A Challenging Case of Severe Cushing’s Syndrome in the Course of Metastatic Thymic Neuroendocrine Carcinoma with a Synchronous Adrenal Tumor

Ectopic ACTH syndrome (EAS) remains one of the most demanding diagnostic and therapeutic challenges for endocrinologists. Thymic neuroendocrine tumors account for 5%–10% of all EAS cases. We report a unique case of a 31-year-old woman with severe EAS caused by primary metastatic combined large-cell neuroendocrine carcinoma and atypical carcinoid of the thymus. The patient presented with severe hypercortisolemia, which was successfully controlled with continuous etomidate infusion. Complex imaging initially failed to detect thymic lesion; however, it revealed a large, inhomogeneous, metabolically active left adrenal mass infiltrating the diaphragm, suspected of primary disease origin. The patient underwent unilateral adrenalectomy, which resulted in hypercortisolemia resolve. The pathology report showed an adenoma with adrenal infarction and necrosis. The thymic tumor was eventually revealed a few weeks later on follow-up imaging studies. Due to local invasion and rapid progression, only partial resection of the thymic tumor was possible, and the patient was started on radio- and chemotherapy.

1 Introduction

Endogenous Cushing’s syndrome (CS) is a rare endocrine condition caused by excess cortisol production with an annual incidence of 0.2–5 cases per million people (1). Adrenocorticotropin (ACTH) hypersecretion of nonpituitary tumors leading to ectopic ACTH syndrome (EAS) accounts for 9%–18% of ACTH-dependent CS cases (12) and represents one of the most common paraneoplastic syndromes (34). Neuroendocrine tumors (NETs) of various locations, degrees of histological differentiation, and aggressiveness potential can lead to EAS; however, most frequently, they derive from the foregut, with the well-differentiated bronchial NET being the most common one in recent series (56). NETs of the thymus (NETTs) represent up to 5% of all thymic tumors, with an incidence of 0.02 per 100,000 people per year in the Caucasian population (78). Up to 50% of the hormonally active NETTs present with ACTH hypersecretion (8) that account for 5%–10% of EAS cases (59). They usually behave aggressively with regional invasion and early distant metastases and lead to the rapid development of severe hypercortisolism (SH), which worsens the initial poor prognosis (811).

Herein, we present a unique case of a patient with EAS caused by a primary metastatic, ACTH-secreting thymic large-cell neuroendocrine carcinoma (LCNEC) with an atypical carcinoid (AC) component with rapid progression, which initially failed to be visualized in imaging studies. Moreover, the diagnostic process was even more difficult because of the co-presence of an adrenal lesion suspected of malignancy on imaging studies and to be the primary origin of the disease.

2 Case report

In April 2019, a 31-year-old previously healthy woman presented to the Emergency Department with a 3-week history of progressing fatigue, muscle weakness, exercise intolerance, headaches, progressive hypertension, generalized swelling, polyuria, polydipsia, and nycturia. Due to the reported symptoms, the patient had previously consulted a family doctor, who initiated oral potassium supplementation because of hypokalemia (2.8 mmol/L) found in basic laboratory tests. On physical examination, the patient presented with significant peripheral pitting edema, high blood pressure (170/100 mmHg), tachycardia (170 beats/minute), and acne lesions on the face, back, and chest. The initial laboratory tests at the Emergency Department showed the following: leukocytosis (13.95 × 109/L) with neutrophilia (12.98 × 109/L) and lymphopenia (0.27 × 109/L), hypochloremic metabolic alkalosis (pH 7.52; HCO3, 38.3 mmol/L; and Cl, 91 mmol/L), hyperglycemia (478 mg/dL), and profound hypokalemia (2.2 mmol/L). Initial laboratory findings are summarized in Table 1A.

Table 1a

www.frontiersin.org

Table 1A Summary of initial laboratory test performed at Emergency Department.

Within the Emergency Department, the patient was considered as a patient with newly diagnosed diabetes and was referred to the Endocrinology Department, where she was started on insulin therapy along with aggressive hypokalemia repletion, antihypertensive treatment, and preventive heparin anticoagulation. Given the overall clinical presentation and resistance to initiated treatment, aggressive CS was quickly suspected. During the first days of hospitalization, the patient also developed agitation with paranoid symptoms; thus, the psychiatrist was consulted, and the patient was additionally started on antipsychotic treatment.

The hormonal evaluation revealed SH with high concentrations of morning (78.2 μg/dL; reference range, 3.7–19.4) and midnight (69.1 μg/dL; reference range, < 5.4) serum cortisol, 24 h urinary free cortisol (UFC) excretion exceeding 65 times the upper reference limit (11,587.5 μg/24 h; reference range, 4.6–176.0), and hyperandrogenemia (testosterone, 6.3 ng/mL; reference range, 0.06–0.8; DHEA-S, 853.2 μg/dL; reference range, 95.8–511.7). ACTH level was markedly elevated (963.7 pg/mL; reference range, 6.0–48.0), confirming ACTH-dependent CS. No dynamic hormonal testing was performed, considering the severe state of the patient. The hormonal findings are summarized in Table 1B.

Table 1b

www.frontiersin.org

Table 1B Summary of initial hormonal assessment in presented patient.

To control SH, continuous etomidate infusion was initiated with significant improvement in the patient’s general condition, edema reduction and normalization of blood pressure, glycemia, and potassium level with a decrease in the need for antihypertensive and insulin treatment, mineralocorticoid receptor blockade, and potassium supplementation. Pituitary magnetic resonance imaging (MRI) revealed no lesion. ACTH-dependent SH with negative pituitary imaging and short duration with rapid progression of symptoms were highly suggestive of EAS. Computed tomography (CT) of the chest, abdomen, and pelvis was performed and revealed a left, inhomogeneous, solid adrenal mass measuring 80 mm ×56 mm ×39 mm of 25 Hounsfield units adjacent to/infiltrating the left dome of the diaphragm, hyperplasia of the right adrenal gland, and numerous sclerotic bone lesions concerning for metastases (Figure 1).

Figure 1

www.frontiersin.org

Figure 1 Frontal (A) and axial (B) CT scans of the abdomen presenting a large lesion of the left adrenal gland. 18F-FDG-PET-CT presenting high metabolic activity of the left adrenal gland lesion (C1, C2) and metastatic bone lesions in the spine and sternum (D).

Plasma and urine metanephrines, renin, and aldosterone levels were within the reference range (Table 1B). However, an excessively elevated concentration of chromogranin A (CgA) was observed (13,835.0 ng/mL; reference range, < 100), which firmly suggested the presence of a NET. The patient underwent whole-body SPECT-CT somatostatin receptor scintigraphy (SRS) with 99mTc-octreotate, which showed no evidence of somatostatin receptor overexpression. Subsequently, whole-body 18F-fluorodeoxyglucose (FDG)-positron emission tomography (PET)-CT was performed and revealed that the left adrenal gland lesion previously found on CT scan is partially metabolically active [maximum standardized uptake value (SUVmax), 3.2] and suspicious of malignancy; the right adrenal gland presented diffused 18F-FDG uptake (SUVmax, 2.6) likely resulting from excessive ACTH overstimulation. 18FDG-PET-CT also showed multifocal metabolically active sclerotic bone lesions in the spine, ribs, clavicles, scapules, sternum, pelvis, femurs, and humerus. A SUVmax 2.1 area (not correlated with CT imaging) was also found in the anterior mediastinum, which has been considered primarily as a residual thymus with physiological FDG uptake (Figure 1).

The clinical presentation was highly suggestive of generalized malignancy with ectopic ACTH secretion with the potential origin in the left adrenal gland. The patient was discussed at a multidisciplinary team meeting and was decided to undergo left-sided open adrenalectomy. After the surgery, the symptoms of hypercortisolemia resolved, the etomidate infusion could have been stopped, and the patient did not require further use of antihypertensive and insulin treatment, mineralocorticoid receptor blockade, and potassium supplementation. Furthermore, postoperatively, a significant decline in cortisol levels was noted (4.08 μg/dL), and the patient was transitioned to oral hydrocortisone. ACTH level also dropped (312.0 pg/mL 2 h after morning dose of oral hydrocortisone); however, it still remained significantly elevated. While waiting for the histopathological result, the patient underwent additional colonoscopy, gastroscopy, and bronchoscopy, but no other potential cancer origin was found. We consulted with the oncologist, and active surveillance was recommended until the histopathological examination results were obtained. The patient was discharged after 5 weeks of hospitalization, awaiting the result of the histopathological examination, in good general condition, requiring only hydrocortisone substitution.

The histopathological examination revealed an adrenal adenoma with the domination of adrenal infarction and necrosis. Immunohistochemistry (IHC) showed the following: CgA (+), EMA (−), synaptophysin (−), S100 (−), CKAE1/AE3 (+), RCC (−), melan-A (−), Ki-67 positive in single adrenal cells. Although its diagnosis was unlikely, according to the histopathology and IHC, it was not possible to clearly exclude the adrenocortical cancer (ACC). However, given ACTH-dependent hypercortisolemia and significantly elevated CgA concentration, an undetected neuroendocrine tumor was considered first.

A follow-up 18F-FDG-PET-CT performed after 8 weeks revealed a metabolically active mass (SUVmax, 9.3) in the superior anterior mediastinum in the thymus location (Figure 2A). CT (Figure 2B) and the subsequently performed MRI (Figure 2C) of the chest confirmed mediastinal mass measuring 42 mm × 33 mm, adjacent to the trachea and superior vena cava, encircling the ascending aorta and aortic arch, most likely corresponding to the invasive thymic malignancy.

Figure 2

www.frontiersin.org

Figure 2 Eight-week follow-up 18F-FDG-PET-CT presenting metabolically active lesion in the superior anterior mediastinum in the thymus location (A1, A2). Axial CT (B) and MRI (C) scans of the chest presenting mediastinal mass corresponding to the invasive thymic malignancy.

The patient was qualified for a thoracic surgery; however, due to the local invasion, it was only possible to perform a partial thymectomy (August 2019). After the procedure, the ACTH concentration dropped but not significantly (537.3 pg/mL before and 446.0 pg/mL after the surgery). A histopathology report revealed thymic LCNEC with AC component extensively infiltrating the surgical margins. On IHC, the tumor stained positive for CgA, synaptophysin, and CD56, and weakly for ACTH; the Ki-67 index was 40%, p53 expression was 70% (Figure 3). The final diagnosis was TNM stage IVB (pT2NxM1b), Masaoka–Koga stage III ACTH-secreting combined thymic LCNC and AC.

Figure 3

www.frontiersin.org

Figure 3 Large-cell neuroendocrine carcinoma of the thymus with atypical carcinoid morphology, pathological diagnosis. (A) Microscopic image of the primary tumor, HE staining. (B) Weak positive IHC staining for ACTH. (C) Microscopic image of the brain metastasis, HE staining. (D) Weak positive IHC staining for ACTH of the brain metastasis. (E) Electron microscopic image of cancer cells; the material was taken from a paraffin block, which resulted in a poorly preserved ultrastructure. Visible neuroendocrine granules concentrated around the cell nucleus (×17,500). (F) Ultrastructural image of the neuroendocrine granules with a diameter of (×24,500).

