2025 年 72 巻 6 号 p. 649-662
Hypopituitarism, characterized by reduced secretion of pituitary hormones, profoundly impacts systemic metabolic homeostasis and quality of life. Its etiology ranges from congenital anomalies in pituitary development to acquired conditions involving inflammation and autoimmune processes. Despite advances in understanding its pathogenesis, diagnostic challenges persist, particularly in cases with complex extra-pituitary manifestations or novel genetic variations. Congenital hypopituitarism often stems from disruptions in transcription factors and signaling pathways critical for pituitary organogenesis. Emerging studies employing next-generation sequencing and developmental biology techniques have revealed new genetic loci and mechanisms implicated in combined pituitary hormone deficiency. However, the pathogenesis of most congenital cases remains elusive, underscoring the need for functional and phenotypic analyses of novel variants. Acquired hypopituitarism, frequently associated with pituitary tumors or systemic diseases, has also been increasingly linked to autoimmune mechanisms. Notably, the concept of paraneoplastic autoimmune hypophysitis has emerged, highlighting malignancy-driven immune responses as a novel etiological framework. Investigations into immune checkpoint inhibitor-related hypophysitis and anti-PIT-1 hypophysitis exemplify the intricate interplay between tumor immunity and endocrine dysfunction, suggesting shared mechanisms involving ectopic antigen expression and autoimmunity. This review synthesizes recent insights into the genetic, developmental, and immunological underpinnings of hypopituitarism. By exploring both congenital and acquired etiologies, we aim to bridge gaps in the current understanding of this complex disorder and provide a foundation for improved diagnostic and therapeutic strategies. Future perspectives emphasize the integration of advanced genetic tools, deeper exploration of tumor-immunity interactions, and a heightened focus on extra-pituitary phenotypes to refine clinical practice and enhance patient outcomes.
Hypopituitarism is a condition characterized by the reduced secretion of one or more of the hormones produced by the pituitary gland [1]. The pituitary gland regulates critical body functions by secreting hormones that control other glands and various physiological processes. Hypopituitarism leads to systemic metabolic dysregulation, including disturbances in lipid metabolism, bone metabolism, and electrolyte balance. As a result, hypopituitarism is associated with both a reduced quality of life and a poor prognosis [2]. The development of an adrenal crisis, particularly in the context of secondary adrenal insufficiency, can be life-threatening [3]. Accurate diagnosis of hypopituitarism is necessary to improve the prognosis and quality of life of affected patients.
The etiology of hypopituitarism can be classified as congenital or acquired and further categorized as affecting the pituitary gland alone or in combination with diseases of other organs. Accurate evaluation of concomitant diseases as a cause of hypopituitarism may be the key to diagnosing systemic or other organ abnormalities. Therefore, it is important to know about pituitary development, related molecules—including transcription factors—and possible background diseases. Recent advances have revealed novel mechanisms underlying the pathogenesis of hypopituitarism [1].
Hypopituitarism is frequently accompanied by other systemic and localized diseases. This has two implications: the possibility of other diseases as a cause of hypopituitarism and the possibility of a systemic disease associated with reduced hormone secretion. This is true for both congenital and acquired cases and must be carefully considered in clinical evaluation. In this review, we focus on recent progress in understanding the pathogenesis of hypopituitarism by congenital and acquired etiology, especially on extra-pituitary phenotypes and background pathogenesis.
The incidence of combined pituitary hormone deficiency (CPHD) is around 1 in 4,000 live births [4]. Pituitary organogenesis and hormone cell specification are regulated by several transcription factors and signaling pathways. The pituitary placode (adenohypophyseal) is a thickening of the oral ectoderm that invaginates to form Rathke’s pouch. The infundibulum (pituitary stalk) and posterior lobe (neurohypophysis) are formed from the neural ectoderm of the ventral diencephalon. These distinct primordia combine to form the pituitary gland. Several signaling molecules, such as fibroblast growth factors (FGF) 8, FGF10, and bone morphogenetic protein (BMP) 4, originating from the ventral diencephalon’s infundibular region, promote the induction and proliferation of Rathke’s pouch [5]. From the early patterning phase of pituitary development, multiple transcription factors coordinate and contribute to the differentiation and specification of pituitary hormone-producing cells.
