A case of membranoproliferative glomerulonephritis-type IgA nephropathy modified by nintedanib

Abstract

A 73-year-old woman had been treated with nintedanib for 40 months for idiopathic pulmonary fibrosis. She presented to our hospital with lower extremity edema since a month and nephrotic syndrome (serum albumin 2.0 mg/dL, urine protein 7.97 g/g Cr). We performed a renal biopsy and diagnosed membranoproliferative glomerulonephritis (MPGN)-type IgA nephropathy (IgAN). We hypothesized that nintedanib-induced microangiopathy modified subclinical IgAN and caused the appearance of MPGN-type IgAN and nephrotic syndrome. After the discontinuation of nintedanib, the urinary protein level improved to 0.5 g/g Cr, and hematuria disappeared.

Introduction

The prognosis of patients with cancer has improved with recent progress in cancer chemotherapy. However, drug-induced renal injury caused by anticancer drugs, especially molecular targeted drugs, has become an increasing problem in onco-nephrology [1]. Among these molecular targeted drugs, the inhibition of vascular endothelial growth factor (VEGF) has been reported to cause hypertension and urinary protein accumulation via glomerular endothelial cell injury [2]. It is unclear how these molecularly targeted drugs modify existing subclinical renal diseases. Nintedanib is a potent triple angiokinase inhibitor of vascular endothelial growth factor receptor (VEGFR), fibroblast growth factor receptor (FGFR), and platelet-derived growth factor receptor (PDGFR) [3] and has been approved for the treatment of idiopathic pulmonary fibrosis in Japan [4]. Herein, we report a case of membranoproliferative glomerulonephritis (MPGN)-type IgA nephropathy (IgAN) modified by nintedanib.

Case report

A 73-year-old woman with idiopathic pulmonary fibrosis and hypothyroidism regularly visited the respiratory department of our hospital. Diffuse reticular shadows were observed throughout the lungs (Fig. 1). She had a history of smoking 20 cigarettes per day for 23 years. Nintedanib (150 mg/day) was initiated to treat idiopathic pulmonary fibrosis. Thirty-eight months after starting nintedanib, the patient developed a nontuberculous mycobacterial infection, and treatment with clarithromycin (400 mg/day), rifampicin (300 mg/day), and ethambutol (500 mg/day) was initiated. The patient was also on diltiazem hydrochloride (100 mg/day) for angina pectoris, furosemide (40 mg/day) for edema, metoclopramide (5 mg/day) for nausea, ambroxol hydrochloride (45 mg/day) for idiopathic pulmonary fibrosis, and bilastine (20 mg/day) for allergic rhinitis. The patient presented to a local clinic 39 months after starting nintedanib treatment with lower leg edema and weight gain. Forty months after starting nintedanib, her serum albumin was 2.4 g/dL, and proteinuria was 5.5 g/gCr, which was suggestive of nephrotic syndrome; she was then referred to our department. At the initial consultation at our department, her body mass index was 16.5 kg/m² with no fever, mild hypertension (blood pressure, 156/107 mmHg), hypoxemia (SpO2, 95% on room air), or dyspnea on light exercise. Significant fine crackles were observed bilaterally and pitting edema was observed in the extremities without purpura. Laboratory tests are listed in Table 1.

Fig. 1.

Chest radiograph at admission

Table 1.

Laboratory examination on admission

Parameter	Level
White blood cell count, per mm3	8780
Red blood cell count, per mm3	394 × 104
Hemoglobin, g/dL	12.6
Platelet count, per mm3	24.0 × 104
Blood urea nitrogen, mg/dL	23.0
Creatinine, mg/dL	0.71
eGFR, ml/min/1.73m2	60.9
Uric acid, mg/dL	6.4
Total protein, g/dL	6.0
Albumin, g/dL	2.0
Sodium, mEq/L	138
Potassium, mEq/L	4.3
Chloride, mEq/L	104
Calcium, mg/dL	8.6
Phosphorus, mg/dL	4.4
CRP, mg/dL	0.52
AST, U/L	38
ALT, U/L	20
HDL cholesterol, mg/dL	98
LDL cholesterol, mg/dL	114
Triglyceride, mg/dl	84
Hemoglobin A1c, %	5.7
Immunoglobulin G, mg/dL	1625
Immunoglobulin A, mg/dL	421
Immunoglobulin M, mg/dL	59
Complement 3, mg/dL	116
Complement 4, mg/dL	26
CH50, /mL	57.7
TSH, µIU/mL	12.300
Free T3, pg/mL	1.92
Free T4, ng/dL	0.84
Antinuclear antibody, fold	< 40
MPO-ANCA, U/mL	< 1.0
PR3-ANCA, U/mL	< 1.0
Prothrombin Time, %	101
Prothrombin Time-International Normalized Ratio	1.00
Activated partial thromboplastin time, sec	23.0
D-dimer, µg/mL	3.2
HBs-Ag	Negative
HCV-Ab	Negative
Cryoglobulin	Negative
Proteinuria by dipstick test	4 +
Hematuria by dipstick test	3 +
Urine protein/Cr ratio, g/gCr	7.97
Urine red blood cell, /HPF	> 100
Urine cast, /WF	1–4
Urine β2 microglobulin, µg/L	359
Urine N-acetyl-beta-glucosaminidase, U/L	46.3

