This week on Biotech Nation, Dr. Martin Borch Jensen and Dr. Francisco LePort from Gordian Biotechnology share a bold new approach to fighting diseases of aging before they take hold. Instead of testing one drug at a time, their technology allows scientists to test hundreds of potential treatments at once, making drug discovery faster and more precise. This could change how we treat conditions like arthritis, heart failure, and fatty liver disease.
Decoding Diseases of Aging...Dr. Martin Borch Jensen and Dr. Francisco LePort, Gordian Biotechnology
Agency begins with knowing what to do – Martin Borch Jensen
Watch Martin Borch Jensen, Co-Founder and Chief Scientific Officer of Gordian Biotechnology, articulate a transformative approach to therapeutic discovery for complex diseases of aging. Borch Jensen challenges the traditional limitations of cell-based models by demonstrating how “mosaic screening” enables scientists to study disease within the context of living systems, ensuring that interventions are grounded in physiological truth. Highlights include an analysis of where disease truly “lives”—whether in the cell, the interface, or the organ—and a deep dive into how Gordian’s platform utilizes single-cell transcriptomics as a phenotypic readout for therapeutic effects.
This presentation explores the “How” and “Why” of creating target abundance to enable broad genetic agency. By utilizing a pooled in vivo screening platform, Gordian Biotechnology can test hundreds of gene targets simultaneously within an animal model that reflects the complexity of human disease, such as osteoarthritis or heart failure. Borch Jensen details the integration of AAV delivery mechanisms with single-cell gene expression measurement to bypass the “lies” often found in simplified in vitro environments. By connecting cellular data directly to clinical measurements like heart function and histology, Gordian is paving the way for a future where medical agency is no longer limited by a lack of known targets.
I. Executive Summary
The provided transcript focuses on Gordian’s application of in vivo single-cell screening to identify novel therapeutic targets for complex, age-related diseases (such as osteoarthritis and heart failure). The core thesis posits that complex diseases cannot be adequately modeled in vitro because the pathology often resides in the systemic interplay between diverse cell types (e.g., immune-fibroblast interactions) rather than within isolated cellular dysfunction.
The primary biological argument centers on Gordian’s proprietary technology platform: “pooled in vivo screening.” This methodology involves injecting adeno-associated virus (AAV) libraries encoding various genetic perturbations (targeting hundreds or thousands of genes up or down) into living animal models that have spontaneously developed the target disease (e.g., age-induced osteoarthritis in horses or mice). Single-cell transcriptomics are subsequently utilized as a phenotypic readout to reconstruct cellular pathways and determine which genetic interventions drive diseased cells back toward a healthy state.
The speaker highlights the computational challenge of ensuring transcriptomic data accurately predicts macro-physiological outcomes. Gordian addresses this by training predictive models on paired datasets linking transcriptomic profiles to clinical endpoints (e.g., echocardiography for diastolic function or histological collagen production in cartilage explants). The transcript concludes by demonstrating preclinical efficacy (Level D evidence) for two proprietary gene therapy targets (Omen 12 and 13) that successfully preserved or restored cartilage in mice with spontaneous osteoarthritis.
While the in vivo screening platform represents a sophisticated, high-throughput approach to target discovery, the presented data is strictly preclinical. A substantial translational gap exists between successfully identifying transcriptomic rescue in mosaic animal tissues and validating safe, systemic, and effective gene therapies in human clinical trials.
II. Insight Bullets
-
Disease Localization Paradigm: Complex diseases (like fibrosis or aging) often manifest at the interface between interacting cell types (e.g., immune activation triggering fibroblast collagen production) rather than originating solely within a single dysregulated cell.
-
In Vitro Limitations: Studying complex diseases in isolated cell cultures fails to capture the systemic, physiological environment necessary to identify valid therapeutic targets.
-
In Vivo Mosaic Screening: Gordian utilizes sub-saturating doses of AAV libraries to create a mosaic tissue environment within a living animal, allowing for the simultaneous in vivo testing of hundreds to thousands of genetic perturbations.
-
Spontaneous Disease Models: To maximize translational relevance, screening is performed in animal models that naturally develop the disease over time (e.g., aged horses or mice with spontaneous osteoarthritis) rather than relying on artificially induced models.
-
Transcriptomics as Phenotype: Single-cell RNA sequencing (transcriptomics) serves as the primary readout to assess how specific genetic perturbations alter cellular pathways in the disease environment.
-
Computational Phenotype Linking: Gordian employs proprietary paired datasets to map transcriptomic changes to actual physiological outcomes (e.g., heart function via echocardiography), ensuring cellular data predicts macro-level disease modification.
-
Preclinical OA Rescue: Specific proprietary gene therapy targets (identified via the platform as Omen 12 and 13) demonstrated the ability to rescue cartilage degradation in aged murine models of spontaneous osteoarthritis.
-
Target Abundance: The ultimate goal of the platform is to generate a massive, validated library of therapeutic targets for complex diseases, shifting medical treatment from a paradigm of scarcity (“getting lucky” that a treatment exists) to one of engineered abundance (genetic agency).
