MEN4 variant of unknown significance in two generations of patients presenting with Cushing disease

Abstract

Cushing disease is the most common form of endogenous hypercortisolism and can be associated with familial endocrine syndromes, including multiple endocrine neoplasia type 4 (MEN4). We report a case of a father and daughter who both presented with Cushing disease and were found to have the same variant of unknown significance in the CDKN1B gene (c.280 C>T, p.P94S), the gene responsible for MEN4. They were treated surgically and are in remission from Cushing disease. This case raises the possibility that the variant could be pathogenic, despite in silico predictions to the contrary.

Introduction

Cushing disease is the most common form of endogenous hypercortisolism. Most forms are sporadic, but some inherited causes include multiple endocrine neoplasia type 1 or type 4 (MEN1 or MEN4, respectively), Carney complex, and familial isolated pituitary adenomas [1].

MEN4 is a rare endocrine syndrome with an autosomal dominant inheritance pattern, characterized by haploinsufficiency in the CDKN1B gene. CDKN1B encodes a cyclin-dependent kinase inhibitor that regulates the G1→S phase transition by inhibiting CDK2/cyclin E and CDK4/6 complexes. It is a tumor suppressor and is implicated in cell cycle control, endocrine tumor predisposition, and growth regulation. MEN4 is characterized by an increased risk of endocrine tumors, including parathyroid, anterior pituitary, and gastro-entero-pancreatic neuroendocrine tumors [1]. A study of mostly children with Cushing disease identified CDKN1B variants in up to 2.6% of individuals [2]. Given MEN4's low suspected prevalence (<1 per million persons) and mild presentation that tends to present at an older age, there are currently no specific guidelines for screening or monitoring patients who are carriers of mutations in the CDKN1B gene [3]. Additionally, due to the limited number of documented MEN4 cases, there are several variants of unknown significance (VUS), making the diagnosis of MEN4 more challenging.

Here, we describe 2 first-degree relatives—a father and daughter—who both presented with Cushing disease and were found to have the same VUS in the MEN4 gene, raising the possibility that it is pathogenic.

Case presentation

Case 1

In 2016, a 62-year-old male with hypertension presented with worsening type 2 diabetes and insomnia. Physical exam findings were notable for central obesity, mild plethora, and proximal muscle weakness in the upper extremities. Upon review of a prior abdominal computed tomography (CT) scan, he was noted to have bilateral adrenal hyperplasia. Biochemical testing demonstrated an elevated 24-hour urine free cortisol (178.0 µg/24 hours (491 nmol/24 hours) [reference range: 4-50 µg/24 hours (SI: 11-138 nmol/24 hours)]) with an adrenocorticotropic hormone (ACTH) level of 45 pg/mL (9.9 pmol/L) (reference range: 6-50 pg/mL [SI: 1.3-11 pmol/L]). An overnight 1 mg dexamethasone suppression test resulted in a nonsuppressed cortisol (28 µg/dL (773 nmol/L) [reference range: ≤ 1.8 µg/dL (SI: ≤ 50 nmol/L)]) despite an adequate dexamethasone level. Given the ACTH-dependence of the hypercortisolemia, a magnetic resonance imaging (MRI) scan of the pituitary was performed and showed a possible right sided microadenoma (∼5 mm). Inferior petrosal sinus sampling (IPSS) was subsequently performed and demonstrated centralization in the setting of hypercortisolemia at the time of the IPSS (Table 1). Therefore, the patient subsequently underwent surgical resection via an endoscopic endonasal approach. Both the right and left sides of the gland were explored during surgery and suspicious areas were resected from both the right and left sides of the gland. Nadir serum cortisol levels were >60 µg/dL (>1655 nmol/L); therefore, the patient underwent repeat surgical exploration 3 days later. Again, both the right and left sides were re-explored during this second surgery, and ultimately an area on the left was resected that was positive for adenoma on cryosection. After the second surgery, the serum cortisol nadir was 3 µg/dL (83 nmol/L), and glucocorticoid replacement was initiated. Immunohistochemistry showed a pituitary adenoma with strong CAM5 staining, patchy ACTH staining, and with minimal luteinizing hormone, follicle stimulating hormone (FSH), thyroid stimulating hormone (TSH), growth hormone, or prolactin staining.

