Proteinase 3 (PR3)-antineutrophil cytoplasmic antibody (ANCA)-associated vasculitic neuropathy in diffuse cutaneous systemic sclerosis: a rare duo

Abstract

Systemic sclerosis (SSc) is characterised by non-inflammatory vasculopathy, autoimmunity and widespread fibrosis. While the presence of antineutrophil cytoplasmic antibodies (ANCAs) has been reported in SSc, their association with ANCA-associated vasculitis is exceedingly rare. Myeloperoxidase ANCA is more common than proteinase-3 ANCA, and glomerulonephritis is the most common clinical presentation of ANCA-associated vasculitis in SSc. ANCAs have been associated with the adverse disease outcomes in SSc, including higher mortality per recent reports. A 65-year-old man with diffuse cutaneous SSc for 6 years presented with new-onset peripheral neuropathy. Workup revealed a positive proteinase-3 and cytoplasmic ANCA, and histopathology confirmed an inflammatory vasculitic neuropathy. The patient was successfully treated with rituximab. Our case highlights the importance of checking ANCA in SSc at baseline, given the risk of disease-related complications, even years later. Tissue biopsy is often warranted for confirmation of vasculitis and prompt treatment can optimise long-term outcomes.

Background

Systemic sclerosis (SSc) is a complex systemic disease characterised by progressive fibrosis of the skin and internal organs, a non-inflammatory vasculopathy and immune system activation. A broad array of autoantibodies characterises SSc, and has diagnostic, clinical and prognostic significance. Anticentromere antibodies are typically associated with limited cutaneous SSc (lcSSc) and pulmonary arterial hypertension (PAH), anti-Scl-70 antibody with diffuse cutaneous SSc (dcSSc) and interstitial lung disease (ILD), and RNA polymerase III antibody with SSc renal crisis and malignancy. While antineutrophil cytoplasmic antibodies (ANCAs) typically characterise a group of conditions associated with small vessel vasculitis, they have also been reported infrequently in SSc.^1–3 In the large majority of SSc cases, they are not associated with systemic vasculitis and clinical significance is unclear.

Case presentation

A 65-year-old man with SSc presented with a new persistent paresthesia of the lateral right lower leg and dorsolateral right foot with intermittent leg cramping.

Six years prior to current presentation, he presented to our clinic with progressive fatigue for 6 months, dyspnoea on exertion, productive cough, postprandial abdominal pain, anorexia and unintentional weight loss of 15 lbs. He reported a decline in health since Raynaud’s began 4 years prior. He also reported hand stiffness, swelling and diminished dexterity compromising his job as an electrician. He had noticed tightening of the skin over his arms and legs. He reported numbness and tingling over the left cheek with reduced sensation to light touch and temperature, and drooping of the left side of his mouth. Physical examination confirmed sclerodactyly, skin sclerosis involving his dorsal hands extending up to the upper arms, lower legs and trunk, digital pitting and telangiectasias over palms and face. Autoimmune serologies revealed a strongly positive ANA (>12 U), anti-ribonucleoprotein (RNP) antibody and anti-Smith antibody. Anti-Scl70, anticentromere and anti-RNA Pol III antibody were negative. At baseline, a patient had minimal proteinuria of 24 mg/dL (normal <22 mg/dL), which normalised ranging between 10 and 12 mg/dL over the last few years. His renal function was normal (serum creatinine 0.8 mg/dL) and there were no features of scleroderma renal crisis.

He had moderate-to-severe ILD on CT chest (non-specific interstitial pneumonia pattern) with restrictive lung physiology (pulmonary function tests (PFTs) showed a forced vital capacity (FVC) 68% predicted, forced expiratory volume (FEV₁) 77% predicted, FVC/FEV₁ ratio of 87% and diffusing capacity of the lungs for carbon monoxide 57% predicted). Estimated right ventricular systolic pressure was 49 mm Hg. The patient underwent an upper extremity vascular ultrasound with laser Doppler flowmetry, which confirmed typical occlusive digital vasculopathy suggestive of scleroderma along with secondary vasospasm. There was no evidence of proximal upper extremity arterial occlusive disease or other features to support a vasculitis. Nailfold videocapillaroscopy showed an active SSc pattern (figure 1). EMG showed trigeminal neuropathy with a length-dependent sensory and motor peripheral neuropathy.

Figure 1.

Nailfold videocapillaroscopy (NVC, Optilia, x200) showing active SSc NVC pattern with numerous capillary microhaemorrhages, dilated and tortuous capillary loops, and significant pericapillary oedema. SSc, systemic sclerosis.

