Selecting Lead Candidates to Qualify as IND Using Human-Relevant Surrogates
Executive Summary
A next-generation U.S.-based biotech company focusing on “undruggable” targets needed decision-grade, human-relevant insights to select its Investigational New Drug (IND) candidates. Relying on ambiguous animal-model data risked delays, high costs, and uncertainty for out-licensing decisions.
By deploying DART’s human microphysiological models combined with AI/ML-powered modules (TransD and SpheroidD), the company obtained clinically surrogate efficacy data before entering trials. This approach enabled faster IND selection, de-risked licensing discussions, and supported a multimillion-dollar out-licensing deal—all without conducting animal studies.
About the Client
A pioneering biotech organization developing advanced therapeutics for “undruggable” targets, focused on leveraging AI and human-relevant models to accelerate clinical readiness.
- Industry: Life Sciences (Small molecule drug discovery)
- Size: ~50 employees
- Country: USA
Problem Statement
A next-generation biotech company developing therapies for previously undruggable targets needed decision-grade, human-relevant evidence to support IND selection and out-licensing. However, animal-model data was inconclusive, and potential buyers required clinically surrogate evidence reflective of human tumor microenvironments. Conducting early clinical trials was prohibitively expensive, creating a gap between preclinical validation and commercial readiness. The client required a cost-effective, scientifically rigorous solution capable of producing mechanistically validated, human-relevant insights to de-risk IND qualification and accelerate partnership decisions.
Challenges
- Animal model-based data provided ambiguous efficacy outcomes.
- Potential buyers required human-relevant clinical trial-grade insights.
- Conducting early-phase clinical studies was too costly and resource-intensive.
- Lack of validated preclinical surrogates hindered confident IND decision-making.
The Solution
DART introduced a non-animal testing framework that integrates human microphysiological models and AI/ML-driven analytics to decode target-related drug efficacy readouts using its TransD and SpheroidD modules. This approach enabled the client to replicate human tumor biology in vitro, generating clinical surrogate efficacy data that enhanced both scientific and commercial outcomes.
Key features included:
- Human tumor modeling: Simulated tumor microenvironments to decode efficacy and safety profiles.
- AI/ML-powered insights: Delivered mechanistic understanding of target specificity, PK, and PD parameters.
- Clinical surrogate validation: Produced preclinical data with 100% relevance to human biology.
- Pipeline scalability: Allowed seamless in-house integration for broader program expansion.
By combining human biology, machine learning, and mechanistic analytics, DART provided clinically translatable, decision-grade insights, strengthening IND candidate confidence and increasing pipeline valuation.
Impacts
- Neurovirulence risk reports ready for regulatory submission.
- 100% relevance to human nervous system modeling.
- Confirmatory insights supporting both test and reference vaccines.
- Applicable for preclinical studies and routine stock testing.
- Established a sustainable, scalable neurovirulence assessment framework.
Technologies Used
Quantiphi’s DART Platform
TransD and SpheroidD Modules
Human Microphysiological Models
AI/ML-driven Efficacy Decoding
Mechanistic PK/PD Profiling Tools
Results
- Enabled confident lead candidate selection for downstream discovery.
- Generated 100% human-relevant efficacy and safety insights, supporting IND qualification.
- Transferred DART bioassay SOPs to the customer’s in-house workflow for consistent application across expanding programs.
- Enhanced pipeline scalability and commercial readiness, reducing dependency on costly animal or clinical trials.