Background
Modern-day approaches to the treatment of malignancy include the use of therapies that exploit the immune system to increase sensitization to cancer cells. Immune checkpoint inhibitors (ICIs) are a therapeutic modality for malignancy that are included in this approach. The distinct types of ICIs are monoclonal antibodies against CTLA-4, PD-1, and PD-L1. PD-L1 is an antigen expressed on tumor cells that binds PD-1 on T cells, leading to T-cell inhibition and desensitization. PD-1/PD-L1 inhibitors inhibit this specific ligand–receptor interaction, thus suppressing T-cell downregulation.1,2 CTLA-4 inhibitors work through a very similar mechanism: CTLA-4 is a receptor found on T cells that normally binds B7 on presenting host cells, leading to T-cell inhibition.2 By blocking this receptor, T cells remain activated and can become sensitized to cellular antigens more easily. Because of their inherent mechanism, their immune-related adverse events (irAEs) include the dysfunction of several systems, including the endocrine system.3–5
Hypophysitis, defined as a dysfunction of the pituitary gland with resultant hormone deficiencies, specifically secondary adrenal insufficiency (AI), and primary thyroid dysfunction are the most common endocrine irAEs stemming from ICIs.3,4 A meta-analysis of clinical trials reported a hypophysitis frequency of 6.4% with combination ICIs, 3.2% with CTLA-4 inhibitor monotherapy, and <1% with PD-1/PD-L1 inhibitor monotherapy.4 Observational studies have reported frequencies of up to 14% with CTLA-4 inhibitors3,6,7 and 0% to 5% with PD-1/PD-L1 inhibitors1,8; the incidence of adrenalitis (primary AI) in both groups tends to be much lower, at <1%.3 Because ICIs are a relatively new class of drugs, data have been limited in this field. However, in recent years, guidelines from national societies including NCCN9–12 and some expert reviews1,3,5,13,14 have provided practical recommendations for the diagnosis and management of ICI-induced endocrinopathies. Despite this development, many unanswered questions remain regarding the patterns and presentation of hormone dysfunction and the degree of hypopituitarism in patients who develop hypophysitis, the degree of pituitary enlargement on imaging, the possibility of pituitary hormone recovery, the role of high-dose glucocorticoids, and the association with survival in these patients with cancer. The link with survival is important to investigate due to its potential for informing ICI efficacy, but the available literature has shown mixed results.6,15–18 The issue of immortal time bias (ITB) or guarantee-time bias affecting the association of hypophysitis with patient survival has been addressed in only one study.18 Hence, we report a cohort of patients treated with ICIs at a large NCCN Member Institution who developed hypophysitis and secondary AI, and the association with overall survival (OS) before and after addressing ITB. We also describe the diagnostic evaluation and treatment challenges in this population.
Methods
Patient Selection
We performed an Institutional Review Board–approved retrospective cohort study evaluating adult patients with cancer who received an FDA-approved ICI between December 1, 2012, and December 31, 2019. Hypophysitis was defined consistent with previous studies2,8,18,19 and the 2022 NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines) for Management of Immunotherapy-Related Toxicities12: (1) pituitary hormone deficiency (biochemical evidence of deficiency of ≥1 pituitary hormones, including secondary AI, secondary hypothyroidism, prolactin deficiency, growth hormone deficiency, or secondary hypogonadism) with or without characteristic MRI changes suggestive of hypophysitis, or (2) MRI findings suggestive of hypophysitis even in the absence of pituitary hormone deficiencies. For patients already receiving high-dose glucocorticoids for other irAEs, secondary AI due to hypophysitis was diagnosed if the patient required long-term replacement/maintenance doses of glucocorticoids despite adequately tapering off the higher initial glucocorticoid dose.
