Xanomeline + Trospium) in Patients with Schizophrenia: Superior Efficacy to Placebo Across Positive and Negative Symptoms and a Favorable Safety/Tolerability Profile
Total Page:16
File Type:pdf, Size:1020Kb
Phase 2 trial results of KarXT (xanomeline + trospium) in patients with schizophrenia: superior efficacy to placebo across positive and negative symptoms and a favorable safety/tolerability profile STEPHEN BRANNAN, M.D. CMO, KARUNA THERAPEUTICS NEW MEDICINES SESSION ECNP 2020 ANNUAL MEETING Disclosures/COI The presenter, Stephen Brannan, M.D., is a full-time employee of Karuna Therapeutics ECNP 2020 2 Xanomeline is a selective M1/M4 muscarinic receptor agonist ACETYLCHOLINE MUSCARINE XANOMELINE First identified First isolated from Selective M1/M4 neurotransmitter the mushroom muscarinic receptor agonist (Loewi,1921) Amanita muscaria synthesized in the 1990s Ubiquitous to both central Five muscarinic receptor Signal of antipsychotic (CNS) and peripheral subtypes: M1-M5, found efficacy in Alzheimer’s sympathetic & both in central (CNS) and disease and schizophrenia parasympathetic systems peripheral (PNS) synapses shown in the 1990s ECNP 2020 3 KarXT: proprietary lead product candidate xanomeline trospium chloride (muscarinic agonist) KarXT (muscarinic antagonist) • Human POC in double- • Does not meaningfully cross the blind, placebo-controlled blood-brain barrier, limiting trials in schizophrenia and effects to peripheral tissues xanomeline + trospium chloride Alzheimer’s disease • No known metabolic overlap with • Trials enrolled over 800 xanomeline patients including 68 elderly KarXT is designed to maintain efficacy of xanomeline • Generic drug for overactive patients for ≥ 1year bladder used since the 1960s while ameliorating • Exclusively licensed from Eli Lilly its cholinergic AEs POC = proof of concept; Sources: Bodick et al., 1997; Shekhar et al., 2008 ECNP 2020 4 Phase 2 RCT in Schizophrenia Design Overview ❑ Multi-site, randomized, double-blind, placebo-controlled, 5-week, inpatient trial ❑ Enrolled 182 schizophrenia patients in acute psychotic exacerbation • (N=90 on KarXT, N=92 on placebo) ❑ Washout of other antipsychotics before randomization ❑ 1:1 randomization KarXT: placebo over 5-week hospitalization ❑ Ascending up-titration schedule • Days 1-2: 50/20 KarXT BID (50 mg xanomeline/20 mg trospium) • Days 3-7: 100/20 KarXT BID • Days 8-35: 100/20 KarXT BID with optional increase to 125/30 KarXT BID BID=. Twice daily; ClinicalTrials.gov Identifier: NCT03697252 ECNP 2020 5 Robust results across efficacy endpoints KarXT vs. Placebo Prespecified Outcome Measures p-value (week 5) PRIMARY ENDPOINT Change from baseline in PANSS total score <0.0001 SECONDARY ENDPOINTS Change from baseline in PANSS-positive <0.0001 Change from baseline in PANSS-negative <0.001 Change from baseline in PANSS Marder negative symptoms factor <0.001 Change from baseline in CGI-S frequency counts <0.001 Proportion of CGI-S responders (CGI-S ≤ 2) =0.151 All efficacy analyses performed using the modified intent-to-treat (mITT) analysis set, defined as all randomized participants who received at least one dose of study medication and had a baseline and at least one post-baseline PANSS assessment (n=83 KarXT; n=87 placebo) PANSS: Positive and Negative Syndrome Scale; CGI-S: Clinical Global Impression Severity ECNP 2020 6 Change in PANSS total score over 5 weeks ❑ Primary endpoint: 11.6-point improvement at week 5 for KarXT compared with placebo (-17.4 vs. -5.9 points, p<0.0001) ❑ Statistical separation at every assessed time point ❑ Cohen’s d effect size of 0.75 Change from Baseline= least-squares mean, using MMRM analysis on the mITT population (n=83 KarXT; n=87 placebo) ECNP 2020 7 Key secondary outcomes PANSS Positive, Negative, Marder, and CGI-S ❑ Rapid and sustained statistical separation on four prespecified secondary endpoints, with p<0.001 at endpoint (week 5): • PANSS positive subscore • CGI-S frequency counts (categorial variable, not shown here) • PANSS negative subscore • PANSS Marder negative factor score ❑ Prespecified responder analysis was not stat sig at week 5 • Percentage of patients with CGI-S score of 1 or 2 for KarXT compared with placebo: 5.6% vs.1.4%, p=0.151 ❑ Strong efficacy in post hoc analysis of CGI-S as continuous variable Change from Baseline= least-squares mean, using MMRM analysis on the mITT population (n=83 KarXT; n=87 placebo) ECNP 2020 8 *p<0.05; **p<0.01, ***p<0.001, ****p<0.0001 Adverse events (>5%) over course of study Placebo KarXT (N=90) (N=89) Any AE (N,%) 39 (43.3%) 48 (53.9%) AEs ≥ 5% Constipation 3 (3.3%) 15 (16.9%) Nausea 4 (4.4%) 15 (16.9%) Dry mouth 1 (1.1%) 8 (9.0%) Dyspepsia 4 (4.4%) 8 (9.0%) Vomiting 4 (4.4%) 8 (9.0%) Headache 5 (5.6%) 6 (6.7%) Somnolence 4 (4.4%) 5 (5.6%) ❑ Safety population received ≥1 dose study medication ❑ One SAE on KarXT for worsening of psychosis SAE=severe