Cushing’s disease (CD) is the most common form of endogenous hypercortisolism and is characterized by an adrenocorticotropic hormone (ACTH)-secreting pituitary adenoma, with corticotroph tumors representing up to 15% of pituitary neoplasms [ ]. In terms of incidence, a nationwide Swedish study reported 1.6 cases per million persons between 1987 and 2003 and up to 2.0 cases per million between 2005 and 2013 [ ], while an even higher incidence of 6.2 and 7.6 cases per million persons per year in 2009 and 2010, respectively, was obtained from health insurance databases in the USA [ ]. The condition also has a significant female predominance with a 4:1 ratio.

Despite significant progress in terms of accuracy of radiological and biochemical testing, diagnosis of CD is still a challenging process, encompassing baseline and dynamic testing, invasive procedures, and different imaging techniques. Treatment strategies for CD are multifaceted and may include surgical, medical, and, in some cases, radiotherapeutic approaches, however, optimal management requires individualized, multidisciplinary care.

In the clinical suspect of hypercortisolism, several steps need to be taken. The diagnosis requires a stepwise approach, aimed at excluding exogenous Cushing’s syndrome (CS), confirming autonomous cortisol secretion, followed by investigating its origin. While the diagnosis of endogenous CS after exclusion of exogenous (Glucocorticoid) GC use has been previously described, to summarize, three tests are used:

• 1.

  24-hour urinary free cortisol (UFC, average of 2–3 days collection): This test measures the amount of cortisol excreted in the urine over a 24-hour period. Elevated levels typically indicate hypercortisolism. It’s crucial to ensure the collection is complete and accurate, as improper collection can lead to false results. Contraindications to the test are mainly kidney-related, as cortisol excretion and metabolism are altered in renal failure. Overweight or obesity can also yield falsely positive results.

• 2.

  Late-night salivary cortisol (LNSC, two or more tests): Thanks to the physiological midnight nadir of cortisol secretion, measuring cortisol in a salivary sample collected late at night (typically around 11–12:00 p.m.) can reveal autonomous secretion in case of nonsuppressed levels. This test is noninvasive and can be conveniently done at home. Night-shift work or other circadian-disrupting conditions can alter the results.

• 3.

  Overnight dexamethasone suppression test (DST) (Nugent’s test): This test takes advantage of the HPA (hypothalamic-pituitary-adrenal) axis suppression by exogenous GCs. Patients take 1 mg of dexamethasone at 11:00 p.m., and a serum cortisol level is measured at 8:00 a.m. the next morning. Normally, dexamethasone suppresses cortisol production, but in CS, cortisol levels remain elevated above a 1.8 mcg/dL–50 nmol/L threshold. Contraindications are conditions that induce altered levels of cortisol-binding globulin (CBG), such as pregnancy or liver disease, or use of medications altering dexamethasone metabolism, such as CYP3A4 inductors, or malabsorption. False negatives can result from renal protein loss or inhibition of dexamethasone metabolism.

Other tests are serum midnight cortisol measurement, the low-dose 2-day dexamethasone test (LDDT) (Liddle’s test), and administering 0.5 mg of dexamethasone every 6 hours for 48 hours, followed by cortisol measurements.

To investigate the origin of CS, the first step is to measure ACTH levels. ACTH-dependent CS is characterized by normal or elevated ACTH levels, while in adrenal CS, pituitary suppression due to hypercortisolism leads to low levels of ACTH. Once the ACTH dependency is confirmed, further steps will be taken and will be discussed in the following section.

Distinguishing CD from ectopic ACTH secretion (EAS) from a nonpituitary neoplasm and nonneoplastic hypercortisolism (pseudo-CS) poses a significant diagnostic challenge. In fact, all three conditions are characterized by signs and symptoms of cortisol excess and share biochemical and clinical features.

Nonneoplastic hypercortisolism : Several conditions, such as psychiatric disorders, polycystic ovary syndrome (PCOS), obesity, and alcohol use disorder, can chronically activate the HPA axis [ ]. Underlying mechanisms include alterations of the physiological cortisol rhythm, altered stress response, and hyperactive morning cortisol surge in neuropsychiatric disorders [ ]; reduced CBG and sex hormone binding globulin (SHBG) levels in PCOS; and increased corticotropin-releasing hormone (CRH) levels paired with decreased hepatic clearance of cortisol in alcohol use disorder.

