Effectiveness of sodium-glucose transporter 2 inhibitors and Semagluti

Introduction

People with diabetes mellitus (DM) and chronic kidney disease (CKD) have increased in proportion to the growing incidence of DM itself, facing high risk of kidney failure, heart failure, and premature mortality.¹ Sarcopenia is age-related skeletal muscle loss coupled with loss of muscle strength and/or decreased physical performance.² Research indicates that the incidence of sarcopenia spans from 7% to 29.3% in type 2 DM (T2DM),³ with a 2-fold increased risk of sarcopenia relative to the general population.⁴ Individuals with CKD are also twice more likely to experience sarcopenia compared to those without CKD,⁵ with the prevalence of sarcopenia ranging from 4% to 49%.⁶ Moreover, sarcopenia has been linked to an increased risk of mortality in patients with T2DM⁷ and the danger of progression to end-stage kidney disease (ESKD) in CKD population.⁵ A rising incidence of fragility fractures is also observed in T2DM.⁸ With the progressive decline of renal function in T2DM, a complication of CKD-mineral bone disorder (CKD-MBD) occurs, resulting in bone fragility.⁹ Therefore, enhancing muscle quality (both muscle mass and function) and lowering the risk of bone fragility is of the utmost significance for the clinical management of T2DM-CKD.

Most people with T2DM and CKD require lifelong anti-hyperglycemia medication treatment, and sarcopenia and osteoporosis are age-related diseases.¹⁰ Thus, the effect of drugs on muscle quality and bone mineral content (BMC) is a crucial factor to take into account. Sodium-glucose transporter 2 inhibitors (SGLT-2Is), as the first-line therapeutic agent, are recommended by major guidelines in T2DM concurrent CKD.^1,11 To achieve personal glycemic goal or gain additional kidney and heart protection, glucagon-like peptide-1 receptor agonists (GLP-1RAs) are also suggested.^12,13 Semaglutide is a long-acting GLP-1RA approved for the subcutaneous administration on a once-weekly basis for the treatment of T2DM or/and obesity.¹⁴ The available evidence indicates that Semaglutide is the best drug in the GLP-1RAs class for weight loss.¹⁵ Weight loss is expected to result in changes of body composition, and it is essential to evaluate the impact of weight loss on muscle mass and BMC in T2DM and CKD. However, current research on body composition has predominantly focused on individuals with T2DM or obesity, with significant heterogeneity observed across studies.^16–19 This differentiation probably arises from the effects of population, comorbidity, and drug-specific property.¹⁸ Therefore, the potential negative impact of SGLT-2Is and GLP-1ARs on muscle mass have raised concerns.^20,21 To date, there is a lack of research investigating the effects of SGLT-2Is and Semaglutide on body composition in the more susceptible population of T2DM patients with CKD. The resolution of this research gap will provide more precise therapeutic guidance for this high-risk population.

Bioelectrical impedance analysis (BIA) is a straightforward and non-invasive method that is widely utilized for the evaluation of body composition.²² BIA uses a low-level electrical current flowing through the whole body to measure the varying impedance of different tissues, such as muscle (low impedance) and fat (high impedance), thereby estimating the mass of body composition.²² This study retrospectively analyzed BIA-measured body composition data to assess the effects of SGLT-2Is and Semaglutide on muscle mass and bone minerals in T2DM patients with CKD.

Methods

Study Design

This investigation was a retrospective cohort study conducted at a single center, aimed at assessing alterations in body composition and metabolic profiles among people with T2DM concurrent CKD using SGLT-2Is or/and Semaglutide. The study received approval from the ethics committee of West China Hospital and adhered to the principles outlined in the Declaration of Helsinki (Ethical Number: 20231196). Given the retrospective design, informed consent was waived. All data were fully de-identified before analysis to ensure patient privacy.

Patients

The research was performed on individuals diagnosed with T2DM-CKD who received SGLT-2Is or/and Semaglutide treatment in West China Hospital of Sichuan University from November 2021 to June 2022. The diagnostic criteria for T2DM were based on the 2018 American Diabetes Association (ADA) diagnostic criteria.²³ The definition of CKD was estimated glomerular filtration rate (eGFR) <60 mL/min/1.73 m2 or albumin to creatinine ratio (ACR) ≥30 mg/g for lasting 3 months.