adverse event ECNP 2020 9 Rates of adverse events over course of study Most cholinergic and anticholinergic adverse events (AE) associated with KarXT treatment decreased over the course of the trial • Rates of nausea, vomiting and dry mouth consistently decreased to rates near placebo by week 5 • All rated mild to moderate and did not lead to any discontinuation from the study • Constipation and dyspepsia rates remained essentially constant (data not shown) Three out of five of the most common KarXT-related AEs decreased over the course of the trial Data represent % of subjects with AEs as an AE at any point in that study week interval (safety population: n=89 KarXT; n=90 placebo) ECNP 2020 10 KarXT was not associated with the most common problematic AEs of current antipsychotic medications Weight Related Observations KarXT was not associated with any weight-related KarXT (N=89) Placebo (N=90) changes Reported AE of weight increased — number (%) 3 (3.4%) 4 (4.4%) • KarXT similar to placebo in mean change in weight, Weight change from baseline to Week 5 — kg ± SD 1.5 ± 2.8 1.1 ± 3.5 mean change in BMI, % patients with >%7 weight Patients with >7% weight increase at Week 5 — number (%) 2 (2.2%) 5 (5.6%) change, and reported AEs of weight increased BMI change from baseline to Week 5 — kg/m2 ± SD 0.5 ±1.0 0.4 ± 1.2 Sedation and Somnolence KarXT was not associated with somnolence or Reported AE of Somnolence — number (%) 5 (5.6%) 4 (4.4%) sedation Reported AE of Sedation — number (%) 2 (2.2%) 2 (2.2%) • Rates of somnolence and sedation similar to placebo Extrapyramidal Symptoms (EPS) KarXT was not associated with EPS Akathisia — number (%) 3 (3.4%) 0 (0%) • Mean changes similar for KarXT and placebo on the Restlessness — number (%) 0 (0%) 1 (1.1%) Barnes akathisia scale and Simpson-Angus scale Simpson-Angus score mean change -0.1 ± 0.7 -0.1 ± 0.6 from baseline to week 5 • 3 patients who reported akathisia in the KarXT arm; all Barnes akathisia mean change 0.0 ± 0.2 0.0 ± 0.4 resolved spontaneously without changes in study drug from baseline to week 5 and all patients scored a 0 at all time points on the Above analysis on safety population; received ≥1 dose study medication Barnes akathisia scale ECNP 2020 11 Summary of safety and tolerability • Overall AE rate for KarXT compared with placebo (54% vs.43%) • One SAE in KarXT group for worsening psychosis • Overall discontinuation rate similar for KarXT and placebo (20% vs. 21%) • TEAE discontinuation rate the same for KarXT vs placebo (2.2% vs. 2.2%) • 91% of patients escalated to 125/30 dose of KarXT (vs. 97% who escalated on placebo) • AE profile of KarXT differed from available antipsychotics • KarXT most common AEs: constipation, nausea, dry mouth, dyspepsia, and vomiting • All of these mild or moderate in severity • No discontinuations associated with above AEs • KarXT not associated with AEs commonly seen with currently available antipsychotics: • Weight gain and metabolic changes similar to placebo • EPS similar to placebo Above analyses in the safety population (n=89 KarXT; n=90 placebo); SAE= serious adverse event; TEAE= treatment-emergent adverse event ECNP 2020 12 Exploratory endpoint: cognitive performance results Cognitive Test Statistic Value (KarXT vs. placebo) • Analysis: ANCOVA of LS mean change from LS mean CFB 0.13 baseline for KarXT vs. placebo at week 5 in mITT Composite Score p-value 0.11 population with baseline and post-baseline (day Cohen’s d 0.24 LS mean CFB 0.04 35) assessments (n=65 KarXT; n=68 placebo) Detection p-value 0.07 Cohen’s d 0.28 LS mean CFB 4.62 • Composite score calculated from summation of Pediatric Groton p-value 0.47 Maze Learning individual test results; Cohen’s d=0.24 with p=0.11 Cohen’s d 0.11 LS mean CFB 0.01 Identification p-value 0.40 • Patients in the KarXT arm performed better than Cohen’s d 0.13 those in the pbo arm on 5 out of 6 tests, with LS mean CFB 0.90 International p-value 0.35 modest Cohen’s d effect sizes ranging from 0.11- Shopping List Cohen’s d 0.14 0.28; lowest p-value in Detection Test (p=0.07) LS mean CFB 0.02 One-Back Speed p-value 0.35 Cohen’s d 0.16 • Potential confound of psychosis improvement LS mean CFB 0.00 associated with KarXT vs. placebo One-Back Accuracy p-value 0.92 Cohen’s d -0.02 ECNP 2020 13 Subgroup analysis: stratification by baseline impairment Highest Middle Lowest Cognitive mITT Statistic Impairment Impairment Impairment Test (KarXT vs. placebo) population Tertile Tertile Tertile Larger benefit observed with KarXT in patients LS mean CFB 0.13 0.50 0.10 0.02 Composite p-value 0.11 0.02 0.52 0.87 with greater cognitive impairment Score Cohen’s d 0.24 0.83 0.19 0.04 • Post hoc analysis evaluating effect in population broken LS mean CFB 0.04 0.07 0.02 0.03 out into tertiles by baseline composite score (n=22-25 Detection p-value 0.07 0.20 0.50 0.44 per arm per tertile) Cohen’s d 0.28 0.47