eGFR: estimated glomerular filtration rate, CRP: C-reactive protein, AST: aspartate aminotransferase, ALT: alanine aminotransferase, HDL-Chol: high density lipoprotein cholesterol, LDL-Chol: low density lipoprotein cholesterol, CK: creatine kinase, CH50; 50% hemolytic complement activity, TSH: Thyroid stimulating hormone, Free T3; free triiodothyronine, Free T4; free thyroxine, MPO-ANCA; myeloperoxidase anti-neutrophil cytoplasmic antibody, PR3-ANCA; serine proteinase3- anti-neutrophil cytoplasmic antibody, HBs-Ag; hepatitis B virus surface antigen, HCV-Ab; hepatitis C virus anti¬body

Percutaneous kidney biopsy revealed 28 glomeruli, including 5 with global sclerosis, 1 with segmental sclerosis, 1 with a cellular crescent, and 6 with fibrocellular crescents, 2 of which showed fibrinoid necrosis. Three glomeruli had fibrous crescents, and segmental endocapillary proliferation was present. Mild mesangial cell and matrix proliferation was observed (Fig. 2a). The basement membrane showed diffuse segmental duplication and subendothelial deposits (Fig. 2b and c). Two glomeruli exhibited extracapillary fibrin deposits, suggesting capillary necrosis (Fig. 2d). There was moderate intimal thickening of the interlobular and arcuate arteries (Fig. 2e), and moderate to severe arteriolar hyalinosis (Fig. 2f).

Fig. 2.

a Periodic acid Schiff staining showing mild proliferation of mesangial cells and matrix, accompanied by fibrocellular crescent formation b Periodic acid silver methenamine staining showing segmental duplications of the basement membrane (arrows). c Masson trichrome staining showing segmental subendothelial deposits (arrows) d Masson trichrome staining showing fibrinoid necrosis in the capillary (arrows). e Elastica Van Gieson staining showing moderate intimal thickening in the interlobular and arcuate arteries f Periodic acid Schiff staining showing moderate to severe arteriolar hyalinosis

Fluorescent staining showed IgA, IgG, IgM, and C3 deposits in both the mesangial areas and capillary walls (Fig. 3a–e). In addition, the staining for galactose-deficient IgA1 (Gd-IgA1; KM55, #10,777, Immuno-Biological Laboratories, Gunma, Japan), a characteristic antibody for primary IgAN, was detected in the mesangial areas and capillary walls (Fig. 3f). Electron microscopy revealed electron-dense deposits in the paramesangial regions, mesangial matrix, and subendothelium of the capillary loops (Fig. 4a, b), with infiltration of a few inflammatory cells. Swelling of the capillary endothelium, widespread effacement of foot processes, accompanied by thickening of foot process material, and mesangial interposition were also noted in certain areas within the loops (Fig. 4c, d). In the mesangial areas, proliferation of mesangial cells and an increase in the mesangial matrix were observed.

Fig. 3.

Fluorescent staining For IgG (a), IgA (b), IgM (c), C1q (d), C3c (e), and galactose-deficient-IgA1 (f)

Fig. 4.

Electron microscopy findings. a Electron-dense deposits were observed in the paramesangial regions and mesangial matrix. b Deposits were also observed in the subendothelium of the capillary loops. c, d In the capillary loops, swelling of the capillary endothelium, widespread effacement of foot processes, and an increase in foot process material were observed, and mesangial interposition was also observed in some areas (arrows)

Based on the given findings, the patient was diagnosed with MPGN-type IgAN, which was classified as M0, E1, S1, T0, C2 according to the Oxford classification, and as H-grade Ⅲ (A/C) according to the IgAN histological classification of the Japanese Society of Nephrology. We hypothesized that nintedanib-induced microangiopathy modifies subclinical IgAN and causes MPGN-type IgAN and nephrotic syndrome. We assumed that the anti-VEGF effect of nintedanib caused endothelial damage, including swelling of the glomerular endothelium, extensive effacement of foot processes in the glomerular epithelial cells, and hyalinization of the arterioles. We decided to discontinue nintedanib, and 18 months after discontinuation, proteinuria gradually decreased to 0.5 g/gCr and hematuria disappeared (Fig. 5). The serum creatinine levels did not change after the discontinuation of nintedanib. The respiratory symptoms remained unchanged, and treatment for non-tuberculous mycobacterial infection was continued.

