC number: MOHAP/DXB-REC/D.D.A/No.208/2024). Written informed consent for the use of de-identified clinical data was obtained where required by institutional policy.

Study Population and Inclusion and Exclusion Criteria

Eligible participants were: (i) adults (≥ 18 years), (ii) with a confirmed diagnosis of T2DM according to ADA 2023 criteria (fasting plasma glucose ≥ 126 mg/dL, HbA1c ≥ 6.5%, or 2-h plasma glucose ≥ 200 mg/dL during an oral glucose tolerance test [11]), and (iii) who had initiated subcutaneous once-weekly semaglutide between June 2022 and January 2025. Oral semaglutide was not used in this cohort. Exclusion criteria were: (i) type 1 diabetes mellitus or gestational diabetes, (ii) bariatric surgery during the study period, (iii) semaglutide treatment duration < 6 months without follow-up data, and (iv) advanced comorbid conditions precluding follow-up (e.g., metastatic cancer, terminal organ failure). All patients followed the same standard subcutaneous dose titration schedule: 0.25 mg once weekly for 4 weeks, 0.5 mg once weekly for 4 weeks, and then escalation to 1 mg once weekly, which was maintained throughout follow-up. Background glucose-lowering therapy was not modified during follow-up other than the planned semaglutide dose escalation. Changes to lipid-lowering or antihypertensive therapy during the follow-up period were not systematically recorded.

Study Outcomes

The primary outcomes were changes in HbA1c, body weight, and SBP from baseline to 6 and 12 months following initiation of semaglutide. Secondary outcomes were the proportion of patients achieving clinically relevant responder endpoints at 12 months, defined as an HbA1c reduction ≥ 1%, HbA1c < 7.0%, weight loss ≥ 5% and ≥ 10%, and SBP reduction ≥ 5 mmHg. Two composite outcomes were also evaluated: (i) a dual composite endpoint of HbA1c reduction ≥ 1% and weight loss ≥ 5%, and (ii) a triple composite of HbA1c < 7.0%, weight loss ≥ 5%, and SBP reduction ≥ 5 mmHg. Additional exploratory outcomes included longitudinal changes in lipid parameters (LDL cholesterol, HDL cholesterol, and triglycerides).

Statistical Analysis

All analyses were conducted on patients who initiated semaglutide and had at least baseline data recorded. Continuous variables are presented as mean ± standard deviation (SD), and categorical variables as counts and percentages.

Longitudinal changes in HbA1c, body weight, and SBP from baseline to 6 and 12 months were analyzed as observed means and SDs at each time point. Longitudinal changes in the same variables were also analyzed using linear mixed-effects models with a random intercept for each patient to account for within-individual correlation over time and to incorporate all available observations under a missing-at-random assumption, thereby mitigating bias from incomplete follow-up that would arise from a complete-case analysis. Time was modeled as a categorical fixed effect (at baseline, 6 months, and 12 months). For each outcome, models were adjusted for prespecified, clinically relevant baseline covariates including age, sex, baseline BMI, baseline HbA1c, and insulin use at baseline. Model estimates are reported as mean differences (change from baseline at each time point) with 95% CIs and p-values.

For each binary endpoint, the proportion of patients achieving the response at 12 months was calculated with 95% CIs. To explore predictors of response, separate multivariable logistic regression models were fit for (i) HbA1c reduction ≥ 1% and (ii) the dual composite endpoint. Candidate predictors were selected a priori on the basis of clinical relevance: age, sex, baseline BMI, baseline HbA1c, duration of diabetes, insulin use at baseline, ischemic heart disease, metformin use, sodium-glucose cotransporter-2 inhibitor (SGLT2i) use, sulfonylurea use, baseline estimated glomerular filtration rate (eGFR), low-density lipoprotein (LDL) cholesterol, high-density lipoprotein (HDL) cholesterol, triglycerides, systolic and diastolic blood pressure, and presence of ≥ 2 comorbidities (hypertension, dyslipidemia, obesity (BMI ≥ 30 kg/m²), chronic kidney disease, and ischemic heart disease); hypertension and dyslipidemia were excluded as they were collinear with blood lipid profiles and blood pressure measurements. Variables with > 30% missing data were excluded from modeling. Results are presented as odds ratios (ORs) with 95% CIs and p-values.

