IgG4-related membranous glomerulonephritis and generalized lymphadenopathy without pancreatitis: a case report

This atypical presentation of a rare pathology (i.e. the patient never experienced any episode of pancreatitis) made difficult the diagnosis of IgG4-RD. Moreover, IgG4-RD was not well known at the time of the first symptoms in 2004 and PLA2R antibodies were not available in daily practice at that time.

First of all, hematologic malignancies had to be excluded in the presence of lymphadenopathy and biological abnormalities. Moreover, a higher risk of developing lymphoma or neoplasia is highly suspected in IgG4-RD patients [4]. Other systemic diseases such as Sjögren’s syndrome, sarcoidosis and especially systemic lupus erythematous (SLE) were also part of the differential diagnoses. A diagnosis of SLE with associated immune MGN (lupus “full-house” deposits were not found on renal biopsy) and reactive lymphadenopathy had been proposed in 2008 and maintained until 2012 in view of the presence of anti-dsDNA, thrombocytopenia, arthritis and history of cutaneous rash. Other entities occasionally associated with high serum IgG4 level or with IgG4+ plasma cells on biopsy samples were also ruled out (i.e. granulomatosis with polyangiitis (GPA), Castleman’s disease or McDuffie hypocomplementemic urticarial vasculitis [4]).

IgG4-RD is considered as definite in this case because all three comprehensive diagnostic criteria are met: diffuse swelling in lymph nodes, elevated serum IgG4 concentration ≥ 135 mg/dL (2790 mg/dL in 2012) and lymph node infiltration of IgG4 + plasma cells with ratio of IgG4+/IgG+ cells > 40% and >10 IgG4+ plasma cells per HPF [5].

MGN are responsible for 30% of adult nephrotic syndromes. Eighty percent of MGN are idiopathic whereas the other 20% are caused by various pathologies which we searched for such as infections, cancers and systemic diseases. Circulating antibodies to the PLA2R (which is a transmembrane protein located on podocytes) are found in 70% of idiopathic membranous glomerulonephritis (IMGN). These antibodies are of IgG4 subclass but high serum IgG4 levels are not encountered in IMGN. The observation of glomerular deposits of PLA2R antibodies on biopsy seems to be more sensitive than their detection in serum. However, this could not be done here giving the absence of a recent renal biopsy [3, 6].

Kidney involvement in IgG4-RD or IgG4-RKD occurs in approximately 15% of IgG4-RD patients. It has the remarkable property to result in two well distinct types of lesions: first, TIN which is most commonly encountered and second, glomerulopathy. Pyelitis have also been described. Glomerular involvement is very rare (7% of IgG4-RKD). Lymphoplasmocytic infiltrate and storiform pattern of fibrosis need not be present for its diagnosis. MGN is the only glomerular lesion that can be reliably attributed to IgG4-RD. However, other forms of glomerulopathy such as mesangioproliferative glomerulonephritis, IgA nephropathy and membranoproliferative glomerulonephritis have also been described by Saeki et al. [7]. Usually, glomerular lesions coexist with TIN as in our patient’s first renal biopsy. Isolated glomerular involvement in IgG4-RKD is uncommon (none of the biopsies studied by Saeki et al. [7] and in four of the nine biopsies studied by Alexander et al. [8]). Unfortunately, in our case IgG subclass could not be determined retrospectively on the renal biopsies of 2004 and 2008 because of insufficient material. However, anti-PLA2R antibodies were negative in the blood which makes IMGN less likely and points towards IgG4-related glomerulonephritis. This is supported by the good response to GC, the associated TIN in the first renal biopsy, the documentation of IgG4-RD in lymph nodes and very high serum IgG4 levels [2, 3, 7‐10]. Jindal et al. [11] have described a similar case report of IgG4-related MGN diagnosed through nephrotic syndrome and lymphadenopathy. In that case, an AIP was associated (which is more usual) and end-stage renal disease with need for dialysis could not be avoided despite GC and rituximab [11].

Specific diagnostic criteria exist for IgG4-RKD [12]. They were used for our patient and confirmed definite IgG4-RKD although further analyses could not be performed on kidney biopsy to assess the presence of IgG4+ plasma cells. Thus IgG4-related generalized lymphadenopathy and IgG4-related kidney disease with MGN and TIN are definite in this case report according to such criteria. Skin and joint involvement would have needed further investigations to be confirmed. Thrombocytopenia has never been reported before such a manifestation of IgG4-RD. Its etiology in this case report actually remains unclear.