Study Measures
All evaluations and treatments were at the discretion of the treating physician. Chart review was conducted by study members to identify the demographic, clinical, laboratory, and radiologic data. Thyroid-stimulating hormone (TSH), free thyroxine, total triiodothyronine, follicle-stimulating hormone, luteinizing hormone, prolactin, and estradiol were tested using Roche cobas immunoassays; adrenocorticotrophic hormone (ACTH) was tested using an Immulite chemiluminescent immunoassay (Siemens Healthineers); serum cortisol was tested using a Beckman Coulter competitive binding immunoassay; and testosterone (total and free) and insulin-like growth factor 1 (IGF-1) were tested using liquid chromatography–mass spectrometry. Thyroid irAEs were diagnosed by protocol-based thyroid function testing before each infusion of ICI or if clinical presentation warranted investigation. Secondary hypothyroidism was diagnosed by the presence of low free thyroxine with low or inappropriately normal TSH, tested on 2 separate occasions to exclude nonthyroidal illness and recovering thyroiditis with TSH lag. Secondary AI was defined as a low 8:00 am serum cortisol level and low or inappropriately normal ACTH levels or an inadequate response to a 250 mcg ACTH stimulation test. However, ACTH stimulation tests can be normal in recent-onset secondary AI, and therefore a normal stimulation test did not exclude secondary AI. Secondary hypogonadism was defined as low serum estrogen (in premenopausal women) or testosterone (in men) level and a low or inappropriately normal follicle-stimulating hormone level. Prolactin deficiency was defined as low serum prolactin, and growth hormone deficiency was defined as low serum IGF-1.
Statistics
Descriptive statistics were used to summarize patient characteristics. Continuous variables were compared between groups using nonparametric Wilcoxon rank sum tests, and categorical variables were compared using Fisher exact tests. To address ITB resulting from the time-dependent nature of hypophysitis status, univariate Kaplan-Meier tests followed by multivariable extended Cox regression models incorporating time-dependent variables were used to assess the effects of hypophysitis status on patient OS. Hazard ratios and 95% confidence intervals were calculated, and P values were derived from the models. The assumption of proportional hazards of Cox models for other covariates were assessed using goodness-of-fit tests. A P value <.05 was considered statistically significant.
Results
Study Population Characteristics
A total of 839 patients with cancer received ICIs, of whom 54 received CTLA-4 inhibitor monotherapy, 734 received PD-1/PD-L1 inhibitor monotherapy, and 51 received a combination of CTLA-4 and PD-1/PD-L1 inhibitors. Within the overall cohort, 16 (1.9%) patients developed hypophysitis; in 15 of these patients it was the first endocrinopathy encountered. Median age at initiation of ICI therapy for the entire cohort was 64.9 years. Age at initiation of ICI therapy was significantly lower among the hypophysitis group (57.4 years) compared with the nonhypophysitis group (65.1 years; P=.01). Men comprised 56.6% of the cohort, with no significant difference between the hypophysitis and nonhypophysitis groups (Table 1).
Baseline Characteristics of Patients Before ICI Initiation
When the groups were stratified by malignancy type, patients with melanoma had the highest frequency of hypophysitis (9/157), followed by patients with renal cell carcinoma (4/75) (Table 1). Hypophysitis occurred most frequently after combination ICI therapy (13.7%) compared with CTLA-4 inhibitors (1.9%) and PD-1/PD-L1 inhibitor monotherapy (1.1%; P<.0001) (Table 2).
Frequency of Hypophysitis Categorized by ICI Therapy Type
Natural History and Management of Hypophysitis
In the entire cohort, the median time from ICI therapy to the diagnosis of hypophysitis was 7 months (range, 0.7–29.5 months). All patients with hypophysitis (n=16) had laboratory evidence of secondary AI. Secondary hypothyroidism occurred in 9 (56.3%) of these patients, secondary hypogonadism in 1 (6.25%), and growth hormone deficiency in 2 (12.5%). Pituitary gland enlargement was noted in 6 of 13 (46.2%) patients who underwent MRI of the brain within 3 months of hypophysitis. Pituitary enlargement was seen in 5 of 7 (71.4%) patients after CTLA-4 inhibitor monotherapy or combination therapy compared with 1 of 6 (16.7%) patients after PD-1/PD-L1 inhibitor monotherapy. Follow-up MRI showed resolution of pituitary enlargement in all 6 patients. Of the 16 patients with hypophysitis, 5 (31.3%) received high-dose glucocorticoids: 4 experienced mass effect as defined by the presence of headache and/or diplopia and 1 without headache or diplopia prophylactically received high-dose glucocorticoids following MRI demonstrating hypophysitis out of concern for the development of mass effect. Of these 5 patients who received glucocorticoids, 3 (60%) received combination CTLA-4/PD-1 inhibitor therapy, whereas the remainder received PD-1 inhibitor monotherapy.
