Abstract
Objective
One of the most severe endocrine side effects of immune checkpoint inhibitors (ICI) is hypophysitis leading to adrenal insufficiency. Recovery is rare, although it has been reported after high-dose glucocorticoid treatment. This is the first randomised study to evaluate whether hormonal recovery differs in patients treated with high-dose glucocorticoids versus glucocorticoid replacement therapy.
Design/Methods
In this single-centre, open, randomised controlled study, patients with ICI associated hypophysitis were randomised 1:1 to high-dose glucocorticoid treatment (1 mg/kg of prednisolone for two weeks, followed by tapering until week 7 and a switch to hydrocortisone 20 mg total daily dose in week 8) or glucocorticoid replacement therapy (hydrocortisone 20 mg total daily dose) over 8 weeks. The primary outcome was the frequency of hormonal axes recovery.
Results
Between 17th April 2019 and 16th September 2022, 18 out of the 20 randomised patients finished the trial; eight completed high-dose, 10 glucocorticoid replacement. Nine patients presented with hyponatraemia, two had typical changes on MRI, 12 had isolated adrenal insufficiency, and six had an additional hormone deficiency. None of the patients in neither group experienced a recovery in adrenal function. One patient in each group showed amelioration of hypogonadism. There was a significant, unfavourable treatment effect of high-dose treatment on HbA1c (mean treatment effect 5.16, 95% confidence interval 0.31 to 10.02, p = 0.039).
Conclusions
High-dose glucocorticoid treatment was not effective in restoring adrenal function and leads to adverse effects on glucose metabolism. We therefore do not recommend its use for the treatment of ICI associated hypophysitis, except for compressive symptoms.
Keywords: Immune checkpoint inhibitor associated hypophysitis, endocrine adverse events, hypopituitarism, adrenal insufficiency
Background
Immune checkpoint inhibitors (ICI) have considerably improved the efficacy of anticancer treatment of various cancer types. Side effects of ICI treatment include, amongst others, the effect on endocrine glands, the most severe and potentially life-threatening of which is ICI-associated hypophysitis. It is defined as a functional deficit in one or more pituitary axes, potentially accompanied by subtle magnetic resonance imaging (MRI) abnormalities, developing in a patient with cancer receiving ICI therapy [1]. The development of pituitary immune-related adverse events has been associated with therapeutic efficacy as it predicted better prognosis for both non-small cell lung cancer and malignant melanoma patients [2,3]. The incidence of hypophysitis ranges from 0.4% after programmed cell death protein 1 (PD-1) blocking antibodies up to 6.4% with combination therapies [4] and was as high as 13.6% after cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4) blocking antibodies in some reports with a time to onset between 9.3 and 25.8 weeks [5]. In anti-CTLA-4-induced hypophysitis, ACTH deficiency is most common and is often accompanied by other pituitary insufficiencies, while anti-PD-1-induced hypophysitis often presents with isolated ACTH deficiency [6].
Presenting symptoms include fatigue, nausea, loss of appetite, myalgias/arthralgias, malaise, dizziness, mild cognitive defects, and sometimes headache [1,5]. These symptoms can be discrete and easily confused with symptoms of the underlying cancer, warranting regular, repeated laboratory screening, especially in patients with new or increasing signs and symptoms such as fatigue and weakness [6]. As hyponatraemia tends to be one of the most common presenting laboratory findings (63%) [6], sodium levels can possibly aid in establishing the diagnosis, especially if hormonal analyses are not readily available. After clinical and laboratory diagnosis of hypophysitis, magnetic resonance imaging (MRI) of the pituitary gland should be performed, potentially showing mild to moderate diffuse enlargement of the pituitary gland, with either homogenous or heterogeneous enhancement after contrast administration [7,8].
Prompt diagnosis facilitated by regular screening is key to avert the risk of adrenal crisis and to improve quality of life by ensuring adequate treatment. High-dose glucocorticoids were used to treat ICI associated hypophysitis in the first years after introduction of IC. Glucocorticoid replacement has been carried out in recent years and is also recommended by previously published guidelines [6] for several reasons. Firstly, higher doses of glucocorticoids have been associated with reduced survival compared to lower doses, suggesting a potential negative effect on the efficacy of ICI [9]. Secondly, high dose glucocorticoid treatment has more potential adverse events than replacement alone including risk of diabetes, osteoporosis, etc. Thirdly, recovery of hypopituitarism is unlikely, regardless of the treatment used, although several cases have been reported after high-dose glucocorticoids [7,10]. If recovery was observed, however, this was mostly the case for the thyroid and gonadal axes, and only rarely for secondary adrenal insufficiency [7,11]. Of note, all of these data were derived from retrospective studies, whereas no randomised controlled trial on this issue has been published to date.
The aim of this single-centre, open, randomised study was therefore to compare high-dose glucocorticoid treatment with glucocorticoid replacement in patients with ICI associated hypophysitis with regard to endocrine and metabolic outcomes. The primary outcome measure was the frequency of resolution of hormonal deficiencies, hypothesising no difference between groups. Resolution of a particular endocrinopathy was defined as normalisation of levels of the primary target gland hormone and the corresponding pituitary hormone following discontinuation of hormone replacement as defined by the Endocrine Society Guidelines [12]. Secondary outcome measures include potential adverse effects of high dose glucocorticoid therapy, determined by the measurement of parameters of glucose (fasting glucose, glycated haemoglobin (HbA1c)) and lipid metabolism as well as bone turnover markers.
