Abstract
Objective:
Glucagon-like peptide-1 receptor agonists (GLP-1RA) have cardioprotective and kidney-protective benefits among patients with type 2 diabetes (T2D). We sought to determine whether GLP-1RA use improves cardiovascular (CV) and kidney outcomes among patients with systemic lupus erythematosus (SLE) and lupus nephritis (LN).
Methods:
We emulated a pragmatic target trial to evaluate the impact of GLP-1RA vs. comparator hypoglycemic agents, dipeptidyl peptidase 4 inhibitors (DPP4i), on CV and kidney outcomes among patients with SLE and T2D using a large, US multi-center electronic health record database. We used propensity score overlap weighting to emulate randomization between treatment groups. Outcomes included major adverse cardiovascular events, venous thrombosis (VTE), kidney disease progression (eGFR decline ≥ 30% or new-onset end-stage kidney disease), and all-cause mortality. We used Cox regression to compare hazard ratios (HR) based on the weighted populations. In a secondary analysis, we only included patients with LN.
Results:
There were 910 and 1004 initiators of GLP-1RA and DPP4i, respectively, including 267 and 324 patients with LN, respectively. Baseline covariates were balanced after propensity score overlap weighting. The risks of MACE (HR 0.66 [95% CI 0.48-0.91]), VTE (HR 0.49 [0.24-0.97]), kidney disease progression (HR 0.77 [0.60-0.98]), and all-cause mortality (HR 0.26 [CI 0.10-0.68]) were lower with GLP-1RA vs. DPP4i use. GLP-1RA use was similarly associated with lower risks of MACE and kidney disease progression among patients with LN.
Conclusion:
We found lower risks of adverse CV and kidney outcomes and mortality with GLP-1RA use compared with DPP4i use among patients with lupus and T2D.
Introduction
Systemic lupus erythematosus (SLE) is a heterogenous autoimmune disease that disproportionately impacts women of childbearing age and racial and ethnic minority groups, and it is associated with serious morbidity and premature mortality.1 Lupus nephritis (LN) is a severe manifestation of SLE which occurs in up to 50% of patients, placing them at significantly increased risk for chronic kidney disease (CKD) and cardiometabolic disease1,2. Despite recent advances in treatment, patients with lupus nephritis have a 10-30% risk of developing end-stage kidney disease (ESKD). Furthermore, atherosclerotic cardiovascular disease is the most common cause of mortality among patients with SLE, who have over a two-fold higher incidence of myocardial infarction (MI) and cerebrovascular accidents (CVA) than those without SLE. These adverse outcomes are driven by a combination of uncontrolled lupus disease activity, chronic tissue damage including proteinuric nephropathy, and treatment-related glucocorticoid toxicity, which exacerbates conventional cardiovascular risk factors. Additional strategies are needed to mitigate the profound morbidity associated with cardiovascular disease and kidney disease progression.
Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) are a class of anti-obesity and hypoglycemic agents that have been found to prevent major adverse cardiac events (MACE) in patients with overweight/obesity and type 2 diabetes (T2D), and to reduce CKD progression in patients with diabetic nephropathy3-5. The emerging evidence of nephroprotective benefits has demonstrated proteinuria reduction and a reduction in estimated glomerular filtration rate (eGFR) decline with GLP-1 RA use.5,6 We postulated that GLP-1RAs, may have similar cardioprotective and nephroprotective effects among patients with SLE and LN.
We utilized a large, multi-center United States electronic health record (EHR) database to emulate a pragmatic target trial7 to evaluate whether GLP-1RAs improve kidney and cardiovascular outcomes in patients with SLE and T2D, compared to dipeptidyl peptidase 4 inhibitors (DPP4i), a comparator class of hypoglycemic agents. DPP4i have been found to have a neutral impact on the risk of cardiovascular events and have not demonstrated nephroprotective benefits.8
Patients and Methods
Data Source and Study Design
We identified patients ≥18 years of age who met administrative definition criteria for SLE and T2D in TriNetX, a federated EHR database consisting of patients from academic medical centers, community hospitals, and outpatient clinics in the United States. This database includes demographics, biometrics, medications, laboratory tests, procedures, clinical encounters, diagnostic and procedure codes, and vital status, as previously described.9 Patients were identified as having SLE based on the presence of at least two corresponding International Classification of Diseases ninth (ICD-9) or tenth (ICD-10) revision codes (ICD-9 710.9 or ICD-10 M32.* excluding M32.0) at least 30 days, but no more than two years, apart. Patients with SLE were further characterized as having LN if they had ≥1 ICD-10 code for lupus nephritis (M32.14 or M32.15) or ≥2 nephritis-related codes (ICD-9 580-586, 791.0; ICD-10 N00, N03, N04, N05, N17, N18, R80.9).
