What are the key takeaways from this Beyond Biotech episode?

Key insights include: **Model selection risk:** Reaching for convenient, overexpressing mouse models rather than disease-relevant ones is a primary driver of CNS drug failure. Less than 10% of CNS candidates entering Phase 1 ever reach market. Biotechs under budget and timeline pressure should prioritize model fit over speed, as mismatched models generate efficacy data that becomes irrelevant in the clinic.; **Multi-pathology model design:** GEM PharmaTech's F83T Alzheimer's model combines APP and PSEN1 mutations for amyloid pathology with a MAPT mutation for tau pathology, capturing neuroinflammation and memory decline simultaneously. Drug developers should demand models that replicate multiple co-occurring disease mechanisms rather than single-pathway constructs that miss the full clinical picture.; **Disease staging alignment:** Most preclinical studies treat animals before symptoms appear, while human trials enroll patients with established pathology. Drug developers should design studies across multiple disease stages and align animal intervention timing with the patient population their clinical trial will actually enroll, or risk flattening efficacy signals entirely.

What did Better Mouse Models discuss on Beyond Biotech?

Brandy Wilkinson, CEO of GEM PharmaTech, and neuroscience pipeline leader Ricky Feng explain how the company's library of over 25,000 genetically engineered mouse models addresses neurology's sub-10% clinical success rate by building preclinical tools that mirror human disease biology, biomarkers, and disease staging. Key topics include: **Model selection risk:** Reaching for convenient, overexpressing mouse models rather than disease-relevant ones is a primary driver of CNS drug failure. Less than 10% of CNS candidates entering Phase 1 ever reach market. Biotechs under budget and timeline pressure should prioritize model fit over speed, as mismatched models generate efficacy data that becomes irrelevant in the clinic.; **Multi-pathology model design:** GEM PharmaTech's F83T Alzheimer's model combines APP and PSEN1 mutations for amyloid pathology with a MAPT mutation for tau pathology, capturing neuroinflammation and memory decline simultaneously. Drug developers should demand models that replicate multiple co-occurring disease mechanisms rather than single-pathway constructs that miss the full clinical picture..

How long is this episode of Beyond Biotech?

This episode is 40 minutes long. SignalCast provides an AI-generated summary so you can get the key insights in about 3 minutes.

Beyond Biotech

De-risking neurology drug development with better mouse models

June 26, 2026

40 min episode · 2 min read

Better Mouse Models

Episode

40 min

Read time

2 min

Topics

Leadership, Design & UX, Software Development

AI-Generated Summary

Published Jun 26, 2026

Key Takeaways

✓Model selection risk: Reaching for convenient, overexpressing mouse models rather than disease-relevant ones is a primary driver of CNS drug failure. Less than 10% of CNS candidates entering Phase 1 ever reach market. Biotechs under budget and timeline pressure should prioritize model fit over speed, as mismatched models generate efficacy data that becomes irrelevant in the clinic.
✓Multi-pathology model design: GEM PharmaTech's F83T Alzheimer's model combines APP and PSEN1 mutations for amyloid pathology with a MAPT mutation for tau pathology, capturing neuroinflammation and memory decline simultaneously. Drug developers should demand models that replicate multiple co-occurring disease mechanisms rather than single-pathway constructs that miss the full clinical picture.
✓Disease staging alignment: Most preclinical studies treat animals before symptoms appear, while human trials enroll patients with established pathology. Drug developers should design studies across multiple disease stages and align animal intervention timing with the patient population their clinical trial will actually enroll, or risk flattening efficacy signals entirely.
✓Blood-brain barrier humanization: Only 0.1% of injected biologic doses typically reach the brain. GEM PharmaTech's humanized transferrin receptor mouse model tests whether receptor-mediated transcytosis strategies actually achieve sufficient brain exposure before costly CSF or human imaging studies. Additional humanized models targeting CD98HC and IGF1R are in development to cover a wider range of delivery mechanisms.
✓Integrated biomarker readouts: Linking behavioral test results directly to molecular and pathological data from the same animal cohort strengthens go/no-go decisions. GEM PharmaTech collects CSF and blood samples post-behavior testing to measure the same diagnostic markers used in human trials, such as phosphorylated tau ratios, creating a dataset that speaks the same language as clinical endpoints.

What It Covers

Brandy Wilkinson, CEO of GEM PharmaTech, and neuroscience pipeline leader Ricky Feng explain how the company's library of over 25,000 genetically engineered mouse models addresses neurology's sub-10% clinical success rate by building preclinical tools that mirror human disease biology, biomarkers, and disease staging.