Abnormalities in genes related to the ventral diencephalon cause holoprosencephaly (HPE) and septo-optic dysplasia. These abnormalities also cause hypopituitarism accompanied by pituitary stalk interruption syndrome (PSIS). PSIS is characterized by a thin or interrupted pituitary stalk connecting the hypothalamus to the pituitary gland, along with hypoplasia or aplasia of the anterior lobe [6]. PSIS is a mild phenotype of HPE and is thought to exist on a spectrum of midline defects [7]. Midline defects involve the central nervous system, primarily causing brain and eye abnormalities [8], such as microcephaly, septo-optic dysplasia, optic nerve hypoplasia, coloboma, and Chiari malformations. With the exception of CPHD caused by molecules specific to the pituitary gland, such as POU Class 1 Homeobox 1 (POU1F1, also known as PIT-1) and PROP paired-like homeobox 1 (PROP1), many genetic abnormalities are often associated with the aforementioned craniofacial abnormalities. However, it should be noted that in CPHD cases caused by midline defect-related abnormalities, only cases with very mild extra-hypopituitary phenotypes may have been reported, as the number of identified cases remains small.
The causative genes for CPHD were initially identified as pituitary hormones themselves, pituitary-specific transcription factors, or genes thought to be involved in pituitary formation and the ventral diencephalon. In recent years, advances in next-generation sequencing (NGS) and other genetic analysis methods, including molecular inversion probes, have led to the identification of additional causative genes [6, 9-60] (Table 1). However, analyses of numerous cases are often limited to variants of uncertain significance (VUS), and more detailed evaluations, including the accumulation of additional cases, are considered necessary [10, 19, 61]. There are also genetic variants that present with symptoms unrelated to pituitary hormone deficiencies or craniofacial abnormalities. Therefore, when CPHD is diagnosed, laboratory data and physical signs other than craniofacial abnormalities should be evaluated. Several CPHD-related genes have been reported to exhibit phenotypes other than those affecting pituitary hormones and craniofacial structures. Therefore, attention to these specific phenotypes in cases of CPHD may improve prognosis and provide helpful insights for genetic diagnosis.
Gene name | full name | reported case of CPHD with PSIS | Craniofacial and eye abnormalities* | Representative non-pituitary hormonal related and non-craniofacial involvement | Representative references |
---|---|---|---|---|---|
ARNT2 | aryl hydrocarbon receptor nuclear translocator 2 | + | + | kidney and urinary tract abnormalities | [11] |
B3GAT3 | beta-1,3-glucuronyltransferase 3 | – | + | developmental defect Larsen-like syndrome: short stature, skeletal deformities, and congenital heart defects recurrent ketotic hypoglycaemia |
[12] |
BLM | BLM RecQ like helicase | – | + | Bloom’s syndrome: predisposition to cancer, sun-sensitive skin rash, immune deficiency, and increased insulin resistance | [13] |
BMP4 | bone morphogenetic protein 4 | – | + | sclerosed nodules at the hands, short meta carpalial, and azoospermia | [14] |
BRAF | B-Raf proto-oncogene, serine/threonine kinase | + | + | structural cardiac abnormalities | [15] |
CDON | cell adhesion molecule-related/down-regulated | + | – | — | [16] |
CHD2 | chromodomain helicase DNA binding protein 2 | + | + | — | [17] |
CHD7 | chromodomain helicase DNA binding protein 7 | + | + | CHD7 disorder (CHARGE syndrome: coloboma, heart defect, choanal atresia, retarded growth and development, genital hypoplasia, ear anomaly) | [18] |
CSNK2A1 | casein kinase 2 alpha 1 | + | + | Okur-Chung neurodevelopmental syndrome: delayed language development, motor delay, intellectual disability, generalized hypotonia, difficulty feeding, and nonspecific dysmorphic facial features | [19] |
FGF8 | fibroblast growth factor 8 | – | + | — | [20] |
FGFR1 | fibroblast growth factor receptor 1 | – | + | — | [21] |
FOXA2 | forkhead box A2 | + | – | hyperinsulinemia, pancreatic dysgenesis, and biliary abnormalities | [22] |