Table 1.

Inferior petrosal sinus sampling (IPSS)

Time before or after CRH administration	Father
	Peripheral ACTH Ref. range: 9-46 pg/mL (SI: 2-10 pmol/L)	Left IPS ACTH Ref. range: 9-46 pg/mL (SI: 2-10 pmol/L)	Right IPS ACTH Ref. range: 9-46 pg/mL (SI: 2-10 pmol/L)
−15 minutes	29 pg/mL (SI: 6.4 pmol/L)	502 pg/mL (SI: 110.5 pmol/L)	509 pg/mL (SI: 112.1 pmol/L)
−5 minutes	29 pg/mL (SI: 6.4 pmol/L)	162 pg/mL (SI: 35.7 pmol/L)	99 pg/mL (SI: 21.8 pmol/L)
0 minutes (CRH administered)	24 pg/mL (SI: 5.3 pmol/L)	32 pg/mL (SI: 7.0 pmol/L)	201 pg/mL (SI: 44.3 pmol/L)
+ 2 minutes	37 pg/mL (SI: 8.1 pmol/L)	490 pg/mL (SI: 107.9 pmol/L)	633 pg/mL (SI: 139.4 pmol/L)
+ 5 minutes	71 pg/mL (SI: 15.6 pmol/L)	960 pg/mL (SI: 211.4 pmol/L)	4341 pg/mL (SI: 955.9 pmol/L)
+ 10 minutes	84 pg/mL (SI: 18.5 pmol/L)	3259 pg/mL (SI: 717.6 pmol/L)	1407 pg/mL (SI: 309.8 pmol/L)

Time before or after DDAVP administration	Daughter
	Peripheral ACTH Ref. range: 7.2-63.3 pg/mL (SI: 1.6-13.9 pmol/L)	Left IPS ACTH Ref. range: 7.2-63.3 pg/mL (SI: 1.6-13.9 pmol/L)	Right IPS ACTH Ref. range: 7.2-63.3 pg/mL (SI: 1.6-13.9 pmol/L)
−10 minutes	5.9 pg/mL (SI: 1.3 pmol/L)	92.2 pg/mL (SI: 20.3 pmol/L)	39.4 pg/mL (SI: 8.7 pmol/L)
−3 minutes	5.2 pg/mL (SI: 1.1 pmol/L)	107.5 pg/mL (SI: 23.7 pmol/L)	30.4 pg/mL (SI: 6.7 pmol/L)
0 minutes (DDAVP administered)	3.6 pg/mL (SI: 0.79 pmol/L)	54.5 pg/mL (SI: 12.0 pmol/L)	17.3 pg/mL (SI: 3.8 pmol/L)
+3 minutes	3.0 pg/mL (SI: 0.66 pmol/L)	62 pg/mL (SI: 13.7 pmol/L)	11.2 pg/mL (SI: 2.5 pmol/L)
+5 minutes	3.4 pg/mL (SI: 0.75 pmol/L)
+10 minutes	3.8 pg/mL (SI: 0.84 pmol/L)	59 pg/mL (SI: 13.0 pmol/L)	15.3 pg/mL (SI: 3.4 pmol/L)
+15 minutes	3.6 pg/mL (SI: 0.79 pmol/L)	127 pg/mL (SI: 28.0 pmol/L)	31.1 pg/mL (SI: 6.8 pmol/L)

Hypercortisolemia was documented at the time of testing for both patients and given the ≥ 2/1 central to peripheral ACTH ratio pre-CRH and pre-DDAVP as well as the ≥ 3/1 central to peripheral ACTH ratio post-CRH and post-DDAVP, both studies were consistent with a central source of ACTH production. Of note, the father received CRH, and the daughter received DDAVP.

