Effects of Therapies on Proteinuria and eGFR in IgA Nephropathy

Visual Abstract

Abstract

Key Points

• Existing and novel therapies reduce proteinuria and slow eGFR rate of decline in IgA nephropathy, with some heterogeneity across drug classes.

• Benefits and limitations of each drug class must be considered, and longer-term Phase 3 data on several therapies are awaited.

• Emerging data support the potential of novel therapies to improve kidney outcomes in IgA nephropathy in the near future.

Background

Substantial advances have been made in therapeutics for IgA nephropathy. We conducted a systematic review and meta-analysis to evaluate the comparative efficacy and safety of existing and novel IgA nephropathy therapies.

Methods

We searched MEDLINE and Embase databases from inception to May 21, 2025, for Phase 2b and 3 multicenter, randomized, placebo-controlled trials enrolling patients with IgA nephropathy that reported treatment effects on proteinuria and/or eGFR slope. Trials were categorized into four drug classes: Nonimmunologic therapies, corticosteroids, B-cell modulating agents, and complement inhibitors. Treatment effects on proteinuria and eGFR slope were pooled overall and by drug class using random-effects meta-analysis.

Results

Fourteen trials were identified of which proteinuria and eGFR outcomes were available in 13 trials (93%) and seven trials (50%), respectively. Pooled data demonstrated all therapies reduced proteinuria, although the magnitude of effect varied across classes: −34% with nonimmunologic therapies, −51% with corticosteroids, −45% with B-cell modulating agents, and −35% with complement inhibitors (P heterogeneity <0.001). Data from trials reporting eGFR slope over a minimum of 12 months indicated benefits for all drug classes, but again with some evidence that effects varied by class. The absolute and relative effect on eGFR slope was 1.1 ml/min per 1.73 m² per year and −28% with nonimmunologic therapies, 2.3 ml/min per 1.73 m² per year and −52% with corticosteroids, and 4.3 ml/min per 1.73 m² per year and −73% with B-cell modulating agents (P heterogeneity = 0.03). Corticosteroids, particularly at higher doses, increased the risk of serious adverse events, but other drug classes were generally well tolerated. Longer-term data on clinical kidney outcomes and safety are awaited.

Conclusions

All four drug classes improve kidney outcomes in IgA nephropathy, with some evidence of differential effects on proteinuria and eGFR slope. The varying mechanisms and effects of different therapies suggest a potential for combination therapy, although selection of the optimal combination of therapies for individuals remain to be determined.

Introduction

Immunoglobulin A (IgA) nephropathy is the most common primary glomerular disease worldwide.¹ Historically, the focus of treatment has centered on maximal supportive care with renin-angiotensin-system (RAS) blockade. The 2021 Kidney Disease Improving Global Outcomes guidelines suggest treatment with systemic corticosteroids in a select cohort of patients with IgA nephropathy at high risk of kidney failure despite 3–6 months of optimal supportive care.² Until recently, there has been a paucity of effective targeted therapeutic agents to delay disease progression.

Significant advancements have since been made in the understanding and treatment of IgA nephropathy, driven by insights into its multihit immunogenic pathogenesis and implementation of more efficient clinical trial design and conduct. Large trial data support the use of change in proteinuria as a surrogate end point to evaluate treatment efficacy in IgA nephropathy, with eGFR slope, typically over 2 years, as a confirmatory end point.^3,4 Consequently, these end points have been accepted by regulatory agencies for use in late phase drug registration trials to obtain accelerated regulatory approval under the rare and orphan diseases category. The use of 9-month proteinuria reduction and 2-year eGFR slope as primary efficacy end points has thus revolutionized the therapeutic pipeline for IgA nephropathy, substantially reducing study duration and sample size, and leading to the exponential rise of novel therapeutic agents.^5–7

In the past 3 years, the Therapeutic Effect of Steroids in IgA Nephropathy Global (TESTING) study of corticosteroids has been reported⁵; four novel therapies have received full or conditional approval for use in IgA nephropathy (targeted release formulation [TRF]-budesonide, sparsentan, atrasentan and iptacopan)^6–9; various other agents have demonstrated promising effects on proteinuria and eGFR in Phase 2 trials,^10–12 and almost a dozen Phase 3 trials have commenced.^13,14

We therefore conducted a systematic review and meta-analysis to quantitatively summarize the effect of these agents on proteinuria and eGFR slope compared with supportive care in individuals with IgA nephropathy.

Methods

Data Sources and Study Selection

We conducted and reported this systematic review and meta-analysis according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis statement.¹⁵ We identified studies through a systematic literature search of MEDLINE and Embase databases from inception to May 21, 2025. The search strategy was based on terms related to IgA nephropathy and randomized controlled trials. Full details for the search strategy are provided in Supplemental Table 1. Where available, personal enquiry was sought to obtain additional unpublished data.

We included multicenter Phase 2b and 3 randomized controlled trials which (1) included adults with IgA nephropathy established on supportive care including maximally tolerated RAS blockade, (2) reported treatment effects in dedicated IgA nephropathy cohorts, (3) reported the relative change in proteinuria from baseline to 6-to-12 months, and/or (4) reported eGFR slope over a minimum 12-month period. Where there was overlap in the cohort analyzed across two or more publications, we used data from the most contemporary analysis. When an investigational therapy had progressed from Phase 2 to Phase 3 evaluation, we used data from the larger, more contemporary population to focus on the most reliable and stable treatment effect. We excluded single-center trials in select populations, and those evaluating the effect of mycophenolate mofetil, leflunomide, hydroxychloroquine, fostamatinib, and tonsillectomy because of insufficient trial size and duration. We also excluded trials in which the control arm received an active comparator outside of standard of care. Trials were categorized into four drug classes according to the mechanism of action: Nonimmunologic therapies, corticosteroids, B-cell modulating agents, and complement inhibitors.

Data Collection and Quality Assessment

Two authors (Kim and Neuen) independently screened and reviewed titles, abstracts, and relevant full-text articles to identify eligible studies. Data extraction was conducted independently by two authors (Kim and Neuen) using a standardized form. Discrepancies in abstract screening and extracted data were resolved through consensus discussion with a third author (Wong). Study quality was assessed independently by two authors (Kim and Neuen) using version 2 of the Cochrane risk of bias tool.