As part of the cancer multidisciplinary team meeting, the patient was qualified for postoperative radiotherapy (RTH) for the thymus bed and residual mass of the mediastinal tumor. Between October and December 2019, the patient received 64 Gy/t in 2 Gy/t fraction doses. Unfortunately, the ACTH concentration increased during the RTH to the maximum observed value of 1,021.2 pg/mL, which suggested further progression of the disease. Follow-up 18F-FDG-PET-CT (January 2020) showed a thymic mass measuring 31 mm × 20 mm, less metabolically active (SUVmax, 4.3) than initially. It also revealed a new 18F-FDG-avid lesion in the pancreas tail in the left temporal lobe (SUVmax, 12.7). The brain MRI confirmed the presence of metastasis measuring 30 × 24 × 18 mm. The patient was qualified for postoperative chemotherapy (CTH) according to the ADOC regimen (cisplatin, doxorubicin, vincristine, and cyclophosphamide) for aggressive thymic tumors (January 2020–May 2020). CTH resulted in disease partial response, and ACTH concentration dropped to 192.0 pg/mL. In June 2020, the patient underwent a craniotomy with left non-radical temporal tumor resection and received additional RTH. The disease was stable for almost 10 months; however, the follow-up 18F-FDG-PET-CT on May 2021 showed new active lesions in the right lung, pancreas, left iliopsoas muscle, and left breast. ACTH level at that time increased to 655.5 pg/mL. The patient was introduced to the PE regimen CTH (cisplatin and etoposide, June 2021–September 2021) with a short-term partial response. Because of the further disease progression, the patient was started on palliative CTH. More than 3.5 years after the first hospitalization, in November 2022, the patient passed away.

3 Discussion

In this paper, we present a unique case of a patient with ACTH-secreting combined thymic LCNEC with AC component, primary manifested as severe CS. NETTs constitute approximately 2%–5% of thymic tumors (79), representing approximately 2% of all mediastinal tumors (8). NETTS are typically diagnosed with a mean age of 55, with a clear male predominance (male-to-female ratio, 3:1) (7). EAS with ACTH secretion occurs in up to 50% of hormonally active NETTS (8). NETTs associated with EAS appear in younger populations below age of 40 compared to overall NETTs and are also more prevalent in male individuals (albeit in a lower proportion when compared to non-EAS NETTs) (9). EAS-related NETTs have a worse outcome than biochemically inactive thymic tumors, since they usually have an aggressive course, with early regional invasion, distant metastasis, and high mortality (811). Patients with EAS typically present with rapid-onset, severe CS, including resistant hypertension, hyperglycemia, profound and refractory hypokalemia with metabolic alkalosis, generalized edema, and proximal muscle weakness (46). SH, which occurs in approximately 60% of patients with hormonally active NETTs secreting ACTH (and approximately 80% in the case of ACTH-secreting thymic carcinomas), significantly worsens the initial poor prognosis (9). In the systematic review by Guerrero Pérez et al., mortality in patients with advanced disease was approximately 55%, and the median time between diagnosis and death was 38 months (9). NETTs are typically large tumors that could manifest with neoplastic mass effect (1112); however, only up to 10% of patients with ACTH-secreting NETTs present with local compressive symptoms (9).

In EAS, the progression of hypercortisolemia is typically accelerated, and patients with very rapid SH onset may not present with typical cushingoid features. During the initial assessment at the Emergency Department, the patient was overlooked—the physician’s attention was captured by diabetes mellitus, but the clinical features have not been linked to CS. It highlights the need for a high CS clinical suspicion in case of SH. The presence of profound hypokalemia in combination with hyperglycemia and resistant hypertension with edema is a clue that should prompt diagnosis (45).

In the Endocrinology Department, the patient was quickly suspected of aggressive CS. Serum cortisol, UFC, and ACTH were dramatically increased. The patient was started on etomidate infusion to control hypercortisolemia. Etomidate is considered the most potent and effective agent for rapidly inhibiting cortisol overproduction (1314). Indeed, the patient’s clinical condition notably improved after only a few days of etomidate therapy with edema reduction and normalization of blood pressure, glycemia, and kalemia.

In the presented patient, the severity of hypercortisolemia with negative pituitary MRI image and positive whole-body CT imaging were compatible with EAS. Whole-body CT and functional imaging highly suggested a generalized malignancy with a potential origin in the left adrenal gland. The possibility of metastatic pheochromocytoma was considered—the concentration of CgA was significantly elevated, but urine and plasma metanephrines were negative. The ACC was also taken into consideration. Nevertheless, only one case of ACC potentially related to EAS was reported (15).

However, it is puzzling how only one of the adrenal glands was ACTH overstimulated, and the function of the second one seemed to be inhibited. A significant decline in cortisol concentration after exclusive unilateral adrenalectomy indicated that the left adrenal gland tumor could indeed have been the primary origin of malignancy and CS itself. On the other hand, ACTH level remained significantly elevated, although it dropped more than twofold compared to baseline. It was considered that the persistently elevated (but markedly lower) ACTH concentration was associated with the presence of metastases or could (albeit partially) result from the pituitary response to a significant decrease in cortisol concentration. Of course, metastatic ectopic ACTH-secreting tumor of unknown origin was also considered at that time.

Surprisingly, the histopathology examination revealed an adrenal adenoma with the domination of necrosis due to the adrenal infarction. It cannot be ruled out that the patient had a previously undiagnosed adrenal adenoma, and even short-term but dramatic ACTH hyperstimulation led it to its significant growth and provoked an adrenal infarction, imitating a malignancy in the imagery evaluation. Differentiating benign and malignant adrenal lesions based on 18F-FDG-PET-CT has a high diagnostic accuracy (1618); however, metabolically active adenomas may present with increased FDG uptake and mimic malignancy (19). In addition, adrenal hemorrhage and necrosis can present with increased activity on 18F-FDG-PET-CT (20). Thus, it seems that the increased 18F-FDG avidity of the left adrenal lesion with foci of intratumoral necrosis was directly related to dramatically elevated ACTH concentrations and adrenal overstimulation.

CT scan located the EAS-related NETTs in 97.8% of cases in the aforementioned systematic review by Guerrero Pérez et al. (9). In the presented patient, imagery diagnostic initially failed to visualize the thymic tumor. There are only few reports on non-diagnostic chest CT or MRI evaluation in patients with EAS NETTs (2123); however, in the presented cases, NETTs were found on SRS, which is contrary to our report. The first 18F-FDG-PET-CT localized an area of 2.1 SUVmax in the anterior mediastinum, initially considered a residual thymus because of the physiological FDG uptake. It is also unique how the PET-CT scan, chest CT, and MRI revealed a highly 18F-FDG-avid, large, invasive mediastinal mass just a few weeks after the baseline assessment. Imagery and functional studies were additionally retrospectively assessed by independent radiologists and nuclear medics to exclude a possible oversight during the initial analysis. However, it was maintained that there was no clear evidence of a thymic neoplasm at baseline.

Thymic carcinomas present with high FDG uptake, typically with SUVmax > 7 (24), SUV max values <4 as being most consistent with benign thymic processes (25). On the other hand, there is a marked overlap in FDG uptake between physiological thymic FDG uptake and thymic neoplasia in the literature, indicating that 18F-FDG-PET-CT has a limited ability to assess the thymus and an equivocal role in the differentiation of a normal thymus from thymic neoplasia (26). Among all described cases of EAS-related NETTs in the literature in which 18F-FDG-PET-CT was indicated as one of the diagnostic step, the primary tumor was visualized in all of them (2738). The primary NETT SUVmax was reported only in five of them ranging from 2.48 to 12.0 (2731); in the remaining ones, 18F-FDG avidity was reported from mild to high. However, in all mentioned cases (besides one (31), where no information about radiological chest imaging was reported), the NETT was previously visualized on chest CT, and 18F-FDG-PET-CT was performed to assess the disease staging rather than to find EAS origin.

On IHC, the thymic tumor stained weekly positive for ACTH in contrast to dramatically elevated plasma ACTH concentration. There seems to be a negative correlation between ACTH immunoreactivity and the neuroendocrine tumor malignancy potential. Moreover, the diagnosis of EAS is not ruled out in the case of primary tumor negative ACTH IHC staining. Less differentiated neuroendocrine tumors are believed to secrete ACTH rapidly and might also lose the ability to store ACTH in the secretory granules, thus leaving for typical techniques insufficient ACTH amounts stored to be stained (3940). In addition, the tumor might secrete various biologically active ACTH precursors that are negative on IHC.

Complete NETT resection is the only curative option and the strongest factor for overall survival (74142). In the case of a subtotal resected tumor, RTH and CTH are considered (4244); however, there is no consensus and guidelines for the optimal postoperative strategy, mainly due to the rarity of the disease. Systematic therapies are also used as palliative treatment in case of unresectable, metastatic, and recurrent NETTs (4244). Besides the primary metastatic disease, the presented patient underwent a partial resection with macroscopic residual tumor (R2) and was then qualified for postoperative RTH and CTH. After the R2 resection, postoperative RTH may be combined sequentially or concurrently with CTH (45). RTH was not clearly effective, as the follow-up 18F-FDG-PET-CT showed new metastases, including an extensive metastasis to the left temporal lobe. Several CTH regimens have been used in patients with NETTs (424445). CTH response rates in metastatic poorly differentiated NETTs are 30%–50%, with progression-free survival rates of 6–9 months (45). The presented patient was introduced to an ADOC regimen, which resulted in a partial response. However, the patient started second-line and later palliative CTH because of the further disease progression.

4 Conclusions

We present a unique and challenging case of malignant, primary metastatic NETT initially manifesting with severe EAS and not visible in initial imaging studies in a patient with coexisting adrenal tumor suspected of malignancy and primary disease origin. The presented case highlights that the diagnosis and management of EAS remain challenging; it requires a high clinical suspicion, rapid hypercortisolemia control with symptomatic treatment of cortisol-induced comorbidities, and simultaneously complex imaging studies to determine the primary source of the ACTH hypersecretion. The treatment of choice is resection of ACTH-secreting NET; however, it may not be possible in patients with initially occult or metastatic disease. Malignant NETTs with ectopic CS are extremely rare, and their management has to be individualized in every case, requiring a multidisciplinary approach. Regardless, the prognosis remains poor due to the aggressiveness of the disease.

Data availability statement

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.

Ethics statement

The studies were conducted in accordance with the local legislation and institutional requirements. The patient gave an oral consent for publication while alive. Written informed consent for publication was obtained from the patient’s parents.

Author contributions

LD: Resources, Writing – review & editing, Writing – original draft, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. AW-L: Writing – review & editing, Supervision, Resources, Methodology, Data curation, Conceptualization. MM: Writing – review & editing, Resources, Data curation, Conceptualization. PW: Writing – review & editing, Supervision, Resources, Methodology, Data curation.

Funding

The author(s) declare financial support was received for the research, authorship, and/or publication of this article. The publication fee was covered by the Medical University of Warsaw.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

1. Lacroix A, Feelders RA, Stratakis CA, Nieman LK. Cushing’s syndrome. Lancet. (2015) 386:913–27. doi: 10.1016/S0140-6736(14)61375-1

PubMed Abstract | CrossRef Full Text | Google Scholar

2. Feelders R, Sharma S, Nieman L. Cushing’s syndrome: epidemiology and developments in disease management. CLEP. (2015) 281:281–93. doi: 10.2147/CLEP.S44336

CrossRef Full Text | Google Scholar

3. Dimitriadis GK, Angelousi A, Weickert MO, Randeva HS, Kaltsas G, Grossman A. Paraneoplastic endocrine syndromes. Endocrine-Related Cancer. (2017) 24:R173–90. doi: 10.1530/ERC-17-0036

PubMed Abstract | CrossRef Full Text | Google Scholar

4. Witek P, Witek J, Zieliński G, Podgajny Z, Kamiński G. Ectopic Cushing’s syndrome in light of modern diagnostic techniques and treatment options. Neuro Endocrinol Lett. (2015) 36:201–8.