Individuals who underwent body composition measurements via BIA both prior to and following 6 months treatment were considered for inclusion. We excluded people if they were (1) undergoing dialysis; (2) with malignant tumors, hepatic cirrhosis, or severe infections; (3) in pregnancy; (4) treated with glucocorticoids; (5) without follow-up data over 6 months.

Data Collection

Patient information was obtained from the hospital information system and supplemented by telephone follow-up, involving demographic, medical and medication history, laboratory, and body composition data. Metabolic indicators, including glycosylated hemoglobin, fasting blood sugar (FBS), serum albumin, hemoglobin, ACR, uric acid, triglycerides, total cholesterol, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol etc, were all measured using standard laboratory methods. The eGFR was computed utilizing the formula established by Chronic Kidney Disease Epidemiology Collaboration.²⁴ Body composition was evaluated using the direct segmental multi-frequency bioelectrical impedance analysis technique (Inbody770, Biospace, Korea). Patients removed their shoes and wore only close-fitting clothes, and a professional technician performed the measurement procedure. The BIA equipment estimated the total muscle mass, total fat mass, total body water, BMC, etc. The calculation methods of the derivative indicators are as follows. Skeletal muscle index (SMI) was determined by dividing the appendicular lean mass by the square of individual’s height. Fat mass index (FMI) was calculated by dividing the total fat mass by the square of the height. Fat-to-muscle ratio (FMR) was total fat mass divided by total muscle mass.

Outcomes

The main outcome of the study was the change in body composition parameters relative to baseline after 6 months. The secondary outcome was the change in metabolic indicators relative to baseline after 6 months.

Statistical Analysis

Statistical analysis was performed by R language 4.4.0. Categorical variables were expressed as absolute and relative frequency (n, %). Quantitative variables were reported as the mean ± standard deviation (SD) when normally distributed, and as medians with interquartile ranges (IQR) in cases of skewed distribution. Means were compared using a matched t-test for within-patient pre-to-post treatment, and a two-sample t-test for two groups. Medians were compared using a Wilcoxon signed-rank test for within-patient pre-to-post treatment, and a Wilcoxon rank sum test for two groups. Chi-squared test or Fisher’s Exact was used to compare the categorical variables. ANCOVA model was used to analyze the differences between two groups with adjustment for age and CKD stages for clinical variables, and age and BMI for body composition variables. We further conducted a subgroup analysis among people using SGLT-2Is only. The differences within subgroups were explored by one-way ANCOVA analysis. A two-tailed P-value <0.05 was deemed statistically significant.

Results

Baseline Features of Patients

Seventy-three people were included as the overall population, with 61 people only using SGLT-2Is and 12 people using SGLT-2Is and Semaglutide (Figure 1). Regarding the use of SGLT-2Is, most people were on dapagliflozin and 3 (4.9%) people were on empagliflozin. People initiated Semaglutide almost for obesity at least after 3 months of SGLT-2Is use. The maintenance dose of Semaglutide was 0.5–1.0 mg per week. Among 73 patients with T2DM and CKD, 4 (5.5%) people were biopsy-confirmed IgA nephropathy, and 1 (1.4%) were biopsy-confirmed membranous nephropathy.

The baseline clinical characteristics are summarized in Table 1. The average age of included patients was 51.3 ± 9.6 (SD) years, with 82.2% male. The median DM duration was 7.7 ± 5.5 (SD) years. A total of 78.1% of the patients had hypertension. The mean of eGFR was 67.3 ± 24.9 (SD) mL/min/1.73 m2. The median of ACR was 267 (IQR, 98.3–623.2) mg/g. Among them, 20.5% of the subjects were prescribed insulin. The body composition metrics for all the patients were detailed in Table 2. The mean of BMI was 25.9 ± 3.5 (SD) kg/m2. The mean of SMI was 7.6 ± 0.9 (SD) kg/m2.

Table 1 Baseline Clinical Characteristics of Patients with T2DM and CKD Using SGLT-2Is or Using SGLT-2Is Plus Semaglutide

Table 2 Baseline Body Composition Characteristics of Patients with T2DM and CKD Using SGLT-2Is or Using SGLT-2Is Plus Semaglutide

Compared people in the SGLT-2Is group, people in the SGLT-2Is plus Semaglutide group were generally younger, demonstrated elevated levels of ACR and triglyceride, and exhibited a significantly higher level of BMI, SMI, FMI, BMC, total body water, visceral fat area, and upper limb muscle circumference.

