3.1 Block and replace (B&R) strategy definition3.1.1 What is the definition of the B&R strategy? What is its objective?

B&R involves using high osilodrostat doses, increased rapidly, with the aim to reach as far as possible hypocortisolism and introduce secondarily steroid supplementation.

We suggest defining a “primary” B&R strategy as one that is decided a priori, when the drug is first introduced. This differs from a strategy in which a gradual increase in drug dosage is likely to lead to adrenal insufficiency. In such cases, steroid supplementation (e.g., with hydrocortisone) is added rather than reducing the drug dosage [16]. We propose referring this latter approach as “secondary” B&R strategy.

3.1.2 What is the target level for cortisol?

There is a lack of agreement regarding the target levels of cortisol. Nevertheless, the key aim is to achieve complete adrenal insufficiency (i.e., a cortisol level at 08.00 am that meets the definition of adrenal insufficiency and is as low as possible), ensuring that the evening cortisol level is also suppressed.

Once adrenal insufficiency is achieved, biochemical follow-up can be simplified to measuring morning cortisol concentration to confirm that insufficiency is maintained. It is important to note that, due to the urinary excretion of cortisol provided by hydrocortisone, 24-hour UFC is not suitable to assess endogenous cortisol production and thus unsuitable for use in the B&R strategy in patients treated with hydrocortisone.

3.2 Patient profiles for block and replace (B&R) and considerations for hospitalization3.2.1 Who could benefit from B&R strategy?

We suggest using a primary B&R strategy in patients for whom rapid control of hypercortisolism is mandatory (severe life-threatening hypercortisolism). Even if some cases of CD and adrenal tumors present with very high levels of cortisol, this situation would mainly concern patients with ectopic ACTH secretion.

In addition, we suggest to use a primary B&R for any level of UFC in patients at high risk of adrenal insufficiency during the course of the treatment: e.g., patients with concurrent cancer treatment or patients with inflammatory bowel diseases and chronic diarrhea; patients who live far from an expert center and for whom regular follow-up could be a challenge; patients with a cyclical/highly fluctuating hypercortisolism requiring multiple and unpredictable adjustments in the drug dosage to maintain normal cortisol levels [25].

3.2.2 Should patients treated with B&R be hospitalized?

Hospitalization is strongly recommended in patients with intense hypercortisolism which represents a life-threatening condition and requires symptomatic treatments for comorbidities. A high starting dose (see later) is usually introduced and increased very rapidly. Close clinical and biochemical monitoring in these fragile patients is mandatory to evaluate the efficiency of the treatment and the timing to add the glucocorticoid replacement.

The need for, and duration of, hospital stay should be discussed on a case-by-case basis by an expert team. In that case, patients usually require cortisol monitoring every 2 to 3 days.

3.3 Block and replace (B&R) strategy management: induction, dose adjustments, and monitoring (Fig. 2)Fig. 2The alternative text for this image may have been generated using AI.

Suggested practical algorithm for the block & replace (B&R) strategy using osilodrostat

3.3.1 How is the induction dose determined?

Based on retrospective study of patients treated for ectopic adrenocorticotropin syndrome in France, Dormoy et al.. proposed three different dosing regimens according to clinical severity (life-threatening comorbidities) and UFC levels [26]. Osilodrostat could be started at a dose of 10 to 30 mg/day if the UFC is between 5 and 10 x upper limit of normal (ULN) with or without severe comorbidities, or 30 to 60 mg/day if the UFC is greater than x 10 ULN, associated with life-threatening complications. However, this single algorithm is not supported by strong evidence studies.

A correlation between baseline UFC levels and the dosage required to control hypercortisolism has not been demonstrated, and no strong data in the literature support the concept that the starting dose should be correlated with the level of UFC and the severity of comorbidities when a B&R strategy is considered. However, based on our experience, we strongly believe that the classical starting dose of osilodrostat (2 mg BID) would not be sufficient in these patients, for whom rapid control of hypercortisolism is essential. Similar to the recommendations written for other steroidogenesis inhibitors, there is no evidence-based data to indicate the starting dose. As reported by Dormoy et al. [26], initiating treatment with less than 10 mg/day is associated with a longer time to reach hypercortisolism control, which suggests starting at a minimum dose of 10 mg/day. However, as the individual sensitivity to the cortisol-lowering effects of the drugs varies, the final dose required to achieve adrenal blockade cannot be predicted a priori.