PubMed Abstract | Google Scholar

5. Young J, Haissaguerre M, Viera-Pinto O, Chabre O, Baudin E, Tabarin A. MANAGEMENT OF ENDOCRINE DISEASE: Cushing’s syndrome due to ectopic ACTH secretion: an expert operational opinion. Eur J Endocrinol. (2020) 182:R29–58. doi: 10.1530/EJE-19-0877

PubMed Abstract | CrossRef Full Text | Google Scholar

6. Paleń-Tytko JE, Przybylik-Mazurek EM, Rzepka EJ, Pach DM, Sowa-Staszczak AS, Gilis-Januszewska A, et al. Ectopic ACTH syndrome of different origin—Diagnostic approach and clinical outcome. Exp One Clin Centre PloS One. (2020) 15:e0242679. doi: 10.1371/journal.pone.0242679

CrossRef Full Text | Google Scholar

7. Gaur P, Leary C, Yao JC. Thymic neuroendocrine tumors: A SEER database analysis of 160 patients. Ann Surg. (2010) 251:1117–21. doi: 10.1097/SLA.0b013e3181dd4ec4

PubMed Abstract | CrossRef Full Text | Google Scholar

8. Jia R, Sulentic P, Xu J-M, Grossman AB. Thymic neuroendocrine neoplasms: biological behaviour and therapy. Neuroendocrinology. (2017) 105:105–14. doi: 10.1159/000472255

PubMed Abstract | CrossRef Full Text | Google Scholar

9. Guerrero-Pérez F, Peiró I, Marengo AP, Teulé A, Ruffinelli JC, Llatjos R, et al. Ectopic Cushing’s syndrome due to thymic neuroendocrine tumours: a systematic review. Rev Endocr Metab Disord. (2021) 22:1041–56. doi: 10.1007/s11154-021-09660-2

PubMed Abstract | CrossRef Full Text | Google Scholar

10. Neary NM, Lopez-Chavez A, Abel BS, Boyce AM, Schaub N, Kwong K, et al. Neuroendocrine ACTH-producing tumor of the thymus—Experience with 12 patients over 25 years. J Clin Endocrinol Metab. (2012) 97:2223–30. doi: 10.1210/jc.2011-3355

PubMed Abstract | CrossRef Full Text | Google Scholar

11. Filosso PL, Yao X, Ahmad U, Zhan Y, Huang J, Ruffini E, et al. Outcome of primary neuroendocrine tumors of the thymus: A joint analysis of the International Thymic Malignancy Interest Group and the European Society of Thoracic Surgeons databases. J Thorac Cardiovasc Surg. (2015) 149:103–109.e2. doi: 10.1016/j.jtcvs.2014.08.061

PubMed Abstract | CrossRef Full Text | Google Scholar

12. Lausi PO, Refai M, Filosso PL, Ruffini E, Oliaro A, Guerrera F, et al. Thymic neuroendocrine tumors. Thorac Surg Clinics. (2014) 24:327–32. doi: 10.1016/j.thorsurg.2014.05.007

CrossRef Full Text | Google Scholar

13. Preda VA, Sen J, Karavitaki N, Grossman AB. THERAPY IN ENDOCRINE DISEASE: Etomidate in the management of hypercortisolaemia in Cushing’s syndrome: a review. Eur J Endocrinol. (2012) 167:137–43. doi: 10.1530/EJE-12-0274

PubMed Abstract | CrossRef Full Text | Google Scholar

14. Varlamov EV, Han AJ, Fleseriu M. Updates in adrenal steroidogenesis inhibitors for Cushing’s syndrome – A practical guide. Best Pract Res Clin Endocrinol Metab. (2021) 35:101490. doi: 10.1016/j.beem.2021.101490

PubMed Abstract | CrossRef Full Text | Google Scholar

15. Dilrukshi MDSA, Wickramarachchi AW, Abeyaratne DDK, Shine B, Jafar-Mohammadi B, Somasundaram NP. An adrenocortical carcinoma associated with non-islet cell tumor hypoglycemia and aberrant ACTH production. Case Rep Endocrinol. (2020) 2020:1–6. doi: 10.1155/2020/2025631

CrossRef Full Text | Google Scholar

16. Boland GWL, Dwamena BA, Jagtiani Sangwaiya M, Goehler AG, Blake MA, Hahn PF, et al. Characterization of adrenal masses by using FDG PET: A systematic review and meta-analysis of diagnostic test performance. Radiology. (2011) 259:117–26. doi: 10.1148/radiol.11100569

PubMed Abstract | CrossRef Full Text | Google Scholar

17. Kim S-J, Lee S-W, Pak K, Kim I-J, Kim K. Diagnostic accuracy of 18 F-FDG PET or PET/CT for the characterization of adrenal masses: a systematic review and meta-analysis. BJR. (2018) 91(1086):20170520. doi: 10.1259/bjr.20170520

PubMed Abstract | CrossRef Full Text | Google Scholar

18. Schaafsma M, Berends AMA, Links TP, Brouwers AH, Kerstens MN. The diagnostic value of 18F-FDG PET/CT scan in characterizing adrenal tumors. J Clin Endocrinol Metab. (2023) 108:2435–45. doi: 10.1210/clinem/dgad138

PubMed Abstract | CrossRef Full Text | Google Scholar

19. Akkuş G, Güney IB, Ok F, Evran M, Izol V, Erdoğan Ş, et al. Diagnostic efficacy of 18F-FDG PET/CT in patients with adrenal incidentaloma. Endocrine Connections. (2019) 8:838–45. doi: 10.1530/EC-19-0204

PubMed Abstract | CrossRef Full Text | Google Scholar

20. Dong A, Cui Y, Wang Y, Zuo C, Bai Y. 18 F-FDG PET/CT of adrenal lesions. Am J Roentgenol. (2014) 203:245–52. doi: 10.2214/AJR.13.11793

CrossRef Full Text | Google Scholar

21. Silva F, Vázquez-Sellés J, Aguilö F, Vázquez G, Flores C. Recurrent ectopic adrenocorticotropic hormone producing thymic carcinoid detected with octreotide imaging. Clin Nucl Med. (1999) 24:109–10. doi: 10.1097/00003072-199902000-00007

PubMed Abstract | CrossRef Full Text | Google Scholar

22. Blumenthal EZ, Muszkat M, Pe’er J, Ticho U. Corticosteroid-induced glaucoma attributable to an adrenocorticotropin-secreting Malignant carcinoid tumor of the thymus. Am J Ophthalmol. (1999) 128:100–1. doi: 10.1016/S0002-9394(99)00052-5

PubMed Abstract | CrossRef Full Text | Google Scholar

23. Sankri-Tarbichi AG, Saydain G. REDUCED LUNG VOLUMES IN A RARE CASE OF CUSHING SYNDROME DUE TO THYMIC CARCINOID OF THE LUNG. Chest. (2006) 130:324S. doi: 10.1378/chest.130.4_MeetingAbstracts.324S-c

CrossRef Full Text | Google Scholar

24. Sharma P, Singhal A, Kumar A, Bal C, Malhotra A, Kumar R. Evaluation of thymic tumors with 18F-FDG PET-CT: A pictorial review. Acta Radiol. (2013) 54:14–21. doi: 10.1258/ar.2012.120536

PubMed Abstract | CrossRef Full Text | Google Scholar

25. Jerushalmi J, Frenkel A, Bar-Shalom R, Khoury J, Israel O. Physiologic thymic uptake of 18 F-FDG in children and young adults: A PET/CT evaluation of incidence, patterns, and relationship to treatment. J Nucl Med. (2009) 50:849–53. doi: 10.2967/jnumed.108.058586

PubMed Abstract | CrossRef Full Text | Google Scholar

26. Lococo F, Chiappetta M, Triumbari EKA, Evangelista J, Congedo MT, Pizzuto DA, et al. Current roles of PET/CT in thymic epithelial tumours: which evidences and which prospects? A pictorial review. Cancers. (2021) 13:6091. doi: 10.3390/cancers13236091

PubMed Abstract | CrossRef Full Text | Google Scholar

27. Sekiguchi Y, Miyamoto Y, Kasahara I, Hara Y, Tani Y, Doi M, et al. Ectopic ACTH syndrome caused by desmopressin-responsive thymic neuroendocrine tumor. Endocr J. (2015) 62:441–7. doi: 10.1507/endocrj.EJ14-0455

PubMed Abstract | CrossRef Full Text | Google Scholar

28. Duran C. Ectopic cushing’s syndrome due to ACTH secreting atypic thymic carcinoid tumor. North Clin Istanbul. (2018) 6(1):85–8. doi: 10.14744/nci.2018.53244

CrossRef Full Text | Google Scholar

29. Fujiwara W, Haruki T, Kidokoro Y, Ohno T, Yurugi Y, Miwa K, et al. Cushing’s syndrome caused by ACTH-producing thymic typical carcinoid with local invasion and regional lymph node metastasis: a case report. Surg Case Rep. (2018) 4:55. doi: 10.1186/s40792-018-0459-7

PubMed Abstract | CrossRef Full Text | Google Scholar

30. Li APZ, Sathyanarayan S, Diaz-Cano S, Arshad S, Drakou EE, Vincent RP, et al. Multiple electrolyte disturbances as the presenting feature of multiple endocrine neoplasia type 1 (MEN-1). Endocrinol Diabetes Metab Case Rep. (2022) 2022:21–0207. doi: 10.1530/EDM-21-0207

PubMed Abstract | CrossRef Full Text | Google Scholar

31. Xu H, Zhang M, Zhai G, Zhang M, Ning G, Li B. The role of integrated 18F-FDG PET/CT in identification of ectopic ACTH secretion tumors. Endocr. (2009) 36:385–91. doi: 10.1007/s12020-009-9247-2

CrossRef Full Text | Google Scholar

32. Chang L, Chen K, Cheng M, Lin C, Wang H, Sung C, et al. Endoscopic ultrasound ablation in a patient with multiple metastatic pancreatic tumors from adrenocorticotropic hormone-producing thymic neuroendocrine neoplasm. Digestive Endoscopy. (2021) 33:458–63. doi: 10.1111/den.13752

PubMed Abstract | CrossRef Full Text | Google Scholar

33. Baranowska-Jurkun A, Szychlińska M, Matuszewski W, Modzelewski R, Bandurska-Stankiewicz E. ACTH-dependent hypercortisolemia in a patient with a pituitary microadenoma and an atypical carcinoid tumour of the thymus. Medicina. (2019) 55:759. doi: 10.3390/medicina55120759

PubMed Abstract | CrossRef Full Text | Google Scholar

34. Lawrence L, Zhang P, Choi H, Ahmad U, Arrossi V, Purysko A, et al. A unique case of ectopic Cushing’s syndrome from a thymic neuroendocrine carcinoma. Endocrinol Diabetes Metab Case Rep. (2019) 2019:EDM190002. doi: 10.1530/EDM-19-0002

PubMed Abstract | CrossRef Full Text | Google Scholar

35. Araujo Castro M, Palacios García N, Aller Pardo J, Izquierdo Alvarez C, Armengod Grao L, Estrada García J. Ectopic Cushing syndrome: Report of 9 cases. Endocrinología Diabetes y Nutrición (English ed). (2018) 65:255–64. doi: 10.1016/j.endien.2018.05.006

CrossRef Full Text | Google Scholar

36. Pallais JC, Fenves AZ, Lu MT, Glomski K. Case 18-2018: A 45-year-old woman with hypertension, fatigue, and altered mental status. N Engl J Med. (2018) 378:2322–33. doi: 10.1056/NEJMcpc1802825

