Histopathological findings in the landscape of IgG4- related pathology in patients with pituitary dysfunction: Review of six cases

IgG4- related hypophysitis (IgG4- RH) is increasingly being reported as an isolated entity or, less frequently, as a manifestation of a multiorgan IgG4- related disease (IgG4- RD), in which typical histopathology is a cornerstone for the diagnosis. We aimed to describe the histopathological changes in the surgical specimens from patients with clinical signs of pituitary disease that fulfilled the current diagnostic criteria for IgG4- RH. Histopathological features were correlated with clinical and radiological findings. Of 19 patients with pituitary dysfunction and inflammatory changes in the surgical pituitary specimen operated on during 2011- 2019, we identified five patients with typical IgG4- related pathology (lymphoplasmacytic infiltration with more than 10 IgG4- positive plasma cells per one high power microscopic field, representing at least 40% of all plasma cells and at least focal storiform fibrosis). One patient with diabetes insipidus and pachymeningitis with IgG4- related changes in a biopsy from the dura was also included. Additional histopatho logical changes that typically are not part of the IgG4- RH were observed: Rathke's cleft cyst in four and granulomatous changes in two patients. One patient had an elevated serum IgG4 level and systemic manifestations that could be associated with the systemic IgG4- RD. Our findings indicate that pure IgG4- RH is uncommon. All patients with pitui tary dysfunction, beyond typical IgG4- related pathology, had other pathological findings that could trigger the secondary IgG4- response. Both primary pathology and secondary IgG4- related features should be reported in patients with pituitary dysfunction because their co- occurrence may cause atypical clinical and imaging features, and unexpected response to surgical and pharmacological treatment. The current criteria for the diagnosis of IgG4- RH can lead to overdiagnosis of IgG4- RH if additional pathological changes are not taken into consideration. The classification criteria of IgG4- RD proposed by the American College of Rheumatology/European League Against Rheumatism could help classify patients more properly as IgG4- RH if applied to the pituitary gland. pituitary inflammation, Rathke's cleft


| INTRODUC TI ON
IgG4-related disease (IgG4-RD) is a multiorgan chronic fibroinflammatory disease with a broad spectrum of clinical manifestations related to IgG4-mediated immune dysregulation. [1][2][3][4][5] The most prevalent manifestations are submandibular sialadenitis, dacryoadenitis, autoimmune pancreatitis and retroperitoneal fibrosis. However, any organ can be affected, including the pituitary gland. 2,4,5 The main histological features, such as lymphoplasmacytic infiltration with a high proportion of IgG4-bearing plasma cells, usually in admixture with some eosinophils, storiform fibrosis and occasional obliterative phlebitis, are common for all organs, 3 although organ-specific variations may occur. 4,6 Inflammation and fibrotic reaction result in a tumour mass in the affected organ and frequent organ dysfunction. 3,5 Histopathological analysis of the biopsy specimen combined with immunohistochemical analysis by using antibodies toward IgG4 + and IgG + plasma cells is a cornerstone in the diagnosis of IgG4-RD. Because the IgG4 + plasma cells can occur in a significant number in different pathological conditions 7 and can vary from organ to organ, 6 requirements regarding the number of IgG4 + plasma cells and the ratio between the IgG4 + plasma cells and a total population of IgG + cells are included in the diagnostic criteria. 6 Histopathological diagnosis may be challenging in advanced phases of the disease when fibrosis is prominent and inflammatory cell infiltration is sparse. 3 Elevated serum IgG4 levels may be useful, although it is not a reliable diagnostic marker of the disease. 3,8 Careful multidisciplinary correlation of histopathological, serological, clinical and imaging findings improves IgG-RD diagnostic accuracy in an individual patient. 5,6 Pituitary involvement based on clinical and imaging grounds has previously been reported in patients with systemic IgG4-RD. [9][10][11] Soon after, in 2007, the first biopsy-proven case of IgG4-related hypophysitis (IgG4-RH) in a patient with the systemic IgG4-RD was published. 12 In 2011, Leporati et al 13 proposed the criteria for the diagnosis of IgG4-RH and emphasised the importance of histopathological proof of the disease, putting it as a single criterion sufficient for diagnosis. As the entity became better known among physicians, an increasing number of cases have been reported. [14][15][16] However, only about 30 of the published cases of IgG4-RH so far have been histologically confirmed in pituitary biopsy. [12][13][14][16][17][18][19][20][21][22][23][24][25][26] Thus, the whole spectrum of IgG4-related histopathological changes in the pituitary gland and sellar region is still unknown.
Leporati et al 13 suggested the presence of more than 10 IgG4 + plasma cells per high power field (HPF) as histological evidence of the IgG4-RH. According to the Consensus statement on the pathology of IgG4-RD, 6 the pathological diagnosis of IgG4-RD requires the presence of two of the three major histopathological features: (i) dense lymphoplasmacytic infiltrate with an increased number of IgG4 + plasma cells; (ii) fibrosis with at least focal storiform pattern; and (iii) obliterative phlebitis. Although the cut-off for the number of IgG4 + plasma cells required for diagnosis varies in different organs, the IgG4 + /IgG + ratio of more than 40% is recommended for all organs. 6 Histological criteria including the presence of more than 10 IgG4 + cells per HPF and IgG4 + /IgG + ratio of more than 40% are also included in the clinical guidelines for the diagnosis of IgG4-RH by the Japan Endocrine Society. 27 Recently, the American College of Rheumatology and the European League Against Rheumatism (ACR/ EULAR) published a more complex set of inclusion and exclusion criteria for the classification of IgG4-RD in the frequently affected organs, not including the pituitary gland. 28 In the present study, we aimed to describe the histopathological changes in the surgical specimens from patients with clinical signs of pituitary disease that fulfilled the criteria for IgG4-RH.
Histopathological features were correlated with clinical and radiological findings. We discuss the potential triggers of IgG4inflammatory response and the clinical and diagnostic impacts of IgG4-related changes in patients with pituitary involvement.

