Are Selective Estrogen Receptor Beta Agonists Potential Therapeutics for Schizophrenia?

Abstract

Background Estrogen therapies, such as estradiol, have shown promise as therapeutics for schizophrenia; however, safety and tolerability concerns, including feminization effects in men and cancer and stroke risk in pre-menopausal women, may limit their broader use. Estradiol binds to both the estrogen alpha (ERA) and beta (ERB) receptors. ERB receptors appear not to mediate many of the concerning side effects of estrogen therapies. In addition, beta receptors have unique localization in cortical regions (i.e., hippocampus), and improve social behaviors and cognition in some animal models, which has led to interests in these compounds for testing in schizophrenia. To our knowledge, there have been no previous clinical trials of selective ERB agonists in schizophrenia. LY500307 is a highly selective agent for beta receptors without effects on estrogen alpha receptors when doses are constrained. Doses that are too high may engage alpha receptors but the alpha engaging threshold dose has not been fully determined in patient groups. The purpose of this dose-response study was to determine: ERB selectivity doses of LY500307 (i.e., without engaging alpha receptors); safety and tolerability; brain target engagement; and effects on cognition and symptoms. Methods A two-staged, double-blind, 8-week, adjunctive to APDs, adaptive phase 1b/2a trial design was conducted in men with schizophrenia (women were not included because of the lack of toxicology, safety, phase 1 and clinical data supporting use in this population). Three LY500307 doses and placebo were evaluated: 25 mg/day, 75 mg/day, and 150 mg/day. The primary markers for estrogen beta receptor selectivity was lack of effects on total testosterone levels (TT) and no feminization signs. Target engagement was assessed with an N-back working memory fMRI task and the electrophysiology measure mismatch negativity (MMN). Cognitive effects were assessed by the MCCB Composite score. Negative and total symptoms were assessed by the NSA-16 and PANSS, respectively. The primary analyses included all subjects and compared the slope from the three LY500307 dosing arms to the placebo slope in order to evaluate the dose responses. The linear mixed model with random intercept was employed and secondary analyses assessed differences between mean changes of the two higher dose arms combined (75 mg and 150 mg) versus placebo. Results Ninety-four patients were randomized across the placebo and three LY500307 dosing arms. There were no effects on plasma TT levels and no evidence of feminization, suggesting all doses were selective for the beta receptor. No significant adverse events were observed. There were no significant differences between the slopes of the three drug doses versus placebo on the brain target engagement variables (fMRI/N-back: F=0.24, p=0.868; MMN (Duration): F=1.08, p=0.358; MMN (Frequency): F=0.89, p=0.446) or on the cognitive/symptom measures (MCCB composite: F=0.87, p=0.458; NSA-16: F=1.79, p=0.148; and PANSS Total: F=0.69, p=0.558.) Secondary analyses also failed to show any significant effects of LY500307 versus placebo on any of the study variables. Discussion Conclusions: This study indicates that the ERB agonist LY500307 was selective, safe, and well tolerated in patients with schizophrenia. This selective ERB agonist, however, failed to demonstrate any significant effects on brain targets, cognition, negative and total symptoms. Potential issues related to dosing and characteristics of the patient population will be discussed. These data suggest that estrogen alpha receptor activation may be necessary to yield positive results in this patient population. Future studies are needed to confirm these findings.