mAbs are established targeted therapies for several diseases, including hematological and solid malignancies. These agents have shown a favorable toxicity profile, but, despite their high selectivity, new typical side-effects have emerged. In cancer patients, pituitary dysfunction may be mainly due to brain metastases or primary tumors and to related surgery and radiotherapy. Anticancer agents may induce hypopituitarism in patients cured for childhood cancers. These agents infrequently affect pituitary function in adult cancer patients. Notably, hypophysitis, a previously very rare disease, has emerged as a distinctive side-effect of ipilimumab and tremelimumab, two mAbs inhibiting the cytotoxic T-lymphocyte antigen-4 receptor, being occasionally seen with nivolumab, another immune checkpoint inhibitor. Enhanced antitumor immunity is the suggested mechanism of action of these drugs and autoimmunity the presumptive mechanism of their toxicity. Recently, ipilimumab has been licensed for the treatment of patients affected by metastatic melanoma. With the expanding use of these drugs, hypophysitis will be progressively encountered by oncologists and endocrinologists in clinical practice. The optimal management of this potentially life-threatening adverse event needs a rapid and timely diagnostic and therapeutic intervention. Hypopituitarism caused by these agents is rarely reversible, requiring prolonged or lifelong substitutive hormonal treatment. Further studies are needed to clarify several clinical and pathogenic aspects of this new form of secondary pituitary dysfunction.
Invited author: Salvatore M. Corsello is Associate Professor of Endocrinology at the Catholic University in Rome. He is the author of around 90 peer-reviewed articles and more than 300 other publications, with a research focus on pituitary and thyroid diseases. Profiles of the co-authors are presented in the Acknowledgements section
In the last years, the increasing understanding of molecular mechanisms of diseases has allowed remarkable improvements in the diagnosis and treatment of several conditions, including cancer. Many of the disease-specific or pathogenic molecules are revealed as detectable and even ‘druggable targets’ by highly selective proteins, such as mAbs (1, 2).
The availability of these synthetic proteins has lead to the creation of numerous new diagnostic applications and clear improvements in those already existing (3). Immunofluorescence, immunohistochemistry, or epitope-specific immunoblotting are only a few examples of the current techniques based on mAbs.
Moreover, a number of mAbs are currently used for the pathogenic treatment of rheumatologic diseases and other conditions sustained by unwanted immune/inflammatory response (e.g. arthritis, inflammatory bowel diseases, acute rejection of kidney transplants, etc.) (4, 5, 6).
MAbs, either as single drugs or combined with cytotoxics, have improved clinical outcomes in patients affected by various hematological and solid malignancies (7, 8). These agents exert their anticancer activity by different mechanisms of action, including the alteration of signal transduction in the downstream intracellular pathways and the neutralization of soluble ligands (7). Moreover, mAbs may activate the antibody-dependent or the complement-mediated cytotoxicity and may target the immunosuppressive crosstalk between immune system and cancer cells, inducing a tumor-specific immune response (7). MAbs can even deliver cytotoxic drug or radionuclide in the proximity of the target cells (9).
Toxicity profiles described with mAbs considerably differ from those seen with cytotoxic agents and are usually mild to moderate, although life-threatening adverse events may occur (7). Unusual toxicities have been associated with a specific class of mAbs, becoming characteristic of that class of drug. Skin toxicities may be induced by anti-epidermal growth factor receptor mAbs, approved for the treatment of advanced colorectal cancer and head and neck cancer (8, 10). Hypertension, thromboembolic events, hemorrhage, proteinuria, wound leakage, fistulas, and bowel perforation are associated with bevacizumab, an anti-vascular endothelial growth factor mAb, approved for the treatment of patients affected by advanced colorectal, breast, lung, kidney, and ovarian cancers and glioblastoma (8, 11, 12). Symptomatic and asymptomatic cardiotoxicity may be induced by trastuzumab, an anti-HER2 mAb, approved for patients with HER2-positive breast and advanced gastric cancer (8, 13). Pneumonia, tuberculosis, demyelination, drug-induced lupus, hepatotoxicity, or lymphomas are seen amongst patients treated with the anti-tumor necrosis factor α (TNF-α) mAbs (14). Recently, hypophysitis has been recognized as an adverse event triggered by some mAbs acting as immune checkpoint inhibitors (15, 16).
Herein, the epidemiological and clinical data on pituitary dysfunction induced by mAbs are reviewed, focusing on the suggested pathogenic mechanisms sustaining hypophysitis induced by anti-cytotoxic T-lymphocyte antigen-4 (CTLA4) mAbs and by other immunoregulatory anticancer mAbs.
Drug-induced pituitary dysfunction
Various anticancer treatments may induce selective deficit of pituitary hormones (17). Usually, these endocrine side-effects manifest and progress subclinically (17, 18). In cancer patients, irradiation of brain primary tumors or metastases may often cause various degrees of pituitary damage (18). Pituitary dysfunction may also occur even when the radiation field does not apparently involve the pituitary or hypothalamic area (18). Anticancer agents may induce hypopituitarism in patients cured for childhood cancers (17). These agents infrequently affect pituitary function in adult cancer patients. Single cases of hypopituitarism have been attributed to interferons (with or without ribavidin) in patients affected by hepatitis C (19, 20, 21, 22, 23), but not in cancer patients. Conversely, pituitary dysfunction is a relatively common side-effect induced by new mAbs that inhibit specific immune checkpoints (24). Pituitary disorders induced by these agents have been reported as ‘hypopituitarism’ or ‘hypophysitis’ with different degrees of toxicity (Table 1). The mAbs subclass of immune checkpoint inhibitors includes mAbs targeting different receptors, including the CTLA4 and the programed death 1 (PD1) or the PD1 ligand 1 (L1) (PD1-L1) (25).