We specified and emulated the protocol of a pragmatic target trial to compare the impact of GLP-1RA and DPP4i use on cardiovascular and kidney outcomes among patients with SLE and T2D and among patients with LN and T2D (Supplemental Table 1).7
This study was approved by the Mass General Brigham Institutional Review Board, and informed consent was waived.
Eligibility Criteria and Treatment Assignment
We identified patients with SLE who initiated a GLP-1RA or DPP4i between October 2006, when the first DPP4i was approved, and August 2021, the end of available follow-up data. GLP-1RAs included exenatide, albiglutide, dulaglutide, liraglutide, and semaglutide, and DPP4is included sitagliptin, saxagliptin, linagliptin, and alogliptin. Of note, the GLP-1RA albiglutide was discontinued in 2017, and tirzepatide, a glucose-dependent insulinotropic polypeptide and GLP-1RA, was not included since it was approved by the Food and Drug Administration (FDA) in 2022. During this study period, GLP-1RAs or DPP4is were only approved by the FDA for the treatment of T2D, and we excluded patients without a diagnosis of T2D. Patients were excluded from the study if they used the comparator medication or a sodium-glucose cotransporter-2 inhibitor (SGLT2i) in the year prior to the index date of treatment initiation, had a diagnosis of ESKD or type 1 diabetes prior to the index date, or had missing covariate data, except for hemoglobin A1C (HgA1C), which was imputed.
We classified individuals into the treatment groups, initiation of GLP-1RA or DPP4i, according to the strategy that their data were compatible with at baseline. We used propensity score overlap weighting to emulate randomization of treatment assignment and balance potential confounders.
Covariates
Covariates were assessed in the 12 months prior to the index date and included demographics, geographic region, comorbidities including CKD, heart failure, cardiovascular disease, tobacco use, obesity, and the Charlson comorbidity index10, medications for SLE and T2D, and healthcare utilization, including outpatient visits and emergency room or inpatient encounters. We also assessed the presence of LN, SLE severity assessed using a severity index modified from an administrative algorithm based on ICD codes and medication use9,11, diabetic complications, and hemoglobin A1c. We used multiple imputation for missing values of hemoglobin A1c.
Outcomes
Our primary outcomes included major adverse cardiac events (MACE) and kidney disease progression. MACE was defined as a composite of ischemic stroke, myocardial infarction, heart failure, and cardiovascular death. MACE events were identified by ICD codes occurring during inpatient encounters. The definition of cardiovascular death was adapted from a validated claims-based algorithm and included death in the month of an inpatient encounter for cardiovascular disease as well as out of hospital death without a diagnosis of cancer in the 365 days prior to and including the date of death and without diagnoses of infection or trauma in the 45 days prior to and including the date of death.12 Kidney disease progression was defined as a decrease in estimated glomerular filtration rate (eGFR) by >30% or new-onset ESKD. We also assessed the individual components of MACE, venous thromboembolism (VTE, including pulmonary embolism or deep vein thrombosis), all-cause death, and a control outcome of genital infection, which was expected to be null.
Statistical Analysis
Descriptive statistics of baseline characteristics between the two treatment groups were performed before and after propensity score overlap weighting and compared using standardized mean differences. We conducted a primary as-treated analysis where patients were followed from the index date of treatment initiation until the outcome of interest, disenrollment from the database, death, discontinuation of assigned treatment, initiation of the comparator treatment, or end of the study period (August 2021). We conducted a secondary intention-to-treat analysis, where patients were followed until the earliest of the outcome, death, or end of the study period.