Fig. 5.

Clinical course. The patient stopped taking nintedanib for a total of 8 days; 5 days prior to and 3 days following the biopsy Cr, creatinine; Alb, albumin; HPF, high power field

Discussion

Here, we report a case of MPGN-type IgAN caused by nintedanib treatment. We hypothesized that nintedanib-induced microangiopathy and foot process effacement modify subclinical IgAN and cause MPGN-type IgAN.

VEGF signaling in the kidneys plays a crucial role in maintaining the function of glomerular capillary endothelial cells [5]; and signaling between podocytes and glomerular endothelial cells is tightly regulated. Therefore, VEGF inhibition can lead to renal endothelial injury, causing proteinuria, hypertension, thrombotic microangiopathy (TMA), minimal change disease and focal segmental glomerulosclerosis [5]. Bevacizumab has been reported to cause adverse effects such as hypertension and dose-dependent proteinuria [6, 7], which are associated with renal microangiopathy [8]. Monoclonal antibodies against vascular endothelial growth factor A (VEGFA), such as bevacizumab, inhibit the binding of VEGFA to vascular endothelial growth factor receptor 2 (VEGFR2). Impairment of VEGF function leads to loss of glomerular endothelial cell fenestrations, which promotes microvascular injury, and the development of thrombotic microangiopathy (TMA) [9]. Tyrosine kinase inhibitors (TKIs), such as nintedanib, inhibit downstream signaling of VEGFR, PDGFR, FGFR, and epidermal growth factor receptor [5]. TKIs act on renin inhibitory activator (RelA), which promotes transcriptional activation and suppresses c-maf inducing protein (c-mip) activity upon nuclear translocation within the cell [10]. As a result, ReIA is unable to translocate into the nucleus, and remains in the cytoplasm. Therefore, TKI administration impairs RelA function, leading to c-mip overexpression, and subsequent podocyte injury [11].

Drug-induced IgAN has been reported with infliximab, adalimumab (tumor necrosis factor-α inhibitors), pembrolizumab, nivolumab (immune checkpoint inhibitors), and bevacizumab (anti-VEGF inhibitors) [12]. However, there are no reported cases of drug-induced IgAN associated with nintedanib. IgAN is hypothesized to develop from the appearance of Gd-IgA1 in serum, followed by the production of autoantibodies, with genetic predisposition as an underlying factor. Drug-induced IgAN caused by nivolumab [13] is believed to be caused by immune fluctuations. In our case, it is unlikely that nintedanib induced IgAN de novo, but rather modified existing IgAN through endothelial damage, resulting in an MPGN-like pathology. Although a pre- and post-treatment renal biopsy would have been informative, the pathology before treatment could not be confirmed. Compared to previous cases, only two, including this case, exhibited severe hematuria [4, 14–16]. In the present case, although hematuria was not evaluated before nintedanib administration, nephritis was not clinically evident. Drug-induced IgAN has been the primary focus in the classification of secondary IgAN, and has been reported in several previous studies [12]. In our case, the patient was Gd-IgA1 positive, suggesting underlying primary IgAN. We believe that nintedanib did not induce IgAN onset but rather activated pre-existing subclinical IgA deposition. In transplanted kidneys, IgA deposition with no urinary abnormalities occurs prior to the onset of IgAN [17]. In the present case, IgA deposition with no urinary abnormalities may have progressed to IgAN with nintedanib treatment, and then reverted to IgA deposition, with an improvement in microangiopathy, after nintedanib discontinuation. Given the recent increase in the aging population, we believe that the complex pathogenesis of molecular targeted drug modifications, as well as subclinical kidney disease, will be on the rise.

Drug-induced TMA is the third most common cause of TMA, representing 10–13% of all TMA cases and 20–30% of secondary cases, following pregnancy-associated and infection-related TMAs [18]. Multiple drugs have been reported to be involved in the development of secondary TMA, and multikinase inhibitors with anti-VEGF activity, including nintedanib, have been strongly associated with TMA [19]. To the best of our knowledge, this is the fourth reported case of TMA attributed to nintedanib. Previous reports on nintedanib-induced glomerular microangiopathy are presented in Table 2. Three of the four cases presented with nephrotic syndrome associated with TMA [4, 14–16]. Patients with TMA and no subclinical renal disease showed improvement after discontinuation of nintedanib. In the present case, we decided to discontinue the use of nintedanib without administering additional immunosuppressive therapy, due to concerns that this might exacerbate the non-tuberculous mycobacterial infection. After drug discontinuation, proteinuria decreased, and edema and hematuria resolved. In the long run, we chose to aim for partial remission type I rather than to target complete remission.

