Phase 3 Trial of Crinecerfont in Pediatric Congenital Adrenal Hyperplasia

Background: Children with classic congenital adrenal hyperplasia (CAH) due to 21-hydroxylase deficiency require treatment with glucocorticoids, usually at supraphysiologic doses, to address cortisol insufficiency and reduce excess adrenal androgens. However, such treatment confers a predisposition to glucocorticoid-related complications. In 2-week phase 2 trials, patients with CAH who received crinecerfont, a new oral corticotropin-releasing factor type 1 receptor antagonist, had decreases in androstenedione levels.

Methods: In this phase 3, multinational, randomized trial, we assigned pediatric participants with CAH, in a 2:1 ratio, to receive crinecerfont or placebo for 28 weeks. A stable glucocorticoid dose was maintained for 4 weeks, and the dose was then adjusted to a target of 8.0 to 10.0 mg per square meter of body-surface area per day (hydrocortisone dose equivalents), provided that the androstenedione level was controlled (≤120% of the baseline level or within the reference range). The primary efficacy end point was the change in the androstenedione level from baseline to week 4. A key secondary end point was the percent change in the glucocorticoid dose from baseline to week 28 while androstenedione control was maintained.

Results: A total of 103 participants underwent randomization, of whom 69 were assigned to crinecerfont and 34 to placebo; 100 (97%) remained in the trial at 28 weeks. At baseline, the mean glucocorticoid dose was 16.4 mg per square meter per day, and the mean androstenedione level was 431 ng per deciliter (15.0 nmol/liter). At week 4, androstenedione was substantially reduced in the crinecerfont group (-197 ng per deciliter [-6.9 nmol/liter]) but increased in the placebo group (71 ng per deciliter [2.5 nmol/liter]) (least-squares mean difference [LSMD], -268 ng per deciliter [-9.3 nmol/liter]; P<0.001); the observed mean androstenedione value, obtained before the morning glucocorticoid dose, was 208 ng per deciliter (7.3 nmol/liter) in the crinecerfont group, as compared with 545 ng per deciliter (19.0 nmol/liter) in the placebo group. At week 28, the mean glucocorticoid dose had decreased (while androstenedione control was maintained) by 18.0% with crinecerfont but increased by 5.6% with placebo (LSMD, -23.5 percentage points; P<0.001). Headache, pyrexia, and vomiting were the most common adverse events.

Conclusions: In this phase 3 trial, crinecerfont was superior to placebo in reducing elevated androstenedione levels in pediatric participants with CAH and was also associated with a decrease in the glucocorticoid dose from supraphysiologic to physiologic levels while androstenedione control was maintained. (Funded by Neurocrine Biosciences; CAHtalyst Pediatric ClinicalTrials.gov number, NCT04806451.).