There is evidence for the association of environmental triggers in autoimmune diseases. For instance, silica exposure has been linked to development of rheumatoid arthritis, systemic sclerosis, ANCA-associated vasculitides, and others [1–3]. Further, the association of smoking and several autoimmune diseases is well established [4]. A recent study in a Dutch IgG4-RD cohort consisting of IgG4-related cholangitis and pancreatitis patients reported that 88 % were blue-collar workers chronically exposed to solvents, oils, industrial and metal dusts, and others. These findings were similar in another IgG4-RD cohort from Oxford reported in the same study [5]. The role of occupational exposure and smoking has been investigated in a large retroperitoneal fibrosis (RPF) cohort form Italy. Asbestos and smoking were both independently and additively associated with RPF in this study, with relatively high odds ratios ranging from 2.93 to 4.22 [6]. Indeed, asbestos exposure has already been linked to RPF in another study from Finland [7]. However, RPF may only be partly associated with IgG4-RD, and therefore caution is mandatory to generalize these findings to IgG4-RD.

Currently, there are no genetic association studies in IgG4-RD cohorts. There is however evidence for genetic associations of, e.g., HLA system in chronic periaortitis/retroperitoneal fibrosis and autoimmune pancreatitis patients discussed in detail in Chap. 2.

In IgG4-related sialoadenitis, pancreatitis, and cholangitis, studies have found a shift of the immune system towards a Th2 response, which is discussed in detail later. Moderate tissue and blood eosinophilia in IgG4-RD is not uncommon, and serum IgE elevation is also frequently observed [8]. While it was initially thought as possibly pathogenic for the disease itself, a recent study challenges this view. In a cohort of 70 biopsy-proven IgG4-RD patients, 31 % had a background of atopic disease, most commonly allergic rhinoconjunctivitis [9]. This is not different from the expected background frequency of atopic diseases in the USA according to the authors. IgE elevation and eosinophilia were much more frequent in patients with various atopic diseases. Also, circulating Th2 memory cells were found preferentially in the latter patients [10]. Thus, the Th2 immune response may be present in IgG4-RD patients with an additional atopic disease. Whether atopic disease is causally linked to IgG4-RD or not remains to be defined. In other clinical IgG4-RD cohorts, allergic diseases have however been described to occur more frequently (more than 50 % of patients) [11, 12].

Alterations in the adaptive immune system have been investigated in detail in several studies. In biopsy specimens from IgG4-related pancreatocholangitis and other lesions, high expression of cytokines associated with a regulatory T-cell and Th2-cell phenotype such as IL-4, IL-5, IL-10, IL-13, and TGF-beta was noted [13]. These cytokines could therefore drive the major fibrotic disease phenotype and also induce the immunoglobulin class switch to IgG4 production [13]. As discussed above, the frequent comorbid atopic diseases and elevated IgE levels in IgG4-RD patients are well in line with this theory. Another study also found increased IL-4 and IL-5 expression in bile of IgG4-related cholangitis patients. These cytokines disturbed barrier function of bile epithelial cells in vitro [14]. In IgG4-related sialoadenitis, increased local expression of, e.g., IL-4, IL-5, IL-10, and TGF-beta was also found when compared to disease controls such as Sjögren’s syndrome [15].

However, whether these cytokines are truly produced by T cells is yet unclear. Moreover, other studies reported different findings challenging this theory. In IgG4-related sialoadenitis, increased expression of local Th1 and cytotoxic T cells but not Th2 cells was found as evidenced by intracellular cytokine flow cytometry. Also, increased IL-17 tissue expression was found [16]. Della Torre et al. investigated T-cell polarization in peripheral blood of IgG4-RD patients. They found increased Th2 immune responses only in IgG4-RD patients with concomitant atopic disorders, questioning a true relationship with IgG4-RD itself [10].

Very recently, the same authors extended their findings on T-cell polarization in IgG4-RD [17]. CD4 + CD27^loCD62L^lo cells, indicative of effector memory T cells usually arising from chronic antigen stimulation, were highly increased in active IgG4-RD. Using a variety of analytical methods, including gene expression profiling, these cells seem to be of a modified Th1 type showing increased expression of T-bet along with SLAMF7 and 2B4. Interestingly, these cells have potent cytolytic activity in vitro after stimulation. These CD4 + SLAMF7+ cytotoxic T cells (CTLs) but not Th2 cells were oligoclonally expanded in active IgG4-RD patients. CD4 + SLAMF7+ CTLs were increased in peripheral blood samples but also abundant in tissue lesions of IgG4-RD patients, outnumbering Th2 cells. In vitro, these cells can produce profibrotic cytokines such as TGF-beta and IL-1beta. In patients successfully treated with rituximab, the concentration of these cells in the peripheral blood successfully decreased [17].