The incidence of pituitary dysfunction in studies on mAbs inhibiting immune checkpoints.
|References||Type of trial, treatment schedule||Clinical setting||Number of patients||Incidence of pituitary dysfunction||Other endocrine adverse events|
|(36)||Phase I trial||Pretreated stage IV melanoma||56||G3–4 hypophysitis: one pt (1.8%)||Not reported|
|3 mg/kg every 3 weeks, or 3 mg/kg→dose reduced to 1 mg/kg every 3 weeks+vaccination with modified HLA-A*0201-restricted peptides from gp100 MAA|
|(37)||Phase I–II trial||Pretreated stage IV melanoma||139||G3–4 hypophysitis: 13 pts (9%)||G1–2 hypothyroidism: three pts (2%)|
|3 mg/kg+peptide vaccinations or intra-patient dose escalation→peptides (HLA-A*0201 status)|
|(38)||Phase I trial||mRCC||61||G3–4 hypopituitarism: two pts (3.3%)||G3–4 primary adrenal insufficiency: one pt (1.6%)|
|3 mg/kg→1 mg/kg or all doses at 3 mg/kg q 3 weeks|
|(15)||Retrospective study||Pretreated stage IV melanoma and RCC||113||G3–4 hypophysitis: 8/163 pts (4.9%), 6/113 melanoma pts (5%), 2/50 RCC pts (4%)||Not reported|
|3 mg/kg every 3 weeks||50|
|(39)||Phase I–II trial||Pretreated stage IV melanoma||46||G3–4 hypophysitis: eight pts (17%); (5 mg/kg: one pt), (9 mg/kg: seven pts)||G3–4 hypothyroidism: one pt (2.2%)|
|3→9 mg/kg (intra-patient dose escalation)|
|(77)||Phase II trial||Metastatic pancreas ADC||27||G2–3 hypopituitarism: one pt (3.7%)||Not reported|
|3 mg/kg every 3 weeks×four for a maximum two courses|
|(78)||Phase I trial||Relapsed/refractory B-cell NHL||18||G1–2 hypophysitis: one pt (6%)||Not reported|
|3 mg/kg→monthly 1 mg/kg×3 months (dose level 1), → escalation to 3 mg/kg monthly×4 months (dose level 2)|
|(29)||Phase III trial||Pretreated unresectable stage III/IV melanoma||676||Hypophysitis: G3, two pts (1.5%) in ipilimumab arm; G3, two pts (0.5%) in combination arm; 0 in gp100 arm;||G1–2 hypothyroidism: two pts (1.5%) in ipilimumab arm; G3, one pt (0.3%); G1/2, five pts (1.6%) in the combined arm; G1/2, two pts (1.5%) in gp100 arm|
|3 mg/kg every 3 weeks for four doses with or without gp100 vs gp100 alone||Hypopituitarism: G3–4, two pts (1.6%); G1/2, one pt in ipilimumab arm; G3, two pts (0.8%); G1–2, one pt in the combination arm||Adrenal insufficiency: G1–2, two pts (1.5%) in ipilimumab arm; G3, two pts (0.5%); G1–2 one pt (0.3) in the combination arm|
|(79)||Phase II trial||Pretreated unresectable stage III/IV melanoma||217||Endocrine-IRAEs were globally reported:|
|0.3, 3, 10 mg/kg q 3 weeks for four cycles (induction)→q 3 months (maintenance)||Grade 3–4: 0.3 mg/kg: 0 pts; 3 mg/kg: two pts (1%); 10 mg/kg: one pt (0.5%)|
|In the 3 mg/kg group, hypopituitarism lead to treatment withdrawal|
|(43)||Phase I trial||Unresectable stage III/IV melanoma (naïve/not)||21||Hypophysitis (grade not specified): three pts (14%)||Thyroiditis (grade not specified): four pts (19%)|
|10 mg/kg q 3 weeks×4→q 3 months+bevacizumab 7.5 mg/kg (Cohort 1) or 15 mg/kg (cohort 2) q 3 weeks|
|(80)||Phase II trial||Pretreated, unresectable stage III/IV melanoma||155||Endocrine-IRAEs were globally reported: G3: two pts (1.3%); G4: 0 pts|
|10 mg/kg q 3 weeks for four cycles (induction)→q 3 months (maintenance)|
|(50)||Phase II trial compassionate use||Refractory melanoma||53||G2–3 hypophysitis with adrenal insufficiency: two pts (4%)||G2 hypothyroidism: one pt (2%)|
|10 mg/kg q 3 weeks for four doses→every 12 weeks in case of CB|
|(41)||Phase I trial 1–10 mg/kg escalating dose+PSA–Tricom vaccine||Chemotherapy refractory/naïve mHRPC||30||Panhypophysitis: four pts (13.3%) (5 mg/kg: G2, two pts; 10 mg/kg: G3, one pt, G2, one pt)||Hypothyroidism: four pts (13.3%) (5 mg/kg: G2, two pts; 10 mg/kg: G2, two pts); Adrenal insufficiency: three pts (10%) (5 mg/kg: G3, one pt; 10 mg/kg: G3, one pt; G2, one pt)|
|(46)||Phase IIb trial||Stage III/IV NSCLC||204||G3 hypophysitis/hypopituitarism: one pt (0.5%) in the concurrent ipilimumab arm||Not reported|
|Paclitaxel 175 mg/m2+carboplatin (AUC 6)±placebo/ipilimumab|
|(81)||Phase II trial||Pretreated, unresectable stage III/IV melanoma||EAP; 830||Endocrinopathies: 4%|
|10 mg/kg i.v. q 3 weeks up to four doses (induction)→10 mg/kg q 12 weeks (maintenance)|
|(42)||Phase I trial||CT-naïve mCRPC||28||Hypophysitis: G2, one pt (3.6%); G3, two pts (7.1%) at 3.0 mg/kg dose level; G2, two pts (7.1%); G3, two pts (7.1%) at the 5.0 mg/kg dose level||Not reported|
|0.3, 1, 3, 5 mg/kg escalating dose+GVAX|
|(40)||Phase II trial||Resected stage IIIc/IV melanoma||75||Hypophysitis: 11 (15%) pts||Hypothyroidism (G1): one (1.3%) pt|
|3 or 10 mg/kg every 6–8 weeks for 12 months→maintenance ipilimumab at 10 mg/kg every 3 months. In HLAA*0201+ pts, ipilimumab infusions were accompanied by three separate subcutaneous vaccine (tyrosinase, gp100, MART-1)|
