Saxagliptin Combined with Liraglutide vs. Saxagliptin Alone on Microinflammation and Adipokines in Obese T2DM Patients with Poor Metformin Response

Summary

This protocol describes a retrospective cohort method to evaluate combined saxagliptin and liraglutide therapy on inflammation and adipokine profiles in obese type 2 diabetes patients with poor response to metformin.

Abstract

Metformin is the first-line therapy for obese patients with type 2 diabetes mellitus (T2DM); however, a substantial proportion of patients exhibit inadequate glycemic control despite optimal dosing. This protocol describes a retrospective cohort approach to evaluate the effects of saxagliptin combined with liraglutide on microinflammation and adipokine profiles in obese T2DM patients with poor response to metformin. A total of 116 eligible patients were identified from electronic medical records and assigned to receive either saxagliptin monotherapy or saxagliptin combined with liraglutide based on predefined clinical criteria. The protocol includes standardized procedures for patient selection, group allocation, metabolic assessment, and laboratory measurement of inflammatory markers, adipokines, and metabolic parameters.

Representative outcomes demonstrate that both treatment regimens are associated with improvements in glycemic control, lipid metabolism, and insulin resistance. The combination regimen shows greater reductions in inflammatory markers and more favorable modulation of adipokine profiles compared to monotherapy, while maintaining a comparable safety profile. These findings illustrate the application of this protocol for evaluating multi-target metabolic interventions in real-world clinical settings.

This method provides a reproducible framework for assessing combined pharmacological strategies in obese T2DM patients with suboptimal response to first-line therapy and may support further clinical and translational investigations.

Key words: Saxagliptin; Liraglutide; Obesity-related type 2 diabetes; inflammatory response; Microinflammatory response

Introduction

T2DM is a global epidemic, with its incidence increasing year by year and posing a major public health threat. Epidemiological surveys indicate approximately 463 million diabetes patients worldwide, over 90% of whom have T2DM. Obesity is one of the most important risk factors for T2DM¹. In China, rapid economic development and lifestyle changes have led to a sharp rise in obesity‑related T2DM, creating a “dual disease burden”. These patients often present not only with significant glucose metabolism disorders but also with lipid abnormalities, insulin resistance (IR), chronic low‑grade inflammation, and dysregulated adipokine secretion, together constituting a complex metabolic syndrome^2,3,4. Metformin is the first‑line oral antidiabetic agent for obese T2DM patients due to its efficacy in reducing hepatic gluconeogenesis, improving peripheral insulin sensitivity, and its favorable safety profile. However, approximately 20–45% of patients do not achieve adequate glycemic control (HbA1c < 7.0%) despite taking metformin at a dose of ≥ 1500 mg·day^-1 for ≥ 12 weeks. Such patients are considered to have a “poor metformin response” and require add‑on therapy. The mechanisms underlying poor response include severe insulin resistance, impaired incretin effect, and progressive β‑cell dysfunction^5,6.

The pathophysiology of obese T2DM involves chronic low‑grade inflammation (microinflammation) and adipokine dysregulation, which are key links between obesity, insulin resistance (IR), and β‑cell dysfunction^7,8. In obesity, adipose tissue hypoxia and stress promote macrophage infiltration and pro‑inflammatory factor release (e.g., TNF‑α, IL‑6, CRP), leading to “adipose tissue inflammation”⁹. This state exacerbates IR and glucose dysregulation by interfering with insulin signaling, impairing β‑cell function, and enhancing hepatic gluconeogenesis, forming a vicious cycle^10,11. Concurrently, altered adipokine profiles (decreased adiponectin, increased resistin) further worsen metabolic disorders and cardiovascular risk^12,13. Thus, effective treatment for obese T2DM must not only control glycemia but also target IR, restore adipokine balance, and alleviate microinflammation as a multi‑domain intervention. This approach is particularly suitable for evaluating multi-target therapeutic strategies in patients with complex metabolic dysregulation who do not respond adequately to first-line therapy.

The glucagon-like peptide-1 (GLP-1) receptor agonist liraglutide and the dipeptidyl peptidase-4 (DPP-4) inhibitor saxagliptin are currently two types of new hypoglycemic drugs commonly used in clinical practice. Both exert their effects by enhancing the activity of endogenous GLP-1, but their mechanisms of action and effects are different^14,15. Liraglutide, as a GLP-1 receptor agonist, has multiple effects: it delays gastric emptying, inhibits the appetite center, promotes insulin secretion, inhibits glucagon release, and may have a direct anti-inflammatory effect^16,17,18. Clinical studies have confirmed that liraglutide not only effectively lowers blood sugar but also significantly reduces weight, improves insulin resistance and lipid profile, and has a positive effect on liver fat content and non-alcoholic fatty liver disease (NAFLD)¹⁹. Saxagliptin, as a DPP-4 inhibitor, prolongs the activity of GLP-1 by inhibiting its degradation, promotes insulin secretion in a glucose concentration-dependent manner, inhibits glucagon release, has a neutral weight effect, and has excellent safety^20,21. It is worth noting that although both of these drugs act on the GLP-1 system, their mechanisms of action are complementary: liraglutide provides exogenous GLP-1 receptor activation, while saxagliptin enhances endogenous GLP-1 activity by inhibiting DPP-4. Theoretically, the combination of the two may produce a synergistic effect, controlling blood sugar while more comprehensively improving IR, weight, and metabolic abnormalities.

