How successful are AI-discovered drugs in clinical trials? A first analysis and emerging lessons.

AI techniques are making inroads into the field of drug discovery. As a result, a growing number of drugs and vaccines have been discovered using AI. However, questions remain about the success of these molecules in clinical trials. To address these questions, we conducted a first analysis of the clinical pipelines of AI-native Biotech companies. In Phase I we find AI-discovered molecules have an 80-90% success rate, substantially higher than historic industry averages. This suggests, we argue, that AI is highly capable of designing or identifying molecules with drug-like properties. In Phase II the success rate is ∼40%, albeit on a limited sample size, comparable to historic industry averages. Our findings highlight early signs of the clinical potential of AI-discovered molecules.

Amyloid precursor protein and amyloid precursor-like protein 2 have distinct roles in modulating myelination, demyelination, and remyelination of axons.

The identification of factors that regulate myelination provides important insight into the molecular mechanisms that coordinate nervous system development and myelin regeneration after injury. In this study, we investigated the role of amyloid precursor protein (APP) and its paralogue amyloid precursor-like protein 2 (APLP2) in myelination using APP and APLP2 knockout (KO) mice. Given that BACE1 regulates myelination and myelin sheath thickness in both the peripheral and central nervous systems, we sought to determine if APP and APLP2, as alternate BACE1 substrates, also modulate myelination, and therefore provide a better understanding of the events regulating axonal myelination. In the peripheral nervous system, we identified that adult, but not juvenile KO mice, have lower densities of myelinated axons in their sciatic nerves while in the central nervous system, axons within both the optic nerves and corpus callosum of both KO mice were significantly hypomyelinated compared to wild-type (WT) controls. Biochemical analysis demonstrated significant increases in BACE1 and myelin oligodendrocyte glycoprotein and decreased NRG1 and proteolipid protein levels in both KO brain tissue. The acute cuprizone model of demyelination/remyelination revealed that whereas axons in the corpus callosum of WT and APLP2-KO mice underwent similar degrees of demyelination and subsequent remyelination, the myelinated callosal axons in APP-KO mice were less susceptible to cuprizone-induced demyelination and showed a failure in remyelination after cuprizone withdrawal. These data identified APP and APLP2 as modulators of normal myelination and demyelination/remyelination conditions. Deletion of APP and APLP2 identifies novel interplays between the BACE1 substrates in the regulation of myelination.

Leukemia inhibitory factor protects axons in experimental autoimmune encephalomyelitis via an oligodendrocyte-independent mechanism.

Leukemia inhibitory factor (LIF) and Ciliary Neurotrophic factor (CNTF) are members of the interleukin-6 family of cytokines, defined by use of the gp130 molecule as an obligate receptor. In the murine experimental autoimmune encephalomyelitis (EAE) model, antagonism of LIF and genetic deletion of CNTF worsen disease. The potential mechanism of action of these cytokines in EAE is complex, as gp130 is expressed by all neural cells, and could involve immuno-modulation, reduction of oligodendrocyte injury, neuronal protection, or a combination of these actions. In this study we aim to investigate whether the beneficial effects of CNTF/LIF signalling in EAE are associated with axonal protection; and whether this requires signalling through oligodendrocytes. We induced MOG35₋55 EAE in CNTF, LIF and double knockout mice. On a CNTF null background, LIF knockout was associated with increased EAE severity (EAE grade 2.1±0.14 vs 2.6±0.19; P<0.05). These mice also showed increased axonal damage relative to LIF heterozygous mice, as indicated by decreased optic nerve parallel diffusivity on MRI (1540±207 µm²-/s vs 1310±175 µm²-/s; P<0.05), and optic nerve (-12.5%) and spinal cord (-16%) axon densities; and increased serum neurofilament-H levels (2.5 fold increase). No differences in inflammatory cell numbers or peripheral auto-immune T-cell priming were evident. Oligodendrocyte-targeted gp130 knockout mice showed that disruption of CNTF/LIF signalling in these cells has no effect on acute EAE severity. These studies demonstrate that endogenous CNTF and LIF act centrally to protect axons from acute inflammatory destruction via an oligodendrocyte-independent mechanism.

Experimental autoimmune encephalomyelitis (EAE) IN C57Bl/6 mice is not associated with astrogliosis.

The C57Bl/6 mouse is the preferred host for the maintenance of gene deletion mutants and holds a unique place in investigations of cytokine/chemokine networks in neuroinflammation. It is also susceptible to experimental autoimmune encephalomyelitis (EAE), a multiple sclerosis (MS)-like disease commonly used to assess potential MS therapies. Investigations of glial reactivity in EAE have revealed hitherto undescribed astroglial responses in this model, characterized by progressively diminishing glial fibrillary acidic protein and aquaporin-4 immunostaining, from early disease. These observations show that astrocyte responses vary with the EAE paradigm and are an important pathological criterion for disease mapping and therapy evaluation.