Data Synthesis and Statistical Analyses

We compared the inclusion and exclusion criteria, baseline characteristics (proportion of women, race, mean age, median proteinuria, mean eGFR, time between kidney biopsy to randomization, proportion of RAS inhibitor use), and mean follow-up duration across included trials.

For each study, we extracted (1) the percentage change in proteinuria from baseline to 6 to 12 months with treatment versus control, (2) the difference in annualized rate of eGFR decline (total eGFR slope) between treatment and control arms, and (3) safety data including serious adverse events (SAEs) and adverse events leading to treatment discontinuation. We aimed to pool effects on proteinuria at 9 months and eGFR slope at 2 years to align with the US Food and Drug Administration (FDA) guidance for the evaluation of therapies in IgA nephropathy. Where 6- to 12-month proteinuria data were not reported, we used the treatment effect over the duration of the trial to make maximal use of available randomized data. Where effects on proteinuria were reported at multiple time points, we prioritized the same time point within each therapeutic class to minimize bias. Where annualized rate of eGFR decline was not reported, we divided the overall rate of eGFR decline by the follow-up duration in years. As total eGFR slope is an absolute measure of effect dependent on the background rate of eGFR decline which varied across the studied populations, we also summarized the relative difference in eGFR slope between treatment and control arms, expressed as a percentage. This was done by dividing the absolute effect and its 95% confidence interval (CI) by the eGFR slope in the control arm as has been done in previous trials and large-scale collaborative meta-analysis.^16,17 Where data were presented graphically without corresponding numerical effect size and/or 95% CIs, we used web-based image extraction software, WebPlotDigitizer,¹⁸ to estimate treatment effects.

For the quantitative synthesis, where both Phase 2 and Phase 3 data were available, we used Phase 3 data for the most reliable estimate of treatment effect on proteinuria and eGFR slope. Phase 2 data were additionally summarized descriptively when the corresponding information was not reported in the Phase 3 trial. Treatment effects on the change in proteinuria and absolute and relative difference in eGFR slope were pooled overall and by drug class using random-effects meta-analysis. I² statistics and P heterogeneity values obtained from the same model were used to assess between-study variability. Heterogeneity between drug classes was evaluated using a Q-test. All analyses were performed using R (version 4.1.3; R Foundation for Statistical Computing, Vienna, Austria).

Results

In total, we identified 648 articles, of which 565 were screened after duplicate records were removed. Thirty-two articles were included for full-text review, and 12 trials met the inclusion criteria (Supplemental Figure 1). Unpublished data were obtained for three of these trials through collaborative efforts (TESTING,⁵ DAPA-CKD,¹⁹ and Telitacicept¹¹). Additional data for two of these trials were obtained through other sources: Post hoc analysis of the STOP-IgAN trial^4,20 and updated Phase 3 data from the ORIGIN trial obtained from a webcast presentation on the sponsor website delivered after the date of the literature search. Two additional studies were included in the meta-analysis through other sources: a prespecified secondary analysis of the EMPA-KIDNEY trial²¹ and prespecified interim analyses of the ongoing VISIONARY trial, which reported data after the date of the literature search.²² Proteinuria data from the VISIONARY trial was obtained from the primary presentation at the 62nd European Renal Association Congress.

In total, 14 trials with 3843 participants with IgA nephropathy were included in the analysis (Table 1). Of these, proteinuria and eGFR outcomes were available in 13 trials (93%) and seven trials (50%), respectively (Table 2). The follow-up duration of the studies ranged from a minimum of 28 weeks to a maximum mean of 4.2 years (SD 2.1). Overall, 37% of the participants were female, and the mean age was 43 years. The mean baseline eGFR was 60 ml/min per 1.73 m², and percentage of RAS inhibitors use was high. Racial representation was variable. All included studies had an overall low-moderate risk of bias (2Supplemental Table 2).

Table 1.

Baseline characteristics of multicenter randomized controlled trials studying the effect of different treatments on kidney outcomes in individuals with IgA nephropathy established on standard of care with maximal tolerated renin-angiotensin system blockade