We conducted survival analyses with Cox proportional hazard regression to compare incidence rates for each outcome and assess hazard ratios with 95% confidence intervals (CIs) for the weighted populations. We also performed a secondary analysis restricted to patients with LN at baseline. We had full data access, and all analyses were conducted using SAS 9.4 (SAS Institute Inc., Cary, NC, USA).
Results
Baseline Characteristics
There were 910 initiators of GLP-1RA and 1004 initiators of DPP4i with SLE and T2D, of which 267 and 324 patients had LN, respectively (Table 1). After propensity score overlap weighting, the treatment groups were balanced across all covariates. Patients were mostly female (92%) with a mean age of 55 years. Nearly half of patients were non-Hispanic White, and 35% were Black. Nearly half used glucocorticoids, and 38% used an angiotensin converting enzyme inhibitor (ACEi) or angiotensin receptor blocker (ARB). Additionally, 33% of patients had chronic kidney disease (CKD) stage ≥ 3, 15% had heart failure, and 35% were obese. The average hemoglobin A1C was 7.6%.
Table 1.
Baseline Characteristics of GLP-1 Receptor Agonist and DPP4 Inhibitor Initiators with Systemic Lupus Erythematosus and Diabetes
| Before Propensity Score Overlap Weighting |
After Propensity Score Overlap Weighting |
|||||
|---|---|---|---|---|---|---|
| GLP-1RA | DPP4i | Std. Diff. | GLP-1RA | DPP4i | Std. Diff. | |
| N | 910 | 1004 | ||||
| Age, mean (SD) | 53.0 (11.3) | 58.2 (12.0) | 0.45 | 55.0 | 55.0 | <0.01 |
| Female, n (%) | 841 (92.4) | 918 (91.4) | 0.04 | 92.3 | 92.3 | <0.01 |
| Race and Ethnicity, n (%) | 0.12 | <0.01 | ||||
| Asian | 6 (0.7) | 19 (1.9) | 0.9 | 0.9 | ||
| Hispanic | 71 (7.8) | 104 (10.4) | 9.0 | 9.0 | ||
| Black | 332 (36.5) | 335 (33.4) | 34.7 | 34.7 | ||
| Other | 63 (6.9) | 67 (6.7) | 7.2 | 7.2 | ||
| White, Non-Hispanic | 438 (48.1) | 479 (47.7) | 48.2 | 48.2 | ||
| Geographic Region, n (%) | 0.08 | <0.01 | ||||
| East | 202 (22.2) | 200 (19.9) | 20.7 | 20.7 | ||
| Midwest | 168 (18.5) | 159 (15.8) | 17.0 | 17.0 | ||
| South | 461 (50.7) | 546 (54.4) | 53.5 | 53.5 | ||
| West | 79 (8.7) | 99 (9.9) | 8.9 | 8.9 | ||
| Lupus nephritis, n (%) | 267 (29.3) | 324 (32.3) | 0.06 | 29.3 | 29.3 | <0.01 |
| SLE Severity Index, n (%) | 0.03 | <0.01 | ||||
| Mild | 317 (34.8) | 343 (34.2) | 35.8 | 35.8 | ||
| Moderate | 375 (41.2) | 415 (41.3) | 40.4 | 40.4 | ||
| Severe | 218 (24.0) | 246 (24.5) | 23.9 | 23.9 | ||
| Charlson Comorbidity Index, mean (SD) | 1.5 (1.4) | 1.8 (1.8) | 0.18 | 1.6 | 1.6 | <0.01 |
| Comorbidities, n (%) | ||||||
| CKD stage ≥3 | 293 (32.2) | 366 (36.5) | 0.09 | 32.7 | 32.7 | <0.01 |
| Heart Failure | 118 (13.0) | 181 (18.0) | 0.14 | 14.9 | 14.9 | <0.01 |
| Cardiovascular disease | 212 (23.3) | 238 (23.7) | 0.01 | 23.1 | 23.1 | <0.01 |
| Obesity | 416 (45.7) | 275 (27.4) | 0.39 | 35.4 | 35.4 | <0.01 |