PubMed Abstract | CrossRef Full Text | Google Scholar

37. Kyriacou A, Stepien K, Issa B. Urinary steroid metabolites in a case of florid Ectopic Cushing’s syndrome and clinical correlations. Hormones (2016) 15(4):540–7. doi: 10.14310/horm.2002.1695

PubMed Abstract | CrossRef Full Text | Google Scholar

38. Wong TW. Long-term treatment with metyrapone in a man with ectopic cushing syndrome. JCEM Case Rep. (2022) 1:luac008. doi: 10.1210/jcemcr/luac008

PubMed Abstract | CrossRef Full Text | Google Scholar

39. Alexandraki KI, Grossman AB. The ectopic ACTH syndrome. Rev Endocr Metab Disord. (2010) 11:117–26. doi: 10.1007/s11154-010-9139-z

PubMed Abstract | CrossRef Full Text | Google Scholar

40. Takeuchi M, Sato J, Manaka K, Tanaka M, Matsui H, Sato Y, et al. Molecular analysis and literature-based hypothesis of an immunonegative prostate small cell carcinoma causing ectopic ACTH syndrome. Endocr J. (2019) 66:547–54. doi: 10.1507/endocrj.EJ18-0563

PubMed Abstract | CrossRef Full Text | Google Scholar

41. Crona J, Bjrklund P, Welin S, Kozlovacki G, berg K, Granberg D. Treatment, prognostic markers and survival in thymic neuroendocrine tumours. A study from a single tertiary referral centre. Lung Cancer. (2013) 79:289–93. doi: 10.1016/j.lungcan.2012.12.001

PubMed Abstract | CrossRef Full Text | Google Scholar

42. Girard N. Neuroendocrine tumors of the thymus: the oncologist point of view. J Thorac Dis. (2017) 9:S1491–500. doi: 10.21037/jtd.2017.08.18

PubMed Abstract | CrossRef Full Text | Google Scholar

43. Ruffini E, Venuta F. Management of thymic tumors: a European perspective. J Thorac Dis. (2014) 6:S228–237. doi: 10.3978/j.issn.2072-1439.2014.04.19

PubMed Abstract | CrossRef Full Text | Google Scholar

44. Zucali PA, De Vincenzo F, Perrino M, Digiacomo N, Cordua N, D’Antonio F, et al. Systemic treatments for thymic tumors: a narrative review. Mediastinum. (2021) 5:24–4. doi: 10.21037/med-21-11

PubMed Abstract | CrossRef Full Text | Google Scholar

45. Girard N, Lal R, Wakelee H, Riely GJ, Loehrer PJ. Chemotherapy definitions and policies for thymic Malignancies. J Thorac Oncol. (2011) 6:S1749–55. doi: 10.1097/JTO.0b013e31821ea5f7

PubMed Abstract | CrossRef Full Text | Google Scholar

Keywords: adrenal tumor, cortisol, ectopic ACTH syndrome, etomidate, thymic neuroendocrine carcinoma

Citation: Dzialach L, Wojciechowska-Luzniak A, Maksymowicz M and Witek P (2024) Case report: A challenging case of severe Cushing’s syndrome in the course of metastatic thymic neuroendocrine carcinoma with a synchronous adrenal tumor. Front. Endocrinol. 15:1399930. doi: 10.3389/fendo.2024.1399930

Received: 12 March 2024; Accepted: 24 May 2024;
Published: 14 June 2024.

Edited by:

Vincent Geenen, University of Liège, Belgium

Reviewed by:

Mara Carsote, Carol Davila University of Medicine and Pharmacy, Romania
Aleksandra Gilis-Januszewska, Jagiellonian University Medical College, Poland

Copyright © 2024 Dzialach, Wojciechowska-Luzniak, Maksymowicz and Witek. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

*Correspondence: Lukasz Dzialach, lukasz.dzialach@wum.edu.pl

Evaluating the usefulness of plasma chromogranin A measurement in cyclic ACTH-dependent Cushing’s syndrome

Abstract

Cushing’s syndrome, a clinical condition characterized by hypercortisolemia, exhibits distinct clinical signs and is associated with cyclic cortisol secretion in some patients. The clinical presentation of cyclic Cushing’s syndrome can be ambiguous and its diagnosis is often challenging.

We experienced a 72-year-old woman with cyclic ACTH-dependent Cushing’s syndrome caused by a pulmonary carcinoid tumor. Diagnosis was challenging because of the extended trough periods, and the responsible lesion was initially unidentified. A subsequent follow-up computed tomography revealed a pulmonary lesion, and ectopic ACTH secretion from this lesion was confirmed by pulmonary artery sampling. Despite the short peak secretion period of ACTH (approximately one week), immunostaining of the surgically removed tumor confirmed ACTH positivity. Interestingly, stored plasma chromogranin A levels were elevated during both peak and trough periods.

The experience in evaluating this patient prompted us to investigate the potential use of plasma chromogranin A as a diagnostic marker of ACTH-dependent Cushing’s syndrome. A retrospective study was conducted to determine the efficacy of plasma chromogranin A in three patients with ectopic ACTH syndrome (EAS), including the present case, and six patients with Cushing’s disease (CD) who visited our hospital between 2018 and 2021. Notably, plasma chromogranin A levels were higher in patients with EAS than in those with CD. Additionally, a chromogranin A level in the present case during the trough phase was lower than that in the peak phase, and was similar to those in CD patients. The measurement of plasma chromogranin A levels could aid in differentiating EAS from CD.

Keywords: ACTH-dependent Cushing’s syndromeCyclicCarcinoidPulmonary arterial samplingChromogranin A

From https://www.jstage.jst.go.jp/article/endocrj/advpub/0/advpub_EJ24-0128/_article

Unveiling the Uncommon: Cushing’s Syndrome (CS) Masquerading as Severe Hypokalemia

Abstract

Cushing’s syndrome (CS) arises from an excess of endogenous or exogenous cortisol, with Cushing’s disease specifically implicating a pituitary adenoma and exaggerated adrenocorticotropic hormone (ACTH) production. Typically, Cushing’s disease presents with characteristic symptoms such as weight gain, central obesity, moon face, and buffalo hump.

This case report presents an unusual manifestation of CS in a 48-year-old male with a history of hypertension, where severe hypokalemia was the primary presentation. Initial complaints included bilateral leg swelling, muscle weakness, occasional shortness of breath, and a general feeling of not feeling well. Subsequent investigations revealed hypokalemia, metabolic alkalosis, and an abnormal response to dexamethasone suppression, raising concerns about hypercortisolism. Further tests, including 24-hour urinary free cortisol and ACTH testing, confirmed significant elevations. Brain magnetic resonance imaging (MRI) identified a pituitary macroadenoma, necessitating neurosurgical intervention.

This case underscores the rarity of CS presenting with severe hypokalemia, highlighting the diagnostic challenges and the crucial role of a collaborative approach in managing such intricate cases.

Introduction

Cushing’s syndrome (CS), characterized by excessive cortisol production, is well-known for its diverse and often conspicuous clinical manifestations. Cushing’s disease is a subset of CS resulting from a pituitary adenoma overproducing adrenocorticotropic hormone (ACTH), leading to heightened cortisol secretion. The classic presentation involves a spectrum of symptoms such as weight gain, central obesity, muscle weakness, and mood alterations [1].

Despite its classic presentation, CS can demonstrate diverse and atypical features, challenging conventional diagnostic paradigms. This case report sheds light on a rare manifestation of CS, where severe hypokalemia was the primary clinical indicator. Notably, instances of CS prominently manifesting through severe hypokalemia are scarce in the literature [1,2].

Through this exploration, we aim to provide valuable insights into the diagnostic intricacies of atypical CS presentations, underscore the significance of a comprehensive workup, and emphasize the collaborative approach essential for managing such uncommon hormonal disorders.

Case Presentation

A 48-year-old male with a history of hypertension presented to his primary care physician with complaints of bilateral leg swelling, occasional shortness of breath, dizziness, and a general feeling of malaise persisting for 10 days. The patient reported increased water intake and urinary frequency without dysuria. The patient was diagnosed with hypertension eight months ago. He experienced progressive muscle weakness over two months, hindering his ability to perform daily activities, including using the bathroom. The primary care physician initiated a blood workup that revealed severe hypokalemia with a potassium level of 1.3 mmol/L (reference range: 3.6 to 5.2 mmol/L), prompting referral to the hospital.

Upon admission, the patient was hypertensive with a blood pressure of 180/103 mmHg, a heart rate of 71 beats/minute, a respiratory rate of 18 breaths/minute, and an oxygen saturation of 96% on room air. Physical examination revealed fine tremors, bilateral 2+ pitting edema in the lower extremities up to mid-shin, abdominal distension with normal bowel sounds, and bilateral reduced air entry in the bases of the lungs on auscultation. The blood work showed the following findings (Table 1).

Parameter Result Reference Range
Potassium (K) 1.8 mmol/L 3.5-5.0 mmol/L
Sodium (Na) 144 mmol/L 135-145 mmol/L
Magnesium (Mg) 1.3 mg/dL 1.7-2.2 mg/dL
Hemoglobin (Hb) 15.5 g/dL 13.8-17.2 g/dL
White blood cell count (WBC) 13,000 x 103/µL 4.5 to 11.0 × 109/L
Platelets 131,000 x 109/L 150-450 x 109/L
pH 7.57 7.35-7.45
Bicarbonate (HCO3) 46 mmol/L 22-26 mmol/L
Lactic acid 4.2 mmol/L 0.5-2.0 mmol/L
Table 1: Blood work findings

In order to correct the electrolyte imbalances, the patient received intravenous (IV) magnesium and potassium replacement and was later transitioned to oral. The patient was also started on normal saline at 100 cc per hour. To further investigate the complaint of shortness of breath, the patient underwent a chest X-ray, which revealed bilateral multilobar pneumonia (Figure 1). He was subsequently treated with ceftriaxone (1 g IV daily) and clarithromycin (500 mg twice daily) for seven days.

A-chest-X-ray-revealing-(arrows)-bilateral-multilobar-pneumonia
Figure 1: A chest X-ray revealing (arrows) bilateral multilobar pneumonia

With persistent abdominal pain and lactic acidosis, a computed tomography (CT) scan abdomen and pelvis with contrast was conducted, revealing a psoas muscle hematoma. Subsequent magnetic resonance imaging (MRI) depicted an 8×8 cm hematoma involving the left psoas and iliacus muscles. The interventional radiologist performed drainage of the hematoma involving the left psoas and iliacus muscles (Figure 2).

Magnetic-resonance-imaging-(MRI)-depicting-an-8x8-cm-hematoma-(arrow)-involving-the-left-psoas-and-iliacus-muscles
Figure 2: Magnetic resonance imaging (MRI) depicting an 8×8 cm hematoma (arrow) involving the left psoas and iliacus muscles

In light of the concurrent presence of hypokalemia, hypertension, and metabolic alkalosis, there arose concerns about Conn’s syndrome, prompting consultation with endocrinology. Their recommended workup for Conn’s syndrome included assessments of the aldosterone-renin ratio and random cortisol levels. The results unveiled an aldosterone level below 60 pmol/L (reference range: 190 to 830 pmol/L in SI units) and a plasma renin level of 0.2 pmol/L (reference range: 0.7 to 3.3 mcg/L/hr in SI units). Notably, the aldosterone-renin ratio was low, conclusively ruling out Conn’s syndrome. The random cortisol level was notably elevated at 1334 nmol/L (reference range: 140 to 690 nmol/L).