Study Grouped by Drug Class	Methodology		Baseline Characteristics									Primary End Point
	Treatment	Inclusion Criteria	Study Size (No., total [tx])	Age (mean, yr [SD])	Female (%)	Race	Proteinuria (median, g/d [IQR] unless Otherwise Specified)	eGFR (mean, ml/min per 1.73 m2 [SD])	Time Since Kidney Biopsy/Diagnosis	RASi Use (%)	Follow-Up (yr unless otherwise specified)
Nonimmunologic therapies
DAPA-CKDa	Dapagliflozin 10 mg/d for study duration versus placebo	eGFR 25–75 uACR 200–5000 mg/g	254 (129)	50 (13)	32	59% Asian 40% White	uACR 888 mg/g (515–1541)	44 (12)	b	100	2.1 (0.03–3.2) median (min–max)	Composite of a sustained ≥50% decline in eGFR, ESKD, or death from kidney disease or cardiovascular cause
EMPA-KIDNEYc	Empagliflozin 10 mg/d for study duration versus placebo	eGFR 20–<45 or eGFR 45−<90 and uACR >200 mg/g	817 (413)	51 (13)	35	54% Asian 44% White	uACR 662 mg/g (331–1265)	43 (18)	b	94	2 (1.5–2.4) median (IQR)	Composite of kidney disease progression or cardiovascular death
PROTECTd	Sparsentan 400 mg/d for study duration versus irbesartan 300 mg/d	eGFR ≥30 Proteinuria >1 g/d	404 (202)	46 (12)	30	29% Asian 67% White	1.8 (1.3–2.8)	57 (24)	4.0-yr median	100	2 study period	Change from baseline proteinuria at 36 wk
ALIGNe	Atrasentan 0.75 mg/d for 132 wk versus placebo	eGFR ≥30 Proteinuria ≥1 g/d	270 (135)	45 (12)	41	57% Asian 36% White	1.43	59	5.6-yr mean	99	Ongoing	Change from baseline proteinuria at 36 wk
Corticosteroids
STOP-IgANf	Corticosteroids IV methylprednisolone 1g+PO prednisolone 0.5 mg/kg per 48 h tapered down for 6 mo or Cyclophosphamide 1.5 mg/kg per d+azathioprine 1.5 mg/kg per d+prednisolone tapered down for 36 mo versus placebo	eGFR ≥30 Proteinuria 0.75–3.5 g/d	162 (82)	44 (13)	21	100% White	1.7 (0.75) mean (SD)	59 (27)	b	98	3 study period	Hierarchical endpoint of full clinical remission and decrease in eGFR of at least 15 ml/min per 1.73 m2 from baseline
TESTINGg	Methylprednisolone 0.6–0.8 mg/kg per d or 0.4 mg/kg per d tapered down for 6–9 mo versus placebo	eGFR 20–120 Proteinuria ≥1 g/d	503 (257)	38 (11)	39	76% Chinese 5% White	1.96	62 (24)	5-mo median	99	4.2 (2.1) mean (SD)	Composite of a sustained 40% decline in eGFR, ESKD or death because of kidney disease
NefIgArdh	Nefecon 16 mg/d for 9 mo versus placebo	eGFR 35–90 Proteinuria ≥1 g/d or uPCR ≥0.8 g/g	364 (182)	43	34	23% Asian 76% White	2.23 (1.58–3.21)	55 (46–70) median (IQR)	2.5-yr median	98	2 study period	Time weighted average of eGFR over 2 yr
B-Cell modulating agents
ORIGINi	Atacicept 150 mg weekly versus placebo	eGFR ≥30 Proteinuria >1.0 g/d or uPCR >1 g/g	203 (106)	40 (18–72) median (range)	43	55% Asian 43% White	uPCR 1.7 g/g (1.0) mean (SD)	65 (28)	2.5-yr mean	b	Ongoing	Change from baseline proteinuria at 36 wk
ENVISIONj	Sibeprenlimab IV 2 mg/kg or 4 mg/kg or 6 mg/kg monthly for 12 mo versus placebo	eGFR ≥30 Proteinuria ≥1 g/d or uPCR ≥0.75 g/g	155 (117)	39 (18–73) median (range)	43	74% Asian 23% White	1.86	62	565-ds median	98	12-mo study period	Change from baseline proteinuria at 12 mo
VISIONARYk	Sibeprenlimab SC 400 mg monthly versus placebo	eGFR ≥30 Proteinuria ≥1 g/d or uPCR ≥0.75 g/g	320 (152)	43	38	59% Asian 38% White	uPCR 1.2 g/g	63 (25)	1.6 yr	97	Ongoing	Change from baseline proteinuria at 9 mo
Lv et al. 2023l	Telitacicept 160 mg or 240 mg weekly for 24 wk versus placebo	eGFR >35 Proteinuria ≥0.75 g/d	44 (30)	37 (9)	48	100% Chinese	1.66 (1.22–2.28)	79	1564-ds median	b	28-wk study period	Change from baseline proteinuria at 24 wk
Complement inhibitors
Barratt et al. 2024m	Cemdisiran 600 mg every 4 wk for 32 wk versus placebo	eGFR ≥30 Proteinuria ≥ 1g/d	31 (22)	40 (10)	48	52% Asian 39% White	2.4	58 mean of median values	2.2-yr mean of median values	97	36-wk study period	Change from baseline proteinuria at 32 wk
SANCTUARYn	Ravulizumab versus placebo for 26 wk followed by open-label Ravulizumab until week 50	eGFR ≥30 Proteinuria ≥ 1 g/d	66 (43)	40 (11)	46	21% Asian 71% White	2.4 (1.6-3.6)	74 (30)	2.5-yr median	ACEi 41 ARB 62	26-wk initial evaluation with ongoing OLE	Change from baseline proteinuria at 26 wk
APPLAUSE-IgANo	Iptacopan 200 mg twice daily versus placebo	eGFR ≥30 Proteinuria ≥ 1 g/d	250 (125)	39 (12)	48	51% Asian	uPCR 1.85 g/g	64 (26)	1.0-yr median	99	Ongoing	Change from baseline proteinuria at 9 mo

Continuous variables expressed as mean values with standard deviations or median values with interquartile range, as reported in study results where available. Where the overall mean/median is not available (i.e., reported separately for treatment and control arms), the mean of the two values was taken. Proteinuria reported as total 24-hour proteinuria in g/day unless otherwise specified. Baseline eGFR values rounded to the nearest whole number.

ACEi, angiotensin converting enzyme inhibitor; ARB, angiotensin receptor II, blocker; CI, confidence interval; eGFR, (ml/min per 1.73 m²); IV, intravenous; n, number; OLE, open label extension; RASi, renin-angiotensin system inhibitor; SC, subcutaneous; TESTING, Therapeutic Effect of Steroids in IgA Nephropathy Global; tx, treatment arm; uACR, urine albumin-to-creatinine ratio; UPCR, urine protein-to-creatinine ratio.

^aPrespecified subgroup analysis including unpublished data for biopsy-proven IgA nephropathy cohort of the DAPA-CKD trial. Wheeler et al. Kidney Int 2021; 100:215–224.

^bNot reported.

^cOnly trial included in this meta-analysis where biopsy proven IgA nephropathy was not an inclusion criterion. 95.7% participants had a kidney biopsy; however, time between kidney biopsy and randomization was not reported. The EMPA-KIDNEY Collaborative Group. Lancet Diabetes Endocrinol 2024; 12: 51–60.

^dPrespecified interim analysis and final 2-year study results. Heerspink et al. Lancet 2023; 401 (10,388):1584–1594 and Rovin et al. Lancet 2023; doi:10.1016/S0140-6736(23) 02,302–4.

^ePrespecified interim analysis of the ongoing ALIGN study. Heerspink et al. NEJM 2024; doi: 10.1056/NEJMoa2409415.

^fRauen et al. NEJM 2015; 373:2225–36.

^gLv and Wong et al. JAMA 2022; 327(19):1888–98.

^hLafayette et al. Lancet 2023; 402:859–70.

ⁱResults from the ongoing ORIGIN 3 study. Methods presented at the World Congress of Nephrology, Buenos Aires 2024. Results for the prespecified interim analysis presented as a webcast available on the ORIGIN trial sponsor website. Accessed online on June 7th, 2025: https://ir.veratx.com/news-releases/news-release-details/vera-therapeutics-announces-atacicept-achieved-46-proteinuria.

^jBaseline proteinuria and eGFR calculated as the mean of the reported median values for the pooled sibeprenlimab and placebo groups. Phase 2 ENVISION trial led to the ongoing Phase 3 VISIONARY trial. Mathur et al. NEJM 2024; doi:10.1056/NEJMoa2305635.