Predefined subgroup analyses examined whether HbA1c change at 12 months differed by (i) baseline BMI category (< 35 versus ≥ 35 kg/m²), (ii) baseline HbA1c category (< 8.0%, 8.0–8.9%, ≥ 9.0%), and (iii) insulin use at baseline (yes/no). Unadjusted mean changes and SDs were calculated within each subgroup. Multivariable linear regression models were fit with 12-month HbA1c change as the dependent variable and the subgroup variable of interest as the main exposure, adjusting for age, sex, baseline HbA1c, and baseline BMI; models including insulin use as the subgroup did not additionally adjust for insulin to avoid collinearity.

A p-value < 0.05 was considered significant. Analyses were performed using R v4.5.2 (R Foundation for Statistical Computing, Vienna, Austria), with mixed-effects models fit using the lme4 and lmerTest packages and descriptive and regression tables generated using gtsummary. Data were visualized using ggplot2.

Results

Study Population

In total, 278 patients initiated semaglutide and were included in the analysis. Baseline demographic and clinical characteristics of the study population are shown in Table 1. The mean age of the baseline study population was 54 ± 11 years, and 65% were female. Most patients were of UAE nationality (172/278, 62%) but, as expected for the unique and diverse demographics of the UAE [26], there was wider representation from the World Health Organization (WHO) Eastern Mediterranean Region (89%), with smaller proportions from Africa (6%), South-East Asia (5%), and the Americas (< 1%). The mean duration of T2DM was 6.4 ± 6.1 years, baseline BMI was high at 36 ± 7 kg/m², and mean HbA1c was 7.99 ± 1.95%.

Table 1.

Baseline demographic and clinical characteristics of the study population

Characteristic	Total (N = 278)
Age (years), mean (SD)	53.6 (11.2)
Sex, n/N (%)
Female	181/278 (65%)
Male	97/278 (35%)
WHO region, n/N (%)
Africa	16/278 (5.8%)
Americas	1/278 (0.4%)
Eastern Mediterranean	248/278 (89%)
Southeast Asia	13/278 (4.7%)
Duration of diabetes (years), mean (SD)	6.4 (6.1)
Missing	60
Duration of diabetes (grouped), n/N (%)
< 5	105/218 (48%)
> 10	53/218 (24%)
5–10	60/218 (28%)
BMI at baseline (kg/m2), mean (SD)	35.8 (6.9)
Missing	3
HbA1c at baseline (%), mean (SD)	7.99 (1.95)
Missing	4
Hypertension, n/N (%)	140/253 (55%)
Dyslipidemia, n/N (%)	186/269 (69%)
Ischemic heart disease, n/N (%)	42/238 (18%)
Insulin use at baseline, n/N (%)	68/242 (28%)
Metformin at baseline, n/N (%)	227/263 (86%)
SGLT2 inhibitor at baseline, n/N (%)	182/258 (71%)
Sulfonylurea at baseline, n/N (%)	39/237 (16%)
eGFR at baseline (mL/min/1.73m2), mean (SD)	91 (24)
LDL cholesterol at baseline (mmol/L), mean (SD)	2.65 (1.10)
HDL cholesterol at baseline (mmol/L), mean (SD)	1.40 (0.34)
Triglycerides at baseline (mmol/L), mean (SD)	1.71 (1.42)
Systolic BP at baseline (mmHg), mean (SD)	132 (17)
Diastolic BP at baseline (mmHg), mean (SD)	74 (12)
≥ 2 comorbidities, n/N (%)
< 2	146/271 (54%)
≥ 2	125/271 (46%)

BMI, body mass index; BP, blood pressure; eGFR, estimated glomerular filtration rate; HbA1c, glycated hemoglobin; HDL, high-density lipoprotein; LDL, low-density lipoprotein; SD, standard deviation; SGLT2, sodium-glucose cotransporter-2; WHO, World Health Organization

Hypertension, dyslipidemia, and ischemic heart disease were common and present in 55%, 69%, and 18% of participants, respectively. At baseline, 86% of patients were using metformin, 71% SGLT2i, 16% sulfonylureas, and 28% insulin.