III. Adversarial Claims & Evidence Table
Note: Live search tool execution is disabled by system parameters for this interaction. Evidence grades and verdicts are synthesized from the current validated clinical and scientific database regarding standard preclinical discovery pipelines.
| Claim from Video | Speaker’s Evidence | Scientific Reality (Current Data) | Evidence Grade | Verdict |
|---|---|---|---|---|
| Complex diseases are best studied in vivo rather than in isolated cell cultures. | Conceptual argument regarding intercellular disease mechanisms (e.g., fibrosis). | It is widely accepted in the scientific community that in vitro models fail to recapitulate the complex microenvironment, immune signaling, and mechanical stressors of intact tissues, limiting translational success. | Level E (Scientific Consensus) | Strong Support |
| Pooled in vivo AAV screening coupled with single-cell transcriptomics can identify viable therapeutic targets. | Internal company platform data; mention of mosaic tissue analysis. | In vivo CRISPR/AAV screening combined with single-cell RNA-seq (Perturb-seq or similar methodologies) is an established, cutting-edge discovery tool validated in recent high-impact literature. | Level D | Plausible |
| Transcriptomic data can reliably predict macro-physiological outcomes (e.g., heart function, collagen production). | Internal scatter plots showing correlation between screen predictions and clinical readouts (explants, echocardiography). | While transcriptomic signatures can strongly correlate with physiological states, predicting complex organ-level function solely from single-cell RNA data requires highly robust, extensively validated machine learning models. Generalizability across distinct human populations remains unproven. | Level D | Speculative (Requires clinical validation) |
| Gordian’s targets (Omen 12/13) rescue spontaneous osteoarthritis. | Internal preclinical data slide showing cartilage improvement in 2-year-old mice. | AAV-mediated gene therapy for OA is an active area of preclinical research (e.g., targeting IL-1Ra or TGF-β). However, the specific targets (Omen 12/13) are proprietary, and efficacy/safety in humans is completely unknown. | Level D | Plausible (Translational Gap) |
Export to Sheets
IV. Actionable Protocol (Prioritized)
Note: The transcript primarily details a preclinical drug discovery platform rather than actionable medical interventions or lifestyle protocols for the end-user. Therefore, the protocol section focuses on the implications for researchers and biotech investors.
High Confidence Tier (Validated Methodologies)
- Adopt Spontaneous Animal Models: When studying diseases of aging (like osteoarthritis), prioritize animal models that develop the pathology naturally over time, rather than relying on chemically or surgically induced models, to improve clinical translation.
- Incorporate Microenvironment Context: Shift target discovery efforts for complex diseases away from monocultures and toward 3D organoids, co-cultures, or in vivo models that preserve immune-stromal interactions.
Experimental Tier (Cutting-Edge Research Protocols)
- Implement In Vivo Perturb-Seq: Utilize pooled viral libraries (AAV or lentivirus) for high-throughput genetic screening directly within the living tissue of interest, followed by single-cell transcriptomic readouts to map pathway responses.
- Develop Multi-Modal Predictive Models: Invest in computational biology pipelines that strictly pair single-cell RNA sequencing data with functional physiological readouts (e.g., imaging, histology) to train predictive algorithms for target selection.
Red Flag Zone (Translational Risks)
- Over-Reliance on Transcriptomics: Assuming that a transcriptomic shift toward a “healthy profile” in vitro or in an animal model will automatically translate to a functional cure in humans without rigorous systemic safety and physiological validation.
- Premature Clinical Extrapolation: Equating successful localized tissue rescue in a mosaic animal model (Level D) with imminent therapeutic viability in humans (Level A/B).
V. Technical Mechanism Breakdown
1. AAV Vector Delivery & Mosaicism Adeno-Associated Viruses (AAVs) are utilized as delivery vehicles because they can efficiently transduce both dividing and non-dividing cells with low immunogenicity. In Gordian’s platform, a “pooled library” of AAVs—where different viruses carry distinct genetic payloads (e.g., shRNA, CRISPR guides, or transgenes)—is injected. A “sub-saturating dose” is deliberately chosen so that only a fraction of the cells in the target tissue are transduced. This creates a “mosaic” tissue where perturbed cells are situated immediately adjacent to unperturbed (control) cells within the exact same physiological disease microenvironment, allowing for highly controlled in vivocomparisons.
2. Single-Cell Transcriptomics as a Phenotype Instead of looking at a single macro-readout (like tissue swelling), the platform uses single-cell RNA sequencing (scRNA-seq). After the in vivo perturbation, the tissue is dissociated into single cells. The mRNA from each cell is sequenced, providing a comprehensive snapshot of which genes are turned on or off. Because the AAV payload includes a unique barcode, researchers can match the specific genetic perturbation a cell received with its resulting transcriptomic profile.
3. Computational Linking to Physiology A cell might express RNA associated with a “healthy” state, but that doesn’t guarantee the whole organ functions better. To bridge this, Gordian builds predictive models. They take paired data—for example, the transcriptomic profile of a diseased rat heart and the actual pumping efficiency of that exact same heart measured via echocardiography (diastolic function). Machine learning algorithms are trained to recognize which specific patterns of RNA expression (gene sets) reliably predict improved organ function, allowing the researchers to filter the massive screening data for targets that matter physiologically.