Vivodyne

The journey of a potential new medicine from laboratory discovery to a patient’s bedside is, by many measures, fundamentally broken. We live in an era of unprecedented scientific understanding, yet this progress is consistently confronted by a stark reality: an astonishing 94% of drugs that demonstrate promise in preclinical animal testing ultimately fail when administered to humans. This is not merely a statistical hurdle; it is a systemic crisis of translation that costs society an estimated $90 billion each year in wasted resources and, more critically, incalculable human suffering and lost potential.

For decades, the cornerstone of preclinical drug development has been the animal model. Therapies are optimized in mice, rats, or even primates, under the prevailing assumption that these biological proxies will adequately predict human responses. Yet, time and again, this assumption shatters in the crucible of human clinical trials. Drugs deemed safe and effective in animals can manifest unexpected toxicities or prove inert in humans, their failures often rooted in human-specific physiological nuances that animal models, by their very nature, cannot capture. The human liver endothelium may partition a biologic differently; human cytokine signaling may react in unforeseen ways; a human protease might degrade a drug with greater efficiency than its animal counterpart. The catalogue of species-specific differences is vast and consequential.

This crisis of translation means that patients wait longer for effective treatments, and some vital medicines may never reach them, filtered out by models that falsely predict failure. Furthermore, the transformative promise of machine learning in revolutionizing drug discovery is itself constrained by the scarcity of large-scale, physiologically accurate human data.

AI models thrive on such data, yet they are starved. Attempts to bridge this gap with traditional in vitro models — such as 2D cell cultures, simple organoids, or early-generation “organs-on-chips” — have largely fallen short, hampered by limited biological realism, lack of vascularization, poor scalability, and reliance on artisanal, labor-intensive processes. Fundamentally, the engine of life sciences R&D remains an anachronism in an age of exponential technological advancement.

The story of Bertrand Might poignantly illustrates the human cost of this broken system. His symptoms began soon after birth — constant shaking, no eye contact, stalled development. After years of misdiagnoses, genetic sequencing in 2011 revealed Bertrand as the first person ever diagnosed with NGLY1 Deficiency, a genetic mutation preventing his cells from clearing waste.

Unfortunately, a diagnosis was only the beginning of further uncertainty. There was no known treatment. No scientist had studied the condition; our medical system had nothing to offer. Bertrand’s parents, Cristina and Matthew, took matters into their own hands.

Matthew, a computer scientist, immersed himself in rare disease research, launching a foundation and building a global patient registry. Realizing that developing a new drug from scratch would take a decade Bertrand didn’t have, Matthew pivoted to drug repurposing, leveraging his expertise to simulate and screen existing FDA-approved medications. Eventually, a common heartburn drug showed a sliver of hope, reducing Bertrand’s seizures. It wasn’t a cure, but it was something wrestled from a system not built to serve him—the result of years of relentless effort few families could replicate.

Bertrand’s story, while unique, is tragically common in its essence. Over 300 million people worldwide suffer from rare diseases. In the US, a disease is classified as rare if it affects fewer than 200,000 people. While the number of patients for any one of the 10,000 known rare diseases may be small, collectively they represent a profound failure of our healthcare system. These diagnoses—muscular dystrophies, metabolic disorders, genetic syndromes—are often debilitating and lifelong. Most are caused by a single, well-understood mutation. And yet, fewer than 5% have an FDA-approved treatment. The economics of drug development, costing nearly $2 billion and over a decade per drug, drive the industry’s focus towards conditions affecting millions, leaving patients like Bertrand underserved and highlighting an urgent need for a fundamentally better approach to discovering treatments.

Helena is a major investor in Vivodyne, a company founded on the premise that the future of medicine must be built on human data, from the very inception of a drug program to its final validation. Vivodyne seeks to improve the current model of drug discovery by creating a new status quo: human data, before human trials. The company has engineered a platform that discovers, develops, and de-risks new therapeutics by testing them on thousands of lab-grown, functional, and vascularized human organ tissues — automating the generation of human physiological insights at a scale and fidelity previously unimaginable.