Change of Metabolic Indicators After 6 Months Compared to the Baseline

The changes of metabolic indicators after 6 months were summarized in Table 3. For SGLT-2Is group, serum albumin showed a subtle rise (P = 0.025). For SGLT-2Is plus Semaglutide group, serum albumin was also increased (P = 0.016). The levels of FBS (−0.88 ± 1.23 (SD) mmol/L, P=0.03) and triglyceride (−0.98, (IQR, −1.63–0) mmol/L, P = 0.042) were significantly declined. ACR showed a declining tendency (−200 (IQR, −515– −33) mg/g, P = 0.06). After the adjustment of covariates by ANCOVA models, the increase of serum albumin and uric acid and the reduction of total cholesterol were significant in SGLT-2Is plus Semaglutide group compared with SGLT-2Is groups.

Table 3 Change From Baseline to Over 6 months in Clinical and Body Composition Parameters

Change of Body Composition After 6 Months Compared to the Baseline

The changes of body composition after 6 months were presented in Table 3. There was a notable decrease for SGLT-2Is group in total fat mass by −0.6 kg (IQR, 1.7–0.7), total body water by −0.6 kg ± 1.2 (SD) (with a subsequent reduction in extracellular water), total muscle mass by −0.4 kg ± 0.8 (SD), and BMC by −0.03 kg ± 0.1 (SD). SMI, FMI, visceral fat area, and upper limb muscle circumference also decreased (P<0.05). FMR showed no significant change.

For SGLT-2Is plus Semaglutide group, a significant reduction was observed in total fat mass (−0.9, IQR (−3.7–0.4) kg), total body water (−1.5 ±1.4 (SD) kg), total muscle mass (−1.1 ±1.2 (SD) kg), and BMC (−0.1 ±0.11 (SD) kg). Additionally, SMI showed significant declines (P=0.001), and FMR remained stable.

In both the t-test and the ANCOVA test to correct for confounding factors, the SGLT-2Is plus Semaglutide group had more reductions in BMI, FMI and SMI than the SGLT-2Is group (Table 3, Figure 2A–C, Figure S1A and S1C). There was no significant difference in the proportion of total fat mass to weight between the two groups, but the proportion of total muscle mass to weight was increased and higher in SGLT-2Is plus Semaglutide group (Table 3, Figure 2B and D, Figure S1B and S1D). FMR indicated no statistically significant difference between the two groups, with approximately 50% of patients experiencing an increase and 50% experiencing a decrease (Table 3, Figure 2E and F and Figure S1E).

Figure 2 Changes in fat and muscle parameters after 6 months of SGLT-2Is or SGLT-2Is plus Semaglutide treatment in T2DM and CKD. Abbreviations: SGLT-2Is, sodium-glucose co-transporter inhibitors; T2DM, type 2 diabetes mellitus; CKD, chronic kidney disease; FMI, fat mass index; SMI, skeletal muscle index. Notes: average change of (A) FMI; (B) total fat mass/weight; (C) SMI; (D) total muscle mass/weight; (E) fat-to-muscle ratio; (F) cumulative change of fat-to-muscle ratio. P-value was using ANCOVA model adjusted for age and body mass index. Bold represents statistical significance.

Subgroup Analysis of Body Composition Changes Within People Only Using SGLT-2Is

Given the significant changes in body composition observed after 6 months of SGLT-2Is treatment, we further explored whether these changes differed across subgroups stratified by age, sex and other covariates (of tertiles). The changes in BMI and FMI were consistent across all subgroups (Figure 3A and B). Although not statistically significant, a greater reduction in SMI was observed in the subgroup with higher baseline SMI (Figure 3C). BMC appeared to decrease less in the subgroup with higher FMI (Figure 3D). The extracellular water ratio showed a greater reduction in the group with higher ACR (Figure 3E). Detailed numerical data can be found in Table S1–7.

Figure 3 Continued.

Figure 3 Change in body compositions after 6 months by different subgroups in patient with T2DM and CKD only using SGLT-2Is. Abbreviations: SGLT-2Is, sodium-glucose co-transporter inhibitors; BMI, body mass index; FMI, fat mass index; SMI, skeletal muscle index; BMC, bone mineral content. Notes: (A) subgroup analysis of BMI; (B) subgroup analysis of FMI; (C) subgroup analysis of SMI; (D) subgroup analysis of BMC; (E) subgroup analysis of extracellular water rate. The differences within subgroups were explored by one-way ANCOVA analysis. Bold represents statistical significance.

