Latus Bio, Inc. (Latus), a biotechnology company engineering gene therapies at scale, today announced the publication of a peer-reviewed study in Nature Communications, titled “Computational modeling and preclinical evaluation support targeting somatic instability for Huntington’s disease treatment.” The study was authored by multiple Latus scientists and led by co-corresponding authors Jang-Ho J. Cha, M.D., Ph.D. and Beverly L. Davidson, Ph.D.

The publication focuses on MSH3, a DNA mismatch repair protein, that is the key pathogenic driver of somatic instability (SI) and causally linked to Huntington’s disease (HD) onset and progression. The authors introduce a computational model built from human natural history and human neuronal-somatic-expansion data that simulates the long-term impact of suppressing MSH3 and, hence, SI, within MSNs. It predicts years of potential therapeutic benefit across HD patient ages and stages. The study further integrates model-predicted thresholds for therapeutic activity with validating preclinical data in support of LTS-201, the company’s investigational adeno-associated virus (AAV) gene therapy designed to reduce MSH3 expression in vulnerable striatal and cortical neurons. LTS-201 combines AAV-DB-3, a previously described capsid engineered for efficient targeting of MSNs and layer-5/6 cortical projection neurons, with an artificial microRNA designed for highly selective lowering of MSH3 expression.

“Genetic studies increasingly point to somatic instability as the central driver of Huntington’s disease progression,” said Jang-Ho J. Cha, M.D., Ph.D., Chief Science and Medical Officer of Latus Bio. “This work establishes a translational framework that quantitatively links MSH3 lowering to reductions in somatic instability and to years of predicted clinical benefit in HD. Our investigational LTS-201 therapy achieves dose-dependent target engagement of MSH3 in preclinical models at levels that are potentially meaningful.”

Key Highlights from the Publication

  • Introduction of a computational model capable of predicting the impact of SI-targeted therapies in HD across a spectrum of inherited CAG-repeat lengths and potential ages of therapeutic intervention.

  • Quantitative correlations that link levels of MSH3 knockdown (KD) and percentages of targeted MSNs to years of expected delay in HD progression.

  • Computational simulations that show deeper MSH3 KD in smaller percentages of MSNs yields greater benefits than shallower KD in higher percentages of on-target cells, offering potential minimum thresholds for translational therapeutic development.

  • Model-derived predictions that deep and durable MSH3 KD in a substantial percentage of MSNs should translate to years of benefit across HD patient ages and stages, with earlier intervention expected to provide the greatest benefit (measurable in decades).

  • Model-predicted correlations between the levels of MSH3 KD and reductions in the rates of cUHDRS decline, ranging from approximately 50% to more than 120%, accompanying predicted delays in motor symptom onset relative to natural history.

  • Demonstrations that LTS-201 results in deep, dose-dependent MSH3 mRNA KD, ranging from 48% to 94%, in >90% of MSNs in nonhuman primates.

  • Preclinical validation that LTS-201-mediated MSH3 KD results in dose-dependent reductions in somatic instability in the striatum of HdhQ111 HD mice, providing a robust in vivo proof-of-mechanism.

“This study highlights the potential for integrating predictive computational modeling with gene therapy engineering to guide the development of targeted disease-modifying therapies,” said Beverly L. Davidson, Ph.D., Chair of the Scientific Advisory Board of Latus Bio.

The authors note that the modeling framework may also be adaptable to other repeat expansion diseases, including X-linked dystonia-parkinsonism and myotonic dystrophy type 1, where MSH3 has been similarly implicated as a disease modifier. The publication builds on prior work describing AAV-DB-3 – an engineered capsid optimized for robust transduction of basal ganglia and cortical projection neurons in nonhuman primates (Leib et al., Nat. Commun. 2025).

About Latus Bio

Latus is a biotechnology company dedicated to addressing devastating CNS and peripheral diseases via innovative and scalable gene therapies. The Company is advancing a broad therapeutics pipeline based on novel AAV capsid variants with potency and specificity. Latus is powered by a diverse team of visionary scientists, experienced clinicians, and leading industry executives. The Company has offices in Philadelphia, PA and in the Seaport in Boston, MA. For more information, visit www.latusbio.com.

About LTS-201

LTS-201 (AAV.DB3.miMSH3) is a preclinical stage, investigational, gene therapy candidate that encodes an engineered microRNA to reduce the expression of the DNA repair enzyme MSH3. Its design is based on AAV-DB3 – a novel and proprietary AAV capsid variant that targets medium spiny neurons and cortical projection neurons in the central nervous system after administration into the deep brain. Preclinical data suggest that a single low-dose injection of LTS-201 reduces MSH3 expression in a high percentage of HD target cells to dramatically slow somatic repeat expansion in mutant HTT. It is intended to slow or halt HD progression while preserving wildtype HTT protein levels. If proven safe and efficacious in future clinical studies, LTS-201 has the potential to provide benefits across the spectrum of HD patients and could enable facile administration at many neurosurgical centers.

About Huntington’s Disease (HD)

Huntington’s disease is an inherited neurodegenerative disorder affecting greater than 100,000 patients in major gene therapy markets. It results from a decades-long expansion of DNA repeats in the Huntingtin gene (HTT). Extensively validated by human data, MSH3 is the key pathogenic driver of the somatic repeat expansion that results in HD onset and progression. Affected patients develop normally but subsequently experience progressive cognitive, neuropsychiatric, and motor changes that result from the accumulation of mutant HTT protein with >150 elongated repeats, leading to the loss of medium spiny neurons in the deep brain (caudate and putamen) as well as projection neurons in the cerebral cortex. Currently, there are no approved disease-modifying therapies.

Media gallery

About The Author
Business Wire