Vivodyne’s mission is to make the development of new medicines dramatically more efficient, predictable, and ultimately, more successful. By allowing researchers to iterate, optimize, and validate drug candidates directly in human-relevant systems before the astronomical expense and ethical burden of clinical trials, Vivodyne offers a path to significantly cut the 94% failure rate, with the goal of increasing the number of life-saving therapies that reach patients.

Vivodyne’s disruption emerges from the integration of three pillars.

Engineered Human Tissues with Unprecedented Fidelity

First, Vivodyne engineers living, highly organized human tissues of unprecedented fidelity.

At the core of its platform are lab-grown human tissues cultivated from patient-derived primary human cells. These are not mere clusters of cells; they are complex, 3D structures, each containing hundreds of thousands to half a million cells, rendering them over a thousand times larger and physiologically relevant than typical organoids. Critically, Vivodyne’s tissues undergo developmental self-assembly, where controlled environmental cues guide cells to organize as they would in the human body, forming native multi-scale architecture comprised of organ-specific cell types, and, crucially, perfusable blood vessel networks.

This vascularization is essential, enabling realistic nutrient and drug delivery, waste removal, and immune cell ingress — functions indispensable for accurate biological modeling. Vivodyne has developed a comprehensive portfolio of over 22 proprietary human tissue models. These include bone marrow, airway tissues, solid tumors, placenta, liver, and pancreas, spanning oncology, immunity, and fibrosis, with fidelity consistently validated against native human organs and documented in over 25 publications in leading journals.

The HIVE: Vivodyne’s Automated Laboratory 

Second, Vivodyne deploys these tissues in their end-to-end automated laboratory – the HIVE, an autonomous robotic system that transforms biological research from a manual, artisanal craft into an industrial-scale, data-driven science.

Each HIVE system is a completely self-contained, BSL2+ (Biosafety Level 2+) human testing laboratory, roughly the size of an office cubicle. It fully automates every stage of experimentation: the precise injection of cells into proprietary Tissue Disks, the cultivation of up to 10,000 independent tissues to steady-state maturity, the administration of complex drug regimens, live and terminal 3D confocal imaging for tissue-scale phenomics with single-cell resolution, secretory sampling for proteomics, and tissue extraction for assays like single-cell sequencing.

This end-to-end, unattended automation operates for weeks, executing vast experimental arrays with unparalleled consistency and speed. With a rapidly expanding fleet, Vivodyne’s vision is to surpass annual human testing capacity of U.S. clinical trials, creating a “human data-center” that turns what was previously artisanal lab work into a scalable discovery engine.

Analysis of AI-Scale Human Data

The Vivodyne platform is architected to produce a torrent of rich, multi-modal data, capturing the effect of drugs and disease on human tissue phenotype, function, and gene signaling with single-cell resolution. This data generation is managed and interpreted by a sophisticated, cloud-hosted software stack.

HIVE Mind™ enables scientists to design complex, combinatorial studies with an intuitive drag-and-drop interface. HIVE Control™ and its advanced Scheduler then act as the “operating system,” translating these high-level study designs into hundreds of thousands of optimized, low-level robotic actions. All raw and derived data are funneled into the Data Depot™, a unified data mesh infrastructure. Finally, Vivodyne’s Analysis Pipeline™, leveraging advanced algorithms, computer vision, and machine learning models, automates the transformation of this raw data into quantifiable, actionable insights, enabling deep phenomic feature extraction, biomarker identification, and the prediction of clinical outcomes.

Helena is an investor in Vivodyne’s Series A financing, a commitment born from our conviction that their technology represents a turning point in humanity’s age-old quest to cure disease. Vivodyne is not merely improving an existing process; the comapny seeks to architect a fundamentally new, human-centric infrastructure for medical discovery that addresses one of the most expensive and impactful failures in modern science. We are excited to support Vivodyne’s vision to reduce the 94% clinical trial failure rate, tackling a multi-billion-dollar inefficiency and a profound barrier to delivering life-saving drugs to patients more rapidly and reliably. At scale, we believe technology platforms like Vivodyne offer a tangible approach to de-risk therapeutic development.

