From Cell State Reprogramming To The Clinic: Turning The Regulatory Genome Into RNA Medicines (Pt. 2)

A conversation with Samir Ounzain, Ph.D., CEO & Scientific Cofounder, HAYA Therapeutics

In Part 1 of his conversation with Life Science Connect Acquisition Editor Michael Soloway, HAYA Therapeutics CEO and scientific cofounder Samir Ounzain, Ph.D., argued that the future of RNA medicine may lie not in targeting individual proteins but in reprogramming the regulatory networks that determine cellular identity. That vision is beginning to move beyond theory. In Part 2, Ounzain discusses how regulatory genome biology is being translated into drug discovery, from identifying high-value long non-coding RNA (lncRNA) targets and leveraging artificial intelligence to building clinically relevant therapeutics for fibrosis and other chronic diseases. He also shares why he believes the industry is approaching an inflection point, where cell state reprogramming could emerge as a new therapeutic paradigm alongside traditional protein-targeted medicines.

Fibrosis remains a major unmet medical need across multiple therapeutic areas. What has made fibrotic diseases so difficult to treat, and what insights has regulatory genome biology revealed about the underlying drivers of these conditions?

SO: Fibrosis is hard to treat because it isn’t caused by a single broken protein but by tissue sentinel cells called fibroblasts getting stuck in overdrive. While fibroblasts normally sense damage and coordinate healing, in chronic disease, they become trapped in a maladaptive state that continuously drives scarring long after the original trigger has passed. Traditional drug discovery has largely tried to interrupt fibrosis by targeting broadly expressed signaling proteins. While relevant, their systemic inhibition can create narrow therapeutic windows and dose-limiting toxicities.

Regulatory genome biology offers more precision by mapping fibroblast cell states across tissues and diseases. We can identify and target regulatory RNAs that stabilize fibrosis-driving cell states in a specific cellular and tissue context.

The lncRNA WISPER is a good example. It is a cardiac fibroblast-enriched lncRNA that coordinates fibrotic gene expression programs in the heart. In patient tissue from hypertrophic cardiomyopathy, WISPER is elevated and associated with fibrosis severity. Our lead investigational clinical candidate, HTX-001, is designed to target WISPER and intended to release the pathological fibroblast state, while preserving normal biology elsewhere. It is now being evaluated in a Phase 1 clinical trial.

Drug discovery efforts have historically struggled to identify targets that are both biologically relevant and therapeutically actionable. How can regulatory genome approaches improve target identification and validation?

SO: The regulatory genome approach allows us to begin with the disease-driving cell state rather than with a predefined target class or a single differentially expressed gene. The challenge is identifying regulatory nodes that are causal and clinically actionable.

At HAYA, we address this through HAYAtlas, our tissue- and disease-resolved map of the regulatory genome. It integrates multimodal human single-cell data sets, genomic, epigenomic, transcriptomics, 3D genome conformation, lncRNA networks, and transposable-element activity. We then use computational and machine learning approaches to triangulate across multiple biological layers and model systems to identify high-hierarchy control points with stronger evidence of causality.

We also build actionability into target selection from the outset. That includes tissue-, cell-type-, and disease-state specificity; biological relevance; genetics and patient tissue evidence; translatable pharmacodynamic biomarkers; cross-species functional conservation; and the ability to design a potent, selective RNA therapeutic.

Ultimately, computational prediction is only the beginning. We validate targets through a rigorous direct perturbation assessment in vitro and in in vivo-relevant models. The aim is to create a coherent causal chain from human disease biology to regulatory target, to therapeutic intervention, to biomarker, with the ultimate goal of delivering clinical benefit.

One of the challenges often discussed in RNA therapeutics is delivery. How do delivery considerations differ when developing therapies that target lncRNAs and regulatory pathways compared with other RNA-based modalities?

SO: Delivery in RNA therapeutics comes down to two key questions: can the drug reach the relevant cell, and can it act selectively once it gets there?

Many disease-driving lncRNAs are restricted to particular tissues, cell types, and active pathological states. Its biology per se delivers a layer of precision. So even if a therapeutic ASO distributes to clearance organs, the intended on-target pharmacology is limited when the disease-driving lncRNA is absent or expressed only at very low levels in healthy tissue. This reduces on-target toxicity outside diseased tissue.

In addition, ASOs are well suited to many lncRNA targets because they can enter relevant non-parenchymal cells, including fibroblasts, and engage nuclear RNAs without necessarily requiring complex delivery vehicles. This has allowed us to develop HTX-001 as a systemically administered, unconjugated gapmer ASO targeting WISPER in cardiac fibroblasts.

Advanced delivery technologies may further improve exposure and reduce dose over time, but for appropriately selected lncRNAs, intrinsic tissue- and disease-state specificity can create therapeutic precision without requiring highly engineered delivery from the outset.

Advances in genomics, computational biology, and artificial intelligence are generating unprecedented amounts of biological data. How are these technologies accelerating our understanding of the regulatory genome and helping identify new therapeutic opportunities?

SO: These technologies have been fundamental in allowing us to move beyond cataloging genes and proteins toward understanding how the regulatory genome controls cell identity and disease. The real opportunity comes from integrating these different layers of biology by leveraging computational biology and AI to find the regulatory nodes most likely to drive a pathological cell state rather than simply reflect it.