As for other endocrine-specific irAEs in the 16 patients who developed hypophysitis, 6 (37.5%) developed primary thyroid dysfunction and 2 (12.5%) developed insulin-deficient diabetes with low C-peptide levels.
Factors Affecting OS
During a median follow-up of 19.4 months, 432 (51.49%) patients died. Median OS was significantly higher in the hypophysitis group (>55.7 months) compared with the nonhypophysitis group (15.9 months; 95% CI, 13.2–19.5 months; P=.003) using traditional Kaplan-Meier survival analysis (Figure 1). Multiple extended Cox regression analyses after adjusting for other covariates, including age at ICI initiation, sex, race, and cancer type (melanoma vs lung cancer vs other), and using hypophysitis as a time-dependent covariate to account for ITB did not show a significant association of hypophysitis occurrence with OS (hazard ratio, 0.21; 95% CI, 0.03–1.49; P=.12) (Table 3).
Kaplan-Meier plots for survival in patients with cancer with and without hypophysitis (P=.003), after treatment with ICIs.
Abbreviation: ICI, immune checkpoint inhibitor.
Citation: Journal of the National Comprehensive Cancer Network 21, 3; 10.6004/jnccn.2022.7098
Kaplan-Meier plots for survival in patients with cancer with and without hypophysitis (P=.003), after treatment with ICIs.
Abbreviation: ICI, immune checkpoint inhibitor.
Citation: Journal of the National Comprehensive Cancer Network 21, 3; 10.6004/jnccn.2022.7098
Kaplan-Meier plots for survival in patients with cancer with and without hypophysitis (P=.003), after treatment with ICIs.
Abbreviation: ICI, immune checkpoint inhibitor.
Citation: Journal of the National Comprehensive Cancer Network 21, 3; 10.6004/jnccn.2022.7098
Multiple Extended Cox Regression Model After Accounting for ITB for Factors Affecting OS in Patients Receiving ICIs
Discussion
In this second largest cancer cohort treated with multiple ICIs at a tertiary care center, we showed that hypophysitis occurs most frequently after a combination of CTLA-4 and PD-1/PD-L1 inhibitors, is characterized by secondary AI in all cases, may not have typical MRI pituitary enlargement when caused by PD-1/PD-L1 inhibitors, and appears to have a survival benefit but not after accounting for ITB. The overall incidence of hypophysitis in our cohort was 1.9% (16/839), being highest after combination ICI therapy and least after PD-1/PD-L1 inhibitor monotherapy. These data are consistent with previous studies in which PD-1/PD-L1 inhibitors have been associated with a lower incidence of hypophysitis relative to CTLA-4 inhibitors.2,3,8,18,20 This may be due in part to CTLA-4 expression on pituitary cells, and with binding of anti–CTLA-4 binding of anti–CTLA-4 receptor has been shown to activate the immune complement system, which then leads to inflammation of the pituitary gland.21
The median time to diagnosis for hypophysitis after initiation of ICI therapy was 7 months (range, 0.7–29.5 months). Given that this study is a retrospective review, it includes some data generated before standardized screening protocols1,5,11–13 had been published or implemented in practice. This wide range could represent variability in presentation of hypophysitis or delay in diagnosis due to the nonspecific findings and symptoms. In addition, in the earlier years of ICI use, there was also more high-dose glucocorticoid therapy utilized as treatment for multiple irAEs, which could have delayed diagnosis until after a patient was weaned off high-dose glucocorticoids. Future prospective studies should be performed with standardized screening protocols to better ascertain the time to development of hypophysitis, especially given the urgency to treatment necessary with secondary AI.