Furthermore, a low-dose dexamethasone suppression test was undertaken due to the high cortisol levels. Following the administration of 1 mg of dexamethasone at 10 p.m., cortisol levels were measured at 9 p.m., 3 a.m., and 9 a.m. the following day. The results unveiled a persistently elevated cortisol level surpassing 1655 nmol/L, signaling an abnormal response to dexamethasone suppression and raising concerns about a hypercortisolism disorder, such as CS.

In the intricate progression of this case, the investigation delved deeper with a 24-hour urinary free cortisol level, revealing a significant elevation at 521 mcg/day (reference range: 10 to 55 mcg/day). Subsequent testing of ACTH portrayed a markedly elevated level of 445 ng/L, distinctly exceeding the normal reference range of 7.2 to 63.3 ng/L. A high-dose 8 mg dexamethasone test was performed to ascertain the source of excess ACTH production. The baseline serum cortisol levels before the high-dose dexamethasone suppression test were 1404 nmol/L, which decreased to 612 nmol/L afterward, strongly suggesting the source of excess ACTH production to be in the pituitary gland.

A CT scan of the adrenal glands ruled out adrenal mass, while an MRI of the brain uncovered a 1.3×1.3×3.2 cm pituitary macroadenoma (Figure 3), leading to compression of adjacent structures. Neurosurgery was consulted, and they recommended surgical removal of the macroadenoma due to the tumor size and potential complications. The patient was referred to a tertiary care hospital for pituitary adenoma removal.

Magnetic-resonance-imaging-(MRI)-of-the-brain-depicting-a-1.3x1.3x3.2-cm-pituitary-macroadenoma-(star)
Figure 3: Magnetic resonance imaging (MRI) of the brain depicting a 1.3×1.3×3.2 cm pituitary macroadenoma (star)

Discussion

CS represents a complex endocrine disorder characterized by excessive cortisol production. While the classic presentation of CS includes weight gain, central obesity, and muscle weakness, our case highlights an uncommon initial manifestation: severe hypokalemia. This atypical presentation underscores the diverse clinical spectrum of CS and the challenges it poses in diagnosis and management [1,2].

While CS typically presents with the classic symptoms mentioned above, severe hypokalemia as the initial manifestation is exceedingly rare. Hypokalemia in CS often results from excess cortisol-mediated activation of mineralocorticoid receptors, leading to increased urinary potassium excretion and renal potassium wasting. Additionally, metabolic alkalosis secondary to cortisol excess further exacerbates hypokalemia [3,4].

Diagnosing a case of Cushing’s disease typically commences with a thorough examination of the patient’s medical history and a comprehensive physical assessment aimed at identifying characteristic manifestations such as central obesity, facial rounding, proximal muscle weakness, and increased susceptibility to bruising. Essential to confirming the diagnosis are laboratory examinations, which involve measuring cortisol levels through various tests, including 24-hour urinary free cortisol testing, late-night salivary cortisol testing, and dexamethasone suppression tests. Furthermore, assessing plasma ACTH levels aids in distinguishing between pituitary-dependent and non-pituitary causes of CS. Integral to the diagnostic process are imaging modalities such as MRI of the pituitary gland, which facilitate the visualization of adenomas and the determination of their size and precise location [1-4].

Treatment for Cushing’s disease primarily entails surgical removal of the pituitary adenoma via transsphenoidal surgery, with the aim of excising the tumor and restoring normal pituitary function. In cases where surgical intervention is unsuitable or unsuccessful, pharmacological therapies employing medications such as cabergoline (a dopamine receptor agonist) or pasireotide (a somatostatin analogue) may be considered to suppress ACTH secretion and regulate cortisol levels. Additionally, radiation therapy, whether conventional or stereotactic radiosurgery, serves as a supplementary or alternative treatment approach to reduce tumor dimensions and mitigate ACTH production [5,6]. To assess the effectiveness of treatment, manage any problem, and assure long-term illness remission, diligent long-term follow-up and monitoring are essential. Collaborative multidisciplinary care involving specialists such as endocrinologists, neurosurgeons, and other healthcare professionals is pivotal in optimizing patient outcomes and enhancing overall quality of life [2,4].

The prognosis of CS largely depends on the underlying cause, stage of the disease, and efficacy of treatment. Early recognition and prompt intervention are essential for improving outcomes and minimizing long-term complications. Surgical resection of the adrenal or pituitary tumor can lead to remission of CS in the majority of cases. However, recurrence rates vary depending on factors such as tumor size, invasiveness, and completeness of resection [2,3]. Long-term follow-up with endocrinologists is crucial for monitoring disease recurrence, assessing hormonal function, and managing comorbidities associated with CS.

Conclusions

In conclusion, our case report highlights the rarity of severe hypokalemia as the initial presentation of CS. This unique presentation underscores the diverse clinical manifestations of CS and emphasizes the diagnostic challenges encountered in clinical practice. A multidisciplinary approach involving endocrinologists, neurosurgeons, and radiologists is essential for the timely diagnosis and management of CS. Early recognition, prompt intervention, and long-term follow-up are essential for optimizing outcomes and improving the quality of life for patients with this endocrine disorder.

References

  1. Nieman LK, Biller BM, Findling JW, Newell-Price J, Savage MO, Stewart PM, Montori VM: The diagnosis of Cushing’s syndrome: an Endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab. 2008, 93:1526-40. 10.1210/jc.2008-0125
  2. Newell-Price J, Bertagna X, Grossman AB, Nieman LK: Cushing’s syndrome. Lancet. 2006, 367:1605-17. 10.1016/S0140-6736(06)68699-6
  3. Torpy DJ, Mullen N, Ilias I, Nieman LK: Association of hypertension and hypokalemia with Cushing’s syndrome caused by ectopic ACTH secretion: a series of 58 cases. Ann N Y Acad Sci. 2002, 970:134-44. 10.1111/j.1749-6632.2002.tb04419.x
  4. Elias C, Oliveira D, Silva MM, Lourenço P: Cushing’s syndrome behind hypokalemia and severe infection: a case report. Cureus. 2022, 14:e32486. 10.7759/cureus.32486
  5. Fleseriu M, Petersenn S: Medical therapy for Cushing’s disease: adrenal steroidogenesis inhibitors and glucocorticoid receptor blockers. Pituitary. 2015, 18:245-52. 10.1007/s11102-014-0627-0
  6. Pivonello R, De Leo M, Cozzolino A, Colao A: The treatment of Cushing’s disease. Endocr Rev. 2015, 36:385-486. 10.1210/er.2013-1048

Is Cushing Syndrome More Common in the US Than We Think?

I think members of the Cushing’s Help boards have been saying this forever!  Cushing’s isn’t all that rare.  Just rarely diagnosed,

 

BOSTON — The prevalence of Cushing syndrome (CS) in the United States may be considerably higher than currently appreciated, new data from a single US institution suggest.

In contrast to estimates of 1 to 3 cases per million patient-years from population-based European studies, researchers at the University of Wisconsin, Milwaukee, estimated that the incidence of CS in Wisconsin is a minimum of 7.2 cases per million patient-years. What’s more, contrary to all previous studies, they found that adrenal Cushing syndrome was more common than pituitary adrenocorticotropic hormone (ACTH)– secreting tumors (Cushing disease), and that fewer than half of individuals with adrenal Cushing syndrome had classic physical features of hypercortisolism, such as weight gain, round face, excessive hair growth, and stretch marks.

“Cases are absolutely being missed…. Clinicians should realize that cortisol excess is not rare. It may not be common, but it needs to be considered in patients with any constellation of features that are seen in cortisol excess,” study investigator Ty B. Carroll, MD, Associate Professor of Medicine, Endocrinology and Molecular Medicine, and the Endocrine Fellowship Program Director at Medical College of Wisconsin in Milwaukee, told Medscape Medical News.

There are several contributing factors, he noted, “including the obesity and diabetes epidemics which make some clinical features of cortisol excess more common and less notable. Providers get used to seeing patients with some features of cortisol excess and don’t think to screen. The consequence of this is more difficult-to-control diabetes and hypertension, more advance metabolic bone disease, and likely more advanced cardiovascular disease, all resulting from extended exposure to cortisol excess,” he said.

Are Milder Cases the Ones Being Missed?

Asked to comment, session moderator Sharon L. Wardlaw, MD, professor of medicine at Columbia University College of Physicians and Surgeons, New York City, said “When we talk about Cushing [syndrome], we usually think of pituitary ACTH as more [common], followed by adrenal adenomas, and then ectopic. But they’re seeing more adrenal adenoma…we are probably diagnosing this a little more now.”

She also suggested that the Wisconsin group may have a lower threshold for diagnosing the milder cortisol elevation seen with adrenal Cushing syndrome. “If you screen for Cushing with a dexamethasone suppression test…[i]f you have autonomous secretion by the adrenal, you don’t suppress as much…. When you measure 24-hour urinary cortisol, it may be normal. So you’re in this in-between [state]…. Maybe in Wisconsin they’re diagnosing it more. Or, maybe it’s just being underdiagnosed in other places.”

She also pointed out that “you can’t diagnose it unless you think of it. I’m not so sure that with these mild cases it’s so much that it’s more common, but maybe it’s like thyroid nodules, where we didn’t know about it until everybody started getting all of these CT scans. We’re now seeing all these incidental thyroid nodules…I don’t think we’re missing florid Cushing.”

However, Wardlaw said, it’s probably worthwhile to detect even milder hypercortisolism because it could still have long-term damaging effects, including osteoporosis, muscle weakness, glucose intolerance, and frailty. “You could do something about it and normalize it if you found it. I think that would be the reason to do it.”

Is Wisconsin Representative of Cushing Everywhere?

Carroll presented the findings at the annual meeting of the Endocrine Society. He began by noting that most of the previous CS incidence studies, with estimates of 1.2-3.2 cases per million per year, come from European data published from 1994 to 2019 and collected as far back as 1955. The method of acquisition of patients and the definitions of confirmed cases varied widely in those studies, which reported CS etiologies of ACTH-secreting neoplasms (pituitary or ectopic) in 75%-85% and adrenal-dependent cortisol excess in 15%-20%.

The current study included data from clinic records between May 1, 2017, and December 31, 2022, of Wisconsin residents newly diagnosed with and treated for CS. The CS diagnosis was established with standard guideline-supported biochemical testing and appropriate imaging. Patients with exogenous and non-neoplastic hypercortisolism and those who did not receive therapy for CS were excluded.

A total of 185 patients (73% female, 27% male) were identified from 27 of the total 72 counties in Wisconsin, representing a population of 4.5 million. On the basis of the total 5.9 million population of Wisconsin, the incidence of CS in the state works out to 7.2 cases per million population per year, Carroll said.

However, data from the Wisconsin Hospital Association show that the University of Wisconsin’s Milwaukee facility treated just about half of patients in the state who are discharged from the hospital with a diagnosis of CS during 2019-2023. “So…that means that an actual or approximate incidence of 14-15 cases per million per year rather than the 7.2 cases that we produce,” he said.

Etiologies were 60% adrenal (111 patients), 36.8% pituitary (68 patients), and 3.2% ectopic (6 patients). Those proportions were similar between genders.

On biochemical testing, values for late-night salivary cortisol, dexamethasone suppression, and urinary free cortisol were highest for the ectopic group (3.189 µg/dL, 42.5 µg/dL, and 1514.2 µg/24 h, respectively) and lowest for the adrenal group (0.236 µg/dL, 6.5 µg/dL, and 64.2 µg/24 h, respectively). All differences between groups were highly statistically significant, at P < .0001, Carroll noted.

