In the late 1940s, researchers observed that the insecticide dichlorodiphenyldichloroethane (DDD) caused selective necrosis in the adrenal cortex of dogs, primarily affecting the zona fasciculata and reticularis [9]. Mitotane, a contaminant in crude DDD preparation, was found to have potent adrenolytic properties. This drug was first administered to humans for the chemotherapeutic treatment of adrenocortical carcinomas, where it demonstrated adrenolytic effects [10], and has since been approved for the symptomatic treatment of advanced adrenocortical carcinoma (unresectable, metastatic, or recurrent). Additionally, mitotane has been used off-label in some countries, notably France, for the treatment of CD and EAS [11-14]. Mitotane has a gradual onset of action that unfolds over several weeks, is cleared slowly after discontinuation, and exerts a long-lasting effect that can persist for months or even years. A retrospective study [12] reported that disease control, defined as normalization of 24-hour urinary free cortisol (UFC), was achieved in 72% (48/67) of patients with CD after a median time of 6.7 months (range, 5.2 to 8.2), with a mean plasma mitotane concentration of 10.5±8.9 mg/L. Patients who attained a plasma mitotane level above 8.5 mg/L were found to have normalized UFC levels. This suggests that the therapeutic plasma concentration of mitotane for achieving an anticortisol effect in ACTH-dependent CS is lower than that required for the anti-tumor effect in adrenocortical carcinoma, which is above 14 mg/L. For the remaining 19 patients, mitotane was discontinued due to ineffectiveness in 10 cases and poor tolerability in nine cases. Among the patients who had well-controlled disease and stopped mitotane without receiving any other treatment, 71% (17/24) experienced a recurrence of hypercortisolism after a median of 13.2 months (range, 5.0 to 67.9) following treatment discontinuation. Mitotane can serve as an alternative to bilateral adrenalectomy if it is initially combined with fast-acting drugs in severe CS [14], and it can enable the subsequent identification of initially occult tumors [13].

Given the prevalence of adrenal insufficiency, glucocorticoid replacement therapy is typically administered using a “block-and-replace” strategy, often with hydrocortisone. The glucocorticoids are prescribed at a higher dose than what is normally used for adrenal insufficiency due to the enzyme-inducing effect of mitotane, which leads to increased catabolism of the replacement therapy [15]. Mitotane elevates cortisol-binding globulin (CBG) levels, causing serum cortisol measurements to overestimate the amount of active (free) cortisol. This can lead to an underestimation of the risk of adrenal insufficiency. Additionally, some assays may underestimate cortisol levels due to their inability to completely liberate cortisol from CBG, potentially leading to an overestimation of adrenal insufficiency risk [16]. Cross-reactivity with 11-deoxycortisol necessitates specific cortisol dosages [7]. Mitotane is known to have numerous drug interactions, which require careful selection of any concomitant therapies [17]. Its teratogenic and abortifacient effects contraindicate pregnancy as long as mitotane is detectable in the blood. In addition to adrenal insufficiency, side effects are extensive and include gastrointestinal problems (such as anorexia and nausea), hepatic and neurologic symptoms (including dizziness, apathy, general weakness, and more rarely ataxia and tremor), metabolic disturbances (dyslipidemia), endocrine problems (male hypogonadism and ovarian macrocysts), hematologic conditions (neutropenia), and dermatologic toxicities.

Despite its effectiveness as a steroidogenesis inhibitor and adrenolytic drug, the use of mitotane is currently limited in CS conditions other than advanced adrenocortical carcinoma due to its specific kinetics, highly variable bioavailability that requires close monitoring, and numerous (and sometimes serious) side effects.

