Inmune Bio – XPro for Alzheimer’s Disease Phase 2

Questionable but Colorful Inmune Bio Graphic

Executive Summary

Inmune Bio has a phase 2 clinical readout in late June from XPro in Alzheimer’s Disease. Given the short timeframe for cognitive improvement, poor interaction with known AD etiology, and questionable pharmacokinetics, the readout will likely be negative. However, betting on a phase 2 trial is inherently very risky as companies have a penchant for spinning negative results into positive results. For example, Cassava Sciences, which faked its Phase 2 results, shot up after it announced “positive” Phase 2 results. Subgroup analysis is common, and most biotech companies refuse to kill Alzheimer’s programs before a spectacular failure in phase 3. We also want to give a shout out to Paurana Capital and their report on Inmune, which covers many of the points made in this article in more detail. It is worth a read. Check out their article at https://www.pauranacapital.ca/publications.

Given this, we think the best options are going long the September/July puts or manufacturing a synthetic put with higher downside exposure. We are short several September/July puts and synthetic puts.

Background Information

Alzheimer’s Disease

Despite billions of dollars and decades of research, the understanding of the disease etiology of Alzheimer’s Disease (AD) is poor. It is a complex, multifactorial disease, and it can be difficult to differentiate between different forms of dementia, neurodegenerative brain diseases, and AD. There are many current hypotheses about the pathogenesis of AD:

1. Amyloid Cascade Hypothesis

• The most established theory.

• Involves the accumulation of amyloid-β (Aβ) peptides, especially Aβ42, which aggregate into plaques.

• Aβ is derived from the amyloid precursor protein (APP) via cleavage by β-secretase and γ-secretase.

• Aggregated Aβ is thought to be neurotoxic, triggering downstream events such as:

  • Tau hyperphosphorylation

  • Neuroinflammation

  • Synaptic dysfunction

2. Tau Pathology

• Tau is a microtubule-associated protein. In AD, it becomes hyperphosphorylated and forms neurofibrillary tangles (NFTs).

• This disrupts neuronal transport and leads to cell death.

• The spread of tau pathology (starting in the entorhinal cortex and moving through the brain) correlates better with clinical symptoms than amyloid burden.

3. Neuroinflammation

• Activated microglia and astrocytes respond to Aβ and tau, releasing cytokines (e.g., IL-1β, TNF-α).

• Chronic inflammation contributes to synapse loss and neuron damage.

• Genes like TREM2, involved in microglial activation, have been implicated in AD risk.

4. Cholinergic Hypothesis

• AD patients show a loss of acetylcholine-producing neurons in the basal forebrain.

• This led to the development of cholinesterase inhibitors (e.g., donepezil) as symptomatic treatments.

• This is now seen as a downstream effect rather than a primary cause.

5. Genetic Factors

• Early-onset AD (before age 65) is often due to mutations in:

  • APP

  • PSEN1 (Presenilin-1)

  • PSEN2 (Presenilin-2)

• Late-onset AD is associated with:

  • APOE ε4 allele – increases risk and lowers the age of onset.

  • Other genome-wide association study (GWAS) hits: CLU, PICALM, BIN1, etc.

6. Mitochondrial Dysfunction and Oxidative Stress

• Impaired energy metabolism and increased production of reactive oxygen species (ROS) contribute to neuronal damage.

• Mitochondrial DNA damage and deficits in oxidative phosphorylation are observed in AD brains.

7. Vascular Contributions

• Vascular injury (e.g., small vessel disease) can reduce cerebral perfusion and Aβ clearance.

• There is an overlap between vascular dementia and AD pathology.

The recent decade of drug development and commercialization in AD has focused on the Amyloid Beta hypothesis. However, despite molecular success in current monoclonal antibodies in removing plaques, cognitive improvement shows minimal difference compared to placebo.

My literature review of AD underscores the importance of metabolism in AD. There is a much lower incidence of AD in physically active people. Additionally, many papers indicate abnormal glucose metabolism of brain cells. No current or investigational treatment aims to address the metabolic component of AD.