Methods
Study design
This open, randomised, controlled trial was performed at the Division of Endocrinology and Diabetology (Figure 1), adhering to the CONSORT statement. Patients treated with ICI were regularly monitored by their treating physicians for endocrine adverse events according to our standard operating procedure based on [13] (see Supplementary file).
Figure 1.
Study flow chart.
Toxicity grading based on the criteria defined by Corsello and colleagues [14], with modification, as used in Min [15] (see Supplementary file) ranging from 1 to 5: 1. asymptomatic, 2. symptomatic, 3. severe symptoms; 4. life-threatening consequences; 5. Death. Study participants were randomly assigned 1:1 to either treatment. MRI of the pituitary gland was performed after clinical and laboratory diagnosis of hypophysitis.
ICI treatment was continued in all patients after the diagnosis of hypophysitis. In the replacement group, ICI was continued according to schedule [7], in the high-dose glucocorticoid group, continuation of ICI treatment was postponed until a physiological glucocorticoid equivalent dose was reached to account for the possibility of a detrimental effect of high-dose glucocorticoid treatment on immune therapy efficacy [16].
Definition of hypopituitarism
The definition of hypopituitarism was based on current guidelines. As suggested by Fleseriu et al. 2016 [12], serum cortisol levels measured at 8 to 9 am were used as the first-line test for diagnosing central adrenal insufficiency. Provided there was no suppression of endogenous cortisol secretion by exogenous glucocorticoid therapy such as dexamethasone, a cortisol level of <3 µg/dL (= 30 ng/mL = 83 nmol/L) was considered indicative of and a cortisol level >15 µg/dL (= 150 ng/mL = 414 nmol/L) excluded adrenal insufficiency. When morning cortisol levels ranged between 3 and 15 µg/dL, a corticotropin stimulation test was performed. Peak cortisol levels of <18.1 µg/dL (= 181 ng/mL = 500 nmol/L) at 30 or 60 min were considered indicative of adrenal insufficiency [12]. The diagnosis of central hypothyroidism was established in light of a free thyroxine (fT4) level below the laboratory reference range in conjunction with a low, normal, or mildly elevated thyroid stimulating hormone (TSH) [12]. Central hypogonadism in males was defined in the presence of low serum testosterone levels and features of testosterone deficiency and/or impaired spermatogenesis; in premenopausal females, when serum oestrogens were low together with impaired ovulation with oligomenorrhea or amenorrhea; in postmenopausal women, in the absence of high serum follicle-stimulating hormone (FSH) and (luteinising hormone) LH [12]. Growth hormone (GH) deficiency was diagnosed based on reduced insulin-like growth factor 1 (IGF-1) levels, accepting insufficient sensitivity and specificity for the diagnosis of GH deficiency in adults [17,18]. However, we abstained from dynamic testing due to the lack of therapeutic consequence, as our patients’ underlying malignant comorbidity would have posed a contraindication to GH treatment. Lastly, diabetes insipidus was ruled out in the absence of polyuria (more than 50 mL/kg of body weight/24 h, 3.5 L/d in a 70-kg person), and a urine osmolarity of greater than 600 mOsmol/L together with high serum osmolarity (>295 mOsmol/L) [12].
Laboratory measurements
Laboratory parameters were measured at the Endocrinology Lab Platform of the Division of Endocrinology and Diabetology, the Clinical Institute of Medical and Chemical Laboratory Diagnostics and at the laboratory of the Division of Rheumatology and Immunology. For further details on the assays used, please see the Supplementary file.
Study population
Twenty patients, i.e. 10 patients in each group were meant to be included in the study, a usual sample size for a pilot RCT. No formal sample size and power calculation was performed. Inclusion criteria were treatment with one ICI alone or a combination of CTLA-4-, PD-1, PD-L1 antibodies, age ≥ 18 years, and diagnosis of ICI associated hypophysitis based on clinical, hormonal and/or radiologic findings. Exclusion criteria were preexisting glucocorticoid therapy as well as pre-existing pituitary or adrenal diseases.
Randomisation
Randomisation was carried out using a web-based software (http://www.randomizer.at) with good clinical practice and compliance as confirmed by the Austrian Agency for Health and Food Safety. Eligible participants were openly randomly assigned in a 1:1 ratio to receive either high-dose glucocorticoid therapy or glucocorticoid replacement for eight weeks.
Intervention
High-dose glucocorticoid treatment consisted of prednisolone at a dose of 1 mg/kg for two weeks followed by a taper (50 mg once daily (OD), 37.5 mg OD, 25 mg OD, 12.5 mg OD, and 6.25 mg OD for one week, respectively) to a physiological replacement dose (hydrocortisone 20 mg total daily dose divided into two doses, 15 mg in the morning, 5 mg at lunchtime) in week eight. Evidence for optimal glucocorticoid tapering in other indications is generally scarce [19–21]. The chosen high-dose glucocorticoid dose was based on previously published experience [7,10]. Patients in the glucocorticoid replacement group received treatment with hydrocortisone at a total daily dose of 20 mg divided into two doses (15 mg in the morning, 5 mg at lunchtime) from day 1 onwards, i.e. a physiological replacement dose for adrenal insufficiency.
Where applicable, thyroid hormone replacement was commenced 3–4 days after glucocorticoid therapy had been started to avoid development of adrenal crisis that can be precipitated by replacing thyroid hormones first or concomitantly with glucocorticoids. Where applicable, testosterone/oestradiol replacement was after four weeks, at study visit 2, in case of persisting hypogonadism. All patients were instructed to follow sick day rules and were equipped with a hydrocortisone emergency kit following the Austrian consensus on the management and prevention of adrenal crisis [22].