3.3.2 How quickly can the dose be increased?

There is no data in the literature on this specific point. As the objective is to achieve rapid hypercortisolism control, it is suggested to increase the dose by 10 mg/day every 72 h until morning serum cortisol level decreases below 200 to 250 nmol/l. The recommended 2-week titration schedule for osilodrostat should not be applied in life-threatening situations.

3.3.3 How can post-cortisol normalization be adjusted? Universal reduction, criteria-based modification, or fixed maintenance?

The possibility to maintain control of hypercortisolism while reducing the dose during long-term treatments has been demonstrated in published case reports [17].

Recently, intriguing and unexplained cases were reported. Patients with overt and/or intense hypercortisolism developed persistent adrenal insufficiency lasting 1.5 to 9 months, following the discontinuation of osilodrostat after 6 to 13 months of treatment. Interestingly, some of these patients were treated with relatively low doses (2 to 10 mg/day) of osilodrostat [25, 27, 28].

These observations support an enhanced effect of isturisa during chronic use and maintaining the B&R strategy while reducing osilodrostat dosage during long-term treatment. This could be obtained by gradual dose reduction after a sufficient period of osilodrostat-induced adrenal insufficiency (e.g., 4 months), with close monitoring of 08:00 AM plasma cortisol to confirm sustained adrenal insufficiency. A dose reduction scheme could be proposed. As an example, and based on the patient’s profile, the “rule of the 4 " could be applied: i.e., a decrease by 1/4 of the dose with monitoring 4 weeks later.

However, decreasing the dose should be performed cautiously (i.e., only once the clinical situation has greatly improved). We recommend that the dose decrease should not be performed in fragile patients for whom a recurrence of hypercortisolism would be highly detrimental.

Additionally, clinicians should always consider that the patient remains at risk of adrenal insufficiency.

3.3.4 How can B&R-treated patients be monitored? Is 08:00 am pre-hydrocortisone cortisol monitoring sufficient or additional monitoring at later time necessary?

There is no consensus on the frequency of these checks, but the expert group proposes that they be carried out every three months.

Given the half-life of hydrocortisone, early cortisol monitoring should be done before taking hydrocortisone supplementation. This rule is not valid for patients treated with other glucocorticoids, but prednisone may interfere with serum cortisol immunoassays.

3.4 Titration strategy: patient profiles, induction, rate of dose increase, and hospitalization considerations (Fig. 3)Fig. 3The alternative text for this image may have been generated using AI.

Suggested practical algorithm for the titration strategy using osilodrostat

3.4.1 Who could benefit from the titration strategy?

This would mainly concern patients for whom urgent control of hypercortisolism is not required: i.e., patients with the classic mid-term exposure to hypercortisolism without life-threatening comorbidities and moderate hypersecretion (arbitrarily defined as UFC < 5 ULN).

The main drawback of the titration strategy is the delay required to achieve eucortisolism. Its main advantages include limiting the risk of adrenal insufficiency and ensuring the use of the lowest effective dose of osilodrostat.

3.4.2 What is the induction dose? What level of cortisol should be targeted?

Based on the characteristics of the drug in the Summary of Product Characteristics (SmPC) (https://www.ema.europa.eu/en/medicines) and the doses given in LINC-3 and LINC-4 [20, 22], the usual first-line recommended dose is 2 mg BID. Note that this dose might be lower in patients from Asia, as described by Tanaka et al. [29]. However, this dose should be adapted according to the biological intensity and clinical situation.

The cortisol response to osilodrostat varies widely among individuals. Some patients may achieve control with moderate doses despite high baseline UFC, whereas others with only mildly increased UFC may not [20, 22]. In the early titration phase in LINC-3 and LINC-4, the rate of hypoadrenalism was up to 51% and 27% respectively [20, 22]. Therefore, regardless of the starting dose, given the potency of the drug and the possibility of a strong individual response, we recommend that all patients should be systematically educated about the clinical signs of adrenal insufficiency. Moreover, each patient should be provided with a written prescription for hydrocortisone, a kit of injectable hydrocortisone to be used if adrenal insufficiency occurs and a card indicating the risk of acute adrenal insufficiency. We also suggest that the measurement of 08:00 AM serum cortisol should be performed with UFC (e.g., 8 to 10 days after starting the treatment or increasing the drug dosage).

