Glucocorticoid Withdrawal Syndrome following treatment of endogenous Cushing Syndrome

Abstract

Purpose:

Literature regarding endogenous Cushing syndrome (CS) largely focuses on the challenges of diagnosis, subtyping, and treatment. The enigmatic phenomenon of glucocorticoid withdrawal syndrome (GWS), due to rapid reduction in cortisol exposure following treatment of CS, is less commonly discussed but also difficult to manage. We highlight the clinical approach to navigating patients from GWS and adrenal insufficiency to full hypothalamic-pituitary-adrenal (HPA) axis recovery.

Methods:

We review the literature on the pathogenesis of GWS and its clinical presentation. We provide strategies for glucocorticoid dosing and tapering, HPA axis testing, as well as pharmacotherapy and ancillary treatments for GWS symptom management.

Results:

GWS can be difficult to differentiate from adrenal insufficiency and CS recurrence, which complicates glucocorticoid dosing and tapering regimens. Monitoring for HPA axis recovery requires both clinical and biochemical assessments. The most important intervention is reassurance to patients that GWS symptoms portend a favorable prognosis of sustained remission from CS, and GWS typically resolves as the HPA axis recovers. GWS also occurs during medical management of CS, and gradual dose titration based primarily on symptoms is essential to maintain adherence and to eventually achieve disease control. Myopathy and neurocognitive dysfunction can be chronic complications of CS that do not completely recover.

Conclusions:

Due to limited data, no guidelines have been developed for management of GWS. Nevertheless, this article provides overarching themes derived from published literature plus expert opinion and experience. Future studies are needed to better understand the pathophysiology of GWS to guide more targeted and optimal treatments.

Introduction

Endogenous neoplastic hypercortisolism – Cushing syndrome (CS) – is one of the most challenging diagnostic and management problems in clinical endocrinology. CS may be due to either a pituitary tumor (Cushing disease, CD), or a non-pituitary (ectopic) tumor secreting ACTH. ACTH-independent hypercortisolism due to unilateral or bilateral adrenal nodular disease has been increasingly recognized as an important cause of CS. Regardless of the cause of CS, the clinical manifestations are protean and include a myriad of clinical, biochemical, neurocognitive, and neuropsychiatric abnormalities. The catabolic state of hypercortisolism causes signs and symptoms including skin fragility, bruising, delayed healing, violaceous striae, muscle weakness, and low bone mass with fragility fractures. Other clinical features include weight gain, fatigue, depression, difficulty concentrating, insomnia, facial plethora, and fat redistribution to the head and neck with resultant supraclavicular and dorsocervical fullness[1]. Metabolic consequences of hypercortisolism including hypertension, diabetes, and dyslipidemia are common. In addition, women often experience hirsutism and menstrual irregularity, while men may have hypogonadism.

Management options of CS include surgery, medications, and radiation. The preferred first line treatment, regardless of source, is surgery, which offers the potential for remission[2,3,4]. The primary literature, reviews, and clinical practice guidelines for CS have traditionally focused on the diagnosis, subtyping, and surgical approach to CS. This bias derives first from the profound diagnostic challenge posed in the evaluation of cortisol production and dynamics, given that circulating cortisol follows a circadian rhythm, exhibits extensive protein binding and metabolism, and rises acutely with stress. CD and ectopic ACTH syndrome may be difficult to distinguish clinically and biochemically, and inferior petrosal sinus sampling is required in many patients to resolve this differential diagnosis. Ectopic ACTH-producing tumors can also be small, and these tumors can escape localization despite the best current methods. Although diagnosis and initial surgical remission can be achieved in the majority of patient with CS at experienced centers, up to 50% of patients with CD will require additional therapies after unsuccessful primary surgeries or recurrence up to many years later[5]. For patients who do not achieve surgical cure or who are not surgical candidates, several medical treatment options are now available. Pharmacotherapies directed at the pituitary include pasireotide[67] (FDA approved) and cabergoline[8]. Adrenal steroidogenesis inhibitors such as osilodrostat[9] (FDA approved), metyrapone[10], levoketoconazole[11] (FDA approved) and ketoconazole[12], as well as the glucocorticoid antagonist, mifepristone[13] (FDA approved), are now widely used to treat CS. Pituitary radiotherapy is an additional treatment option for CD but can take months to years to lower cortisol production. Bilateral adrenalectomy (BLA) provides immediate, reliable correction of hypercortisolism but mandates life-long corticosteroid replacement therapy, and, in patients with CD, may be complicated by corticotroph tumor progression syndrome in 25–40% of patients[14].

After successful surgery for CS, the rapid onset of adrenal insufficiency (AI) is anticipated and usually portends a favorable prognosis [15,16,17,18]; however, despite the use of post-operative corticosteroid replacement, the rapid reduction in cortisol exposure often results in an enigmatic phenomenon referred to as the glucocorticoid withdrawal syndrome (GWS). This article addresses the clinical presentation and the pathogenesis of GWS, as well as its distinction from AI. When available, appropriate references are provided. Statements and guidance provided without references are derived from expert opinion and experience.

Clinical Presentation and Pathogenesis of GWS

GWS occurs following withdrawal of supraphysiologic exposure to either exogenous or endogenous glucocorticoids of at least several months duration[19]. After surgical cure of endogenous CS, GWS is usually characterized by biochemical evidence of hypothalamic-pituitary-adrenal (HPA) axis suppression with many signs and symptoms consistent with cortisol deficiency despite the use of supraphysiologic glucocorticoid replacement therapy. The degree of physical or psychologic glucocorticoid dependence experienced by patients may not correlate with the degree of HPA axis suppression[2021].

GWS symptom onset is typically 3–10 days postoperatively, often after the patient has been discharged from the hospital. The first symptoms of GWS vary but usually consist of myalgias, muscle weakness, fatigue, and hypersomnolence. Anorexia, nausea, and abdominal discomfort are common, but vomiting should raise concern for hyponatremia, cerebrospinal fluid leak, hydrocephalus, or other perioperative complications. Mood changes develop more gradually and range from mood swings to depression, and the fatigue with myalgias can exacerbate mood changes. An atypical depressive disorder has been described in many patients after CD surgery[22]. Weight loss should ensue in most patients but gradually and proportionate to the reduction in glucocorticoid exposure. It is important to complete a thorough symptom review and physical exam at postoperative visits, as the differentiation between GWS and bona fide AI – and even between GWS and recurrence of CS – can be challenging (Fig. 1). All three conditions are associated with symptoms of myalgias, weakness, and fatigue; however, rapid weight loss, hypoglycemia, and hypotension are suggestive of AI and the need for an increase in the glucocorticoid dose. In parallel, hypersomnia is more suggestive of GWS, while insomnia is more associated with recurrence of CS. Given the anticipation of GWS onset shortly after discharge and the potential for hyponatremia during this time, a widely employed strategy is a generous glucocorticoid dose for the first 2–3 weeks, at least until the first postoperative outpatient visit (Table 1).

Fig. 1

figure 1

Overlapping clinical features of Cushing syndrome (CS), glucocorticoid withdrawal syndrome (GWS), and adrenal insufficiency (AI)

Table 1 Glucocorticoid Therapy Options After Surgery for CS

The mechanisms responsible for the precipitation of the GWS after surgery for CS and the variability in its manifestations are not completely understood, yet alterations in the regulation of cortisol and cortisol-responsive genes appear to contribute. Down-regulation of corticotropin-releasing hormone (CRH) and proopiomelanocortin (POMC) expression, combined with up-regulation of cytokines and prostaglandins are likely to be important components of GWS. Low CRH has been associated with atypical depression[23], and CRH levels in cerebrospinal fluid of patients with CD are significantly lower compared to healthy subjects[24]. CRH suppression gradually resolves after surgical cure over 12 months during glucocorticoid replacement[25], illustrative of the slow recovery process. The expression of POMC, the ACTH precursor molecule, is also suppressed with chronic glucocorticoid exposure[26], and the normalization of POMC-associated peptides mirrors HPA axis recovery[19]. In the acute phase of glucocorticoid withdrawal, interleukins IL-6 and IL-1β, as well as tumor-necrosis factor alpha (TNFα) have been observed to rise[27], suggesting that glucocorticoid-mediated suppression of cytokines and prostaglandins is then released in GWS, and these cytokines induce the associated flu-like symptoms. Glucocorticoid replacement with dexamethasone 0.5 mg/d reduced but did not normalize IL-6 after 4–5 days[27], consistent with resistance to suppression during GWS.

Acute Care: Perioperative Planning, Coaching, and Management

For patients with CD, transsphenoidal surgery performed by an experienced surgeon achieves remission in about 80% of pituitary microadenomas and 60% of macroadenomas[28,29,30,31]. Post-operative AI and GWS are some of the most challenging phases of management for endocrinologists and one of the most disheartening for CS patients. Many patients report feeling unprepared for the postsurgical recovery process[32]. For these reasons, it is important to prepare the patient prior to surgery for the difficult months ahead, and the same considerations apply to the commencement of medical therapies, as will be discussed later. On the one hand, more potent glucocorticoids and higher doses reliably mitigate symptoms, but on the other hand, substitution of exogenous for endogenous CS delays recovery of the HPA axis and perpetuates CS-related co-morbidities. Limited data that compare management strategies preclude evidence-based decisions, yet some themes can be derived from expert opinion and extensive experience from CS centers.

In centers dedicated to the management of CS, surgeons and endocrinologists work closely together through all phases of the process. Although the goal of primary surgery for CD is adenoma resection, the tumor might not be found and/or removed completely after initial exploration. To prepare for this possibility, the surgeon should determine in advance with the patient and endocrinologist what to do next in this situation – dissect further, perform a hypophysectomy or hemi-hypophysectomy, or stop the operation. The plan for perioperative testing and glucocorticoid treatment varies widely among centers. The conundrum faced in the immediate perioperative period is that withholding glucocorticoids allows for rapid testing and demonstration of remission; however, complete resection of the causative tumor causes AI from prolonged suppression of the HPA axis and concerns for acute decompensation. Abundant evidence has shown that post-pituitary adenomectomy patients are not at risk for an adrenal crisis when monitored closely in an intensive care unit or equivalent setting[33]. Many studies have confirmed that post-operative AI almost always suggests a remission of CD[15,16,17,1834]. A standard protocol includes securing serum electrolytes and cortisol, plasma ACTH, capillary blood glucose, blood pressure, and urine specific gravity every 6 h for 24–48 h while withholding all glucocorticoids. Consecutive serum cortisol values less than 2–5 µg/dL (we use < 3 µg/dL) are sufficient to document successful tumor resection and to begin glucocorticoid therapy[35]. Post-operative signs and symptoms of AI including vomiting, hyponatremia, hypoglycemia, and hypotension should also mandate immediate glucocorticoid support. Although not clinically useful in the immediate post-operative period, some investigators have shown that low ACTH and DHEAS levels may be better predictors of long-term remission than serum cortisol[36]. A similar strategy for the management of possible post-operative AI/GWS following unilateral adrenalectomy for nodular adrenal disease has recently been reported. A post-operative day 1 basal cortisol and its response to cosyntropin stimulation can reliably segregate those patients with HPA axis suppression requiring cortisol replacement from those with an intact HPA axis who do not need to be discharged with glucocorticoid therapy[37].