Key Questions Answered

•Model selection risk: Reaching for convenient, overexpressing mouse models rather than disease-relevant ones is a primary driver of CNS drug failure. Less than 10% of CNS candidates entering Phase 1 ever reach market. Biotechs under budget and timeline pressure should prioritize model fit over speed, as mismatched models generate efficacy data that becomes irrelevant in the clinic.
•Multi-pathology model design: GEM PharmaTech's F83T Alzheimer's model combines APP and PSEN1 mutations for amyloid pathology with a MAPT mutation for tau pathology, capturing neuroinflammation and memory decline simultaneously. Drug developers should demand models that replicate multiple co-occurring disease mechanisms rather than single-pathway constructs that miss the full clinical picture.
•Disease staging alignment: Most preclinical studies treat animals before symptoms appear, while human trials enroll patients with established pathology. Drug developers should design studies across multiple disease stages and align animal intervention timing with the patient population their clinical trial will actually enroll, or risk flattening efficacy signals entirely.
•Blood-brain barrier humanization: Only 0.1% of injected biologic doses typically reach the brain. GEM PharmaTech's humanized transferrin receptor mouse model tests whether receptor-mediated transcytosis strategies actually achieve sufficient brain exposure before costly CSF or human imaging studies. Additional humanized models targeting CD98HC and IGF1R are in development to cover a wider range of delivery mechanisms.
•Integrated biomarker readouts: Linking behavioral test results directly to molecular and pathological data from the same animal cohort strengthens go/no-go decisions. GEM PharmaTech collects CSF and blood samples post-behavior testing to measure the same diagnostic markers used in human trials, such as phosphorylated tau ratios, creating a dataset that speaks the same language as clinical endpoints.

Notable Moment

Ricky Feng draws a striking analogy for why early-stage brain disease remains so poorly understood: studying neurodegeneration from patient data alone is like trying to understand an entire movie by watching only its final five minutes, because pathological cascades run silently for years before any clinical presentation.

Know someone who'd find this useful?

Episode Transcript

Hello, and welcome to Beyond Biotech, the weekly podcast from La BioTech. I'm Dylan Kysain, and this is episode 203 for the podcast. Today, I'm delighted to welcome Brandy Wilkinson, CEO of GEM PharmaTech, and Ricky Feng, the company's neuroscience pipeline leader. Brandy brings more than two decades of preclinical operations leadership from Covance, Explora Bio Labs, and the Jackson Laboratory, and now steers GEM PharmaTech's North American growth. Ricky oversees a neuroscience portfolio, translating clinical insights into genetically engineered mouse models that are proving far more predictive than traditional platforms. Today, we're focusing on a persistent challenge in drug development, neurology's stubbornly high clinical failure rates. Gem PharmaTech is tackling this head on with proprietary models for Alzheimer's, Parkinson's, and blood brain barrier transport that better mirror human disease biology. We explore why mouse models matter more in neurology than in other fields, the design principles behind these next generation tools, and how smarter preclinical partnerships can help therapeutic developers derisk programs earlier. Brandy, you earned your PhD in neuroscience. You built a career leading large scale preclinical operations. Tell me a little bit about this journey. What drew you to GEM Pharmatech? Oh, yeah. Thank you for having me. It's a pleasure to be here. And sure, it's it's been quite a winding road. Like you you said, I started with a PhD in neuroscience, focused, my postdoc research, primarily in Alzheimer's disease. So I'm really a scientist at heart. But, over time, I took a detour, decided I I really like the operational and business side of life science. Really interested in building things, scaling teams, you know, solving problems for for the researchers. So that shift took me into the, preclinical CRO world where I was lucky enough to have some really formative opportunities, really great mentors and things that got me into directing preclinical oncology programs, working closely with biotech and pharma on everything from when PDX models first came out to immunotherapy platforms as those things evolved. And each stop sort of added a new layer of either scientific depth or or business development, client relationships that grew and sometimes continued. And I found myself increasingly drawn to that challenge of building, and starting, businesses and scaling them. What really drew me to GEM Pharmatech, though, at you know, as my career grew is I really felt it's somewhat of a natural chapter in what I've been doing before, but it was really the scientific rigor that, GEM PharmaTech has and the platform that they had built. And it's an incredibly strong team that science is incredibly exciting and good and and something I was really interested in being a part of to grow something meaningful and, really backed by a scientific foundation that I'm I'm extremely proud to be a part of. Brandy, maybe give the listeners a quick overview of what Gem PharmaTech does, what it what its mission is. Yeah. Thanks. GemPharmatech was actually founded in 2017. So still a relatively …

Get the full transcript (5,944 words) + summary by email — free

One-time email with the complete transcript and AI summary of this episode. No account needed.

One email, no spam. We’ll also show you what SignalCast does.

Browse all Beyond Biotech transcripts →