Ketoconazole is an imidazole derivate, originally developed as an antifungal agent. It inhibits several steps in adrenal and gonadal steroidogenesis, particularly in the testis [18]. Cortisol synthesis is blocked at the levels of cytochrome P450 family 11 subfamily A member 1 (CYP11A1; cholesterol side-chain cleavage enzyme), cytochrome P450 family 17 subfamily A member 1 (CYP17A1; 17α-hydroxylase/17,20 lyase), and cytochrome P450 family 11 subfamily B member 1 (CYP11B1; 11β-hydroxylase) (Fig. 1). Ketoconazole monotherapy has demonstrated rapid action and prolonged beneficial effects in retrospective series [19-23]. In the largest retrospective multicenter study on this drug [23], disease control, as indicated by UFC normalization, was achieved in 49% (97/197) of patients (including 35 undergoing radiotherapy), with a mean follow-up of 20.6 months (range, 0.03 to 135) and a mean daily dose of 600 mg (range, 200 to 1,200). Notably, approximately two-thirds of the patients whose condition was not controlled did not receive the maximum dose, despite demonstrating good tolerance of the drug. In a subgroup treated for over 24 months, 65% (33/51) achieved normal UFC levels, but nearly 12% (6/51) experienced a relapse. Due to intolerance, 21.5% (41/190) of patients discontinued treatment.

Ketoconazole necessitates dosing twice or three times daily and relies on gastric acidity for its absorption. It is commonly administered using a dose titration regimen, calibrating the dosage to achieve cortisol levels within the normal range. Adrenal insufficiency has been reported in 6% of cases [24].

The most feared adverse effect of ketoconazole is hepatotoxicity, with rare cases of fatal hepatitis reported. Ketoconazole should not be initiated or continued if liver enzyme levels exceed three times the normal range. Clinicians must strictly monitor liver function at the start of treatment and after each dose adjustment. This entails weekly monitoring for at least the first month, followed by monthly monitoring for up to 6 months and routine checks thereafter. Patients should also be informed about the signs of liver toxicity [24,25]. With these precautions in place, fewer than 20% of patients experience hepatotoxicity, which is typically moderate (with liver function markers less than five times the normal range) and reversible upon timely cessation of the drug [23,25,26]. Gastrointestinal problems are among the other common adverse effects. Although ketoconazole can reduce testosterone production by the testes, side effects related to male hypogonadism are rare. Clinicians should be aware of the potential for multiple drug interactions, as many medications are contraindicated or necessitate careful monitoring when administered with ketoconazole. These include drugs that are metabolized through or that modulate the activity of the cytochrome P450 family 3 subfamily A member 4 (CYP3A4) enzymatic pathway, those that are substrates of the P-glycoprotein efflux pump, and those that carry a risk of QT prolongation. Additionally, it is necessary to address modifiable risk factors for QT prolongation, consider patient-specific risk factors, and monitor electrocardiogram findings.

Ketoconazole is approved for use in Europe to treat CS, but it is not authorized in many other countries. However, levoketoconazole, the levorotatory enantiomer of ketoconazole, received approval in the United States in 2021 for treating CS in patients for whom surgery is not an option or has not been curative. This approval followed the results of two studies. The Study Of levoketocoNazole In CS (SONICS) study was a single-arm, open-label trial consisting of three phases: a dose titration phase lasting 2 to 21 weeks, a 6-month maintenance phase, and a 6-month extended evaluation phase, involving patients with CS [27-29]. Of the 94 patients enrolled, 47 discontinued treatment for various reasons, including 16 who did so due to adverse events. Disease control, as measured by normalization of UFC levels, was achieved in 63 patients (67%) at the end of the titration phase, in 29 patients (30.9%) at the end of the maintenance phase, and in 16 patients (17%) at the end of the extended evaluation phase. The LevOketoconazole to fill a Gap In CS (LOGICS) study included an open-label dose titration and maintenance phase lasting up to 19 weeks, followed by an 8-week double-blind, placebo-controlled, randomized withdrawal phase [30]. Of the 39 patients who had normal UFC levels at the start of the randomized withdrawal phase, 18 were assigned to the placebo group and 21 to the levoketoconazole group. At the conclusion of the randomized withdrawal phase, a loss of UFC response was observed in 94.4% of the placebo group and 47.6% of the levoketoconazole group. Prespecified adverse events of special interest included hepatotoxicity, with transaminase levels at least three times normal values (10.7%), QT interval prolongation (10.7%), and adrenal insufficiency (9.5%).