Glycemic Outcomes

HbA1c measurements were available for 274 patients at baseline, 216 (79%) at 6 months, and 157 (57%) at 12 months. Mixed-effects modeling was therefore used to incorporate all available data and to minimize bias from incomplete follow-up. The observed mean HbA1c improved from 7.99 ± 1.95% at baseline (n = 274) to 7.18 ± 1.72% at 6 months (n = 216) and 7.22 ± 1.82% at 12 months (n = 157) (Fig. 1A).

Fig. 1.

Cardiometabolic outcomes at baseline, 6 months, and 12 months following initiation of semaglutide. Mean values (± standard error) are shown for: A HbA1c, B body weight, C systolic blood pressure, D LDL cholesterol, E HDL cholesterol, and F triglycerides. All outcomes represent unadjusted longitudinal summaries. HbA1c, glycated hemoglobin; HDL, high-density lipoprotein; LDL, low-density lipoprotein

In the adjusted linear mixed model (Supplementary Table 1), semaglutide therapy was associated with significant decreases in HbA1c at both 6 months (mean difference −0.89%; 95% CI −1.07 to −0.70; p < 0.001) and 12 months (−0.71%; 95% CI −0.91 to −0.50; p < 0.001) compared with baseline. Baseline HbA1c was a strong independent predictor of the magnitude of improvement (p < 0.001).

Weight Outcomes

There was a decrease in weight over the follow-up period from 94.2 ± 20.6 kg at baseline (n = 275) to 90.0 ± 19.5 kg at 6 months (n = 244) and 89.5 ± 20.1 kg at 12 months (n = 171) (Fig. 1B). The mixed-effects model estimated a 3.09 kg reduction at 6 months (95% CI −4.19 to −2.00; p < 0.001) and 4.77 kg at 12 months (95% CI −6.00 to −3.53; p < 0.001) (Supplementary Table 2). Male sex and higher BMI were associated with higher absolute weight but not with differential response. HbA1c and insulin use were not associated with weight changes over time.

Blood Pressure Outcomes

Mean systolic blood pressure decreased from 131.7 ± 17.3 mmHg at baseline (n = 274) to 128.5 ± 17.4 mmHg at 6 months (n = 243) and 129.6 ± 17.6 mmHg at 12 months (n = 171) (Fig. 1C). In the mixed-effects model (Supplementary Table 3), SBP reduced by 3.56 mmHg at 6 months (95% CI −6.17 to −0.96; p = 0.007), with a nonsignificant reduction of 2.06 mmHg at 12 months (95% CI −4.97 to 0.84; p = 0.16). Older age, male sex, and higher BMI were independently associated with higher systolic pressure over time.

Lipid Outcomes

In exploratory analyses (Fig. 1D–F), mean LDL cholesterol decreased from 2.65 mmol/L at baseline to 2.37 mmol/L at 6 months and 2.48 mmol/L at 12 months; HDL cholesterol increased from 1.40 mmol/L at baseline to 1.45 mmol/L at 12 months (not significant); and triglycerides showed no significant change over follow-up. In adjusted mixed-effects models (Supplementary Table 4), LDL was significantly lower at 6 months (−0.27 mmol/L; 95% CI −0.42 to −0.11; p < 0.001) and remained so at 12 months (−0.17 mmol/L; p = 0.050). Older age and male sex were associated with lower LDL levels, whereas higher baseline BMI and higher baseline HbA1c were associated with higher LDL levels; insulin use at baseline was also associated with slightly lower LDL. Male sex and higher baseline HbA1c were associated with lower HDL levels, while higher baseline HbA1c was associated with higher triglyceride levels.

Responder Outcomes

At 12 months, 39% of patients achieved a ≥ 1% reduction in HbA1c (61/156; 95% CI 31.4% to 47.2%), and 55% achieved HbA1c < 7% (87/157; 95% CI 47.3% to 63.3%). Overall, 43% achieved a ≥ 5% weight loss (74/171; 95% CI 35.7% to 51.1%) and 20% achieved a ≥ 10% loss (35/171; 95% CI 14.7% to 27.3%). A ≥ 5 mmHg reduction in systolic BP was observed in 45% (77/171; 95% CI 37.4% to 52.8%) of cases (Table 2).

Table 2.

