Greenstone Biosciences wins NHLBI R61 award to identify small-molecule treatments for DMD heart disease

Greenstone Biosciences, Inc. has received a $625,968 grant from the National Heart, Lung, and Blood Institute to identify potential small-molecule treatments for myocardial fibrosis and dilated cardiomyopathy associated with Duchenne muscular dystrophy. The R61 award, numbered R61HL179715, has a performance period running from August 1, 2026, to May 31, 2028 and will support early target identification, computational screening and laboratory validation rather than human clinical trials. The Palo Alto biotechnology company intends to combine patient-derived induced pluripotent stem cells, cardiac organoids, proteomics and artificial intelligence to search for compounds capable of preventing or reducing fibrotic damage in the heart. The central opportunity is a substantial treatment gap in Duchenne muscular dystrophy, although the programme remains several scientific and financing milestones away from producing a clinical candidate.

The Greenstone Biosciences NIH grant is important partly because therapeutic development in Duchenne muscular dystrophy has historically concentrated on skeletal muscle function, dystrophin production and mutation-specific approaches. Cardiac disease, however, becomes increasingly consequential as patients live longer, with progressive myocardial fibrosis contributing to ventricular dysfunction, dilated cardiomyopathy and heart failure. United States Food and Drug Administration reviewers stated in July 2026 that no therapy had been approved specifically for Duchenne cardiomyopathy and that existing care relies mainly on medicines adapted from general heart-failure treatment.

Greenstone Biosciences is seeking to approach the problem from an earlier point in the development chain. Rather than beginning with a predetermined drug target, the company plans to examine interactions between Duchenne patient-derived cardiomyocytes and cardiac fibroblasts, identify disease pathways through proteomics and then use computational tools to screen potential compounds. That creates an opportunity to uncover biological mechanisms that may be missed by conventional animal models, but the value of the approach will depend on whether laboratory findings translate into measurable therapeutic effects.

What does Greenstone Biosciences’ $625,968 NIH R61 grant actually finance?

The current award finances the R61 portion of the National Heart, Lung, and Blood Institute’s Catalyze programme. This phase is intended to support target identification, target validation, compound screening and preliminary product or lead-series discovery for heart, lung, blood and sleep disorders. It should not be interpreted as funding for an approved drug, a clinical trial or even a completed preclinical candidate.

Greenstone Biosciences plans to begin by using unbiased proteomics to study communication between induced pluripotent stem cell-derived cardiomyocytes and cardiac fibroblasts from Duchenne muscular dystrophy patients. Fibroblasts are particularly relevant because their activation contributes to the production and accumulation of extracellular matrix material that stiffens and scars cardiac tissue.

Once a potential target is selected, Greenstone Biosciences intends to employ molecular docking and molecular-dynamics simulations to screen compounds computationally. The company also plans to use SyntheMol, a generative artificial intelligence system, to design molecules from a chemical space estimated at approximately 30 billion possible structures. A separate bioactive compound screen is expected to include existing substances as well as nutritional and dietary compounds that could influence cardiac fibrosis.

The scale of the virtual search is eye-catching, but molecule count is not the decisive measure of progress. Large computational libraries can generate extensive lists of theoretical binders, while relatively few compounds demonstrate the necessary combination of potency, selectivity, solubility, metabolic stability, safety and manufacturability. Greenstone Biosciences must therefore show that its artificial intelligence tools improve the quality of candidate selection rather than merely expanding the volume of predictions.

The most useful outputs from the R61 phase would be a reproducible disease target, validated assays and several chemically tractable hits that consistently reduce fibrosis-related activity across patient-derived models. These results would create the basis for moving into compound optimisation, more extensive safety work and animal testing.

Why is Duchenne muscular dystrophy cardiac fibrosis still a major treatment gap?

Duchenne muscular dystrophy is caused by variants in the DMD gene that result in absent or severely deficient dystrophin. Without adequate dystrophin, skeletal, respiratory and cardiac muscle cells become vulnerable to repeated contraction-related injury. In the heart, this damage can produce inflammation, cell loss, replacement fibrosis and progressive ventricular dysfunction.

The disease affects approximately one in every 3,500 male births, according to the Greenstone Biosciences grant abstract and previous clinical literature. Although corticosteroids and modern supportive care can delay disease progression, patients may experience intolerance, inadequate response or continued cardiac deterioration.

Current cardiac management generally includes renin-angiotensin system inhibitors, mineralocorticoid receptor antagonists and beta blockers. These medicines can help delay deterioration or manage established heart failure, but they were not originally developed to correct the molecular processes responsible for Duchenne-related myocardial fibrosis. The evidence base in Duchenne muscular dystrophy also remains smaller than that available for more common forms of heart failure.

The regulatory experience of Capricor Therapeutics illustrates the difficulty of demonstrating cardiac benefit in this population. In July 2026, an FDA advisory committee voted nine to three that the available evidence did not provide substantial evidence of effectiveness for deramiocel in Duchenne cardiomyopathy. Panel members raised concerns about missing data, analytical changes and whether the enrolled population had sufficient cardiac impairment for the trial to demonstrate a meaningful effect. The FDA’s decision on the application was expected by August 22, 2026.