GLI2 | GLI family zinc finger 2 | + | + | polydactyly, cryptorchidism, mega cisterna magna | [23] |
GLI3 | GLI family zinc finger 3 | – | + | Greig cephalo-polysyndactyly syndrome: abnormal development of the limbs, head, and face and includes polydactyly, macrocephaly, and hypertelorism Pallister Hall syndrome: hypothalamic hamartoma and bifid epiglottis | [24] |
GNAO1 | G protein subunit alpha O1 | – | – | GNAO1-related disorder phenotypes (infantile-onset developmental and epileptic encephalopathy, severe developmental delay and/or intellectual disability, and/or an early-onset movement disorder) | [25] |
GPR161 | G protein-coupled receptor 161 | + | – | — | [26] |
HESX1 | HESX homeobox 1 | – | + | — | [27] |
HHIP | hedgehog interacting protein | – | + | — | [28] |
HNRNPU | heterogeneous nuclear ribonucleoprotein U | – | + | epilepsy | [29] |
IGSF1 | immunoglobulin superfamily member 1 | – | – | macroorchidism | [30] |
IGSF10 | immunoglobulin superfamily member 10 | + | – | — | [31] |
LAMB2 | laminin subunit beta 2 | – | + | albuminuria, congenital nephrosis, and optical abnormalities | [32] |
LHX3 | LIM homeobox 3 | + | + | respiratory difficulties, skeletal abnormalities, hearing impairment, and vestibular function | [33, 34] |
LHX4 | LIM homeobox 4 | + | + | respiratory disease and genital malformations | [34, 35] |
L1CAM | L1 cell-adhesion molecule | – | + | CRASH syndrome: corpus callosum hypoplasia, retardation, adducted thumbs, spasticity, and hydrocephalus | [36] |
MAGEL2 | MAGE family member L2 | – | + | Schaaf–Yang syndrome: hypotonia, feeding difficulties during infancy, global developmental delay, and sleep apnea | [36] |
MIR17HG | miR-17-92a-1 cluster host gene | – | + | Feingold syndrome type 2: microcephaly, learning disabilities, and digital anomalies | [37] |
NFKB2 | nuclear factor kappa B subunit 2 | +; thin stalk | – | DAVID (deficient anterior pituitary with variable immune deficiency) syndrome: primary hypogammaglobulinemia, recurrent infection, and autoimmune manifestations | [38] |
NKX2.1 | NK2 homeobox 1 | – | + | Brain–lung–thyroid syndrome: primary hypothyroidism, respiratory distress, and neurological disturbances | [39] |
OTX2 | orthodenticle homeobox 2 | + | + | — | [40] |
PAX6 | paired box 6 | + | + | — | [41] |
PNPLA6 | patatin like phospholipase domain containing 6 | – | + | Broad spectrum of neurological disorders: Gordon-Holmes syndrome, Boucher-Neuhäuser syndrome, Laurence-Moon syndrome, and Oliver-McFarlane syndrome: progressive spastic paraplegia, distal muscle-wasting phenotype, and neurodegenerative condition | [42] |
POLR3A | RNA polymerase III subunit A | – | + | leukodystrophy syndrome named 4H syndrome: hypomyelination, hypogonadotropic hypogonadism, and hypodontia | [43] |
POU1F1 | POU class 1 homeobox 1 | – | – | — | [44] |
PROKR2 | prokineticin receptor 2 | + | + | — | [45] |
PROP1 | PROP paired-like homeobox 1 | + | – | — | [46] |
RBM28 | RNA binding motif protein 28 | – | + | alopecia, neurological defects, and endocrinopathy (ANE) syndrome | [47] |
RNPC3 | RNA binding region (RNP1, RRM) containing 3 | – | + | congenital cataracts | [48] |
ROBO1 | roundabout guidance receptor 1 | + | + | — | [49] |
SALL4 | spalt like transcription factor 4 | – | + | Duane-radial ray syndrome (Okihiro syndrome), acro-renal-ocular syndrome, and Holt-Oram syndrome: congenital upper limb defect, radial hypoplasia and kidney dystopia | [50] |
SEMA3A | semaphorin 3A | + | + | heart, pelvic genitourinary dysplasia, and skeletal abnormalities | [51] |
SHH | sonic hedgehog signaling molecule | + | + | — | [52] |
SIX3 | SIX homeobox 3 | + | – | — | [53] |
SMCHD1 | structural maintenance of chromosomes flexible hinge domain containing 1 | + | +: Bosma arhinia microphthalmia syndrome (microphthalmia and absence of a nose) | — | [54] |
SOX2 | SRY-box transcription factor 2 | – | + | intellectual disabilities, esophageal atresia, genital abnormalities, and sensorineural hearing loss | [55] |
SOX3 | SRY-box transcription factor 3 | – | + | — | [56] |
TBC1D32 | TBC1 domain family member 32 | + | + | polydactyly, developmental delay, rhizomelic shortening, and respiratory insufficiency | [57] |