3.4.3 How can the rate of dose be increased?

According to the SmPC (https://www.ema.europa.eu/en/medicines), the dose of osilodrostat should be increased every 2 to 3 weeks depending on the patient’s general condition and the urgency of disease control. Available data suggest that the drug reaches its maximum therapeutic effect 3 to 4 weeks after each dose adjustment. Consequently, in patients with more florid hypercortisolism, a faster titration schedule (e.g., weekly dose increases) may be considered, acknowledging that the plateau of therapeutic effect of each dosage change is probably obtained after 1 month of treatment.

The speed of titration may influence the prevalence of adrenal insufficiency during the titration period. In LINC-4, where the dose was increased every 3 weeks (and not every 2 weeks as stated in the SmPC and as done in LINC-3), with sequential increments of 2, 5, 10, 20 mg BID up to a maximum dose of 30 mg BID, the rate of adrenal insufficiency was lower than in LINC-3, suggesting that the slower titration in LINC-4 may have contributed to this difference [20, 22].

3.4.4 How should titration patients be monitored?

Eucortisolism is the goal of the titration strategy and is usually considered achieved when UFC is normalized. However, some data suggest that UFC alone is not sufficient to consider eucortisolism and other parameters should also be monitored. The pasireotide-LAR study suggests that patients with both a normal mean 24-hour UFC and late-night salivary cortisol concentration (LNSC) have better improvements in metabolic parameters and blood pressure than patients with UFC normalized alone [30].

Therefore, the second step is to normalize LNSC without inducing adrenal insufficiency defined using 08.00 AM cortisol levels. This may be a challenging task. The number of LNSC necessary to assess the nadir of cortisol remains debatable owing to the huge variability of this parameter in some patients [25, 30]. However, we could quite reasonably assume that 2 or 3 LNSC measures which reach the objective are compatible with clinical practice.

Similar monitoring should be maintained over the long term, although the frequency of monitoring may decrease over time. Given the risk of delayed adrenal insufficiency, in the short-term, at least monthly monitoring is required to ensure proper control of hypercortisolism and to detect adrenal insufficiency. It is worth noting that osilodrostat has a delayed onset of action, with a profile different from that of other steroidogenesis inhibitors, which justifies the need for prolonged regular monitoring of cortisol levels.

In patients on a stable long-term dose, provided they are informed about the clinical signs of adrenal insufficiency, hormonal measurements should be performed every 2/3 months. After approximately 6 months, one may consider that a steady state has been reached, making it reasonable to switch to every four months. However, some patients may exhibit frequent and significant spontaneous fluctuations in cortisol production (high variability) [25]. These patients should be identified as this variability may persist during medical treatment and patients therefore require close monitoring [31]. Alternatively, a B&R strategy may be considered for these patients as discussed later.

Another essential aspect of effective monitoring and patient safety is enhancing patient education and awareness. This includes providing information on the clinical signs of adrenal insufficiency and situations requiring increased supplementation; ensuring access to the Adrenal Insufficiency Risk Card; prescribing hydrocortisone; educating on sick day rules; training patients in self-administration of subcutaneous hydrocortisone.

3.4.5 Should glucocorticoid replacement be discontinued after adrenal insufficiency recovery in patients treated with a titration strategy?

In patients in whom titration leads to adrenal insufficiency [20, 22], a transient withdrawal of or decrease in osilodrostat dosage may be considered associated with hydrocortisone supplementation. Once the “acute” phase is over, two different options should be considered: (1) long-term glucocorticoid maintenance as a “secondary” B&R strategy or (2) reducing the dose of osilodrostat and withdrawing hydrocortisone as soon as morning serum cortisol levels return to normal (i.e. above 350 to 500 nmol/l). However, delayed adrenal insufficiency has also been reported with a stable dose of osilodrostat over the mid- to long-term [17, 25, 27]. Therefore, regular monitoring of blood cortisol (even on stable doses) and education of patients are necessary.

3.5 Glucocorticoid replacement for block and replace (B&R) and titration strategies3.5.1 When should glucocorticoid replacement be initiated during the B&R strategy? On day 0? Later?

Glucocorticoid initiation should be based on the chosen strategy. In patients with B&R, we suggest that glucocorticoids be started when 08.00 AM cortisol reaches the lower limit of normal range (LLN) of the assay, while considering the risk of cross reactivity with 11deoxycortisol [32] (for more details, see 6.2).

The objective is to achieve the lowest possible cortisol concentration consistent with adrenal insufficiency. Glucocorticoid supplementation is usually introduced when serum cortisol levels reach the LLN. However, as a safety precaution, some members of the panel propose to initiate glucocorticoid treatment at low doses simultaneously with osilodrostat.

