Anti-neutrophil Cytoplasmic Antibody-negative Eosinophilic Granulomatosis with Polyangiitis Complicated with Peripheral Neuropathy that Underwent Remission Induction with Mepolizumab Monotherapy

Abstract

A 72-year-old woman was admitted to our hospital with numbness in her lower extremities and hypereosinophilia. She was diagnosed with eosinophilic granulomatosis with polyangiitis (EGPA). On admission, she was suspected of being complicated with pneumonia and sepsis; therefore, treatment with mepolizumab monotherapy was begun, resulting in partial improvement. After the possibility of a complicating infection was ruled out, corticosteroids were initiated, followed by intravenous gamma globulin therapy. Although the induction of remission of EGPA with mepolizumab monotherapy is not usually recommended, induction with mepolizumab monotherapy may be an option in terms of safety and clinical efficacy in some cases.

Introduction

Eosinophilic granulomatosis with polyangiitis (EGPA) is a disease characterized by bronchial asthma and sinusitis as prodromal symptoms, followed by an increased eosinophil count and damage to various organs (1). Although remission induction therapy with corticosteroids and immunosuppressive drugs or biologics is recommended for the treatment of EGPA (2), there is no clear evidence for remission induction therapy in the presence of concomitant infection.

Mepolizumab, a humanized anti-interleukin (IL)-5 monoclonal antibody, is effective in reducing disease recurrence and the dose of corticosteroids administered in patients with EGPA for whom conventional therapy has proven ineffective (3). Its use was approved in Japan in 2018. The 2021 American College of Rheumatology (ACR) guidelines for the management of anti-neutrophil cytoplasmic antibody (ANCA)-associated vasculitis also recommended mepolizumab be administered in combination with corticosteroids for remission induction therapy for active non-severe EGPA (2). Still, remission induction with mepolizumab monotherapy is not common.

We herein report a patient with EGPA who was initially suspected of being complicated with pneumonia and sepsis and experienced rapidly progressive peripheral neuropathy. Remission was induced in the patient with mepolizumab monotherapy, which did not worsen the infection. This approach resulted in partial improvement of the peripheral neuropathy symptoms and subsequent remission with the addition of corticosteroids.

Case Report

A 72-year-old woman was referred to our hospital with a fever of 38°C, an elevated peripheral blood eosinophil count, and infiltrative shadows in both lung fields on chest computed tomography (CT). At the initial examination, the patient was conscious with a body temperature of 38.6°C, blood pressure of 172/92 mmHg, a pulse rate of 83/min, and saturation of percutaneous oxygen (SpO₂) of 98% in room air. Subcutaneous hemorrhaging and purpura were observed in her extremities. Neurological findings revealed no abnormalities in the central nervous system. Barre's sign was negative on both sides, and the results of manual muscle testing (MMT) were as follows: biceps: 5/5; triceps: 5/5; wrist flexors: 5/5; wrist extensors: 4/4; iliopsoas: 4-/4, quadriceps: 5/5; tibialis anterior: 1/1; and triceps femoris: 1/1 (left/right). An abnormal sensation was observed in her upper extremities, both hands, and her lower extremities beyond the wrists. Pain sensation was decreased at the dorsal surfaces of both hands and lower legs, and vibration sensation was decreased in her lower limbs.

Blood tests showed a white blood cell (WBC) count of 17,700 /μL (31% neutrophils, 62.0% eosinophils, 3.0% monocytes, 4.0% lymphocytes), hemoglobin level of 11.1 g/dL, platelet count of 33.1×10³/μL and marked eosinophilia. The erythrocyte sedimentation rate was 93 mm/h, and the C-reactive protein (CRP) level was 10.36 mg/dL, indicating an elevated inflammatory change. The creatinine kinase (CK) level was 63 U/L with normal myogenic enzyme levels, and the renal function was normal with a blood urea nitrogen (BUN) level of 16.2 mg/dL and creatinine (Cr) level of 0.60 mg/dL. Immunoglobulin (Ig) G, IgA, and IgM levels were 2,005, 229, and 236 mg/dL, respectively, and the IgE level was 2,181 IU/mL, with elevated IgG and IgE levels noted. Procalcitonin (PCT) levels were markedly elevated at 37.87 ng/mL. The fractional exhaled nitric oxide (FeNO) concentration was 18 ppb, which was almost normal. No significant bacteria were detected in venous blood or sputum cultures, although neutrophils were significantly increased in sputum.