Accordingly, DST, LDDT, and UFC results can be altered in all these conditions, and further testing is necessary. Two different dynamic tests are used to differentiate nonneoplastic and neoplastic hypercortisolism: the paired LDDT–CRH test (Dex-CRH) and the desmopressin test.

The first is based on the measurement of CRH-stimulated cortisol after 2 days of dexamethasone suppression, based on the observed increased response to CRH in corticotroph adenomas. It was first introduced in 1993 [ ], but further studies did not entirely confirm the reliability of its performance in identifying neoplastic hypercortisolism, even though the new cutoff of 15-minute cortisol levels has shown good results when compared to other tests [ ]. Variability in CRH formulation, cortisol and ACTH immunoassays, and characteristics of patients could contribute to the heterogeneity of results from different studies [ ].

The second consists of measuring ACTH and cortisol levels after desmopressin infusion, since due to the presence of hyperexpression of V2 and V1b (V3) vasopressin receptors on corticotroph adenomas, a marked increase in ACTH release can be stimulated by desmopressin [ , ].

Even though both tests have been evaluated independently with good specificity for CD, concordance between the two gives the best results in terms of avoiding false positives and detecting neoplastic hypercortisolism [ ].

Ectopic ACTH secretion : Acute and severe presentation of biochemical and clinical signs of CS is usually characteristic of EAS, but CD can be characterized by severe hypercortisolism as well, and the two conditions often have overlapping features. In the presence of life-threatening hypercortisolism, rapid diagnosis and treatment of complications are advised by guidelines [ ], with 2–3 ACTH determinations and UFC as the mainstays of a first-line evaluation. Severe hypokalemia, accelerated progression, and catabolic signs are usually suggestive for EAS rather than CD. UFC of more than 5-fold the upper limit of normality is also suggestive for an ectopic secretion. However, dynamic tests are often necessary [ ]. Most, but not all, nonpituitary neuroendocrine ACTH-secreting tumors lose CRH and vasopressin receptors, along with sensitivity to exogenous glucocorticoids. This is why CRH and desmopressin tests, along with high-dose DST, are used as differential diagnostic tools between CD and EAS. It is worth mentioning that well-differentiated neuroendocrine tumors (NETs) can show similar response to corticotroph adenomas.

Sensitivity and specificity of the CRH test for EAS across different studies were analyzed in a recent consensus [ ] with values ranging from 59% to 93% and from 70% to 100%, respectively. The desmopressin test has also been used in EAS, even though an aberrant expression of V3 receptors has been described, and false positives have been reported [ ]. Concordance between the tests seems to ultimately achieve the best performance.

On the other hand, the HDDST (High Dose Dexamethasone Suppression Test) seems to be losing ground, since its reliability has been challenged in several case series and cohorts, especially in well-differentiated NETs. A recent large retrospective study has confirmed the outperformance of the CRH test compared to HDDT [ ]. Interestingly, proposed cutoff values for CD diagnosis—a ≥35% ACTH and ≥20% cortisol increase for the CRH test and a ≥50% cortisol suppression for the HDDT test—showed sensitivity and specificity values ≤90%. Increasing thresholds to 40% for ACTH and 22% for cortisol in the CRH test and to ≥69% for the HDDT achieved the best results based on combined sensitivity and specificity. Lastly, the authors compared the diagnostic power of combined CRH and HDDT tests; however, the performance depended on whether a negative test—in case of discordant results—was interpreted as positive or negative. For example, if a positive CRH test and a negative HDDT test were considered compatible with the diagnosis of CD, the performance of the combined tests was similar to the CRH alone. On the other hand, interpreting any negative result in combined testing as suggestive for EAS increased specificity and positive predictive value, and was therefore the preferred approach. A “cautionary” approach, combined with the increased thresholds, achieved a 99% specificity in detecting EAS.

The differential diagnosis of ACTH-dependent CS remains a challenge, requiring a combination of clinical evaluation, biochemical testing, imaging, and, in selected cases, invasive procedures. Despite the progress in noninvasive diagnostic modalities, imaging and standard biochemical tests sometimes fail to provide definitive localization of the source of excess ACTH production. In such cases, bilateral inferior petrosal sinus sampling (BIPSS) represents the gold standard for distinguishing between CD and EAS [ ]. This procedure, which involves the catheterization of the inferior petrosal sinuses and measurement of ACTH levels with or without stimulation with CRH or desmopressin, offers high diagnostic accuracy. However, its invasive nature, potential risks, and limited availability require a careful assessment of its indications and limitations.