At present, research on the combination of saxagliptin and liraglutide for the treatment of obese T2DM patients with suboptimal metformin efficacy is relatively limited, especially the impact of this combination regimen on the micro-inflammatory state and the lipid factor profile has not been fully clarified. Existing studies have shown that liraglutide monotherapy can significantly reduce the levels of AST, ALT, TB, TBA, TC, TG, and FFA in T2DM patients with NAFLD, and increase HDL-C²². Another study shows that liraglutide combined with metformin in the treatment of obese type 2 diabetes can significantly improve the micro-inflammatory state of patients and reduce the levels of TNF-α and IL-6²³. However, the effects of saxagliptin combined with liraglutide on adipokines (such as adiponectin, resistin, Vaspin, Chemerin) and the synergistic mechanism of their regulation in micro-inflammatory responses still need further exploration. The specific effects of saxagliptin combined with liraglutide on adipokine regulation and microinflammatory pathways in obese T2DM patients remain largely unknown.

This study aims to investigate the effects of saxagliptin combined with liraglutide on micro-inflammatory responses and adipokines in obese T2DM patients with suboptimal metformin efficacy. The goal of this protocol is to provide a standardized retrospective approach to evaluate combined pharmacological interventions targeting metabolic and inflammatory pathways in obese T2DM patients. A total of 116 patients meeting the criteria were included, and they were divided into the control group and the combined group according to the intervention method. Key outcomes included anthropometric measures, glucose and lipid metabolism, adipokine profiles, inflammatory markers, and safety. We hypothesize that combination therapy offers superior metabolic and anti-inflammatory effects over monotherapy. We hypothesize that combination therapy offers superior metabolic and anti-inflammatory effects over monotherapy. This study will provide a more optimized combined treatment plan for obese T2DM patients with suboptimal metformin efficacy and provide new clinical evidence for understanding the drug mechanism.

Protocol

1. Ethical Approval

1. Obtain written approval from the institutional ethics committee before starting any study procedure.
2. Archive the approval document (Huangpu District Dapuqiao Community Health Center Ethics Review Committee, Approval Number: 2022‑032‑A, dated March 12th, 2022). Include this number in the final publication.

2. Patient Identification and Eligibility Screening

1. Retrieve electronic medical records (EMR) of all patients diagnosed with type 2 diabetes mellitus (T2DM) who visited the hospital between May 2022 and December 2023 (Refer to Figure 1 for work-flow).
2. Apply the following filter in the EMR query interface: BMI ≥ 28 kg·m⁻², metformin dose ≥ 1500 mg·day⁻¹, and treatment duration ≥ 12 weeks.
3. Generate a preliminary patient list weekly. Export the list as a comma‑separated values (CSV) file containing age, sex, BMI, HbA1c, and current medications.
4. Assign two trained research physicians (each with ≥ 3 years of endocrinology experience) to independently screen all candidate records.
5. Mark a patient as “initially qualified” if both physicians agree that all inclusion criteria are met and none of the exclusion criteria apply.
   1. Inclusion criteria²⁴: T2DM diagnosis, age ≥ 18 years, BMI ≥ 24 kg·m⁻², metformin monotherapy (≥ 1500 mg·day⁻¹) for ≥12 weeks, and most recent HbA1c ≥ 7.5%.
   2. Exclusion criteria²⁵: type 1 diabetes, gestational or secondary diabetes; active autoimmune or chronic inflammatory diseases; acute infection or major surgery within 4 weeks; active malignancy (except basal cell skin cancer); untreated thyroid dysfunction; ALT or AST > 3× upper normal limit; total bilirubin > 2× upper normal limit; eGFR < 45 mL·min⁻¹·1.73 m⁻²; use of systemic glucocorticoids (> 7 days), high‑dose NSAIDs (> 3 days), GLP‑1 agonists, DPP‑4 inhibitors (other than study drugs), SGLT2 inhibitors, thiazolidinediones, insulin, or weight‑loss medications within 3 months; pregnancy or lactation; severe mental illness; alcohol/drug abuse within 1 year; known allergy to saxagliptin or liraglutide; history of pancreatitis; personal or family history of MEN‑2 or medullary thyroid carcinoma.
6. Refer cases with inconsistent assessments to a third senior endocrinologist (≥ 10 years of experience). Ask the senior endocrinologist to make the final eligibility decision and document the reason for disagreement.
7. Compile the final list of eligible patients. Confirm that 116 patients meet all criteria from the initial pool of 328 patients. Archive all screening records, eligibility assessments, and decision logs in a password‑protected folder for future reference.