All patients with active IgG4-RD and a subset of asymptomatic patients require treatment, sometimes urgently. GC are currently the main first-line treatment with initial good clinical and biological response, especially when a low degree of fibrosis is present. However, relapses are very common as seen in this case report and maintenance GC is often required after induction therapy. Patients who relapse for the first time after successful induction should be inducted again by GC. Introduction of steroid-sparing immunosuppressive agents should be considered for following relapses due to the frequency of side effects associated with long term GC therapy [13]. Mycophenolate mofetil, methotrexate or azathioprine have shown some efficacy in IgG4-related pancreatitis [1, 14]. Rituximab is one of the most promising treatment of relapsing IgG4-related diseases. Rituximab is an anti-CD20 chimeric antibody that acts by depleting CD20 positive B-lymphocytes. Good response has been seen even without initial or concomitant GC therapy and it allowed to discontinue GC in patients with steroid dependence in several published cases. Serum IgG4 level is also lowered by this treatment. Clinical and biological response to rituximab is maintained even after B cell reconstitution [1, 3, 15]. However, in the case reported by Jindal et al. [11], renal function did not improve after rituximab therapy perhaps because its late introduction. Further investigations are needed to assess the exact position and optimal regimen of rituximab, in IgG4-RKD [11].

It is interesting to note that the patient received CYP in 2008 for immune thrombocytopenia. This treatment was followed by 3 years without any manifestation of the IgG4-RD, in the absence of maintenance GC. CYP was also given with GC to a patient with IgG4-related TIN in Australia, and similarly allowed clinical and radiological remission as well as stabilization of renal failure [14]. Alexander et al. [8] have described a good response of proteinuria in one patient with IgG4-related MGN treated by prednisolone and CYP. More experience with CYP is required to confirm its effectiveness in IgG4-RD.

IgG4-RD follow-up is based on biological markers such as serum CH50, C3c, C4, IgG and IgG4 values. In case of relapsing IgG4-RD, a lowering of serum complement level is classically noticed (especially in IgG4-RKD) as well as increased IgG and IgG4 levels, as shown in Fig. 3 [2]. However, recent studies showed that these markers are poor predictors of the need for additional treatment. More attention should be paid to clinical manifestations [3]. PET/CT has a valuable role in mapping lesions, staging extent of the disease and guiding biopsy. Some studies showed that it could also be useful in monitoring response to treatment and identifying disease relapse earlier than other imaging techniques [4].

IgG4-RD pathophysiology is not completely understood yet. Excess of T helper 2 (Th2) cells and regulatory T (Treg) cells has been described in involved organs. This could explain fibrosis (since transforming growth factor-beta is secreted by Treg) and increased production of IgG4 (because cytokines secreted by Th2 and Treg stimulate IgG4 class switch). Excess of Th2 can also explain IgE excess and hyper eosinophilia [1, 16]. This immune imbalance could be due to abnormal reaction against pathogenic agents (i.e. Helicobacter pylori) or to commensals belonging to normal microbiota with continuous presentation of antigen by B cells. This could explain clinical improvement observed following B-cell depletion by rituximab [3]. In genetic studies, higher prevalence of HLA-DRB10405 and HLA-DQB10401 has been reported in Japanese patients with IgG4-RD [1].

An important question remains: are IgG4 pathogenic or not in IgG4-RD? The recent discovery of serum IgG4-negative IgG4-RD raises questions about the causative role of the IgG4 molecule [2]. ANA are frequently found in IgG4-RD patients as seen in this case report and are suspected to be pathogenic but do not belong to the IgG4 subclass [17]. Moreover, IgG4 is a relatively non-inflammatory antibody because it undergoes heavy-chain exchange leading to monovalent but bispecific antibody with reduced ability to crosslink antigens and form immune complexes. The amino acid sequence in its CH2 domain also limits IgG4 interaction with C1q and Fcγ receptors [3, 18]. Nevertheless, recent studies show the capacity of IgG4 to form aggregates which can cause inflammatory lesions in the absence of antigen [19]. So, IgG4-RD can be considered either as an allergic disease where IgG4 are only a marker of Th2 and Treg excess in tissues [1] or can be seen as an immune disorder where IgG4 could play by itself in the pathology [7].