BIPSS remains the most reliable diagnostic tool when conventional tests fail to provide a clear diagnosis. In patients with ACTH-dependent hypercortisolism, distinguishing between pituitary and ectopic sources is essential to determine the appropriate therapeutic approach. Imaging techniques such as magnetic resonance imaging (MRI) frequently lack sensitivity, with studies showing that up to 50% of ACTH-secreting pituitary adenomas are not detectable on standard scans [ ]. Given that pituitary microadenomas may be very small or indistinguishable from incidental findings, BIPSS provides a direct assessment of ACTH secretion from the pituitary, helping to confirm or rule out CD. Several studies have demonstrated the high sensitivity and specificity of BIPSS in differentiating CD from EAS. The test relies on the measurement of an ACTH gradient between the inferior petrosal sinus and peripheral blood, and conventional ratios were ≥2 at baseline and ≥3 after CRH stimulation [ ]. A relatively recent systematic review and meta-analysis of over 1600 patients confirmed the high sensitivity and specificity of the test [ ], while a recent retrospective study has proposed new cutoffs of ≥1.9 at baseline and ≥2.1 5 minutes after CRH stimulation [ ], even though benefits on overall specificity were minor. In cases where imaging fails to identify an adenoma, BIPSS can be invaluable in confirming the presence of a corticotroph tumor.

Another important advantage of BIPSS is its role in preventing unnecessary pituitary surgery. Misdiagnosis may lead to incorrect transsphenoidal surgery (TSS); establishing a definitive diagnosis through BIPSS can prevent patients from undergoing an ineffective and invasive procedure. Furthermore, in patients with suspected EAS but without an identifiable tumor on imaging, BIPSS can provide clarity and direct further investigative efforts. Despite its advantages, BIPSS is an invasive procedure that carries risks and limitations. First and foremost, it requires highly specialized expertise, as successful catheterization of the inferior petrosal sinuses is technically demanding. This factor limits the availability of the test, as only centers with experienced interventional radiology teams can perform the procedure with high accuracy and minimal complications [ ].

In terms of procedural risks, BIPSS is associated with vascular complications, including thrombosis, hemorrhage, and, in rare cases, stroke. Catheterization of the venous system carries an inherent risk of thrombotic events, particularly in patients with hypercoagulability due to chronic hypercortisolism. Though uncommon, cases of dural sinus thrombosis following BIPSS have been reported, emphasizing the need for meticulous patient selection and postprocedural monitoring [ ].

Another important limitation of BIPSS is the potential for false-negative and false-positive results. False negatives may occur if catheterization is unsuccessful or if ACTH secretion is episodic, as seen in cyclic CD, where fluctuations in hormone levels may affect diagnostic results [ ]. Similarly, false positives can arise in cases where ectopic tumors retain partial CRH receptor expression, leading to an exaggerated response to CRH stimulation and an erroneous diagnosis of CD [ ]. These limitations highlight the necessity of correlating BIPSS findings with clinical presentation and other biochemical markers. Moreover, the cost and resource-intensive nature of BIPSS make it less accessible in many regions. The test requires the use of expensive CRH formulations, multiple ACTH assays, and specialized radiological expertise, factors that contribute to substantial financial and logistical burdens. Given the decline in availability of CRH in some countries, alternative stimulants such as desmopressin are increasingly employed [ ].

Given the complexity of diagnosing ACTH-dependent CS, the decision to perform BIPSS should be based on a comprehensive assessment of the patient’s clinical and biochemical profile. In cases where noninvasive tests provide conflicting results or where imaging fails to localize a pituitary adenoma, BIPSS remains an invaluable tool. However, its use should be limited to centers with appropriate expertise, and patients should be thoroughly informed of the potential risks and benefits. Recent studies have explored potential refinements to BIPSS, including the simultaneous measurement of prolactin levels, which has shown promise in improving diagnostic accuracy by confirming successful catheter placement [ ]. Additionally, novel noninvasive biomarkers and advances in molecular imaging may eventually reduce the reliance on invasive testing, though current methodologies remain insufficient to replace BIPSS in challenging diagnostic cases.

In conclusion, while BIPSS offers unparalleled accuracy in distinguishing between pituitary and ectopic ACTH sources, its invasive nature, potential complications, and limited availability necessitate careful patient selection. As research advances, efforts should focus on optimizing noninvasive alternatives while maintaining access to BIPSS for patients in whom it remains clinically essential.