| Diabetic complications | ||||||
| Diabetic nephropathy | 134 (14.7) | 154 (15.3) | 0.02 | 13.7 | 13.7 | <0.01 |
| Diabetic retinopathy | 90 (9.9) | 61 (6.1) | 0.14 | 7.4 | 7.4 | <0.01 |
| Diabetic neuropathy | 246 (27.0) | 199 (19.8) | 0.17 | 21.8 | 21.8 | <0.01 |
| Diabetic vascular disease | 61 (6.7) | 53 (5.3) | 0.06 | 5.1 | 5.1 | <0.01 |
| Tobacco use, n (%) | 246 (27.0) | 246 (24.5) | 0.06 | 24.8 | 24.8 | <0.01 |
| SLE Medication use, n (%) | ||||||
| Glucocorticoids | 468 (51.4) | 498 (49.6) | 0.04 | 49.4 | 49.4 | <0.01 |
| Hydroxychloroquine | 366 (40.2) | 351 (35.0) | 0.11 | 38.3 | 38.3 | <0.01 |
| Azathioprine | 65 (7.1) | 60 (6.0) | 0.05 | 6.1 | 6.1 | <0.01 |
| Methotrexate | 85 (9.3) | 80 (8.0) | 0.05 | 8.6 | 8.6 | <0.01 |
| Mycophenolate | 93 (10.2) | 112 (11.2) | 0.0303 | 10.6 | 10.6 | <0.01 |
| Belimumab | 32 (3.5) | 14 (1.4) | 0.1374 | 2.1 | 2.1 | <0.01 |
| Other immunosuppressant | 80 (8.8) | 79 (7.9) | 0.0334 | 8.6 | 8.6 | <0.01 |
| Other Medications, n (%) | <0.01 | |||||
| ACEi or ARB | 366 (40.2) | 407 (40.5) | 0.0065 | 38.4 | 38.4 | <0.01 |
| Insulin use | 356 (39.1) | 324 (32.3) | 0.1433 | 32.9 | 32.9 | <0.01 |
| Metformin use | 356 (39.1) | 368 (36.7) | 0.0509 | 35.2 | 35.2 | <0.01 |
| Sulfonylurea use | 109 (12.0) | 170 (16.9) | 0.1412 | 13.0 | 13.0 | <0.01 |
| Other hypoglycemic medication | 28 (3.1) | 40 (4.0) | 0.0492 | 3.0 | 3.0 | <0.01 |
| eGFR, mean (SD) | 81.5 (27.0) | 74.7 (28.6) | 0.2416 | 79.0 | 79.0 | <0.01 |
| Healthcare Utilization | ||||||
| Outpatient visits, median (IQR) | 9 (2-17) | 7 (1-17) | 0.1010 | 7 (2-17) | 7 (2-17) | <0.01 |
| ER/Inpatient visits, n (%) | 332 (36.5) | 415 (41.3) | 0.0996 | 37.8 | 37.8 | <0.01 |
| Hemoglobin A1c (%) | 7.8 (2.5) | 7.7 (3.0) | 0.0310 | 7.6 | 7.6 | <0.01 |
CKD, chronic kidney disease; CVD, cardiovascular disease; ACEi, angiotensin converting enzyme inhibitor; ARB, angiotensin receptor blocker; eGFR, estimated glomerular filtration rate; IQR (inter-quartile range); ER, emergency room
Outcomes in all Patients with Systemic Lupus Erythematosus
In the primary per-protocol analysis, GLP-1RA initiators with SLE had lower risks of MACE (adjusted hazard ratio [aHR] 0.66 [95% CI 0.48-0.91]) and kidney disease progression (aHR 0.77 [95% CI 0.60-0.98]) compared to initiators of DPP4i (Table 2). There was no significant difference in the aHR for individual components of myocardial infarction, ischemic stroke, or heart failure hospitalization, but the incidence of myocardial infarction and heart failure tended to be lower with GLP-1 RA use. The risk of VTE was lower with GLP-1RA than DPP4i initiation (aHR 0.49 [95% CI 0.24-0.97]), as was the risk of all-cause death during the study period. (aHR 0.26 [95% CI 0.10-0.68]). There was no difference in the risk of the control outcome, genital infection, between the treatment groups, as expected.
Table 2.