Another study linked IL-21 overproduction to the pathogenesis of IgG4-RD [18]. IL-21 is an important cytokine for the formation of germinal centers, the latter being frequently found in IgG4-related sialoadenitis. IL-21 was overexpressed in minor salivary glands from IgG4-RD patients. Both molecules indicating Th2 (IL-4, CCR4, and cMaf) and T follicular helper cell (Tfh) response were found in the same lesions representing possible IL-21 producers in IgG4-related sialoadenitis.

There is also evidence for alterations of regulatory T cells in IgG4-RD [13]. Miyoshi et al. analyzed the frequency of CD4 + CD25 high T cells and CD4 + CD25 + CD45RA+ (naïve) T cells in the peripheral blood of autoimmune pancreatitis patients. They found increased frequencies of Tregs, while naïve regulatory T cells were decreased [19]. Tregs can also be found locally in IgG4-related pancreatitis lesions as evidenced by immunohistochemical methods [20]. Another study also identified increased numbers of Tregs in the peripheral blood of IgG-related pancreatitis patients. They also described increased frequencies of IL-10 producing ICOS+ Tregs in these patients. Possibly, these cells could drive a B-cell response towards IgG4 production and represent an attempt to limit an uncontrolled inflammatory response [21].

The efficacy of the anti-CD20 antibody rituximab in refractory IgG4-RD patients put B cells central in the pathogenesis of the disease [18]. Indeed, B cells are abundant in tissue lesions of IgG4-RD patients. Cytokines found to be produced in IgG4-RD patients such as IL-4, IL-10, and others could drive B-cell differentiation towards IgG4 production. In IgG4-related cholangitis, the B-cell receptor (BCR) repertoire was investigated in peripheral blood and tissue. In IgG4-RD but not in controls, IgG4+ B cell clones were identified that were selectively suppressed upon immunosuppressive therapy. These clones seemed to have undergone affinity maturation as evidenced by somatic hypermutation [22]. Well fitting, oligoclonal intrathecal IgG4 production has been described in IgG4-related pachymeningitis suggesting locally residing IgG4-producing plasma cells [23].

Though it is still unknown which antigens drive this process, a prominent process towards expansion of specific B cells and finally differentiation to IgG4-producing plasmablasts and plasma cells is evident. Plasmablasts are derived from B cells, produce antibodies, and still retain the ability to proliferate while having a short half-life. Plasmablasts can finally differentiate into plasma cells. Circulating plasmablasts, rare in healthy individuals, have been described in a variety of immune-mediated diseases including rheumatoid arthritis and systemic lupus erythematosus. In a recent study, a significant expansion of CD19 + CD27 + CD20-CD38^hi plasmablasts in the peripheral blood of active IgG4-RD patients was found [24]. These plasmablasts were largely surface IgG4-positive, and next-generation sequencing revealed oligoclonal expansion in all tested samples. Interestingly, numbers of circulating plasmablasts seemed to be independent of serum IgG4 levels [25].

After administration of rituximab in IgG4-RD patients, not only CD20+ B cells were depleted but also plasmablast frequency decreased in the peripheral blood [23]. There was a high variability on the timing of repopulation of clonally divergent plasmablasts in the blood of the treated patients partially correlating with disease relapse. Somatic hypermutation was observed both before and after rituximab therapy in IgG4-RD patients, implying importance of T helper cell responses in this disease. CD20 depletion presumably depletes precursors of IgG4-producing plasmablasts and short-lived plasma cells. However, elevated serum IgG4 levels are frequently observed after therapy in patients, which suggests residual plasma cells in the bone marrow and/or affected tissue of patients after therapy.

