Plasma cell hepatitis (PCH) is a pathohistological finding characterized by lymphoplasmacytic portal and lobular inflammation with prominent plasma cells and often with the presence of interface hepatitis, perivenulitis and centrilobular necrosis (1–4). While the term “PCH” was originally used to describe autoimmune hepatitis (AIH), its use has extended to other plasma cell (PC)-rich necroinflammatory disorders including recurrent AIH (rAIH) and PC-rich rejection (PCR) in liver allograft (2, 5–7). Accordingly, the Banff Working Group defines that AIH represents PCH of the native liver, while rAIH and PCR are clinical entities of PCH occurring in the post-LT setting (8–12).

PCH often results in the development of hepatic fibrosis if not promptly and adequately treated with potent immunosuppressants (IS) such as corticosteroids (CS), antimetabolites, and calcineurin inhibitors (13–16). Moreover, IS often fail to sufficiently control hepatic necroinflammation, which eventually leads to liver failure (7, 14, 17, 18). Furthermore, the long-term administration of IS is associated with significant morbidity, including the development of opportunistic infections and neoplasms (19).

Since histopathological and serological assessment do not distinguish between AIH, rAIH, and PCR, their diagnosis is entirely reliant on clinical context, which is coupled with challenges, perhaps ambiguity (8, 12, 20–23). This is especially relevant in differentiating between rAIH and PCR. By convention, PCR refers to PCH in individuals who underwent transplantation for diseases other than AIH (8). Conversely, rAIH refers to PCH occurring in patients transplanted for AIH. Thus, based upon current convention, differentiating between rAIH and PCR may not be plausible in circumstances where pre-LT diagnosis was uncertain (e.g., acute liver failure of unknown etiology or cryptogenic cirrhosis). Furthermore, current PCH classification scheme excludes individuals transplanted for AIH from receiving a diagnosis of PCR. Thus, it remains elusive if rAIH and PCR represent distinctive or overlapping clinical entities. Similarly, there has not been evidence demonstrating that the onset of rAIH is mediated through the recurrence of immunopathology underlying AIH in the native liver, making the nomenclature of rAIH potentially deceiving. Ultimately, the fundamental issue regarding the current classification of PCH is the substantial degree of uncertainty as to whether each disease—as currently classified—represents a unique immunologic phenomenon.

PC, the terminally differentiated B cells, play a major role in the regulation of humoral immunity through the production of immunoglobulin (Ig). PC exhibit highly diverse immunomodulatory effects depending on the classes and subclasses of Ig production such as IgG, A, and M as well as IgG1, 2, 3, and 4, respectively (24). Hence, the Ig classes and subclasses expressed in the infiltrating PC would associate with, at least in part, the immunopathological presentations of PC-mediated disorders (24, 25). In particular, inflammatory disorders with a pronounced infiltration of IgG4-PC have been known to manifest marked tissue fibrosis and favorable response to IS (26). Therefore, chronic inflammatory disorders with IgG4-PC infiltration have emerged as a unique clinical entity, namely IgG4-related diseases (IgG4-RD) (27). The pancreas was the first organ in which IgG4-RD was recognized, namely autoimmune pancreatitis; thereafter this disease entity has been known to affect multiple organs, including the liver parenchyma (27, 28).

Prior studies have demonstrated the infiltration of IgG4-PC in the liver tissue of PCR and native-liver AIH, with a PCR subtype demonstrating significant enrichment (28, 29). Consequently, the Banff Working Group recommends use of IgG4 immunostaining in the evaluation of post-LT PCH (12). However, this recommendation lacks a specific threshold for IgG4-PC positivity and does not provide guidance with respect to its clinical relevance. One potential reason for this is that there has not been a comprehensive study that cross-sectionally compares the degree of IgG4-PC infiltration between PCH types with a standardized quantification method, significantly limiting its practical use. In particular, the degree of IgG4-PC infiltration in rAIH has not previously been studied. Thus, it remains elusive whether assessment of IgG4 immunostaining may be of diagnostic and therapeutic relevance in the post-LT setting—particularly in differentiating between rAIH and PCR. The primary aim of this study is to characterize PCH diseases by objectively determining the IgG4-PC positivity and evaluate for associations with clinical presentations and outcomes.