TCF7L1 | transcription factor 7 like 1 | – | + | polydactyly, polyhydramnios, heart defect, hepatopathy, and nephropathy | [58] |
TGIF1 | TGFB induced factor homeobox 1 | + | + | — | [52] |
TTC26 | tetratricopeptide repeat domain containing protein 26 | + | + | BRENS (biliary, renal, neurological, skeletal) syndrome: hexadactyly, severe neonatal cholestasis, and involvement of the brain, heart, and kidney | [59] |
WDR11 | WD repeat domain 11 | + | + | urinary tract abnormalities and cardiac anomaly | [60] |
* when syndromes including non-pituitary hormonal related and non-craniofacial involvement were proposed, the name of syndrome and its phenotypes were described in the right column
Recent attempts in developmental biology have focused on identifying genes in embryonic lethal mice that are important for pituitary development and may be responsible for congenital hypopituitarism [62]. However, most new causative genes have been reported only in case studies; even in recent comprehensive genetic analyses, the cause and pathogenesis of most CPHD cases (84.2%) remain unclear [63]. It goes without saying that, even for known causative genes, it is essential to perform functional confirmation when identifying new variants [40]. Identifying novel causative genes and new genetic variation sites has contributed to further elucidating the pituitary formation process. For example, a recent study with a mouse model revealed the role of SIX homeobox 3 (SIX3) in the formation of the pituitary gland and hypothalamus during the early phase of the embryo [53]. An example of a novel genetic variation is found in variants in the β-isoform (repressive form)-specific region of the POU1F1 gene. Two independent groups reported that these variants in the β-isoform-specific region of the POU1F1 gene cause pituitary deficiency due to dominant β-isoform expression [64, 65]. Phenotypic analysis in mice with the same gene mutation further confirmed this phenomenon [66].
These genetic analyses have not only revealed the genes responsible for CPHD but also the process of pituitary formation. Clinical manifestations depend on which hormones are deficient but may include growth failure, reproductive dysfunction, hypothyroidism, adrenal insufficiency, and metabolic abnormalities. Diagnosis requires a combination of hormone testing, imaging studies, and, in some cases, genetic analysis. Treatment involves hormone replacement therapy tailored to the specific deficiencies present. Recent advances in genetic testing have expanded our understanding of the molecular basis of congenital hypopituitarism, while new insights into autoimmune mechanisms have shed light on some acquired forms. Ongoing research aims to understand better the genetic and molecular mechanisms underlying pituitary development and function in order to improve diagnostic and therapeutic approaches for this challenging condition. In addition, immune responses to some pituitary-related molecules, including transcription factors and hormones, previously mentioned are responsible for acquired hypopituitarism. This content is discussed in the next section.
Various conditions such as pituitary tumors, surgery, irradiation, infarction, autoimmunity, trauma, infection, hemochromatosis, granulomatous disease, and histiocytosis can cause acquired hypopituitarism [67]. Although the predominant cause of hypopituitarism is pituitary tumors, the underlying background may include other systemic diseases and an appropriate differential diagnosis of the pathogenesis of hypopituitarism is required. In addition, hypopituitarism associated with pharmaceutical agents, especially immune checkpoint inhibitors (ICIs), has been increasingly reported in recent years. Among the various causes of acquired hypopituitarism, the pathogenesis of the autoimmune etiologies of hypopituitarism has been increasingly elucidated. Hypophysitis is a rare condition characterized by inflammation of the pituitary gland [68, 69]. Based on etiology, hypophysitis is classified into two groups: primary and secondary (Table 2).
