Osilodrostat for the treatment of Cushing's disease: efficacy, stability, and persistence
2022; Elsevier BV; Volume: 10; Issue: 6 Linguagem: Inglês
10.1016/s2213-8587(22)00134-6
ISSN2213-8595
AutoresOlivier Chabre, Marie Müller, Justine Cristante, Jean‐Luc Cracowski, Emmanuel Gay,
Tópico(s)Adrenal Hormones and Disorders
ResumoWe read with interest the consensus statement of the Pituitary Society on diagnosis and management of Cushing's disease,1Fleseriu M Auchus R Bancos I et al.Consensus on diagnosis and management of Cushing's disease: a guideline update.Lancet Diabetes Endocrinol. 2021; 9: 847-875Summary Full Text Full Text PDF PubMed Scopus (54) Google Scholar which provides very useful new recommendations, including on neurosurgical or medical treatment of Cushing's disease. We have a comment on table 2, which summarises the efficacy of all drugs used for medical treatment of Cushing's disease. In our opinion, the 86% efficacy of osilodrostat is excessively enthusiastic. Efficacy of a drug is defined as the performance of an intervention under ideal and controlled circumstances. For many of the drugs presented in table 2 of the consensus statement,1Fleseriu M Auchus R Bancos I et al.Consensus on diagnosis and management of Cushing's disease: a guideline update.Lancet Diabetes Endocrinol. 2021; 9: 847-875Summary Full Text Full Text PDF PubMed Scopus (54) Google Scholar unlike with most drug development trials, there is no pure placebo comparator, which is acceptable if one assumes that no patient with Cushing's disease is expected to be controlled without treatment. So, in table 2 of the consensus statement, efficacy corresponds to the percentage of patients who showed disease control with the drug; the criterion used for disease control was normalisation of urinary free cortisol (UFC; less than or equal to the upper limit of normal). For Osilodrostat, however, the LINC 3 study,2Pivonello R Fleseriu M Newell-Price J et al.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. 2020; 8: 748-761Summary Full Text Full Text PDF PubMed Scopus (50) Google Scholar which had a somewhat complex design including four phases for a total of 48 weeks, did provide a placebo comparator in the third phase, which was a 10-week randomised withdrawal phase. It is important to understand that in LINC 3,2Pivonello R Fleseriu M Newell-Price J et al.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. 2020; 8: 748-761Summary Full Text Full Text PDF PubMed Scopus (50) Google Scholar this randomised withdrawal of osilodrostat was performed not in the whole population of 137 treated patients but only in 71 patients, who were all previously treated before randomisation and then carefully selected for their complete response to osilodrostat (100% of these 71 patients were controlled by osilodrostat). 36 of these 71 patients were then randomly assigned to continue the drug for 10 more weeks: 31 (86%) of 36 patients maintained their response, compared with only ten (29%) of 35 patients assigned to the placebo group. The primary objective of LINC 32Pivonello R Fleseriu M Newell-Price J et al.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. 2020; 8: 748-761Summary Full Text Full Text PDF PubMed Scopus (50) Google Scholar was to compare the efficacy of osilodrostat versus placebo at the end of the third phase, and its primary endpoint was the proportion of participants who had been randomly assigned to treatment or placebo who showed a complete response. So, the primary outcome of LINC 3 was the maintenance of a complete response in 86% of patients in the osilodrostat group versus 29% in the placebo group. Because 86% is significantly different to 29%, LINC 3 showed that osilodrostat is efficacious. However, 86% is not a measure of efficacy in the whole population of 137 patients who received osilodrostat. This efficacy measure was one of the prespecified secondary endpoints of LINC 3,2Pivonello R Fleseriu M Newell-Price J et al.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. 2020; 8: 748-761Summary Full Text Full Text PDF PubMed Scopus (50) Google Scholar defined as the rate of complete response at weeks 12, 24, and 48, and these rates are clearly provided in the appendix of the LINC 3 publication2Pivonello R Fleseriu M Newell-Price J et al.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. 2020; 8: 748-761Summary Full Text Full Text PDF PubMed Scopus (50) Google Scholar (71·5%, 68%, and 66%, respectively). Therefore, in our opinion, 86% can be understood as a measure of the stability of the response to osilodrostat in a subgroup of patients who were first selected because they showed a complete response, but it is not the efficacy of osilodrostat in the whole population, which was 71·5%, 68%, or 66%, depending on the duration of treatment. Interestingly, in table 2 of the consensus statement,1Fleseriu M Auchus R Bancos I et al.Consensus on diagnosis and management of Cushing's disease: a guideline update.Lancet Diabetes Endocrinol. 