3.5.2 Which replacement agent should be used? At what dosage?

Hydrocortisone is a more physiological drug and is the substitutive treatment of reference. Also, hydrocortisone is a short-acting drug which is easy to use and has a limited side-effect profile. However, its main disadvantage is that it cross-reacts with plasma cortisol in immunoassays. Therefore, serum cortisol level should be measured before the first morning intake, ensuring that the previous dose was taken no later than the late afternoon the day before.

Dexamethasone is a convenient alternative: it is easy to use, administered once daily, and has a long-lasting profile. It does not cross-react with plasma cortisol in most assays. In the context of B&R strategy, some of us suggest initiating dexamethasone at 0.5 mg/day, while maintaining regular monitoring of plasma cortisol. Dexamethasone may be switched to hydrocortisone once low cortisol levels are obtained.

Some of us have also proposed a compromise using prednisone. However, clinicians should be aware that cross-reactions with cortisol has been observed in some immunoassays [32].

The dose of glucocorticoids should be the same as for adrenal insufficiency: i.e., not more than 8 to 10 mg/m² hydrocortisone per day (which could correspond to half a tablet of dexamethasone: i.e., 0.25 mg).

As mentioned previously, patients on B&R strategy should be given the same rules of therapeutic education as those given to patients with adrenal insufficiency although the risk of adrenal crisis is reduced compared to patients on titration strategy; they already take daily steroid supplementation [33].

3.6 Specific monitoring of CS treated with Osilodrostat3.6.1 Should CS monitoring be driven using MRI? Or according to plasma ACTH levels?

Follow-up with pituitary magnetic resonance imaging (MRI) is only relevant in CD. In LINC-4 [22], 5.5% of patients (4/73) showed an increase in tumor size at 48 weeks. The rate was similar in LINC-3 [20] but increased to 12.2% after 72 weeks of treatment [34]. The increase was moderate, with a median change in tumor volume from baseline to week 72 of 1.0 mm3 (range: −74.7 to 1268.5). In LINC-4, 29.6% of patients with MRI at baseline and week 72 had a > 20% decrease, and 37% had a > 20% increase in tumor size. Importantly, pituitary adenoma remnants in CD are small (< 10 mm) in most cases, and a relative increase > 20% may not have clinical consequences. However, Fontaine-Sylvestre et al. reported a case of significant corticotrophic tumor progression (from 3 to 14 mm) in a 40-year-old woman with persistent CD following transsphenoidal surgery treated for 4 years with a stable dose of osilodrostat, accompanied by a large increase in ACTH (from 73 to 500 pmol/l) [35]. Notably, in LINC-3, ACTH levels increased mainly during the first 12 weeks of the titration phase and then remained stable. In LINC-4, ACTH levels increased progressively during the first 72 weeks of treatment (17 to 65 pmol/l) and then remained stable). Despite this, our experience and literature data (LINC-4) do not show an obvious correlation between ACTH and growth, and it is better to rely on systematic MRI follow-up than indication to perform MRI guided by ACTH levels.

We therefore suggest that pituitary MRI should be performed every year (and probably sooner in the case of rapid ACTH increase, high Ki67, the onset of visual disturbances or significant post-surgical remnant volume). The duration of monitoring should be at least 7 years as stated in the guidelines on Nelson’s syndrome which recommended a yearly MRI during the first 3 years, and then MRI every 2 years for 4 years [36].

3.6.2 Should dosing precursors be monitored?

The benefit of monitoring precursors is questionable.

Osilodrostat is a dual inhibitor of CYP11B1 and CYP11B2, resulting in increased levels of 11-deoxycortisol and 11-deoxycorticosterone [37]. 11-deoxycorticosterone has mineralocorticoid activity, but its measurement is not useful in clinical practice. Serum cortisol levels might be falsely increased due to cross-reactions with 11-deoxycortisol in some immunoassays. It is therefore essential to know which assay is used to make sure that the cortisol levels measure is valid [32]. Measurement of cortisol with Liquid Chromatography-Tandem Mass spectrometry (LC-MS/MS) is the best option but some immunoassays with low cross-reactivity are also usable [32]. A decrease in 11-deoxycortisol concentrations over time, after initial increase, has been described in the LINC-3 and LINC-4 extension studies. However, to date, this measurement has no known implications in clinical practice and routine use in practice is questionable.