He met the ACR/EULAR 2013 criteria for SSc, was diagnosed with dcSSc complicated by ILD and peripheral neuropathy. He did not have any features clinically to suggest lupus or myositis. Mycophenolate mofetil (2 g daily) was initiated with good response over the next several months; the dose was later reduced to 500 mg two times a day due to gastrointestinal side effects. Skin sclerosis and dyspnoea improved, cough resolved, weight and appetite normalised. CT chest and PFTs stabilised. Symptoms of trigeminal and peripheral neuropathy also resolved completely and overall he felt significantly better. For the following 3 years, he was stable with no significant flares of disease with the exception of progressive gastrointestinal involvement including gastro-oesophageal reflux disease, recurrent esophagitis and diminished small bowel transit.

In 2016, he was enrolled into a phase III trial evaluating the efficacy and safety of nintedanib in patients with SSc-ILD.⁴ Over the course of the study, his study medication dose (placebo/nintedanib) was reduced for diarrhoea, but otherwise continued to show improvement in pulmonary symptoms and functional capacity. In August 2018, 2 years after participating in the study, he began the open-label extension trial and received nintedanib 100 mg two times a day. Within a few weeks, he reported intermittent right lower leg cramping. It was generally occurring at rest and lasting for few minutes. Over next 5 weeks, symptoms continued progressing and began occurring more frequently. He presented with a new persistent paresthesia of the lateral right lower leg and the dorsolateral right foot with intermittent cramping after activity. No weakness or foot drop was reported. A neurology referral was made.

Investigations

EMG showed moderate length dependent large fibre peripheral neuropathy with axonal features. Differential diagnosis for peripheral neuropathy included SSc-related peripheral nerve involvement, metabolic aetiologies of neuropathy, neoplastic or paraneoplastic neuropathy, drug-induced nerve dysfunction, versus another aetiology for peripheral nerve involvement.

Further evaluation demonstrated stable SSc including ILD on CT chest. No evidence of malignancy was evident on chest CT and patient was up to date on age-appropriate malignancy screening. Serum protein electrophoresis demonstrated polyclonal hypergammaglobulinaemia, without monoclonal protein. Autoimmune serologies were unchanged in comparison to baseline, besides positive proteinase 3 (PR3) ANCA of 6.1 and cytoplasmic (c-ANCA) 1:32. With positive ANCAs, a small to medium vessel vasculitis was expected and workup was directed to assess for this. Neuropathy is not typically a feature of large vessel vasculitis such as temporal arteritis. The patient did not report any symptoms of large vessel vasculitides such as headache, scalp tenderness, jaw claudication, carotidynia or constitutional symptoms.

A sural nerve biopsy (figure 2) was performed and revealed perivascular inflammatory infiltration of a large arteriole with vessel wall disruption and destruction, leucocytoclasia and associated haemosiderin deposition. It also revealed moderate-to-severe decreased density of myelinated nerve fibres.

Figure 2.

Sural nerve biopsy: arrow sign demonstrates large arteriole showing leucocytoclasia and prominent inflammatory infiltration and vessel wall disruption along the centre of the longitudinal section of the sural nerve, concerning for inflammatory necrotising vasculitis; as shown in H&E stain (A) and Masson’s trichrome stain (B). Haemosiderin deposition, haemosiderin engulfed macrophages (amber), haemosiderin in the epineurium and perineurium (turquoise) was demonstrated by Congo red stain (C) and Turnbull’s blue stain (D).

Treatment

Overall, the findings were suggestive of a large arteriole vasculitis as can be seen in ANCA (PR3/c-ANCA)-associated vasculitic neuropathy. Given that a possible drug-induced vasculitis was difficult to rule out conclusively, nintedanib was held briefly but restarted after reassurance was provided by neurology. He was started on rituximab, 1000 mg X 2 doses over 2 weeks.

Outcome and follow-up

The patient showed improvement in neuropathic symptoms over the next several weeks after starting rituximab. Cramps and paresthesia improved, and his pulmonary symptoms were stable. The patient did not note any decline in neurological function after restarting nintedanib open-label therapy, making the likelihood of a drug-induced aetiology unlikely.

Discussion

ANCAs are autoantibodies directed against enzymes within primary granules of neutrophils and lysosomes in monocytes. There are three distinct subtypes based on immunofluorescence patterns on ethanol-fixed neutrophils; c-ANCA directed against PR3, perinuclear (p-ANCA) directed against myeloperoxidase (MPO) or atypical ANCA directed against other antigens (elastase, lactoferrin, etc).⁵

ANCAs typically characterise and are involved in the pathogenesis of ANCA-associated vasculitis (AAV), a group of conditions (that include granulomatosis with polyangiitis, microscopic polyangiitis and eosinophilic granulomatosis with polyangiitis) characterised by small to medium vessel vasculitis that can impact multiple organs, commonly the lungs and kidneys, but also joints, skin and nerves.⁶ ANCAs have been also reported in other autoimmune diseases with or without evidence of vasculitis, such as MPO-p-ANCA in rheumatoid arthritis, systemic lupus erythematosus, Sjogren’s syndrome, inflammatory myositis and inflammatory bowel disease.⁷ Atypical ANCA has been reported in ulcerative colitis and Crohn’s disease.