Once remission is achieved, exogenous glucocorticoid replacement should be initiated and maintained during the months required for HPA axis recovery. Several glucocorticoids and dosing options are available (Table 1), and the initial dose is generally 3- to 4-fold higher than the physiologic range and graded based on age, comorbidities, and severity of disease. Fludrocortisone acetate should also be initiated following BLA for patients who receive glucocorticoids other than hydrocortisone, the only glucocorticoid with mineralocorticoid activity. By comparison, post-BLA patients receiving supraphysiologic hydrocortisone doses usually do not need mineralocorticoid support until their dose is tapered to near physiologic replacement. In the acute postoperative period, several medical comorbidities accompanying CS may reverse rapidly and require medication adjustments[35]. In particular, insulin and oral hypoglycemic drugs, potassium-sparing diuretics such as spironolactone, and other cardiovascular drugs are typically tapered or discontinued as glucose counter-regulation and electrolyte balance change rapidly upon cortisol reduction. Due to the high risk of postoperative venous thromboembolism[38,39,40], prophylaxis is frequently recommended and continued for several weeks after discharge. Posterior pituitary manipulation can disturb water balance and result in serum sodium alterations, including transient or permanent central diabetes insipidus, and in rare cases the triphasic response of diabetes insipidus, followed by syndrome of inappropriate secretion of antidiuretic hormone (SIADH), and finally permanent diabetes insipidus[4142]. In the first week or two after discharge, the most common cause for readmission is hyponatremia[4344], although the mechanisms responsible for this transient SIADH state are not known. For this reason, patients should be instructed to drink only when thirsty and not as an alternative to solid foods or for social reasons for 7–10 days after the surgery. Both diabetes insipidus and SIADH may not manifest for weeks after surgery; consequently, serum sodium should be monitored after hospital discharge as well [42].

Subacute Care: The GWS and HPA Axis Recovery

When managing GWS symptoms, it is important to repeatedly emphasize to the patient that not only are GWS symptoms to be expected, but in fact these manifestations portend a favorable prognosis of sustained remission from CS. The most important treatment intervention is frequent reassurance to the patient that GWS typically resolves as the HPA axis recovers. Family members must be included in the conversation to help provide as much support as possible, as patients report that support from family and friends is the most helpful coping mechanism during the recovery process[32]. When appropriate, it may be necessary to provide the patient with temporary disability documentation, since GWS symptoms may be so severe to preclude gainful employment. The patient must know that the myalgias reflect the body’s attempts to repair the muscle damage, similar to the soreness experienced the day after resistance weight training, and these aches will eventually subside. Due to the challenges of differentiating between GWS and AI, a higher glucocorticoid dose can be briefly trialed to assess if this increased glucocorticoid exposure improves symptoms, but late-day dosing should be avoided to support recovery of the circadian rhythm. In parallel, the patient should be encouraged to adequately rest, particularly going to sleep early but limiting daytime sleep to short naps.

Several other classes of medications can be trialed to target specific patient symptoms (Table 2). Antidepressants such as fluoxetine, sertraline, and trazodone might help to improve mood, sleep and appetite. A non-steroidal anti-inflammatory medication to address the musculoskeletal discomfort might be used early in the GWS, with the cyclooxygenase type 2 (COX-2) inhibitor celecoxib (100–200 mg once or twice daily) preferred when several weeks of daily treatment is needed, generally not more than 3 months. With anorexia and reduced food intake, adequate protein intake is necessary to allow muscle recovery. Egg whites, nuts, and lean meats are nutritionally dense and generally easy to tolerate despite poor appetite.

Table 2 Pharmacotherapy and Ancillary Treatment Options for GWS Symptoms

Following surgical remission, the duration of glucocorticoid taper can vary from 6 to 12 months or more, depending on age, severity of disease, and duration of disease [4546]. Monitoring for HPA axis recovery involves both clinical and biochemical assessments. Since the HPA axis is likely to remain suppressed with prolonged supraphysiologic glucocorticoid replacement, the first goal is to shift from all-day dosing to a circadian schedule as soon as possible, such as hydrocortisone 20 mg on rising and 10 mg in the early afternoon by 2–6 weeks after surgery. The advantages of hydrocortisone include rapid absorption for symptom mitigation, the ability to measure serum cortisol as a measure of drug exposure when helpful, and the relatively short half-life [47], which ensures a glucocorticoid-free period in the early morning when it is most critical to avoid prolonged HPA axis suppression and to enhance recovery. The second goal, which should not be attempted until GWS symptoms – particularly the anorexia and myalgias – are considerably improved, is to limit replacement to a single morning dose.

Biochemical assessment should begin once patients are taking a physiologic dose of glucocorticoid replacement (total daily dose of hydrocortisone 15 to 20 mg per day) and clinically feel well enough to begin the final stage to discontinuation of glucocorticoid replacement (Fig. 2). Biochemical evaluation begins with basal testing, and dynamic assessment of adrenal function might be necessary to confirm completion of recovery. For basal testing, patients should not take their afternoon hydrocortisone dose (if prescribed) the day before testing and then have a blood draw by 0830 prior to the morning hydrocortisone dose on the day of testing. While a serum cortisol alone is adequate to taper hydrocortisone, a simultaneous plasma ACTH assists in gauging the state of HPA axis recovery. Often the ACTH and cortisol rise gradually in parallel, but sometimes the ACTH rises above the normal range despite a low cortisol, which indicates recovery of the hypothalamus (CRH neuron) and pituitary corticotrophs in advance of adrenal function. Serum DHEAS can remain suppressed for months to years after cortisol normalization, and a low DHEAS does not indicate continued HPA axis suppression. A rapid rise in DHEAS, in contrast, is concerning for disease recurrence, but a slow drift to a measurable amount in parallel with the cortisol rise is consistent with HPA axis recovery. Periodic assessment of electrolytes is prudent to screen for hyponatremia and hypo- or hyperkalemia as medications are changed, particularly diuretics. Hypercalcemia that is parathyroid-hormone independent might be observed during the recovery phase, probably related to the rise in cytokines that accompany resolution of hypercortisolemia[4849].

Fig. 2

figure 2

Glucocorticoid withdrawal algorithm. TDD, total daily dose

Basal testing is performed at 4- to 6-week intervals during glucocorticoid replacement. A rule of thumb is that the AM cortisol in µg/dL plus the morning dose of hydrocortisone in milligrams should sum to 15–20. Thus, once endogenous cortisol production is measurable, the hydrocortisone dose should be not more than 20 mg on arising. Once the AM cortisol rises to near 5 and then 10 µg/dL, the AM hydrocortisone dose is dropped to 15 and then 10 mg, respectively. Once the AM cortisol is 12–14 µg/dL, recovery is essentially complete, and the morning hydrocortisone dose is dropped to 5 mg for 4–6 weeks and then stopped or held for dynamic testing (Fig. 2). A clinical pearl related to HPA axis recovery is that patients who state that they are finally feeling better and getting over the GWS usually have started to make some endogenous cortisol, yet not enough to stop glucocorticoid tapering. Nevertheless, a smidgeon of endogenous cortisol production with the waning of GWS symptoms is a harbinger that HPA axis recovery is imminent. If basal testing is equivocal, dynamic testing might be necessary. The gold standard testing for central AI is the insulin tolerance test, which is rarely used, and metyrapone testing might be employed once the basal cortisol is > 10 µg/dL. Although designed to test for primary adrenal insufficiency, the cosyntropin stimulation test is often employed in this setting due to greater availability, simplicity, and safety than insulin or metyrapone testing. The duration of full HPA axis recovery can be highly variable depending on the individual and postoperative glucocorticoid dosing[50].

GWS During Medical Management of CS

Patients who are not surgical candidates or do not have successful remission of CS following surgery may be offered medical treatment or BLA. After BLA, the GWS will ensue without eventual recovery of the HPA axis, so glucocorticoids are tapered until a chronic physiologic replacement dose is reached as described previously. With medical management, patients might also experience GWS, particularly at the onset of treatment. Therefore, patients must be counseled that the typical symptoms of fatigue, myalgias, and anorexia are not only possible but indeed expected, rather than “side effects” of the medication, with two caveats. First, as described for glucocorticoid replacement following surgical remission, the endocrinologist must distinguish GWS from AI due to over-treatment of CS. The same parameters of vomiting, hypotension, and hypoglycemia favor inadequate cortisol exposure and the need for dose reduction or treatment pause and/or supplementation with a potent glucocorticoid such as dexamethasone to reverse an acute event. Second, known adverse effects of the specific drug in use should be considered and excluded. The quandary of distinguishing GWS from over-treatment raises an important principle of medical management: under-dose initially and gauge primarily the severity of GWS symptoms in the first several days. The initial goal of medical therapy is not to rapidly achieve normal cortisol milieu, but rather to “dial in” just enough inhibition of cortisol production or receptor antagonism to precipitate mild to moderate GWS symptoms. Once GWS symptoms appear and/or a typical dose of the medication is achieved, further assessments, including glucose, serum cortisol and/or UFC (except when treated with mifepristone), clinical appearance, and body weight are conducted while the dose is maintained constant until GWS symptoms begin to dissipate. If the patient is not experiencing adequate clinical and/or biochemical benefit from the medication in the absence of GWS symptoms, the dose is gradually raised incrementally. This iterative process might require periodic dose reduction or perhaps even temporarily discontinuing the medication if the patient’s daily living activities are affected at any point in the process.

For several medications, a block-and-replacement strategy is an option[3], particularly for very compliant patients for whom a priority is placed on avoidance of over-treatment. This strategy resembles thionamide-plus-levothyroxine therapy for the treatment of Graves disease. The patient is given both a generous dose of medication to completely block endogenous glucocorticoid production, plus simultaneous exogenous glucocorticoid therapy, titrated to replacement dose or greater. This approach allows for greater control over glucocorticoid exposure and low risk of AI, as long as the patient always takes both medications each day. Long-acting pasireotide, for example, would not be an appropriate drug for the block-and-replace strategy. Based on the drug mechanism of action, this block-and-replace strategy is feasible with ketoconazole or levoketoconazole, the 11β-hydroxylase inhibitors osilodrostat and metyrapone, and the adrenolytic agent mitotane (the latter three are off-label uses). Alternatively, the patient might be given a double replacement dose of glucocorticoid to take only if symptoms concerning for over-treatment occur, and the medical therapy for hypercortisolemia is then interrupted until the patient communicates with the endocrinologist.

Treatment monitoring with medical management includes biochemical and symptom assessment. For all medications other than mifepristone, normalization of 24-hour UFC is the minimal goal [2]. Basal morning cortisol and late-night salivary cortisol may be more challenging to interpret in the setting of diurnal rhythm loss characteristic of CS. Because mifepristone blocks glucocorticoid receptors, ACTH and cortisol increase with treatment for most forms of CS; dose titration therefore relies on assessment of clinical features, glycemia, body weight, and other metabolic parameters [2]. For occult tumors, periodic imaging to screen for a surgical target and/or tumor regrowth is prudent, and a pause in treatment for repeat surgery might be indicated.

The End Game: Comprehensive Recovery for the Patient with CS

Besides navigating the GWS and shepherding recovery of the HPA axis, recovery from co-morbidities of CS must be addressed to the extent possible. Hypertension, hyperglycemia, hypokalemia, and dyslipidemia often improve substantially but do not always resolve. Insomnia, skin thinning and bruising, and risk of thrombosis also generally resolve, and associated treatments might be discontinued. Although there is usually an improvement in bone density and decreased fracture risk following correction of CS, anabolic and/or anti-resorptive therapies may be warranted in some patients. The deformities of vertebral compression fractures may be permanent, and some authors have recommended the use of vertebroplasty for symptom relief[51]. Violaceous striae and chronic skin tears might heal with hyperpigmentation, leaving “the scars of Cushing’s,” which can persist for a lifetime. These milestones or minor victories can be used as evidence of healing and encouragement for the patient during the dark days of the GWS, and these changes herald further improvements. Fat redistribution and significant weight loss take some weeks to manifest and usually follow next.