At HAYA, we are biology-native and AI-enabled. We use these technologies to model disease-relevant cell states, prioritize high-confidence regulatory targets, and predict how their modulation could shift cells toward healthier function. The aim is to move from observing biological complexity to identifying precise therapeutically actionable control points within it.

What are some of the biggest scientific misconceptions surrounding non-coding RNAs (nRNAs) and the regulatory genome today?

SO: The most persistent misconception is that molecular biology can still be adequately explained by a simple DNA–RNA–protein hierarchy, with RNA acting mainly as an intermediate between genetic information and protein function. Rather, RNA is a central operating and organizational layer.

A second misconception is that ncRNAs are merely transcriptional noise. Their low abundance, rapid evolution, and exquisite tissue- and cell state specificity are often presented as reasons to dismiss them. In reality, those characteristics can be precisely what make them powerful regulators.

Looking across the broader pharmaceutical industry, what signals suggest that regulatory genome medicine is moving from an emerging scientific concept toward a clinically validated therapeutic category?

SO: The first is clinical translation. HAYA has advanced HTX-001, a first-in-class investigational ASO targeting WISPER, into Phase 1 development in non-obstructive hypertrophic cardiomyopathy and the program has received FDA Fast Track designation. The field is now moving into human testing.

The second signal is strategic commitment and growing investments from venture capitalists and large pharmaceutical companies like our multiyear collaboration with Eli Lilly. This reflects a broader recognition that the regulatory genome may contain therapeutically actionable drivers that conventional protein-centric discovery has missed and reflects a broader acceptance that the cell state itself can be a therapeutic substrate.

Taken together, these developments suggest that regulatory genome medicine is progressing toward a clinically testable category. The decisive step will be human evidence showing that these therapies can safely reprogram disease-driving cell states and deliver meaningful benefit to patients.

What milestones will you be watching over the next three to five years that could define the future of regulatory genome therapeutics?

The defining milestone will be showing that regulatory genome therapies can safely alter the intended disease-driving cell state and deliver meaningful clinical benefit.

A second milestone will be repeatability. The field will become a true therapeutic category when regulatory genome control points can be identified and translated across different tissues, diseases, and biological contexts through a reproducible discovery and development process.

The third milestone is mechanism-matched biomarkers. We need these to connect target engagement, cellular reprogramming, and clinical outcomes.

Over the next few years, I expect the central question to shift from whether RNA can regulate cell state to how precisely, safely, and broadly we can use it to program healthier biology in patients.

As researchers, investors, and biopharma leaders evaluate the next wave of innovation in RNA medicine, what opportunities do you believe are still being underestimated?

Today, much of the field is focused on silencing or modulating endogenous RNAs. But as we understand how lncRNAs organize chromatin, recruit molecular machinery, and control cell state transitions, we can begin reverse-engineering those mechanisms to design programmable synthetic lncRNA therapies.

A second opportunity is the use of RNA as a measurement layer. Circulating regulatory RNAs can provide a more direct readout of tissue-specific pathology, target engagement, and cell state change than many conventional biomarkers with potential to improve clinical translation.

The broader opportunity is about learning the regulatory language of the cell well enough to read it, modulate it, and eventually write it.

Looking ahead, do you envision a future where the industry moves beyond a protein-centric model altogether, or will protein-targeting and regulatory genome approaches ultimately coexist as complementary strategies?

The future is not about replacing protein-targeting medicines but moving beyond the assumption that proteins are always the most appropriate point of intervention.

Protein-directed therapies will remain highly effective where disease is driven by a discrete protein defect or a single dysregulated pathway. However, for many common and chronic conditions driven by maladaptive cell states, cellular reprogramming approaches that act on the regulatory genome offer a far more direct way to target the underlying cause of disease.

These two approaches will coexist and become increasingly integrated. Regulatory genome therapies may reset the pathological cell state, while protein-targeting medicines provide complementary control.

The future will not be protein-centric or RNA-centric; it will be causality-centric.

If you missed Part 1, read it today.

About The Expert

Samir Ounzain, Ph.D., is a molecular biologist and entrepreneur with over 20 years of experience exploring the regulatory “dark matter” of the human genome. As CEO and cofounder of HAYA Therapeutics, he is pioneering a new class of precision RNA medicines that target long non-coding RNAs (lncRNAs) to reprogram the root causes of disease — at the level of cell state. Under his leadership, HAYA has grown from foundational discovery to platform-enabled execution, securing top-tier venture backing and forging transformative partnerships, including one of the largest collaborations to date in the regulatory genome space with Eli Lilly and Company. This alliance is focused on uncovering novel lncRNA targets for the development of new therapies in obesity and related metabolic diseases. Ounzain’s vision is to build a future where medicine is programmable, preventative, and patient-centric — powered by real-time interaction with the regulatory genome. He was named among the top innovators on The Power List 2025 in the Advanced Medicine category by Medicine Maker. Before cofounding HAYA, Ounzain was a project leader and research fellow at Lausanne University Hospital (CHUV), where he led the discovery of hundreds of novel cardiac-enriched lncRNAs, including CARMEN, Meteor, and WISPER — the latter forming the scientific and therapeutic foundation for HAYA’s lead program.