Cardiovascular and Kidney Outcomes Associated with GLP-1 Receptor Agonist versus DPP4 Inhibitor Use in Systemic Lupus Erythematosus
| Events, n | Follow-up Time, Years |
Incidence Rate (per 1000 Person Years) |
Hazard Ratio (95% CI) |
||||
|---|---|---|---|---|---|---|---|
| GLP-1RA | DPP4i | GLP-1RA | DPP4i | GLP-1RA | DPP4i | ||
| Per-Protocol | |||||||
| Cardiovascular outcomes | |||||||
| MACE | 32 | 42 | 1.3 | 1.2 | 61.1 | 92.9 | 0.66 (0.48-0.91) |
| Myocardial infarction | 3 | 6 | 1.4 | 1.3 | 5.6 | 12.8 | 0.43 (0.16-1.14) |
| Stroke | 8 | 6 | 1.4 | 1.3 | 13.9 | 12.5 | 1.08 (0.51-2.31) |
| Heart Failure | 23 | 26 | 1.4 | 1.2 | 44.1 | 55.7 | 0.78 (0.53-1.15) |
| VTE | 7 | 13 | 1.4 | 1.2 | 12.6 | 26.5 | 0.49 (0.24-0.97) |
| Kidney outcome | |||||||
| eGFR decline by ≥30% or new ESKD | 58 | 68 | 1.3 | 1.1 | 117.5 | 158.1 | 0.77 (0.60-0.98) |
| All-cause death | 3 | 11 | 1.4 | 1.3 | 5.6 | 21.6 | 0.26 (0.10-0.68) |
| Control outcome | |||||||
| Genital Infection | 27 | 24 | 1.3 | 1.2 | 52.2 | 52.4 | 1.02 (0.69-1.50) |
| Intention-to-treat | |||||||
| Cardiovascular outcomes | |||||||
| MACE | 54 | 63 | 2.7 | 2.5 | 51.3 | 64.7 | 0.80 (0.62-1.03) |
| Myocardial infarction | 10 | 11 | 3.0 | 2.8 | 8.5 | 10.3 | 0.81 (0.44-1.47) |
| Stroke | 13 | 13 | 3.0 | 2.8 | 11.1 | 11.6 | 0.99 (0.57-1.72) |
| Heart Failure | 40 | 41 | 2.8 | 2.7 | 36.9 | 40.0 | 0.93 (0.69-1.26) |
| VTE | 10 | 19 | 3.0 | 2.8 | 8.9 | 17.5 | 0.53 (0.30-0.94) |
| Kidney outcome | |||||||
| eGFR decline by ≥30% or new ESKD | 92 | 108 | 2.5 | 2.2 | 95.2 | 124.3 | 0.79 (0.65-0.96) |
| All-cause death | 10 | 22 | 3.0 | 2.9 | 8.4 | 19.5 | 0.42 (0.25-0.73) |
| Control outcome | |||||||
| Genital Infection | 46 | 44 | 2.7 | 2.6 | 42.8 | 43.8 | 1.01 (0.75-1.36) |
MACE, major adverse cardiac events; VTE, venous thromboembolism; eGFR, estimated glomerular filtration rate; ESKD, end-stage kidney disease. MACE includes myocardial infarction, ischemic stroke, or heart failure hospitalization.
Outcomes in Patients with Lupus Nephritis
In the secondary analysis of patients with LN, GLP-1RA initiation compared to DPP4i use was similarly associated with lower risks of MACE (aHR 0.64 [95% CI 0.41-0.98]) and kidney disease progression (aHR 0.70 [95% CI 0.49-1.00], Table 3). The effect sizes for VTE and all-cause death did not reach statistical significance in the LN subgroup.
Table 3.