This represents the first study to comprehensively evaluate the degree of IgG4-PC infiltration across types of PCH in the non- and post-transplant setting. Notably, our results showed a high degree of IgG4-PC infiltration more frequently associated with the clinical diagnosis of PCR than tnAIH, fAIH, and rAIH. In particular, markedly high IgG4-Positivity (≥0.500) was only found in PCR cases. This implies that IgG4-PC infiltration to this extent is not a general immune response to a liver allograft, but rather appears to be a unique immunopathological fingerprint of PCR (Figure 2). We also note that there are cases of PCR with minimal or absent of IgG4-PC infiltration. Accordingly, the overlapping profiles of rAIH and some cases of PCR (IgG4-Positivity <0.500) raises the possibility a shared immunopathology in these two separate clinical entities (Figure 2). These findings may support prior hypotheses that a pathophysiologically-distinct de novo AIH exists separately from PCR (33). Moreover, there are some cases with relatively high degree of IgG4-PC infiltration (IgG4-positivity >0.20) in AIH cases, especially in tnAIH group. These observations collectively indicate that each PCH disease consists of heterogeneous immunophenotypes. Additionally, we found that differentiation of PCH by IgG4-positivity may be of direct clinical relevance in the post-LT setting. The group with markedly high IgG4-Positivity (≥0.500) appears to demonstrate a more rapid response to directed IS. Consequently, our study suggests that evaluation of IgG4-Positivity may serve as a valuable diagnostic approach in the post-LT setting with corresponding implications in management. Additionally, the lack of rAIH cases with markedly high IgG4-Positivity in our study suggests that IgG4-Positivity may be of diagnostic utility when pre-LT diagnoses is unclear or unknown. Specifically, an IgG4-Positivity threshold of ≥0.500 may effectively rule out a diagnosis of rAIH. This is of potential therapeutic relevance as rAIH has been known to require aggressive immunosuppression to prevent graft loss (14).

To date, there remain substantial gaps in understanding the underlying pathophysiology of PCH, which significantly hinders clinical management of PCH diseases. In particular, there is concern as to whether each PCH disease represents a distinctive condition or a cluster of broader, overlapping conditions than current classification schemes. The latter case might, at least in part, be the basis for the variable degree of response to IS among and within the PCH diseases. Current management strategies of PCH entirely rely on the empiric use of a variety of IS, which may result in significant toxicity and morbidity such as, but not limited to, the development of infection and malignancy. While the current empiric management strategy is capable of inducing remission in a large proportion of patients, the prognosis of individuals with PCH remains highly variable. In the non-LT setting, transplant-free survival of AIH is 91% and 70% at 10 and 20 years from the initial diagnosis, respectively, with up to 40% progressing to cirrhosis despite treatment (18, 34). Similarly, rAIH is responsible for 1% of all deaths at 5 years post-transplantation due to graft failure and is associated with increased risk of death from infection (14). With regards to PCR, the overall prognosis in adults is not well defined, although studies in pediatric populations demonstrate no significant impact on the prognosis (35, 36). Though PCR typically responds well to increased doses of corticosteroids, it is understood that non- or under-treated patients eventually progress to graft loss (13). As use of IS also carries its own risks (19, 37), an establishment of a specific, molecular-targeted therapeutic strategy that abrogates off-target toxicities provides an opportunity to improve patient outcomes.

Given advancement in understanding of IgG4-RD, there has been increasing interest in the involvement of IgG4-PC in liver diseases. The pioneer work examined the liver tissue of autoimmune pancreatitis patients with liver enzyme abnormalities, in which the dense infiltration of IgG4-PC in the liver parenchyma was observed in nearly all cases (28). Since the histopathological findings also exhibited multiple features commonly seen in AIH, further studies investigating the presence of IgG4-PC in the liver tissue of patients with AIH were performed (38–40). These studies found an infiltration of IgG4-PC in a small proportion of cases based on the criteria of >5 or 10 IgG4-PC per high power field (HPF). In contrast, more pronounced degree of IgG4-PC infiltration, >25 IgG4-PC per HPF was noted in cases of PCR (29). While these findings led to substantial excitement into the field, the definition of the “IgG4-PC infiltration” determined in a binary fashion (e.g. “positive” or “negative”) or the number of cells staining positive for IgG4 per HPF might not serve as an objective indicator since it lacks the consideration of the total number of PC aggregated in the corresponding foci. To date, the lack of a standardized evaluation method for IgG4-PC infiltration has severely limited its clinical utility—despite expert acknowledgement of its importance in prior consensus statements (12). To overcome this fundamental issue, we established a rigorous approach to better evaluate the degree of IgG4-PC infiltration. To this end, we determined the frequency of IgG4-PC by normalizing the number of IgG4-PC over the total number of PCs in the corresponding portal tracts using serially sectioned slides (IgG4-Positivity). In addition, we sought to minimize the risk of data interpretation bias resulting from the potential sampling error by enrolling only samples that have at least 8 portal tracts, of which the foci with the representative tracts were used to evaluate the degree of IgG4-PC infiltration. We believe our quantification system allows objective cross-sectional evaluation of IgG4-PC infiltration.