^kTime from biopsy to randomization reported as the mean of the median time for intervention and control arms. Results from the ongoing Phase 3 VISIONARY trial presented by Perkovic et al. Late Breaking Clinical Trials. 62nd European Renal Association Congress, Vienna June 6, 2025.

^lLv et al. Kidney Int Reports 2023; 8:499–506.

^meGFR and time since kidney biopsy calculated as the mean of the median values reported for the placebo and treatment arms. Barratt et al. CJASN 2024; doi:10.2215/CJN.0000000000000384.

ⁿResults from the initial double-blind placebo-control evaluation period of the SANCTUARY trial. Lafayette et al. JASN 2024; doi: 10.1681/ASN.0000000534.

^oPrespecified interim analysis of the APPLAUSE-IgAN trial. Perkovic et al. NEJM 2024; doi: 10.1056/NEJMoa2410316.

Table 2.

Comparison of findings from multicenter randomized controlled trials of the effect of different treatments on change in proteinuria from baseline and annual eGFR slope of decline in individuals with IgA nephropathy established on standard of care with maximal tolerated renin-angiotensin-system blockade

Study Grouped by Drug Class	Intervention	Change in Proteinuria from Baseline (% [95% CI])				Annual Total eGFR Slope (ml/min per 1.73 m2/yr [95% CI])
		Treatment	Control	Treatment Versus Control	Time Perioda	Treatment	Control	Treatment Versus Control	Time Periodb
Nonimmunologic Therapies
DAPA-CKDc	Dapagliflozin versus placebo	−32.4 (−41.7 to −21.8)	−3.7 (−16.8 to 11.5)	−35.4 (−56.2 to −14.7)	12 mo	−3.49 (SE 0.5)	−4.88 (SE 0.50)	1.39 (SE 0.71)	2.1 yr
EMPA-KIDNEYd	Empagliflozin versus placebo	e	e	−24 (−33 to −13)	2 yr	−2.87 (SE 0.22)	−4.01 (SE 0.22)	1.14 (0.54 to 1.75)	2 yr
PROTECTf	Sparsentan versus irbesartan	−49.8 (−55 to −44)	−15.1 (−23.7 to −5.4)	−41 (−49 to −31)	36 wk	−2.9 (−3.6 to −2.2)	−3.9 (−4.6 to −3.1)	1.0 (−0.03 to 1.94)	110 wk
ALIGNg	Atrasentan versus placebo	−38.1 (−43.9 to −31.7)	−3.1 (−12.4 to 7.3)	−36.1 (−44.6 to −26.4)	36 wk	e	e	e	e
Corticosteroids
STOP-IgANh	Immunosuppression including Corticosteroids versus placebo	−58.0 (e)	−27.0 (e)	−43 (−55 to −27)	12 mo	−1.4 (SD 4.7)	−1.57 (SD 4.1)	0.7 (−0.62 to 2.06)	2 yr
TESTINGi	Methylprednisolone versus placebo	−62.0 (−67.0 to −56.1)	−14.8 (−26.8 to −0.8)	−55.4 (−62.1 to −47.4)	12 mo	−0.08 (−1.60 to 1.44)	−3.51 (−5.04 to −1.98)	3.43 (1.28 to 5.58)	2 yr
TESTING  Reduced dosei	Reduced-dose methylprednisolone versus placebo	−65.0 (−72.0 to −56.3)	−17.6 (−34.2 to 3.3)	−57.6 (−67.2 to −45.1)	12 mo	1.95 (−0.25 to 4.15)	−2.22 (−4.46 to 0.03)	4.17 (1.03 to 7.31)	2 yr
TESTING  Full dosei	Full-dose methylprednisolone versus placebo	−60.1 (−66.3;−52.8)	−14.2 (−28.9 to 3.5)	−53.5 (−62.5 to −42.4)	12 mo	−2.40 (−4.43 to −0.38)	−4.97 (−6.99 to −2.94)	2.56 (−0.30 to 5.42)	2 yr
NefIgArdj	Nefecon versus placebo	−51.3 (−56.2 to −45.9)	−3.2 (−12.8 to 7.5)	−49.7 (−56.6 to −41.6)	12 mo	−3.06 (e)	−6.00 (e)	2.95 (1.67 to 4.58)	2 yr
B-cell modulating agents
ORIGINk	Atacicept versus placebo	−46 (SE 3)	−7 (SE 7)	−42 (−57 to −27)	36 wk	2.6 (SE 2.37)	−3.2 (SE 2.35)	5.9 (−0.75 to 12.49)	36 wk
ENVISIONl	Sibeprenlimab IV 2mg/kg monthly versus placebo	−47.2 (SE 8.2)	−20.0 (SE 12.6)	−34.0 (−56.2 to −0.4)	12 mo	−4.1 (SE 1.7)	−5.9 (SE 1.7)	1.81 (−2.8 to 6.4)	12 mo
ENVISIONl	Sibeprenlimab IV 4mg/kg monthly versus placebo	−58.8 (SE 6.1)	−20.0 (SE 12.6)	−48.5 (−65.4 to −23.2)	12 mo	0.1 (SE 1.6)	−5.9 (SE 1.7)	5.96 (1.5 to 10.4)	12 mo
ENVISIONl	Sibeprenlimab 8mg/kg monthly versus placebo	−62.0 (SE 5.7)	−20.0 (SE 12.6)	−52.5 (−68.4 to −28.8)	12 mo	−0.8 (SE 1.6)	−5.9 (SE 1.7)	5.08 (0.5 to 9.6)	12 mo
VISIONARYm	Sibeprenlimab SC 400mg monthly versus placebo	−50.2 (e)	2.1 (e)	−51.2 (−58.2 to −42.9)	9 mo	e	e	e	e
Lv et al. 2023n	Telitacicept 160mg weekly versus placebo	−24.7 (−41.2 to −8.2)	−0.8 (−19.4 to 17.8)	−23.9 (−48.6 to 0.8)	24 wk	8.04 (0.52 to 15.56)	−12.29 (−20.9 to −3.67)	20.33 (8.91 to 31.76)	24 wk
Lv et al. 2023n	Telitacicept 240mg weekly versus placebo	−49.3 (−66.9 to −31.6)	−0.8 (−19.4 to 17.8)	−48.5 (−73.9 to −23.0)	24 wk	3.82 (−4.21 to 11.85)	−12.29 (−20.9 to −3.67)	16.12 (4.35 to 27.88)	24 wk
Complement Inhibitors
Barratt et al. 2024o	Cemdisiran versus placebo	−31.4 (−46.2 to −12.6)	9.5 (−26.6 to 63.5)	−37.4 (−61.0 to 0.5)	32 wk	−6.76 (SD 10.92)	−11.90 (SD 9.33)	5.14 (e)	36 wk
SANCTUARYp	Ravulizumab versus placebo	−41.9 (−50.2 to −32.0)	−16.8 (−31.8 to 1.6)	−30.1 (−43.5 to −13.7)	26 wk	−1.35 (−5.51 to 2.81)	−6.74 (−12.37 to −1.1)	e	26 wk
APPLAUSE-IgANq	Iptacopan versus placebo	−43.8 (−50.9 to −35.8)	−9.0 (−20.8 to 4.6)	−38.3 (−48.6 to −26)	9 mo	e	e	e	e