Cardiovascular and Kidney Outcomes Associated with GLP-1 Receptor Agonists versus DPP4 Inhibitor Use in Lupus Nephritis
| Outcomes | Events, n | Follow-up Time, Years |
Incidence Rate (per 1000 Person Years) |
Hazard Ratio (95% CI) |
|||
|---|---|---|---|---|---|---|---|
| GLP-1RA | DPP4i | GLP-1RA | DPP4i | GLP-1RA | DPP4i | ||
| Per-Protocol | |||||||
| Cardiovascular outcomes | |||||||
| MACE | 18 | 23 | 1.2 | 1.0 | 122.2 | 193.0 | 0.64 (0.41-0.98) |
| Myocardial infarction | 2 | 5 | 1.4 | 1.2 | 11.0 | 32.7 | 0.34 (0.10-1.09) |
| Stroke | 2 | 4 | 1.4 | 1.2 | 14.7 | 25.3 | 0.58 (0.18-1.85) |
| Heart Failure | 15 | 16 | 1.2 | 1.1 | 98.5 | 124.9 | 0.78 (0.48-1.28) |
| VTE | 4 | 5 | 1.3 | 1.2 | 24.9 | 38.3 | 0.62 (0.25-1.56) |
| Kidney outcome | |||||||
| eGFR decline by ≥30% or new ESKD | 28 | 35 | 1.1 | 0.9 | 213.3 | 325.2 | 0.70 (0.49-1.00) |
| All-cause death | 3 | 5 | 1.4 | 1.2 | 16.2 | 36.5 | 0.46 (0.16-1.37) |
| Control outcome | |||||||
| Genital Infection | 9 | 8 | 1.3 | 1.1 | 56.6 | 62.9 | 0.90 (0.48-1.68) |
| Intention-to-treat | |||||||
| Cardiovascular outcomes | |||||||
| MACE | 25 | 30 | 2.1 | 1.9 | 97.7 | 132.7 | 0.75 (0.52-1.09) |
| Myocardial infarction | 3 | 8 | 2.5 | 2.3 | 11.1 | 29.0 | 0.39 (0.17-0.91) |
| Stroke | 4 | 5 | 2.5 | 2.3 | 12.9 | 18.8 | 0.70 (0.28-1.73) |
| Heart Failure | 20 | 23 | 2.2 | 2.0 | 79.2 | 93.7 | 0.85 (0.56-1.29) |
| VTE | 5 | 7 | 2.4 | 2.3 | 16.3 | 25.5 | 0.66 (0.29-1.51) |
| Kidney outcomes | |||||||
| eGFR decline by ≥30% or new ESKD | 38 | 52 | 1.9 | 1.5 | 171.8 | 285.0 | 0.63 (0.47-0.85) |
| All-cause death | 4 | 11 | 2.5 | 2.4 | 12.2 | 39.3 | 0.32 (0.13-0.79) |
| Control outcome | |||||||
| Genital Infection | 10 | 12 | 2.3 | 2.2 | 36.6 | 48.8 | 0.78 (0.44-1.39) |
MACE, major adverse cardiac events; VTE, venous thromboembolism; eGFR, estimated glomerular filtration rate; ESKD, end-stage kidney disease. MACE outcome includes combined myocardial infarction, ischemic stroke, or heart failure hospitalization. Heart failure outcome includes heart failure hospitalization.
In the intention-to-treat analyses, the direction of effect of GLP-1RA use was similar across outcomes with some attenuation, with lower risks of kidney disease progression and all-cause death with GLP-1 RA initiation compared with DPP4i initiation for all patients with SLE (progression of kidney disease aHR 0.79 [0.65-0.96], all-cause death aHR 0.42 [0.25-0.73]) and patients with LN (progression of kidney disease aHR 0.63 [0.47-0.85], all-cause death aHR 0.32 [0.13-0.79]).
Discussion
This emulated target trial using observational data demonstrated that GLP-1RA use was associated with a lower risk of MACE, kidney disease progression, VTE, and all-cause death compared with DPP4i use among patients with SLE and comorbid T2D. Similar benefits were also demonstrated among patients with LN. This is the first observational study, to our knowledge, demonstrating the potential cardioprotective, nephroprotective, and mortality benefits of GLP-1RAs for patients with SLE and LN, who have been largely excluded from participating in randomized clinical trials for GLP-1RAs3,5.