In the glucocorticoid replacement group, ICI was re-administered after general conditions were stabilised by hormone replacement therapy. In the high-dose glucocorticoid group, ICI was continued once the prednisolone dose was lower than the threshold level of a total daily dose of 7.5 mg.
Blood glucose management
See Supplementary file.
Data analysis
Continuous data following a normal distribution are displayed as means with standard deviations, parameters with a skewed distribution as medians with interquartile ranges and categorical data as percentages. Where appropriate, skewed variables were ln-transformed for parametric analyses. To compare between groups the unpaired Student t-test, the Mann-Whitney-U-test, or the chi-square test were used. Treatment effects with 95% confidence interval and p-values were calculated by ANCOVA for group differences at follow-up with adjustment for baseline values. Analyses were carried out according to the intention-to-treat principle with no data imputation and inclusion of all participants with baseline and follow-up values of the respective outcome variable. A p-value <0.05 was considered statistically significant. Statistical analyses were performed with SPSS version 23 (IBM Corp., Armonk, NY, USA).
Ethical issues
The ethics committee of the Medical University of Graz has approved the study (ethics committee number 29-452 ex 16/17). The study was carried out following the principles of the Declaration of Helsinki. Written informed consent was obtained from all study participants before inclusion in the study. The trial was registered at www.clinicaltrialregister.eu (EudraCT number 2017-001915-35 AT 20170508).
Results
Of the patients screened for this study between April 2019 and September 2022, 20 were randomised to either high-dose glucocorticoid treatment (10 patients) or glucocorticoid replacement (10 patients) for eight weeks. The last visit was completed in November 2022. Two patients in the high-dose glucocorticoid group discontinued early before initiation of treatment and were thus not included in our analyses. Overall, 11 of the 18 patients were women and 11 had malignant melanoma, two had non-small cell lung cancer and one patient had squamous cell carcinoma of the skin, renal cell carcinoma, breast cancer, gastric or colorectal cancer, respectively. Six out of 18 patients received nivolumab alone, five were treated with a combination of ipilimumab and nivolumab, one patient received nivolumab plus relatlimab, five patients received pembrolizumab and one patient atezolizumab. Three patients received ICI therapy as adjuvant, 15 as palliative treatment, two as second line, the other patients as first line option. Baseline characteristics of the study participants are shown in Table 1. Three patients were on hydromorphone treatment before the diagnosis of hypophysitis and throughout the entire study without major changes in hydromorphone dose. All three patients showed a documented rapid and pronounced decline in cortisol and ACTH levels over the course of 3–4 weeks after the initiation of immune therapy, pointing towards ICI-associated hypophysitis.
Table 1.
Overview of clinical data of all study participants.
| Replacement group | High-dose GC group | |
|---|---|---|
| Sex (number of females) | 6 (60%) | 5 (63%) |
| Age at diagnosis (years) | 64 ± 12.6 | 68.3 ± 5.7 |
| Time on ICI to onset of hypophysitis (months) | 5.5 (2.8-15.3) | 5 (2.5-11.3) |
| Isolated adrenal insufficiency (number of patients) | 6 (60%) | 6 (75%) |
| BMI (kg/m2) | 28.0 ± 4.8 | 29.6 ± 4.6 |
| Waist-hip ratio | 0.99 ± 0.1 | 0.97 ± 0.1 |
| Systolic blood pressure (mmHg) | 124 (110-143) | 139 (110-140) |
| Diastolic blood pressure (mmHg) | 81 (70-84) | 91 (67-93) |
ICI = immune checkpoint inhibitor, GC = glucocortiocoid, BMI = body mass index.
Clinical symptoms of hypophysitis were fatigue/asthenia in seventeen, and weight loss in seven patients. Twelve patients complained of loss of appetite and one patient was completely asymptomatic. Half of the patients had hyponatraemia on laboratory findings which resolved in all patients at visit 2 after initiation of glucocorticoid replacement or treatment. One patient was asymptomatic regarding hypophysitis (grade 1 according to toxicity grading), 10 patients had grade 2, six patients had grade 3, and one patient had grade 4 hypophysitis. Changes typical of hypophysitis on MRI, which was performed 3 ± 3.5 weeks after visit 1, were found in two patients. Two patients had pituitary microadenomas on MRI. All other patients had an inconspicuous pituitary gland on MRI.
Twelve patients had isolated adrenal insufficiency, one patient had thyroid hormone deficiency, four patients had gonadotropic deficiencies, and one had gonadotropic and somatotropic deficiency in addition to adrenal insufficiency (Figure 2). In two patients with hypogonadism, one patient on high-dose glucocorticoid treatment had resolution of LH, FSH and total testosterone. In that patient, only free testosterone remained slightly below the reference range during follow-up, so that no testosterone replacement therapy was initiated. The second patient with hypogonadism was on glucocorticoid replacement therapy and complete resolution of hypogonadotropic hypogonadism was observed during follow-up. ACTH and cortisol levels between groups did not differ significantly at baseline, and there was no significant treatment effect of high-dose glucocorticoid therapy on these parameters after eight weeks (Table 2). As for the primary endpoint, none of the patients, neither in the glucocorticoid replacement nor in the high-dose glucocorticoid group, experienced a recovery of adrenal insufficiency (Figure 2).