The localization of ACTH-secreting pituitary adenomas in CD is crucial for determining appropriate treatment strategies. MRI remains the gold standard for identifying pituitary adenomas, but its sensitivity is variable, and alternative imaging techniques, including positron emission tomography (PET) and hybrid imaging methods, are increasingly being investigated.

MRI is the most widely used imaging technique for detecting pituitary adenomas in CD. The standard 1.5T MRI has a sensitivity of approximately 50% for detecting ACTH-secreting pituitary microadenomas, while 3T MRI improves this detection rate significantly [ ]. A systematic review reported an overall MRI sensitivity of 73.4%, with slightly lower sensitivity (70.6%) for detecting microadenomas, demonstrating that while MRI is useful, it fails to identify adenomas in a significant proportion of patients [ ].

Technical refinements, including thin-section acquisitions and the use of gadolinium contrast, particularly in dynamic sequences, enhance the ability to visualize adenomas. Fluid-attenuated inversion recovery (FLAIR) and newer 3D spoiled gradient-echo (SPGR) sequences may further improve sensitivity, though these techniques remain complementary to conventional spin-echo sequences [ ]. Additionally, ultra-high-field 7T MRI has been explored for its ability to improve lesion detection, but its availability is currently limited to research settings [ ]. Despite advances in MRI technology, a substantial proportion of ACTH-secreting adenomas remain undetected, leading to so-called MRI-negative CD. In such cases, additional imaging modalities are required to confirm the presence of a pituitary lesion or to explore EAS.

Functional imaging with PET has been proposed as a complementary tool for identifying pituitary adenomas, particularly in cases where MRI is inconclusive. Studies have demonstrated that ¹⁸ F-fluorodeoxyglucose ( ¹⁸ F-FDG) PET/CT can localize corticotroph adenomas with a sensitivity comparable to that of MRI [ ]. However, prior stimulation with CRH has been shown to enhance ¹⁸ F-FDG uptake, thereby improving the detection of microadenomas [ ]. Another promising PET tracer, ¹¹ C-methionine, offers a higher specificity for pituitary adenomas compared to ¹⁸ F-FDG, as it targets protein synthesis pathways that are upregulated in adenomatous cells [ ]. Hybrid imaging with PET/MRI has been explored as a way to combine the functional insights of PET with the anatomical precision of MRI. In a study involving patients with recurrent or persistent hypercortisolism, PET/MRI successfully localized adenomas in individuals with negative or equivocal MRI findings [ ].

Several alternative imaging techniques have been evaluated for their role in localizing pituitary adenomas. Gallium-68 DOTATATE PET/CT, which detects somatostatin receptor (SSTR) expression, has been investigated for the identification of ectopic ACTH-secreting tumors but has limited utility in pituitary adenomas due to the low SSTR density in these lesions [ ]. Similarly, imaging targeting CRH receptors on corticotroph tumors has been proposed, though further studies are needed before this approach becomes widely available [ ]. Another innovation involves intraoperative MRI (iMRI), which allows for real-time imaging during TSS to confirm the extent of tumor resection. iMRI has been shown to improve gross total resection rates, thereby enhancing surgical outcomes [ ]. However, this technology is not yet widely available and remains limited to specialized centers.

Despite the improvements in imaging technology, certain challenges remain in the detection and localization of ACTH-secreting adenomas. First, the small size of these tumors and their potential for heterogeneous enhancement patterns make their identification difficult. Second, the presence of pituitary incidentalomas may confound interpretation, especially in cases of equivocal MRI findings [ ].

Looking forward, the development of molecular imaging techniques that specifically target corticotroph tumor receptors, along with machine learning-assisted image analysis, holds promise for improving diagnostic accuracy. Additionally, radiomics-based imaging approaches, which analyze quantitative features from imaging data, may provide novel insights into tumor characterization and localization [ ].

In conclusion, MRI remains the cornerstone of imaging in CD, with sensitivity enhanced by high-field scanners, contrast-enhanced sequences, and refined imaging protocols. However, MRI-negative cases necessitate the use of functional imaging modalities such as PET/CT and PET/MRI to improve localization. iMRI may further optimize surgical outcomes, and emerging molecular imaging strategies offer potential for future advancements. As technology continues to evolve, a multimodal approach that integrates anatomical and functional imaging will likely yield the highest diagnostic accuracy for CD.