Macrophages are multifunctional cells of the immune system potentially involved in IgG4-RD pathogenesis. Currently, macrophages are grossly divided into classically activated M1 macrophages and M2 macrophages. M1 macrophages are generally regarded as proinflammatory, while M2 macrophages display a different phenotype: they are usually found as resident tissue macrophages and have been linked to wound healing and repair processes. M2 macrophages can produce significant amounts of IL-4, IL-10, and TGF-beta and thus could play a role in fibroinflammatory diseases such as IgG4-RD. In IgG4-related sialoadenitis, local macrophage infiltration preferentially of the M2 type could be demonstrated by immunohistochemistry [26]. These cells seem to produce for instance CCL18, a chemokine linked to fibrotic diseases such as pulmonary fibrosis and systemic sclerosis. Further, macrophage infiltration positively correlated with the severity of fibrosis in affected tissues.

In peripheral blood mononuclear cells (PBMCs) from healthy donors, activation of toll-like receptors (TLRs) and nucleotide-binding oligomerization domain-containing protein 2 (NOD-2) induced significant IgG4 production in vitro [27]. NOD2 activation was specifically able to trigger IgG4 production by B cells from healthy donors when these were cocultured with monocytes from IgG4-related pancreatitis patients. IgG4 production was dependent on monocyte production of B-cell activating factor (BAFF). These findings were independent of functional T-cell signaling in this in vitro system.

Mast cells have been described as important mediators of anaphylactic reactions and in allergic disorders. However, mast cells are currently considered to be involved in classical autoimmune diseases such as rheumatoid arthritis and bullous pemphigoid. Tissue mast cells have been found in IgG4-related sialoadenitis recently. Moreover, these cells colocalized with IL-4, IL-5, IL-10, and TGF-beta expression [28]. The same group reported recently that mast cells in IgG4-related sialadenitis may also produce IL-13 [29]. Interestingly, long-term treatment with an antihistamine drug was associated with regression of IgG4-related dacryoadenitis in one patient without immunosuppressive treatment [30]. In retroperitoneal fibrosis (RPF), tissue mast cell infiltration was well documented in a recent study [31]. Eotaxin/CCL11, a mediator involved in mast cell chemotaxis, is increased in serum of RPF patients and is locally produced in the inflamed tissue. Tissue infiltrating mast cells also seem to express the CCL11 receptor CCR3.

Basophils, the least abundant granulocyte type in peripheral blood, are involved in the pathogenesis of allergic and other disorders. Recently, basophils have also been implicated in pathogenesis of fibrosis. In an animal model of cardiac allograft fibrosis, basophils were important for myofibroblast activation leading to cardiac fibrosis [32]. A recent study demonstrated that activation of TLRs in basophils from healthy donors could induce IgG4 production by B cells through BAFF production in vitro. Basophils from IgG4-RD patients were even more effective in inducing an IgG4 response in B cells from healthy donors. This phenomenon was associated with increased IL-13 and BAFF production in vitro.

Immunoglobulin 4 (IgG4) is the least abundant IgG subclass in serum of healthy individuals accounting usually for less than 5 % of total IgG. Its physiological relevance is currently unclear because IgG4 is not able to significantly activate the complement system. IgG4 antibodies are unique because of the possibility to exchange Fab arms between different IgG4 molecules resulting in so-called bi-specific antibodies. However, there is no evidence of the formation of immune complexes using this bi-specificity, and therefore the relevance of this finding is unclear. On the other hand, IgG4 is associated with Th2 immune response and allergic diseases. Further, induction of blocking allergen-specific IgG4 antibodies correlates with successful desensitization in immunotherapy [33]. Although the name IgG4-RD suggests a central role for IgG4 antibodies in the disease, this is currently unproven. There are a variety of diseases such as eosinophilic granulomatosis with polyangiitis, Castleman disease, and others that commonly have elevated serum IgG4 levels [34, 35]. Conversely, active IgG4-RD may present with normal IgG4 levels. Also, serum IgG4 levels usually decline with clinical response upon immunosuppressive treatment, but this is not specific and relapses may occur with normal serum IgG4 levels [36]. Moreover, although IgG4 does not seem to activate complement, decreased complement C3 and C4 levels suggesting complement activation are not uncommon in IgG4-RD [37].

Recently, an experimental study of passive transfer of IgG1 and IgG4 antibodies from IgG4-RD patients to neonatal mice showed pathogenicity in terms of pancreatic damage. Transfer of patient IgG but not control IgG induced pancreatic neutrophil infiltration, hemorrhage, and acinar cell necrosis. However, in this study, IgG1 antibodies were more harmful than IgG4 antibodies. Also, IgG4 antibodies acted counteractive in combination and attenuated damage. However, the histopathological findings in this model were not reminiscent of type 1 autoimmune pancreatitis, and further research is necessary [38].