Greenstone Biosciences is developing a small-molecule discovery strategy rather than Capricor Therapeutics’ cell-therapy approach, and the two programmes are at very different stages. Nevertheless, the deramiocel review highlights a broader industry challenge. A candidate must not only influence fibrosis in laboratory models but eventually demonstrate a credible relationship between biological activity, cardiac imaging measures, functional outcomes and benefits meaningful to patients.

How could patient-derived iPSCs and generative AI improve early drug discovery?

Induced pluripotent stem cells can be generated from human donor cells and differentiated into specialised cell types such as cardiomyocytes and cardiac fibroblasts. When these cells come from patients carrying Duchenne muscular dystrophy mutations, they may retain disease-relevant genetic characteristics that allow researchers to recreate aspects of the condition in laboratory systems.

Greenstone Biosciences says its biobank includes induced pluripotent stem cell lines from more than 2,500 donors, including rare and orphan disease populations. The company offers derived cardiomyocytes, fibroblasts, endothelial cells and cardiac organoids, alongside high-throughput drug-screening and toxicity-testing services.

This infrastructure could allow Greenstone Biosciences to test compounds across genetically diverse patient-derived cells rather than relying on a single engineered cell line. That matters because Duchenne muscular dystrophy progression is heterogeneous, and a compound that appears effective in one model may not perform consistently across different mutations, biological backgrounds or disease stages.

The use of three-dimensional cardiac organoids may provide an additional level of complexity. Organoids can reproduce some interactions among cardiomyocytes, fibroblasts and other cardiac cell types that are difficult to observe in two-dimensional cultures. They still cannot reproduce the entire physiology of a human heart, including systemic circulation, immune responses, metabolism and long-term mechanical stress, but they may provide more relevant screening conditions than simplified assays.

Artificial intelligence can complement these biological models by prioritising targets, predicting compound-target interactions, proposing chemical structures and estimating properties associated with absorption, distribution, metabolism, excretion and toxicity. The commercial proposition is straightforward: better early prediction could reduce the number of unsuitable compounds carried into expensive preclinical programmes.

The scientific requirement is more demanding. Predictions must be prospectively validated, results must be reproducible and selected compounds must outperform conventional screening alternatives. Without that evidence, artificial intelligence remains an enabling tool rather than an independently validated source of therapeutic value.

What evidence does Greenstone already have from its earlier Duchenne research?

The new NIH award builds on a $675,000 grant previously awarded by the California Institute for Regenerative Medicine for Duchenne muscular dystrophy drug discovery. Under that programme, Greenstone Biosciences generated, differentiated and characterised patient-derived cardiac cells that reproduced disease-associated phenotypes. The programme’s stated activities included screening approximately 8,000 small molecules, evaluating a candidate identified as GSB-D010 and validating potential drugs in three-dimensional cellular and mouse models.

The California Institute for Regenerative Medicine grant is listed as closed. Its reported progress supports the view that Greenstone Biosciences can generate and characterise Duchenne patient-derived cardiomyocytes, which reduces part of the platform-development risk surrounding the new award. It does not, however, establish that the company has identified a clinically effective treatment or that any previously screened compound has advanced towards human testing.

The distinction matters because platform validation and product validation are separate commercial milestones. A company may possess high-quality disease models that generate research revenue or partnership interest without successfully developing a proprietary medicine. Conversely, a drug candidate emerging from the platform could produce substantially greater value but would require a longer development timeline, additional capital and regulatory engagement.

Greenstone Biosciences has also received earlier federal funding for an ethnically diverse induced pluripotent stem cell-derived cardiomyocyte panel and for an artificial intelligence and organoid-based myotonic dystrophy project. Including the latest award, United States Department of Health and Human Services records show approximately $3.67 million in grant actions associated with the company since 2023.

This funding history suggests that government agencies see value in the company’s human-relevant disease-modelling capabilities. The stronger proof point would be evidence that repeated public funding has produced intellectual property, proprietary targets or therapeutic candidates that attract follow-on investment from pharmaceutical companies or specialist biotechnology investors.

Why does the R33 transition matter for financing, partnerships and execution?

The Catalyze award uses a phased R61 and R33 structure. Greenstone Biosciences must achieve agreed scientific milestones before the National Heart, Lung, and Blood Institute will consider releasing funding for the R33 phase. Progression also requires evidence of non-federal matching capital, continued intellectual-property and regulatory planning and the identification of at least one accelerator partner.

The National Heart, Lung, and Blood Institute recommends non-federal matching funds equal to at least 25% of the federal direct costs requested for the R33 phase. Proof is not required to activate the initial R61 award, but it must be supplied before R33 funds are released. This requirement effectively tests whether the project can attract external conviction once early technical results become available.

During the R33 phase, Greenstone Biosciences intends to synthesise and characterise selected compounds, assess properties such as lipophilicity, acidity and solubility, and test safety and efficacy in cardiac organoids. The company would then evaluate candidates in the D2-mdx Duchenne mouse model and conduct initial pharmacokinetic and toxicity studies.