The myopathy from CS is an example of a co-morbidity that rarely improves without targeted treatment, and the German Cushing’s Registry has provided evidence for chronic muscle dysfunction following cure of CS[52]. Recent data indicate that a low IGF-1 after curative surgery is associated with long-term myopathy [53]. This persistent myopathy is a common source of chronic fatigue following HPA axis recovery, which is unresponsive to glucocorticoids. For these reasons, an important ancillary modality is physical therapy, and an ideal time to initiate this treatment is at the first signs of HPA axis recovery when the GWS symptoms have subsided. A complete evaluation from an experienced physical therapist should focus on core and proximal muscle strength, balance, and other factors that limit function. Exercises targeting these factors (stand on one foot, sit-to-stand, straight-arm raises with 1- to 5-pound weights) rather than traditional gym exercises (arm curls, bench press, treadmill) are necessary to restore functional status and avoid frustration and injury when the patient is not yet prepared for the latter stages of recovery. Professional supervision of this initial phase is a critical component of the recovery process, and failure to attend to musculoskeletal rehabilitation – as would be routine following survival of a critical illness – risks long-term morbidities from a curable disease.

Patients with CS often complain of cognitive defects, which usually improve but may not completely recover following treatment[5455]. Glucocorticoids are toxic to the hippocampus, and both rats treated with high-dose corticosterone and patients with CD experience reductions in hippocampal volume, which does not completely return to normal even with correction of hypercortisolemia[5657]. Because the hippocampus is an important brain region for memory, the main complaint is impaired formation of new memories and recall of recent events. When significant cognitive dysfunction persists, a formal neuropsychologic testing session is prudent, both to screen for additional sources of memory loss (degenerative brain diseases) and to identify aspects that might be amenable to functional management approaches. Cognitive therapy can be effective for mental health and overall disease coping strategies as well.

Finally, for patients undergoing transsphenoidal surgery for CD, complications associated with pituitary surgeries in general should also be considered. Anterior pituitary hormone axes should be assessed biochemically and symptomatically for hypothyroidism and hypogonadism, as hypopituitarism is an independent predictor of decreased quality of life after surgical cure [58]. Hypopituitarism can not only complicate the assessment of GWS with overlapping symptoms such as fatigue, but treatment of hypopituitarism can also be important for GWS recovery. Prior to initiating physical therapy, testosterone replacement in male patients with hypogonadism should be optimized. Hypothyroidism can contribute to hyponatremia and can also slow the metabolism of glucocorticoids. Therefore, optimizing the treatment of hypothyroidism and hypogonadism prior to completing glucocorticoid taper is prudent. Growth hormone deficiency may also be evaluated in symptomatic patients in the setting of other anterior pituitary hormone deficiencies, although formal evaluation is best delayed for at least 6–12 months when HPA axis recovery has occurred or at least the glucocorticoid dose is reduced to a physiologic range [2].

Summary and Final Thoughts

After a diagnosis of CS has been well established, a multidisciplinary team of endocrinologists and surgeons must design the best treatment strategy for the patient. Expectations and possible adverse side effects of surgery or pharmacotherapy should be reviewed with the patient. The GWS is a very difficult concept for patients to understand. It seems inconceivable to them that they could possibly feel worse (and that this is a good omen) six weeks after resolution of their hypercortisolism than they do pre-operatively; however, there are no studies that address whether comprehensive pre-operative patient education regarding GWS has any impact on the patient’s post-operative perception and outcome after successful surgery. An addiction metaphor is sometimes helpful: the patient’s body and brain has become addicted to steroids (cortisol) and after steroids are abruptly reduced, their body and brain are dysphoric — much like removal of any other addictive substance (e.g., opioids, alcohol, nicotine). The patient and their care team need to know that this treatment odyssey will be a marathon, not a sprint. It may take as long as 12–18 months for patients to have full HPA axis recovery, regression of GWS, and, most importantly, resolution of the devastating effects of chronic excessive glucocorticoid exposure.

Conclusions

GWS following surgery or during medical treatment of CS can be challenging to manage. There are currently no standard guidelines for management of GWS, but various available medical and ancillary therapies are discussed here. Studies are needed to better understand the pathophysiology of GWS to guide more targeted treatments. There may be yet unrecognized steroids produced by the adrenal glands, the withdrawal of which contributes to GWS symptoms[59]. Future observational and interventional studies would be beneficial for identifying optimal management options.

References

  1. Carroll TB, Findling JW (2010) The diagnosis of Cushing’s syndrome. Rev Endocr Metab Disord 11:147–153. https://doi.org/10.1007/s11154-010-9143-3

    Article PubMed Google Scholar

  2. Fleseriu M, Auchus R, Bancos I et al (2021) Consensus on diagnosis and management of Cushing’s disease: a guideline update. Lancet Diabetes Endocrinol 9:847–875. https://doi.org/10.1016/S2213-8587(21)00235-7

    Article PubMed Google Scholar

  3. Nieman LK, Biller BMK, Findling JW et al (2015) Treatment of Cushing’s Syndrome: An Endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab 100:2807–2831. https://doi.org/10.1210/jc.2015-1818

    CAS Article PubMed PubMed Central Google Scholar

  4. Biller BMK, Grossman AB, Stewart PM et al (2008) Treatment of adrenocorticotropin-dependent Cushing’s syndrome: a consensus statement. J Clin Endocrinol Metab 93:2454–2462. https://doi.org/10.1210/jc.2007-2734

    CAS Article PubMed PubMed Central Google Scholar

  5. Geer EB, Shafiq I, Gordon MB et al (2017) BIOCHEMICAL CONTROL DURING LONG-TERM FOLLOW-UP OF 230 ADULT PATIENTS WITH CUSHING DISEASE: A MULTICENTER RETROSPECTIVE STUDY. Endocr Pract 23:962–970. https://doi.org/10.4158/EP171787.OR

    Article PubMed Google Scholar

  6. Colao A, Petersenn S, Newell-Price J et al (2012) A 12-Month Phase 3 Study of Pasireotide in Cushing’s Disease. N Engl J Med 366:914–924. https://doi.org/10.1056/NEJMoa1105743

    CAS Article PubMed Google Scholar

  7. Lacroix A, Gu F, Gallardo W et al (2018) Efficacy and safety of once-monthly pasireotide in Cushing’s disease: a 12 month clinical trial. Lancet Diabetes Endocrinol 6:17–26. https://doi.org/10.1016/S2213-8587(17)30326-1

    CAS Article PubMed Google Scholar

  8. Pivonello R, De Martino MC, Cappabianca P et al (2009) The medical treatment of Cushing’s disease: effectiveness of chronic treatment with the dopamine agonist cabergoline in patients unsuccessfully treated by surgery. J Clin Endocrinol Metab 94:223–230. https://doi.org/10.1210/jc.2008-1533

    CAS Article PubMed Google Scholar

  9. Pivonello R, Fleseriu M, Newell-Price J et al (2020) Efficacy and safety of osilodrostat in patients with Cushing’s disease (LINC 3): a multicentre phase III study with a double-blind, randomised withdrawal phase. Lancet Diabetes Endocrinol 8:748–761. https://doi.org/10.1016/S2213-8587(20)30240-0

    CAS Article PubMed Google Scholar

  10. Ceccato F, Zilio M, Barbot M et al (2018) Metyrapone treatment in Cushing’s syndrome: a real-life study. Endocrine 62:701–711. https://doi.org/10.1007/s12020-018-1675-4

    CAS Article PubMed Google Scholar

  11. Fleseriu M, Pivonello R, Elenkova A et al (2019) Efficacy and safety of levoketoconazole in the treatment of endogenous Cushing’s syndrome (SONICS): a phase 3, multicentre, open-label, single-arm trial. Lancet Diabetes Endocrinol 7:855–865. https://doi.org/10.1016/S2213-8587(19)30313-4

    CAS Article PubMed Google Scholar

  12. Castinetti F, Guignat L, Giraud P et al (2014) Ketoconazole in Cushing’s disease: is it worth a try? J Clin Endocrinol Metab 99:1623–1630. https://doi.org/10.1210/jc.2013-3628

    CAS Article PubMed Google Scholar

  13. Fleseriu M, Biller BMK, Findling JW et al (2012) Mifepristone, a glucocorticoid receptor antagonist, produces clinical and metabolic benefits in patients with Cushing’s syndrome. J Clin Endocrinol Metab 97:2039–2049. https://doi.org/10.1210/jc.2011-3350

    CAS Article PubMed Google Scholar

  14. Reincke M, Albani A, Assie G et al (2021) Corticotroph tumor progression after bilateral adrenalectomy (Nelson’s syndrome): systematic review and expert consensus recommendations. Eur J Endocrinol 184:P1–P16. https://doi.org/10.1530/EJE-20-1088

    CAS Article PubMed PubMed Central Google Scholar

  15. Lindsay JR, Oldfield EH, Stratakis CA, Nieman LK (2011) The Postoperative Basal Cortisol and CRH Tests for Prediction of Long-Term Remission from Cushing’s Disease after Transsphenoidal Surgery. J Clin Endocrinol Metab 96:2057–2064. https://doi.org/10.1210/jc.2011-0456

    CAS Article PubMed PubMed Central Google Scholar

  16. Hameed N, Yedinak CG, Brzana J et al (2013) Remission rate after transsphenoidal surgery in patients with pathologically confirmed Cushing’s disease, the role of cortisol, ACTH assessment and immediate reoperation: a large single center experience. Pituitary 16:452–458. https://doi.org/10.1007/s11102-012-0455-z

    CAS Article PubMed Google Scholar

  17. Ramm-Pettersen J, Halvorsen H, Evang JA et al (2015) Low immediate postoperative serum-cortisol nadir predicts the short-term, but not long-term, remission after pituitary surgery for Cushing’s disease. BMC Endocr Disord 15:62. https://doi.org/10.1186/s12902-015-0055-9

    CAS Article PubMed PubMed Central Google Scholar

  18. Ironside N, Chatain G, Asuzu D et al (2018) Earlier post-operative hypocortisolemia may predict durable remission from Cushing’s disease. Eur J Endocrinol 178:255–263. https://doi.org/10.1530/EJE-17-0873

    CAS Article PubMed PubMed Central Google Scholar

  19. Hochberg Z, Pacak K, Chrousos GP (2003) Endocrine Withdrawal Syndromes. Endocr Rev 24:523–538. https://doi.org/10.1210/er.2001-0014

    Article PubMed Google Scholar

  20. Dixon RB, Christy NP (1980) On the various forms of corticosteroid withdrawal syndrome. Am J Med 68:224–230. https://doi.org/10.1016/0002-9343(80)90358-7

    CAS Article PubMed Google Scholar

  21. AMATRUDA TT ND JR (1965) Certain Endocrine and Metabolic Facets of the Steroid Withdrawal Syndrome. J Clin Endocrinol Metab 25:1207–1217. https://doi.org/10.1210/jcem-25-9-1207

    Article PubMed Google Scholar

  22. Dorn LD, Burgess ES, Friedman TC et al (1997) The Longitudinal Course of Psychopathology in Cushing’s Syndrome after Correction of Hypercortisolism. J Clin Endocrinol Metab 82:912–919. https://doi.org/10.1210/jcem.82.3.3834

    CAS Article PubMed Google Scholar

  23. Chrousos GP, Gold PW (1992) The Concepts of Stress and Stress System Disorders: Overview of Physical and Behavioral Homeostasis. JAMA 267:1244–1252. https://doi.org/10.1001/jama.1992.03480090092034