Figure 2.
Hormone deficiencies at baseline and after 8 weeks in the high-dose versus the replacement group. AI: adrenal insufficiency, hypogon: hypogonadism, GH def: growth hormone deficiency.
Table 2.
Laboratory parameters of participants in the high-dose glucocorticoid and the glucocorticoid replacement group at baseline, after four and eight weeks (i.e. Visit 1, 2, and 3).
| Replacement group |
High-dose GC group |
|||||||
|---|---|---|---|---|---|---|---|---|
| Visit 1 | Visit 2 | Visit 3 | Visit 1 | Visit 2 | Visit 3 | Treatment effect | p-value (V1 to V3) | |
| Anthropometric parameters | ||||||||
| BMI | 28.0 ± 4.8 | 28.0 ± 5.2 | 28.0 ± 5.1 | 28.4 ± 3.4 | 28.7 ± 3.6 | 29.2 ± 3.4 | 0.87 (−0.18 to 1.91) | 0.097 |
| Waist−hip ratio | 0.99 ± 0.1 | 0.96 ± 0.10 | 0.93 ± 0.08 | 0.97 ± 0.1 | 0.96 ± 0.06 | 0.97 ± 0.05 | 0.04 (−0.03 to 0.11) | 0.221 |
| Laboratory parameters | ||||||||
| ACTH (pg/mL) | 5.0 (5.0-10.2) | 5.0 (5.0-5.8) | 5.0 (5.0-7.4) | 5.0 (5.0-8.2) | 5,0 (5,0-5,0) | 5,0 (5,0-6,8) | −0.30 (−2.49 to 1.90) | 0.747 |
| Cortisol (µg/dL) | 1.7 (0.5-11.7) | 1.0 (0.5-3.1) | 0.9 (0.5-2.8) | 2.0 (1.4-8.4) | 1.7 (1.0-16.5) | 0.9 (0.6-1.1) | 0.14 (−1.04 to 1.32) | 0.575 |
| Sodium (mmol/L) | 138 (130-140) | 141 (139-142) | 140 (140-141) | 131 (128-136) | 141 (138-142) | 142 (140-142) | 0.93 (−0.83 to 2.70) | 0.275 |
| Fasting glucose | 86 (80-102) | 98 (88-106) | 90 (87-96) | 87 (75-107) | 91 (77-113) | 95 (76-100) | −5.8 (−17.9 to 6.2) | 0.378 |
| HbA1c (mmol/mol) | 38 (34-42) | 38 (38-40) | 38 (33-40) | 38 (36-42) | 42 (38-44) | 42 (38-44) | 5.16 (0.31 to 10.02) | 0.039 |
| Total cholesterol (mg/dl) | 170 (146-227) | 202 (166-225) | 218 (162-238) | 158 (137-230) | 228 (205-261) | 180 (165-244) | −9.93 (−38.46 to 18.60) | 0.523 |
| HDL cholesterol (mg/dl) | 44 (25-51) | 51 (40-71) | 57 (40-74) | 31 (20-48) | 102 (79-122) | 60 (52-67) | 2.66 (−14.12 to 19.44) | 0.779 |
| LDL cholesterol (mg/dl) | 99 (74-150) | 115 (88-135) | 101 (87-155) | 102 (80-156) | 95 (82-118) | 101 (92-140) | −10.59 (−36.39 to 15.21) | 0.491 |
| Triglycerides (mg/dl) | 129 ± 34 | 146 ± 44 | 156 ± 82 | 176 ± 73 | 156 ± 88 | 156 ± 69 | −34.5 (−110.0 to 40.9) | 0.345 |
| Bone-specific alkaline phosphatase (μg/L) | 17.6 (12,8-21,8) | 15.0 (12,3-24,5) | 16,1 (11,9-24,0) | 15,4 (10,0-18,9) | 14,3 (12,7-19,1) | 12,7 (11,6-14,9) | −1.91 (−6.31 to 2.49 | 0.233 |
| Osteocalcin (ng/mL) | 25,9 ± 10,2 | 33,7 ± 14,7 | 35,1 ± 18,8 | 20,3 ± 15,1 | 8,8 ± 2,3 | 20,4 ± 6,6 | −11.99 (−26.97 to 2.98) | 0.108 |
| P1NP (ng/mL) | 98,4 (57,0-123,1) | 96,3 (57,7-125,1) | 91,9 (43,6-129,9) | 60,0 (37,5-98,3) | 22,9 (19,8-41,1) | 50,1 (38,6-57,1) | −17.75 (−46.30 to 10.81) | 0.332 |
| ß-Crosslaps (ng/mL) | 0,63 (0,41-0,74) | 0,59 (0,38-0,82) | 0,65 (0,29-0,75) | 0,31 (0,23-0,36) | 0,37 (0,20-0,60) | 0,34 (0,27-0,61) | −0.03 (−0.22 to 0.16) | 0.633 |
Data are shown as median with interquartile ranges or as mean with standard deviations. Treatment effects with 95% confidence interval and p-values were calculated by ANCOVA for group differences at follow-up with adjustment for baseline values. For skewed variables logarithmically transformed values were used in ANCOVA but untransformed values are shown in the table. GC = glucocorticoid, BMI = body mass index, ACTH = adrenocorticotropic hormone, HbA1c = glycaeted haemoglobin, HDL = high-density lipoprotein, LDL = low-density lipoprotein, P1NP = N-terminal propeptide pf type I procollagen; p-values for total cholesterol change from baseline to visit 3 in the replacement group: p = 0.033; in the high-dose group: p = 0.025. p-values for HDL cholesterol change from baseline to visit 3 in the replacement group: p = 0.032; in the high-dose group: p = 0.012.