eGFR, (ml/min per 1.73 m²); CI, confidence interval; IV, intravenous; SC, subcutaneous; TESTING, Therapeutic Effect of Steroids in IgA Nephropathy Global

^aTime period at which change in proteinuria from baseline was analyzed as per the study.

^bTime period over which the total eGFR slope was analyzed as per the study.

^cPost hoc analysis of the biopsy-proven IgA nephropathy cohort of the DAPA-CKD trial. Total eGFR slope analyzed over study duration (median follow-up of 2.1 years) reported as mean values with corresponding SE.

^dAvailable results from prespecified subgroup analysis of the EMPA-KIDNEY trial over the entire study period which included a median follow-up period of 2 years.

^eNot reported.

^fProteinuria results from Heerspink et al. Lancet 2023. Total annual eGFR slopes over the full double-blind treatment period from Rovin et al. Lancet 2023.

^gPrespecified interim analysis of the ongoing ALIGN study.

^hAnnual total eGFR slope in the treatment and control arms estimated by dividing the reported absolute eGFR change at 36 months by three from Rauen et al. NEJM 2015. Post hoc analysis results for treatment effect on change in proteinuria and eGFR from Thompson et al. cJASN 2019; 14:469–481 and Inker et al. JASN 2019; 30:1735–1745, respectively.

ⁱPost hoc analysis of the TESTING trial, including separate analyses on the reduced-dose and full-dose cohorts to further interrogate the effect of methylprednisolone on proteinuria and eGFR at specific time points to facilitate comparison with the other novel agents. The relative percentage change from baseline proteinuria at 12 months was estimated for each treatment arm with a mixed model for repeated measures using log-transformed proteinuria, to account for the skewed distribution of the values. Nonlinearity of eGFR slope was addressed by using a two-slope, mixed effects, linear spline model with a knot at month 3 over 2 years.

^jResults as reported in the 2-year NefIgArd trial from Lafayette et al. NEJM 2023; 402: 859–70.

^kTreatment effect on proteinuria reported on the ORIGIN 3 trial sponsor webcast and variance estimated using web-based image extraction software (WebPlotDigitizer). Treatment effect on annual eGFR slope based on 36-week follow-up data from the ORIGIN Phase 2b trial from Lafayette et al. KI 2024; 105:1306–1315. eGFR data not included in meta-analysis because of time period of <12 months.

^lProteinuria data not included in meta-analysis because of availability of Phase 3 data from the VISIONARY study outlined below. eGFR results from the Phase 2 ENVISION trial for each treatment dose.

^mPrespecified interim analysis of the ongoing VISIONARY study.

ⁿPost hoc analyses of Lv et al. KI Reports. Treatment effects on change in proteinuria and eGFR, with annual eGFR slope data estimated from 24-week follow-up data. eGFR data not included in meta-analysis because of time period of <12 months.

^oResults from Barratt et al. CJASN 2024; 19(4):452–462 and poster presented at the American Society of Nephrology Kidney Week 2022. Proteinuria effects reported with corresponding 90% confidence intervals. eGFR data not included in meta-analysis because of time period of <12 months.

^pResults for the initial evaluation period of the ongoing SANCTUARY study. 90% confidence intervals reported for the relative placebo-adjusted treatment effect on proteinuria reduction while 95% confidence intervals reported for the change in proteinuria for each treatment arm. Annual eGFR slope data estimated based on 26-week follow-up data.

^qPrespecified interim analysis results of the ongoing APPLAUSE-IgAN study from Perkovic et al. NEJM 2024 and presentation at the European Renal Association Congress 2024.

Effects on Proteinuria

Data indicate that clinically meaningful reductions in proteinuria have been observed across all therapeutic classes studied, although the magnitude of this effect may vary (P heterogeneity <0.001; Figure 1). Pooled data from recent trials found that the relative percentage change from baseline was −34% (95% CI, −42 to −26) with nonimmunologic therapies, −51% (95% CI, −56 to −46) with corticosteroids, −45% (95% CI, −55 to −35) with B-cell modulating agents, and −35% (95% CI, −46 to −25) with complement inhibitors. These effects were consistent across trials within each drug class (all P heterogeneity ≥ 0.10).

Figure 1.

Comparison of the effect of different therapeutic agents on proteinuria change from baseline in IgA nephropathy. Random-effects meta-analysis of the effect of treatment versus control on relative change in proteinuria from baseline as a percentage at a time point (months). Subgroup analysis by drug class with Q-test and corresponding P value to detect heterogeneity between groups. Empagliflozin proteinuria change analyzed from baseline to end of trial follow-up (median 2 years). Ravulizumab proteinuria change analyzed at 26 weeks from baseline. *90% confidence intervals presented as per available published data. CI, confidence interval; n, number; TESTING, Therapeutic Effect of Steroids in IgA Nephropathy Global.