    CAS Article PubMed Google Scholar

  24. Kling MA, Roy A, Doran AR et al (1991) Cerebrospinal fluid immunoreactive corticotropin-releasing hormone and adrenocorticotropin secretion in Cushing’s disease and major depression: potential clinical implications. J Clin Endocrinol Metab 72:260–271. https://doi.org/10.1210/jcem-72-2-260

    CAS Article PubMed Google Scholar

  25. Gomez MT, Magiakou MA, Mastorakos G, Chrousos GP (1993) The pituitary corticotroph is not the rate limiting step in the postoperative recovery of the hypothalamic-pituitary-adrenal axis in patients with Cushing syndrome. J Clin Endocrinol Metab 77:173–177. https://doi.org/10.1210/jcem.77.1.8392083

    CAS Article PubMed Google Scholar

  26. Young EA, Kwak SP, Kottak J (1995) Negative feedback regulation following administration of chronic exogenous corticosterone. J Neuroendocrinol 7:37–45. https://doi.org/10.1111/j.1365-2826.1995.tb00665.x

    CAS Article PubMed Google Scholar

  27. Papanicolaou DA, Tsigos C, Oldfield EH, Chrousos GP (1996) Acute glucocorticoid deficiency is associated with plasma elevations of interleukin-6: does the latter participate in the symptomatology of the steroid withdrawal syndrome and adrenal insufficiency? J Clin Endocrinol Metab 81:2303–2306. https://doi.org/10.1210/jcem.81.6.8964868

    CAS Article PubMed Google Scholar

  28. Ciric I, Zhao J-C, Du H et al (2012) Transsphenoidal surgery for Cushing disease: experience with 136 patients. Neurosurgery 70:70–80 discussion 80–81. https://doi.org/10.1227/NEU.0b013e31822dda2c

    Article PubMed Google Scholar

  29. Alexandraki KI, Kaltsas GA, Isidori AM et al (2013) Long-term remission and recurrence rates in Cushing’s disease: predictive factors in a single-centre study. Eur J Endocrinol 168:639–648. https://doi.org/10.1530/EJE-12-0921

    CAS Article PubMed Google Scholar

  30. Capatina C, Hinojosa-Amaya JM, Poiana C, Fleseriu M (2020) Management of patients with persistent or recurrent Cushing’s disease after initial pituitary surgery. Expert Rev Endocrinol Metab 15:321–339. https://doi.org/10.1080/17446651.2020.1802243

    CAS Article PubMed Google Scholar

  31. Stroud A, Dhaliwal P, Alvarado R et al (2020) Outcomes of pituitary surgery for Cushing’s disease: a systematic review and meta-analysis. Pituitary 23:595–609. https://doi.org/10.1007/s11102-020-01066-8

    Article PubMed Google Scholar

  32. Acree R, Miller CM, Abel BS et al (2021) Patient and Provider Perspectives on Postsurgical Recovery of Cushing Syndrome. J Endocr Soc 5:bvab109. https://doi.org/10.1210/jendso/bvab109

    Article PubMed PubMed Central Google Scholar

  33. AbdelMannan D, Selman WR, Arafah BM (2010) Peri-operative management of Cushing’s disease. Rev Endocr Metab Disord 11:127–134. https://doi.org/10.1007/s11154-010-9140-6

    Article PubMed Google Scholar

  34. Costenaro F, Rodrigues TC, Rollin GAF et al (2014) Evaluation of Cushing’s disease remission after transsphenoidal surgery based on early serum cortisol dynamics. Clin Endocrinol (Oxf) 80:411–418. https://doi.org/10.1111/cen.12300

    CAS Article Google Scholar

  35. Varlamov EV, Vila G, Fleseriu M (2022) Perioperative Management of a Patient With Cushing Disease. J Endocr Soc 6:bvac010. https://doi.org/10.1210/jendso/bvac010

    Article PubMed PubMed Central Google Scholar

  36. El Asmar N, Rajpal A, Selman WR, Arafah BM (2018) The Value of Perioperative Levels of ACTH, DHEA, and DHEA-S and Tumor Size in Predicting Recurrence of Cushing Disease. J Clin Endocrinol Metab 103:477–485. https://doi.org/10.1210/jc.2017-01797

    Article PubMed Google Scholar

  37. DeLozier OM, Dream SY, Findling JW et al (2022) Selective Glucocorticoid Replacement Following Unilateral Adrenalectomy for Hypercortisolism and Primary Aldosteronism. J Clin Endocrinol Metab 107:e538–e547. https://doi.org/10.1210/clinem/dgab698

    Article PubMed Google Scholar

  38. Stuijver DJF, van Zaane B, Feelders RA et al (2011) Incidence of venous thromboembolism in patients with Cushing’s syndrome: a multicenter cohort study. J Clin Endocrinol Metab 96:3525–3532. https://doi.org/10.1210/jc.2011-1661

    CAS Article PubMed Google Scholar

  39. van der Pas R, Leebeek FWG, Hofland LJ et al (2013) Hypercoagulability in Cushing’s syndrome: prevalence, pathogenesis and treatment. Clin Endocrinol (Oxf) 78:481–488. https://doi.org/10.1111/cen.12094

    CAS Article Google Scholar

  40. van der Pas R, de Bruin C, Leebeek FWG et al (2012) The hypercoagulable state in Cushing’s disease is associated with increased levels of procoagulant factors and impaired fibrinolysis, but is not reversible after short-term biochemical remission induced by medical therapy. J Clin Endocrinol Metab 97:1303–1310. https://doi.org/10.1210/jc.2011-2753

    CAS Article PubMed Google Scholar

  41. Kristof RA, Rother M, Neuloh G, Klingmüller D (2009) Incidence, clinical manifestations, and course of water and electrolyte metabolism disturbances following transsphenoidal pituitary adenoma surgery: a prospective observational study: Clinical article. J Neurosurg 111:555–562. https://doi.org/10.3171/2008.9.JNS08191

    Article PubMed Google Scholar

  42. Yuen KCJ, Ajmal A, Correa R, Little AS (2019) Sodium Perturbations After Pituitary Surgery. Neurosurg Clin 30:515–524. https://doi.org/10.1016/j.nec.2019.05.011

    Article Google Scholar

  43. Ghiam MK, Chyou DE, Dable CL et al (2021) 30-Day Readmissions and Coordination of Care Following Endoscopic Transsphenoidal Pituitary Surgery: Experience with 409 Patients. J Neurol Surg Part B Skull Base. https://doi.org/10.1055/s-0041-1729980

    Article Google Scholar

  44. Bohl MA, Ahmad S, Jahnke H et al (2016) Delayed Hyponatremia Is the Most Common Cause of 30-Day Unplanned Readmission After Transsphenoidal Surgery for Pituitary Tumors. Neurosurgery 78:84–90. https://doi.org/10.1227/NEU.0000000000001003

    Article PubMed Google Scholar

  45. Doherty GM, Nieman LK, Cutler GB et al (1990) Time to recovery of the hypothalamic-pituitary-adrenal axis after curative resection of adrenal tumors in patients with Cushing’s syndrome. Surgery 108:1085–1090

    CAS PubMed Google Scholar

  46. Sippel RS, Elaraj DM, Kebebew E et al (2008) Waiting for change: Symptom resolution after adrenalectomy for Cushing’s syndrome. Surgery 144:1054–1061. https://doi.org/10.1016/j.surg.2008.08.024

    Article PubMed Google Scholar

  47. Derendorf H, Möllmann H, Barth J et al (1991) Pharmacokinetics and Oral Bioavailability of Hydrocortisone. J Clin Pharmacol 31:473–476. https://doi.org/10.1002/j.1552-4604.1991.tb01906.x

    CAS Article PubMed Google Scholar

  48. Suzuki K, Nonaka K, Ichihara K et al (1986) Hypercalcemia in Glucocorticoid Withdrawal. Endocrinol Jpn 33:203–209. https://doi.org/10.1507/endocrj1954.33.203

    CAS Article PubMed Google Scholar

  49. Oyama Y, Iwafuchi Y, Narita I (2021) A case of hypercalcemia because of adrenal insufficiency induced by glucocorticoid withdrawal in a patient undergoing hemodialysis. CEN Case Rep. https://doi.org/10.1007/s13730-021-00619-5

    Article PubMed PubMed Central Google Scholar

  50. Berr CM, Di Dalmazi G, Osswald A et al (2015) Time to Recovery of Adrenal Function After Curative Surgery for Cushing’s Syndrome Depends on Etiology. J Clin Endocrinol Metab 100:1300–1308. https://doi.org/10.1210/jc.2014-3632

    CAS Article PubMed Google Scholar

  51. Gad HEM, Ismail AM (2020) The role of vertebroplasty in steroid-induced vertebral osteoporotic fractures. Egypt Spine J 35:41–52. https://doi.org/10.21608/esj.2020.34844.1140

    Article Google Scholar

  52. Vogel F, Braun LT, Rubinstein G et al (2020) Persisting Muscle Dysfunction in Cushing’s Syndrome Despite Biochemical Remission. J Clin Endocrinol Metab 105:e4490–e4498. https://doi.org/10.1210/clinem/dgaa625

    Article PubMed Central Google Scholar

  53. Vogel F, Braun L, Rubinstein G et al (2021) Patients with low IGF-I after curative surgery for Cushing’s syndrome have an adverse long-term outcome of hypercortisolism-induced myopathy. Eur J Endocrinol 184:813–821. https://doi.org/10.1530/EJE-20-1285

    CAS Article PubMed Google Scholar

  54. Andela CD, van Haalen FM, Ragnarsson O et al (2015) MECHANISMS IN ENDOCRINOLOGY: Cushing’s syndrome causes irreversible effects on the human brain: a systematic review of structural and functional magnetic resonance imaging studies. Eur J Endocrinol 173:R1–R14. https://doi.org/10.1530/EJE-14-1101

    CAS Article PubMed Google Scholar

  55. Bride MM, Crespo I, Webb SM, Valassi E (2021) Quality of life in Cushing’s syndrome. Best Pract Res Clin Endocrinol Metab 35:101505. https://doi.org/10.1016/j.beem.2021.101505

    CAS Article PubMed Google Scholar

  56. Starkman MN, Gebarski SS, Berent S, Schteingart DE (1992) Hippocampal formation volume, memory dysfunction, and cortisol levels in patients with Cushing’s syndrome. Biol Psychiatry 32:756–765. https://doi.org/10.1016/0006-3223(92)90079-F

    CAS Article PubMed Google Scholar

  57. McEwen BS, Gould EA, Sakai RR (1992) The Vulnerability of the Hippocampus to Protective and Destructive Effects of Glucocorticoids in Relation to Stress. Br J Psychiatry 160:18–23. https://doi.org/10.1192/S0007125000296645

    Article Google Scholar

  58. van Aken MO, Pereira AM, Biermasz NR et al (2005) Quality of Life in Patients after Long-Term Biochemical Cure of Cushing’s Disease. J Clin Endocrinol Metab 90:3279–3286. https://doi.org/10.1210/jc.2004-1375

    CAS Article PubMed Google Scholar

  59. Zorumski CF, Paul SM, Izumi Y et al (2013) Neurosteroids, stress and depression: potential therapeutic opportunities. Neurosci Biobehav Rev 37:109–122. https://doi.org/10.1016/j.neubiorev.2012.10.005

    CAS Article PubMed Google Scholar

Download references

Acknowledgements

We thank Recordati Rare Diseases for their support with literature review and figure preparation to the authors’ designs.