3. Group Allocation (Based on Existing Clinical Records)

1. Classify each eligible patient according to the treatment regimen documented in the EMR. Assign patients receiving saxagliptin alone (5 mg·day⁻¹) to the control group (n = 58).
2. Assign patients receiving saxagliptin (5 mg·day⁻¹) plus liraglutide to the combination group (n = 58).
3. Verify that liraglutide prescription was based on institutional criteria documented in the EMR: HbA1c ≥ 8.5% despite metformin ≥ 1500 mg·day⁻¹ for ≥ 12 weeks, or BMI ≥ 30 kg·m⁻², or HOMA‑IR ≥ 3.5, or central obesity (waist circumference ≥ 90 cm for males, ≥ 85 cm for females). If none of these criteria are met, the patient should have been assigned to the control group.

4. Baseline Assessment (Week 0)

1. Instruct the patient to fast overnight (≥ 8 h, water allowed) and to wear light clothing without shoes or socks.
2. Measure height using a calibrated wall‑mounted stadiometer (range 60‑200 cm, accuracy 0.5 cm). Ask the patient to stand upright with heels together, occiput and back touching the pillar. Read the value to the nearest 0.5 cm.
3. Measure weight using a calibrated electronic scale (max 150 kg, accuracy 0.1 kg). Ask the patient to stand in the center of the scale. Record the value after the reading stabilizes.
4. Measure waist circumference with a standard soft tape measure (accuracy 0.1 cm). Locate the midpoint between the anterior superior iliac spine and the lower edge of the 12th rib. Wrap the tape horizontally around the abdomen. Take the reading at the end of normal exhalation.
5. Measure hip circumference at the maximum circumference of the buttocks using the same tape measure. Record to 0.1 cm.
6. Calculate waist‑to‑hip ratio (WHR) = waist circumference (cm) ÷ hip circumference (cm). Repeat all anthropometric measurements twice and record the average.
7. Assess physical activity using the International Physical Activity Questionnaire Short Form (IPAQ‑SF). Ask the patient to recall activities over the past 7 days.
   1. Calculate walking MET‑min/week = 3.3 × daily walking min × walking days per week. Calculate moderate‑intensity MET‑min/week = 4.0 × daily moderate activity min × activity days per week.
   2. Calculate high‑intensity MET‑min/week = 8.0 × daily high‑intensity activity min × activity days per week. Sum the three values to obtain total MET‑min/week. Classify as: sedentary (< 600), mild (600‑2999), moderate (3000‑5999), or severe (≥ 6000).​

Pause Point 1: After completing baseline assessments (week 0) and before assigning the treatment plan, data has been collected and verified. The patient may leave, and the intervention can start at the next visit.

5. Dietary Planning (Actionable Steps for the Dietitian)

1. Open the nutrition assessment software (“Dietary Analysis and Meal Planning System”, version 2.0 or higher). Select “File” → “New Patient Record”.
2. Input the patient’s height (cm, to 0.5 cm), weight (kg, to 0.1 kg), age (years, integer), and gender (male/female) into the corresponding fields.
3. Select the Mifflin‑St Jeor equation from the “Tools → Basal Metabolic Rate Calculator” menu.
   1. For males: BMR (kcal·day⁻¹) = 10 × weight (kg) + 6.25 × height (cm) – 5 × age (years) + 5.
   2. For females: BMR = 10 × weight (kg) + 6.25 × height (cm) – 5 × age (years) – 161.
4. Multiply the BMR by the activity coefficient obtained from the IPAQ‑SF classification: sedentary ×1.2, mild ×1.375, moderate ×1.55, severe ×1.725. This gives resting energy expenditure (REE).
5. Set the daily calorie target = REE × 0.90 (mild calorie restriction). If the result falls outside 25‑30 kcal·kg⁻¹ of ideal body weight, adjust to the nearest boundary value²⁶.
6. Navigate to “Plan” → “Generate Dietary Plan”. Set macronutrient ratios: carbohydrates 50‑55%, protein 15‑20%, fat 25‑30% (use sliders)²⁷.
7. In output settings, select “Food Exchange Portion List” and a 7‑day recurring menu. Confirm exchange definitions: 1 portion grains = 15 g carbohydrates; 1 portion meat/eggs = 7 g protein; 1 portion fat = 5 g fat.
8. Click “Generate”. Provide the patient with a printed paper version (A4) and, if available, share a synchronized version to the patient’s mobile application (“Dietary Assistant”, version 1.0) using the patient’s unique study number.
9. Instruct the patient to record all meals and snacks daily using the mobile app or a standardized paper diary (see Appendix A). Assign a research nurse to review dietary records weekly (by phone, 10‑15 min). If the deviation from the prescribed calorie target exceeds 10% for two consecutive weeks, repeat steps 5.1‑5.9 to re‑educate the patient.