The role of medical therapy in treating patients with CS is increasingly recognized as significant, particularly for patients with persistent or recurrent hypercortisolism postsurgery or those who are not suitable candidates for surgery, or as a bridging treatment during the waiting period for radiation techniques to take effect [ ]. In certain instances, preoperative medical therapy is used to manage hypercortisolism, thereby improving the patient’s clinical status before surgery. Medical intervention is especially crucial for acutely ill patients, such as those suffering from severe infections, recent cardiovascular events, or acute psychosis. These patients typically require stabilization through medical means aimed at controlling hypercortisolism before they can safely undergo tumor-directed surgery. The spectrum of medical treatments for CD includes steroidogenesis inhibitors, agents targeting corticotropin-secreting tumors, and medications that inhibit the glucocorticoid receptor. Recently, the field has seen significant advancements with the development of novel pharmacological compounds. However, the absence of head-to-head studies means there is limited information available on the superiority of one drug over another.

Steroidogenesis inhibitors target the adrenal gland, blocking one or more enzymes involved in steroidogenesis, which leads to the production of aldosterone, cortisol, and testosterone. To date, ketoconazole, levoketoconazole, metyrapone, osilodrostat, mitotane, and etomidate are the available drugs used in clinical practice for all causes of CS.

Pituitary-directed drugs have become a cornerstone of medical treatment in CD based on their ability to target different receptors expressed by corticotroph tumors, thus reducing ACTH secretion and exerting mass shrinkage. To date, pasireotide remains the only approved pituitary-directed drug for the treatment of CD, whereas cabergoline is widely used in clinical practice as an off-label alternative. Temozolomide is currently the only formally recommended option for aggressive and/or metastatic corticotroph tumors following the failure of conventional, multimodal treatment.

RT is primarily used as adjuvant therapy for patients with persistent or recurrent disease after TSS [ , ] or in cases of aggressive tumor growth [ ]. Treatment modalities employed in clinical practice include conventional RT, consisting of administration of small daily fractions over 6 weeks with a cumulative dose of 45–50 Gy, or stereotactic radiotherapy (SRT), administered as either single or several sessions through different modalities, including photon source (Gamma Knife, CyberKnife, and linear accelerators) or proton particles. The choice of radiation modality should include convenience for patients and proximity to critical structures. Gamma Knife radiosurgery (GKRS), administered as a single dose or a few fractions of approximately 20 Gy, is generally regarded as the more convenient option [ ], especially in case of a clearly identifiable lesion on MRI [ ], as fewer treatment sessions are required. Following RT, the rate of tumor control, defined by decreased or stable tumor volume on MRI, is generally very high, reaching up to 95% of treated patients [ , ]. Hormonal remission, generally defined by the normalization of UFC levels, has been documented in up to two-thirds of cases treated with conventional external-beam RT [ , ]. Of note, stereotactic radiosurgery (SRS) has been associated with even higher biochemical remission rates, occurring in 80% of cases and being maintained in the long term in 57% [ , ]. Conversely, whole-sellar RT might be the preferable option whenever a clear biological target is not identified, showing comparable initial and long-term remission rates to MRI-positive cases [ ]. SRS may also serve as a primary treatment option in selected cases, namely for patients with high surgical risk or those who decline surgery. In this setting, the 5-year endocrine remission rate was found to be 81% out of 41 patients [ ]. However, long-term monitoring is essential, as both recurrence and tumor progression have been reported following RT.

Importantly, normalization of hypercortisolism usually occurs within 3 years of radiation treatment, with a median time to remission of approximately 15 months [ ]. Given the latency between RT administration and initial remission, adjuvant medical therapy is often required to control hypercortisolism, with periodic discontinuations to assess treatment efficacy [ ]. To date, clear predictors of RT efficacy have yet to be identified, with preliminary data suggesting a potential role of age, high Ki67 index [ ], and the expertise of the treatment team [ ]. Conversely, cortisol-lowering medical therapies, such as ketoconazole [ , ] and cabergoline [ ], have been suggested to negatively affect RT outcomes, but definitive evidence is still lacking. According to a recent meta-analysis, recurrence rates have been reported to be as high as 16%–18% in previously controlled patients in the two largest available series, whereas smaller series did not describe the late failure of GKRS [ ].

New-onset hypopituitarism is the most frequent side effect of RT regardless of treatment modality, occurring in up to 50% of cases [ , ], with hypothyroidism being the most frequent deficiency [ ]. Conversely, severe neurologic complications, including stroke, cranial nerve damage, and risk of secondary brain tumor, are uncommon in radiation-naïve patients and even more so in those treated with SRS [ ].