Responder and composite outcomes at 12 months

Outcome	N = 278
HbA1c reduction ≥ 1%
Yes	61/156 (39%)
HbA1c < 7.0%
Yes	87/157 (55%)
Weight loss ≥ 5%
Yes	74/171 (43%)
Weight loss ≥ 10%
Yes	35/171 (20%)
SBP reduction ≥ 5 mmHg
Yes	77/171 (45%)
Dual composite: HbA1c ≥ 1% reduction +  ≥ 5% weight loss
Yes	35/151 (23%)
Triple composite: HbA1c < 7%, ≥ 5% weight loss, ≥ 5 mmHg SBP reduction
Yes	17/150 (11%)

Data are presented as n/N (%)

HbA1c, glycated haemoglobin; SBP, systolic blood pressure

The dual composite endpoint (≥ 1% HbA1c reduction plus ≥ 5% weight loss) was achieved by 23% (35/151; 95% CI 16.7% to 30.7%) of participants, while 11% (17/150; 95% CI 6.7% to 17.5%) met the stricter triple composite endpoint (HbA1c < 7%, ≥ 5% weight loss, ≥ 5 mmHg SBP reduction) (Table 2).

Predictors of Response

In logistic regression models (Table 3), baseline HbA1c predicted a ≥ 1% HbA1c reduction (OR 1.85; 95% CI 1.26 to 2.88; p = 0.003). No other demographic or cardiometabolic factors were reliably associated with glycemic response. Sulfonylurea use was associated with both HbA1c and composite response but with wide confidence intervals. There were no strong or clinically interpretable predictors of a composite response.

Table 3.

Multivariable predictors of glycemic and composite response at 12 months

Characteristic	HbA1c response ≥ 1%			Dual composite response
	OR	95% CI	p-Value	OR	95% CI	p-Value
Age (years)	0.99	0.92, 1.1	0.869	0.91	0.81, 1.0	0.086
Sex
F	–	–		–	–
M	1.0	0.33, 3.3	0.943	1.5	0.37, 6.6	0.560
Baseline BMI (kg/m2)	0.91	0.82, 1.0	0.053	0.96	0.86, 1.1	0.432
Baseline HbA1c (%)	1.8	1.3, 2.9	0.003	1.1	0.73, 1.7	0.633
Duration of diabetes (years)	0.91	0.78, 1.0	0.178	0.97	0.81, 1.2	0.718
Insulin at baseline
No	–	–		–	–
Yes	0.46	0.11, 1.7	0.263	0.44	0.07, 2.3	0.348
Ischemic heart disease
No	–	–		–	–
Yes	0.41	0.07, 2.1	0.289	0.86	0.09, 6.9	0.892
Metformin
No	–	–		–	–
Yes	1.3	0.35, 5.5	0.684	2.4	0.42, 22	0.361
SGLT2 inhibitor
No	–	–		–	–
Yes	0.64	0.19, 2.0	0.443	0.68	0.16, 2.7	0.579
Sulfonylurea
No	–	–		–	–
Yes	4.9	1.1, 25	0.041	5.5	1.0, 32	0.047
eGFR (mL/min/1.73m2)	0.99	0.96, 1.0	0.501	0.98	0.95, 1.0	0.184
LDL cholesterol (mmol/L)	1.4	0.86, 2.4	0.176	1.3	0.65, 2.5	0.459
HDL cholesterol (mmol/L)	0.13	0.01, 0.96	0.055	0.02	0.00, 0.27	0.007
Triglycerides (mmol/L)	0.45	0.20, 0.93	0.043	0.45	0.17, 1.0	0.082
Systolic BP (mmHg)	1.0	0.96, 1.0	0.998	1.0	0.99, 1.1	0.132
Diastolic BP (mmHg)	1.0	0.96, 1.1	0.571	0.96	0.89, 1.0	0.310
≥ 2 comorbidities
< 2	–	–		–	–
≥ 2	1.5	0.46, 5.1	0.517	2.5	0.60, 12	0.218

BMI, body mass index; BP, blood pressure; CI, confidence interval; eGFR, estimated glomerular filtration rate; HbA1c, glycated hemoglobin; HDL, high-density lipoprotein; LDL, low-density lipoprotein; OR, odds ratio; SGLT2, sodium-glucose cotransporter-2

Subgroup Analyses

Reductions in HbA1c at 12 months were similar in individuals with BMI < 35 kg/m² (−0.76 ± 1.50%) and ≥ 35 kg/m² (−0.69 ± 1.36%). Adjusted analyses confirmed no significant effect of BMI category on HbA1c change (p = 0.98) (Supplementary Table 5).