Classic physical features of CS were present in 91% of people with pituitary CS and 100% of those ectopic CS but just 44% of individuals with adrenal CS. “We found that adrenal-dependent disease was the most common form of Cushing syndrome. It frequently presented without classic physical features that may be due to the milder biochemical presentation,” he concluded.

Carroll reports consulting and investigator fees from Corcept Therapeutics. Wardlaw has no disclosures. 

Miriam E. Tucker is a freelance journalist based in the Washington DC area. She is a regular contributor to Medscape, with other work appearing in The Washington Post, NPR’s Shots blog, and Diatribe. She is on X (formerly Twitter) @MiriamETucker.

1

Credit

Lead image: Designer491/Dreamstime

Medscape Medical News © 2024 WebMD, LLCSend comments and news tips to news@medscape.net.

Cite this: Is Cushing Syndrome More Common in the US Than We Think? – Medscape – June 07, 2024.

Epigenetic Mechanisms Modulated by Glucocorticoids With a Focus on Cushing Syndrome

Abstract

In Cushing syndrome (CS), prolonged exposure to high cortisol levels results in a wide range of devastating effects causing multisystem morbidity. Despite the efficacy of treatment leading to disease remission and clinical improvement, hypercortisolism-induced complications may persist. Since glucocorticoids use the epigenetic machinery as a mechanism of action to modulate gene expression, the persistence of some comorbidities may be mediated by hypercortisolism-induced long-lasting epigenetic changes. Additionally, glucocorticoids influence microRNA expression, which is an important epigenetic regulator as it modulates gene expression without changing the DNA sequence. Evidence suggests that chronically elevated glucocorticoid levels may induce aberrant microRNA expression which may impact several cellular processes resulting in cardiometabolic disorders.

The present article reviews the evidence on epigenetic changes induced by (long-term) glucocorticoid exposure. Key aspects of some glucocorticoid-target genes and their implications in the context of CS are described. Lastly, the effects of epigenetic drugs influencing glucocorticoid effects are discussed for their ability to be potentially used as adjunctive therapy in CS.

In Cushing syndrome (CS), adrenocorticotropic hormone (ACTH) hypersecretion by a pituitary adenoma or an ectopic source, or autonomous cortisol hypersecretion by an adrenal tumor, induces chronic endogenous hypercortisolism with loss of the cortisol circadian rhythm (1). CS is more prevalent in women than men and frequently occurs in the fourth to sixth decades of life (2).

Glucocorticoids (GC) have extensive physiological actions and regulate up to 20% of the expressed genome, mainly related to the immune system, metabolic homeostasis, and cognition. Therefore, the prolonged exposure to high cortisol levels results in a wide range of devastating effects, including major changes in body composition (obesity, muscle atrophy, osteoporosis), neuropsychiatric disturbances (impaired cognition, depression, sleep disturbances), the metabolic syndrome (obesity, hypertension, insulin resistance, and dyslipidemia), hypercoagulability, and immune suppression (34). The consequences of hypercortisolism lead to compromised quality of life and increased mortality rate (5). The mortality rate in patients with CS is 4 times higher than the healthy control population (6). Risk factors such as obesity, diabetes, and hypertension contribute to the increased risk of myocardial infarction, stroke, and cardiac insufficiency. As a result, cardiovascular disease is the leading cause of the premature death in CS (5). Infectious disease is also an important cause of death in CS (5). Therefore, prompt treatment to control hypercortisolism is imperative to prevent complications and an increased mortality rate.

Despite the efficacy of treatment leading to disease remission, the clinical burden of CS improves, but does not completely revert, in every patient (7). Indeed, obesity, neuropsychiatric disturbances, hypertension, diabetes, and osteoporosis persist in a substantial number of biochemically cured patients. For instance, in a study involving 118 CS patients in remission for about 7.8 years (median), resolution of comorbidities such as diabetes occurred in only 36% of cases, hypertension in 23% of cases, and depression in 52% of the cases (8). It has been proposed that epigenetic changes as a consequence of hypercortisolism is a mechanism of the persistence of some comorbidities (9-12).

Epigenetics is a reversible process that modifies gene expression without any alterations in DNA sequence; frequently it is mediated by histone modification and DNA methylation together with microRNAs (13-15). GCs use the epigenetic machinery as a mechanism of action to regulate gene expression in physiological circumstances, such as metabolic actions and stress response. Its networks involve DNA and histone modifying enzymes, such as DNA methyltransferases (DNMTs), histone acetyltransferases (HATs), and histone deacetylases (HDACs) (16). (Fig. 1) The DNA methylation process catalyzed by DNMTs is usually associated with downregulation of gene expression (17). Histone modifications catalyzed by HAT enzymes induce gene transcription, while those by HDAC enzymes induce transcriptional repression (17). Drugs interfering with these enzymes (so-called epigenetic drugs) may affect the GC genomic actions confirming the interaction between GC and the epigenetic system (1819). Furthermore, GC can modulate HDAC and DNMT expression and activity (161920). Based on these data it might be speculated that in CS, epigenetic modifications induced by long-term GC exposure plays a role in the development of the disease-specific morbidity (910).

Figure 1.

Glucocorticoid (GC) and its epigenetic machinery. GC through its receptor interacts with DNA and histone modifying enzymes, such as DNA methyltransferases (DNMTs), histone acetyl transferases (HATs), and histone deacetylases (HDAC) to modulate gene expression.

In this review we provide an overview of epigenetic aspects of GC action in physiological conditions and in the context of CS. We start with a detailed characterization of how GC, using the epigenetic system, can change chromatin structure in order to activate or silence gene expression. (Fig. 2) Subsequently, we describe the role of epigenetic mechanisms in the regulation of expression of several GC-target genes related to CS. Finally, we present the current evidence of epigenetic changes caused by the long-term of GC exposure and the potential use of epidrugs influencing GC actions.

Figure 2.

Epigenetic mechanisms of the glucocorticoid action to regulate gene expression. The GR is located in cytoplasm in a multi-protein complex; after GC binding, GR dissociates from the multi-protein complex, crosses the nuclear membrane, dimerizes, and binds to the GRE of the target gene. One of the mechanisms of action of GC is through the recruitment of co-regulators together with epigenetic enzymes, such as HAT, to change the chromatin structure, resulting in activation of gene transcription. Also, GR decreases gene expression by tethering other transcriptional factors and recruiting HDAC2, causing histone deacetylation, which leads to a repressed chromatin. GC can cause hypomethylation through downregulation in the expression of DNMT1. Abbreviations: Ac, acetylation; DNMT1, DNA methyltransferase 1; GC, glucocorticoid; GR, glucocorticoid receptor; GRE, glucocorticoid responsive elements; HAT, histone acetyltransferase; HDAC, histone deacetylases; Me: methylation.

Search Strategy

A search of the PubMed database was conducted using the advanced search builder tool for articles in the English language on the following terms “glucocorticoids,” “glucocorticoid receptor,” “Cushing,” “hypercortisolism,” “epigenetic,” “DNA methylation,” “histone deacetylase,” “histone acetyltransferase,” “microRNA” “fkbp5,” “clock genes,” and “POMC.” Moreover, references were identified directly from the articles included in this manuscript. The articles were selected by the authors after being carefully analyzed regarding their importance and impact.

Epigenetic Aspects of Genomic Action of Glucocorticoids

GCs regulate gene expression positively or negatively. GC-responsive genes include genes encoding for proteins associated with inflammation, metabolic processes, blood pressure and fluid homeostasis, apoptosis, cell cycle progression, circadian rhythm, and intracellular signaling (21).

The GC actions are cell type–specific (22). For instance, in an in vitro study, the comparison of GC-expressed genes between 2 cell lines, corticotroph (AtT20) and mammary (3134) cell lines, showed a different set of GC-regulated genes, revealing the cell type–specific nature of GC effects (23). GC function depends on the accessibility of glucocorticoid receptor (GR)-binding sites in the DNA of the target tissue, which in turn is mostly established during cell differentiation. Therefore, different chromatin organization explains the distinct GR-binding sites among different tissues (222425). The chromatin accessibility is determined by histone modifications such as acetylation, methylation, phosphorylation, and/or DNA methylation, processes that are both dynamic and reversible (26).

Furthermore, gene expression is regulated in a GC-concentration-dependent manner which is tissue-specific. Only a few genes can be upregulated or downregulated at low concentrations of GC. For example, a dose of dexamethasone (Dex) as low as 0.5 nM selectively activated PER1 (period 1, transcription factor related to circadian rhythm) expression in lung cancer (A549) cells (2127). Additionally, continuous GC exposure or pulsed GC (cortisol fluctuation during circadian rhythm) may cause different responses with respect to gene expression (2628). For example, constant treatment with corticosterone induced higher levels of PER1 clock gene mRNA expression compared with pulsatile treatment, as demonstrated in an in vitro study using 3134 cell line (28).

The time course for gene expression in response to Dex is fast, with repression occurring slightly slower compared to activation. Half of activated and repressed genes are detected within, respectively, about 40 minutes and 53 minutes following Dex exposure (21).

In short, the transcriptional output in response to GC depends on cell type, as well as on the duration and intensity of GC exposure (21242627). GCs act as a transcriptional regulatory factor resulting in activating or repressing the expression of genes. The GC exerts its function through binding to corticosteroid receptors, specifically, the mineralocorticoid receptor and the GR, members of the nuclear receptor superfamily (2930).

Glucocorticoid Receptor

The GR is located in the cytoplasm in a chaperone complex which includes heat-shock proteins (70 and 90) and immunophilins (such as FK506 binding protein [FKBP5]). Cortisol diffuses across the cell membrane and binds with high affinity to the GR. The activated GR bound to GC dissociates of the multi-protein complex and is transferred to the nucleus, where it ultimately regulates gene expression (2631).

GR is a transcription factor encoded by nuclear receptor subfamily 3, group C member 1 (NR3C1) gene, located in chromosome 5, and consisting of 9 exons. It is composed of 3 major functional domains, namely a DNA binding domain (DBD), the C-terminal ligand-binding domain (LBD) and the N-terminal domain (NTB). The LBD recognizes and joins the GC. NTB contains an activation function-1 (AF1) which connects with co-regulators and the members of the general transcription machinery to activate target genes. The DBD comprises 2 zinc fingers motifs that are able to identify and bind to glucocorticoid responsive elements (GREs) (3233).

GRα is the most expressed and functionally active GR. GRβ is another isoform which is the result of an alternative splicing in exon 9 of the GR transcript. The difference between the 2 isoforms is the distinct ligand-binding domain in GRβ. This variance prevents the GRβ from binding to GC. In fact, the GRβ counteracts GRα function by interfering with its binding to a GRE in the target gene, and GRβ expression is associated with GC resistance (32). In addition, GRβ has its own transcriptional activity which is independent and distinct from GRα (34).

Another splice variant of human GR, GRγ, is associated with GC resistance in lung cell carcinoma and childhood acute lymphoblastic leukemia (3335). There is an additional amino acid (arginine) in the DBD of the GRγ that reduces, by about half, the capacity to activate or suppress the transcription of the target gene, as compared with GRα (32). One study identified GRγ in a small series of corticotroph adenomas (36).

Glucocorticoid Mechanism of Action

The GR-GC complex induces or represses gene expression directly by binding to DNA, indirectly by tethering other transcription factors or yet in a composite manner that consists in binding DNA in association with binding to other co-regulators (3537).