Metyrapone is a pyridine derivative that inhibits the activity of 11β-hydroxylase (CYP11B1), blocking the conversion of the biologically inactive 11-deoxycortisol to cortisol. It also inhibits aldosterone synthase (cytochrome P450 family 11 subfamily B member 2 [CYP11B2]) and, to a lesser extent, other enzymes involved in steroidogenesis (Fig. 1) [31]. Metyrapone acts rapidly and has a prolonged beneficial effect when used alone, as demonstrated in retrospective series [22,32-35]. In the largest retrospective multicenter study on this drug [36], disease control was achieved in 43% of the 164 patients on metyrapone monotherapy, as measured by UFC normalization, and in 55% of patients as determined by the average of multiple daily serum cortisol levels, with a median dosage of 1,425 mg per day (range, 500 to 4,000). Among a subgroup of 38 patients treated for more than 6 months, 64% (9/14) achieved normal UFC levels, and 72% (18/25) attained a normal cortisol day curve. In the prospective phase III/IV PROMPT study, disease control, based on UFC normalization, was achieved in 48.6% (17/35) of patients at week 36 [37].

A block-and-replace regimen (with a replacement dose of glucocorticoid) or a titration regimen can be employed, involving multiple daily doses (ranging from twice to six times a day). Timed evening doses have been successfully used to reset the cortisol rhythm to normal in a small prospective, open-label, controlled study of patients with adrenal adenomas causing mild autonomous cortisol secretion [38].

Side effects of metyrapone include dizziness within an hour of administration, gastrointestinal problems (which may be minimized by taking the medication with food), clinical hyperandrogenism in women due to the accumulation of androgenic precursors driven by ACTH, hypertension, edema, and hypokalemia resulting from the accumulation of mineralocorticoid precursors also driven by ACTH [24]. Metyrapone may also prolong the QT interval, as noted in the precautions mentioned earlier. Cortisol levels must be monitored using assays that do not cross-react with 11-deoxycortisol, which can become markedly elevated, particularly in patients with CD. Cortisol-specific immunoassays or mass spectrometry-based methods are recommended [39].

Osilodrostat, initially developed as a potential antihypertensive drug, inhibits 11β-hydroxylase (CYP11B1) and aldosterone synthase (CYP11B2), as well as other steroidogenesis enzymes to a lesser extent (Fig. 1) [31,40]. Unlike older medications, the efficacy and safety of osilodrostat have been evaluated in randomized, double-blind, placebo-controlled trials, including a drug withdrawal phase, in patients with CD [41]. For instance, in the LCI IN Cushing’s (LINC) 4 study—a randomized, double-blind, placebo-controlled trial—76.2% (32/42) of patients without a history of pituitary irradiation achieved disease control, as indicated by normalization of UFC, after 12 weeks of treatment, compared to only 4.5% (1/22) in the placebo group [42]. With progressive titration, UFC normalization was observed within 5 weeks in 58% of treated patients. This control was sustained over the long term. Specifically, it was maintained in 69% of patients at week 48, following 36 weeks of open-label osilodrostat treatment [42], and in 54.8% at week 72 of an optional expansion period, with a median daily dose of 4.6 mg (interquartile range, 3.7 to 9.2 mg per day) during the overall study period [43]. No correlation was observed between baseline mean UFC levels and the osilodrostat dose required to achieve UFC normalization. The efficacy of osilodrostat was also demonstrated for other causes of CS in a phase 2 single-arm open-label study, which involved five patients with adrenal adenoma, one with bilateral macronodular adrenocortical disease (BMAD), and three with EAS [44]. Additionally, its effectiveness has been reported in case studies [45-47] and small retrospective series [48-51]. It is essential to monitor cortisol levels using laboratory methods that do not exhibit cross-reactivity with 11-deoxycortisolespecially in patients with CD.