| Selection of clinical cases
In total, 19 patients operated on during the period 2011-2019 with inflammatory changes of any type in pituitary specimens, and one patient with pituitary dysfunction and inflammation in the dura, were identified through the search in the local pathology database Sympathy at the Department of Clinical Pathology, Uppsala University Hospital. All of the specimens were re-evaluated. In six cases with lymphoplasmacytic infiltration and fibrosis, suggestive of the IgG4-related process, additional immunohistochemical analyses were performed. The specimens were evaluated to confirm that the cases fulfilled both the criteria of Leporati et al 13 for IgG4-RH (> 10 IgG4 + plasma cells per HPF) and the international guidelines for the diagnosis of IgG4-RD (dense lymphoplasmacytic infiltration with an increased number of IgG4 immunoreactive cells, representing at least 40% of all plasma cells, fibrosis with at least focal storiform character). 6 The remaining 14 patients with pituitary specimens without the IgG4-related changes were excluded. In these patients, lymphocytic hypophysitis was diagnosed in three cases, granulomatous inflammation without additional pathology in two cases, lymphocytic inflammation secondary to a cystic process in six cases and granulomatous inflammation secondary to a cystic process in three cases.
The cystic process represented Rathke's cleft cyst in four cases, craniopharyngioma in one case and epithelial fragments that could not be classified more specifically in four cases.
The study was approved by the Regional Ethical Committee in Uppsala (Dnr 2016/126). All six patients with IgG4-related pathology presented here signed the informed consent to participate in the study.

| Clinical data and laboratory analysis
Detailed clinical data regarding the patients' age, gender, clinical symptoms, results of the relevant laboratory tests, and response to surgical and pharmacological therapy were collected from the electronic patient records.

| Magnetic resonance imaging (MRI) work-up
Magnetic resonance images before and after surgery for each patient were re-examined by an experienced neuroradiologist (JW). MRI examinations were performed at different hospitals and with slightly different techniques. In the five cases with sellar pathology (patients 1-5), all examinations included T1-weighted turbospin-echo sequences before and after gadolinium administration in the coronal and sagittal planes, as well as T2-weighted turbospin-echo sequence in the sagittal or coronal plane. The slice thickness varied between two and four millimeters. Four patients were examined at 1.5 T (patients 1-4) and one at 3 T (patient 5).
In the case of pachymeningitis (patient 6), the examination was performed at 3 T using a routine brain protocol, with the diagnosis mainly based on 3D T1-weighted TSE imaging after gadolinium administration. MRI examinations before surgery and post-surgical MRIs, including the last available examination, were evaluated.
The pituitary mass lesion size is reported in three dimensions (depth × width × height).

| Surgery
All patients with pituitary mass underwent transsphenoidal pituitary surgery (TSS) by an experienced neurosurgeon (OG). The main indication for surgery was a space-occupying pituitary mass that compromised visual functions. One patient who had headaches, diplopia and pituitary dysfunction without identifiable sellar mass lesion, and subsequently developed pachymeningitis, underwent only a limited surgical procedure to obtain a meningeal biopsy.