Table 2.

Reports of nintedanib-induced glomerular microangiopathy

Author	Year	Age	Sex	Complaint	U-RBC	Duration	Diagnosis	Treatment	Outcome
Ismail [4]	2017	59	Female	Acute kidney injury	> 100/HPF	4 months	Anti-GBM Nephritis	Steroid pulse, IVCY + PEx	Hemodialysis
Hasegawa [14]	2020	68	Male	Nephrotic syndrome	10–14/HPF	10 months	Glomerular microangiopathy	Nintedanib discontinuation	ICR-1
Inoue [15]	2020	45	Male	Nephrotic syndrome	N/A	36 months	Thrombotic microangiopathy	Nintedanib discontinuation	CR
Fujita [16]	2021	83	Male	Nephrotic syndrome	0/HPF	10 months	Thrombotic microangiopathy	Nintedanib discontinuation	CR
Kunisho (This Case)	2024	73	Female	Nephrotic syndrome	> 100/HPF	40 months	Glomerular microangiopathy  + IgA deposition	Nintedanib discontinuation	ICR-1

RBC; red blood cells, GBM; glomerular basement membrane, IVCY + PEx; Intravenous cyclophosphamide and therapeutic plasma exchange, ICR-1; Incomplete remission type 1, CR; Complete remission

Although several cases of drug-induced kidney injury caused by nintedanib have been reported [4, 14–16], none of these reports mention arteriolar hyalinosis. However, endothelial swelling of arterioles caused by anti-VEGF inhibitors has been reported [20]. Chronic calcineurin inhibitor (CNI)-associated arteriolopathy (CAA) due to CNI toxicity causes swelling and vacuolization of smooth muscle cells in the arteriolar media. As the condition progresses, severe degeneration and necrosis of smooth muscle cells occur, leading to plasma protein leakage, and deposition in a globular or spindle-shaped pattern [21]. Microangiopathy caused by anti-VEGF inhibitors resembles CAA in that both demonstrate plasma protein deposits within the medial layer of arterioles. However, a key difference is that CAA is characterized by degeneration and necrosis of medial smooth muscle cells. Additionally, CAA is associated with tubulointerstitial fibrosis and tubular atrophy, which are not observed in the present case.

Both acute and chronic glomerular lesions in IgAN are associated with the loss of glomerular endothelial cells and capillaries. Particularly, glomerular necrosis is characterized by the disappearance of endothelial cells and fibrin exudation, suggesting a possible link to the development of proteinuria and/or hematuria. Morphological features such as subendothelial space widening, mesangial interposition, and separation of endothelial cells from the glomerular basement membrane are thought to represent early manifestations of endothelial injury. These endothelial abnormalities may contribute to disease progression, activity, chronicity, and the advancement of chronic kidney disease in patients with IgAN [22]. IgA-dominant MPGN is considered distinct from primary IgAN in both clinical presentation and histopathological features. Unlike primary IgAN, IgA-dominant MPGN typically shows diffuse MPGN-like patterns with prominent subendothelial deposits and no mesangial deposits [23]. Clinically, compared to typical IgAN, MPGN-pattern IgAN more frequently presents with nephrotic syndrome and is associated with a poorer prognosis [24]. In the differential diagnosis of IgA-dominant MPGN, other conditions such as IgA vasculitis, lupus nephritis, and IgA-dominant infection-related glomerulonephritis should be considered. However, these possibilities were ruled out in this case. In this case, deposits of IgA and C3 were observed primarily in mesangial areas, consistent with the typical pattern of IgAN. We hypothesize that IgAN was present before the administration of nintedanib, and that the endothelial damage induced by the drug led to capillary endothelial swelling and mesangial interposition, forming spaces in the subendothelial region, where dense deposits subsequently accumulated or migrated. This resulted in a diagnosis of MPGN-type IgAN. However, there is no evidence to support this hypothesis, and future research is needed to clarify this phenomenon. In addition, IgM deposition was considered a nonspecific finding for IgAN.

In conclusion, we report a case of IgAN exacerbated by the administration of nintedanib. Nintedanib-induced microangiopathy was diagnosed based on renal histopathology and information from previously reported cases. The improvement in proteinuria following the discontinuation of nintedanib further confirmed our diagnosis. Therefore, close monitoring of urinary abnormalities and blood pressure should be performed when administering molecular targeted drugs that affect VEGF levels, such as nintedanib.

Declarations

Conflict of interest

The authors declare no conflict of interest.

Ethical approval

All procedures involving human participants were performed in accordance with the ethical standards of the institutional and/or national research committee and the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.

Informed consent

Informed consent was obtained from an individual participant included in the study.