Treatments:

There are two categories of treatments for AD: Symptomatic treatments and Disease-Modifying Therapies.

FDA-Approved Symptomatic Treatments

These manage cognitive symptoms but do not slow disease progression.

1. Cholinesterase Inhibitors

• Mechanism: Prevent breakdown of acetylcholine to boost neurotransmission.

• Approved Drugs:

  • Donepezil (Aricept) – all stages of AD

  • Rivastigmine (Exelon) – mild to moderate AD

  • Galantamine (Razadyne) – mild to moderate AD

2. NMDA Receptor Antagonist

• Mechanism: Regulates glutamate to prevent excitotoxicity.

• Approved Drug:

  • Memantine (Namenda) – moderate to severe AD

3. Combination Therapy

• Donepezil + Memantine (Namzaric) – moderate to severe AD

FDA-Approved Disease-Modifying Therapies (DMTs)

These aim to alter the underlying pathophysiology, particularly by targeting amyloid-beta plaques.

4. Anti-Amyloid Monoclonal Antibodies

a. Aducanumab (Aduhelm) – Biogen

• Approved: June 2021 (via accelerated approval)

• Target: Aβ plaques

• Notes: Approval was highly controversial due to unclear clinical benefits; commercial use is limited.

b. Lecanemab (Leqembi) – Eisai/Biogen

• Approved: Full traditional approval in July 2023

• Target: Soluble protofibrils of amyloid-beta

• Indication: Mild cognitive impairment (MCI) or mild AD with confirmed amyloid pathology

• Clinical Evidence: CLARITY-AD trial showed ~27% reduction in cognitive decline (CDR-SB score) over 18 months

c. Donanemab (Kisunla) – Eli Lilly

• Approved: March 2025

• Target: N-terminal pyroglutamate Aβ plaques

• Indication: Early symptomatic AD (MCI or mild dementia with amyloid positivity)

• Clinical Evidence: TRAILBLAZER-ALZ 2 trial showed 35–40% slowing in clinical decline over 18 months

Development History in AD

The development history of AD is marred by many failures, with molecules targeting many different possible etiologies of AD. One of the more recent (and profitable for us) failures was Cassava Sciences’ simufilam, a molecule that targeted none of the known etiologies for AD.

From 2015-2020, many amyloid-targeting MAB, such as solanezumab, bapineuzumab, and crenezumab, failed in late-stage clinical trials due to lack of efficacy. BACED inhibitor trials were halted due to worsening cognitive decline and safety issues. Molecules targeting the Tau protein have had mixed results in early-stage trials.

There is limited data on molecules that directly target neuroinflammation in AD. NSAIDs have failed to show any benefit in randomized trials. Atuzaginstat, which targets the inflammatory pathogen P.gingivalis (the same pathogen responsible for periodontitis), failed to show any benefit. Mastinib, a tyrosine kinase inhibitor that targets microglia/mast cell modulation, seemed to show modest benefit in one subgroup in their phase 2 trials. Likewise, NE3107 by BioVie, an ERK/NF-kB inhibitor, hinted at a modest benefit in APOE E4 non-carriers. However, subgroup analysis is inherently faulty and almost always results in failed late-stage clinical trials and further sunk costs. Thus far, the clinical results from molecules that target neuroinflammation have been poor. Given the interlinked nature of inflammation and immune cascades, it is unlikely XPro will show superior results compared to these other molecules.

Inmune Bio and XPro

TNF exists in both soluble and transmembrane forms and binds to two TNF receptors, TNRF1 and TNFR2. These two receptors differ in their expression profiles and ligand affinities. TNFR1 binds to solTNF, while TNFR2 binds to tmTNF.

XPro

XPro1595 is an investigational, selective soluble tumor necrosis factor (solTNF) antagonist. Currently, available TNF inhibitors (infliximab, etanercept, etc.) inhibit both solTNF and tmTNF. This leads to serious side effects. The idea behind XPro1595 is to inhibit only solTNF, which will theoretically reduce proinflammatory immune cell migration and rescue impaired long-term potentiation in the AD brain.