It is widely established that GLP-1RAs reduce cardiovascular risk and attenuate progression of kidney disease among those with type II diabetes3,5,13,14; however, recent studies suggest GLP-1RAs also confer similar benefits to other populations with elevated risks of CV events and/or progressive CKD4,6. In the SELECT trial, which included patients with at least one CV risk factor, a body mass index (BMI) of 27 or greater, and no prior diagnosis of diabetes, the use of the GLP-1 RA semaglutide, compared to placebo, reduced the incidence of death from CV disease, nonfatal myocardial infarction, or nonfatal stroke4. Many patients with SLE have traditional CV risk factors, in addition to the elevated CV risk related to SLE itself, so many patients will now have indications for GLP-1RA use and may have considerable potential benefit. Although our emulated trial could not isolate the effect of GLP-1RA on patients with SLE without T2D, we observed substantial reductions in the risk of MACE for patients with both conditions. Additionally, the kidney-protective effects of GLP-1RA likely extend beyond treatment of patients with diabetic nephropathy. The SMART trial recently demonstrated that weekly administration of semaglutide led to appreciable reduction in albuminuria in a cohort of patients with non-diabetic CKD, including 25 patients with chronic glomerulonephritis.6 Persistent proteinuria is a major risk factor for progressive CKD in LN; therefore, reducing proteinuria is a potential mechanism of benefit for GLP-1RA for patients with LN. We found a reduction in eGFR decline/new onset ESKD in patients with LN and T2D. Future studies are needed to determine the impact of GLP-1 RA on CV events, proteinuria, and kidney disease progression among patients with SLE without T2D. Further, future studies should assess the impact GLP-1 RA use for patients undergoing standard immunosuppression treatment for lupus nephritis.
While research is ongoing to further identify specific mechanisms contributing to the cardio- and kidney-protective properties of GLP1-RAs aside from glucose-lowering and hemodynamic effects, studies in murine models and humans suggest these agents potentially possess immunomodulatory properties15. GLP1-RAs have been shown in murine models to regulate lymphocyte proliferation and maintenance of regulatory T cells, and studies in humans have demonstrated that GLP-1RAs decrease production of inflammatory cytokines and oxidative stress biomarkers.15,16 It remains unclear to what extent these mechanisms are mediated through weight reduction or separate, direct processes. However, these posited immune mechanisms of GLP-1RAs warrant further investigation in patients with SLE and LN.
This study is strengthened by the employment of the emulated target trial framework to assess causal associations from observational data, with the specific use of propensity score overlap weighting to mitigate confounding. Limitations of our study include the relatively small sample size and the potential for misclassification that is inherent to observational, EHR data. This may include misclassification of medication exposure, as we lacked pharmacy dispensing records and could not assess medication adherence to prescribed treatment. Lastly, we note that the inability to investigate the efficacy of GLP-1RA among SLE patients without T2D serves as another limitation.
To conclude, our findings suggest that GLP-1RAs confer substantial cardioprotective and kidney-protective benefits to patients with SLE and LN with comorbid T2D, who are inherently predisposed to a markedly disproportionate risk of atherosclerotic CV disease and chronic kidney disease. These support a potential role for GLP1-RAs in the management and treatment of patients with SLE and LN, although additional studies are needed to assess the safety of GLP-1RAs in this population. Furthermore, patients with SLE should be included in future randomized-trials involving GLP-1RAs, and future studies should evaluate the efficacy of GLP-1RA use among patients with SLE and LN without concomitant diabetes. Potential barriers to accessing these medications, including cost, should also be addressed to avoid exacerbating healthcare disparities. Lastly, basic science and translational studies are needed to help provide mechanistic insight for these findings, which may elucidate novel indications and benefits.
Supplementary Material
Financial Support:
AJ is supported by the National Institutes of Health (NIH)/National Institutes of Arthritis and Musculoskeletal Diseases (NIAMS) K23-AR-079040 and the Rheumatology Research Foundation K Supplement. HKC is supported by NIAMS/NIH P50 AR060772 and R01-AR-065944.
Footnotes
Conflicts of Interest:
The authors declare no conflicts of interest.