HbA1c reductions were higher in patients with higher baseline HbA1c. HbA1c decreased by −1.51 ± 1.91% in those with HbA1c ≥ 9%, compared with −0.45 ± 1.11% in those with HbA1c < 8%. This pattern persisted in adjusted models (p < 0.001), consistent with a baseline-dependent glycemic response in which absolute HbA1c reduction is greater in patients with higher baseline glycemia (Supplementary Table 5).

HbA1c reductions were similar in insulin users (−0.88 ± 1.28%) and non-users (−0.62 ± 1.53%). In adjusted analyses, insulin use was not associated with differential HbA1c change (p = 0.33) (Supplementary Table 5).

Adverse Events

Adverse events were not systematically recorded in the clinical records, precluding quantitative analysis of incidence rates. However, consistent with the established safety profile of semaglutide, the most frequently documented side-effects were transient gastrointestinal symptoms such as nausea, vomiting, and dyspepsia. No clearly documented unexpected safety concerns were identified in the available clinical notes.

Discussion

In this large cohort of adults with T2DM from the UAE, weekly semaglutide use was associated with improvements in glycemic control, body weight, and systolic blood pressure over 12 months. These findings are consistent with existing evidence but provide important new information on the real-world efficacy of semaglutide in MENA populations, who carry a high burden of obesity, T2DM, and cardiometabolic disease.

The magnitude of observed glycemic and weight improvements (mean reductions in HbA1c of ~0.9% at 6 months and ~0.7% at 12 months, and a mean weight loss approaching 5 kg at 12 months) is consistent with real-world studies from Europe and North America and findings from the SUSTAIN clinical trials. Trials of semaglutide 0.5–1.0 mg have demonstrated HbA1c reductions of 1.0–1.8% and 3–6 kg weight loss over 30–56 weeks, depending on baseline glycemia, BMI, and background therapy [17, 27, 28]. The smaller observed effect sizes and modest responses observed here probably reflect typical real-world practice including variable dosing and adherence differences, patterns reported in other healthcare systems [24, 25] and known to impact the degree of efficacy of drug therapy observed in real-world settings compared with clinical trials [29]. Similarly, a recent systematic review of real-world evidence demonstrated wide but clinically meaningful improvements in HbA1c (−0.3% to −3.4%) and body weight (−0.6 kg to −8.4 kg) with once-weekly semaglutide, with substantial but variable proportions of patients achieving HbA1c < 7% and ≥ 5% weight loss in routine practice. The observed changes in HbA1c and weight observed here fall well within the reported real-world ranges, supporting the generalizability of semaglutide effectiveness outside clinical trials and extending this evidence to Middle Eastern populations. The observed 3.6 mmHg reduction in systolic blood pressure at 6 months is also consistent with published data, with SBP improvements typically falling within a 2–5 mmHg range and correlating with weight loss [30]. These reductions are important, as even small reductions in SBP can meaningfully reduce cardiovascular risk at the population level [31].

Our findings are also consistent with other real-world evidence from the Gulf region. Ahmed et al. recently reported 6-month outcomes for a small cohort of 73 patients with T2DM treated with weekly subcutaneous semaglutide and recorded a 1.07% decrease in HbA1c and 4.5 kg reduction in weight, similar to the magnitude of effect seen here [32]. In a retrospective cohort study from Saudi Arabia, Alarfaj [33] reported that 42% of patients with obesity and T2DM achieved the same composite endpoint of ≥ 1% HbA1c reduction and ≥ 5% weight loss after 12 months, together with significant improvements in glycemia, body weight, and lipid parameters. Although only 23% of our cohort achieved this composite endpoint, the direction and magnitude of individual glycemic and weight responses were comparable, particularly when accounting for differences in baseline HbA1c, obesity severity, insulin use, and real-world treatment heterogeneity. Of note, Alarfaj [33] also identified higher baseline HbA1c as a strong predictor of glycemic response, supporting a consistent baseline-dependent glycemic response across MENA populations. Clinically, prioritizing semaglutide in patients with poorer baseline glycemic control may result in the largest absolute reductions in HbA1c, although the relative benefit and the broader cardiometabolic effects of semaglutide are not confined to this subgroup. However, we did not detect lower baseline BMI and insulin non-use [33] as response predictors, probably owing to population heterogeneity. Nevertheless, our data and other real-world studies support the use of semaglutide in MENA populations.

