Background

NDX-1017 is a brain-penetrant small molecule that activates signaling via the hepatocyte growth factor (HGF)/MET receptor system (Moebius and Church, 2023). HGF promotes proliferation and survival of neurons, enhances hippocampal synaptic plasticity, and boosts learning and memory (Akimoto et al., 2004; Kato et al., 2012). NDX-1017 is being developed as a subcutaneous once-daily injection.

Hippocampal expression of MET receptors is diminished in Alzheimer’s disease (Hamasaki et al., 2014). Conversely, expression of HGF by gene transfer improved Aβ-induced cognitive deficits in mice, and prevented motor symptoms and neuron loss in a rat model of parkinsonism (Takeuchi et al., 2008; Koike et al., 2006).

Direct, intrathecal administration of hepatocyte growth factor itself has been studied in a rat model of ALS and a marmoset model of traumatic brain injury, and is being investigated clinically for ALS in Japan (Aoki et al., 2019). 

Fosgonimeton is a prodrug formulation of a small molecule originally identified in a screen for positive modulators of HGF/MET signaling (Johnston et al., 2023). The small molecule, fosgonimeton-active metabolite (fosgo-AM), was reported to be neurotrophic and neuroprotective in primary rat neurons, and to reverse scopolamine-induced learning deficits in rats. To improve its pharmacokinetics, the company developed the prodrug formulation, which generated higher blood and brain levels of fosgo-AM than were possible with direct delivery of the metabolite. In a mouse model of cognitive impairment due to acute inflammation, fosgonimeton partly reversed cognitive deficits. It was also reported to prevent Aβ42 toxicity and protect dopamine neurons against 6-OHDA (2023 AD/PD conference poster, Reda et al., 2024).

This drug may be related to N-hexanoic-Tyr-Ile-(6) aminohexanoic amide, aka Dihexa, U.S. Patent 8598118B2. This small-molecule, brain-penetrant angiotensin IV analog was invented by researchers at Washington State University, Pullman. Dihexa activates HGF/MET signaling to stimulate dendritic arborization and synaptogenesis, and improves learning/memory in aged rats, and in rats with scopolamine-induced amnesia (ScienceDaily news story; McCoy et al., 2013; Benoist et al., 2014; reviewed in Wright and Harding, 2015). 

M3 Biotechnology was founded to commercialize Dihexa (see WSU article). The company told Alzforum that further development led to NDX-1017. M3 Biotechnology in 2019 was renamed Athira Pharma, and NDX-107 in 2020 was renamed ATH-1017 and, in March 2022, fosgonimeton.

Findings

In October 2017, the company began a first-in-human, Phase 1 safety study of single- and multiple-ascending doses in 88 healthy people and AD patients. Funded by ADDF and others, the placebo-controlled trial tested doses from 2 to 90 mg, injected subcutaneously. Outcomes were drug-related adverse events and pharmacokinetics.

Results were presented at the 2019 CTAD conference (Dec 2019 conference news). In the single-ascending-dose portion, 48 healthy men received up to 90 mg and no drug-related adverse events were seen. Likewise, multiple dosing of 29 healthy elderly and 11 AD patients with up to 80 or 40 mg/day, respectively, for nine days raised no safety issues, with dose-proportional pharmacokinetics in all groups.

Quantitative electroencephalography (qEEG) and event-related potential (ERP) measures were evaluated as noninvasive markers of brain penetration and target engagement by NDX-1017. EEG gamma power is associated with learning, memory, and executive functions, while ERP P300 latency is a marker of cognitive processing speed. In people with AD, gamma power is reduced, and P300 latency is lengthened. In the trial, NDX-1017 was reported to cause a dose-related increase in gamma power in healthy people; in the AD group, multiple dosing led to improvements in gamma power and p300 latency. The trial included no cognitive assessments. Trial results were published after peer review (Hua et al., 2022; NeurologyLive).

In September 2020, the company began a Phase 2 trial called LIFT-AD. It planned to enroll 300 participants with clinically diagnosed mild to moderate AD, and compare two doses of drug to placebo, given once daily using prefilled syringes, for six months. The primary outcome will be the Global Statistical Test, a combination of scores from the ADAS-Cog11 and the ADCS-Clinical Global Impression of Change. Secondary endpoints will include ADAS-Cog11, ADCS CGIC, and ADCS-ADL. In early 2022, the company increased enrollment to 420, and reclassified the trial to Phase 3. In 2023, enrollment was further increased to 475, with completion expected in January 2024.