A skin biopsy of the purpura was also performed, showing destroyed blood vessels in the dermis, with vasculitis, marked eosinophilic infiltration, and fibrinoid necrosis in the surrounding area (Fig. 1). She had a history of bronchial asthma, eosinophilia in the peripheral blood (WBC 17,700 /μL, Eo 62%; 10,974 /μL), a fever, mononeuritis multiplex, and purpura with vasculitis of eosinophilic infiltration. She was diagnosed with EGPA based on the four EGPA criteria of ACR in 1990 (4).

Figure 1.

Histopathological findings of a skin biopsy of the purpura. Right side: 40×. Left side: 100×.

On admission, chest CT showed infiltration shadows in the middle and lower lobes of both lungs, and pneumonia was strongly suspected (Fig. 2a). Magnetic resonance imaging (MRI) of her head showed no evidence of hypertrophic pachymeningitis, cerebral infarction, or sinusitis. Electrocardiography (ECG) showed a normal sinus rhythm with no findings suggesting ischemia or pericarditis. Echocardiography showed a left ventricular ejection fraction (LVEF) of 60.9% with a normal function and no asynergy. Upper and lower gastrointestinal endoscopy revealed no significant findings. The nerve conduction velocity was measured, revealing a decrease in the motor distribution of the left median nerve and right and left tibial nerves. There was also a decrease in the sensory distribution of the right and left median, ulnar, and sural nerves, suggesting axonal damage, especially in the motor distribution of the right and left tibial nerves and the sensory distribution of the sural nerves, with no response (Fig. 3, Table). The skin perfusion pressure (SPP), an indicator of peripheral circulatory disturbance, was measured at the left sural nerve and showed 28 mmHg, an indicator of severe ischemia. The Birmingham Vasculitis Score (BVAS) (5) was 20 points, and the Five Factor Score (FFS) (6) was 2 points.

Figure 2.

Computed tomography findings. a: On admission. b: After antibiotic therapy.

Figure 3.

Nerve conduction velocity findings. MCV: motor nerve conduction velocity, SCV: sensory nerve conduction velocity, N: nerve

Table.

Nerve Conduction Velocity Findings.

Moter nerves	MCV (m/s)	CMAP amplitude (mV)
Left median nerve	3.8	1.8↓
Right median nerve	3.3	9.7
Left ulner nerve	2.8	5.6
Right ulner nerve	2.8	4.7
Left tibial nerve	NR	NR
Right tibial nerve	NR	NR

Sensory nerves	SCV (m/s)	SNAP amplitude (μV)
Left median nerve	2.8	18.5↓
Right median nerve	2.9	17.4↓
Left ulner nerve	2.8	8.6↓
Right ulner nerve	2.9	7.6↓
Left sural nerve	NR	NR
Right sural nerve	NR	NR

MCV: motor nerve conduction velocity, SCV: sensory nerve conduction velocity, CMAP: compound muscle action potential, SNAP: sensory nerve action potential, SPP: skin perfusion pressure

On admission, the patient had a fever, significantly increased neutrophils in sputum, cough, markedly elevated CRP and procalcitonin levels, and infiltrating shadow on chest CT. Although no significant bacteria were detected in the venous blood and sputum culture, pneumonia and sepsis were strongly suspected as complications, and intravenous administration of antibiotics (1 g/day of meropenem and 2 g/day of vancomycin) were started. Patients with a poor prognosis and an FFS of 2 points and rapid progression of mononeuritis multiplex need early administration of immunosuppressive agents, such as high-dose glucocorticoids and/or intravenous cyclophosphamide therapy, which could not be given to the present patient because of uncontrolled complications of bacterial infection in addition to myeloperoxidase (MPO)-ANCA-negative EGPA. Therefore, mepolizumab monotherapy was started on the third day of treatment, after obtaining permission for off-label use at our hospital and after receiving informed consent from the patient.