In our study, 13 of 16 patients with hypophysitis had pituitary MRIs performed within 3 months of hypophysitis and pituitary hormone deficiency; 46% of these patients showed typical pituitary enlargement, which was more frequent after CTLA-4 inhibitor or combination therapy compared with PD-1/PD-L1 inhibitor monotherapy. Previous studies have indicated a variable presence of MRI changes associated with the hormonal diagnosis of hypophysitis, with some showing MRI changes in 62% to 77% of patients with hypophysitis related to ICI therapy,18,22 being more common with CTLA-4 inhibitor or combination therapy compared with PD-1/PD-L1 inhibitor monotherapy.2,3,8,18 The different frequency of MRI findings of hypophysitis in our study is likely due to differing population composition across studies regarding ICI class. In our study, half the cases of hypophysitis occurred after PD-1/PD-L1 inhibitor monotherapy, likely accounting for the lower frequency of typical MRI findings of pituitary gland enlargement and enhancement compared with other studies.18,22 All cases of pituitary enlargement resolved during follow-up MRI imaging, confirming that it was indeed due to hypophysitis as opposed to pituitary metastases. The 2022 NCCN Guidelines for ICI-induced hypophysitis recommend obtaining an MRI if the patient is symptomatic during treatment.12 In our opinion, MRI findings would add diagnostic value when the hypophysitis process is in early stages and hormonal evaluation is borderline, especially if the patient received CTLA-4 inhibitors. This could potentially facilitate early detection and treatment of hormonal deficiency. Therefore, we recommend consideration of pituitary MRI even in asymptomatic patients with evidence of pituitary hormone deficiencies, especially when treated with a CTLA-4 inhibitor. Pituitary gland enlargement needs to be followed with an MRI in 3 months to evaluate for resolution, because persistent enlargement suggests pituitary metastases as opposed to hypophysitis.
In our study, 5 patients received high-dose glucocorticoids, 4 of whom were reported to have clinical evidence of mass effect. This was done before subsequent studies demonstrated a lack of clear benefit of high-dose glucocorticoid use on resolution of pituitary enlargement from hypophysitis,2,15,23,24 with one study reporting no effect on survival23 and another reporting reduced survival.15 Hence, we have updated our practice to limit the use of high-dose glucocorticoids to only patients with severe and unresolving mass effects from hypophysitis.
An obstacle in the diagnosis of ICI-induced hypophysitis is exogenous high-dose glucocorticoid use in typical oncologic treatment regimens. In our study, 6 patients appeared to have secondary AI, but they were also on glucocorticoids, making it impossible to distinguish between the responses to the agents; 4 were excluded because of this, but 2 also had secondary hypothyroidism and therefore remained in the hypophysitis group. This demonstrates the importance of further pituitary evaluation to differentiate secondary AI due to exogenous glucocorticoids versus hypophysitis. The NCCN Guidelines also reflect this notion and recommend screening asymptomatic patients for dysfunction of the pituitary–thyroid and pituitary–gonadal axes in addition to the pituitary–adrenal axis.12 Because hypophysitis and secondary AI present acutely before chronic atrophy of the adrenal cortex, an ACTH stimulation test is not reliable in the initial diagnostic workup of ICI-induced hypophysitis, and consequently may result in a normal and falsely reassuring result.1,9,12 It is best to screen with an 8:00 am serum cortisol and ACTH. If a patient is on glucocorticoids for other irAEs, then a slow taper off glucocorticoids is needed. Once below maintenance glucocorticoid doses, further biochemical testing should be performed. Persistent secondary AI would suggest hypophysitis because that caused by exogenous glucocorticoids usually resolves after a few months.