There were four completed trials with follow-up beyond 2 years: DAPA-CKD, EMPA-KIDNEY, STOP-IgAN, and TESTING. Although the proteinuria-lowering effects of dapagliflozin at 12 months were comparatively modest, the effects of both dapagliflozin and empagliflozin were sustained throughout the study period.^19,21 Conversely, despite early proteinuria-lowering effects of corticosteroids seen in the STOP-IgAN and TESTING studies, reductions in proteinuria attenuated over time, with no discernible difference between treatment and placebo by 36 months.^5,23

Effects on eGFR Slope

Similar to effects on proteinuria, all therapeutic classes improved eGFR slope; however, the magnitude of benefit, both in relative and absolute terms, appeared to vary across different drug classes (both P heterogeneity = 0.03; Figure 2). Data on eGFR slope over 2 years remain limited to nonimmunologic therapies (DAPA-CKD, EMPA-KIDNEY, PROTECT) and corticosteroids (STOP-IgAN, TESTING, NefIgArd).

Figure 2.

Comparison of the effect of different therapeutic agents on annual total eGFR slope in IgA nephropathy. Random-effects meta-analysis of the effect of treatment versus control on absolute difference in annualized eGFR slope in ml/min per 1.73 m²/year and relative change in eGFR slope difference as a percentage of the control eGFR slope, analyzed over a time period (year) in clinical trials conducted in IgA nephropathy with minimum 1-year follow-up data. Subgroup analysis by drug class with Q-test and corresponding P value to detect heterogeneity between groups. Slope data for dapagliflozin and empagliflozin analyzed over the study duration of their respective studies. Slope data for sparsentan analyzed over 110 weeks. Tx, treatment,

Pooled data indicated that nonimmunologic therapies improved eGFR slope by 1.1 ml/min per 1.73 m² per year (95% CI, 0.7 to 1.6), an effect consistent across trials of sodium-glucose cotransporter-2 inhibitors (SGLT2i) and sparsentan (P heterogeneity = 0.90). In relative terms, this represented a −28% (95% CI, −40 to −16) improvement in eGFR slope, again consistent across trials (Figure 2).

Corticosteroids improved eGFR slope by 2.3 ml/min per 1.73 m² per year (95% CI, 0.8 to 3.8) in absolute terms. There was some evidence that this magnitude of effect varied by trial (P heterogeneity = 0.06) with the smallest effect observed in STOP-IgAN, likely because the background rate of eGFR decline was slowest in this trial. The largest effect on both absolute and relative difference in eGFR rate of decline was observed with reduced-dose methylprednisolone. Taking into account differences in the rate of eGFR decline in the placebo arms across trials, corticosteroids improved eGFR slope by −52% (95% CI, −74 to −31) which was consistent across trials (P heterogeneity = 0.31).

Currently, sibeprenlimab is the only novel immunological therapy to have eGFR slope data over a minimum of 12 months. Pooled results from the ENVISION study found that sibeprenlimab improved the rate of eGFR decline by 4.3 ml/min per 1.73 m² per year (95% CI, 1.7 to 6.9) and −73% (95% CI, −118 to −29) compared with control. Results for other B-cell modulating agents (atacicept and telitacicept) and complement inhibitors (cemdisiran and ravulizumab) on eGFR slope over 6–9 months are summarized in Table 2. Over a 96-week treatment, the rate of eGFR decline was −0.6 ml/min per 1.73 m²/year with atacicept; however, there was no placebo control included in the open label extension which constituted 60 of the 96 weeks observation period.²⁴

Safety and Net Effects

Safety outcomes for various agents are summarized in Table 3. SGLT2i have demonstrated favorable safety profiles in individuals IgA nephropathy, with no significant difference in treatment-emergent SAEs compared with placebo including over longer-term follow-up. Similarly, there were no differences in SAEs or adverse events leading to treatment discontinuation between dual endothelin angiotensin receptor antagonists and endothelin receptor antagonists (ERAs) and the control arms.^25,26

Table 3.

Comparison of safety outcomes reported in multicenter randomized controlled trials of different treatments studied in individuals with IgA nephropathy

Study Grouped by Drug Class	Intervention	Serious Adverse Events, n (%)		Adverse Events Leading to Treatment Discontinuation, n (%)
		Treatment	Control	Treatment	Control
Nonimmunologic therapies
DAPA-CKD	Dapagliflozin versus placebo	22 (16)	34 (26)	6 (4)	7 (5)
EMPA-KIDNEYa	Empagliflozin versus placebo	b	b	b	b
PROTECT	Sparsentan versus irbesartan	75 (37)	71 (35)	21 (10)	18 (9)
ALIGN	Atrasentan versus placebo	10 (6)	11 (7)	6 (4)	6 (4)
Corticosteroids
STOP-IgAN	Immunosuppression including corticosteroids versus placebo	29 (35)	21 (26)	b	b
TESTING  Reduced dose	Reduced-dose methylprednisolone versus placebo	6 (5)	3 (3)	2 (2)	1 (1)
TESTING  Full dose	Full-dose methylprednisolone versus placebo	22 (16)	4 (3)	8 (6)	2 (2)
NefIgArdc	Nefecon versus placebo	18 (10)	9 (5)	17 (9)	9 (2)
B-cell modulating agents
ORIGIN	Atacicept versus placebo	1 (1)	11 (5)	2 (1)	8 (4)
VISIONARY	Sibeprenlimab SC versus placebo	6 (4)	9 (5)	1 (1)	4 (2)
Lv et al. 2023	Telitacicept versus placebo	3 (10)	1 (1)	0 (0)	0 (0)
Complement inhibitors
Barratt et al. 2024	Cemdisiran versus placebo	1 (5)	9 (0)	0 (0)	0 (0)
SANCTUARY	Ravulizumab versus placebo	1 (2)	0 (0)	0 (0)	0 (0)
APPLAUSE-IgAN	Iptacopan versus placebo	18 (8)	22 (5)	6 (3)	6 (3)

n, number; SC, subcutaneous; TESTING, Therapeutic Effect of Steroids in IgA Nephropathy Global.

^aSafety data for dedicated IgA nephropathy cohort in the EMPA-KIDNEY trial not reported.

^bNot reported.

^cSafety data during the 9-month treatment period as reported in the 2-year NefIgArd study.

The evidence to date indicates that corticosteroids improve kidney outcomes in IgA nephropathy at the cost of an increased risk of SAEs, with the greatest risk observed with full-dose methylprednisolone, followed by reduced-dose methylprednisolone and TRF-budesonide. Although TRF-budesonide is considered the safest corticosteroid therapy, it was still associated with double the rate of SAEs compared with placebo (10% versus 5%) leading to a higher rate of discontinuation because of treatment-emergent AE (9% versus 2%) in the NefIgArd study (Table 3). There are no safety data for prolonged therapy beyond 9 months or repeated dosing of TRF-budesonide.