We observed a significant, unfavourable treatment effect of high-dose glucocorticoid treatment on HbA1c (Table 2). Six patients had prediabetes at baseline, one had pre-existing metformin treatment. One patient had diabetes without treatment but required the initiation of antidiabetic treatment during high-dose glucocorticoid treatment. There was no treatment effect on fasting glucose, parameters of lipid or bone metabolism. When analysing both groups separately, in both the replacement as well as the high-dose glucocorticoid group, patients showed significant increases of total and HDL cholesterol (see Table 2). Only one patient was on lipid-lowering treatment (simvastatin) from the baseline visit onward, which remained unchanged throughout the study. In the high-dose group, OC was significantly decreased at visit 2 compared to baseline (20.3 ± 15.1 ng/mL at baseline vs. 8.8 ± 2.3 ng/mL after four weeks, p = 0.043), which was not the case in the replacement group.
Discussion
This is the first RCT published to date in patients with ICI associated hypophysitis showing that after eight weeks of treatment there was no difference in the remission of hypopituitarism between the high-dose glucocorticoid versus the glucocorticoid replacement group. As none of the patients in both study arms showed any improvement in their adrenal function we conclude that high-dose glucocorticoid therapy is not superior as compared to replacement with hydrocortisone regarding recovery of adrenal function in patients with ICI associated hypophysitis. This finding substantiates previously published guidelines on the management of endocrine adverse events, suggesting high-dose glucocorticoid treatment only in cases of optic chiasm compression or other severe compression symptoms [15,23].
Persistent adrenal insufficiency in ICI hypophysitis patients is in line with previously published data [15,23]. In our study, the only alleged remission observed was resolution of hypogonadism in two patients – one in either group. Gonadal axis recovery has previously been reported to range between 11% and 57% [11,13,14,24–26]. However, interpretation of gonadal and thyroidal axes is challenging due to the influence of a state of illness, often found in cancer patients, potentially leading to hypopituitarism-like laboratory constellations [24]. Clinical amelioration can be followed by improvement of hormone levels, complicating the discrimination between true improvement in hypophysitis and simple recovery from the underlying illness [13,24]. A watch and wait strategy is therefore often warranted to distinguish transient from true hypogonadism, especially in light of often unspecific symptoms at presentation of hypophysitis.
Unspecific symptoms together with the absence of readily available hormone parameters render diagnosis of hypophysitis challenging in clinical practice. In line with previously published literature on the topic [5,15,23], half of our patients presented with new-onset hyponatraemia. A decrease in sodium levels can possibly be a helpful, widely and quickly available aid in the diagnosis of hypophysitis in patients receiving ICI therapy, especially since ACTH and cortisol are usually only measurable during normal working hours. Newly diagnosed hyponatraemia might thus perhaps be helpful in clinical decision making to initiate glucocorticoid replacement therapy even before the availability of cortisol levels, always bearing in mind other potential causes of hyponatraemia of course. Pending results of hormone analyses should never delay the initiation of appropriate treatment as adrenal insufficiency can be a potentially life-threatening condition.
After establishing a clinical and laboratory diagnosis of hypopituitarism, pituitary imaging is warranted in patients with hypophysitis. Two patients had microadenoma, and only two out of 18 patients had typical signs of hypophysitis on MRI. All others had normal MRI, which is in line with previously published data showing MRI abnormalities in only 18% of cases [6]. The median onset time of pituitary enlargement ranges around 1 week before biochemical evidence of hypopituitarism [15]. The pituitary gland is believed to decrease in size over the following ∼4–12 weeks, and subsequently pituitary atrophy may develop [7,8,11,25,27,28]. As imaging is often only performed weeks after clinical and laboratory suspicion of hypophysitis due to its limited availability, transient pituitary changes might already have resolved by that time [7] or may have been subtle to begin with. Baseline MRI for comparison are rarely available. Therefore, an inconspicuous MRI does not rule out hypophysitis. Nonetheless, the performance of an MRI is mandatory to rule out other causes of hypopituitarism such as adenoma or metastases.
Presenting symptoms, time of onset and radiological appearances might differ between patients receiving CTLA-4-antibodies and PD-1/PD-L1-anbibodies [1,29]. The number of patients we describe here is too small, though, to discriminate between these groups and to draw any conclusions on possible differences.
While the relatively low sample size is a limitation of our study, our findings still have major implications for clinical practice. Considering the potentially vast adverse effects of high-dose glucocorticoid treatment, the effect size of high-dose glucocorticoid treatment would have to be significant already in a small cohort of patients treated to justify its use. As we did not see any benefit in any of the eight patients on high-dose glucocorticoid treatment and documented cases with remission of hypopituitarism after high-dose glucocorticoid in the literature are scarce and with questionable diagnoses to begin with, we can reasonably conclude that glucocorticoid replacement is the go-to therapeutic approach of ICI associated hypophysitis.
Adverse events associated with high-dose glucocorticoid therapy may be musculoskeletal, endocrine, cardiovascular, gastrointestinal, neuropsychiatric, dermatologic, or immunologic in nature [29–31]. Despite the short treatment duration, we were able to observe unfavourable effects on glucose metabolism after eight weeks as well as subtle effects on parameters of bone metabolism, i.e. osteocalcin, after 4 weeks already. Beyond its role in bone metabolism, low levels of osteocalcin could also be directly linked to an adverse glucose metabolism [32]. The fact that we did not observe significant treatment effects on fasting glucose may be related to the small sample size of our study and to the fact that anti-diabetic treatment was initiated in one patient with pre-existing diabetes, attenuating possible significant effects of high-dose glucocorticoid treatment.