|(51)||Phase I trial 0.01–15 mg/kg every 90 days||Metastatic malignancies||39||G2 hypopituitarism: one pt (2.5%)(15 mg/kg)||G1 hypothyroidism: one pt (2.5%) G1 hyperthyroidism: one pt (2.5%) (15 mg/kg)|
|(82)||Phase II trial||Refractory or relapsed melanoma||251||G3–4 hypophysitis: one pt (0.4%)||Thyroid disorders: eight pts (3.2%)|
|15 mg/kg i.v. every 90 days|
|(31)||Phase III trial 15 mg/kg i.v. every 90 days (arm A) vs CT (arm B)||Naïve unresectable stage IIIc/IV melanoma||325 (arm A)||‘Hypothalamic or pituitary gland disorders’: six pts (2%) No case in arm B||‘Primary thyroid disorders’: 17 (5%) vs one (2%) pt in arm B Primary adrenal insufficiency: four pts (1%) vs no case in arm B|
|(67)||Phase I trial Nivolumab 0.1–10 mg/kg every 2 weeks (escalating dose)||Various advanced solid malignancies||296||Hypophysitis: one pt (<1%)||Thyroiditis: <1%. Adrenal insufficiency: 0 TSH increase: nine pts (3%) (0.1 mg/kg: two pts G1–2; 1 mg/kg: two pts G1–2; 3 mg/kg: two pts G1–2; 10 mg/kg: three pts G1–2+1 pt G3–4) Hypothyroidism: seven pts (2%) (0.3 mg/kg: one pt G1–2; 1 mg/kg: two pts G1–2; 3 mg/kg: one pt G1–2; 10 mg/kg: two pts G1–2+one pt G3–4) Hyperthyroidism: three pts (<1%) (0.1 mg/kg: one pt G1–2; 3 mg/kg: one pt G1–2; 10 mg/kg: one pt G1–2+one pt G3–4)|
AEs, adverse events; CT, cytotoxic chemotherapy; DLT, dose limiting toxicity; EAP, expanded access program; ED-SCLC, extensive disease-small-cell lung cancer; G, toxicity grade according to CTC-NCI criteria; G1, mild; G2, moderate; G3, severe; G4, life threatening; GVAX, granulocyte–macrophage colony-stimulating factor gene-transfected tumor cell vaccine; HRPC, hormone resistant prostate cancer; IRAEs, immune-related adverse events; NSCLC, nonsmall cell lung cancer; Pretr, pretreated; PSA, prostate specific antigen; Pt(s), patient(s).
mAbs inducing pituitary dysfunction
CTLA4 is a key immune checkpoint molecule that downregulates T-cell activation and proliferation (26). CTLA4 contributes to control autoimmunity and in the presence of cancer it limits the expansion of tumor-specific effector T-cells, favoring cancer immune tolerance (26). CTLA4 blockade with an anti-CTLA4 mAb (anti-CTLA4-mAb) results in the removal of the negative immune-modulatory effect of CTLA4. This action elicits therapeutic benefit (27), but it may also remove CTLA4-mediated protection from autoimmunity, and is responsible for a new spectrum of autoimmune–inflammatory side-effects, classified as immune-related adverse events (IRAEs) (28). Two mAbs inhibiting CTLA4 activity have been evaluated for clinical use: ipilimumab and tremelimumab. In two phase III trials, ipilimumab, a fully human anti-CTLA4 IgG1-mAb, improved for the first time the overall survival in patients with metastatic melanoma (29, 30). Based on this exciting finding, in March 2011, ipilimumab was approved by the FDA as first- and second-line monotherapy for unresectable or metastatic melanoma, followed by the EMA approval of the drug for previously treated advanced melanoma (8). Tremelimumab, another anti-CTLA4-mAb, did not show similar efficacy (31). The toxicity profile of anti-CTLA4-mAbs appears characteristic compared with other anticancer agents. IRAEs frequently involved the gastrointestinal tract, liver, skin, and anterior pituitary. Rarer IRAEs may involve thyroid, adrenals, central and peripheral nervous system (i.e. aseptic meningitis, Guillan-Barré syndrome, myastenia gravis, optic neuropathy), eyes (episcleritis or uveitis), lung, pancreas, heart, kidney, and bone marrow (red blood cells aplasia) (27). These toxicities are occasionally (if any) encountered with other immunomodulating agents (i.e. interferons) or with cytotoxic drugs. The CTLA4-mAbs delivered dose seems to influence the frequency and severity of IRAEs (27). IRAEs induced by anti-CTLA4-mAbs were mild to moderate in most patients, and the recovery was obtained by medical treatments as indicated in protocol guidelines (32). However, more severe grade of IRAEs, particularly severe or life-threatening diarrhea/colitis/diverticulitis, hepatitis, and hypophysitis and even drug-related deaths (2.2%) were reported (33, 34). In the majority of cases, the onset of IRAEs seems to depend on the organ damaged by the drug (33, 35). Skin IRAEs are typically reported after an average of 3–4 weeks, those affecting the gastrointestinal tract and liver occur after 6 or 7 weeks, while endocrine adverse events tend to occur after 9 weeks (32, 35). The difference in IRAEs time of onset remains unexplained (33). Skin and endocrine disorders are usually more persistent. Gastrointestinal symptoms resolve more rapidly, although were most frequently associated with severe complications (bowel perforation/obstruction or peritonitis) if not readily recognized (33). Similarly hypophysitis, if not promptly diagnosed and properly treated, may be life threatening due to secondary adrenal insufficiency.