There was no increased risk of SAEs or AEs leading to treatment discontinuation in the ORIGIN and VISIONARY trials for B-cell modulating agents. There was a greater proportion of SAEs experienced with telitacicept compared with placebo; however, the sample size was small and total number of events was low. There was a numerically increased incidence of SAEs with complement inhibitors although there were very few events, with no difference in AEs leading to treatment discontinuation over a short follow-up period (6 to 9 months).

Discussion

The therapeutic landscape for IgA nephropathy has undergone transformation, with multiple proven and emerging evidence-based therapies that offer major potential to delay or even halt disease progression. In this systematic review and meta-analysis of the available randomized evidence we summarized the efficacy and safety of multiple proven and emerging evidence-based therapies that offer major therapeutic potential for individuals with IgA nephropathy. Available data indicate that nonimmunologic therapies reduce proteinuria and eGFR decline with a favorable safety profile. Although corticosteroids reduce short-term proteinuria and improve eGFR decline to a greater extent than nonimmunologic therapies, they are also associated with increased SAEs, particularly related to infection. A reduced-dose or targeted-release formulation is likely to mitigate, but not eliminate, these safety concerns. Early data on B-cell modulating agents and complement inhibitors are encouraging, but confirmatory findings, including safety data, from larger Phase 3 trials are awaited.

SGLT2i are now considered standard of care in international guidelines for patients with proteinuric CKD.²⁷ Post hoc analyses of large SGLT2i trials have demonstrated improved kidney outcomes in patients with IgA nephropathy.^19,21 In a pooled analysis of 1087 patients with IgA nephropathy from the DAPA-CKD and EMPA-KIDNEY trials, SGLT2i reduced the risk of 50% decline in eGFR or kidney failure by 51% (HR, 0.49; 95% CI, 0.32 to 0.74).²⁸ We observed that SGLT2i and other nonimmunologic therapies including dual endothelin angiotensin receptor antagonist and ERA appeared to have a lower magnitude of effect on proteinuria and eGFR decline compared with immunologic therapies, which may reflect the fact that nonimmunologic therapies do not directly target immune complex formation or glomerular inflammation. Despite this, the favorable safety profile and broader cardio- and reno-protective effects of SGLT2i carry significant benefits particularly for patients at high risk of side effects from immunosuppressive therapies.

Autoimmunity, characterized by the production of galactose-deficient IgA1 autoantibodies, and immune complex formation and deposition, is central to the pathogenesis of IgA nephropathy. Modulating the immune system and reducing glomerular inflammation with corticosteroids has therefore been a mainstay of treatment. Results from key trials of corticosteroids have been mixed^5,6,23; however, we found clear improvements in proteinuria and eGFR outcomes in IgA nephropathy overall. Although the observed eGFR slope benefits may be partially confounded by steroid-induced sarcopenia (artificially elevating eGFR based on serum creatinine), clear benefits were demonstrated on clinical kidney outcomes including kidney failure alone in the TESTING trial. Use of corticosteroids has largely been limited by concerns about toxicities and their broader immunosuppressive effects rather than a targeted disease-specific mechanism. Harm minimization is a key component in treatment decisions which may be particularly relevant for individuals who are frail, older, or more predisposed to the side effects of corticosteroids. This is reflected in the 2021 Kidney Disease Improving Global Outcomes treatment guidelines where corticosteroids are not recommended in those with advanced age, an eGFR <30 ml/min per 1.73 m², metabolic syndrome, or latent infection.² Indeed, post hoc analyses of the TESTING trial suggest harms may outweigh benefits in patients with advanced CKD.²⁹ If corticosteroids are used, TRF-budesonide or a reduced-dose regimen as used in the TESTING trial combined with Pneumocystis jirovecii pneumonia prophylaxis may mitigate infection risk and should be considered.³⁰

There have been multiple promising agents developed to inhibit a proliferation inducing ligand (APRIL) and B-cell activating factor, two key mediators of B-cell maturation, survival, and proliferation in IgA nephropathy.³¹ Early proteinuria and eGFR data are especially promising across a number of Phase 2 and ongoing Phase 3 trials.^10,11,24,32 Indeed, we observed that effects on eGFR decline were largest with B-cell directed therapies, highlighting the potential impact of targeting the underlying immunological basis of the disease. Importantly, safety and tolerability of these agents has appeared promising thus far. Current data indicate that cessation of some of these agents result in a rebound rise in galactose-deficient IgA1,¹⁰ suggesting that longer duration of therapy (e.g.,>12 months) may be required to maintain disease-modifying benefits. As IgA nephropathy often affects young adults, the ramifications of long-term treatment and potential side effects such as infection, hypogammaglobulinemia, and malignancy must therefore be carefully evaluated in ongoing trials.

Involvement of the complement system in glomerular injury in IgA nephropathy has been well documented for decades and is often associated with disease activity.³³ For this reason, several therapeutic agents targeting the complement system are currently under development and investigation for use in IgA nephropathy. Iptacopan, a factor B inhibitor, has been shown to reduce proteinuria compared with placebo at 9 months and was well tolerated, leading to its recent accelerated FDA approval⁸ while awaiting confirmatory results from the ongoing Phase 3 APPLAUSE-IgAN study.³⁴ Other agents aiming to inhibit complement are being evaluated in early Phase trials with encouraging preliminary results on proteinuria reduction,^12,35 but with no robust data on eGFR slope to date. The confirmation of benefit on clinical kidney outcomes, clinical translation, and safety of complement inhibitors is still largely awaited. Importantly, complement inhibition places patients at an increased risk of opportunistic infections, particularly meningococcal infection. Although some of these can be prevented with appropriate vaccines, additional long-term follow-up of treatment-emergent side effects is critical.

Future Directions and Unanswered Questions

As we enter a new era in IgA nephropathy therapeutics, clinicians are provided with ever expanding opportunities to use different agents and deliver highly personalized patient-tailored treatment. The four-hit hypothesis of IgA nephropathy supports a multifaceted treatment approach. Taken together, the future of IgA nephropathy treatment may resemble other glomerulonephritides involving induction therapy with corticosteroid-sparing complement inhibition to suppress glomerular inflammation while B-cell modulating therapy is introduced, with background optimal supportive therapies to mitigate ongoing maladaptive responses to glomerular injury. This is, of course, a simplification of a highly heterogeneous condition, and many important therapeutic questions remain unanswered. As data on more novel therapies continue to be generated, addressing these questions is likely to become more pressing.