Abbreviations: ACTH, adrenocorticotropic hormone; CRH, corticotropin-releasing hormone; DDAVP, 1-deamino-8-D-arginine vasopressin; IPS, inferior petrosal sinus.

Case 2

Approximately 5 years later, the patient’s 40-year-old daughter presented for evaluation of weight gain. She had a distant history of normo-calcemic nephrolithiasis, but there was no biochemical evidence of hyperparathyroidism on laboratory evaluation (parathyroid hormone 60 pg/mL (60 ng/L) [reference range: 18-88 pg/mL (SI: 18-88 ng/L)], calcium 9.7 mg/dL (2.4 mmol/L) [reference range: 8.4-10.2 mg/dL (SI: 2.1-2.55 mmol/L)]). She was also diagnosed with primary hypothyroidism in her mid-thirties after presenting with secondary amenorrhea, for which she was started on levothyroxine and subsequently successfully conceived. Approximately 2 years after the birth of her second child, she started aggressive lifestyle modification for weight loss, but she was unsuccessful and gained 50 pounds. She also reported excessive sweating, proximal muscle weakness, and poor sleep.

Diagnostic assessment

On initial examination, the daughter was hypertensive (blood pressure: 151/80 mmHg) with a confirmed weight gain of 20 pounds over 6 months based on weights documented in the medical record. On exam, she had moon facies, mild plethora, supraclavicular fullness, and a dorsocervical pad. Her proximal muscle strength was 4+/5 in her upper and lower extremities.

Biochemical testing for hypercortisolemia was notable for 2 elevated late-night salivary cortisol levels (0.126 and 0.094 µg/dL (3.48 and 2.59 nmol/L) [reference range: ≤ 0.09 µg/dL (SI: < 2.5 nmol/L)]) and an elevated 24-hour urine free cortisol [74 µg/24 hours (204 nmol/24 hours)]. An overnight 1 mg dexamethasone suppression test showed an unsuppressed cortisol of 4.3 µg/dL (118.6 nmol/L). Of note, her hypothyroidism was noted to be well-controlled on levothyroxine (TSH 1.09 µU/mL (1.09 mU/L), free thyroxine (fT4) 1.08 ng/dL (13.9 pmol/L) [reference range: TSH 0.3-5.0 µU/mL (SI: 0.3-5.0 mU/L), fT4 0.89-1.78 ng/dL (SI: 11.5-23 pmol/L)]).

ACTH measured using the Roche Elecsys assay was 25.8 pg/mL (5.68 pmol/L) [reference range: 7.2-63.3 pg/mL (SI: 1.6-13.9 pmol/L)], indicating ACTH-dependent hypercortisolemia. Pituitary MRI did not demonstrate a discrete lesion. IPSS was subsequently performed; hypercortisolemia at the time of the testing was confirmed, and the results were consistent with a central source (Table 1), although peripheral ACTH levels were noted to be quite low at the time of testing.

Given that both father and daughter had a history of Cushing disease, genetic evaluation was performed using Invitae's Hereditary Neuroendocrine Tumors and Adrenocortical Carcinoma panel. For both patients, genetic testing showed a VUS in the CDKN1B gene, c.280 C>T (p.P94S), which is the gene associated with MEN4.