Funding

XH is supported by grant T32DK07245 from the National Institutes of Diabetes and Digestive and Kidney Diseases.

Author information

Affiliations

  1. Department of Internal Medicine, Division of Metabolism, Endocrinology and Diabetes, University of Michigan Medical School, Ann Arbor, MI, USA

    Xin He & Richard J. Auchus

  2. Department of Medicine, Division of Endocrinology and Molecular Medicine, Medical College of Wisconsin, Milwaukee, WI, USA

    James W. Findling

  3. Endocrinology Center and Clinics, Medical College of Wisconsin, Milwaukee, WI, USA

    James W. Findling

  4. Department of Pharmacology, University of Michigan Medical School, Ann Arbor, MI, USA

    Richard J. Auchus

  5. Lieutenant Colonel Charles S. Kettles Veterans Affairs Ann Arbor Healthcare System, Ann Arbor, MI, USA

    Richard J. Auchus

Contributions

All authors contributed to the manuscript conception, design, and content. All authors read, edited, and approved the final manuscript.

Corresponding author

Correspondence to Richard J. Auchus.

Ethics declarations

Financial Interests

Dr. Auchus has received research support from Novartis Pharmaceuticals, Corcept Therapeutics, Spruce Biosciences, and Neurocrine Biosciences and has served as a consultant for Corcept Therapeutics, Janssen Pharmaceuticals, Novartis Pharmaceuticals, Quest Diagnostics, Adrenas Therapeutics, Crinetics Pharmaceuticals, PhaseBio Pharmaceuticals, OMass Therapeutics, Recordati Rare Diseases, Strongbridge Biopharma, and H Lundbeck A/S. Dr. Findling has received research support from Novartis Pharmaceuticals and has served as a consultant for Corcept Therapeutics and Recordati Rare Diseases.

Human Subjects and Animals

No human subjects or animals were used to collect data for this manuscript.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic Supplementary Material

Below is the link to the electronic supplementary material.

About this article

Cite this article

He, X., Findling, J.W. & Auchus, R.J. Glucocorticoid Withdrawal Syndrome following treatment of endogenous Cushing Syndrome. Pituitary (2022). https://doi.org/10.1007/s11102-022-01218-y

Download citation

From https://link.springer.com/article/10.1007/s11102-022-01218-y

Levoketoconazole improves cortisol control in endogenous Cushing’s syndrome

Compared with placebo, levoketoconazole improved cortisol control and serum cholesterol levels for adults with endogenous Cushing’s syndrome, according to results from the LOGICS study presented here.

Safety and efficacy of levoketoconazole (Recorlev, Xeris Biopharma) for treatment of Cushing’s syndrome were established in the pivotal phase 3, open-label SONICS study. The phase 3, double-blind LOGICS study sought to demonstrate the drug specificity of levoketoconazole in normalizing mean urinary free cortisol (mUFC) level.

“Treatment with levoketoconazole benefited patients with Cushing’s syndrome of different etiologies and a wide range in UFC elevations at baseline by frequent normalization of UFC,” Ilan Shimon, MD, professor at the Sackler Faculty of Medicine at Tel Aviv University and associate dean of the Faculty of Medicine at Rabin Medical Center and director of the Institute of Endocrinology in Israel, told Healio. “This is a valuable Cushing’s study as it includes a placebo-controlled randomized withdrawal phase.”

LOGICS participants were drawn from a cohort of 79 adults with Cushing’s syndrome with a baseline mUFC at least 1.5 times the upper limit of normal who participated in a single-arm, open-label titration and maintenance phase of approximately 14 to 19 weeks. Researchers randomly assigned 39 of those participants plus five from SONICS who had normalized mUFC levels on stable doses of levoketoconazole for at least 4 weeks to continue to receive the medication (n = 22) or to receive placebo with withdrawal of the medication (n = 22) for 8 weeks. At the end of the withdrawal period, all participants received levoketoconazole for 8 more weeks. Primary endpoint was proportion of participants who lost mUFC normalization during the randomized withdrawal period, and secondary endpoints included proportion with normalized mUFC and changes in total and LDL cholesterol at the end of the restoration period.

During the withdrawal period, 95.5% of participants receiving placebo vs. 40.9% of those receiving levoketoconazole experienced loss of mUFC response, for a treatment difference of –54.5% (95% CI, –75.7 to –27.4; P = .0002). At the end of the withdrawal period, 4.5% of participants receiving placebo vs. 50% of those receiving levoketoconazole maintained normalized mUFC, for a treatment difference of 45.5% (95% CI, 19.2-67.9; P = .0015).

Among participants who had received placebo and lost mUFC response, 60% regained normalized mUFC at the end of the restoration period.

During the withdrawal period, participants in the placebo group had increases of 0.9 mmol/L in total cholesterol and 0.6 mmol/L in LDL cholesterol vs. decreases of 0.04 mmol/L (P = .0004) and 0.006 mmol/L (P = .0056), respectively, for the levoketoconazole group. The increases seen in the placebo group were reversed when participants restarted the medication.

The most common adverse events with levoketoconazole were nausea (29%) and hypokalemia (26%). Prespecified adverse events of special interest were liver-related (10.7%), QT interval prolongation (10.7%) and adrenal insufficiency (9.5%).

“This study has led to the FDA decision to approve levoketoconazole for the treatment of Cushing’s syndrome after surgical failure or if surgery is not possible,” Shimon said.

From https://www.healio.com/news/endocrinology/20220512/logics-levoketoconazole-improves-cortisol-control-in-endogenous-cushings-syndrome

Osilodrostat Resolves Most Physical Signs of Cushing’s Disease

— More than half of patients saw physical manifestations fully resolve by week 72

by Kristen Monaco, Staff Writer, MedPage Today May 16, 2022

SAN DIEGO — Osilodrostat (Isturisa) improved many physical features associated with Cushing’s disease, according to additional findings from the phase III LINC-3 study.

Among 137 adults with Cushing’s disease, a 39.5% improvement in central obesity scores was observed from baseline to week 72 with osilodrostat, reported Alberto Pedroncelli, MD, PhD, of Recordati AG in Basel, Switzerland.

Not only was central obesity the most common physical manifestation associated with hypercortisolism among these Cushing’s disease patients, but it was also more frequently rated as severe at baseline, Pedroncelli explained during the American Association of Clinical Endocrinology (AACE) annual meeting.

Osilodrostat treatment also led to a 34.9% improvement in proximal muscle atrophy at week 72, along with a 34.4% improvement in hirsutism scores.

By week 72, nearly all physical manifestations of hypercortisolism saw significant improvement — marked by more than 50% of patients scoring these physical traits as nonexistent:

  • Dorsal fat pat: 50.6%
  • Central obesity: 30.6%
  • Supraclavicular fat pad: 51.8%
  • Facial rubor: 64.7%
  • Hirsutism in women: 53.1%
  • Proximal muscle atrophy: 61.2%
  • Striae: 63.5%
  • Ecchymoses: 87.1%

Most of these physical manifestation improvements were notable soon after treatment initiation with osilodrostat, Pedroncelli pointed out.

When stratified according to testosterone levels, hirsutism scores remained either stable or improved in the majority of patients who had normal or above normal testosterone levels. More women with normal testosterone levels over time experienced improvements in hirsutism versus those with levels above the upper limit of normal, who mostly remained stable.

Osilodrostat is an oral agent that was first FDA approved in March 2020 for adults with Cushing’s disease who either cannot undergo pituitary gland surgery or have undergone the surgery but still have the disease. Available in 1 mg, 5 mg, and 10 mg film-coated tablets, the drug acts as a potent oral 11-beta-hydroxylase inhibitor — the enzyme involved in the last step of cortisol synthesis.

Osilodrostat is taken orally twice daily, once in the morning and once in the evening.

Approval was based upon findings from the LINC-3 and LINC-4 trials, which found osilodrostat was able to normalize cortisol levels in 53% of patients, based on mean 24-hour urinary free cortisol (UFC) concentrations. During an initial 10-week randomization phase, 86% of patients maintained their complete cortisol response if they remained on osilodrostat versus only 29% of those who were switched to placebo.

As expected, 77.4% of the 137 adults included in the trial were women. The median participant age was 40 and about 47 months had passed since their initial diagnosis. A total of 87.6% underwent previous pituitary surgery and 16.1% underwent previous pituitary irradiation. At baseline, median and mean 24-hour UFC levels were 3.5 nmol and 7.3 nmol, respectively, based on two or three urine samples.

Participants had an average body weight of 176.4 lb, body mass index (BMI) of 30, and 41 in waist circumference at baseline. Throughout the trial, all measures dropped, reaching the nadir at week 72: body weight of 165 lb, BMI of 27, and 37.8 in waist circumference.

The most common side effects reported with the agent include adrenal insufficiency, fatigue, nausea, headache, and edema.

  • author['full_name']

    Kristen Monaco is a staff writer, focusing on endocrinology, psychiatry, and nephrology news. Based out of the New York City office, she’s worked at the company since 2015.

Disclosures

The study was supported by Recordati AG.

Pedroncelli reported employment with Recordati.

Copeptin Levels Before and After Transsphenoidal Surgery for Cushing Disease: A Potential Early Marker of Remission

Abstract

Context

Arginine-vasopressin and CRH act synergistically to stimulate secretion of ACTH. There is evidence that glucocorticoids act via negative feedback to suppress arginine-vasopressin secretion.

Objective

Our hypothesis was that a postoperative increase in plasma copeptin may serve as a marker of remission of Cushing disease (CD).

Design

Plasma copeptin was obtained in patients with CD before and daily on postoperative days 1 through 8 after transsphenoidal surgery. Peak postoperative copeptin levels and Δcopeptin values were compared among those in remission vs no remission.

Results

Forty-four patients (64% female, aged 7-55 years) were included, and 19 developed neither diabetes insipidus (DI) or syndrome of inappropriate anti-diuresis (SIADH). Thirty-three had follow-up at least 3 months postoperatively. There was no difference in peak postoperative copeptin in remission (6.1 pmol/L [4.3-12.1]) vs no remission (7.3 pmol/L [5.4-8.4], P = 0.88). Excluding those who developed DI or SIADH, there was no difference in peak postoperative copeptin in remission (10.2 pmol/L [6.9-21.0]) vs no remission (5.4 pmol/L [4.6-7.3], P = 0.20). However, a higher peak postoperative copeptin level was found in those in remission (14.6 pmol/L [±10.9] vs 5.8 (±1.4), P = 0.03]) with parametric testing. There was no difference in the Δcopeptin by remission status.

Conclusions

A difference in peak postoperative plasma copeptin as an early marker to predict remission of CD was not consistently present, although the data point to the need for a larger sample size to further evaluate this. However, the utility of this test may be limited to those who develop neither DI nor SIADH postoperatively.

Arginine vasopressin (AVP) and CRH act synergistically as the primary stimuli for secretion of ACTH, leading to release of cortisol [12]. The role of AVP in the hypothalamic-pituitary-adrenal (HPA) axis is via release from the parvocellular neurons of the paraventricular nuclei (and possibly also from the magnocellular neurons of the paraventricular and supraoptic nuclei), the secretion of which is stimulated by stress [3-6]. AVP release results in both independent stimulation of ACTH release and potentiation of the effects of CRH [37-9]. Additionally, there is evidence that glucocorticoids act by way of negative feedback to suppress AVP secretion [1011-20]. Further, parvocellular neurons of the hypothalamic paraventricular nuclei have been shown to increase AVP production and neurosecretory granule size after adrenalectomy, and inappropriately elevated plasma AVP has been reported in the setting of adrenal insufficiency with normalization of plasma AVP after glucocorticoid administration [21-24]. This relationship of AVP and its effect on the HPA axis has been used in the diagnostic evaluation of Cushing syndrome (CS) [14] and evaluation of remission after transsphenoidal surgery (TSS) in Cushing disease (CD) by administration of desmopressin [25].