Adrenal insufficiency was reported in 54% (74/137) and 26% (19/73) of patients in the LINC 3 [28,29] and LINC 4 [43] studies, respectively. This included six (4.4%) and three (4.1%) adverse events classified as grade 3 (requiring hospitalization) or 4 (life-threatening/disabling). Adrenal insufficiency occurred not only during the titration phase but also during long-term treatment at a stable dose of osilodrostat, affecting 26.5% of patients after 72 weeks in LINC 3 and 8.5% of patients after 60 weeks in LINC 4. Furthermore, cases have been observed of prolonged control and even sustained adrenal insufficiency after discontinuation of osilodrostat, leading to the hypothesis of a potential cytotoxic effect after extended exposure. In LINC 3, UFC levels remained normalized 8 weeks after osilodrostat withdrawal in 29.4% (10/34) of patients who were randomized to placebo following a 24-week period on osilodrostat. Persistent UFC normalization or low morning serum cortisol levels were observed from 6 weeks to over 16 months after osilodrostat cessation in five patients who had been treated for 6 to 16 months. Despite elevated ACTH concentrations, low dehydroepiandrosterone sulfate (DHEA-S) and low-to-normal 11-deoxycortisol levels were observed [52,53]. Adrenal shrinkage was noted on computed tomography (CT) after 3 months of treatment in a patient with persistent ectopic ACTH secretion [54].

Given its long half-life and the risk of delayed adrenal insufficiency, a block-and-replace regimen—comprising twice-daily dosing of osilodrostat and systematic glucocorticoid supplementation—is commonly employed in current practice [51] and is even recommended [55]. The initial dose and the rate of escalation should be tailored to the patient, with a high starting dose (15 to 30 mg twice a day) and rapid increase (over a few days) for hospitalized patients with severe hypercortisolism and/or life-threatening complications [49,51]. Conversely, a low starting dose (1 or 2 mg twice daily) and gradual escalation (in increments of at least 2 weeks) are advised for other cases. Plasma drug monitoring may be particularly useful during the initiation phase or in instances of poor compliance [56]. Once low cortisol levels have been sustained for several weeks, the dose can likely be reduced to the minimum required to maintain adrenal blockade, or even temporarily discontinued, at least in some patients [53]. Patients must be informed of the potential prolonged effects of the drug, necessitating careful compliance with glucocorticoid replacement therapy and appropriate adjustments in the event of acute illness, even if osilodrostat is discontinued.

Aside from adrenal insufficiency, the most frequently reported adverse effects included gastrointestinal problems, arthralgia, dizziness, hypertension, and elevated blood testosterone levels in female patients. However, side effects associated with adrenal precursors appear to be less frequent over the long term, coinciding with the reduction of their levels, even as ACTH stimulation persists [39,43]. Symptomatic mineralocorticoid deficiency may also develop, necessitating fludrocortisone supplementation [52,53].

Osilodrostat moderately inhibits cytochrome P450 family 1 subfamily A member 2 (CYP1A2) and cytochrome P450 family 2 subfamily C member 19 (CYP2C19) and weakly inhibits CYP3A4 and cytochrome P450 family 2 subfamily D member 6 (CYP2D6) [57]. It often increases the QT interval at high plasma concentrations. Therefore, careful attention should be paid to co-prescribed medicines. Additionally, it is essential to address modifiable risk factors, consider patient-related risk factors for QT prolongation, and monitor electrocardiogram signals.