Anti-CTLA4-mAbs and pituitary dysfunction
Anti-CTLA4-mAbs have been reported to induce pituitary dysfunction (hypophysitis, anti-CTLA4-H, and/or hypopituitarism) with a variable incidence (0–17%) of 4.5% in the largest trial (29). The reasons explaining the wide range of incidence are still unknown. The onset of the pituitary damage secondary to anti-CTLA4-mAbs is independent from the tumor types, having been reported in patients with various solid and hematologic malignancies (kidney, prostate, pancreatic, lung cancer, melanoma, lymphoma). However, as ipilimumab and tremelimumab have been mostly studied in patients with kidney and prostate cancer and melanoma, the majority of data comes from these settings. Notably, not in all trials, a surveillance for endocrine toxicity (i.e. adrenocorticotropin, cortisol, thyroid-stimulating hormone, and free thyroxine serum levels) was required by protocol, and therefore some cases may have been missed. The dosage of the drugs seems to influence the rate of anti-CTLA4-H. In initial small studies on patients affected by metastatic melanoma with lower doses of the drug (1–3 mg/kg), a lower incidence of hypophysitis (1.8%) was reported (36). However, in a larger phase II study including 139 advanced/metastatic cutaneous melanoma patients receiving 1–3 mg/kg with or without a peptide vaccine, severe to life-threatening hypophysitis was recorded in 9% of patients (37). Two out of 61 patients (3.3%) with metastatic renal cell cancer who received ipilimumab (1–3 mg/kg) experienced hypophysitis (38). In a trial evaluating ipilimumab (3–9 mg/kg intra-patient dose escalation) in patients affected by cutaneous melanoma who had previously received other treatments, eight out of 46 (17%) patients had hypophysitis (one patient at the dose of 5 mg/kg and seven patients at the dose of 9 mg/kg) (39). In the first published trial evaluating toxicity and tolerability of ipilimumab in patients with radically resected stage IIIc/IV melanoma (adjuvant setting) with doses of 3 and 10 mg/kg (at the extended dosing interval of 6–8 weeks), 11 out of 75 (15%) patients had G1–3 hypophysitis. In this trial, patients who resulted HLAA*0201-positive received three separate subcutaneous vaccines (tyrosinase, gp100, MART1) in addition to ipilimumab. Approximately 60% were pretreated with other drugs active against melanoma. The authors did not observe a significant difference in toxicity when comparing the two ipilimumab dosing regimens. It is unclear what the influence of the administration of vaccines is in triggering hypophysitis. In this report, six (8%) patients required replacement glucocorticoids due to hypophysitis for longer than 2 years (40). In the phase III trial leading to the FDA approval of ipilimumab (676 pretreated patients with unresectable stage III/IV melanoma receiving the drug as a single agent or combined with the peptide vaccine gp100 or gp100 alone), ‘hypophysitis’ was reported in two (1.5%) patients in ipilimumab only and in two (0.5%) patients in the combination arm. Furthermore, drug-related ‘hypopituitarism’ was reported in six (1.2%) patients (two in the ipilimumab arm and two in the combination arm respectively) (29).
In a phase I trial on patients with metastatic castration-resistant prostate cancer (mCRPC), ipilimumab (1–10 mg/kg) was administered in combination with an anti-prostate-specific antigen (PSA) vaccine (41). Hypophysitis occurred in four out of 30 (13.3%) patients on the higher doses (5 mg/kg, two patients; 10 mg/kg, two patient) (41). In another mCRPC study of ipilimumab (0.3–5 mg/kg) combined with a prostate cancer cell vaccine, moderate to severe hypophysitis was diagnosed in all three patients at the dose of 3 mg/kg, in two of 16 (13%) patients of the 3 mg/kg expansion cohort, and in two of three patients at the higher dose (5 mg/kg) (42). Whether the higher incidence of hypophysitis seen in these trials might be attributed to the higher drug dose, to the concomitant administration of vaccine or both, remains to be elucidated. Preliminary results from the only trial evaluating ipilimumab (10 mg/kg) administered in combination with bevacizumab reported high rates of hypophysitis (14%) (43).
Hypophysitis was never diagnosed when ipilimumab, even at a dose of 10 mg/kg, was combined with chemotherapy in ant studies (30, 44, 45) except one (46). Similarly, hypophysitis was not reported in patients affected by advanced cutaneous melanoma with brain radiotherapy-pretreated metastases (47). The presumptive explanation of these data might reside in the immune cell depletion induced by cytotoxic chemotherapy and radiotherapy. Hypophysitis was not reported among patients with metastatic melanoma who were involved in two of three expanded access programs of ipilimumab as a single agent (3 mg/kg) (48, 49). In the other trial, the incidence of hypophysitis was 4% (50).
Tremelimumab (15 mg/kg) has also been reported to induce hypophysitis, apparently at a lower frequency (0.4–5% of patients). In the first phase I study of tremelimumab, one out of 39 (5%) patients with solid malignancies developed panhypopituitarism due to hypophysitis (51). In the largest phase III trial, 655 untreated patients affected by unresectable stage IIIc or IV melanoma randomly received tremelimumab (15 mg/kg once every 90 days) or chemotherapy (temozolomide or dacarbazine). In the tremelimumab arm, endocrine toxicities occurred in 7.4% of patients (31). ‘Hypothalamic or pituitary gland disorders’ were reported in 2% of patients, while primary thyroid disorders and primary adrenal insufficiency was diagnosed in 5 and 1% of patients respectively (31). However, data on pituitary dysfunction induced by tremelimumab are quantitatively more limited when compared with ipilimumab. This is the consequence of the negative results on tumor control seen in the tremelimumab pivotal phase III trial that have greatly lessened the clinical development of the drug.
Clinical features of anti-CTLA4-H
Anti-CTLA4-H seems to occur mainly in males. This is the opposite of classic lymphocytic hypophysitis (LyH) (52, 53). Presenting symptoms are not specific, including headache, fatigue, weakness, confusion, hallucinations, memory loss, labile moods, insomnia, anorexia, temperature intolerance, and chills (15, 28, 38). Sometimes, erectile dysfunction and loss of libido were also present (54). Visual impairment, including diplopia may occur, but less frequently compared with LyH (52, 54). In patients receiving ipilimumab at standard doses, symptoms occurred at a median time of 11 weeks after initiation of therapy (before the fourth dose), suggesting a possible cumulative effect (54). However, the onset of hypophysitis has been reported as early as 4 weeks from the first infusion (55). When pituitary hormone levels were evaluated at the onset of symptoms, variable degrees of hypopituitarism were detected (56). In patients with anti-CTLA4-H, the pituitary–thyroid and pituitary–adrenal axes are almost invariably damaged (54, 56). Also, hypogonadotropic hypogonadism is present in most male patients (83–87%) (54, 56). However, as patients who received anti-CTLA4-mAbs were affected by an advanced stage disease, hormonal abnormalities may also result from sickness-induced hypogonadism and/or sick euthyroid syndrome. In a short case series, three out of five patients showed low serum IGF1 levels (56). Prolactin levels may be elevated or low in ∼25% of patients (15, 39). Sporadically, the posterior pituitary was affected with diabetes insipidus (55) or SIADH (57).