6. Structured Exercise Plan

1. Instruct the patient to perform moderate‑intensity walking (or slow jogging) for 10–15 min, starting 1 h after each main meal (breakfast, lunch, dinner). Total daily duration must be ≥ 30 min.
2. Provide a Bluetooth heart rate wristband (model: universal wrist optical heart rate recorder, sampling frequency 1 Hz). Calculate target heart rate range: (220 – age) × 60% to (220 – age) × 70%.
3. Ask the patient to wear the wristband during every exercise session. Record start time, duration, average heart rate, and subjective fatigue score (RPE, 6‑20 scale) in a paper exercise log (Appendix B).
4. Every Saturday by 8:00 PM the patient connects the wristband to a dedicated research tablet. Open the data synchronization software, select “Device → Sync Data”, and download 7‑day min‑by‑min heart rate data.
5. On Sunday evening, the research nurse compares the log with the downloaded data.
   NOTE: Allowable error: exercise duration ± 5 min, average heart rate ± 5 bpm. If inconsistent, mark the exercise session as “unverified”.
6. Conduct follow‑up visits or phone calls every two weeks (weeks 2,4,6,8,10,12,14,16). Open the past 14‑day exercise data on the tablet. Ask the patient about reasons for non‑adherence (e.g., weather, discomfort, forgetfulness) and record answers in Appendix C.
7. If the percentage of time with heart rate within target range is < 70% for two consecutive follow‑ups, or total weekly exercise duration < 210 min, arrange an additional 30‑minute face‑to‑face coaching session with a rehabilitation therapist.

7. Pharmacological Intervention

1. Saxagliptin Administration (Both Groups)
   1. Instruct the patient to take one 5 mg saxagliptin tablet (National Drug Approval H20193008) orally every morning at least 30 min before breakfast, or at the same fixed time each day regardless of meals.
   2. Swallow the tablet whole with at least 200 mL of warm water. Provide a labeled pillbox marked with days of the week.
   3. Ask the patient to tick the medication log (paper, Appendix D) immediately after each dose and record the administration time (h: min).
   4. At follow‑up visits (weeks 4,8,12,16), the research nurse counts the remaining tablets in the pillbox. A second nurse independently repeats the count. If the two counts differ (≥ 1 tablet), perform a third count and confirm jointly.
   5. Calculate medication compliance (%) = (actual number of doses taken / prescribed number of doses) × 100%. Prescribed doses for 4 weeks = 28. If compliance < 80% or > 120%, record the reason (e.g., missed dose, over‑dosage) in the case report form.
2. Liraglutide Administration (Combination Group Only)
   CAUTION: liraglutide may cause nausea, vomiting, and hypoglycemia. Do not use in patients with personal or family history of medullary thyroid carcinoma or MEN‑2.
   1. Use a pre‑filled injection pen (National Drug Approval J20160005, 3 mL containing 18 mg liraglutide, i.e., 6 mg·mL⁻¹). Attach a new 32G×4 mm needle before each injection.
   2. Train the patient and one family member using a needleless training pen until they can independently perform two consecutive error‑free full injection procedures.
      NOTE: The training must cover washing hands, injection site selection and rotation, needle installation, dose setting (0.6 mg), 90‑degree insertion, pressing the button and holding for 6 seconds, needle withdrawal, and safe disposal.
   3. Rotate injection sites daily according to a fixed 8‑day cycle. Keep each injection ≥ 2 cm away from previous points. Day 1 – upper left abdomen, Day 2 – lower left abdomen, Day 3 – upper right abdomen, Day 4 – lower right abdomen, Day 5 – left thigh, Day 6 – right thigh, Day 7 – left arm, Day 8 – right arm. Keep each injection ≥2 cm away from previous points.
   4. Start with 0.6 mg once daily. Maintain this dose for at least 7 days. Pause Point 2: Before week 4 dose escalation assessment for the combination group. The patient has taken 0.6 mg of liraglutide for one week. Pause to evaluate tolerance; the protocol can be resumed after tolerance confirmation. After 7 days, assess tolerance. Tolerance is good if gastrointestinal adverse reaction score ≤ 2 and duration < 3 days.
   5. If tolerance is good, increase the dose to 1.2 mg once daily on the first day of week 2. Maintain for 7 days, then reassess tolerance. If tolerance remains good, increase to the target maintenance dose of 1.8 mg once daily on the first day of week 3²⁸.
   6. If intolerable adverse reactions occur during escalation, revert to the last tolerated dose and do not attempt further escalation. Continue the 1.8 mg maintenance dose (or the highest tolerated dose) until week.

