Background

NE3107 is a derivative of β-androstenetriol, a naturally occurring adrenal sterol metabolite. The parent compound has been chemically modified to increase oral bioavailability and stability. It has anti-inflammatory and insulin-sensitizing actions, and enters the brain. NE3107 binds to ERK1/2, kinases involved in inflammatory signaling and insulin responses. It does not interact with nuclear steroid hormone receptors, and has no androgenic or estrogenic function. It is not immunosuppressive.

In preclinical work, NE3107 demonstrated anti-inflammatory activity in vitro, and was neuroprotective in animal models of optic neuritis and glaucoma, where it reduced microglia activation (e.g., Lambert et al., 2017; Khan et al., 2014). In the glaucoma model, NE3107 selectively inhibited pro-inflammatory ERK/NK-kB signaling without affecting the enzyme’s homeostatic actions. It eased insulin resistance in diabetic rodents (Wang et al., 2010; Lu et al., 2010). In Parkinson’s disease mouse models, the compound was reported to protect against dopaminergic neuron loss (Nicoletti et al., 2012). The company claims that it improved the efficacy of L-dopa and decreased dyskinesia in Parkinsonian marmosets.

Multiple Phase 1 studies, and one Phase 2 study, have been completed in a range of indications including impaired glucose tolerance, Type 2 diabetes, ulcerative colitis, and rheumatoid arthritis. Where results were reported, the drug showed anti-inflammatory and insulin-sensitizing activities (Reading et al., 2013; Reading et al., 2013). Pharmacokinetics were dose-proportional from 2 to 100 mg. Most side effects were reported to be mild or moderate, and no serious adverse events were attributed to the drug.

Findings

In August 2021, BioVie began a Phase 3 trial in people with mild to moderate probable AD. The study planned to randomize 316 people to 20 mg NE3107 twice daily or placebo, for 30 weeks. The coprimary outcomes were change on the ADAS-Cog12 and ADCS-CGIC. Secondary endpoints include measures of neuropsychological deficits, functional performance, and glycemic control. A substudy will evaluate exploratory endpoints of volumetric magnetic resonance imaging and FDG-PET glucose uptake. The study, at 36 locations in the United States, finished in September 2023. The rationale and protocol are published (Reading et al., 2021).

In January 2022, an open-label, exploratory study began to examine changes in cognition and CNS biomarkers of AD, inflammation, and metabolic state in patients treated with NE3107. The 23 participants had MCI or mild dementia, with a primary complaint of memory impairment, and at least one abnormal brain imaging result. They took 20 mg twice a day for 12 weeks, and were evaluated by multiple MRI modalities to assess brain glutathione levels, dendritic density, functional connectivity, and neurovascular coupling. Secondary outcomes were standard scales of cognition including the Quick Dementia Rating Scale, MoCA, ADAS-Cog12, MMSE, CDR, ADCOMS, and fasting glucose. The trial ran by invitation at a single center in Los Angeles, and finished in August 2022. According to results presented by the company  at the 2022 CTAD, imaging studies showed normalization of cerebral perfusion, as well as increased functional connectivity in several networks involving the nucleus basalis of Meynert, the precuneus, and the hippocampus after treatment. Increased brain glutathione suggested a lessening of oxidative stress. Participants' cognitive function was reported to have improved in four of six tests, with changes reaching statistical significance in a prespecified subgroup analysis of 17 patients with MMSE scores above 20. About 60 percent of participants were reported to have had reductions in the inflammatory serum biomarker TNFα, and CSF pTau and Aβ42, that correlated with ADCOMS or ADAS-Cog12 scores. Imaging improvements were said to correlate with better ADCOMS scores.

Also in January 2022, BioVie treated the first patient in a Phase 1/2 trial testing safety, tolerability, and pharmacokinetics of NE3107 in people with Parkinson’s disease. Forty participants on dopamine replacement were taking 20 mg NE3107 or placebo twice daily for 28 days. The primary objective was to evaluate drug-drug interaction with levodopa, as required by the U.S. FDA. A dozen other endpoints included the Unified Parkinson Disease Rating Scale Parts 1-3, total ON and OFF time, nonmotor symptoms, dyskinesia, and NE3107 pharmacokinetics. The study, at 16 centers in the U.S., was fully enrolled as of October 2022. On December 5, BioVie announced positive topline results on the UPDRS motor score. Additional data was presented at meetings in March and August 2023 (see 2022 Globe Newswire press release; 2023 BioVie pree release; 2023 BioVie press release). 

In August 2023, the company began an exploratory trial to assess the effects of NE3107 on sleep and fatigue in five patients with traumatic brain injury.

In November 2023, BioVie announced results of the Phase 3 Alzheimer’s trial (press release; company slides). The analysis covered data from 57 patients out of the 439 enrolled. The company had to disqualify nearly half of the study sites for protocol violations, including suspected data falsification at some sites. In the resulting subset of patients, BioVie said that treatment resulted in trends toward improvement on co-primary endpoints of ADAS-Cog12 and CDR-CB, and multiple secondary endpoints. On biomarker measures, BioVie reported NE3107-treated patients to have shown a flattening of NfL and GFAP levels, compared to increases in placebo. Treatment also reportedly improved epigenetic/DNA methylation markers of aging. The company announced it would work with the FDA to use an adaptive trial design and continue enrolling patients to increase the trial's statistical power.

For details on NE3107/HE3286 trials, see clinicaltrials.gov.