Likewise, there was no significant treatment effect on lipid metabolism, possibly related to the short treatment course and limited patient number. Interestingly though, in within group comparisons, there was a significant rise in total and HDL cholesterol in both groups following the initiation of treatment. Similar findings have been previously reported in secondary adrenal insufficiency patients, who had lower HDL-cholesterol levels as compared to non-secondary adrenal insufficiency patients [33]. Initiation of glucocorticoid replacement increased HDL-cholesterol in a dose-dependent manner [33,34]. As serum HDL-cholesterol correlates inversely with all-cause and stroke-related mortality and low levels are predictors of cardiovascular events in the general population [35–38], the reduced levels in untreated adrenal insufficiency patients might hypothetically contribute to their two-fold increased risk of cardiovascular disease [39,40]. Nevertheless, we have to acknowledge that due to the low sample size of our study with a missing power calculation, the results of our analyses should only be considered as exploratory. Further, statistically significant findings should be cautiously interpreted in light of multiple tests performed with risk of statistical type 1 errors. While acknowledging all above mentioned limitations of our study, we would like to emphasise its strengths. These are, firstly, the prospective, randomised, controlled nature of the study – the only trial of its kind comparing the two treatment forms of ICI associated hypophysitis, thus substantiating previously published guidelines on the topic [6,41]. Secondly, patients treated with ICI underwent regular and structured screening for endocrine adverse events based on our standard operating procedure, ensuring early diagnosis. In case of conspicuous results, immediate endocrine work-up, e.g. an ACTH stimulation test, was performed and assessed by an endocrinologist to ensure correct diagnosis. Upon diagnosis, patients were quickly included in the study and randomised to account for the possibility that if at all only rapidly initiated high-dose glucocorticoid treatment might lead to hypopituitarism remission. Thirdly, this study emphasizes previous retrospective data suggesting that high-dose glucocorticoids do not have an effect in resolving adrenal insufficiency.
In conclusion, this randomised controlled trial has demonstrated that resolution of hypopituitarism, especially adrenal insufficiency, in ICI associated hypophysitis is highly unlikely, emphasising that the treatment of choice is glucocorticoid replacement. High-dose glucocorticoid treatment should be reserved for compression symptoms only, especially in light of its potentially detrimental side effects on glucose, bone and lipid metabolism.
Supplementary Material
Acknowledgment
We thank our study nurse Roswitha Gumpold for her support in carrying out the study visits and the team in the Endocrinology Lab Platform for carrying out the laboratory analyses.
Funding Statement
This study was funded by the Anniversary Fund of the Oesterreichische Nationalbank (OeNB), grant number 18377.
Authors contributions
Verena Theiler-Schwetz: conceptualisation, methodology, validation, formal analysis, resources, data curation, writing—original draft, writing—review and editing, visualisation, project administration, funding acquisition. Christian Trummer: investigation, validation, writing—review and editing. Lisa Schmitt: investigation, writing—review and editing. Angelika Terbuch: investigation, writing—review and editing. Barbara Obermayer-Pietsch: investigation, resources, writing—review and editing. Erika Richtig: investigation, writing—review and editing, supervision. Stefan Pilz: conceptualisation, methodology, validation, formal analysis, writing—review and editing, supervision. All authors have read and approved the final manuscript.
Disclosure statement
No potential competing interest was reported by the authors.
Ethics approval and consent to participate
The ethics committee of the Medical University of Graz has approved the study. The study was carried out following the principles of the Declaration of Helsinki. Written informed consent was obtained from all study participants before inclusion in the study.
Data availability statement
The data that support the findings of this study are available from the corresponding author, upon reasonable request.