Copeptin makes up the C-terminal portion of the AVP precursor pre-pro-AVP. Copeptin is released from the posterior pituitary in stoichiometric amounts with AVP, and because of its longer half-life in circulation, it is a stable surrogate marker of AVP secretion [26-28]. Plasma copeptin has been studied in various conditions of the anterior pituitary. In a study by Lewandowski et al, plasma copeptin was measured after administration of CRH in assessment of HPA-axis function in patients with a variety of pituitary diseases. An increase in plasma copeptin was observed only in healthy subjects but not in those with pituitary disease who had an appropriately stimulated serum cortisol, and the authors concluded that copeptin may be a sensitive marker to reveal subtle alterations in the regulation of pituitary function [7]. Although in this study and others, plasma copeptin was assessed after pituitary surgery, it has not, to the best of our knowledge, been studied as a marker of remission of CD before and after pituitary surgery [729].

In this study, plasma copeptin levels were assessed as a surrogate of AVP secretion before and after TSS for treatment of CD. Because there is evidence that glucocorticoids exert negative feedback on AVP, we hypothesized that there would be a greater postoperative increase in plasma copeptin in those with CD in remission after TSS resulting from resolution of hypercortisolemia and resultant hypocortisolemia compared with those not in remission with persistent hypercortisolemia and continued negative feedback. In other words, we hypothesized that an increase in copeptin could be an early marker of remission of CD after TSS. We aimed to complete this assessment by comparison of the peak postoperative copeptin and change in copeptin from preoperative to peak postoperative copeptin for those in remission vs not in remission postoperatively.

Subjects and Methods

Subjects

Adult and pediatric patients with CD who presented at the Eunice Kennedy Shriver National Institute of Child Health and Human Development under protocol 97-CH-0076 and underwent TSS between March 2016 and July 2019 were included in the study. Exclusion criteria included a prior TSS within 6 weeks of the preoperative plasma copeptin sample or a preoperative diagnosis of diabetes insipidus, renal disease, or cardiac failure. Written informed consent was provided by patients aged 18 years and older and by legal guardians for patients aged < 18 years to participate in this study. Written informed assent was provided by patients aged 7 years to < 18 years. The 97-CH-0076 study (Investigation of Pituitary Tumors and Related Hypothalamic Disorders) has been approved by the Eunice Kennedy Shriver National Institute of Child Health and Human Development institutional review board.

Clinical and Biochemical Data

Clinical data were extracted from electronic medical records. Age, sex, body weight, body mass index (BMI), pubertal stage (in pediatric patients only), and history of prior TSS were obtained preoperatively during the admission for TSS. Clinical data obtained postoperatively included TSS date, histology, development of central diabetes insipidus (DI) or (SIADH), time from TSS to most recent follow-up, and clinical remission status at postoperative follow-up.

Preoperatively, serum sodium, 24-hour urinary free cortisol (UFC), UFC times the upper limit of normal (UFC × ULN), midnight (MN) serum cortisol, MN plasma ACTH, and 8 AM plasma ACTH were collected. Postoperatively, serum sodium, serum and urine osmolality, urine specific gravity, serum cortisol, and plasma ACTH were collected. For serum cortisol values < 1 mcg/dL, a value of 0.5 mcg/dL was assigned for the analyses; for plasma ACTH levels < 5 pg/mL, a value of 2.5 pg/mL was assigned.

Additionally, plasma copeptin levels were obtained preoperatively and on postoperative days (PODs) 1 through 8 after TSS at 8:00 AM. Peak postoperative copeptin was the highest plasma copeptin on PODs 1 through 8. The delta copeptin (Δcopeptin) was determined by subtracting the preoperative copeptin from the peak postoperative copeptin; hence, a positive change indicated a postoperative increase in plasma copeptin. Plasma copeptin was measured using an automated immunofluorescent sandwich assay on the BRAHMS Kryptor Compact PLUS Copeptin-proAVP. The limit of detection for the assay was 1.58 pmol/L, 5.7% intra-assay coefficient of variation, and 11.2% inter-assay coefficient of variation, with a lower limit of analytical measurement of 2.8 pmol/L. For those with multiple preoperative plasma copeptin values within days before surgery, an average of preoperative copeptin levels was used for analyses.

Diagnosis of CD was based on guidelines published by the Endocrine Society and as previously described for the adult and pediatric populations [3031]; diagnosis was further confirmed by either histologic identification of an ACTH-secreting pituitary adenoma in the resected tumor specimen, decrease in cortisol and ACTH levels postoperatively, and/or clinical remission after TSS at follow-up evaluation. All patients were treated with TSS at the National Institutes of Health Clinical Center by the same neurosurgeon. Remission after surgical therapy was based on serum cortisol of < 5 μg/dL during the immediate postoperative period, improvement of clinical signs and symptoms of cortisol excess at postoperative follow up, nonelevated 24-hour UFC at postoperative follow-up, nonelevated midnight serum cortisol at postoperative follow up when available, and continued requirement for glucocorticoid replacement at 3 to 6 months’ postoperative follow-up.

Diagnosis of SIADH was based on development of hyponatremia (serum sodium < 135 mmol/L) and oliguria (urine output < 0.5 mL/kg/h). Diagnosis of DI was determined by development of hypernatremia (serum sodium > 145 mmol/L), dilute polyuria (urine output > 4 mL/kg/h), elevated serum osmolality, and low urine osmolality.

Statistical Analyses

Results are presented as median (interquartile range [IQR], calculated as 25th percentile-75th percentile) or mean ± SD, as appropriate, and frequency (percentage). Where appropriate, we compared results using parametric or nonparametric testing; however, the median (IQR) and the mean ± SD were both reported to allow for comparisons with the appropriate testing noted. Subgroup analyses were completed comparing those who developed water balance disorders included patients who developed DI only (but not SIADH), those who developed SIADH only (but not DI), and those with no water balance disorder; hence, for these subgroup analyses, those who developed both DI and SIADH postoperatively (n = 4) were excluded. Preoperative copeptin, peak postoperative copeptin, and Δcopeptin were compared between those with and without remission at follow-up, using either t test or Wilcoxon rank-sum test, depending on the distribution of data. These were done in all patients combined, as well as within each subgroup. The same tests were used for comparing other continuous variables (eg, age, BMI SD score [SDS], cortisol excess measures) between those with and without remission. Categorical data (eg, sex, Tanner stage) were analyzed using the Fisher exact test. Comparisons of copeptin levels among the subgroups (DI, SIADH, neither) were carried out using mixed models and the Kruskal-Wallis test, as appropriate. Post hoc pairwise comparisons were adjusted for multiplicity using the Bonferroni correction, and as applicable, only corrected P values are reported. Mixed models for repeated measures also analyzed copeptin, serum sodium, and cortisol data for PODs 1 through 8. In addition, maximum likelihood estimation (GENMOD) procedures analyzed the effects of copeptin and serum sodium on the remission at follow-up. Correlation analyses were done with Spearman ρ. All analyses were tested for the potential confounding effects of age, sex, BMI SDS, and pubertal status, and were adjusted accordingly. For plasma copeptin reported as < 2.8 pmol/L, a value of 1.4 pmol/L (midpoint of 0 and 2.8 pmol/L) was used; sensitivity analyses repeated all relevant comparisons using the threshold limit of 2.8 pmol/L instead of 1.4 pmol/L. Odds ratios (OR) and 95% CIs, other magnitudes of the effect, data variability, and 2-sided P values provided the statistical evidence for the conclusions. Statistical analyses were performed in SAS version 9.4 software (SAS Institute, Inc, Cary, NC).

Results

Patient Characteristics

Forty-four adult and pediatric patients, aged 7 to 55 years (77.2% were < 18 years old), with CD were included in the study. The cohort included 28 female patients (64%), and the median BMI SDS was 2.2 (1.1-2.5). Thirty-four percent (15/44) had prior pituitary surgery (none within the prior 6 weeks). Seventy-five percent (33/44) had postoperative follow-up evaluations available, with median follow-up of 13.5 months (11.3-16.0). Of those 33 patients, 85% were determined to be in remission at follow-up. Comparing those in remission vs no remission, there was no difference in age, sex, BMI SDS, pubertal status (in pediatric ages only), preoperative measures of cortisol excess (UFC × ULN, PM serum cortisol, MN plasma ACTH, AM plasma ACTH), duration of follow-up, or development of DI or SIADH. There was a lower postoperative serum cortisol nadir in those in remission at follow-up compared with those not in remission at follow-up, as expected, because a postoperative serum cortisol < 5 μg/dL was included in defining remission status. Postoperatively, 8/44 (18%) developed DI, 13/44 (30%) developed SIADH, 4/44 (9%) developed both DI and SIADH, and 19/44 (43%) developed no water balance disorder (Table 1). There were no differences by remission status when assessing these subgroups (ie, DI, SIADH, and no water balance disorder) separately.

 

Table 1.

Demographic and clinical characteristics of subjects

All subjects, n = 44 All subjects by remission status, n = 33 All subjects by remission status, excluding those with DI or SIADH, n = 13
Remission, n = 28 No remission, n = 5 P Remission,
n = 10
No remission, n = 3 P
Age, median (range), y 14.5 (7-55) 17.4 ± 10.7
14.5 (12.5-17.5)
15.6 ± 13.2
11.0 (9.0-12.0)
0.11 13.7 ± 3.1
14.0 (13.0-15.0)
19.7 ± 16.8
11.0 (9.0-39.0)
0.60a
Sex
Female
28 (64%) 22 (78.6%) 3 (60.0%) 0.57 9 (90.0%) 2 (66.7%) 0.42
BMI SDS 2.2 (1.1-2.5) 1.7 ± 1.0
2.0 (0.9-2.5)
2.2 ± 0.4
2.2 (2.1-2.3)
0.70 1.7 ± 1.1
2.0 (0.7-2.5)
2.0 ± 0.4
2.1 (1.5-2.3)
0.65a
Pubertal status
Female (n = 19) (n = 15) (n = 2) 0.51 (n = 8) (n = 1) 0.44
  Tanner 1-2 6 4 (26.7%) 1 (50.0%) 3 (37.5%) 1 (25.0%)
  Tanner 3-5 13 11 (73.3%) 1 (50.0%) 5 (62.5%) 0
Male (n = 14) (n = 5) (n = 2) (n = 1) (n = 1)
Testicular volume < 12, mL 10 4 (80.0%) 2 (10.00%) 1 (100.0%) 1 (100.0%)
Testicular volume ≥ 12, mL 4 1 (20.0%) 0 1.0 0 0
Preoperative UFC ULN 3.3 (1.2-6.1) 4.9 ± 6.1
2.6 (1.0-7.6)
3.2 ± 1.3
3.7 (3.0-3.9)
0.70 7.2 ± 8.4
3.9 (1.8-9.1)
3.8 ± 0.7
3.9 (3.0-4.4)
0.93
Preoperative PM cortisol 11.9 (9.2-14.8) 13.3 ± 4.7
12.2 (9.2-16.8)
10.8 ± 2.1
11.5 (9.0-11.6)
0.30 13.3 ± 6.0
11.2 (8.4-16.5)
11.1 ± 2.6
11.6 (8.3-13.6)
0.57a
Preoperative MN ACTH 43.4 (29.3-51.6) 44.2 ± 25.5
46.1 (27.6-50.5)
40.9 ± 15.3
11.5 (9.0-11.6)
0.74 36.6 ± 16.6
37.4 (29.1-48.8)
34.0 ± 9.4
39.3 (23.1-39.5)
0.67
Preoperative AM ACTH 44.6 (31.4-60.5) 46.9 ± 28.9
44.0 (29.8-56.2)
48.6 ± 28.8
58.7 (21.7-60.5)
0.84 35.2 ± 16.2
40.3 (28.0-44.0)
45.4 ± 24.6
58.7 (17.0-60.5)
0.41a
Postoperative cortisol nadir 0.5 (0.5-0.5) 0.7 ± 0.7
0.5 (0.5-0.5)
7.8 ± 6.6
5.2 (2.2-12.3)
<0.001 0.6 ± 0.3
0.5 (0.5-0.5)
8.1 ± 7.9
5.2 (2.1-17.0)
0.003
Duration of follow-up 13.5 (11.3-16.0) 15.3 ± 7.9
14.0 (12.0-16.5)
14.0 ± 13.0
11.0 (6.0-14.0)
0.30 18.6 ± 11.2
15.5 (12.0-27.0)
16.7 ± 17.2
11.0 (3.0-36.0)
0.82a
DI only 8 (18%) 7/8 (87.5%) 1/8 (12.5%) 0.91
SIADH only 13 (30%) 8/9 (88.9%) 1/9 (11.1%)
Neither DI/SIADH 19 (43%) 10/13 (76.9%) 3/13 (23.1%)
Both DI and SIADH 4 (9%) 3/3 (100%) 0/3