The GR has the ability to reorganize the chromatin structure to become more or less accessible to the transcriptional machinery. In the classical mechanism of direct induction of gene expression, the GR dimerizes and binds to a GRE in DNA. The receptor recruits co-regulators, such as CREB binding protein, which has intrinsic histone acetyltransferase (HAT) activity that modifies the chromatin structure from an inactive to an active state. This model, called transactivation, upregulates the expression of some genes related to glucose, protein, and fat metabolism. Gene repression, on the other hand, is accomplished by GR binding to a negative GRE (nGRE) leading to the formation of a chromatin remodeling complex composed by co-repressor factors, such as NCOR1 and SMRT, and histone deacetylases (HDACs), that ultimately turn chromatin less accessible and suppress gene transcription. The gene repression through direct binding events occurs less frequently when compared to gene induction (253538).

Another mechanism of GC action is through binding to other transcription factors (tethering). In case of switching off inflammatory genes, GR binds to transcriptional co-activator molecules, such as CREB binding protein with intrinsic HAT activity, and subsequently recruits HDAC2 to reverse histone acetylation, thus resulting in a suppression of the activated inflammatory gene (39). In the same model, GC interacts with other cofactors, such as the STAT family, to induce chromatin modifications resulting in increased gene expression (26).

Furthermore, the transcriptional dynamics of some genes follow a composite manner. In this model, GR, in conjunction with binding to GRE, also interacts with cofactors in order to enhance or reduce gene expression (35).

GCs can also modulate gene expression by influencing the transcription of epigenetic modifiers. An experimental study demonstrated that GC mediated the upregulation of HDAC2 in rats exposed to chronic stress, which in turn decreased the transcription of histone methyltransferase (Ehmt2) that ultimately upregulated the expression of Nedd4. Nedd4 is a ubiquitin ligase, expression of which has been related to cognitive impairment (40). Additionally, GC was found to interact with another epigenetic eraser, namely JMJD3, a histone demethylase, suppressing its transcription in endothelial cells treated with TNFα that led to decreased expression of other genes related to the blood-brain barrier (41).

GCs have the ability to induce (de)methylation changes in DNA, ultimately affecting gene expression. The DNA methylation process triggered by GC involves the family of DNA methyltransferases (DNMT) and ten-eleven translocation (TET) protein (2042-44). The DNMT, DNMT1, DNMT3A, and DNMT3B are able to transfer a methyl group to a cytosine residue in DNA, forming 5-methylcytosine (5mC), which negatively impacts gene expression. In contrast, TET protein chemically modifies the 5mC to form 5-hydroxymethylcytosine (5hmC), which ultimately leads to unmethylated cytosine, positively influencing gene expression (45).

Glucocorticoids mainly induce loss of methylation events rather than gain of methylation across the genome (1146). The DNA demethylation process can be either active or passive. The active mechanism is linked to the upregulation of TET enzyme expression that follows GC treatment, which was described in retinal and osteocyte cell line model studies (4243). The passive demethylation event involves the downregulation (Fig. 2) or dysfunction of DNMT1. DNMT1 is responsible for maintaining the methylation process in dividing cells (45). In case of GC exposure, GC can cause hypomethylation through downregulation in the expression of DNMT1, a process described in the AtT20 corticotroph tumor cell model, or through GC hindering DNMT activity, particularly DNMT1, as demonstrated in the retinal cell (RPE) line (204244).

Glucocorticoid-Induced Epigenetic Changes

There are several molecular mechanisms connecting GR activation and epigenetic modifications ultimately affecting gene expression (Fig. 2). As described above, GC uses epigenetic machinery, such as DNA and histone modifying enzymes, to restructure the chromatin in order to induce or silence gene transcription (1647).

In an in vitro study using murine AtT20 corticotroph tumor and neuronal cell lines, after chronic GC exposure followed by a recovery period in the absence of GC, the cells retained an “epigenetic memory” with persistence of loss of methylation content in FKBP5 gene but with no increased gene expression at baseline. The functionality of this “epigenetic memory” only became evident in a second exposure to GC, when the cells responded sharply with a more robust expression of FKBP5 gene compared to the cells without previous exposure to GC (44). Another in vitro study, using a human fetal hippocampal cell line, confirmed long-lasting DNA methylation changes induced by GC. The cells were treated for 10 days with dexamethasone, during the proliferative and cell differentiation phases of the cell line, followed by 20 days without any treatment. The second exposure to GC resulted in an enhanced gene expression of a subset of GC-target genes (48). Additionally, using an animal model subjected to chronic stress, a distinct gene expression profile was demonstrated in response to acute GC challenge compared to those without chronic stress history. The proposed mechanism was that chronic stress resulted in GC-induced enduring epigenetic changes in target genes, altering the responsiveness to a subsequent GC exposure (49).

In general, it seems that the majority of differential methylation regions (DMRs) induced by GC are loss of methylation rather than gain of methylation. In an experimental study, an association between hypomethylation and GC exposure was demonstrated in mice previously exposed to high levels of GC. Further analysis demonstrated that the genes linked with DMR were mostly related to metabolism, the immune system, and neurodevelopment (11).

Human studies have also shown that excess of cortisol can induce modifications in DNA methylation. DNA methylation data obtained from whole blood samples from patients with chronic obstructive pulmonary disease (COPD) treated with GC revealed DMR at specific CpG dinucleotides across the genome. These DMR were confirmed by pyrosequencing and annotated to genes, such as SCNN1A, encoding the α subunit of the epithelial sodium channel, GPR97, encoding G protein coupled receptor 97, and LRP3, encoding low-density lipoprotein receptor-related protein 3 (50). Furthermore, it has been proposed that the negative impact of chronic GC exposure on the immune system, which increases the risk of opportunistically infections, may be epigenetically mediated (51). In a clinical study, using whole blood samples, an analysis of genome-wide DNA methylation was performed on patients before and after exposure to GC (51). Long-term GC exposure disrupts, through a persistent modification of the cytosine methylation pattern, the mTORC1 pathway which affects CD4+ T cell biology (51).

Taken together, these data clearly show the interplay between GC signaling and methylation and histone modifications processes suggesting that GC interferes in the epigenetic landscape modulating gene expression. It is possible that most of these GC-induced epigenetic events are dynamic and temporary, while others may persist leading to long-lasting disorders. Further research to provide insight into what makes some events reversible is warranted.

Epigenetic Changes as a Consequence of Long-Term Glucocorticoid Exposure in Cushing Syndrome

The comorbidities associated with CS are associated with increased mortality mainly due to cardiovascular events (52). GC-induced comorbidities in CS may be at least in part epigenetically mediated. Previous study using whole blood methylation profile demonstrated that specific hypomethylated CpG sites induced by GC were associated with Cushing comorbidities, such as hypertension and osteoporosis (46). The study identified a methylator predictor of GC excess which could be used as a biomarker to monitor GC status (46).

The long-term exposure to high cortisol levels may be crucial for the persistence of some morbidities in CS through epigenetic changes. Hypercortisolism-induced persistent changes in visceral adipose tissue gene expression through epigenetic modifications was investigated in a translational study (12). This study combined data from patients with active CS and data from an animal model of CS in active and remitted phase. Interestingly, the study demonstrated long-lasting changes in the transcriptome of adipose tissue that were associated with histone modifications induced by GC. Therefore, these epigenetic fingerprints observed even after the resolution of hypercortisolism may elucidate the mechanism of persistent modifications in gene expression in the visceral adipose tissue (12).

With regard to the persistence of GC-induced DMR, a genome-wide DNA methylation analysis showed a lower average of DNA methylation in patients in remission of CS compared to controls. Interestingly, the most common biologically relevant affected genes were retinoic acid receptors, thyroid hormone receptors, or hormone/nuclear receptors, important genes related to intracellular pathways and regulators of gene expression (9).

In summary, this large body of evidence supports the concept that prolonged GC exposure modulates the epigenetic landscape across the genome by inducing DMR and histone modifications. Some epigenetic modifications are persistent, and this may partially explain the incomplete reversibility of some of CS features following clinical remission.

Glucocorticoid-Target Genes in Cushing Syndrome

A detailed identification and characterization of GC-target genes may shed light in the understanding of the pathophysiology and treatment response in patients with CS. For instance, the GC regulation of pro-opiomelanocortin (POMC) expression as part of the physiologic GC negative feedback may be impaired in Cushing disease (CD), which is an important mechanism for the maintenance of high GC levels (53). Another example is the interaction between GC and clock genes, which may interfere in the loss of the GC circadian rhythm and may contribute to metabolic disorders in CS (54). Furthermore, the suppressive action of GC on drug targets, such as the somatostatin receptor (subtype 2), may influence the efficacy of first-generation somatostatin receptor ligands in normalizing cortisol levels in CD (55). Here we describe how GCs using epigenetic machinery influence the expression of important target genes and their implications in CS.

FKBP5

FK506 binding protein (FKBP5) plays an important role in the regulation of hypothalamic-pituitary-adrenal (HPA) system (56). As part of the GC negative feedback loop, GC binds to hypothalamic and pituitary GR. In the cytoplasm, GR is bound to a multi-protein complex including FKBP5. FKBP5 modulates GR action by decreasing GR binding affinity to GC and by preventing GR translocation from cytoplasm to nucleus (5758). In other words, an increase of FKBP5 expression is inversely correlated with GR activity and results in GC resistance leading to an impaired negative feedback regulation in the HPA axis (59).

FKBP5 is a GC-responsive gene; its upregulation by GC is part of an intracellular negative short-feedback loop (60). The mechanism by which GC regulates FKBP5 expression was shown to include inhibition of DNA methylation (44). In a model for CS, mice treated with corticosterone for 4 weeks had a reduced level of DNA methylation of FKBP5 in DNA extracted from whole blood, which was strongly correlated in a negative manner with GC concentration. Interestingly, a negative correlation was also observed between the degree of FKBP5 gene methylation measured at 4 weeks of GC exposure and the percentage of mice visceral fat (61). Accordingly, previous studies have provided compelling evidence of decreased methylation in the FKBP5 gene in patients with active CS compared to healthy control (1046). Even in patients with CS in remission, previous data have suggested a small decrease in FKBP5 methylation levels compared to healthy controls (910). In an in vitro study, it was demonstrated that, by decreasing DNMT1 expression, GC is able to reduce FKBP5 methylation levels and, therefore, increase its expression (44).

Likewise, FKBP5 mRNA is also sensitive to GC exposure. A time-dependent increase in blood FKBP5 mRNA after single-dose prednisone administration has been demonstrated in healthy humans (62). Accordingly, patients with ACTH-dependent CS had higher blood FKBP5 mRNA levels compared with healthy controls, and after a successful surgery, FKBP5 mRNA returned to baseline levels (63). Furthermore, in another study, blood FKBP5 mRNA was inversely correlated with FKBP5 promoter methylation and positively correlated with 24-hour urine free cortisol (UFC) levels in patients with CS (46). Taken together, this fine-tuning of FKBP5 DNA methylation and mRNA according to the level of GC suggests that FKBP5 can be used as a biomarker to infer the magnitude of GC exposure.

POMC and Corticotropin-Releasing Hormone

The partial resistance of the corticotroph adenoma to GC negative feedback is a hallmark of CD. Indeed, the lack of this inhibitory effect constitutes a method to diagnose CD, that is, with the dexamethasone suppression test. One of the mechanisms related to the insensitivity to GC can be attributed to GR mutations which are, however, rarely found in corticotrophinomas (64). Another mechanism that was uncovered in corticotroph adenomas is an overexpression of the HSP90 chaperone resulting in reduced affinity of GR to its ligand and consequently GR resistance (5365).