Our study adds to the currently limited literature evaluating the link between ICI-induced hypophysitis and patient OS (Table 4). We found a survival benefit incurred by the hypophysitis group on initial analysis; however, this was not significant after using hypophysitis as a time-dependent covariate to address ITB and adjusting for other confounding variables, including age, sex, race, cancer type, and ICI class. Previous studies in this realm have yielded conflicting results (Table 4). Three studies demonstrated a statistically significant survival benefit with the development of hypophysitis.6,15,16 One study demonstrated differences in OS favoring the group with hypophysitis, but this was not statistically significant after utilizing landmark survival analysis to account for hypophysitis as a time-dependent covariate.18 An additional study demonstrated no improvement in OS with the development of hypophysitis17; however, the investigators did not adjust for confounding factors that could impact survival. Variability in the outcomes of studies regarding survival benefit may indicate that the effect size of improved survival between hypophysitis and nonhypophysitis groups is small. This small difference may only become apparent when factors affecting survival are accounted for and the sample is large enough to adequately power a study for this. Another possible explanation is that the survival benefit accrued by those patients who develop hypophysitis is counterbalanced by the morbidity and mortality of AI. This is supported by the larger survival benefit consistently demonstrated in previous studies regarding ICI-induced thyroiditis25 even after accounting for ITB.26 When untreated or suboptimally treated, AI can be life-threatening. It can pose a diagnostic dilemma due to confounding variables, such as exogenous glucocorticoid use and laboratory diagnostics that are more time- and labor-intensive relative to thyroid function testing. Hence, studies in the future with higher numbers of hypophysitis cases and addressing ITB are needed to delineate the link between the development of hypophysitis and patient survival.
Cohort Studies Reporting Association Between Hypophysitis Occurrence and Survival in Patients Receiving ICIs
Although this is a large cohort study, it is limited by its retrospective nature and limited pituitary hormone testing in all patients treated with an ICI. This limited our ability to analyze subtle hormone deficiencies that were not clinically evident. Due to sample size limitation, we could not analyze the effects of radiologic pituitary abnormality, severity of hypophysitis, mass effects, or ICI class on OS. However, we were able to adjust for the most pertinent variables affecting survival via multiple extended Cox regression analysis. Some data were from before standardized regimens were incorporated into practice, specifically affecting the time to diagnosis and use of high-dose glucocorticoids in our cohort. Because all patients with mass effect and pituitary enlargement received high-dose glucocorticoids, and because secondary AI did not resolve in any of the patients, we could not compare the efficacy of high-dose glucocorticoids on the resolution of hypopituitarism, but this has been evaluated in previous studies.2,15,23,24 The number of patients with hypophysitis and the median follow-up duration of 19.4 months was enough to characterize the patients with hypophysitis but precluded us from providing an accurate OS probability in the entire cohort. The multiple types of cancer allowed us to evaluate their association with hypophysitis occurrence and survival but limited the ability to fully characterize hypophysitis within a specific cancer cohort.
Conclusions
Our study demonstrated that the risk of hypophysitis is highest after combination CTLA-4 and PD-1/PD-L1 inhibitor therapy. Secondary AI is the most frequent hormone deficiency, followed in frequency by secondary hypothyroidism. The diagnosis may be missed initially due to nonspecific symptoms or concurrent use of glucocorticoids, highlighting the importance of biochemical screening and clinical suspicion. MRI of the pituitary can aid in the diagnosis in such situations, but it is important to note that PD-1/PD-L1 inhibitor–induced hypophysitis may lack typical pituitary enlargement, only presenting with secondary AI. Finally, the survival benefit of ICI-induced hypophysitis is dampened after accounting for ITB and possibly by the morbidity/mortality of untreated secondary AI, or limited by the small sample size, and hence needs further investigation in multicenter studies.
References
- 1.↑
Kotwal A. Hypophysitis from immune checkpoint inhibitors: challenges in diagnosis and management. Curr Opin Endocrinol Diabetes Obes 2021;28:427–434.
- 2.↑
Di Dalmazi G, Ippolito S, Lupi I, et al. Hypophysitis induced by immune checkpoint inhibitors: a 10-year assessment. Expert Rev Endocrinol Metab 2019;14:381–398.