The presence of ANCA is relatively uncommon in sera of patientswith SSc, ranging around 0%–12% in reports.^1–3 Moreover, even when present their clinical significance is often unclear as majority is not associated with AAV clinically and the overlap of SSc-AAV is exceedingly rare (0.2%–0.8%).^{3 8 9} Moxey et al reported a positive ANCA in 116 of 1303 patients with SSc (8.9%) in the Australian Scleroderma Cohort Study (18 (13.8%) PR3-ANCA, 13 (11.2%) MPO-ANCA).⁹ Only three ANCA positive patients developed AAV (0.2% of the entire SSc cohort).⁹

The overlap of SSc-AAV is reported more commonly among those with lcSSc (79%–88%),^{8 10} those with Scl-70 antibody (59%–77%).^{3 11} and most commonly reported organ involvement is glomerulonephritis (often renal limited vasculitis). While vasculitic peripheral neuropathy has been rarely reported,^8–10 in conjunction with other vasculitic manifestations in SSc-AAV, an isolated vasculitic neuropathy as a manifestation of SSc-AAV overlap is exceedingly uncommon. The large majority of SSc-AAV overlap cases reported in the literature are MPO-p-ANCA positive and presence of PR3-c-ANCA is very uncommon (table 1).

Table 1.

Reports describing vasculitic neuropathy alone or with other features of vasculitis among cases of SSc-AAV overlap syndrome

First author (reference)	Type of study	AAV-SSc overlap (% of total SSc cohort)	Vasculitic neuropathy	Other major manifestations of vasculitis	p-ANCA	c-ANCA	Anti-MPO	Anti-PR3
Our patient	Case report	1	1	None	0/1	1/1	0/1	1/1
Liang10	Case series	14	5	Renal*-7 Alveolar haemorrhage-5 Skin†-6	10/10	0/10	10/10	1/7
Derrett-Smith8	Retrospective cohort	8/2200 (0.4%)	2/8	Renal*-7 Skin†-4	7/8	0/8	7/8	1/8
Quéméneur3	Retrospective cohort	9/1120 (0.8%)	5/9	Renal*-3 Skin†-5	6/7	0/7	8/9	0/9
Moxey9	Cross-sectional	3/1303 (0.23%)	1/3	Renal*-2	1/3	0/3	2/3	0/3

*Renal- glomerulonephritis or interstitial nephritis.

†Skin-cutaneous vasculitis or purpura.

AAV, ANCA-associated vasculitis;ANCA, antineutrophil cytoplasmic antibody;c-ANCA, cytoplasmic-ANCA; MPO, myeloperoxidase;p-ANCA, perinuclear-ANCA; PR3, proteinase 3; SSc, systemic sclerosis.

In another cohort of 2200 patients with SSc,⁸ only eight had AAV (0.4%), seven had lcSSc and all seven patients had MPO-ANCA. Only two of these had vasculitic peripheral neuropathy.⁸ In a report from 2013,³ of 1120 patients with SSc reviewed, nine SSc-AAV overlap were reported, six of seven tested were p-ANCA positive, eight of nine tested MPO-ANCA positive and none had a PR3 or c-ANCA.³ In the study of Moxey et al referenced above,⁹ three SSc-AAV overlap patients had either MPO or p-ANCA while none had PR3 or c-ANCA. Vasculitic neuropathy was reported only in one patient.⁹

Presence of ANCA in SSc has been associated with increased prevalence of renal and pulmonary comorbidity and believed to impart an inflammatory component to the illness.¹² The clinical significance of ANCAs in SSc, besides their association with vasculitis, was recently examined.⁹ ANCA positivity was associated with higher prevalence of ILD (45% vs 22%, p<0.001) even after adjusting for Scl-70 antibodies. Pulmonary embolism was more common in ANCA positive (9% vs 3%, p=0.002) and PR3-ANCA positive patients with SSc (17% vs 3%, p=0.02), even after adjusting for the presence of antiphospholipid antibodies raising suspicion for their possible prothrombotic effect even in the absence of AAV. A higher incidence of malignancies was also noted among ANCA positive patients with SSc (27% vs 19 %, p=0.04),⁹ ANCAs were also associated with increased mortality (p=0.006), after adjusting for age and sex.⁹ ANCA positivity can, thus, be considered a red flag and sign for complex disease course in patients with SSc, prompting a thorough investigation for systemic complications and close follow-up.^{9 12}