Compared with the obesity-focused STEP trial, which assessed semaglutide 2.4 mg in individuals with OAO and demonstrated weight reductions of ~15% at week 68 [34], our observed weight loss was lower, consistent with diabetes-indicated lower-dose therapy. Nevertheless, achieving ≥ 5% weight loss in ~40% and ≥ 10% in ~20% of participants is of real clinical benefit in the UAE, where one in three of the population live with obesity [5], which substantially contributes to the cardiometabolic disease burden [7]. BMI is, however, an imperfect marker of adiposity, and a recent prespecified analysis of the SELECT trial reported that the cardioprotective effects of semaglutide were independent of baseline BMI and weight loss, instead being associated with reductions in waist circumference [35]. This supports waist circumference as a more informative measure of central adiposity in cardiometabolic risk assessment, highlighting the importance of incorporating it into routine clinical evaluation alongside BMI.

Our findings of a significant but small reduction in LDL at 6 months, an upward trend in HDL at 12 months, and no consistent change in triglycerides are consistent with published trials and real-world evidence. For instance, in the SUSTAIN and PIONEER trials, semaglutide produced small reductions in LDL cholesterol, minor increases in HDL cholesterol, and variable changes in triglycerides, often reflecting weight loss rather than a direct pharmacological effect [28, 36]. Although relatively small, these lipid effects can contribute to cardiometabolic improvement when combined with weight loss, blood pressure reduction, and glycemic control. However, the observed LDL reduction should be interpreted with caution given that concurrent changes in lipid-lowering therapy were not systematically captured in our dataset.

This study is strengthened by its real-world nature, reflecting routine practice in the UAE and enhancing generalizability within the region. The use of linear mixed-effects models maximized available data and mitigated bias from incomplete follow-up data. Our use of several, clinically relevant outcomes provided a comprehensive assessment of semaglutide effectiveness. However, the study was limited by the observational design and missing data, particularly at 12 months, which may have introduced bias. Although linear mixed-effects models incorporated all available data under a missing-at-random assumption, residual bias cannot be excluded if loss to follow-up was related to treatment response, tolerability, or treatment discontinuation, and reasons for missing data were not systematically captured in our dataset. Reassuringly, the direction and magnitude of effects were consistent across both observed-mean and adjusted mixed-model analyses. All patients followed the same titration schedule and were maintained on 1 mg once weekly, so although differential dosing does not explain different responses between responders and nonresponders, patient-level adherence data (e.g., missed injections and temporary dose interruptions) were not available. Detailed data on individual statin and antihypertensive agents at baseline were not extracted from the records, although diagnoses of dyslipidemia and hypertension are reported, and these patients received guideline-directed therapy. Although background glucose-lowering therapy was not modified during follow-up, changes to statin or antihypertensive therapy were not systematically captured, and we cannot exclude that newly initiated or intensified lipid-lowering therapy contributed to the observed reduction in LDL cholesterol. Adverse events were not systematically captured in our clinical records, precluding quantitative safety analysis. Waist circumference was not routinely measured during the study period and could not be analyzed; given recent evidence that the cardiovascular benefit of semaglutide tracks more closely with central adiposity than with BMI [35], future prospective evaluations should incorporate waist circumference as a potentially more informative marker of cardiometabolic risk. Finally, although sulfonylurea use was associated with both HbA1c and composite responses, the wide confidence intervals and absence of a physiological rationale suggest that this was likely a chance finding related to small subgroup sizes.