Anti-CTLA4-H presents radiological features similar to classic LyH. At magnetic resonance imaging (MRI), a pituitary swelling appears in most cases (up to 60–100% of baseline size), sometimes with thickening of the stalk (15). The height of the gland in the sagittal view increases from 3.4–6 to 7.7–11.8 mm (15, 54, 57). Contrast enhancement of the pituitary may or may not be homogeneous (58). However, MRI changes appear to be of lesser magnitude compared with those seen in classic LyH (54) and in some cases, the gland may show normal radiologic features (54, 58).
As in patients with classic LyH (51, 52), high-dose glucocorticoids are used to treat anti-CTLA4-H. The treatment is then slowly tapered according to the improvement of the clinical and laboratory picture (32). In almost all patients, acute symptoms disappear a few days after the drug withdrawal and the start of glucocorticoids and hormone replacement therapy (15). The efficacy of glucocorticoids generally parallels with rapid pituitary shrinkage on MRI, but enlargement of the gland may decrease more gradually in a variable period of time (4–12 weeks), despite a more rapid reduction of symptoms (54, 56). Importantly, pituitary hormone deficits may be prolonged or even lifelong, despite the prompt initiation of glucocorticoid therapy. Indeed, hypopituitarism is the only potentially irreversible IRAE induced by anti-CTLA4-mAbs (32, 33). Early discontinuation of glucocorticoid therapy may predispose to relapsing hypophysitis. Recovery of pituitary–thyroid function has been reported in 37–50% of patients (32, 33, 54, 56), whereas pituitary–gonadal axis function recovered in 57% of men (15). Very few patients were able to discontinue glucocorticoid replacement due to persistent secondary adrenal insufficiency (28, 54). Unfortunately, at the onset of anti-CTLA4-H or during its course, it is still difficult to predict which patients will develop persistent hypopituitarism.
Similarly, the protective role of glucocorticoids in reducing the incidence/severity of anti-CTLA4-H remains to be investigated. Remarkably, the antitumor effects of CTLA4 blockade seemed not to be influenced by the treatment of anti-CTLA4-H with high-dose glucocorticoids and replacement therapy (27). Patients who experienced anti-CTLA4-H may be safely retreated, based on individual case evaluations (27).
Other mAbs inhibiting immune checkpoints
PD1 is a key immune-checkpoint receptor expressed by activated T cells (59). It is part of the CD28/CTLA4 family and contributes to the maintenance of T-cell functional silence against their cognate antigens (59). PD1 is similar to CTLA4 in structure, but exerts a distinct biologic function, primarily in peripheral tissues (59, 60). PD1 has two known ligands, PD1-L1 (B7-H1) and PD1-L2 (B7-DC), PD1-L1 being the primary PD1 ligand (59, 60, 61). In contrast to CTLA4 ligands, PD1-L1 is selectively upregulated and expressed on many solid tumors (60, 61, 62) and on cells within the tumor microenvironment in response to inflammation (59, 62, 63). In this context, PD1-L1 inhibits cytokine production and the cytolytic activity of PD1+, tumor-infiltrating CD4+, and CD8+T cells (61, 64, 65). Both PD1 and PD1-L1 play a pivotal role in the ability of tumor cells to evade the host's immune system (59, 60, 61, 62, 63, 64, 65). Human mAbs targeting PD1 or PD1-L1 induce various pharmacological interference, including increase in immune function and antitumor immunity in vitro (62, 64, 66). Similarly, in preclinical models, blockade of the PD1 pathway using mAbs directed against both PD1 and PD1-L1 stimulates cytotoxic activity and inhibits tumor growth (59, 60).
Nivolumab (formerly BMS-936558, MDX-1106) is a fully human IgG4 PD1 mAb. It was evaluated in a phase I trial on 296 patients affected by various advanced types of cancer at the escalating doses of 0.1–10.0 mg/kg every 2 weeks. Cumulative response rates ranged between 18 and 28%. Severe drug-related adverse events were reported in 14% of patients, with three deaths due to pulmonary toxicity. Hypophysitis was reported in one (<1%) case (67).
No cases of pituitary dysfunction were reported in initial trials evaluating lambrolizumab, another humanized IgG4 mAb that acts against PD1 (68) or other mAbs targeting PD1-L1 (i.e. BMS-936559 and MPDL3280A) (69, 70).
In cancer patients, pituitary dysfunction may occur due to a variety of causes. Hypopituitarism may be a consequence of brain radiotherapy and surgery as well as of metastases that infiltrate or compress the gland or hypothalamus or affect their connections. Anticancer drugs may also induce selective deficit of different pituitary hormones (17). However, all these toxicities are on the whole infrequent (17). Until the advent of anti-CTLA4-mAbs, hypophysitis was never induced by anticancer drugs. It was an exceptional adverse event reported in patients who were treated with interferon for viral hepatitis (49, 50, 51, 52, 53). Furthermore, hypophysitis is per se a very rare disease. Up to now, <1000 cases have been reported, although the incidence of this condition may be underestimated (52, 53). Unexpectedly, during the clinical development of anti-CTLA4-mAbs, hypophysitis has emerged as a distinctive side-effect and probably as a new form of secondary hypophysitis. The wide range of anti-CTLA4-H incidence (0–17%) may in part depend on the different drug dosage, being more frequent when higher doses are used. However, it should be emphasized that pituitary hormones monitoring was not required by all trials' protocols. Therefore, transient or mild damage to the pituitary might have been underdiagnosed. It is unclear whether previous anticancer immune therapies may facilitate the onset of pituitary toxicity seen in these patients. Notably, a higher rate of hypophysitis has been reported when ipilimumab was associated with bevacizumab or certain vaccine types, but not with cytotoxic chemotherapy or radiotherapy to melanoma brain metastases (which actually seems to have a protective effect). Recently, hypophysitis is very rare (<1%) or absent in initial trials evaluating other immune checkpoint inhibitors (anti-PD1 and anti-PD-L1 respectively) (67, 68, 69, 70).