The astrocytic response in early experimental autoimmune encephalomyelitis occurs across both the grey and white matter compartments.

An unexpectedly prominent aspect of murine experimental autoimmune encephalomyelitis is pre-onset astrocyte reactivity. Further examination of this phenomenon in the spinal cord demonstrates that grey matter, as well as white matter astrocytes, change their morphology and cell density from the earliest disease manifestation. Comparison of the two compartments reveals that, whereas white matter changes are rostro-caudally consistent, grey matter reactivity is spatially restricted and of varying amplitude between spinal cord levels. These data strongly suggest that in neuroinflammation early, cross-compartmental recruitment of astrocytes occurs, but with different expression patterns.

Astrocyte-associated axonal damage in pre-onset stages of experimental autoimmune encephalomyelitis.

Recent studies of axon-glia and glia-glia communication have emphasized interactivity and interdependence between central nervous system (CNS) components. Concurrently, data from imaging, biochemical, and morphological studies have changed the view of multiple sclerosis (MS) from a neuroinflammatory condition with primary demyelination to one in which cumulative axonal damage drives progression. We therefore studied axonal damage in the context of inflammation and glial responses, from the pre-clinical to onset stage of murine experimental autoimmune encephalomyelitis (EAE), an established MS model. We report three major findings: (1) the first evidence of axonal injury before significant T-cell entry into the parenchyma, (3) coincidence of the earliest manifestation of axonal damage and astrocytic responses, and (3) an association between accumulation of axonal and astrocytic changes and specific forms of MS. These data demonstrate the relationship between the initiation of axonal injury and early inflammation. Significantly, we show that, in common with a growing number of neurodegenerative conditions, the pathology of murine EAE is characterized by early active contribution from astrocytes. This marks a change in the understanding of the role of astrocytes in MS pathogenesis and has important implications for the development of neuroprotective strategies.

Early glial responses in murine models of multiple sclerosis.

Investigations of functional interactions among axons and glia over the last decade have revealed the extent and complexity of glial-neuronal and glial-glial communication during development, adult function and recovery from injury. These data have profound implications for the understanding of central nervous system (CNS) disorders, which until recently, have been classified as either neuronal or glial diseases. Re-evaluation of the pathological processes in a number of conditions has clearly shown involvement of both neurons and glia in early pathology. In multiple sclerosis (MS), the myelin sheath has traditionally been regarded as the primary target. However, recent evidence has clearly demonstrated axonal damage in new lesions. We have addressed the question of the role of axonal pathology in early MS by using well-characterized murine models for the relapsing-remitting (RR) or the primary progressive (PP) forms of the disease. We performed a histopathological survey of the CNS, following induction of the disease, to determine the timing of appearance, as well as the development of lesions. Then we analysed the relationship between inflammation, demyelination and axonal damage together with responses from astrocytes and microglia in each model from the earliest evidence of inflammation. We found that axonal damage begins well ahead of the appearance of motor symptoms. Pathology appears to be more closely related to the degree of inflammation than to demyelination. We also show that early astrocyte responses and the degree of axonal loss are markedly different in the two models and relate to the severity of pathology. These data support the now widely accepted hypothesis that axonal damage begins early in the disease process, but also suggest modulation of axonal loss and disease progression by the astrocytic response.

Herpes simplex virus-specific CD8+ T cells can clear established lytic infections from skin and nerves and can partially limit the early spread of virus after cutaneous inoculation.

HSV infects skin or mucosal epithelium as well as entering the sensory nerves and ganglia. We have used TCR-transgenic T cells specific for the immunodominant class I-restricted determinant from HSV glycoprotein B (gB) combined with a flank zosteriform model of infection to examine the ability of CD8+ T cells to deal with infection. During the course of zosteriform disease, virus rapidly spreads from the primary inoculation site in the skin to sensory dorsal root ganglia and subsequently reappears in the distal flank. Virus begins to be cleared from all sites about 5 days after infection when gB-specific CD8+ T cells first appear within infected tissues. Although activated gB-specific effectors can partially limit virus egress from the skin, they do so only at the earliest times after infection and are ineffective at halting the progression of zosteriform disease once virus has left the inoculation site. In contrast, these same T cells can completely clear ongoing lytic replication if transferred into infected immunocompromised RAG-1-/- mice. Therefore, we propose that the role of CD8+ T cells during the normal course of disease is to clear replicating virus after infection is well established rather than limit the initial spread of HSV from the primary site of inoculation.