Case Study: Insilico's Transformation

Insilico Medicine Featured

in Harvard Business School Case Study on Rentosertib

Harvard Business School Case Study

Historically, business students have had limited resources to bridge the gap between technical biotechnology and strategic drug development. We are pleased to share a recently published Harvard Business School (HBS) case study—the first to explore the impact of Generative AI on the end-to-end drug discovery process through the lens of Insilico Medicine.

The study focuses on a critical strategic choice biotechs face: whether to license their drugs to a partner or develop them internally. We created this resource to show how generative AI is used to speed up every step, from finding new drug targets to predicting if clinical trials will succeed.

AI-Powered Drug Discovery Crash Course

Table of Contents

Module 1: The Broken Pipeline & The AI Intervention

Lesson 1.1: The Efficiency Crisis

The "Leaky Pipeline": Eroom’s Law: Why drug discovery is getting slower and more expensive

Approximately 90% of drug candidates that enter clinical trials fail to reach the market. The failure is attributed primarily to a lack of clinical efficacy and unmanageable toxicity. The industry has lost an estimated $1 trillion in failed drug developments over the past decade.

Lesson 1.2: The Rise of "TechBio"

Differentiating "Biotech" (Biology-first) vs. "TechBio" (Data-first)

Insilico Medicine exemplifies the TechBio evolution. The company utilizes an engineering and data-first approach to drug discovery, leveraging massive data generation and artificial intelligence to create discovery engines capable of producing multiple assets across various indications.

Module 2: Deconstructing the Pharma.AI Platform

Lesson 2.1: Target ID: PandaOmics

How AI Mines Multi-Omics Data: The first step in drug discovery is target identification—isolating the biological origin of a disease.

High Confidence Targets vs. Novel Targets - The AI can identify first-in-class opportunities where no previous drugs or clinical trials exist.

Lesson 2.2: Molecular Generation: Chemistry42

Creating Needles: GANs and "Imagining" New Molecules

Traditional drug discovery has relied on virtual screening, while AI utilizes Generative Adversarial Networks to design new drug-like molecules.

Lesson 2.3: Simulating Clinical Trials: Medicine 42 (inClinico)

Digital Twins - Simulations allow researchers to forecast treatment effects and optimize trial designs.

Module 3: The Rentosertib Story

Lesson 3.1: Transformation of Insilico — From Platform to Pipeline

Insilico Medicine’s transition marked a pivot from a software provider to a full-stack, clinical-stage biotechnology enterprise. The dual-CEO structure ensures focus on both AI platform development and drug development.

Lesson 3.2: Rentosertib – A Proof of Concept for Generative AI

The identification of TNIK for Idiopathic Pulmonary Fibrosis (IPF) has been significant in demonstrating AI's capabilities in drug development.

Lesson 3.3: From Hypothesis to Preclinical Candidate in Under 18 Months

The entire process was completed in approximately 18 months, representing a reduction from the traditional timeline.

Module 4: The Economics of Innovation

Lesson 4.1: The Financial Engine of Transformation

Funding from multiple rounds has helped establish automated infrastructure necessary for scaling up clinical trials.

Lesson 4.2: Partnership Modeling

Insilico has executed high-value out-licensing agreements, with a combined potential value exceeding $2 billion.

Lesson 4.3: The Strategic Dilemma

Insilico adopts a hybrid approach: developing some programs internally while out-licensing others to maintain financing stability.

Conclusion

The significance of Rentosertib extends beyond its primary therapeutic indication, highlighting a broader shift toward platform-centric R&D where drug programs benefit from shared knowledge across diverse fields.