| Histopathological results
In all six specimens, five from the pituitary gland and one from the dura, the diagnostic criteria for IgG4-RH, as defined by Deshpande et al 6 and Leporati et al, 13 were fulfilled. All patients had prominent fibrosis, focally of storiform character, and rich inflammatory cell infiltrate composed of CD138 + and IgG + plasma cells, CD3 + Tlymphocytes, CD4 + T helper cells, CD68 + macrophages and CD20 + B-cells with the admixture of scattered eosinophils. IgG4 + plasma cells varied in number from 30 to more than 50 cells per HPF and represented at least 40% of all CD138 + plasma cells in the respective foci.
Obliterated blood vessels were observed in one specimen; however, the presence of obliterated blood vessels was generally difficult to as-

| Demographics and clinical histories of the patients
The age of the patients at the onset of the symptoms ranged from showed necrotising granulomas in the meningeal biopsy, was diagnosed with chronic pachymeningitis, which could be IgG4 and/or sarcoidosis-related.
The details of clinical, laboratory and treatment data are outlined below and in Table 1.

| Patient 2
A 42-year-old male was admitted to hospital because of headaches.
Additionally, he complained of frequent urination and impaired libido F I G U R E 1 Magnetic resonance imaging (MRI), histopathological and immunohistochemical characteristics of the patients. 1A, Coronal T2-weighted image shows an intra-and suprasellar lesion, which causes upward dislocation of the optic chiasm. The signal intensity pattern is compatible with a solid process with partially high fluid content. 2A, Coronal contrast medium enhanced T1-weighted MRI shows a bilobar peripherally enhancing lesion, with intra-and suprasellar components, with the latter causing upward dislocation of the optic chiasm. The central portions have signal intensity characteristics suggesting fibrosis. 3A, Coronal T2-weighted image shows a predominantly suprasellar cystic lesion, with oedema in the adjacent optic chiasm. 4A, Coronal T2-weighted image shows a cystic lesion occupying a large part of the pituitary gland, causing slight upward bulging but without affection of the optic chiasm. 5A, Coronal T2-weighted image shows a large intra-and suprasellar lesion with high water content. 6A, Sagittal contrast medium enhanced T1-weighted MR image shows thickening and enhancement of the dura anterior to the brain stem. 1B -6B, Rich lymphoplasmacytic infiltration admixed with eosinophils and fibrosis were present in all the cases (haematoxylin and eosin, ×200). 1C -6C, More than 30 IgG4 + cells/HPF (IgG4 IHC, ×200) were present in all the specimens. 1 -6 D, E, Additional lesions not typical for IgG4-RD have been observed in all the patients. Necrotising granulomas with occasional multinuclear giant cells were present in patients 1 and 6 (1D, 6D). Features consistent with Rathke's cyst, such as epithelial fragments (2D, 3D, 4D, 5D) and accumulation of eosinophilic amorphous material suspected for cyst content (2E, 4E), were observed in patients 2-5 (heamatoxylin and eosin, ×200 except for 5D, which has a magnification of ×100). Microphotograph 3C has been reprinted with a different magnification from the Encyclopedia of Endocrine Diseases 29 ( Figure 5D, p. 333) with permission from Elsevier (license number 4858250707676, license date 29 June 2020)

| Patient 3
A she has no anti-inflammatory treatment, but remains on hormone replacement therapy. At the 1-year MRI control, the pituitary gland had normal size with slightly decreased contrast enhancement than the normal gland, which raised suspicion of fibrosis. There were slight signal changes in the optic chiasm and tracts, corresponding to preoperative enhancement areas, possibly representing gliosis.

| Patient 4
After the surgery, the patient was treated with hormone substitution because of persistent panhypopituitarism. His lung fibrosis progressed, despite the treatment with an anti-fibrotic drug pirfenidone. He died 4 years after the pituitary surgery at the age of 73.

| Patient 5
A

| Patient 6
A 57-year-old female with a history of almost 20 years of recur-

| D ISCUSS I ON
In the present study, we reviewed clinical, histopathological and radiological findings in six patients with pituitary dysfunction and IgG4 features in surgical specimens. We found additional, non-IgG4related findings in all the patients.
In accordance with the proposal of Leporati et al 13  However, to induce remission in ANCA-associated vasculitis, more aggressive treatment strategies can be required. 37 Furthermore, besides IgG4-related histopathology, three patients in our cohort had evidence of Rathke's cleft cyst in the surgical specimens and, in one patient, fragments of Rathke's cleft cyst were suspected. Rathke's cleft cyst can cause an inflammation of the pituitary gland, especially when ruptured. 38 The inflammatory reaction surrounding Rathke's cyst can be associated with xanthomatous 39  Diabetes insipidus was present in as many as five of the six pa- tients. An anterior pituitary insufficiency also dominated in our patients, as also frequently reported in the literature. 14

| CON CLUS IONS
The preoperative diagnosis of IgG4-RH is difficult as a result of the lack of specific clinical and imaging features and the frequent ab- and a grant from ALF supported Olivera Casar-Borota.

CO N FLI C T O F I NTE R E S T S
The authors declare that they have no conflicts of interest.

PE E R R E V I E W
The peer review history for this article is available at https://publo ns.com/publo n/10.1111/jne.12942.