In 2006, a doctor named Tobinick authored a paper in which he claimed that perispinal etanercept (Enbrel) improved cognitive function in AD patients. The paper received strong pushback from the medical community, and frankly, few believed his claims. However, the idea seems to be the inspiration behind XPro’s development.

A phase 1 study demonstrated improvements in CSF inflammation biomarkers. (NCT03943264)

XPro1595 was administered subcutaneously once weekly over 12 weeks. Key findings included:

• Reduction in Neuroinflammation: MRI assessments showed a significant decrease in white matter-free water, a biomarker of neuroinflammation, particularly in the arcuate fasciculus, a brain region associated with language and memory.

• CSF Biomarker Improvements: Proteomic analysis of cerebrospinal fluid (CSF) indicated reductions in inflammatory cytokines and changes in proteins related to synaptic function and neurodegeneration.

• Safety and Tolerability: XPro1595 was well-tolerated, and no serious adverse events were reported.

To us, it is still unclear if there is significant blood-brain barrier (BBB) penetration. In a 2018 paper by MacPherson et al. they saw brain differences in immune cell profiles, indicating it crosses the blood-brain barrier. In a 2015 paper on a rat Parkinson model, peripheral dosing of 10 mg/kg (probably around 5 mg based on average rat weight). It should be noted that after administration, the concentration in the CNS was in the ng/mL range, which indicates poor BBB penetrance. However, that amount may be sufficient to exert an effect.

Clinical Trials

Phase 2 Study Design

The phase 2 study investigating XPro is called MINDful.

Study Overview

• Design: Randomized, double-blind, placebo-controlled trial.

• Participants: 208 individuals aged 50–85 diagnosed with either mild cognitive impairment (MCI) or mild AD.

• Intervention: Participants were randomized in a 2:1 ratio to receive either 1.0 mg/kg of XPro1595 or a placebo via subcutaneous injection once weekly for 23 weeks.

• Enrollment Criteria: All participants met at least one of the following inflammation-related biomarkers:

  • High-sensitivity C-reactive protein (hsCRP) > 1.5 mg/L

  • Erythrocyte sedimentation rate (ESR) > 10 mm/hr

  • Hemoglobin A1c (HbA1c) > 6.0%

  • Presence of at least one APOE ε4 allele

Objectives and Endpoints

• Primary Endpoint: Change from baseline in cognitive performance measured by the Early and Mild Alzheimer's Cognitive Composite (EMACC), a tool designed for sensitivity in early AD trials.

• Secondary Endpoints: Assessments include Clinical Dementia Rating–Sum of Boxes (CDR-SB), Everyday Cognition (E-Cog), Activities of Daily Living (ADL), Neuropsychiatric Inventory (NPI), blood biomarkers, and neuroimaging outcomes.

Baseline Findings

• Demographics: The average participant age was 72 years.

• Genetic Risk: Approximately 69.2% of participants were carriers of the APOE ε4 allele.

• Inflammation Markers: About 64.4% met the criteria for more than one enrichment biomarker.

• Screening: The trial had a 72% screen failure rate, primarily due to disease severity as measured by the Mini-Mental State Examination (MMSE).

Analysis and Discussion

The near-term readout in the MINDful study will likely be negative.  In 24 weeks, it is doubtful a cognitive improvement will be noticed. Additionally, the entire mechanism is based on questionable science. TNF-a seems to be downstream of many of the primary indicators of AD. The company does not have much cash runway; this one could get ugly if the trial is negative.

However, like many other early-stage AD trials, it is likely some form of subgroup analysis will be conducted that results in a “positive” outcome. Cassava Sciences faked their phase 2 data before being caught. Betting on phase 2 trials is inherently very risky as companies can spin negative data into positive with many different techniques.

Conclusion

During their recent earnings call, Inmune announced that the readout will be in late June. Therefore, September/July options are best. Naked shorting is too risky, given what happened with Cassava after they announced their faked phase 2 data.  Synthetic puts or at-the-money options seem the best way to bet on this. We are short several September/July puts and a few synthetic puts.