Demographic and clinical characteristics of all subjects (n = 44) with Cushing disease. Data are also presented by remission status for all subjects with postoperative follow-up (n = 33) and by remission status after excluding those who developed DI or SIADH postoperatively with postoperative follow-up (n = 13). Both median (IQR) and mean ± SD reported to allow for comparisons, with P value provided using appropriate testing depending on distribution of data sets. Data are mean ± SD, median (25th-75th IQR), or frequency (percentage) are reported, except for age, which is presented as median (range).

Abbreviations: AM, 7:30-8 PM; BMI, body mass index; DI, diabetes insipidus; IQR, interquartile range; MN, midnight; N/A, not applicable; SDS, SD score; SIADH, syndrome of inappropriate antidiuresis; UFC, urinary free cortisol; ULN, upper limit of normal. p-values below the threshold of 0.05 are in bold.

aP value indicates comparison using parametric testing, as appropriate for normally distributed data.

Preoperative copeptin levels were higher in males (7.0 pmol/L [5.1-9.6]) than in females (4.0 pmol/L [1.4-5.8], P = 0.004) (Fig. 1). Age was inversely correlated with preoperative copeptin (rs = -0.35, P = 0.030) and BMI SDS was positively correlated with preoperative copeptin (rs = 0.54, P < 0.001) (Fig. 2).

 

Figure 1.

Preoperative plasma copeptin and sex. Preoperative plasma copeptin in all patients, comparing by sex. A higher preoperative plasma copeptin was found in males (7.0 pmol/L [5.1-9.6]) than in females (4.0 pmol/L [1.4-5.8], P = 0.004). Horizontal lines = median. Whiskers = 25th and 75th interquartile ranges.

Preoperative plasma copeptin and sex. Preoperative plasma copeptin in all patients, comparing by sex. A higher preoperative plasma copeptin was found in males (7.0 pmol/L [5.1-9.6]) than in females (4.0 pmol/L [1.4-5.8], P = 0.004). Horizontal lines = median. Whiskers = 25th and 75th interquartile ranges.

 

Figure 2.

Preoperative plasma copeptin and BMI SDS. Association of preoperative plasma copeptin and BMI SDS in all patients. A BMI SDS was positively associated with a preoperative plasma copeptin (rs = 0.54, P < 0.001). Shaded area = 95% confidence interval.

Preoperative plasma copeptin and BMI SDS. Association of preoperative plasma copeptin and BMI SDS in all patients. A BMI SDS was positively associated with a preoperative plasma copeptin (rs = 0.54, P < 0.001). Shaded area = 95% confidence interval.

Copeptin Before and After Transsphenoidal Surgery for CD

Among the 33 patients with postoperative follow-up, there was no difference in peak postoperative copeptin for patients in remission vs those not in remission (6.1 pmol/L [4.3-12.1] vs 7.3 pmol/L [5.4-8.4], P = 0.88). There was also no difference in the Δcopeptin for those in remission vs not in remission (2.3 pmol/L [-0.5 to 8.2] vs 0.1 pmol/L [-0.1 to 2.2], P = 0.46) (Fig. 3). Including all subjects, the mean preoperative copeptin was 5.6 pmol/L (±3.4). For patients with follow-up, there was no difference in preoperative copeptin for those in remission (4.8 pmol/L [±2.9]) vs no remission (6.0 pmol/L [±2.0], P = 0.47). POD 1 plasma copeptin ranged from < 2.8 to 11.3 pmol/L.

 

Figure 3.

(A) Peak postoperative plasma copeptin in all patients, comparing those in remission with no remission (6.1 pmol/L [4.3-12.1] vs 7.3 pmol/L [5.4-8.4], P = 0.88). (B) ΔCopeptin (preoperative plasma copeptin subtracted from postoperative peak plasma copeptin) in all patients, comparing those in remission with no remission (2.3 pmol/L [-0.5 to 8.2] vs 0.1 pmol/L [-0.1 to 2.2], P = 0.46). Horizontal lines = median. Whiskers = 25th and 75th interquartile ranges.

(A) Peak postoperative plasma copeptin in all patients, comparing those in remission with no remission (6.1 pmol/L [4.3-12.1] vs 7.3 pmol/L [5.4-8.4], P = 0.88). (B) ΔCopeptin (preoperative plasma copeptin subtracted from postoperative peak plasma copeptin) in all patients, comparing those in remission with no remission (2.3 pmol/L [-0.5 to 8.2] vs 0.1 pmol/L [-0.1 to 2.2], P = 0.46). Horizontal lines = median. Whiskers = 25th and 75th interquartile ranges.

When those who developed DI or SIADH were excluded, there was no difference in peak postoperative copeptin in those in remission vs no remission (10.2 pmol/L [6.9-21.0] vs 5.4 pmol/L [4.6-7.3], P = 0.20). However, because the distribution of the peak postoperative copeptins was borderline normally distributed, parametric testing was also completed for this analysis, which showed a higher peak postoperative copeptin in remission (14.6 pmol/L [±10.9]) vs no remission (5.8 [±1.4], P = 0.03). There was no difference in the Δcopeptin for those in remission vs not in remission (5.1 pmol/L [0.3-19.5] vs 1.1 pmol/L [-0.1 to 2.2], P = 0.39) (Fig. 4). Preoperative copeptin was not different for those in remission (4.7 pmol/L [±2.4]) vs no remission (4.9 pmol/L [±20.3], P = 0.91). There was no association between serum cortisol and plasma copeptin over time postoperatively (Fig. 5).

 

Figure 4.

(A) Peak postoperative plasma copeptin excluding those who developed DI or SIADH, comparing those in remission with no remission (10.2 pmol/L [6.9-21.0] vs 5.4 pmol/L [4.6-7.3], P = 0.20). (B) ΔCopeptin (preoperative plasma copeptin subtracted from postoperative peak plasma copeptin) excluding those who developed DI or SIADH, comparing those in remission with no remission (5.1 pmol/L [0.3-19.5] vs 1.1 pmol/L [-0.1 to 2.2], P = 0.39). Horizontal lines = median. Whiskers = 25th and 75th interquartile ranges.

(A) Peak postoperative plasma copeptin excluding those who developed DI or SIADH, comparing those in remission with no remission (10.2 pmol/L [6.9-21.0] vs 5.4 pmol/L [4.6-7.3], P = 0.20). (B) ΔCopeptin (preoperative plasma copeptin subtracted from postoperative peak plasma copeptin) excluding those who developed DI or SIADH, comparing those in remission with no remission (5.1 pmol/L [0.3-19.5] vs 1.1 pmol/L [-0.1 to 2.2], P = 0.39). Horizontal lines = median. Whiskers = 25th and 75th interquartile ranges.

 

Figure 5.

Plasma copeptin and serum cortisol vs postoperative day for patients who did not develop DI or SIADH. Plasma copeptin (indicated by closed circle) and serum cortisol (indicated by “x”). Results shown as (median, 95% CI).

Plasma copeptin and serum cortisol vs postoperative day for patients who did not develop DI or SIADH. Plasma copeptin (indicated by closed circle) and serum cortisol (indicated by “x”). Results shown as (median, 95% CI).

All analyses here were repeated adjusting for serum sodium, and there were no differences by remission status for preoperative, peak postoperative, or Δcopeptin for all subjects or after excluding those who developed a water balance disorder (data not shown).

Copeptin and Water Balance Disorders

As expected, peak postoperative copeptin appeared to be different among patients who developed DI, SIADH, and those without any fluid balance disorder (P = 0.029), whereas patients with DI had lower median peak postoperative copeptin (4.4 pmol/L [2.4-6.9]) than those who developed no fluid abnormality (10.0 pmol/L [5.4-16.5], P = 0.04), the statistical difference was not present after correction for multiple comparisons (P = 0.13). Peak postoperative copeptin of patients with SIADH was 9.4 pmol/L (6.5-10.4) and did not differ from patients with DI (P = 0.32) or those with no fluid abnormality (P = 1.0). There was a difference in Δcopeptin levels among these subgroups (overall P = 0.043), which appeared to be driven by the lower Δcopeptin in those who developed DI (-1.2 pmol/L [-2.6 to 0.1]) vs in those with neither DI or SIADH (3.1 pmol/L [0-9.6], P = 0.05). However, this pairwise comparison did not reach statistical significance, even before correction for multiple comparisons (P = 0.16) (Fig. 6). Preoperative copeptin levels were also not different among the subgroups (P = 0.54).

 

Figure 6.

(A) Peak postoperative plasma copeptin, comparing those who developed DI, SIADH, or neither (P = 0.029 for comparison of all 3 groups). (B) ∆ Copeptin (preoperative plasma copeptin subtracted from postoperative peak plasma copeptin), comparing those who developed DI, SIADH, or neither (P = 0.043 for comparison of all 3 groups). Horizontal lines = median. Whiskers = 25th and 75th interquartile ranges. Top brackets = pairwise comparisons. P values presented are after Bonferroni correction for multiple comparisons.

(A) Peak postoperative plasma copeptin, comparing those who developed DI, SIADH, or neither (P = 0.029 for comparison of all 3 groups). (B) ∆ Copeptin (preoperative plasma copeptin subtracted from postoperative peak plasma copeptin), comparing those who developed DI, SIADH, or neither (P = 0.043 for comparison of all 3 groups). Horizontal lines = median. Whiskers = 25th and 75th interquartile ranges. Top brackets = pairwise comparisons. P values presented are after Bonferroni correction for multiple comparisons.

Association of Sodium and Copeptin

Longitudinal data, adjusting for subgroups (ie, DI, SIADH, neither), were analyzed. As expected, there was a group difference (P = 0.003) in serum sodium over time (all DI was missing preoperative serum sodium), with the difference being driven by DI vs SIADH (P = 0.007), and SIADH vs neither (P = 0.012). There was no group difference in plasma copeptin over POD by water balance status (P = 0.16) over time (Fig. 7). There was also no effect by remission status at 3 to 6 months for either serum sodium or plasma copeptin.