References
- 1.Di Dalmazi G, Ippolito S, Lupi I, et al. Hypophysitis induced by immune checkpoint inhibitors: a 10-year assessment. Expert Rev Endocrinol Metab. 2019;14(6):381–398. doi: 10.1080/17446651.2019.1701434. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Kobayashi T, Iwama S, Yasuda Y, et al. Pituitary dysfunction induced by immune checkpoint inhibitors is associated with better overall survival in both malignant melanoma and non-small cell lung carcinoma: a prospective study. J Immunother Cancer. 2020;8(2):e000779. doi: 10.1136/jitc-000779. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Postow MA, Chesney J, Pavlick AC, et al. Nivolumab and ipilimumab versus ipilimumab in untreated melanoma. N Engl J Med. 2015;372(21):2006–2017. doi: 10.1056/NEJMoa1414428. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Barroso-Sousa R, Barry WT, Garrido-Castro AC, et al. Incidence of endocrine dysfunction following the use of different immune checkpoint inhibitor regimens: a systematic review and meta-analysis. JAMA Oncol. 2018;4(2):173–182. doi: 10.1001/jamaoncol.2017.3064. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Faje A, Reynolds K, Zubiri L, et al. Hypophysitis secondary to nivolumab and pembrolizumab is a clinical entity distinct from ipilimumab-associated hypophysitis. Eur J Endocrinol. 2019;181(3):211–219. doi: 10.1530/EJE-19-0238. [DOI] [PubMed] [Google Scholar]
- 6.Husebye ES, Castinetti F, Criseno S, et al. Endocrine-related adverse conditions in patients receiving immune checkpoint inhibition: an ESE clinical practice guideline. Eur J Endocrinol. 2022;187(6):G1–G21. doi: 10.1530/EJE-22-0689. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Faje AT, Sullivan R, Lawrence D, et al. Ipilimumab-induced hypophysitis: a detailed longitudinal analysis in a large cohort of patients with metastatic melanoma. J Clin Endocrinol Metab. 2014;99(11):4078–4085. doi: 10.1210/jc.2014-2306. [DOI] [PubMed] [Google Scholar]
- 8.Corsello SM, Salvatori R, Barnabei A, et al. Ipilimumab-induced endocrinopathies: when to start corticosteroids (or not). Cancer Chemother Pharmacol. 2013;72(2):489–490. doi: 10.1007/s00280-013-2213-y. [DOI] [PubMed] [Google Scholar]
- 9.Faje AT, Lawrence D, Flaherty K, et al. High-dose glucocorticoids for the treatment of ipilimumab-induced hypophysitis is associated with reduced survival in patients with melanoma. Cancer. 2018;124(18):3706–3714. doi: 10.1002/cncr.31629. [DOI] [PubMed] [Google Scholar]
- 10.Garon-Czmil J, Petitpain N, Rouby F, et al. Immune check point inhibitors-induced hypophysitis: a retrospective analysis of the French Pharmacovigilance database. Sci Rep. 2019;9(1):19415–19419. doi: 10.1038/s41598-019-56026-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Min L, Vaidya A, Becker C.. Association of ipilimumab therapy for advanced melanoma with secondary adrenal insufficiency: a case series. Endocr Pract. 2012;18(3):351–355. doi: 10.4158/EP11273.OR. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Fleseriu M, Hashim IA, Karavitaki N, et al. Hormonal replacement in hypopituitarism in adults: an endocrine society clinical practice guideline. J Clin Endocrinol Metab. 2016;101(11):3888–3921. doi: 10.1210/jc.2016-2118. [DOI] [PubMed] [Google Scholar]
- 13.Byun DJ, Wolchok JD, Rosenberg LM, et al. Cancer immunotherapy—immune checkpoint blockade and associated endocrinopathies. Nature Review Endocrinol 2017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Corsello SM, Barnabei A, Marchetti P, et al. Endocrine side effects induced by immune checkpoint inhibitors. J Clin Endocrinol Metab. 2013;98(4):1361–1375. doi: 10.1210/jc.2012-4075. [DOI] [PubMed] [Google Scholar]
- 15.Min L, Hodi FS, Giobbie-Hurder A, et al. Systemic high-dose corticosteroid treatment does not improve the outcome of ipilimumab-related hypophysitis: a retrospective cohort study. Clin Cancer Res. 2015;21(4):749–755. doi: 10.1158/1078-0432.CCR-14-2353. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Goodman RS, Johnson DB, Balko JM.. Corticosteroids and cancer immunotherapy. Clin Cancer Res. 2023;29(14):2580–2587. doi: 10.1158/1078-0432.CCR-22-3181. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Biller BMK, Samuels MH, Zagar A, et al. Sensitivity and specificity of six tests for the diagnosis of adult GH deficiency. J Clin Endocrinol Metab. 2002;87(5):2067–2079. doi: 10.1210/jcem.87.5.8509. [DOI] [PubMed] [Google Scholar]
- 18.Tritos NA, Biller BMK.. Current concepts of the diagnosis of adult growth hormone deficiency. Rev Endocr Metab Disord. 2021;22(1):109–116. doi: 10.1007/s11154-020-09594-1. [DOI] [PubMed] [Google Scholar]
- 19.Richter B, Neises G, Clar C.. Glucocorticoid withdrawal schemes in chronic medical disorders. A systematic review. Endocrinol Metab Clin North Am. 2002;31(3):751–778. doi: 10.1016/s0889-8529(02)00008-7. [DOI] [PubMed] [Google Scholar]
- 20.Brignola C, De Simone G, Belloli C, et al. Steroid treatment in active Crohn’s disease: a comparison between two regimens of different duration. Aliment Pharmacol Ther. 1994;8(4):465–468. doi: 10.1111/j.1365-2036.1994.tb00316.x. [DOI] [PubMed] [Google Scholar]
- 21.Hings IM, Filipovich AH, Miller WJ, et al. Prednisone therapy for acute graft-versus-host disease: short- versus long-term treatment. A prospective randomized trial. Transplantation. 1993;56(3):577–580. doi: 10.1097/00007890-199309000-00016. [DOI] [PubMed] [Google Scholar]