8. Blood Sample Collection and Processing

1. Collect all blood samples in the morning after an overnight fast (≥ 8 h). Perform venipuncture at the antecubital fossa using a 21G vacuum blood collection needle and holder.
2. Draw 3 mL of blood into a sodium fluoride/potassium oxalate tube (gray top) for fasting plasma glucose (FPG) and, later, 2‑h postprandial glucose (2hPG). Draw 3 mL of blood into an EDTA tube (purple top) for HbA1c.
3. Draw 4 mL of blood into a coagulation separation tube (yellow or red top) for lipids (TC, TG, HDL‑C, LDL‑C) and fasting insulin. Allow to stand at room temperature (20‑25°C) for 30 min to clot.
4. Draw 5 mL of blood into a coagulation separation tube (yellow top) for inflammatory markers (IL‑6, TNF‑α, ASAA, CRP) and adipokines (adiponectin, Vaspin, Chemerin, resistin).
5. Process all tubes within 60 min after collection. Centrifuge the gray and purple tubes at 4°C, 1500 × g for 10 min (fixed‑angle rotor F‑34‑6‑38, radius 15 cm). Centrifuge the yellow/red tubes at room temperature (20‑25°C) or 22°C if ambient temperature exceeds 25°C, at 1500 × g for 10 min.
6. Immediately separate the serum or plasma. Dispense 500 µL into each 1.5 mL microcentrifuge tube.
7. Label each tube with patient number, collection date, and freezing date. Store immediately at –80°C in a freezer set to -80°C ± 1°C.
8. Maximum storage time before analysis: 6 months. Allow only one freeze‑thaw cycle. To thaw, place the tube at 4 °C for 12–16 h. Do not refreeze; if retesting is needed, use a separate aliquot.
   Pause Point 3: After completing all week‑16 endpoint sample collections, before starting laboratory assays. Serum/plasma samples are stored at -80°C and can be assayed at any later date.

