An All-In-Human Approach to Neurodegenerative Diseases
How an all-in-human approach to treating neurodegenerative diseases could succeed where animal models have failed
By Robert Scannevin, PhD, Chief Scientific Officer, Verge Genomics
In recent years scientists have made amazing advances in our understanding of neurological diseases like amyotrophic lateral sclerosis (ALS) and Parkinson’s disease (PD). The catch? Much of this work has been in animal models and cell systems that don't recapitulate human disease and therefore have not translated to clinical successes and disease-modifying therapies for humans. How can we better translate scientific discovery to the clinic to treat people impacted by these devastating neurological diseases?
This fundamental question drives our team at Verge Genomics. Alice Zhang, our founder and CEO, started Verge to understand, and solve for, the translational gap in the drug discovery process for complex neurodegenerative diseases. While researchers had begun to leverage new targets emerging from genome-wide association studies (GWAS) as intervention points, a wealth of functional genomic data remained largely untapped. We asked: what if we could integrate these multiple layers of data into a single platform built on all-human data, with the goal of identifying disease signatures – the networks of genes that are perturbed in the diseased state compared with the healthy state?
Verge’s all-in-human platform is predicated on the idea that we start with human data to identify targets that will be relevant for treating human diseases. We believe this will increase clinical translatability, enable us to avoid the clinical failures of the past and speed the development of new and better therapies for patients. To understand why I believe that deciphering biology using human datasets is the most significant opportunity in drug discovery and development today, let’s take a deeper dive into the Verge platforms to better understand how we leverage this unique approach.
Target discovery begins with human data
Verge’s platform starts with high quality post mortem brain tissue samples from patients diagnosed with neurodegenerative disease and aged matched healthy control subjects sourced from numerous brain banks and academic institutes around the world. We perform whole genome and RNA sequencing on those samples, then apply artificial intelligence (AI) - and machine learning (ML) - based computational biology techniques to look for networks of genes – not just individual genes – that are dysregulated in tissues from patients but not in healthy controls.
These networks take a systems biology view of disease pathways and incorporate multiple known genetic drivers of disease, such as C9orf72 or TDP-43 mutations in ALS. Rather than aiming to correct a specific mutation, we look for the functional consequences of these genetic drivers, which manifest as a network of dysregulated genes. This approach first allows us to identify disease-relevant pathways directly impacted by genetic drivers, but then expands this view as we gain a deeper understanding of the dysregulation that propagates throughout the network. We do find considerable overlap in dysregulated networks between genetic and sporadic patient samples, which allows us to expand our approach into broader patient populations that have common pathological features.
Deep bioinformatic analysis of these networks allow us to pinpoint master regulator genes that we believe can impact the disease-associated signature, and these “nodes” are nominated as candidate targets. We are completely agnostic in terms of the types of targets we will ultimately develop; but because the list often includes 100 or more candidates, we have had to develop ways to prioritize how we approach validating them.
Our gold standard system for validating targets uses patient-derived cell cultures. To generate those cultures, we start with cells that are plentiful and easily gathered from the patient – typically fibroblasts – and then “reprogram” them into specific types of central nervous system (CNS) cells. The resulting cultures can consist solely of human neurons or are a 3D “brain-in-a-dish” culture containing human neurons and other CNS cell types. We believe this ‘all-in-human’ system enables better predictions about a target’s role in a disease-relevant process. It makes sense that studying targets derived from human patient samples are best evaluated in a system derived from human patient cells rather than in a mouse or other type of cell model in which aspects of a disease process has been artificially introduced.
Modalities and drug discovery models
Once a target is validated, we initiate a medicinal chemistry program to discover and develop small molecule therapeutics. Verge has emphasized small molecule discovery, but our therapeutic approach is modality-agnostic and dictated by the nature of the target. Some targets are better suited to modalities such as antisense oligonucleotides (ASOs) or gene therapy, and for those targets we form strategic partnerships, such as we’ve done with Eli Lilly in ALS.
While our target validation platform focuses on human cell models, we do utilize animal studies as part of the drug discovery process. A critical aspect of this is understanding a drug’s pharmacokinetic and pharmacodynamic properties and evaluating this in animals can help us predict how drugs will be distributed and metabolized in humans and also provide insights into drug safety. These types of studies also help prioritize compounds to identify leads and development candidates.
We also test our candidate compounds in animal models of disease. While these models don’t offer great translatability or predict the ability to cure human disease, they can have utility in understanding target engagement. If administration of a candidate compound can hit the anticipated target in the right quantities for the right period of time and modulate disease-relevant biology in the model system, as measured by biomarkers, this can provide some confidence the compound will have a similar effect in humans. For example, in a mouse model of ALS, our clinical candidate can reduce the levels of a biomarker known as neurofilament light chain (NfL), which is a surrogate marker for neurodegeneration. If we can observe similar lowering of NfL in our clinical studies, this may provide an early indication if our drug is impacting the disease.
Overall, Verge believes our all-in-human platform increases the probability of clinical success, because we are identifying and validating targets and testing compounds in patient tissues and patient-derived CNS cell cultures, supplemented with information from animal models. Our platform is grounded in human biology and the dysregulation that occurs in human disease, which we feel is most relevant to addressing pathological processes in patients. Our approach also identifies targets that regulate disease-relevant networks in a broad population of patients.
As my understanding of the Verge all-in-human platform grew, so did my desire to be a part of this transformative company. Getting to know the visionary leaders and innovative scientists that make up the Verge team further solidified this desire, and I am inspired daily by what we are accomplishing. Neurodegenerative disease research desperately needs to better understand the complex pathobiology that underlies these devastating disorders. The Verge platform is providing these critical insights and discovering disease-relevant targets and therapeutics, and that’s exactly why I’m so excited about our potential to impact human health.