Corcept Therapeutics Incorporated has resubmitted its New Drug Application for relacorilant as a treatment for patients with Cushing’s syndrome, returning the oral glucocorticoid receptor antagonist to the U.S. Food and Drug Administration review process after its earlier rejection. The resubmission includes additional analyses of previously submitted clinical data, with Corcept expecting a six-month regulatory review.
Why the resubmission changes the timeline but does not erase the FDA’s core evidentiary concerns
The most consequential element of the resubmission is not that relacorilant has returned to the FDA, but that Corcept appears to have secured a regulatory path based on additional analyses rather than a new clinical trial. That distinction could save several years of development time and substantial capital if the revised statistical package adequately addresses the regulator’s concerns.
It also creates a much narrower margin for error. Additional analyses can clarify missing data, test the robustness of treatment effects and explain how results change under different statistical assumptions. They cannot create prospectively collected evidence that was absent from the original programme. The review will therefore turn on whether Corcept has demonstrated that the available data are sufficiently reliable, clinically meaningful and representative of the broader Cushing’s syndrome population.
The expected six-month review suggests the resubmission will receive a full regulatory assessment rather than a short administrative update. However, the establishment of a review cycle should not be interpreted as evidence that approval is likely. The FDA routinely accepts resubmissions that require substantive reconsideration, and acceptance would only confirm that Corcept has provided a complete response to the deficiencies identified in the previous review.
The exact indication requested in the revised application will also matter. Corcept’s original application focused on hypertension secondary to hypercortisolism, while the latest announcement describes relacorilant more broadly as a treatment for patients with Cushing’s syndrome. Any expansion or adjustment of the proposed label could change the efficacy threshold, the relevant patient population and the benefit-risk analysis required for approval.
How the GRACE trial design produced relacorilant’s strongest evidence and its largest weakness
The pivotal GRACE trial enrolled 152 patients with endogenous Cushing’s syndrome and hypertension, hyperglycaemia or both. Every participant initially received relacorilant during a 22-week open-label phase. Patients who met predefined response criteria could then enter a 12-week, double-blind withdrawal phase in which they either continued relacorilant or switched to placebo.
This design generated a statistically positive primary endpoint. Patients who continued relacorilant were significantly more likely to maintain blood pressure control than those moved to placebo, supporting the argument that the improvements observed during treatment were drug-related rather than spontaneous fluctuations.
The same enrichment strategy that strengthened the randomized comparison also limited its generalisability. Only patients who tolerated treatment and demonstrated a predefined response were eligible for randomization. Among the patients who entered GRACE with hypertension, a substantial proportion discontinued during the open-label phase, and fewer than half proceeded to the randomized withdrawal portion.
That means the strongest controlled evidence came from a selected group already known to have responded to relacorilant. The FDA concluded that this approach could overestimate the treatment effect expected across the wider intended population. The regulator also questioned the reliability of the open-label blood pressure improvement because of missing data and the absence of a concurrent control group during the initial 22 weeks.
Randomized withdrawal studies are useful in rare diseases where conventional parallel-group trials can be difficult to conduct. They can demonstrate whether withdrawing a therapy causes deterioration among responders. They are less persuasive when regulators need to determine how many unselected patients will benefit after starting treatment, particularly when discontinuation and missing data affect interpretation.
Corcept’s new analyses will need to demonstrate that the GRACE result remains credible under more conservative assumptions about participants who discontinued, did not respond or lacked complete blood pressure measurements. A favourable result confined to treatment responders may still be clinically relevant, but regulators must be able to identify those responders and understand the probability that a typical patient will reach that stage.
Why the GRADIENT result remains difficult despite supportive subgroup findings
GRADIENT was intended to provide controlled supportive evidence in patients with autonomous cortisol secretion or Cushing’s syndrome accompanied by hypertension or abnormal glucose metabolism. Unlike GRACE’s enriched withdrawal design, GRADIENT directly compared relacorilant with placebo over 22 weeks.
The study did not meet its primary endpoint. The difference in average 24-hour systolic blood pressure between the relacorilant and placebo groups was less than one millimetre of mercury and was not statistically significant. This result weakened the argument that GRADIENT independently confirmed the blood pressure benefit suggested by GRACE.
Corcept subsequently identified a subgroup of patients with both Cushing’s syndrome and hypertension in which relacorilant appeared to produce a larger reduction in systolic blood pressure than placebo. The analysis was encouraging, but it was conducted after the main study had failed and involved a small number of participants with substantial missing data.
The FDA’s alternative analysis of that subgroup produced a smaller treatment difference that was not statistically significant. This disagreement over statistical handling is likely to sit at the centre of the new review. Corcept may have developed additional sensitivity analyses, estimands or missing-data models that better explain the discrepancy, but the regulator will assess whether those methods were clinically and statistically justified rather than selected because they generated a favourable outcome.
The resubmission therefore represents a test of whether a failed confirmatory trial can still contribute meaningful evidence when interpreted alongside a positive enriched trial, long-term follow-up and improvements across multiple Cushing’s manifestations. Approval is possible without every study meeting its primary endpoint, but the total evidence must still establish effectiveness through substantial and reproducible findings.
What relacorilant could change within the current Cushing’s syndrome treatment landscape
Cushing’s syndrome is treated through surgery when the source of excess cortisol can be removed, followed by medical therapy when surgery fails, is unsuitable or does not fully control the disorder. Available medicines either reduce cortisol production, act on the tumour driving cortisol excess or block cortisol activity at the glucocorticoid receptor.
Relacorilant belongs to the receptor-blocking category. Its proposed advantage over mifepristone, marketed by Corcept as Korlym, is greater receptor selectivity. Mifepristone also interacts with the progesterone receptor, contributing to reproductive and endometrial effects that can complicate treatment in some patients.