 

Figure 7.

(A) Serum sodium and (B) plasma copeptin by POD and water balance status longitudinal data, adjusting for subgroups (ie, DI, SIADH, neither). Data points at point 0 on the x-axis indicate preoperative values. As expected, there was a group difference (P = 0.003) in serum sodium over time (all with DI were missing preoperative serum sodium), with the difference being driven by DI vs SIADH (P = 0.007), and SIADH vs neither (P = 0.012). There was no group difference in plasma copeptin over POD by water balance status (P = 0.16) over time.

(A) Serum sodium and (B) plasma copeptin by POD and water balance status longitudinal data, adjusting for subgroups (ie, DI, SIADH, neither). Data points at point 0 on the x-axis indicate preoperative values. As expected, there was a group difference (P = 0.003) in serum sodium over time (all with DI were missing preoperative serum sodium), with the difference being driven by DI vs SIADH (P = 0.007), and SIADH vs neither (P = 0.012). There was no group difference in plasma copeptin over POD by water balance status (P = 0.16) over time.

Higher serum sodium levels from PODs 1 through 8 itself decreased the odds of remission (OR, 0.56; 95% CI, 0.42-0.73; P < 0.001) in all CD patients. Copeptin levels from these repeated measures adjusting for serum sodium did not correlate with remission status at 3 to 6 months’ follow-up (P = 0.38). There were no differences in preoperative, peak postoperative, or delta sodium levels by remission vs no remission in all patients and in those with no water balance disorders.

Discussion

AVP and CRH act synergistically to stimulate the secretion of ACTH and ultimately cortisol [12], and there is evidence that glucocorticoids act by way of negative feedback to suppress AVP secretion [1011-20]. Therefore, we hypothesized that a greater postoperative increase in plasma copeptin in those with CD in remission after TSS because of resolution of hypercortisolemia and resultant hypocortisolemia, compared with those not in remission with persistent hypercortisolemia and continued negative feedback, would be observed. Although a clear difference in peak postoperative and Δcopeptin was not observed in this study, a higher peak postoperative copeptin was found in those in remission after excluding those who developed DI/SIADH when analyzing this comparison with parametric testing, and it is possible that we did not have the power to detect a difference by nonparametric testing, given our small sample size. Therefore, postoperative plasma copeptin may be a useful early marker to predict remission of CD after TSS. The utility of this test may be limited to those who do not develop water balance disorders postoperatively. If a true increase in copeptin occurs for those in remission after treatment of CD, it is possible that this could be due to the removal of negative feedback from cortisol excess on pre-pro-AVP secretion, as hypothesized in this study. However, it is also possible that other factors may contribute to an increase in copeptin postoperatively, including from the stress response of surgery and postoperative hypocortisolism and resultant stimulation of pre-pro-AVP secretion from these physical stressors and/or from unrecognized SIADH.

It was anticipated that more severe hypercortisolism to be negatively correlated with preoperative plasma copeptin because of greater negative feedback on AVP. However, no association was found between preoperative plasma copeptin and markers of severity of hypercortisolism (MN cortisol, AM ACTH, UFC × ULN) in this study. Similarly, we would expect that the preoperative plasma copeptin would be lower compared with healthy individuals. However, comparisons of healthy individuals may be difficult because the fluid and osmolality status at the time of the sample could influence the plasma copeptin, and depending on those factors, copeptin could be appropriately low. A healthy control group with whom to compare the preoperative values was not available for this study, and the thirsted state was not standardized for the preoperative copeptin measurements. Future studies could be considered to determine if preoperative plasma copeptin is lower in patients with CD, or other forms of CS, compared with healthy subjects, with all subjects thirsted for an equivalent period. Further, if preoperative plasma copeptin is found to be lower in thirsted subjects with CS than a thirsted healthy control group, the plasma copeptin could potentially be a diagnostic test to lend support for or against the diagnosis of endogenous CS.

In the comparisons of those who developed DI, SIADH, or neither, no difference was found in the Δcopeptin. Peak copeptin was lower in DI compared with those without DI or SIADH (but not different from SIADH). Again, it is possible that there is a lower peak postoperative copeptin and change in copeptin in those with DI, but we may not have had the power to detect this in all of our analyses. These comparisons of copeptin among those with or without water balance disorders postoperatively are somewhat consistent with a prior study showing postoperative copeptin as a good predictor of development of DI, in which a plasma copeptin < 2.5 pmol/L measured on POD 0 accurately identified those who developed DI, and plasma copeptin > 30 pmol/L ruled out the development of DI postoperatively [29]. In the current study, 3 of 6 subjects with DI had a POD 1 plasma copeptin < 2.5 pmol/L, and none had a POD 1 plasma copeptin > 30 pmol/L. However, the study by Winzeler et al found that copeptin measured on POD 0 (within 12 hours after surgery) had the greatest predictive value, and POD 0 plasma copeptin was not available in our study. Further, we used the preoperative, peak, and delta plasma copeptin for analyses, so the early low copeptin levels may not have been captured in our data and analyses.

Additionally, this study revealed that increasing levels of serum sodium have lower odds of remission. Those who have an ACTH-producing adenoma that is not identified by magnetic resonance imaging and visual inspection intraoperatively have lower rates of remission and are more likely to have greater manipulation of the pituitary gland intraoperatively [32-36], and the latter may result in greater damage to the pituitary stalk or posterior pituitary, increasing the risk for development of DI and resultant hypernatremia.

A higher preoperative copeptin was associated with male sex and increasing BMI SDS. Increasing preoperative copeptin was also found in pubertal boys compared with pubertal girls, with no difference in copeptin between prepubertal boys and girls. It is particularly interesting to note that these associations were only in the preoperative plasma copeptin levels, but not the postoperative peak copeptin or Δcopeptin. Because the association of higher plasma in adult males and pubertal males in comparison to adult females and pubertal females, respectively, have been reported by others [2637-40], it raises the question of a change in the association of sex and BMI with plasma copeptin in the postoperative state. An effect of BMI or sex was not found by remission status, so it does not seem that the postoperative hypocortisolemic state for those in remission could explain this loss of association. However, this study may not have been powered to detect this.

Strengths of this study include the prospective nature of the study. Further, this is the first study assessing the utility of copeptin to predict remission after treatment of CD. Limitations of this study include the small sample size because of the rarity of the condition, difficulty in clinically diagnosing DI and SIADH, potential effect of post-TSS fluid balance disorders (particularly for those who may have developed transient partial DI or transient SIADH), lack of long-term follow-up, lack of any postoperative follow-up in 11 of the 44 total subjects, as well the observational nature of the study. Further, it is possible that pubertal status, sex, and BMI may have affected copeptin levels, which may have not been consistently detected because of lack of power. Lack of data on the timing of hydrocortisone replacement is an additional limitation of this study because postoperative glucocorticoid replacement could affect AVP secretion via negative feedback. Additional studies are needed to assess to further assess the role of vasopressin and measurement of copeptin in patients before and after treatment of CD.

A clear difference in peak postoperative plasma copeptin as an early marker to predict remission of CD after TSS was not found. Further studies with larger sample sizes are needed to further evaluate postoperative plasma copeptin as an early marker to predict remission of CD, though the utility of this test may be limited to those who do not develop water balance disorders postoperatively. Future studies comparing copeptin levels before and after treatment of adrenal CS would be of particular interest because this would minimize the risk of postoperative DI or SIADH which also influence copeptin levels. Additionally, comparison of thirsted preoperative plasma copeptin in those with endogenous CS and thirsted plasma copeptin in healthy controls could potentially provide evidence of whether or not preoperative plasma copeptin is lower in patients with CD, or other forms of CS, compared with healthy subjects. Further, if this is found to be true, it could potentially be a diagnostic test to lend support for or against endogenous CS.

Abbreviations

 

  • AVP

    arginine vasopressin

  • BMI

    body mass index

  • CD

    Cushing disease

  • CS

    Cushing syndrome

  • DI

    diabetes insipidus

  • HPA

    hypothalamic-pituitary-adrenal

  • IQR

    interquartile range

  • MN

    midnight

  • OR

    odds ratio

  • POD

    postoperative day

  • SDS

    SD score

  • SIADH

    syndrome of inappropriate antidiuresis

  • TSS

    transsphenoidal surgery

  • UFC

    urinary free cortisol

  • ULN

    upper limit of normal

Acknowledgments

The authors thank the patients and their families for participating in this study.

Funding

This work was supported by the Intramural Research Program, Eunice Kennedy Shriver National Institute of Child Health & Human Development (NICHD), National Institutes of Health.

Disclosures

C.A.S. holds patents on technologies involving PRKAR1A, PDE11A, GPR101, and related genes, and his laboratory has received research funding support by Pfizer Inc. for investigations unrelated to this project. C.A.S. is associated with the following pharmaceutical companies: ELPEN, Inc., H. Lunbeck A/S, and Sync. Inc.

Clinical Trial Information

ClinicalTrials.gov registration no. NCT00001595 (registered November 4, 1999).

Data Availability

Some or all datasets generated during and/or analyzed during the current study are not publicly available but are available from the corresponding author on reasonable request.

Published by Oxford University Press on behalf of the Endocrine Society 2022.
This work is written by (a) US Government employee(s) and is in the public domain in the US.

Management of Adrenocorticotropic Hormone-secreting Neuroendocrine Tumors and the Role of Bilateral Adrenalectomy in Ectopic Cushing Syndrome

Abstract

Background

Neuroendocrine tumors can cause ectopic Cushing syndrome, and most patients have metastatic disease at diagnosis. We identified risk factors for outcome, evaluated ectopic Cushing syndrome management, and explored the role of bilateral adrenalectomy in this population.

Methods

This was a retrospective study including patients with diagnosis of ectopic Cushing Syndrome secondary to neuroendocrine tumors with adrenocorticotropic hormone secretion treated at our quaternary referral center over a 40-year period (1980–2020).

Results

Seventy-six patients were included. Mean age at diagnosis was 46.3 ± 15.8 years. Most patients (N = 61, 80%) had metastases at ectopic Cushing syndrome diagnosis. Average follow-up was 2.9 ± 3.7 years (range, 4 months–17.2 years). Patients with neuroendocrine tumors before ectopic Cushing syndrome had more frequent metastatic disease and resistant ectopic Cushing syndrome. Patients with de novo hyperglycemia, poor neuroendocrine tumor differentiation, and metastatic disease had worse survival. Of those with nonmetastatic disease, 8 (53%) had ectopic Cushing syndrome resolution after neuroendocrine tumor resection, 3 (20%) were medically controlled, and 4 (27%) underwent bilateral adrenalectomy. In patients with metastatic neuroendocrine tumors, hypercortisolism was initially medically managed in 92%, 3% underwent immediate bilateral adrenalectomy, 2% had control after primary neuroendocrine tumor debulking, and 2% were lost to follow-up. Medical treatment resulted in hormonal control in 7 (13%) patients. Of the 49 patients with metastatic disease and medically resistant ectopic Cushing syndrome, 23 ultimately had bilateral adrenalectomy with ectopic Cushing syndrome cure in all.

Conclusion

Patients with neuroendocrine tumors before ectopic Cushing syndrome development were more likely metastatic and had worse survival. De novo hyperglycemia and poor neuroendocrine tumor differentiation were predictive of worse prognosis. Medical control of hypercortisolism is difficult to achieve in patients with neuroendocrine tumors–ectopic Cushing syndrome. Well-selected patients may benefit from bilateral adrenalectomy early in the treatment algorithm, and multidisciplinary management is essential in this complex disease.

Graphical abstract