No case of hypopituitarism/hypophysitis was attributable to mAbs targeting immune mediators (i.e. TNF-α or interleukins), used in patients affected by rheumatologic diseases or other conditions sustained by autoimmunity. This might be explained by the inherent mechanism of action of these mAbs and the therapeutic objective which is aimed at reducing rather than enhancing the immune activation triggering the disease. Infliximab, for example, by targeting TNF-α in inflammatory bowel disease, blocks one of the pathogenic effectors of the disease. On the contrary, anti-CTLA4-mAbs, by unleashing one of the molecules involved both in cancer immune escape and autoregulatory activity of the immune response, induce not only tumor response but also IRAEs. The different incidence of hypophysitis between anti-CTLA4- and anti-PD1/anti-PD1-L1 mAbs may be related to the different role of the molecule targeted by these agents in the context of the negative modulation of immune response. Indeed, CTLA4 is induced in T cells at the time of their initial response to antigen to modulate cell-mediated immune response once triggered. The PD1/PD1-L1 pathway is not involved in initial T-cell activation, but it starts later to regulate inflammatory responses in tissues and tumor microenvironment sustained by effector T-cells, limiting tissue damage related to immune activation (25). Therefore, the mAbs-mediated inhibition of PD1/PD1-L1 pathway presumably acts on the modulatory phase of the process more than in the triggering phase of the autoregulatory mechanism.
The key question remains why anti-CTLA4-mAbs trigger the onset of hypophysitis. Unfortunately, almost all aspects of this condition are unclear. Firstly, the exact incidence is unknown. More importantly, the pathogenic mechanisms of anti-CTLA4-mAbs have been only hypothesized (54, 56). Since the presenting clinical features, including MRI and hormone profile, resemble primary LyH and the causative drugs unleash autoimmunity, the suggested pathogenic mechanism is likely to be autoimmunity. However, as in LyH, neither antigen(s) nor the role of pituitary immune microenvironment favoring the onset of this condition has been unveiled. Furthermore, to date no case CTLA4-H has been proven by pathology. The onset of anti-CTLA4-H has been suggested to result from the activity of anti-pituitary antibodies (15). Although the presence of pituitary antibodies in patients on anti-CTLA4-mAbs has been detected, pituitary-triggering antigen(s) and pathogenic antibodies in anti-CTLA4-H remain to be evaluated. Furthermore, potential factors influencing the onset of anti-CTLA4-H, including age, gender, genetic predisposition (specific polymorphisms of CTLA4 and other genes involved in known autoimmune diseases) have not been elucidated. Overall, the available data show that this condition is more prevalent in males, contrary to most other autoimmune diseases, including autoimmune hypophysitis. However, such male preponderance needs to be confirmed in larger trials. When the potential genetic correlations between CTLA4 haplotypes and responses in patients receiving ipilimumab were evaluated, no correlation emerged (71, 72). Until now, studies evaluating genetic markers as potential predictors of specific toxicities are still lacking.
It has also been reported that the onset of IRAEs is associated with better clinical outcomes of underlying malignancy (73). However, whether the onset of hypophysitis in patients under anti-CTLA4-mAbs (or a specific grade of this toxicity) may predict better survival and/or response remains to be estimated in further trials with appropriate analytical approach (73).
As in primary LyH, patients on anti-CTLA4-mAbs with symptoms suggesting hypophysitis should promptly undergo pituitary MRI and pituitary function assessment. The possibility of pituitary metastasis should be considered in the presence of a sellar mass in this clinical scenario. If metastasis is high in the differential diagnosis, a biopsy should be considered (74). If anti-CTLA4-H is suspected, the drug should be held. Methylprednisolone (1–2 mg/kg per day i.v.) should be given for a few days, followed by prednisone (1–2 mg/kg per day orally), gradually tapered over 4 weeks (24, 32). An alternative high-dose steroid regimen may be 4 mg dexamethasone every 6 h for 7 days, gradually tapered to 0.5 mg/day and then shifting to a replacement dose of hydrocortisone (24, 75). Notably, as MRI findings may be normal, the treatment should be started as soon as possible once hypopituitarism is documented (76). However, several aspects regarding the treatment with glucocorticoids in anti-CTLA4-H are still unclear. Indeed, it is not known whether high-dose corticosteroids are necessary or if lower doses may suffice. Also, the timing of the tapering should be evaluated as, sometimes, hypophysitis relapse follows the reduction of glucocorticoids to replacement dose. Finally, the potential of the precautionary use of steroids (before starting anti-CTLA4-mAbs) in reducing the onset and/or long-term sequelae of hypophysitis/hypopituitarism, especially in preventing prolonged substitutive treatment, remains to be evaluated.
Many of the current unexplained issues regarding anti-CTLA4-H might be clarified by findings that could emerge from a pathology evaluation on pituitary biopsies of patients presenting with this adverse event. However, to the best of our knowledge, no patient with anti-CTLA4-H has undergone a pituitary biopsy. Indeed, to perform a pituitary biopsy in cancer patients with a suspected anti-CTLA4-H may be unethical, as it is not essential for the diagnosis nor for the treatment of this condition (unless pituitary metastasis is in the differential diagnosis).
Alternatively, evaluations could be performed on biopsy material obtained postmortem. However, postmortem material might not guarantee the quality of pathology assessment. Anti-CTLA4-H remains highly interesting as a unique model of research. Indeed, it allows to monitor the variation of the known pituitary antigens and antibodies in a homogeneous cohort of patients (a specific disease and a known pituitary-damaging drug), testing their role as diagnostic and predictive tools. Obviously, this context appears also as a fertile ground to research new pituitary antigens. Pituitary antigens and antibodies could be monitored in these patients at baseline, before each cycle of treatment, and during follow-up. Such a study could help to define a series of important clinical, laboratory, and radiological features aimed at better defining the real incidence, the diagnosis, and the pathogenesis of anti-CTLA4-H. The gained information could be helpful in evaluating the potential existence of a subclinical anti-CTLA4-H, the likely impact of this syndrome on the quality of life, and the possible existence of a subgroup of patients with factors that could predispose to develop anti-CTLA4-H. This might not only reduce the rate of life-threatening complications, such as secondary adrenal insufficiency or other pituitary deficits, but also allow an optimized use of these drugs from a pharmaco-economy point of view.