Primary | Secondary | ||
---|---|---|---|
Lymphocytic (autoimmune) | ●Focal lesions of the suprasellar region | ||
Granulomatous | Germinoma | ||
Xanthogranulomatous | Rathke’s cleft cyst | ||
Necrotizing | Craniopharyngioma | ||
Pituitary tumors/adenomas (PitNET) | |||
Chronic inflammation from peri-pituitary tissues, such as sinusitis | |||
●Systemic disease | |||
Sarcoidosis | |||
Vasculitis (e.g., Granulomatosis with polyangiitis, temporal arteritis) | |||
Autoimmune disease (e.g., SLE, Sjögren’s syndrome) | |||
Infiltrative diseases (e.g., Langerhans cell histiocytosis, Erdheim-Chester disease) | |||
Infection (e.g., tuberculosis, syphilis) | |||
IgG4-related | |||
Drug-induced (e.g., immune checkpoint inhibitor, daclizumab) | |||
Paraneoplastic syndrome |
Primary hypophysitis is considered a rare disease, with an incidence estimated at ~1 in 9 million cases per year [70]. This disease is reported relatively often in Japan. The reason for its high frequency in the country is unknown: geographic or ethnic variations in risk or purely greater awareness of hypophysitis in Japan [71]. Primary hypophysitis is classified into lymphocytic, granulomatous, xanthogranulomatous, and necrotizing types. Lymphocytic hypophysitis (also referred to as autoimmune hypophysitis) is morphologically classified into three subtypes: lymphocytic adenohypophysitis (LAH), lymphocytic infundibulo-neurohypophysitis (LINH), and lymphocytic panhypophysitis (LPH). HLA-DQ8 and DR53 are associated with the development of lymphocytic hypophysitis [72].
Several studies have identified autoantigens recognized by antibodies in autoimmune hypophysitis, such as growth hormone (GH)1, GH2, α-enolase, pituitary gland-specific factors 1a and 2 (PGSF1a, 2), and secretogranin-2 [73, 74]. An autoantibody targeting lactotrophs has also been reported [74, 75]. Additionally, an autoantibody targeting rabphilin-3A was reported as a marker of LINH [76]. The authors also noted that immunization of mice with rabphilin-3A led to neurohypophysitis [77]. A recent study reported circulating antibodies against the glycoprotein hormone alpha chain (the covalent alpha subunit shared by thyroid stimulating hormone (TSH), luteinizing hormone (LH), and follicle stimulating hormone (FSH), also known as chorionic gonadotropin alpha polypeptide or chorionic gonadotropin alpha chain [CGA]) in a case with acquired TSH, LH, and FSH deficiencies [78]. This finding suggests that the pathogenesis and epitopes of autoimmune hypophysitis are not uniform and may vary between individual cases.
Secondary hypophysitis is due to systemic causes, including autoimmune conditions (e.g., Sjögren’s syndrome and systemic lupus erythematosus), sarcoidosis, infiltrative diseases (e.g., Langerhans cell histiocytosis and Erdheim-Chester disease), vasculitis (e.g., temporal arteritis and granulomatosis with polyangiitis), and immunoglobulin G (IgG)4-related diseases [69, 70, 79-84]. This category also comprises drug-induced hypophysitis. For example, a recent study reported that daclizumab, a humanized monoclonal antibody that binds to the interleukin-2-receptor alpha chain and is used to treat relapsing-remitting multiple sclerosis, can cause hypophysitis [85]. In another example, coronavirus disease (COVID)-19 vaccination-induced hypophysitis could also be classified in this category. Circulating anti-pituitary antibodies against corticotrophs, thyrotrophs, gonadotrophs, and folliculo-stellate cells were detected in the patient’s serum [86]. Moreover, ICIs are widely known as a cause of hypophysitis [87]. A recent study showed that transient elevation in plasma adrenocorticotropic hormone (ACTH) occurred before the onset of ICI-related hypophysitis (ICI-RH). This phenomenon does not occur in all ICI-RH cases, but paying attention to the fluctuations may provide a clue for predicting its onset [88].
Within the broader category of hypopituitarism, a new category called paraneoplastic autoimmune hypophysitis has been proposed [89]. This condition is an inflammatory response of the pituitary gland triggered by autoimmune mechanisms associated with malignancies elsewhere in the body [90]. It is classified as “paraneoplastic” because it is linked to cancer, specifically due to the production of antibodies or other immune mechanisms activated by the tumor that cross-react with normal tissues [91]. These results suggest that the possibility of accompanying malignant tumors should be kept in mind when the physician manages acquired hypopituitarism cases with suspected autoimmune etiology. We also need to consider the possibility that some cases previously categorized as primary hypopituitarism may have a paraneoplastic background.