2021; 9: 847-875Summary Full Text Full Text PDF PubMed Scopus (54) Google Scholar the evaluation of another medical treatment, levoketoconazole, is shown as “31% UFC normalisation (primary endpoint), 42% normalisation when using imputed data (comparable with other studies); phase 3 randomised withdrawal study showed that 41% lost response with drug vs 96% with placebo”. This means that 59% of patients maintained their response with the drug compared with 4% on placebo, so that 59% for levoketoconazole is the exact equivalent of 86% for osilodrostat. As the authors, rightly, did not use 59% as a measure of efficacy for levoketoconazole, we believe that they should not use 86% as a measure of efficacy for osilodrostat. We believe clinicians must have the most accurate information regarding the efficacy of medical treatments, especially when they might hesitate in advising medical treatment compared with pituitary surgery, such as in patients with Cushing's disease with a normal pituitary MRI scan, a situation where we strongly favour pituitary surgery,3Cristante J Lefournier V Sturm N et al.Why we should still treat by neurosurgery patients with Cushing's disease and a normal or inconclusive pituitary MRI.J Clin Endocrinol Metab. 2019; 104: 4101-4113Crossref Scopus (17) Google Scholar whereas others prefer medical treatment. Nevertheless, the authors should be commended for their work on table 2 of the consensus statement,1Fleseriu M Auchus R Bancos I et al.Consensus on diagnosis and management of Cushing's disease: a guideline update.Lancet Diabetes Endocrinol. 2021; 9: 847-875Summary Full Text Full Text PDF PubMed Scopus (54) Google Scholar as it raises a puzzling question: how is it possible that the two placebo groups from the two withdrawal studies (LINC 32Pivonello R Fleseriu M Newell-Price J et al.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. 2020; 8: 748-761Summary Full Text Full Text PDF PubMed Scopus (50) Google Scholar with osilodrostat and LOGICS4Zacharieva S, Pivonello R, Elenkova A, et al. Safety and efficacy of levoketoconazole in the treatment of endogenous Cushing's syndrome (LOGICS): results from a double-blind, placebo-controlled, randomized withdrawal study. 2021 Virtual Meeting of the Endocrine Society (ENDO 2021); March 20–23, 2021 (abstr P30-1).Google Scholar with levoketoconazole) are so different (29% remained with disease control in the placebo group of the osilodrostat study compared with only 4% in the placebo group of the levoketoconazole study)? More specifically, how is it possible that, 10 weeks after cessation of osilodrostat, 29% of patients remained with disease control, although the half-life of the molecule is believed to be 4 h (as stated in European Medicines Agency report EMA/65346/2019)? One hypothesis is that osilodrostat might exert a persistent effect on adrenocortical cells. This hypothesis warrants additional research, but it already bears important clinical consequences. Indeed, if one considers stopping a so-called block and replace treatment (osilodrostat plus hydrocortisone) in patients with Cushing's disease that is believed to have evolved towards remission (eg, some years after pituitary radiotherapy), this hypothesis implies that, before stopping hydrocortisone, one should check that the patients are still able to secrete cortisol, so that they do not risk adrenal insufficiency if their adrenal glands have developed a persistent effect of osilodrostat. OC reports personal fees for giving lectures at congress and travel to congress from Novartis, Recordati, HRA Pharma, and Pfizer; and fees for being co-investigator in clinical research studies funded by Novartis, Recordati, and HRA Pharma. All other authors declare no competing interests. Consensus on diagnosis and management of Cushing's disease: a guideline updateCushing's disease requires accurate diagnosis, careful treatment selection, and long-term management to optimise patient outcomes. The Pituitary Society convened a consensus workshop comprising more than 50 academic researchers and clinical experts to discuss the application of recent evidence to clinical practice. In advance of the virtual meeting, data from 2015 to present about screening and diagnosis; surgery, medical, and radiation therapy; and disease-related and treatment-related complications of Cushing's disease summarised in recorded lectures were reviewed by all participants. Full-Text PDF
Referência(s)