Securing an accelerator partner could be as strategically important as obtaining additional grant funding. A pharmaceutical or biotechnology partner could contribute medicinal chemistry, regulatory experience, manufacturing capability and development capital. An investor or translational organisation could provide matching funds and help Greenstone Biosciences prepare for investigational new drug-enabling studies.

The award therefore functions as both scientific financing and a staged commercial filter. Success in the laboratory must be followed by outside capital formation. Failure to secure matching support would indicate that promising scientific results had not yet translated into sufficient commercial confidence.

What technical and regulatory evidence must Greenstone produce before the thesis strengthens?

The first technical challenge is disease-model fidelity. Greenstone Biosciences must demonstrate that the fibrosis-related pathways observed in its induced pluripotent stem cell-derived cells resemble those found in patients and are not artefacts of cell culture, differentiation methods or assay design.

The second challenge is target validation. Proteomic analysis may reveal many proteins and signalling pathways associated with fibrosis, but association does not prove that manipulating a target will safely change disease progression. Targets must be confirmed using independent experiments, multiple patient lines and ideally different laboratory methods.

The third challenge is medicinal chemistry. Compounds identified through molecular docking or generative artificial intelligence may perform poorly when synthesised or tested in biological systems. They may lack sufficient potency, interact with unintended targets or show properties that make oral administration and long-term dosing impractical.

Safety will be especially important because Duchenne muscular dystrophy patients may receive treatment over extended periods and may already use corticosteroids, cardiac drugs and other supportive therapies. Any candidate must therefore have a wide therapeutic window and a manageable interaction profile.

Regulatory strategy presents another challenge. The company will eventually need to determine which patient population to study, when treatment should begin and which endpoints can demonstrate meaningful benefit. Cardiac fibrosis can develop before overt symptoms, creating a potential case for early intervention, but slowly progressing disease also makes clinical trials longer and harder to interpret.

The recent regulatory scrutiny of deramiocel shows that cardiac trial design, baseline disease severity, missing data and endpoint selection can become decisive. Greenstone Biosciences has time to incorporate those lessons because its programme remains at the discovery stage, but that advantage will matter only if regulatory considerations shape candidate development before clinical testing begins.

What would turn this NIH grant from platform validation into a credible drug programme?

The Greenstone Biosciences NIH grant provides non-dilutive capital and strengthens the external validation of a platform built around human induced pluripotent stem cells and computational drug discovery. It also places the company in a clinically significant area where therapeutic progress has not kept pace with the growing recognition of cardiac disease as a major determinant of Duchenne muscular dystrophy outcomes.

What remains unresolved is whether the platform can generate a drug candidate with reproducible biological activity, favourable pharmaceutical properties and a realistic path into clinical development. The difference between a promising screening programme and an investable therapeutic asset will be determined by a relatively small number of measurable events: confirmation of a disease target, selection of credible compounds, validation across diverse patient-derived organoids, transition into R33 funding and recruitment of an external development partner.

The strongest outcome would be a lead series that reduces fibroblast activation and preserves cardiomyocyte function across several Duchenne muscular dystrophy models without introducing meaningful toxicity. Evidence of that kind could support further National Institutes of Health funding, pharmaceutical partnering or private investment for investigational new drug-enabling work.

Conversely, inconsistent effects across patient lines, weak pharmacological properties or failure to attract matching capital would suggest that the platform remains more valuable as a research and screening service than as the foundation of a proprietary therapeutic pipeline. Greenstone Biosciences has gained an important opportunity, but the decisive test begins after the grant announcement, when computational predictions must become reproducible pharmacology.

Key takeaways from the Greenstone Biosciences NIH grant for Duchenne cardiac fibrosis

• Greenstone Biosciences has received a $625,968 National Heart, Lung, and Blood Institute R61 award for Duchenne cardiac fibrosis drug discovery.
• The current award runs from August 1, 2026, through May 31, 2028 and does not finance a clinical trial.
• The programme will use patient-derived cardiomyocytes, cardiac fibroblasts, proteomics, organoids and artificial intelligence.
• Greenstone Biosciences plans to screen conventional compound libraries and use generative artificial intelligence across a chemical space of approximately 30 billion molecules.
• The project targets myocardial fibrosis and dilated cardiomyopathy, areas where no therapy is currently approved specifically for Duchenne cardiac disease.
• Earlier California Institute for Regenerative Medicine funding helped Greenstone Biosciences establish patient-derived Duchenne cardiac models.
• Advancement into the R33 phase depends on scientific milestones, non-federal matching funds and an accelerator partner.
• The R33 phase would support compound synthesis, organoid validation, animal testing and early pharmacokinetic and toxicity work.
• The main uncertainties involve disease-model fidelity, target validation, medicinal chemistry, safety and eventual clinical endpoint selection.
• The next meaningful proof point will be the identification of reproducible targets and drug candidates capable of reducing fibrosis-related activity across multiple patient-derived models.