The concept of paraneoplastic autoimmune hypophysitis emerged from the pathophysiological analysis of anti-PIT-1 hypophysitis. The subsequent discovery of paraneoplastic isolated ACTH deficiency has further solidified this framework. This concept may also shed light on certain pathologies associated with ICI-related hypophysitis.
A. Anti-PIT-1 hypophysitisAnti-PIT-1 hypophysitis (originally called anti-PIT-1 antibody syndrome) was first described in 2011, based on analyses of cases exhibiting acquired deficiencies in GH, prolactin (PRL), and TSH [92]. Circulating anti-PIT-1 antibodies were identified in the sera of these cases, which displayed a hormone deficiency profile akin to that observed in PIT-1 gene mutations, thereby indicating the presence of autoimmunity directed against PIT-1-expressing cells. Immunohistochemical analyses of pituitary tissue from autopsies revealed the persistence of ACTH, LH, and FSH-positive cells, while PIT-1-positive (GH, PRL, and TSH) cells were absent. Moreover, histological examinations of other tissues (gastric mucosa, pancreas, adrenal glands, liver, and thyroid) demonstrated lymphocytic infiltration and structural tissue destruction.
Serum from patients with anti-PIT-1 hypophysitis exhibited no inhibitory effects on pituitary cells. Conversely, it was found that lymphocytes from these patients displayed specific reactivity to the PIT-1 protein [93]. These findings suggest that cell-mediated immunity plays a pivotal role in the onset of anti-PIT-1 hypophysitis. The genetic background regulating the threshold of autoantibody production in B-cells has also been implicated [94]. Subsequent studies established that processed PIT-1 epitopes are presented on the cell surfaces of anterior pituitary cells via major histocompatibility complex (MHC)/human leukocyte antigen (HLA) class I molecules [95]. In HLA class I, HLA-A*24:02:01 and/or A*02:06:01 were present in six out of seven cases that could be examined. Six and two out of seven cases (including ICI-induced anti-PIT-1 hypophysitis, discussed below) exhibited HLA-DR53 and HLA-DQ8, respectively, which are associated with lymphocytic hypophysitis [96, 97]. In addition, alleles associated with pituitary immune-related adverse events [98, 99], including HLA-DR15 and HLA-DQ7, DQ15, and DQ16, were observed in four of seven cases.
However, analyses utilizing PIT-1-positive cells derived from induced pluripotent stem (iPS) cells from both healthy individuals and patients revealed no significant differences in the quantity of PIT-1 epitope presentation, indicating the potential involvement of additional factors [95].
A pivotal moment in the understanding of this disease was marked by the diagnosis of thymoma in the initial three cases of anti-PIT-1 hypophysitis [100]. The ectopic expression of PIT-1 was observed within the thymoma, suggesting its contribution to the onset of this syndrome. Notably, in one case that underwent thymectomy, the anti-PIT-1 antibody titer diminished, accompanied by a reduction in lymphocyte reactivity to the PIT-1 protein. Other malignancies were also noted in later cases identified without thymoma [96, 97, 101]. These malignancies similarly exhibited ectopic expression of the PIT-1 protein, indicating that such ectopic expression, whether in thymomas or other malignancies, may disrupt immune tolerance to PIT-1, thereby leading to the development of autoimmunity against PIT-1 [96, 97]. Based on these observations, diagnostic criteria for this condition are currently being proposed [102], underscoring the necessity of considering potential tumor associations. Recently, cases of this disease have also been documented following the administration of immune checkpoint inhibitors, highlighting the need for vigilance in patients presenting with central hypothyroidism [96, 97].
Anti-PIT-1 hypophysitis should be included in the differential diagnosis of acquired GH deficiency. Clinical manifestations of adult GH deficiency encompass metabolic disturbances associated with visceral obesity, decreased muscle and bone mass, and diminished quality of life [103, 104]. GH replacement therapy aims to ameliorate these symptoms [105]. GH replacement therapy does not increase the risk of malignant tumor development, progression, or recurrence [106]. However, it is generally acknowledged that GH/insulin-like growth factor 1 (IGF-I) may promote tumor cell proliferation, rendering GH replacement therapy contraindicated in patients with active malignancies. When evaluating acquired GH deficiency, it is imperative to consider this syndrome and assess the possibility of tumor associations during replacement therapy.