Regular monitoring of serum potassium is mandatory for patients treated with osilodrostat.

It is worth noting that mineralocorticoid supplementation is usually not required as precursors upstream of the enzymatic blockade accumulate and provide partial mineralocorticoid activity. However, because these precursors have significantly lower bioactivity compared to aldosterone, there remains a risk of mineralocorticoid deficiency, particularly in patients receiving concomitant treatment with spironolactone and eplerenone.

3.6.3 Should androgens be assessed in female patients?

Hyperandrogenism may occur as a side effect of osilodrostat due to its inhibition of the final steps of cortisol and aldosterone synthesis, leading to an increase in ACTH production. In LINC-3, women had a roughly 2-fold increase in testosterone levels from baseline to week 48 (mean testosterone concentrations: 1.3 nmol/l vs. 2.6 nmol/l). Similarly in LINC-4, hyperandrogenism was reported in 24.7% of the 73 patients evaluated in the core phase (48 weeks), with testosterone increasing from 1.1 nmol/l at baseline to 2.5 nmol/l. However, a secondary decrease in testosterone level has been observed during the extension phase of the studies [34, 38].

Measurement of testosterone is recommended in women complaining from hirsutism, but systematic measurement of testosterone is questionable.

Furthermore, measurement of total testosterone may not be a valid assessment of the androgen status as free testosterone may decrease due to elevated Sex Hormone-Binding Globulin (SHBG) levels resulting from weight loss and improvement in insulin sensitivity that follows hypercortisolism control.

In men, the increase in testosterone after 48 weeks of treatment with osilodrostat was not biologically significant. However, measurement of testosterone level in men could be useful in patients with hypogonadism induced by hypercortisolism [39].

3.6.4 Should ionograms be performed?

Because of the increase of 11 deoxycorticosterone [37], which has a mild mineralocorticoid effect, hypokalemia and hypertension can be observed with osilodrostat. In LINC-3, hypokalemia requiring potassium supplementation, spironolactone, and/or dose decrease was reported in 7/137 patients (5.1%) and osilodrostat was discontinued due to worsening of high blood pressure in one patient. In LINC-4, serum potassium remained stable and within the normal range in all patients. We think that ionogram should be performed initially and in parallel to each hormone measurement. This is particularly relevant for patients with severe hypercortisolism or marked hypokalemia at the initiation of osilodrostat, as the drug may rarely worsen hypokalemia via the increase in 11-deoxycorticosterone levels.

Given the effects of osilodrostat on aldosterone synthase, one should be cautious about the long-term use of spironolactone in patients treated with osilodrostat as they may present with a relative mineralocorticoid deficiency when eucortisolism is achieved. This does not preclude the use of spironolactone in the initial stages of treatment, especially in severe hypercortisolism, to control blood pressure and kalemia.

3.6.5 How should monitoring be optimized to improve long-term outcomes?

To date, long-term real-world data on the use of osilodrostat are limited. Available data come from LINC-3 and LINC-4 phase 3 extensions studies, in which treatment outcomes have been reported for duration up to 72 and 96 weeks, respectively [34, 38]. Additional information is expected from ongoing observational studies LINC-6 and LINC-7, which focus on long-term treatment with osilodrostat in patients with CD or CS, respectively.

As previously discussed, we believe that monitoring should not be limited to UFC alone. After UFC normalization, LNSC should also be assessed regularly. Normalization of both markers is likely important to prevent potential long-term consequences of persistent, even mild, hypercortisolism. This hypothesis is supported by the results from a recent analysis pooling data from the 48-week core and extension phases of LINC-3 and LINC-4 to enable a larger sample size and subgroup analyses. For the analysis, patients were stratified based on whether they achieved control of both UFC and LNSC, UFC alone, LNSC alone, or neither. Results showed that patients with both normalized UFC and LNSC experienced better long-term clinical outcomes than the other patients, suggesting that the normalization of both UFC and LNSC optimizes therapeutic benefit [40]. Normalization of LNSC may require a higher dose to be given in the evening than in the morning, ensuring that it does not induce morning adrenal insufficiency. A recently published exploratory study conducted in 16 patients receiving stable treatment with osilodrostat twice daily showed that administering an equivalent daily dose in a single dose at 7 p.m. reduces nighttime salivary cortisol levels and is associated with improved sleep and quality of life. These interesting data need to be confirmed in an independent prospective study [41].