Pituitary dysfunction induced by mAbs is limited to anti-CTLA4 agents and is occasionally seen with a PD1 inhibitor. With the expanding use of ipilimumab, hypophysitis, a previously very rare disease, will be progressively encountered by oncologists and endocrinologists in clinical practice. The key approach to obtain the optimal management of this potentially life-threatening adverse event is a rapid and timely diagnostic and therapeutic intervention. This goal could be obtained with a high level of communication between clinical staff and patients (33). Continued efforts should be made to optimize the therapeutic use of mAbs, as their efficacy may be relatively modest in absolute magnitude compared with their high cost (2). To this aim, discovery of new predictive biomarkers of both activity and toxicity remains a priority.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the review.
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
Co-authors profiles are given below:
Francesco Torino is an Assistant Professor of Medical Oncology at School of Medicine of Tor Vergata University of Rome. His research focuses on the endocrine toxicity of targeted/cytotoxic agents, cancer autophagy and the clinical utility of circulating tumor cells.
Agnese Barnabei is a Senior Assistant at the Endocrinology Unit, “Regina Elena” National Cancer Institute of Rome. Current areas of research include the endocrine toxicity of targeted/cytotoxic agents and innovative therapeutic approaches of thyroid, adrenal and neuroendocrine cancer.
Rosa Maria Paragliola is Senior resident of Endocrinology at the “A. Gemelli” University Hospital in Rome. Her research focus is on pituitary and adrenal dysfunctions, thyroid hormone physiopathology and genetics of endocrine diseases.
Paolo Marchetti is Professor of Medical Oncology, Chief of Medical Oncology at Sant'Andrea Hospital - Rome, and Director of Postgraduate Course in Oncology of the “Sapienza” University of Rome. Principal areas of research include alterations of cancer hormone receptors and their implications on prognosis/therapy; chemoprevention of breast cancer and detection of genetic mutations.
Roberto Salvatori is Professor of Medicine and Medical Director of the Pituitary Center at the John Hopkins University, Baltimore, MD, USA. He focuses his clinical practice on the diagnosis and treatment of pituitary and adrenal tumors and pituitary dysfunction.
LipskyPEvan der HeijdeDMSt. ClairEWFurstDEBreedveldFCKaldenJRSmolenJSWeismanMEmeryPFeldmannM. Infliximab and methotrexate in the treatment of rheumatoid arthritis. Anti-Tumor Necrosis Factor Trial in Rheumatoid Arthritis with Concomitant Therapy Study Group. New England Journal of Medicine20003431594–1602. (doi:10.1056/NEJM200011303432202).
Van AsscheGLewisJDLichtensteinGRLoftusEVOuyangQPanesJSiegelCASandbornWJTravisSPColombelJF. The London position statement of the World Congress of Gastroenterology on Biological Therapy for IBD with the European Crohn's and Colitis Organization: safety. American Journal of Gastroenterology20111061594–1602. (doi:10.1038/ajg.2011.211).
BlansfieldJABeckKETranKYangJCHughesMSKammulaUSRoyalRETopalianSLHaworthLRLevyC. Cytotoxic T-lymphocyte associated antigen-4 blockage can induce autoimmune hypophysitis in patients with metastatic melanoma and renal cancer. Journal of Immunotherapy200528593–598. (doi:10.1097/01.cji.0000178913.41256.06).
RibasAKeffordRMarshallMAPuntCJHaanenJBMarmolMGarbeCGogasHSchachterJLinetteG. Phase III randomized clinical trial comparing tremelimumab with standard-of-care chemotherapy in patients with advanced melanoma. Journal of Clinical Oncology201331616–622. (doi:10.1200/JCO.2012.44.6112).
WeberJSDummerRde PrilVLebbéCHodiFS. MDX010–20 Investigators. Patterns of onset and resolution of immune-related adverse events of special interest with ipilimumab: detailed safety analysis from a phase 3 trial in patients with advanced melanoma. Cancer20131191675–1682. (doi:10.1002/cncr.27969).
VoskensCJGoldingerSMLoquaiCRobertCKaehlerKCBerkingCBergmannTBockmeyerCLEigentlerTFluckM. The price of tumor control: an analysis of rare side effects of anti-CTLA-4 therapy in metastatic melanoma from the ipilimumab network. PLoS ONE20138e53745. (doi:10.1371/journal.pone.0053745).
Lebbe C O'Day SJ Chiarion-Sileni V Gajewski T Pehamberger H Bondarenko I Queirolo P Lundgren S Roman L Verschraegen C et al. Analysis of the onset and resolution of immune-related adverse events during treatment with ipilimumab in patients with metastatic melanoma [abstract]. Paper presented at: Perspectives in Melanoma XII; October 2-4 2008; Scheveningen/The Hague Netherlands. Oral Abstract O-015
AttiaPPhanGQMakerAVRobinsonMRQuezadoMMYangJCSherryRMTopalianSLKammulaUSRoyalRE. Autoimmunity correlates with tumor regression in patients with metastatic melanoma treated with anti-cytotoxic T-lymphocyte antigen-4. Journal of Clinical Oncology2005236043–6053. (doi:10.1200/JCO.2005.06.205).
DowneySGKlapperJASmithFOYangJCSherryRMRoyalREKammulaUSHughesMSAllenTELevyCL. Prognostic factors related to clinical response in patients with metastatic melanoma treated by CTL-associated antigen-4 blockade. Clinical Cancer Research2007136681–6688. (doi:10.1158/1078-0432.CCR-07-0187).
YangJCHughesMKammulaURoyalRSherryRMTopalianSLSuriKBLevyCAllenTMavroukakisS. Ipilimumab (anti-CTLA4 antibody) causes regression of metastatic renal cell cancer associated with enteritis and hypophysitis. Journal of Immunotherapy200730825–830. (doi:10.1097/CJI.0b013e318156e47e).
SarnaikAAYuBYuDMorelliDHallMBogleDYanLTarganSSolomonJNicholG. Extended dose ipilimumab with a peptide vaccine: immune correlates associated with clinical benefit in patients with resected high-risk stage IIIc/IV melanoma. Clinical Cancer Research201117896–906. (doi:10.1158/1078-0432.CCR-10-2463).