B. Paraneoplastic isolated ACTH deficiencyAutoimmunity directed against ACTH-producing cells is thought to be the cause of acquired isolated ACTH deficiency (IAD), as IAD cases are frequently associated with various autoimmune disorders [107]. Rab GDP dissociation inhibitor alpha (Rab GDI alpha) has been reported as a potential autoantigen in IAD [108]. However, the precise mechanisms underlying the onset of IAD remain unclear.
Recent investigations of IAD cases linked to lung cancer revealed ectopic ACTH expression within tumor tissue. The patient’s IgG recognized pro-opiomelanocortin (POMC), although it did not hinder the proliferation or survival of AtT20 cells (a mouse POMC-expressing cell line). However, it was demonstrated that the patient’s lymphocytes exhibited specific reactivity to the POMC protein, suggesting that the immune response directed at ectopic antigens within the tumor may also compromise ACTH-producing cells in the pituitary, aligning with the concept of paraneoplastic syndromes [109]. However, whether these patients exhibited reactivity to other antigens, including the molecule mentioned above, Rab GDI alpha, is a subject for future study.
Prior to this report, several instances of acquired IAD associated with malignancies, such as gastric cancer [110, 111] and acute lymphoblastic leukemia [112], had been documented. Various studies have indicated POMC expression in tumors; for instance, 48% of non-small cell lung cancers show silent POMC expression without the manifestation of Cushing’s syndrome, and POMC expression is also recognized in neuroendocrine tumors that do not elicit ectopic ACTH syndrome [113]. This suggests that the prevalence of tumors exhibiting ectopic POMC, potentially triggering IAD via an immune response, may be underestimated. Particularly in patients with tumor-associated syndromes, it is common for a diagnosis of malignancy to occur post-symptom onset. Thus, it is prudent to remain cautious regarding the potential existence of malignancies in patients with idiopathic IAD. However, not all instances of IAD fall within the category of paraneoplastic hypophysitis, necessitating further research to determine the frequency of tumor associations in IAD and to identify potential malignancies. There has been a case report of IAD associated with tumors lacking ectopic expression of POMC; it remains to be seen whether the tumor in these cases is an accidental complication, a technical problem in the analysis of POMC expression in the tumor tissues, or an antigenic reaction to molecules other than POMC/ACTH [114].
C. Expansion into the mechanisms of immune checkpoint inhibitor-related hypophysitisRecent advancements in cancer immunotherapy have led to numerous reports of ICI-related hypophysitis. In ICI use cases, those who develop immune-related adverse events, including hypopituitarism, are known to be associated with good overall survival [115]. The expression of cytotoxic T-lymphocyte-associated protein 4 (CTLA4) within the pituitary and type 2 hypersensitivity reactions have been implicated in CTLA4 inhibitor-related hypophysitis, while programmed cell death 1 (PD-1) antibodies have been associated with the potential for type 4 hypersensitivity reactions [116]. However, the mechanisms underlying ICI-related hypophysitis remain largely unclear.
As previously noted, many malignancies frequently express POMC. Investigations assessing whether ICI-related hypophysitis shares analogous pathology with paraneoplastic autoimmune ACTH deficiency revealed that anti-corticotropin antibodies were detected in two out of twenty patients (10%) with PD-1/programmed cell death 1- ligand 1 (PD-L1) inhibitor-related hypophysitis [117]. Ectopic POMC expression was identified in the tumor tissues of these cases, suggesting that the immune response to ectopic POMC may contribute to the breakdown of immune tolerance. These findings imply a shared mechanism between “paraneoplastic isolated ACTH deficiency” and “ICI-related hypophysitis.” Specifically, in patients with ectopic POMC expression in silent tumors, the application of ICIs may incite autoimmune responses, potentially culminating in IAD. Recent case reports of ICI-related hypophysitis in the context of large-cell neuroendocrine carcinoma have also postulated that ACTH25–39 may act as an autoantibody epitope [118]. Notably, instances of anti-PIT-1 hypophysitis have been documented following ICI treatment [96, 97]. This suggests that the concept of “paraneoplastic autoimmune hypophysitis” may align with the mechanisms underlying at least some cases of ICI-related hypophysitis.