In this meta-analysis, we found that all drug classes improved kidney outcomes; however, it remains uncertain which agent(s) should be used for which patients. Treatment response in IgA nephropathy has historically been widely variable, which is hypothesized to be influenced by factors such as disease activity or chronicity, kidney biopsy findings, race, and geographic and environmental variations. Other patient factors including age, frailty, or comorbidities may also influence the drug's safety profile, which is equally important. As new drugs continue to be developed, understanding which treatment is most effective and least toxic for each individual must be considered. Head-to-head comparison of these agents is most desirable, yet this is unlikely to occur at least in the short term. However, the data-sharing of prospective patient level clinical trial data has the potential to advance risk prediction tools to aid treatment decisions and personalized care.

The optimal duration of treatment is also unknown and is likely to differ substantially by drug class. Insights from the corticosteroid trials suggest that a short 6- to 9-month course improves kidney outcomes, but their side effect profile restricts prolonged or repeated therapy, and by 36 months, no difference in proteinuria is observed across treatment arms. On the other hand, the SGLT2i trials suggest that ongoing therapy is both safe and efficacious over time. These differences likely reflect the mechanisms of action and treatment regimens of each of the two drugs classes: Continuing SGLT2i is proposed to carry broader hemodynamic and reno-protective effects in individuals with CKD, independent of etiology, whereas a short course of corticosteroids has been postulated to suppress the initial inflammatory process of the disease with a legacy effect lasting 12–24 months.

The lack of long-term data on the emerging immunomodulatory agents lends uncertainty to the safety and efficacy of prolonged treatment duration or repeated courses. Furthermore, the roles of each of these agents in a proposed induction/maintenance treatment regimen must also be delineated.

Another key question is whether these drugs work synergistically and how combination therapy can be used safely and effectively. Data from CKD trials suggest that adding SGLT2i to ERAs mitigate the risks of fluid retention and enhance albuminuria reduction,^36,37 but data on combination therapy with other immunomodulatory drugs in IgA nephropathy remain limited. To this end, collaborative platform clinical trials supported by industry partnerships to study drug interactions and multiple therapies simultaneously are needed. Complex trial designs, statistical methodologies, costs, and establishing industry collaborations may pose as potential challenges. Despite this, such trials are being implemented in CKD,³⁸ demonstrating their feasibility, and these models should serve as valuable frameworks for the study of therapeutics in IgA nephropathy.

Finally, and crucially, the translational potential of these research findings in the treatment of IgA nephropathy must be prioritized. Most of these drugs, including those approved by the FDA, remain inaccessible for large parts of the world. When generating practical international treatment guidelines, drug accessibility and affordability must be considered, and the engagement of industry, policy makers and governments will be critical to lower costs and increase subsidization for these novel agents.

Strengths and Limitations

A key strength of this work is the pooling of available randomized evidence, including unpublished data from multiple trials, to provide a timely and clinically relevant review of the available and upcoming therapies in the IgA nephropathy. However, there are some limitations to acknowledge. First, there is a paucity of long-term and confirmatory Phase 3 eGFR slope data particularly for B-cell modulating agents and complement inhibitors. Second, there was substantial heterogeneity across study populations in terms of ethnicity, baseline eGFR and proteinuria, and background rate of eGFR decline. Consequently, evaluation of comparative effectiveness should be undertaken with appropriate caution. However, given the lack of available head-to-head data, this meta-analysis provides the most comprehensive evaluation of the established and emerging immunological and nonimmunologic therapies for IgA nephropathy to-date. Finally, although we aimed to pool effects on proteinuria at 9 months and eGFR slope at 2 years to align with FDA guidance for the evaluation of therapies in IgA nephropathy, we used data from other time points where needed to make maximal use of available randomized data.

In summary, we found that nonimmunologic therapies, corticosteroids, B-cell modulating agents, and complement inhibitors reduced proteinuria from baseline and slowed eGFR rate of decline compared with supportive care, with heterogeneity across drug classes. Each drug class carries benefits and limitations which must be considered when making treatment decisions, and longer-term confirmatory Phase 3 data on several novel therapies are awaited. Despite knowledge gaps and barriers to treatment access, the recent pace of advancements in IgA nephropathy is unprecedented, with major opportunities to improve outcomes for patients with IgA nephropathy in the near future.

Disclosures

Disclosure forms, as provided by each author, are available with the online version of the article at http://links.lww.com/CJN/C408.

Author Contributions

Conceptualization: Dana Kim, Brendon L. Neuen, Vlado Perkovic, Muh Geot Wong.

Data curation: Dana Kim, Brendon L. Neuen.

Formal analysis: Dana Kim.

Methodology: Dana Kim, Brendon L. Neuen, Vlado Perkovic, Muh Geot Wong.

Resources: Vlado Perkovic, Muh Geot Wong.

Supervision: Brendon L. Neuen, Vlado Perkovic, Muh Geot Wong.

Writing – original draft: Dana Kim.

Writing – review & editing: Dana Kim, Brendon L. Neuen, Vlado Perkovic, Muh Geot Wong.

Funding

None.

Data Availability Statements

Data belong to a third party, and authors are not authorized to share the data.*; Original data generated for the study will be made available upon reasonable request to the corresponding author. Data Type: Clinical Trial Data. Reason for Restricted Access: Data access for the TESTING trial can be requested from Professor Vlado Perkovic via the email: vlado.perkovic@unsw.edu.au. If data access is granted following reasonable request, data can be accessed via a secured environment hosted by The George Institute for Global Health, Australia, as per the study's ethics approval and China Personal Information Protection Law. Identity of Third Party: The authors acknowledge the collaborators of this review and the sharing of data, including contributions from the DAPA-CKD trial (Hiddo J.L Heerspink) and the Telitacicept trial (Jicheng Lv). Reason for Restriction: The authors are not authorized to share data from the DAPA-CKD or Telitacicept trials.

Supplemental Material

This article contains the following supplemental material online at http://links.lww.com/CJN/C409.

Supplemental Table 1. Search strategy.

Supplemental Table 2. Risk of bias assessment.

Supplemental Figure 1. Flow diagram of study selection.