9. Laboratory Assays

1. Inflammatory Markers (IL‑6, TNF‑α, ASAA, CRP by ELISA) assay
   1. Remove serum samples from –80°C and thaw at 4°C for 12–16 h. Equilibrate all reagents and samples to room temperature (20–25°C) for 30 min.
      ​CAUTION: TMB substrate is a potential irritant. Wear gloves and safety glasses. Work in a fume hood or biosafety cabinet when handling.
   2. Prepare standard curves for each analyte using serial dilutions provided in the commercial ELISA kit. Typical ranges: IL‑6 0‑2000 pg·mL^-1 (8 points), TNF‑α 0‑1000 pg·mL^-1, ASAA 0‑500 ng·mL^-1, CRP 0‑100 ng·mL^-1.
      ​Pause Point 4: After adding samples and reagents to the ELISA plate, before the first incubation step. The plate can be sealed and stored at 4 °C overnight (≤ 16 h) without affecting results.
   3. Add 100 µL of standard or sample (pre‑dilute samples 1:2 with sample diluent for ASAA and CRP) to each well of a 96‑well plate.
   4. Cover the plate with a sealing membrane. Incubate at 37°C for 2 h (static, no shaking).
   5. Discard the liquid. Add 300 µL of wash buffer (PBS with 0.05% Tween‑20) per well, soak for 30 seconds, and discard. Repeat the washing 5 times. After the final wash, tap the plate on absorbent paper to remove residual buffer.
   6. Add 100 µL of biotin‑labeled detection antibody to each well (working concentrations: IL‑6 0.5 µg·mL⁻¹, TNF‑α 0.25 µg·mL⁻¹, ASAA 0.2 µg·mL⁻¹, CRP 0.4 µg·mL⁻¹). Incubate at 37°C for 1 h. Wash 5 times as in step 9.1.5.
   7. Add 100 µL of horseradish peroxidase‑labeled streptavidin (diluted 1:200) to each well. Incubate at 37°C for 30 min. Wash 5 times as in step 9.1.5.
   8. Add 100 µL of TMB substrate solution to each well. Incubate in the dark at room temperature (20‑25°C) for exactly 15 min. Add 100 µL of stop solution (2 M H₂SO₄) to each well. Mix gently by tapping the plate.
   9. Read absorbance at 450 nm with a reference wavelength of 570 nm using a microplate reader. Fit a four‑parameter logistic regression curve using the reader’s software (standard concentration on x‑axis, absorbance on y‑axis). Calculate sample concentrations from the curve.
2. Adipokines (Adiponectin, Vaspin, Chemerin, Resistin by ELISA)
   1. Use the same thawing, equilibration, and safety procedures as in steps 9.1.1‑9.1.2. Select commercial ELISA kits specific for each adipokine.
   2. Prepare standard curves: adiponectin 0‑50 µg·mL⁻¹, Vaspin 0‑20 ng·mL⁻¹, Chemerin 0‑50 ng·mL⁻¹, resistin 0‑100 ng·mL⁻¹ (eight points each).
   3. Add 100 µL of standard or sample to each well. For Chemerin and resistin, pre‑dilute samples 1:5 with sample diluent. Cover and incubate at 37°C for 2 h (static). Wash 5 times as in step 9.1.6.
   4. Add 100 µL of detection antibody (adiponectin 1 µg·mL⁻¹, Vaspin 0.5 µg·mL⁻¹, Chemerin 0.3 µg·mL⁻¹, resistin 0.4 µg·mL⁻¹). Incubate at 37°C for 1 h. Wash 5 times. Add 100 µL of enzyme conjugate (diluted 1:200). Incubate at 37°C for 30 min.
   5. Wash 5 times. Add 100 µL TMB, incubate in dark for 15 min. Add 100 µL stop solution (2 M H₂SO₄). Read absorbance at 450/570 nm. Calculate concentrations using four‑parameter logistic regression.
3. Glycemic Indicators (FPG, 2hPG, HbA1c)
   1. For FPG: use the gray‑top tube plasma from the fasting draw. For 2hPG: collect a second gray‑top tube exactly 2 h after the start of a standard 75 g oral glucose tolerance test (OGTT).
   2. Centrifuge both gray‑top tubes as described in step 8.6. Analyze plasma immediately or store at –80°C for up to 1 week.
   3. Set the automatic biochemical analyzer) to the hexokinase method with the following parameters: main wavelength 340 nm, sample volume 3 µL, reagent volume 200 µL, incubation temperature 37 °C, incubation time 5 min.
   4. Calibrate the analyzer using four calibrators (2.5, 5.0, 10.0, 20.0 mmol·L^-1) before the first run of each batch.
   5. Run a quality control sample (normal level: 4.5‑6.0 mmol·L^-1, abnormal level: 10.0‑12.0 mmol·L^-1) after every 20 patient samples.
   6. For HbA1c: use EDTA‑anticoagulated whole blood (purple top). Inject the sample into a high‑performance liquid chromatography (HPLC) analyzer set to: flow rate 1.5 mL·min⁻¹, column temperature 40°C, detection wavelength 415 nm. Report results in NGSP (%) and IFCC (mmol·mol⁻¹) units.
4. Lipid Metabolism Indicators (TC, TG, HDL‑C, LDL‑C)
   1. Use serum from the yellow/red top tube (coagulation separation) processed in step 8.6.
   2. For total cholesterol (TC): set the biochemical analyzer to the CHOD‑PAP method (main wavelength 500 nm, secondary 600 nm, sample 3 µL, reagent 200 µL, 37°C, 10 min).
   3. For triglycerides (TG): set to the GPO‑PAP method (main wavelength 540 nm, secondary 600 nm, sample 3 µL, reagent 200 µL, 37°C, 10 min).
   4. For HDL‑C: use the direct method (heminase inhibition). Main wavelength 600 nm, secondary 700 nm, sample 3 µL, R1 (blocker) 150 µL, R2 (enzyme/chromogen) 50 µL, 37°C, 10 min.
   5. For LDL‑C: use the direct method (selective surfactant inhibition) with the same wavelength settings as for HDL‑C.
   6. Run manufacturer‑provided calibrators (TC 5.17 mmol·L⁻¹, TG 1.69 mmol·L⁻¹, HDL‑C 1.55 mmol·L⁻¹, LDL‑C 3.36 mmol·L⁻¹) and two levels of quality control samples in each batch.
5. Insulin Function Indicators (FINS, HOMA‑β, HOMA‑IR)
   1. Use serum from the yellow/red top tube (coagulation separation) as in step 8.6
   2. Detect fasting insulin (FINS) using an automated chemiluminescence immunoassay analyzer (double‑antibody sandwich method). Set parameters: sample volume 50 µL, capture antibody 100 µL, detection antibody 100 µL, incubation 37°C for 18 min, wash 3 times with 300 µL wash buffer, add 100 µL luminol substrate, measure relative light units.
   3. Fit a four‑parameter logistic curve using six standard concentrations (0, 5, 10, 50, 100, 200 mU·L⁻¹). Run low (10‑15 mU·L⁻¹) and high (60‑80 mU·L⁻¹) quality controls per batch.
   4. Calculate HOMA‑IR = FPG (mmol·L⁻¹) × FINS (mU·L⁻¹) ÷ 22.5.
   5. Calculate HOMA‑β = 20 × FINS (mU·L⁻¹) ÷ (FPG – 3.5).

10. Safety Monitoring and Adverse Event Recording

1. At each follow‑up visit (weeks 4,8,12,16), ask the patient using a standardized questionnaire (Appendix E) about any symptoms experienced in the past 4 weeks: nausea, vomiting, diarrhea, constipation, abdominal pain, loss of appetite, hypoglycemic events (palpitations, tremor, cold sweat, hunger, blood glucose < 3.9 mmol·L⁻¹), dizziness, rash, injection site reactions (redness, pain, induration).
2. Instruct the patient to record any discomfort in the “Adverse Events” column of the medication log (Appendix D) including date of onset, duration, and severity (mild: no effect on daily activities; moderate: partial effect; severe: unable to perform activities).
3. If the patient reports suspected hypoglycemia, ask them to measure fingertip capillary blood glucose immediately using their own glucometer. Confirm hypoglycemia when glucose < 3.9 mmol·L⁻¹ with or without symptoms. Define severe hypoglycemia as glucose < 2.8 mmol·L⁻¹ or requiring assistance from others.
4. Grade each adverse event according to CTCAE. Record the relationship to study drug as: definitely related, likely related, possibly related, possibly unrelated, definitely unrelated.
5. Report any grade ≥ 3 adverse event or any severe hypoglycemic event to the principal investigator and the ethics committee within 24 h.
6. For grade ≥ 3 events, suspend the study drug until the event resolves to ≤ grade 1. If resolution does not occur within 7 days, terminate treatment for that patient and refer to a specialist.