Implementation of new biologics in the treatment of malignant and autoimmune diseases also carries the risk for uncommon side effects such as drug-induced neuropathy or vasculitides and these situations can present additional diagnostic challenges. Nintedanib is a triple tyrosine kinase inhibitor that inhibits the vascular endothelial growth factor receptors, fibroblast growth factor receptors and platelet-derived alpha and beta growth factor receptors. It also blocks intracellular signalling and prevents proliferation, migration and transformation of fibroblasts implicated in the pathogenesis of SSc.¹³ It has well-documented benefit in preventing decline of lung function in IPF, and more recently was also confirmed to demonstrate benefit in SSc-ILD.⁴ Nintedanib has several known side effects from previous reports that include gastrointestinal disturbances, weight loss, decreased appetite or hypertension. Serious but uncommon side effects include arterial thromboembolism, myocardial infarction, gastrointestinal perforation and drug-induced liver disease.⁴ A few studies have reported the association of tyrosine kinase inhibitors with neuropathies; sunitinib and imatinib have been reported to cause an acute polyneuropathy,¹⁴ and cranial sensory disturbances have been reported with gefitinib used in the treatment of malignancies.¹⁵ There is also a report of severe acute disabling motor neuropathy with flaccid tetraparesis in a 59-year-old woman 12 weeks after starting add on therapy with nintedanib for ovarian cystadenocarcinoma.¹³ The patient was also treated with paclitaxel, which has been associated with a slowly developing subacute neuropathy, and carboplatin that has been reported to cause dorsal root ganglionopathy. While the aetiology of neuropathy could not be conclusively determined, the authors suspected an increased vulnerability to neurotoxicity from tyrosine kinase inhibition and recommended a high degree of vigilance by the treating physician.¹³

Our case is unique in several respects. Our patient had dcSSc, which is uncommonly associated with SSc-AAV overlap than lcSSc variant. Our patient was positive for anti-RNP antibody and anti-Smith (which can be seen in patients with SSc, besides mixed connective tissue disease (MCTD)), but negative for Scl-70 which has been typically associated with SSc-AAV overlap. Interestingly, AAV has been reported to develop more frequently among patients with other overlap connective tissue diseases such as MCTD or lupus¹⁰ and it is plausible that his serologies placed him at risk. Our patient also had PR3-c-ANCA with isolated vasculitic neuropathy (without glomerulonephritis, sinus disease or pulmonary manifestations such as capillaritis), a duo which is rare in SSc-AAV overlap cases described thus far in the literature.

In summary, SSc-AAV overlap is very rare, and often requires histopathological proof of vasculitis. While majority of these occur in patients with lcSSc and positive MPO/p-ANCA, they can occur among those with dcSSc and those with positive PR3-c-ANCA. Vasculitic neuropathy is a rare manifestation of SSc-AAV, but with development of new neuropathic symptoms in patients with SSc on stable immunosuppression, it should be considered in the differential diagnosis and evaluation with ANCAs is indicated in the workup. While it is difficult to attribute the patient’s neuropathic symptoms to nintedanib, its increasing utilisation for SSc-ILD in the future will assist in understanding whether neuropathy or a drug-induced vasculitis can develop during its ongoing use. Until then, ongoing vigilance per treating physician is necessary and clinical correlation is advised. Given the recent reports of adverse clinical outcomes among patients with SSc with positive ANCAs, including the higher risk of thromboembolism, malignancy and mortality, it may be worth screening all patients with SSc for ANCA at baseline even in the absence of vasculitis, to ascertain long-term prognosis.

Learning points.

• Majority of antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis (AAV)-systemic sclerosis (SSc) overlap syndrome described thus far in the literature presents among limited cutaneous SSc variant, with Myeloperoxidase-perinuclear-ANCA on laboratory testing and renal involvement being the most common clinical presentation.

• SSc-AAV overlap is very rare, and often requires histopathological proof of vasculitis.

• We describe SSc-AAV overlap in a diffuse cutaneous SSc variant, presenting with a very rare duo of proteinase 3-cytoplasmic-ANCA with isolated vasculitic neuropathy (without glomerulonephritis, sinus disease or pulmonary manifestations such as capillaritis).

• Positive ANCAs among patients with SSc have been associated with adverse clinical outcomes (including a higher risk of thromboembolism, malignancy and mortality) and it may be worth screening all patients with SSc for ANCA at baseline even in the absence of vasculitis, to ascertain long-term prognosis.