Treatment

Like her father, the patient underwent surgical exploration via an endoscopic endonasal approach for a suspected corticotroph adenoma. The surgical exploration was focused on the left side of the gland and suspicious areas were resected and sent for pathologic evaluation. Postoperatively, her serum cortisol levels did not drop (nadir: 14 µg/dL (386 nmol/L)) and the patient underwent a second exploration 4 days later, focusing on the right side of the gland. Suspicious areas on the right side of the gland were resected during this second exploration and postoperatively the serum cortisol nadir was 4 µg/dL (110 nmol/L) suggesting biochemical remission. Pathology did not demonstrate an adenoma from either exploration. The patient was started on glucocorticoid replacement (dexamethasone 0.5 mg PO qAM) and had 2 documented low-normal 24-hour urine cortisol levels [6 µg/24 hours and 8 µg/24 hours (17 nmol/24 hours and 22 nmol/24 hours)], suggestive of biochemical remission. She was titrated off glucocorticoid replacement 6 months after surgery.

Outcome and follow-up

Case 1 and Case 2

Both father and daughter remain in biochemical remission from Cushing disease. They follow regularly with neuroendocrinology for ongoing evaluation and monitoring. Thus far, neither patient has clinical, biochemical, or radiological evidence of hyperparathyroidism or a gastro-entero-pancreatic neuroendocrine tumor on CT or MR imaging. Notably, the child of the daughter is currently being evaluated by pediatric endocrinology for premature adrenarche. She has not undergone genetic testing for the MEN4 mutation.

Discussion

This case report describes 2 related patients who presented with Cushing disease and were successfully treated with pituitary surgery, resulting in biochemical remission. Both were found to carry the same CDKN1B variant, the gene responsible for MEN4, suggesting a possible causative relationship. Twenty-eight pathogenic variations of CDKN1B causing MEN4 are cataloged in ClinVar. However, c.280 C>T is currently classified as a VUS and, therefore, is nondiagnostic in this family. At the current time, neither the father nor the daughter exhibits other features of MEN4, such as parathyroid or gastro-entero-pancreatic tumors, but given the reported relationship with mutations in this gene and Cushing disease, the c.280 C>T variant could indeed be pathogenic.

ClinVar lists 5 reports of the c.280 C>T (p.P94S) variant, but none contain clinical or family information. It is present in population databases (rs146217922, gnomAD 0.006%) [4]. This missense change has also been observed in an individual with primary hyperparathyroidism [5]. In silico analysis of the effect of the p.P94S substitution on protein function using PolyPhen-2 yielded a score of 0.031, indicating it to be benign [6]. The SIFT score is 0.12 (damaging is <0.05), also predicting that the substitution is unlikely to impair protein function [7]. Position 94 lies outside the core CDK-binding inhibitory domain (AAs 25-93), is near a structurally flexible region, and is not within a known catalytic or binding hotspot [8]. Replacing proline with a serine does not introduce a charge nor create steric clash based on molecular modeling. However, proline imposes backbone rigidity, and thus the serine substitution could increase flexibility, alter tertiary structure, and introduce a potential phosphorylation site. Thus, additional multigeneration case reports will be necessary to redefine this variant as likely pathogenic.

Learning points

• MEN4 is a rare endocrine syndrome associated with pathogenic variants in the CDKN1B gene with autosomal dominant inheritance.

• Clinical features of MEN4 are similar to MEN1, however, MEN4 patients tend to present later in life and may have milder disease.

• The CDKN1B c.280 C>T (p.P94S) variant could be pathogenic for MEN4 but requires additional studies.

Contributors

All authors made individual contributions to the authorship. P.F. was involved in the diagnosis and/or management of these patients. S.G. and A.A. were involved in the conception and design of the work, interpretation of data, and drafting of the work. J.V. contributed significantly to the interpretation of the genetic variant and in silico analysis. All authors participated in manuscript editing and reviewed and approved the final draft.

Funding

No public or commercial funding.

Disclosures

P.K.F. receives research support from Corcept Therapeutics, Crinetics Pharmaceuticals and Quest Diagnostics. She has served as an occasional advisory board member for Corcept, Crinetics, Chiesi, and Camurus. She has served as a consultant for Regeneron and Quest. These disclosures are not related to the content of this manuscript.

Informed patient consent for publication

Informed consent was obtained directly from the patients.

Data availability

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