Furthermore, the current AI revolution in medicine is significantly bottlenecked by a lack of reliable data; Vivodyne is positioned to generate vast, complex, and physiologically relevant human datasets necessary to train sophisticated AI models, thereby unlocking new frontiers in predictive biology and drug design.

Vivodyne was co-foundered by Dr. Andrei Georgescu (CEO), whose doctoral work laid Vivodyne’s technical groundwork, and Dr. Dan Huh (CSO), a world-renowned bioengineering professor and the original inventor of “organs-on-a-chip” technology (whose lab’s IP is exclusively licensed to Vivodyne). The company’s management team includes Julie O’Shaughnessy (COO), with experience scaling global operations at AWS; and Dr. Anthony Bahinski (Chief Biotech Officer), previously Global Head of Safety Pharmacology at GSK and a member of FDA scientific advisory boards.

Vivodyne is not just building a better preclinical model. At scale, the company could help construct a future operating system for human biological discovery.

Vivodyne’s longer-term vision imagines a world where the 94% preclinical failure rate is a relic of the past and where new medicines for intractable diseases like Alzheimer’s, complex cancers, and rare genetic disorders are developed with far greater speed and certainty because they were designed, tested, and optimized in human-relevant systems from day one. Vivodyne’s approach points toward a future in which mysteries of human biology are systematically unraveled, not through slow, piecemeal experiments, but through massively parallel, AI-driven interrogation of living human tissues. The result is a world in which the path from biological insight to life-saving cure is dramatically shorter, more efficient, and more human.

Despite his parents’ heroic efforts, Bertrand Might ultimately succumbed to his disease in 2020, at the age of 12. Had he been born a decade later, his story might have unfolded differently. Not because his biology would have changed — but because the tools available to understand it, to test interventions with speed and human relevance, would have. Vivodyne represents a vision of that new toolset. It is not just a faster horse; it’s an update to the internal combustion engine of biological discovery, shrinking the gap between what we know and what we can test.

Vivodyne’s technology holds promise to generate future ripple effects across different challenges in drug development. With more efficient and cheaper ways to test at scale, smaller biotech companies could pursue more breakthrough therapies they may otherwise not be able to afford. Drugs combatting rarer diseases, rather than just blockbuster bets, could become cheaper and faster to bring to market. For centuries, we’ve approached medicine by poking, prodding, and guessing. In making deep, predictive biology accessible on demand, Vivodyne is working to break down the next foundational barrier in drug discovery — a transformation that isn’t distant or theoretical, but already underway.

On April 10, 2025, the U.S. Food and Drug Administration (FDA) announced a comprehensive roadmap aimed at significantly reducing, and ultimately replacing, animal testing in preclinical safety studies.

This strategic shift, building upon the legislative foundation of the FDA Modernization Act 2.0 passed in late 2022 , signals a new era where New Approach Methodologies (NAMs) will take center stage. The FDA’s plan emphasizes the adoption of scientifically validated, human-relevant methods such as organ-on-a-chip systems, advanced in vitro assays, and AI-driven computational modeling to enhance the predictive accuracy of drug development.

The FDA’s initiative is driven by the growing scientific recognition that animal models often poorly predict human outcomes, contributing to the high failure rates and costs in drug development. By prioritizing NAMs, the agency aims to streamline the path for safer and more effective therapies to reach patients faster, while also addressing ethical considerations surrounding animal use. The roadmap outlines a phased approach, initially focusing on areas like monoclonal antibody testing, with the long-term vision of making animal studies the exception rather than the norm within three to five years.

This decisive move by the FDA strongly validates the approach of companies like Vivodyne, which have been at the forefront of developing and deploying such human-relevant testing platforms. As the regulatory landscape realigns to favor these advanced methodologies, Vivodyne is well-positioned to lead the charge, empowering the pharmaceutical industry to develop new medicines with greater precision, efficiency, and ethical consideration, ultimately benefiting global public health.