Cabergoline has not yet been approved for the treatment of CD, despite reports of its capacity to suppress ACTH and/or cortisol levels—as well as induce tumor shrinkage—in case reports and small series, and to achieve long-term remissions in some patients with CD [58,59]. In five open-label prospective studies including a total of 86 patients with CD, disease control—defined as UFC normalization in four studies and midnight cortisol and/or a low-dose dexamethasone suppression test in one study—was achieved in 25% to 40% of patients. However, none of the 20 patients treated with increasing doses up to a median of 5 mg/week (range, 2.5 to 5) over 6 weeks displayed normalization of UFC, salivary, and serum cortisol levels [60-64]. In the largest retrospective multicenter study on this drug [65], of the 53 patients treated with cabergoline monotherapy, five (9%) developed adrenal insufficiency and 16 (30%) achieved UFC normalization (with only half of these also achieving midnight cortisol normalization) within 12 months—11 within the first 3 months and the remainder thereafter. The median dosage was 1.5 mg per week (range, 0.5 to 4.0). Sustained UFC normalization was obtained in 12 patients (23% of the entire cohort and 67% of the complete responders treated for more than a year), lasting for a median of 32.5 months (range, 19 to 105).

Cabergoline is generally well-tolerated, with mild dizziness and gastrointestinal problems being the most common side effects. However, clinicians must remain vigilant regarding rare psychiatric complications, and cardiac valve monitoring is recommended. Patients should be informed of the potential risk of cardiac valve issues and, more broadly, the risk of fibrosis. Screening recommendations for these risks vary across countries. Recent guidelines from the Pituitary Society on prolactinomas advise performing echocardiography before initiating cabergoline treatment, or within the first few months if a long-term regimen with a dose exceeding 2 mg per week is anticipated. Echocardiography should also be conducted promptly upon detection of a heart murmur, every 2 to 3 years for doses above 2 mg per week, and after 5 to 6 years for doses of 2 mg per week or less [66]. Although cabergoline has been used safely in pregnant women with macroprolactinoma, the risk-benefit balance must be considered in patients with CD.

Pasireotide is a second-generation multireceptor-targeted somatostatin analog with a high affinity for somatostatin receptor subtype 5. In corticotroph adenomas more so than somatotroph adenomas, this receptor subtype predominates over somatostatin receptor 2; accordingly, a proof-of-concept study has demonstrated its potential involvement in treating CD [67]. In randomized double-blind studies, disease control—based on UFC normalization—was achieved at month 6 in 15% and 26% of patients administered 600 and 900 μg of pasireotide subcutaneously twice daily, respectively [68]. Additionally, approximately 40% of patients receiving 10 or 30 mg of the drug intramuscularly once a month achieved disease control at month 7 [69]. In the latter study, among patients with evaluable assessments at month 12, 45.7% achieved disease control based on UFC normalization, but only 17.4% attained control based on both UFC and late-night salivary cortisol levels [70]. In the former study, among 53 of 162 patients enrolled with measurable tumor volume, reductions in tumor volume were both dose- and time-dependent. Specifically, a reduction of more than 25% was observed in 16% (4/25) of patients treated with 600 μg twice daily at month 6, increasing to 72% (13/18) in those treated with 900 μg twice daily at month 12 [71].

The primary adverse effect is hyperglycemia, which typically manifests within 1 to 2 months and necessitates supplementary medication in about 75% of patients [24,72]. Additional side effects include gastrointestinal problems and cholelithiasis [24].

In EAS, the first-generation somatostatin analogs octreotide and lanreotide have demonstrated partial and transient efficacy in small series and case reports [73]. Specifically, in cases of EAS resulting from medullary thyroid cancer, an anti-secretory effect has been observed with the use of tyrosine kinase inhibitors, independent of their anti-tumor effect [74].

In BMAD, drugs designed to inhibit the secretion of aberrant ligands or to block illegitimate receptors have demonstrated efficacy in case reports or small series. Notably, long-acting leuprolide acetate has been effective in patients with luteinizing hormone-/human chorionic gonadotropin-dependent CS, and first-generation somatostatin analogs have shown promise in treating food-dependent CS [75].