In addition, the loss of protein expression of either Brg1, ATPase component of the SWI/SNF chromatin remodeling complex, or HDAC2 has been linked to GC resistance in about 50% of some adenomas (66). The trans-repression process on POMC transcription achieved by GC involves both the histone deacetylation enzyme and Brg1. One mechanism of corticotropin-releasing hormone (CRH)-induced POMC expression is through an orphan nuclear receptor (NR) related to NGFI-B (Nur77). NGFI-B binds to the NurRE sequence in the promoter region of POMC gene and recruits a co-activator to mediate its transcription. In a tethering mechanism, the GR directly interacts with NGFI-B to form a trans-repression complex, which contains the GR itself, Brg1, the nuclear receptor, and HDAC2; the latter being essential to block the gene expression through chromatin remodeling process (5366).

In CD, hypercortisolism exerts a negative feedback at CRH secretion from the hypothalamus (67). The mechanism involved in GR-induced suppression of CRH expression is through direct binding to a nGRE in the promoter region of CRH gene and subsequent recruitment of repressor complexes. In a rat hypothalamic cell line, it was demonstrated that Dex-induced CRH repression occurs through coordinated actions of corepressors involving Methyl-CpG-binding protein 2 (MeCP2), HDAC1, and DNA methyltransferase 3B (DNMT3B). Possibly, GR bound to nGRE recruits DNMT3B to the promoter in order to methylate a specific region, subsequently binding MeCP2 on these methylated sites followed by the recruitment of chromatin modify corepressor HDAC1, ultimately resulting in CRH suppression. Another possibility is that 2 independent complexes, one consisting of GR with DNMT3 for the methylation and the other the MeCP2, bound to methylated region, interact with HDAC1 to induce repression (68).

Clock Genes

The clock system and the HPA axis are interconnected regulatory systems. Cortisol circadian rhythm is modulated by the interaction between a central pacemaker, located in the hypothalamic suprachiasmatic nuclei, and the HPA axis (69). At the molecular level, mediators of the clock system and cortisol also communicate with each other, both acting as transcription factors of many genes to influence cellular functions.

In CS, the impact of chronic GC exposure on clock genes expression was recently evaluated using peripheral blood samples from patients with active disease compared with healthy subjects. The circadian rhythm of peripheral clock gene expression (CLOCK, BMAL, PER1-3, and CRY1) was abolished as a result of hypercortisolism, and that may contribute to metabolic disorders observed in Cushing patients (70). Another study, which investigated persistent changes induced by hypercortisolism in visceral adipose tissue, found that the expression of clock genes, such as PER1, remained altered in association with persistent epigenetic changes in both H3K4me3 and H3K27ac induced by hypercortisolism even after the resolution of hypercortisolism (12). This suggests that chronic exposure to GC may induce sustained epigenetic changes that can influence clock genes expression. Nevertheless, further studies are warranted to better elucidate how long-term exposure to GC impacts clock genes expression using the epigenetic machinery.

Glucocorticoid Effects on MicroRNAs

Along with histone modification and DNA methylation, microRNAs (miRNAs) have emerged as an epigenetic mechanism capable of impacting gene expression without changing DNA sequence (15). Interestingly, miRNA expression itself is also under the influence of epigenetic modifications through promoter methylation like any other protein-encoding genes (71).

MicroRNAs are small (about 20-25 nucleotides in length) non-coding RNAs that are important in transcriptional silencing of messenger RNA (mRNA). By partially pairing with mRNA, miRNAs can either induce mRNA degradation or inhibit mRNA translation to protein. MiRNAs regulate the translation of about 50% of the transcriptome, allowing them to play an important role in a wide range of biological functions, such as cell differentiation, proliferation, metabolism, and apoptosis under normal physiological and pathological situations. Some miRNAs can be classified as oncogenes or tumor suppressing genes, and aberrant expression of miRNAs may be implicated in tumor pathogenesis (71-73).

Insight into the regulation of miRNA expression is, therefore, crucial for a better understanding of tumor development and other human diseases, including cardiac, metabolic, and neurological disorders (7374). There are different regulatory mechanisms involved in miRNA expression, including transcriptional factors such as GR-GC. GC may modulate miRNA expression through direct binding to GRE in the promoter region of the host gene, as observed in hemopoietic tumor cells (75). In addition to transcriptional activation, in vascular smooth muscle cells, Dex treatment induces downregulation of DNMT1 and DNMT3a protein levels and reduces the methylation of miRNA-29c promoter, resulting in an increased expression of miRNA-29c (76). Interestingly, it was demonstrated that the increased expression of miRNA-29 family (miRNA-29a, -29b, and -29c) associates with metabolic dysfunction, such as obesity and insulin resistance, which pertains to CS (7778). With regard to metabolic dysfunction, miRNA-379 expression was shown to be upregulated by GC and its overexpression in the liver resulted in elevated levels of serum triglycerides associated with very low-density lipoprotein (VLDL) fraction in mice (79). In obese patients, the level of hepatic miRNA-379 expression was higher compared to nonobese patients and positively correlated with serum cortisol and triglycerides (79). Hence, GC-responsive miRNA may be, at least in part, a mediator to GC-driven metabolic conditions in CS.

In pathological conditions, such as seen in CS, prolonged exposure to an elevated cortisol level results in a wide range of comorbidities. It can be hypothesized that the chronic and excessive glucocorticoid levels may induce an aberrant miRNA expression that might impact several cellular processes related to bone and cardiometabolic disorders. A recent study addressed the impact of hypercortisolism on bone miRNA of patients with active CD compared to patients with nonfunctional pituitary adenomas. Significant changes in bone miRNA expression levels were observed, suggesting that the disruption of miRNA may be partially responsible for reduced bone formation and osteoblastogenesis (80). Similarly, altered expression levels of selected miRNAs related to endothelial biology in patients with CS may point to a contribution to a high incidence of cardiovascular disorders in Cushing patients (81). Therefore, dysregulated miRNAs as a consequence of high cortisol levels may underpin the development and progression of comorbidities related to CS. To the best of our knowledge, it is currently not clear whether miRNA dysregulation persists after resolution of hypercortisolism, thus contributing to the persistence of some comorbidities. This hypothesis needs to be further investigated.

MicroRNA can also be used as a diagnostic tool in CS. A study was performed to identify circulating miRNA as a biomarker to differentiate patients with CS from patients with suspected CS who had failed diagnostic tests (the control group) (82). It was observed that miRNA182-5p was differentially expressed in the CS cohort compared to the control group; therefore, it may be used as a biomarker (82). However, a large cohort is necessary to validate this finding (82). In corticotroph tumors, downregulation of miRNA 16-1 expression was observed relative to normal pituitary tissue (83). In contrast, the plasma level of miRNA16-5p was found to be significantly higher in CD compared to ectopic Cushing (EAS) and healthy controls (84). This finding suggests that miRNA16-5p may be a biomarker capable to differentiate the 2 forms of ACTH-dependent Cushing (84).

Epidrugs and Glucocorticoid Action in Cushing’s Syndrome

The interest in understanding the epigenetic mechanism of GC action in the context of CS is based on reversibility of epi-marks, such as DNA methylation and histone modifications, using epidrugs (8586). The biological characteristics of epigenetic drugs and their target have been extensively explored. Their effectiveness as antitumor drugs have been tested on corticotroph tumors using in vitro studies (87-89). However, a limited number of studies have explored the role of epidrugs as a therapeutic tool in reversing the genomic action of GC in CS, particularly in comorbidities induced by hypercortisolism (9091).

The use of histone deacetylase inhibitors (HDACi) may reduce the genomic action of GC (90-92). It has been demonstrated that the use of the HDAC inhibitor valproic acid increases the acetylation level of GR, consequently attenuating the genomic action of GC. In an experimental Cushing model in rats, the use of valproic acid decreased expression of genes related to lipogenesis, gluconeogenesis, and ion regulators in the kidney that ultimately reduces hepatic steatosis, hyperglycemia, and hypertension in ACTH-infused rats (9091).

More studies evaluating the effects of epidrugs influencing the GC actions are warranted to further elucidate the underlying mechanisms and to explore potential treatment modalities to reverse long-lasting consequences of chronic corticoid exposure.

Conclusions

In physiologic conditions, GC are secreted in pulses following a circadian rhythm pattern, as opposed to a constant, chronic, and high GC exposure in CS. This pathological pattern may account for numerous devastating effects observed in CS (7). Yet, the expressed genome in response to chronic GC exposure may potentially be abnormal, leading to dysregulation in clock genes, among other effects.

GC levels may return to a normal circadian pattern in response to a successful treatment, but with incomplete reversibility of some CS features, which may in part be explained by epigenetic changes. The epigenetic machinery is used by GC to induce dynamic changes in chromatin to modulate gene expression. (Fig. 2) It seems that most of chromatin modifications are reversible, but some may persist resulting in long-term epigenetic changes. (Table 1)

Table 1.

Evidence of interaction between glucocorticoid and epigenetic machinery

Epigenetic changes/epigenetic enzymes Action
Histone acetylation (HAT)
  • Glucocorticoid receptors (GR) recruit co-regulators, such as CREB binding protein (CBP), which has intrinsic histone acetyltransferase (HAT) activity that modifies the chromatin structure from an inactive to an active state (253335).

Histone deacetylation (HDAC)
  • GR recruit histone deacetylases (HDACs) to turn chromatin less accessible and suppress gene transcription (2535).

  • The trans-repression process on POMC transcription achieved by glucocorticoids (GC) involves the histone deacetylation enzyme (HDAC2).

  • GC mediates the upregulation of HDAC2 in rats exposed to chronic stress (40).

Histone demethylase (JMJD3)
  • GC suppress transcription of JMJD3 in endothelial cells treated with TNFα (41).

Histone modifications
  • Using ChIP-seq, a study in mice treated for 5 weeks with corticosterone showed higher levels of histone modifications (H3K4me3, H3K27ac) compared to control mice. In mice after a 10-week washout period, persistence of this epigenetic fingerprint was observed, which was associated with long-lasting changes in gene expression (12).

DNA methylation (DNMT3B) and histone deacetylation (HDAC1)
  • GC mediates CRH downregulation through DNMT3B to the promoter in order to methylate a specific region and recruitment of chromatin modify corepressor HDAC (68).

DNA hypomethylation
  • GC induces downregulation of DNMT1 in AtT20 (mouse corticotroph adenoma cell line) (20).

  • GC induces upregulation of TET enzyme expression which was described in retinal and osteocyte cell line model (4243).

  • An experimental study in mice previously exposed to high levels of GC showed differentially methylated regions (DMR) induced by GC treatment, of which the majority was loss of the methylation (11).

  • Reduced DNA methylation in FKBP5 gene was found in patients in active disease and also in remission state of Cushing syndrome (CS) as compared to a healthy control group (10).

  • A genome-wide DNA methylation analysis showed a lower average of DNA methylation in patients in remission of CS compared to controls (9).

  • A study using whole blood methylation profile demonstrated an association between cortisol excess and DNA hypomethylation in patients with CS (46).

Further studies are needed to elucidate how chronic exposure to GC leads to incomplete reversibility of CS morbidities via sustained modulation of the epigenetic machinery and possibly other mechanisms. Subsequent identification of therapeutic targets may offer new perspective for treatments, for example, with epidrugs, aiming to reverse hypercortisolism-related comorbidities.

Funding

The authors received no financial support for this manuscript.

Disclosures

T.P., R.A.F., and L.J.H. have nothing to declare.

Data Availability

Data sharing is not applicable to this article, as no datasets were generated or analyzed during the current study.