Discussion

Age-related clinical manifestations such as wasting and sarcopenia have been recognized as important risk factors in patients with T2DM and CKD.^25,26 Novel antidiabetic agents exhibit varying efficacy in weight reduction, but their impact on sarcopenia remains debated. Potential drug effects on the human body composition cannot be overlooked. To address this issue, we firstly employed BIA to investigate alterations in body composition before and after treatment of SGLT-2Is and Semaglutide in people with T2DM and CKD.

Our study demonstrated that after 6 months treatment with either SGLT-2Is alone or SGLT-2Is combined with Semaglutide, patients with T2DM and CKD exhibited reduction in total fat mass, total muscle mass, total body water, and BMC, without significance of FMR. When comparing treatments, the combination therapy group showed significantly greater reductions in both total fat mass and total muscle mass than the SGLT-2Is alone group. The combination group also led to a greater decrease in the SMI, a key indicator of sarcopenia. However, the combination group achieved a more significant increase in the proportion of muscle relative to total body weight. The results may suggest a preferential loss of fat than muscle with SGLT-2Is-Semaglutide combination therapy.

Our findings are partially consistent with previous reports of Semaglutide in patients with T2DM or obesity. Semaglutide significantly reduces muscle mass but increases the proportion of muscle mass relative to total body weight, with ambiguous changes in bone. In a substudy of SUSTAIN8 (n = 178), weekly injections of 1 mg Semaglutide for 1 year reduced absolute lean body mass but increased the proportion of lean body mass relative to total body weight in T2DM patients.²⁷ Additionally, a study of 43 obese Chinese patients found no significant changes in SMI or grip strength after 6 weeks of Semaglutide treatment.²⁸ A systematic review of six studies further confirmed that while Semaglutide reduced absolute lean body mass in overweight or obese patients, the percentage of lean to body weight mass significantly increased, suggesting an overall beneficial effect.²⁹ Nevertheless, a study of T2DM patients treated with GLP-1RAs (n = 24 with Semaglutide) for 1 year found a significant reduction in femoral and lumbar bone mineral density (BMD) as measured by dual-energy X-ray absorptiometry (DXA), while radiofrequency echo ultrasound (REMS) and bone turnover markers showed neutral results.³⁰ Moreover, in population at increased risk of fractures (n = 64), 1 year of Semaglutide treatment revealed no significant changes in the bone formative marker, but increase in the bone resorptive marker and inconsistent changes in BMD by sites.³¹

Notably, the present study indicated that 19% of the weight loss induced by Semaglutide is attributed to a reduction in muscle mass. Other studies have suggested that GLP-1RAs can make decline in lean body mass from 20% to 40% weight loss.³² However, patients with T2DM and CKD often exhibit chronic low-grade inflammation and elevated oxidative stress,^33,34 which promote protein breakdown and inhibit muscle synthesis.^35,36 Abnormalities of uremic toxins, vitamin D, and parathyroid hormone in advanced CKD exacerbate muscle loss and bone metabolism disorders.^37–40 A loss of 10% or more of muscle mass is equivalent to the muscle loss associated with 10 or more years of aging.⁴¹ Significant reductions in muscles may adversely affect metabolic health and elevate the likelihood of sarcopenic obesity after discontinuation of treatment.⁴² Therefore, when evaluating the fat-reducing benefits of Semaglutide, the negative impact of muscle and BMC loss in specific populations cannot be overlooked.

Current studies indicate that the weight loss effect of Semaglutide reaches a plateau after 6 to 12 months treatment, with most research on body composition changes limited to this timeframe or shorter.^16–19 However, with the long-term use of antidiabetic drugs, aging, and the progression of CKD, these patients may face a higher risk of muscle loss and bone fragility. Whether the trends in muscle mass and BMC stabilize, improve, or worsen after weight stabilization in T2DM and CKD remains unclear and requires further long-term studies. Additionally, research focusing on different high-risk populations will provide more comprehensive evidence for clinical practice.