- 3.↑
Chang LS, Barroso-Sousa R, Tolaney SM, et al. Endocrine toxicity of cancer immunotherapy targeting immune checkpoints. Endocr Rev 2019;40:17–65.
- 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;4:173–182.
- 5.↑
Wright JJ, Powers AC, Johnson DB. Endocrine toxicities of immune checkpoint inhibitors. Nat Rev Endocrinol 2021;17:389–399.
- 6.↑
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;99:4078–4085.
- 7.↑
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;21:371–381.
- 8.↑
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;181:211–219.
- 9.↑
Girotra M, Hansen A, Farooki A, et al. The current understanding of the endocrine effects from immune checkpoint inhibitors and recommendations for management. JNCI Cancer Spectr 2018;2:pky021.
- 10.↑
Castinetti F, Albarel F, Archambeaud F, et al. French Endocrine Society guidance on endocrine side effects of immunotherapy. Endocr Relat Cancer 2019;26:G1–18.
- 11.↑
Schneider BJ, Naidoo J, Santomasso BD, et al. Management of immune-related adverse events in patients treated with immune checkpoint inhibitor therapy: ASCO guideline update. J Clin Oncol 2021;39:4073–4126.
- 12.↑
Thompson JA, Schneider BJ, Brahmer J, et al. NCCN Clinical Practice Guidelines in Oncology: Management of Immunotherapy-Related Toxicities. Version 1.2022. Accessed October 1, 2022. To view the most recent version, visit https://www.nccn.org
- 13.↑
Quandt Z, Young A, Perdigoto AL, et al. Autoimmune endocrinopathies: an emerging complication of immune checkpoint inhibitors. Annu Rev Med 2021;72:313–330.
- 14.↑
Albarel F, Castinetti F, Brue T. Management of endocrine disease: immune check point inhibitors-induced hypophysitis. Eur J Endocrinol 2019;181:R107–118.
- 15.↑
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;124:3706–3714.
- 16.↑
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:e000779.
- 17.↑
Snyders T, Chakos D, Swami U, et al. Ipilimumab-induced hypophysitis, a single academic center experience. Pituitary 2019;22:488–496.
- 18.↑
Kotwal A, Rouleau SG, Dasari S, et al. Immune checkpoint inhibitor- induced hypophysitis: lessons learnt from a large cancer cohort. J Investig Med 2022;70:939–946.
- 19.↑
Faje A. Immunotherapy and hypophysitis: clinical presentation, treatment, and biologic insights. Pituitary 2016;19:82–92.
- 20.↑
de Filette J, Andreescu CE, Cools F, et al. A systematic review and meta-analysis of endocrine-related adverse events associated with immune checkpoint inhibitors. Horm Metab Res 2019;51:145–156.
- 21.↑
Iwama S, Arima H. Anti-pituitary antibodies as a marker of autoimmunity in pituitary glands. Endocr J 2020;67:1077–1083.
- 22.↑
Nguyen H, Shah K, Waguespack SG, et al. Immune checkpoint inhibitor related hypophysitis: diagnostic criteria and recovery patterns. Endocr Relat Cancer 2021;28:419–431.
- 23.↑
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;21:749–755.
- 24.↑
Lam T, Chan MM, Sweeting AN, et al. Ipilimumab-induced hypophysitis in melanoma patients: an Australian case series. Intern Med J 2015;45:1066–1073.
- 25.↑
Kotwal A, Ryder M. Survival benefit of endocrine dysfunction following immune checkpoint inhibitors for nonthyroidal cancers. Curr Opin Endocrinol Diabetes Obes 2021;28:517–524.
- 26.↑
Cheung YM, Wang W, McGregor B, et al. Associations between immune-related thyroid dysfunction and efficacy of immune checkpoint inhibitors: a systematic review and meta-analysis. Cancer Immunol Immunother 2022;71:1795–1812.
- 27.
Eatrides JM, Weber J, Egan K. Autoimmune hypophysitis is a marker of favorable outcome during treatment of melanoma with ipilimumab. Cancer Res 2015;75(Suppl):Abstract B23.