IgG4 is known to have a unique immunomodulatory effect distinct from other types Ig (24). Other than antigen binding, Ig plays an important role in activation of the immune system through two discrete mechanisms: 1)Ig binding to fragment crystallizable (Fc) receptors (FcR) expressed on the cell surface of various immune cells, which augment the cytotoxic and phagocytic function of immune cells, and 2)Ig interaction with complement component 1q (C1q), resulting in activation of classical complement pathway. Of great interest, IgG4 has been considered a noninflammatory IgG compared with other IgG subclasses due to stronger affinity to the inhibitory FcR, Fcγ receptor IIB, and an inferior capacity in complement pathway activation (41, 42). These notions regarding the “non-inflammatory” characteristics of IgG4 appears irrelevant in PCH, at least in our study, as our observations revealed that IgG4-Positivity is associated with higher degree of hepatic necroinflammation. A potential explanation is that IgG4-PCs may serve an anti-inflammatory role when severe inflammation is present. Interestingly, despite the positive association of IgG4-Posivity with inflammation severity, there was no difference with respect to liver biochemistries at biopsy (Table 4).

It has been shown that the production of IgG4 involves an activation of unique upstream immune pathway, in which interleukin (IL)-4 and IL-10 secreted from CD4⁺ (C-X-C chemokine receptor type-5) CXCR5⁺ T cell, namely Tfh cells, facilitates transdifferentiation of naive B cells to IgG4-PC (43–45). Accordingly, affected organs in IgG4-RD exhibit an abundant infiltration of Tfh cells (45, 46). In addition, the clonal expansion of CD4⁺ T cells with a cytotoxic function (CD4⁺ CTLs), which abundantly express SLAM family member 7 (SLAMF7), granzyme A (GZMA), IL-1β, and TGF-β, is observed in the affected organ of IgG4-RD (47, 48). TGF-β is a potent anti-inflammatory cytokine and contributes to the fibrosis development. Thus, development of fibrosis a hallmark feature of IgG4-RD. However, we observed that infiltration of IgG4-PC was not associated with fibrosis severity. This potential discrepancy between IgG4-RD and PCH with IgG4 PC infiltration may be partially due to the close laboratory monitoring of AIH and post-LT patients, resulting in uptitration of IS before the fibrosis development. Alternatively, the immunosuppressive properties of IgG4 and TGF-β might have a stronger impact on disease presentation than the pro-fibrotic effect of TGF-β in PCH.

In general, IgG4-RD is known to be highly responsive to glucocorticoid-based IS. This notion coincides with the immunosuppressive properties of IgG4 as well as the anti-inflammatory effect of TGF-β secreted by the CD4⁺ CTLs. Accordingly, previous studies by others reported that AIH cases with IgG4-PC infiltration were associated with a comparatively rapid response to steroid therapy to the induction of clinical remission (38, 40). Moreover, the relatively high frequency of IgG4-PC seen in PCR cases in our observation as well as by others (29) may serve as an explanation as to why a less aggressive regimen of IS is generally required for PCR compared to AIH and rAIH (8, 49, 50). This notion is also supported by our finding that markedly high IgG4-Positivity in the post-LT setting (≥0.500) was correlated with rapid improvement of serum AST, for which all cases were in the PCR group (Table 5). Moreover, the strong association of decreasing IgG4-Positivity and CS use is also congruent with existing paradigm of steroid-responsiveness in IgG4-RD (Table 4). These observations suggest that evaluating IgG4-Positivity has a direct relevance to clinical management of PCH, particularly in the post-LT setting, as it may be of prognostic value in assessing expected response to corticosteroid therapy.

In summary, this study is the first to cross-sectionally demonstrate the diversity of the immunophenotypic profile of PCH in the non- and post-transplant setting, in the absence and presence of IS, as defined by IgG4-Positivity—with significant differences across and within disease entities. The highly heterogenous IgG4-Positivity across and within PCH entities indicate that current classification of PCH diseases is insufficient in capturing the immuno-pathophysiology. Hence, we propose refining the PCH classification strategy by incorporating the IgG4-Positivity based stratified categorization, particularly in the post-LT setting; which might inform a immunopathology-specific management strategy and prognostication. Toward this goal, further prospective studies with a larger number of subjects treated with a standardized IS regimen, and long-term follow up, are warranted.