11. Laboratory Safety and Biosafety Procedures

1. Perform all blood handling steps inside a Biosafety Level 2 (BSL‑2) laboratory.
   CAUTION: human blood samples are potentially infectious (HIV, HBV, HCV). Always wear a laboratory coat, disposable latex gloves, protective glasses, and a mask when handling samples.
2. Perform all open handling of serum/plasma (e.g., aliquoting, diluting, ELISA addition) inside a Class II A2 biological safety cabinet.
   1. Before use, verify that the cabinet is operational and the airflow alarm is not active.
   2. Arrange items inside the cabinet with a clear “clean” zone (reagents, pipettes) and “contaminated” zone (sample tubes, used tips).
   3. After finishing, wipe all interior surfaces with 0.5% sodium hypochlorite solution (freshly diluted), then with 70% ethanol. Allow the cabinet to run for 10 additional min for self‑purification.
3. Discard all used needles, blood collection tubes, transfer pipettes, and other sharps immediately into a puncture‑resistant sharps container.
4. Place all contaminated plasticware (microcentrifuge tubes, pipette tips) and gloves into a yellow biohazard waste bag. Autoclave the bag at 121°C for 20 min before disposal by a licensed medical waste company.
5. Keep an eyewash station and a first‑aid kit in the laboratory. In case of a sample spill, cover the spill area with absorbent pads, pour 0.5% sodium hypochlorite solution, let it act for 20 min, then clean up using forceps and dispose in biohazard waste.

12. Sample size calculation

1. Use the G*Power software to estimate the required sample size. Set the significance level (α) to 0.05 and the test power (1 - β) to 90%.
2. Based on the effect size reported in the prior study by Coppenrath V A et al.³⁰, calculate that a minimum of 50 subjects per group is required. To account for potential subject dropout, include 58 research subjects in each group. This sample size ensures the statistical power and reliability of the research conclusions.

13. Statistical Analysis

1. Perform all analyses using SPSS software. Test continuous variables for normality using the Shapiro‑Wilk test.
2. For normally distributed data, express as mean ± standard deviation (\(\bar{x} \pm s\)). Use two‑way repeated measures ANOVA with time (pre, post) as within‑subject factor and group (control, combination) as between‑subject factor. If the time‑group interaction is significant (p < 0.05), perform post‑hoc pairwise comparisons using Bonferroni correction.
3. For categorical variables (adverse events), express as count and percentage [n (%)] and compare between groups using the chi‑square test (χ²). Set statistical significance at two‑tailed p < 0.05³¹.

Results

A total of 116 obese T2DM patients with poor metformin response (HbA1c ≥ 7.5% despite metformin ≥ 1500 mg·day^-1 for ≥ 12 weeks) were retrospectively included (Table 1). Patients were assigned to a control group (saxagliptin alone, n = 58) or a combination group (saxagliptin plus liraglutide, n = 58). Baseline demographic and clinical characteristics were comparable between the two groups (all p > 0.05). The following sections present representative results for metabolic, inflammato...