MadanRAMohebtashMArlenPMVergatiMRauckhorstMSteinbergSMTsangKYPooleDJParnesHLWrightJJ. Ipilimumab and a poxviral vaccine targeting prostate-specific antigen in metastatic castration-resistant prostate cancer: a phase 1 dose-escalation trial. Lancet Oncology201213501–518. (doi:10.1016/S1470-2045(12)70006-2).
van den EertweghAJVersluisJvan den BergHPSantegoetsSJvan MoorselaarRJvan der SluisTMGallHEHardingTCJoossKLowyI. Combined immunotherapy with granulocyte–macrophage colony-stimulating factor-transduced allogeneic prostate cancer cells and ipilimumab in patients with metastatic castration-resistant prostate cancer: a phase 1 dose-escalation trial. Lancet Oncology201213509–517. (doi:10.1016/S1470-2045(12)70007-4).
ReckMBondarenkoILuftASerwatowskiPBarlesiFChackoRSebastianMLuHCuillerotJMLynchTJ. Ipilimumab in combination with paclitaxel and carboplatin as first-line therapy in extensive disease-small-cell lung cancer: results from a randomized, doubleblind, multicenter phase 2 trial. Annals of Oncology20132475–83. (doi:10.1093/annonc/mds213).
Di GiacomoAMAsciertoPAPillaLSantinamiMFerrucciPFGiannarelliDMarascoARivoltiniLSimeoneENicolettiSV. Ipilimumab and fotemustine in patients with advanced melanoma (NIBIT-M1): an open-label, single-arm phase 2 trial. Lancet Oncology201213879–886. (doi:10.1016/S1470-2045(12)70324-8).
LynchTJBondarenkoILuftASantinamiMFerrucciPFGiannarelliDMarascoARivoltiniLSimeoneENicolettiSV. Ipilimumab in combination with paclitaxel and carboplatin as first-line treatment in stage IIIB/IV non-small-cell lung cancer: results from a randomized, double-blind, multicenter phase II study. Journal of Clinical Oncology2012302046–2054. (doi:10.1200/JCO.2011.38.4032).
Di GiacomoAMDanielliRCalabròLBertocciENanniciniCGiannarelliDBalestrazziAVigniFRiversiVMiraccoC. Ipilimumab experience in heavily pretreated patients with melanoma in an expanded access program at the University Hospital of Siena (Italy). Cancer Immunology Immunotherapy201160467–477. (doi:10.1007/s00262-010-0958-2).
WilgenhofSDu FourSVandenbrouckeFEveraertHSalmonILiénardDMarmolVDNeynsB. Single-center experience with ipilimumab in an expanded access program for patients with pretreated advanced melanoma. Journal of Immunotherapy201336215–222. (doi:10.1097/CJI.0b013e31828eed39).
KuGYYuanJPageDBSchroederSEPanageasKSCarvajalRDChapmanPBSchwartzGKAllisonJPWolchokJD. Single-institution experience with ipilimumab in advanced melanoma patients in the compassionate use setting: lymphocyte count after 2 doses correlates with survival. Cancer2010161767–1775. (doi:10.1002/cncr.24951).
RibasACamachoLHLopez-BeresteinGPavlovDBulanhaguiCAMillhamRComin-AnduixBReubenJMSejaEParkerCA. Antitumor activity in melanoma and anti-self responses in a phase I trial with the anti-cytotoxic T lymphocyte-associated antigen 4 monoclonal antibody CP-675,206. Journal of Clinical Oncology2005238968–8977. (doi:10.1200/JCO.2005.01.109).
FreemanGJLongAJIwaiYBourqueKChernovaTNishimuraHFitzLJMalenkovichNOkazakiTByrneMC. Engagement of the PD-1 immunoinhibitory receptor by a novel B7 family member leads to negative regulation of lymphocyte activation. Journal of Experimental Medicine20001921027–1034. (doi:10.1084/jem.192.7.1027).
TaubeJMAndersRAYoungGDXuHSharmaRMcMillerTLChenSKleinAPPardollDMTopalianSL. Colocalization of inflammatory response with B7-H1 expression in human melanocytic lesions supports an adaptive resistance mechanism of immune escape. Science Translational Medicine20124127ra37. (doi:10.1126/scitranslmed.3003689).
BreunisWBTarazona-SantosEChenRKileyMRosenbergSAChanockSJ. Influence of cytotoxic T lymphocyte-associated antigen 4 (CTLA4) common polymorphisms on outcome in treatment of melanoma patients with CTLA-4 blockade. Journal of Immunotherapy200831586–590. (doi:10.1097/CJI.0b013e31817fd8f3).
RoyalRELevyCTurnerKMathurAHughesMKammulaUSSherryRMTopalianSLYangJCLowyI. Phase 2 trial of single agent ipilimumab (anti-CTLA-4) for locally advanced or metastatic pancreatic adenocarcinoma. Journal of Immunotherapy201033828–833. (doi:10.1097/CJI.0b013e3181eec14c).
AnsellSMHurvitzSAKoenigPALaPlantBRKabatBFFernandoDHabermannTMInwardsDJVermaMYamadaR. Phase I study of ipilimumab, an anti-CTLA-4 monoclonal antibody, in patients with relapsed and refractory B cell non-Hodgkin lymphoma. Clinical Cancer Research2009156446–6453. (doi:10.1158/1078-0432.CCR-09-1339).
WolchokJDNeynsBLinetteGNegrierSLutzkyJThomasLWaterfieldWSchadendorfDSmylieMGuthrieTJr. Ipilimumab monotherapy in patients with pretreated advanced melanoma: a randomised, double-blind, multicentre, phase 2, dose-ranging study. Lancet Oncology201011155–164. (doi:10.1016/S1470-2045(09)70334-1).
O'DaySJMaioMChiarion-SileniVGajewskiTFPehambergerHBondarenkoINQueiroloPLundgrenLMikhailovSRomanL. Efficacy and safety of ipilimumab monotherapy in patients with pretreated advanced melanoma: a multicenter single-arm phase II study. Annals of Oncology2010211712–1717. (doi:10.1093/annonc/mdq013).
KirkwoodJMLoriganPHerseyPHauschildARobertCMcDermottDMarshallMAGomez-NavarroJLiangJQBulanhaguiCA. Phase II trial of tremelimumab (CP-675,206) in patients with advanced refractory or relapsed melanoma. Clinical Cancer Research2010161042–1048. (doi:10.1158/1078-0432.CCR-09-2033).