Seven days after starting mepolizumab, the eosinophil count had decreased from 10,924 /μL to 1,833 /μL, and the CRP level had decreased from 17.7 mg/dL to 2.9 mg/dL, while the SPP had increased from 28 to 46 mmHg. Although the nerve conduction velocity could not be rechecked, the BVAS had decreased from 20 to 10 points. After reconfirming negative blood and sputum cultures, the absence of bacteria by bronchoalveolar lavage, and that the infiltrative shadows on chest CT had improved with antibiotic therapy (Fig. 2b), methylprednisolone pulse therapy (methylprednisolone 1,000 mg/day for 3 days) was started on day 8 while continuing treatment with oral prednisolone 40 mg. Intravenous immunoglobulin therapy (IVIG; 1,000 mg/day for 5 days) was administered on day 11 to treat the remaining peripheral neuropathy after obtaining permission for off-label use at our hospital and after receiving informed consent from the patient. Subsequently, the symptoms of peripheral neuropathy gradually subsided, and the BVAS decreased to 4 points. In addition, the SPP rose to 52 mmHg. The patient was transferred to another hospital on day 31 because of a gradual reduction in the dose of the corticosteroid medication and the need for rehabilitation (Fig. 4).

Figure 4.

Clinical course before and after the administration of mepolizumab. Eosino: eosinophil, mPSL: methyl prednisolone, PSL: prednisolone, IVIG: intravenous immunoglobulin, CRP: C-reactive protein, MEPM: meropenem, VCM: vancomycin, BVAS: Birmingham Vasculitis Score, SPP: skin perfusion pressure

Discussion

EGPA is classified as an ANCA-associated vasculitis and is known to cause a markedly elevated eosinophil count and prodromal symptoms, such as asthma and sinusitis, followed by various systemic symptoms (1). In the 2021 ACR guidelines for the management of ANCA-associated vasculitis, the recommendation was to start treatment with high-dose glucocorticoids in combination with cyclophosphamide or rituximab for active severe EGPA. For active non-severe EGPA, treatment was to be done with glucocorticoids in combination with mepolizumab, including immunosuppressive drugs, such as methotrexate, azathioprine, mycophenolate mofetil, and rituximab, or monotherapy with glucocorticoids (2). EGPA is known to cause various types of organ damage, and in Japan, neuropathy with mononeuritis multiplex is very common, being seen in 93% of all cases (7). Nerve bundles in peripheral nerves are nourished by microvessels, and in ANCA-associated vasculitis (including EGPA), nutrient vessels are damaged, causing peripheral nerve ischemia and axonal degeneration, thus resulting in neuropathy. The regeneration of axons with treatment is very slow and may take several years. It is also known that if treatment is initiated too late, neuropathy may not subside, even if blood flow disturbance improves (8). Prolonged peripheral neuropathy is one of the causes of reduced activities of daily living (ADLs) in patients (9). Therefore, EGPA with progressive peripheral neuropathy should be diagnosed as early as possible and treated aggressively to restore the normal function (10). Although the efficacy of IVIG in the treatment of peripheral neuropathy in EGPA has been reported (11,12), the evidence for this is limited.

The pathology of EGPA neuropathy is characterized by necrotizing vasculitis in ANCA-positive cases. However, ANCA-negative cases show marked eosinophil infiltration of capillaries within the neuroepithelium and nerve sheath, suggesting that eosinophils may directly cause neuropathy (13). Therefore, it is necessary to correct blood flow obstruction caused by perivascular eosinophilic infiltration as early as possible to relieve neuropathy in ANCA-negative EGPA. In the present case, early remission induction therapy with glucocorticoids and immunosuppressive drugs was necessary for managing progressive peripheral neuropathy. However, infiltrating shadows were seen in the bilateral lower lung fields on CT at the time of the initial diagnosis, making it important to differentiate between pulmonary involvement of EGPA and bacterial pneumonia.

Suzuki et al. (14) reported that the most common pulmonary lesion in EGPA is bronchial wall thickening, while the other lesions are mainly ground-glass opacity (GGO) appearance and thickening of the interlobular septum. CT at admission showed mainly infiltrating shadows in the bilateral lower lung fields, with no thickening of the interlobular septum and/or a GGO appearance. No bacteria were found in the sputum, but the cells in the sputum showed significantly neutrophil. FeNO was also not high at 18 ppb. On reviewing these results with a respiratory medicine specialist, it was determined that the lung lesions in this case were more likely to be bacterial pneumonia than EGPA lung lesions. Therefore, this case could not be treated immediately with strongly immunosuppressive therapy for these reasons.