References
- 1.Mahajan A, Amelio J, Gairy K, et al. Systemic lupus erythematosus, lupus nephritis and end-stage renal disease: a pragmatic review mapping disease severity and progression. Lupus 2020;29(9):1011–1020. (In eng). DOI: 10.1177/0961203320932219. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Wong CY, Ma BMY, Zhang D, Cheung W, Chan TM, Yap DYH. Cardiovascular risk factors and complications in patients with systemic lupus erythematosus with and without nephritis: a systematic review and meta-analysis. Lupus Sci Med 2024;11(1) (In eng). DOI: 10.1136/lupus-2024-001152. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Sattar N, Lee MMY, Kristensen SL, et al. Cardiovascular, mortality, and kidney outcomes with GLP-1 receptor agonists in patients with type 2 diabetes: a systematic review and meta-analysis of randomised trials. Lancet Diabetes Endocrinol 2021;9(10):653–662. (In eng). DOI: 10.1016/s2213-8587(21)00203-5. [DOI] [PubMed] [Google Scholar]
- 4.Lincoff AM, Brown-Frandsen K, Colhoun HM, et al. Semaglutide and Cardiovascular Outcomes in Obesity without Diabetes. N Engl J Med 2023;389(24):2221–2232. (In eng). DOI: 10.1056/NEJMoa2307563. [DOI] [PubMed] [Google Scholar]
- 5.Perkovic V, Tuttle KR, Rossing P, et al. Effects of semaglutide on chronic kidney disease in patients with type 2 diabetes. New England Journal of Medicine 2024;391(2):109–121. [DOI] [PubMed] [Google Scholar]
- 6.Apperloo EM, Gorriz JL, Soler MJ, et al. Semaglutide in patients with overweight or obesity and chronic kidney disease without diabetes: a randomized double-blind placebo-controlled clinical trial. Nat Med 2024. (In eng). DOI: 10.1038/s41591-024-03327-6. [DOI] [PubMed] [Google Scholar]
- 7.Hernán MA, Robins JM. Using Big Data to Emulate a Target Trial When a Randomized Trial Is Not Available. Am J Epidemiol 2016;183(8):758–64. (In eng). DOI: 10.1093/aje/kwv254. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Giugliano D, Longo M, Signoriello S, et al. The effect of DPP-4 inhibitors, GLP-1 receptor agonists and SGLT-2 inhibitors on cardiorenal outcomes: a network meta-analysis of 23 CVOTs. Cardiovasc Diabetol 2022;21(1):42. (In eng). DOI: 10.1186/s12933-022-01474-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Materne E, Choi H, Zhou B, Costenbader KH, Zhang Y, Jorge A. Comparative Risks of Infection with Belimumab versus Oral Immunosuppressants in Patients with Non-Renal Systemic Lupus Erythematosus. Arthritis & rheumatology (Hoboken, NJ) 2023. (In eng). DOI: 10.1002/art.42620. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987;40(5):373–83. (In eng). DOI: 10.1016/0021-9681(87)90171-8. [DOI] [PubMed] [Google Scholar]
- 11.Garris C, Jhingran P, Bass D, Engel-Nitz NM, Riedel A, Dennis G. Healthcare utilization and cost of systemic lupus erythematosus in a US managed care health plan. J Med Econ 2013;16(5):667–77. (In eng). DOI: 10.3111/13696998.2013.778270. [DOI] [PubMed] [Google Scholar]
- 12.Lix LM, Sobhan S, St-Jean A, et al. Validity of an algorithm to identify cardiovascular deaths from administrative health records: a multi-database population-based cohort study. BMC Health Serv Res 2021;21(1):758. (In eng). DOI: 10.1186/s12913-021-06762-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Rossing P, Bain SC, Bosch-Traberg H, et al. Effect of semaglutide on major adverse cardiovascular events by baseline kidney parameters in participants with type 2 diabetes and at high risk of cardiovascular disease: SUSTAIN 6 and PIONEER 6 post hoc pooled analysis. Cardiovasc Diabetol 2023;22(1):220. (In eng). DOI: 10.1186/s12933-023-01949-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Perkovic V, Tuttle KR, Rossing P, et al. Effects of Semaglutide on Chronic Kidney Disease in Patients with Type 2 Diabetes. N Engl J Med 2024. (In eng). DOI: 10.1056/NEJMoa2403347. [DOI] [PubMed] [Google Scholar]
- 15.Drucker DJ. The cardiovascular biology of glucagon-like peptide-1. Cell metabolism 2016;24(1):15–30. [DOI] [PubMed] [Google Scholar]
- 16.Mehdi SF, Pusapati S, Anwar MS, et al. Glucagon-like peptide-1: a multi-faceted anti-inflammatory agent. Front Immunol 2023;14:1148209. (In eng). DOI: 10.3389/fimmu.2023.1148209. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.