- 22.Pilz S, Krebs M, Bonfig W, et al. Emergency card, emergency medication, and information leaflet for the prevention and treatment of adrenal crisis (Addison crisis): an Austrian consensus document. J Klin Endokrinol Stoffwechs. 2022;15(1):5–27. doi: 10.1007/s41969-022-00155-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Husebye ES, Allolio B, Arlt W, et al. Consensus statement on the diagnosis, treatment and follow-up of patients with primary adrenal insufficiency. J Intern Med. 2014;275(2):104–115. doi: 10.1111/joim.12162. [DOI] [PubMed] [Google Scholar]
- 24.Ryder M, Callahan M, Postow MA, et al. Endocrine-related adverse events following ipilimumab in patients with advanced melanoma: a comprehensive retrospective review from a single institution. Endocr Relat Cancer. 2014;21(2):371–381. doi: 10.1530/ERC-13-0499. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Juszczak A, Gupta A, Karavitaki N, et al. Ipilimumab: a novel immunomodulating therapy causing autoimmune hypophysitis: a case report and review. Eur J Endocrinol. 2012;167(1):1–5. doi: 10.1530/EJE-12-0167. [DOI] [PubMed] [Google Scholar]
- 26.Blansfield JA, Beck KE, Tran K, et al. Cytotoxic T-lymphocyte-associated antigen-4 blockage can induce autoimmune hypophysitis in patients with metastatic melanoma and renal cancer. J Immunother. 2005;28(6):593–598. doi: 10.1097/01.cji.0000178913.41256.06. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Carpenter KJ, Murtagh RD, Lilienfeld H, et al. Ipilimumab-induced hypophysitis: MR imaging findings. AJNR Am J Neuroradiol. 2009;30(9):1751–1753. doi: 10.3174/ajnr.A1623. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Phan GQ, Yang JC, Sherry RM, et al. Cancer regression and autoimmunity induced by cytotoxic T lymphocyte-associated antigen 4 blockade in patients with metastatic melanoma. Proc Natl Acad Sci USA. 2003;100(14):8372–8377. doi: 10.1073/pnas.1533209100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Oray M, Abu Samra K, Ebrahimiadib N, et al. Long-term side effects of glucocorticoids. Expert Opin Drug Saf. 2016;15(4):457–465. doi: 10.1517/14740338.2016.1140743. [DOI] [PubMed] [Google Scholar]
- 30.So H, Lam TO, Meng H, et al. Time and dose-dependent effect of systemic glucocorticoids on major adverse cardiovascular event in patients with rheumatoid arthritis: a population-based study. Ann Rheum Dis. 2023;82(11):1387–1393. doi: 10.1136/ard-2023-224185. [DOI] [PubMed] [Google Scholar]
- 31.Furuta S, Nakagomi D, Kobayashi Y, LoVAS Collaborators ., et al. Effect of reduced-dose vs high-dose glucocorticoids added to rituximab on remission induction in ANCA-associated vasculitis: a randomized clinical trial. Jama. 2021;325(21):2178–2187. doi: 10.1001/jama.2021.6615. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Schwetz V, Pieber T, Obermayer-Pietsch B.. The endocrine role of the skeleton: background and clinical evidence. Eur J Endocrinol. 2012;166(6):959–967. doi: 10.1530/EJE-12-0030. [DOI] [PubMed] [Google Scholar]
- 33.Hayashi R, Tamada D, Murata M, et al. Glucocorticoid replacement affects serum adiponectin levels and HDL-C in patients with secondary adrenal insufficiency. J Clin Endocrinol Metab. 2019;104(12):5814–5822. doi: 10.1210/jc.2019-00420. [DOI] [PubMed] [Google Scholar]
- 34.Werumeus Buning J, Dimova LG, Perton FG, et al. Downregulation of cholesteryl ester transfer protein by glucocorticoids: a randomised study on HDL. Eur J Clin Invest. 2017;47(7):494–503. doi: 10.1111/eci.12770. [DOI] [PubMed] [Google Scholar]
- 35.Okamura T, Hayakawa T, Kadowaki T, NIPPON DATA90 Research Group ., et al. The inverse relationship between serum high-density lipoprotein cholesterol level and all-cause mortality in a 9.6-year follow-up study in the Japanese general population. Atherosclerosis. 2006;184(1):143–150. doi: 10.1016/j.atherosclerosis.2005.03.042. [DOI] [PubMed] [Google Scholar]
- 36.Gordon DJ, Probstfield JL, Garrison RJ, et al. High-density lipoprotein cholesterol and cardiovascular disease. Four prospective American studies. Circulation. 1989;79(1):8–15. doi: 10.1161/01.cir.79.1.8. [DOI] [PubMed] [Google Scholar]
- 37.Barter P, Gotto AM, LaRosa JC, Treating to New Targets Investigators ., et al. HDL cholesterol, very low levels of LDL cholesterol, and cardiovascular events. N Engl J Med. 2007;357(13):1301–1310. doi: 10.1056/NEJMoa064278. [DOI] [PubMed] [Google Scholar]
- 38.Gordon T, Castelli WP, Hjortland MC, et al. High density lipoprotein as a protective factor against coronary heart disease. The Framingham Study. Am J Med. 1977;62(5):707–714. doi: 10.1016/0002-9343(77)90874-9. [DOI] [PubMed] [Google Scholar]
- 39.Bergthorsdottir R, Leonsson-Zachrisson M, Oden A, et al. Premature mortality in patients with Addison’s disease: a population-based study. J Clin Endocrinol Metab. 2006;91(12):4849–4853. doi: 10.1210/jc.2006-0076. [DOI] [PubMed] [Google Scholar]
- 40.Schumaecker MM, Larsen TR, Sane DC.. Cardiac manifestations of adrenal insufficiency. Rev Cardiovasc Med. 2016;17(3-4):131–136. doi: 10.3909/ricm0820. [DOI] [PubMed] [Google Scholar]
- 41.Percik R, Criseno S, Adam S, et al. Diagnostic criteria and proposed management of immune-related endocrinopathies following immune checkpoint inhibitor therapy for cancer. Endocr Connect. 2023;12(5):e220513. Apr 17 doi: 10.1530/EC-22-0513. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are available from the corresponding author, upon reasonable request.