Discussion

With the development of society and the change in dietary patterns, the clinical characteristics of diabetes patients have shown new trends. Among T2DM patients, the traditional “three more and one less” symptoms are no longer common. Data indicate that 44% of T2DM patients did not exhibit typical clinical symptoms, and overweight/obesity has a high prevalence in the T2DM population, often associated with insulin resistance, lipid metabolism disorders, and chronic hyperglycemia³². Metf...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
32G×4 mm disposable needles	BD Medical	328418	For subcutaneous injection of liraglutide; 4 mm length reduces intramuscular injection risk
5810R	Refrigerated centrifuge	Eppendorf	5810R
Adiponectin ELISA kit	R&D Systems	DY1065	For quantitative measurement of human adiponectin in serum; 96-well plate format
AU680	Fully automated biochemical analyzer	Beckman Coulter	AU680
Bluetooth heart rate monitor	Xiaomi	Mi Band 6	For monitoring exercise heart rate; 1 Hz sampling frequency; syncs with research tablet
Chemerin ELISA kit	R&D Systems	DY2325	For detection of human chemerin (RARRES2) in serum; sensitivity <50 pg/mL
Cobas e601	Fully automated chemiluminescence immunoanalyzer	Roche Diagnostics	Cobas e601
C-reactive protein (CRP) ELISA kit	Abcam	ab260078	For high-sensitivity measurement of human CRP; range 0.1–100 ng/mL
Date-labeled medication boxes	MedCenter	M0314	Daily pill organizer with 7 compartments; helps track medication adherence
Dietary Analysis and Meal Planning System	Dietary planning software	Nutritionist Pro	Version 2.0 or higher
EDTA anticoagulant tubes (purple cap)	BD Vacutainer	367861	For HbA1c and complete blood count; 5 mL draw volume; spray-coated EDTA
Electronic weight scale	Tanita	HD-351	For body weight measurement; capacity 150 kg, accuracy 0.1 kg
Exercise log (paper)	In-house	N/A	Custom-designed form to record exercise type, duration, heart rate, and RPE score
Fluoride/oxalate anticoagulant tubes (gray cap)	BD Vacutainer	367835	For glucose measurement; inhibits glycolysis; 3 mL draw volume
Fully automated biochemical analyzer	Beckman Coulter	AU680	For measurement of FPG, 2hPG, TC, TG, HDL-C, LDL-C; throughput up to 800 tests/h
Fully automated chemiluminescence immunoanalyzer	Roche Diagnostics	Cobas e601	For fasting insulin (FINS) measurement; electrochemiluminescence method
G*Power	Heinrich-Heine-Universität Düsseldorf, Germany	RRID:SCR_013726	Free statistical power analysis software. The version used in this study is 3.1.9.6. Official download and information page: https://www.psychologie.hhu.de/arbeitsgruppen/allgemeine-psychologie-und-arbeitspsychologie/gpower. Key reference: Faul, F., Erdfelder, E., Lang, A.-G., & Buchner, A. (2007). G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Methods of Behavioral Research, 39(2), 175-191.
G11	HbA1c analyzer (HPLC)	Tosoh Bioscience	G11
HbA1c analyzer (HPLC)	Tosoh Bioscience	G11	For glycated hemoglobin quantification; NGSP-certified; retention time 2.4 min
IL-6 ELISA kit	R&D Systems	DY206	For human IL-6 detection in serum; sensitivity <0.7 pg/mL
Liraglutide pre-filled injection pen	Novo Nordisk	J20160005	Pre-filled 3 mL (18 mg) pen for subcutaneous injection; delivers 0.6, 1.2, or 1.8 mg/dose
Medication log (paper)	In-house	N/A	Daily record for drug intake, adverse events, and injection site rotation
Mi Band 6	Wearable heart rate monitor	Xiaomi	Mi Band 6
Microplate reader (ELISA)	BioTek	Synergy H1	For absorbance measurement at 450 nm (reference 570 nm); 96-well plate reader
Mobile application (dietary record)	Nutritionist Pro	Mobile App v2.1	For patient self-recording of meals; syncs with nutritionist software
Nutritional assessment software	Nutritionist Pro	v7.0	For calculating BMR, REE, and generating 7-day food exchange plans
Personal protective equipment (lab coat, gloves, goggles, mask)	Various	N/A	BSL-2 requirement; latex gloves, nitrile optional; safety goggles with anti-fog
Refrigerated centrifuge	Eppendorf	5810R	Temperature range -9°C to 40°C; fixed-angle rotor F-34-6-38; used for blood sample centrifugation
Resistin ELISA kit	R&D Systems	DY1359	For detection of human resistin (RETN) in serum; intra-assay CV <5%
SAA ELISA kit	Abcam	ab100634	For human serum amyloid A1 (SAA1) measurement; range 0.15–20 ng/mL
Saxagliptin tablets (5 mg)	Jiangsu Aosaikang Pharmaceutical	H20193008	Oral DPP-4 inhibitor; 5 mg once daily; store at 20-25°C
Serum separator tubes (yellow/red cap)	BD Vacutainer	367988	For serum separation; contains clot activator and gel barrier; 4 mL draw
Sharp container	BD	305790	Puncture-resistant container for safe disposal of needles and lancets
SPSS software	IBM	Version 25.0	For statistical analysis: t-tests, ANOVA, chi-square; Windows 10 compatible
Stadiometer	Seca	213	For height measurement; range 60–200 cm; wall-mounted with movable headpiece
Standard soft tape measure	Seca	201	For waist and hip circumference measurement; accuracy 0.1 cm; non-stretchable
Synergy H1	Microplate reader (ELISA)	BioTek	Synergy H1
TNF-α ELISA kit	R&D Systems	DY210	For human TNF-α detection; sensitivity <2 pg/mL; 96-well plate
Training injection pen (needle-free)	Novo Nordisk	123456 (sample)	Device for patient training; no needle; replicates handling of real injection pen
Vacuum blood collection needle set (21G)	BD Vacutainer	367281	Multi-sample needle; 21G × 1.5 inch; for venipuncture with holder
Vaspin ELISA kit	R&D Systems	DY1206	For human Vaspin (SERPINA12) measurement; assay range 0–20 ng/mL