In addition, PCT was elevated in our case. PCT is known to be increased in cases of bacterial pneumonia and sepsis but not in viral infections or autoimmune diseases. Akikawa et al. (15) reported that PCT is the best biomarker for differentiating bacterial from non-bacterial infections, compared to CRP, IL-6 and endotoxin. The clinical findings show that, in patients diagnosed with bacteremia, the positive rate of a PCT diagnosis is 70.2%, while the positive rate of a blood culture diagnosis is 42.6%; furthermore, PCT ≥2.0 ng/mL indicates a high risk of severe sepsis. Joo et al. (16) also showed that PCT is superior to CRP in differentiating bacterial infections from disease flare-ups in autoimmune diseases. Although no bacteria were detected in the blood culture in this case, the PCT value was extremely high (37.87 ng/dL), so we strongly suspected a combination of bacterial pneumonia and sepsis and therefore decided that the use of high-dose glucocorticoids and immunosuppressive drugs for remission induction therapy was risky.

The efficacy of mepolizumab in treating neuropathy associated with peripheral circulation disorders in ANCA-negative EGPA has been previously reported (17). Furthermore, early introduction of mepolizumab reportedly improves peripheral neuropathy of EGPA at an early stage (18); therefore, in this case, we started treatment with mepolizumab monotherapy to stop the progression of peripheral neuropathy. After treatment with mepolizumab, the peripheral blood eosinophil count decreased markedly, the BVAS improved from 20 to 10 points, the SPP increased from 28 to 46 mmHg, and the patient's subjective symptoms partially improved. The SPP has been reported to be useful in the evaluation of neuropathy due to peripheral circulatory disturbance with ANCA-negative EGPA (17). Mepolizumab monotherapy partially improved the SPP in this patient. After confirming that pneumonia and sepsis had improved, the patient was treated with additional glucocorticoids and IVIG, and her symptoms improved further. We successfully demonstrated a decrease in the peripheral blood eosinophil count and partial improvement in the BVAS and SPP with mepolizumab monotherapy before the use of glucocorticoids and IVIG in our patient. Therefore, we attribute these improvements to mepolizumab. Although there has been one report of induction remission therapy for EGPA with a combination of mepolizumab and IVIG (19), no report exists concerning remission induction with mepolizumab monotherapy. To our knowledge, this is the first report of remission induction with mepolizumab monotherapy.

Mepolizumab is a humanized anti-IL-5 monoclonal antibody that inhibits eosinophil production and proliferation by binding to IL-5 and blocking its interaction with the IL-5 receptor (3). It is not expected to suppress Th1-dominant cytokines, and the frequency of complicated infections is expected to be low. Clinical trials of one-year use of mepolizumab in bronchial asthma reported a lower incidence of upper respiratory tract infections than in the placebo group (20). In addition, a study on the long-term safety of mepolizumab reported that there were no records of serious infections associated with bronchial asthma over a 3.5-year course of using mepolizumab (21). Furthermore, in a systematic review of the safety and efficacy of mepolizumab, no increase in severe infections was reported (22). There was a case in which mepolizumab was used to reduce the dose of steroids in allergic bronchopulmonary aspergillosis associated with nontuberculous mycobacteriosis (23), and the use of mepolizumab in patients with infectious complications is expected to be a treatment option.

In conclusion, we used mepolizumab monotherapy to treat a patient with EGPA complicated with pneumonia and sepsis. This treatment resulted in the early improvement of symptoms of EGPA and infection control. Although the use of mepolizumab as monotherapy is still limited, and we understand that it is difficult to achieve remission with mepolizumab alone, it has proven to be a viable treatment option that should be considered for cases of ANCA-negative EGPA complicated with infection. Similar case reports to the present one will help provide a sounder scientific basis for using mepolizumab as monotherapy.

Author's disclosure of potential Conflicts of Interest (COI).

Noboru Kitamura: Honoraria, GlaxoSmithKline.