Taken together, the observed changes in glycemic, weight, blood pressure, and lipid outcomes indicate multi-factorial cardiometabolic benefit from semaglutide consistent with the mechanisms and outcomes demonstrated in formal clinical trials. The favorable composite cardiometabolic risk profile is consistent with results from the SUSTAIN-6 cardiovascular outcomes trial, in which once-weekly subcutaneous semaglutide 0.5–1.0 mg reduced major adverse cardiovascular events in patients with T2DM at high cardiovascular risk [17]. This consistency in risk-factor improvements suggests that semaglutide may be cardioprotective in our routine practice, informing local management and policy, and providing primary prevention of adverse cardiovascular outcomes.

Conclusions

Here, we show that once-weekly subcutaneous semaglutide is associated with improvements in HbA1c, weight, and systolic blood pressure over 12 months. Given the dual benefits of semaglutide on weight and glycemic control, semaglutide may play an important role in earlier, more aggressive management of T2DM and obesity, especially in individuals with higher baseline HbA1c where cardiovascular risk is increased. These real-world findings support the use of semaglutide as an effective component of comprehensive cardiometabolic risk reduction strategies in UAE adults with T2DM.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contribution

Conceptualization: Muhammad Hamid Siddique Mian, Kholood Abdulla Hasan Abdulla Janahi, and Sobia Siddiq. Data curation: Afnan Tayeb, Fathima Musfira Mohamad Musaffar, Nur Al Sheykh Khalil, Sarah Hisham Al Toubah, Abdul Moen Abdul Karim, Ahlam Almarzooqi, Fardeen Mohammad Sayfoo, Ahmed Hamadeh, and Mohannad Abdulla Hasan Abdulla Janahi. Formal analysis: Muhammad Hamid Siddique Mian, Kholood Abdulla Hasan Abdulla Janahi, and Sobia Siddiq. Funding acquisition: Sobia Siddiq. Investigation: Muhammad Hamid Siddique Mian, Kholood Abdulla Hasan Abdulla Janahi, and Sobia Siddiq. Methodology: Muhammad Hamid Siddique Mian, Kholood Abdulla Hasan Abdulla Janahi, and Sobia Siddiq. Project administration: Sobia Siddiq. Supervision: Muhammad Hamid Siddique Mian, Kholood Abdulla Hasan Abdulla Janahi, and Sobia Siddiq. Writing—original draft: Muhammad Hamid Siddique Mian, Kholood Abdulla Hasan Abdulla Janahi, Afnan Tayeb, Fathima Musfira Mohamad Musaffar, Nur Al Sheykh Khalil, Sarah Hisham Al Toubah, Abdul Moen Abdul Karim, Ahlam Almarzooqi, Fardeen Mohammad Sayfoo, Ahmed Hamadeh, Mohannad Abdulla Hasan Abdulla Janahi, and Sobia Siddiq. Writing—review and editing: Muhammad Hamid Siddique Mian, Kholood Abdulla Hasan Abdulla Janahi, Afnan Tayeb, Fathima Musfira Mohamad Musaffar, Nur Al Sheykh Khalil, Sarah Hisham Al Toubah, Abdul Moen Abdul Karim, Ahlam Almarzooqi, Fardeen Mohammad Sayfoo, Ahmed Hamadeh, Mohannad Abdulla Hasan Abdulla Janahi, and Sobia Siddiq.

Funding

This study was supported by an unrestricted research grant from Novo Nordisk-UAE, who had no role in the study design, data collection, data analysis, data interpretation or writing of the report. All authors had final responsibility for the decision to submit for publication. The journal’s Rapid Service Fee was funded by Novo Nordisk-UAE.

Data Availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Conflicts of Interest

Sobia Siddiq received an unrestricted research grant from Novo Nordisk-UAE, who had no role in the study design, data collection, data analysis, data interpretation, or writing of the report. Muhammad Hamid Siddique Mian, Kholood Abdulla Hasan Abdulla Janahi, Afnan Tayeb, Fathima Musfira Mohamad Musaffar, Nur Al Sheykh Khalil, Sarah Hisham Al Toubah, Abdul Moen Abdul Karim, Ahlam Almarzooqi, Fardeen Mohammad Sayfoo, Ahmed Hamadeh, and Mohannad Abdulla Hasan Abdulla Janahi have nothing to disclose.

Ethical Approval

The Emirates Health Services Research Ethics Committee approved the protocol (REC number: MOHAP/DXB-REC/D.D.A/No.208/2024). Written informed consent for the use of de-identified clinical data was obtained where required by institutional policy.