GLP-1RAs induce weight loss primarily by modulating appetite centers to reduce food intake. However, maintenance of muscle quality during calorie restriction-induced weight loss largely depends on two critical factors: nutritional intake and physical exercise.⁴³ Adequate intake of high-quality protein and micronutrients should be ensured. Clinical studies have shown that compared to exercise intervention alone, the addition of liraglutide not only enhances weight loss but also improves muscle preservation and attenuates BMD reduction.^44,45 Furthermore, some combination drug therapies exhibit potential muscle reserve function. For instance, Bimagrumab, an antibody blockade of activin type II receptor, promote fat loss while preventing muscle wasting during Semaglutide therapy in obese mice.⁴⁶ Safety and efficacy of Bimagrumab and Semaglutide in the overweight is under Phase 2 RCT to examine (NCT05616013). Based on current evidence, it is recommended that patients with T2DM and CKD receiving Semaglutide therapy regularly monitor their body weight. For patients experiencing significant weight reduction, a structured exercise regimen (mainly resistance training) should be emphasized, and protein and micronutrient supplementation should be considered for.

Previous studies on body composition changes associated with SGLT-2Is have primarily focused on individuals with T2DM, mainly showing a reduction in body water or a trend toward fat and muscle loss that lacked statistical significance.^16,17,47,48 Given the larger cohort in our SGLT-2Is group, we conducted an internal subgroup analysis and identified several findings. Among patients treated with SGLT-2Is, those with higher baseline SMI tended to show a greater reduction in SMI after 6 months, while no significant differences were observed across age subgroups. However, a study has reported that patients with T1DM using SGLT-2Is, particularly those with a lower BMI (<23kg/m2) and males over 60 years, experienced more pronounced muscle loss.⁴⁹ The possible explanation of our results may be due to the fact that individuals with greater baseline muscle mass tend to lose a higher proportion of muscle during weight loss for an adaptive metabolic response to reduce energy expenditure.²⁰ In our study, older patients exhibited greater heterogeneity in SMI changes, potentially due to the clinical practice of recommending resistance training to older individuals to preserve sarcopenia, thus resulting in significant interindividual variability in this subgroup. Notably, a case reports an elderly T2DM patient who experienced rapid and severe muscle loss and functional decline while using an SGLT-2Is,⁵⁰ underscoring the need for careful monitoring of muscle loss risks in older individuals receiving SGLT-2Is. In subgroups with higher FMI, the reduction in BMC appeared to be less. Traditionally, obesity has been considered protective against fractures, with an inverse relationship between BMI and fracture risk, although excessively high BMI may attenuate this effect.⁵¹ Adipokines such as adiponectin, secreted by adipose tissue, may contribute to bone protection.⁵² Furthermore, the extracellular water ratio decreased more significantly in subgroups with higher ACR. Because patients with severer proteinuria often exhibit greater sodium and water retention and interstitial edema.⁵³ Consequently, the inhibition of sodium reabsorption by SGLT-2Is may lead to more in extracellular water.

Several limitations exist in this study. First, as a retrospective study based on a small cohort, it only assessed short-term outcomes. The SGLT-2Is plus Semaglutide group had a smaller sample size and higher body weight. However, after adjustment using an ANOVA model, statistically significant results were still observed, offering insights for future research. Additionally, since the use of SGLT-2Is is strongly recommended in patients with T2DM-CKD, this study did not include a control group without SGLT-2Is treatment. Second, the study primarily evaluated muscle mass through body composition analysis without multi-scale indicators such as muscle function and pathology. However, by employing multiple metrics, including SMI, percentage changes in muscle mass relative to body weight, and FMR, this study provided relatively robust evidence to support the findings. Third, although older patients and those with low muscle mass probably received additional recommendations for resistance exercise and food, this study did not collect information on exercise or diets that can influence muscle mass. Future research could further investigate the impact of dietary or exercise on body composition in patients with T2DM-CKD using Semaglutide and evaluate their long-term outcomes.

Conclusion

After 6-month administration of SGLT-2Is or a combination of SGLT-2Is with Semaglutide, patients with T2DM and CKD exhibited reduction in total fat mass, total muscle mass, total body water, and BMC. Compared to the SGLT-2Is group, the combination therapy group showed a greater absolute reduction in muscle mass, but a more increase in muscle mass-to-body weight percent, while the relative ratio of fat to muscle remained unchanged. In the SGLT-2Is group, patients with higher baseline muscle mass experienced greater muscle loss, whereas those with higher fat mass had less bone mineral loss. Further studies with longer treatment periods and a focus on more vulnerable populations are needed.