The causes of hypopituitarism are many and varied, with the majority of congenital cases remaining unexplained. Improvements in analytical techniques such as NGS and research into previously unchartered areas, such as the role of long non-coding RNAs in pituitary development and environmental factors, may dramatically improve diagnostic accuracy [6, 9, 119, 120]. Accurate assessment of extra-pituitary phenotypes in these molecules would also be helpful in improving the accuracy of genetic diagnosis. Moreover, the elucidation of novel molecules involved in pituitary development may lead to the identification of autoantigens implicated in acquired hypopituitarism.
Regarding acquired etiologies, drawing upon the findings to date, the notion of “paraneoplastic autoimmune hypophysitis” has been proposed as a potential mechanism for the onset of acquired pituitary insufficiency (Fig. 1). Although we have classified this concept as one of the diagnoses of secondary hypophysitis (Table 2), we believe it is important to consider the possibility of latent concomitant malignant tumors in cases of primary hypophysitis during clinical practice.
Malignant tumors ectopically express pituitary antigens (PIT-1 or ACTH/POMC). Dendritic cells phagocytize tumor cells, leading to the generation of pituitary-reactive cytotoxic T lymphocytes (CTLs), B cells, and plasma cells. The CTLs attack both pituitary cells and tumor cells.
Current analyses and reported cases have facilitated a provisional classification, as depicted in Fig. 2. However, further investigation is needed to ascertain the proportion of ICI-related hypophysitis attributable to paraneoplastic autoimmune hypophysitis, the appropriateness of equating anti-PIT-1 hypophysitis in thymoma versus non-thymoma cases, and the potential influence of genetic backgrounds such as HLA. The collection and analysis of additional cases are essential to advancing this field.
Some cases of autoimmune-related hypopituitarism (e.g., autoimmune hypophysitis and isolated ACTH deficiency) develop as paraneoplastic autoimmune hypophysitis. A subcategory of anti-PIT-1 hypophysitis includes thymoma-related forms. Whether thymoma-related and non-thymoma-related hypophysitis represent the same disease remains to be determined. ICI-related hypophysitis is also a form of autoimmune hypopituitarism and, in some cases, may occur as paraneoplastic autoimmune hypophysitis. Some ICI-related hypophysitis also occurs as anti-PIT-1 hypophysitis. However, the proportion of ICI-related hypophysitis that occurs as paraneoplastic autoimmune hypophysitis remains unknown.
The concept of “paraneoplastic autoimmune hypophysitis,” derived from the examination of a limited number of cases, represents a novel disease paradigm wherein endocrine dysfunction is clinically evident, yet is underpinned by malignancy with autoimmunity as the fundamental mechanism [90]. Comprehending the interplay between tumors, immunity, and endocrinology is crucial for advancing our understanding of these pathologies. Furthermore, when diagnosing acquired pituitary dysfunction, clinicians must remain vigilant regarding the possibility of malignancies, particularly when the involvement of autoimmunity cannot be excluded as a mechanism for pituitary impairment. Consequently, diagnosing pituitary dysfunction may serve as an early indicator for malignancy detection. While it is widely believed that patients with paraneoplastic syndrome have more favorable prognoses than those without, prospective studies are required to evaluate the incidence of malignancies in acquired pituitary insufficiency and their prognostic implications [115]. Additionally, numerous tumor-bearing conditions exhibit ectopic hormone expression, necessitating further exploration of their relevance to this emerging concept to enhance our understanding of these disorders.
We hope that further elucidation of the pathogenesis of hypopituitarism and the proposal of novel disease concepts will continue (Graphical Abstract).
We thank Prof. Genzo Iguchi (Konan Women’s University/Kobe University), Dr. Ryoko Hidaka-Takeno (Takeno Clinic), Prof. Yutaka Takahashi (Nara Medical University), Prof. Yasuhiko Okimura (Kobe Women’s University), Prof. Wataru Ogawa (Kobe University), Prof. Emer. Kazuo Chihara (Akashi Medical Center), and Prof. Sally Camper (University of Michigan) for their generous mentorship and supervision. We also thank our laboratory members for their fruitful discussions.
The authors have no conflicts of interest directly relevant to the content of this article.
Grant-in-Aid for Scientific Research from the Japan Society for the Promotion of Science (KAKENHI, grant numbers: 17K16165, 21K16370, and 21KK0149 [HB], 21K20933 [KK], and 23K08008 [MY]).