Corcept’s clinical programme has presented relacorilant as capable of improving hypertension, glucose control, body composition, cognition and quality-of-life measures without several adverse effects associated with existing therapies. The absence of progesterone receptor activity could give clinicians another receptor-directed option for patients who may benefit from blocking cortisol’s effects rather than suppressing cortisol production.
That potential differentiation should not be treated as settled. The FDA’s previous review identified probable cases of drug-induced liver injury, including one participant with an exceptionally large elevation in alanine aminotransferase. Although the regulator did not report traditional Hy’s law cases, the liver findings introduced a safety issue that must be considered alongside the uncertain magnitude of efficacy.
The revised application is therefore likely to require a detailed safety update, exposure analysis and discussion of liver monitoring. Even if approved, relacorilant could receive laboratory-testing requirements, warnings, treatment-interruption criteria or restrictions affecting real-world adoption.
Its safety profile must also be evaluated in the context of chronic endocrine treatment. Patients with Cushing’s syndrome may receive daily therapy for prolonged periods. A toxicity that is acceptable under intermittent oncology dosing may carry a different significance when patients are exposed continuously for months or years.
Why relacorilant’s ovarian cancer approval does not resolve the Cushing’s review
Relacorilant has already crossed the FDA approval threshold in another indication. In March 2026, the regulator approved the molecule under the brand name Lifyorli in combination with nab-paclitaxel for certain adults with platinum-resistant ovarian, fallopian tube or primary peritoneal cancer.
That approval provides important validation of relacorilant’s pharmacology, manufacturing controls and ability to deliver clinical benefit. It also means Corcept has established commercial supply, medical affairs infrastructure and post-marketing safety systems for the molecule.
However, the oncology approval does not establish effectiveness in Cushing’s syndrome. The approved cancer regimen uses 150 milligrams on the day before, the day of and the day after each nab-paclitaxel infusion. The endocrine development programme evaluated chronic daily dosing that could reach substantially higher doses.
The therapeutic context is also different. In platinum-resistant ovarian cancer, regulators weighed a demonstrated survival benefit against the risks of treatment in a life-threatening malignancy with limited options. Cushing’s syndrome is also serious and can be fatal, but the acceptable toxicity profile for long-term endocrine therapy is assessed differently, particularly when several other medical treatments are available.
Clinicians should consequently view the oncology approval as evidence that relacorilant can be an effective medicine, not as proof that the Cushing’s application has already cleared its key regulatory questions. The efficacy datasets, dosing schedules, patient populations and benefit-risk calculations remain distinct.
How approval could reshape Corcept’s Cushing’s franchise without immediately replacing Korlym
Corcept already has a substantial commercial position in Cushing’s syndrome through Korlym and an authorized generic version of mifepristone. Its existing commercial infrastructure could allow relacorilant to reach endocrinologists more rapidly than a product launched by a new entrant.
Relacorilant could broaden Corcept’s addressable population if its final label extends beyond the narrower hyperglycaemia-related indication held by Korlym. It could also provide an alternative for patients who discontinue mifepristone because of tolerability, reproductive or monitoring concerns.
The commercial transition would nevertheless be more complex than replacing an older product with a newer one. Physicians have more than a decade of experience with Korlym, while relacorilant would arrive with unresolved questions about patient selection, liver monitoring and the predictability of blood pressure response. Payers may also demand evidence explaining which patients should receive relacorilant instead of lower-cost mifepristone or cortisol synthesis inhibitors.
An approval could initially produce internal product segmentation rather than wholesale conversion. Korlym may remain appropriate for patients with prominent hyperglycaemia who tolerate the medicine, while relacorilant could be positioned for patients requiring broader control of cortisol-related manifestations or a more selective receptor profile.
Corcept’s 2026 revenue guidance of $950 million to $1.05 billion shows that its existing endocrine franchise continues to expand even without relacorilant approval in Cushing’s syndrome. This reduces immediate financing pressure, but it also raises the strategic stakes. Relacorilant represents an opportunity to extend Corcept’s leadership in cortisol modulation while differentiating its portfolio from generic mifepristone competition.
What regulators and clinicians will watch during the expected six-month FDA review
The first milestone will be formal FDA acceptance of the resubmission and assignment of a new action date. The classification of the submission and any disclosure about the revised indication could offer early clues about the scope of the review, although neither would predict the final outcome.
Regulatory observers will focus on whether the additional analyses reconcile the different conclusions reached by Corcept and the FDA regarding GRADIENT. They will also examine how missing data and open-label discontinuations are handled in GRACE, whether the treatment effect remains meaningful under conservative assumptions and whether the totality of evidence establishes effectiveness across the proposed population.
Safety will receive comparable attention. Corcept must explain the liver injury cases, provide updated exposure data across indications and show whether monitoring or risk-mitigation measures can make chronic treatment acceptable. The FDA may also consider whether the approved ovarian cancer experience adds useful safety information, while accounting for the major differences in dosing and patient characteristics.
Clinicians will be watching for a label that clearly identifies appropriate candidates. Relacorilant would be easier to integrate into practice if physicians can determine response using blood pressure, glucose control and broader clinical measures without relying on cortisol concentrations, which can remain elevated during receptor blockade.
The resubmission is a meaningful regulatory recovery for Corcept because it avoids an immediate requirement for another lengthy Phase 3 programme. It is not yet a resolution of the issues that produced the original complete response letter. The next review will determine whether stronger interpretation of existing evidence can substitute for new evidence, and whether relacorilant’s potential clinical advantages outweigh the efficacy uncertainty and liver safety risk identified by the FDA.