In November 2021, another Phase 2 study called ACT-AD began to assess the relationship between ERP changes and cognition in 77 people with mild to moderate AD. Participants received 40 or 70 mg ATH-1017 or placebo daily for six months, with an optional six-month open-label extension. The primary outcome was ERP P300 latency; secondary outcomes included ADAS-Cog11, global clinical change, ADCS-ADL, and drug pharmacokinetics. The trial was completed in May 2022. In June, the company announced the trial had failed to meet its primary endpoint (press release). A prespecified subgroup analysis suggested that use of cholinesterase inhibitors by 60 percent of participants may have diminished the effects of fosgonimeton. A post hoc analysis of people taking only fosgonimeton indicated improvements in ERP P300 latency and ADAS-Cog11 endpoints, the company reported. There were no serious adverse events or deaths in the study. The most common adverse event was injection site reaction. Results were presented at meetings in 2022 (e.g. see AAIC slides). The company presented additional post hoc analyses at the April 2023 American Academy of Neurology conference (slides). They reported that fosgonimeton, in the absence of acetylcholinesterase inhibitors, improved MMSE scores by 1.6 points from baseline; however, the change compared to placebo was not statistically significant. Decreases in biomarkers NfL and GFAP correlated with clinical outcomes in people taking only fosgonimeton. The company presented data suggesting that acetylcholinesterase inhibitors antagonize the effect of fosgonimeton on HGF/MET signaling.

Both ACT-AD and LIFT-AD offer a two-year, open-label safety extension.

In June 2021, the company placed CEO Leen Kawas on temporary leave amid questions about research integrity related to her doctoral dissertation, which laid the foundations for work on ATH-1017 (Seattle Times news story). According to news reports, Washington State University is conducting an investigation of multiple papers from her time as a researcher there, after comments on PubPeer noted irregularities in figures and data (Puget Sound Business Journal news; STAT news). Kawas resigned in October 2021, after an independent investigation found altered images in her dissertation and at least four publications related to her graduate work (company release).

In January 2022, Athira began a Phase 2 trial testing 26 weeks of fosgonimeton on cognition in people with Parkinson’s disease dementia or dementia with Lewy bodies (DLB). Seventy-five participants were to be randomized to 40 or 70 mg fosgonimeton or placebo, injected daily, against a primary outcome of a combined score of the ADAS-Cog 13 and P300 latency. By December 2022, the trial had stopped recruiting at 28 participants, and was expected to finish in May 2023. In a December 2023 press release, the company announced that this trial had failed on the primary endpoint. Results were presented at the March 2024 AD/PD conference (poster). An apparent improvement in the ADAS-Cog13 with 40 mg fosgonimeton compared to placebo was difficult to interpret due to baseline differences between the groups. The treatment appeared safe, but four of 10 participants in the higher dose arm dropped out because of adverse events. The most common side effect was injection site reaction.

From July to October 2022, a study of absorption, metabolism, and excretion of a single dose of [14C]-fosgonimeton was conducted in eight male volunteers.

Athira is developing additional small-molecule modulators of HGF/MET signaling that, unlike fosgonimeton, are orally available (Taylor et al., 2022).

In April 2024, the company completed a Phase 1 trial in 32 healthy adults of ATH-1020 (Berthiaume et al., 2022, poster).

In February 2024, company scientists reported beneficial effects of ATH-1105 in cell culture and in animal models of ALS (Berthiaume et al., 2024). In April 2024, the company began Phase 1 trial to test single and multiple daily doses of ATH-1105 in healthy volunteers in Dallas.

On September 3, 2024, Athira announced that the Phase 2/3 LIFT-AD trial had failed to meet its primary endpoint of improvement in the Global Statistical Test after 26 weeks treatment (press release, slides). Treatment also missed the key secondary endpoints of the ADAS-Cog11 or ADCS-ADL23. The company reported numerical improvements on cognitive and biomarker endpoints, but none were statistically significant. The trial had enrolled 549 participants, but the efficacy analysis covered only 287. Participants who took acetylcholinesterase inhibitors were excluded from the analysis, after results of the ACT-AD study (above) suggested that these drugs blocked the effects of fosgonimeton on HGF/MET signaling. In addition, the higher-dose arm had been discontinued mid-trial because of injection site reactions. Plans for an additional study have been canceled, the company told news sources (Endpoints News). Results presented at the October 2024 CTAD conference reiterated the negative results. Trial participants had milder dementia than intended, and the placebo group did not decline as expected. Both treated and placebo groups improved on the ADAS-Cog11 in the first six weeks, and neither group returned to baseline by the end of the trial. In a subgroup analysis, patients with more severe cognitive impairment had larger numerical improvements on the ADAS-Cog11. Plasma biomarkers NfL, GFAP, pTau181, and pTau217 all stayed stable in the placebo group, and improved on treatment. The change in pTau217 reached statistical significance. Twenty-two percent of enrollees dropped out of the trial due to bothersome injection site reactions. Besides these and brief eosinophilia, other adverse events were mainly mild and similar in placebo and treatment groups.

In September 2024, the company announced it would focus on development of the oral HGF/MET modulator ATH-1105 for ALS (press release).

For details on NDX-1017/ATH-1017, ATH-1020, and ATH-1105 trials, see clinicaltrials.gov.

Clinical Trial Timeline