Failure Analysis
Base Therapeutics died from a fatal combination of Product/Tech Failure and capital inefficiency, rooted in the classic biotech trap: over-investing in computational infrastructure without...
Base Therapeutics \China
Base Therapeutics was a Chinese biotech startup founded in 2021 by Xu Tianhong, focused on developing novel therapeutics leveraging computational biology and AI-driven drug discovery. The company raised $34.5M from prominent investors including Baidu and Great Eagle VC, positioning itself at the intersection of China's booming AI sector and the global race for precision medicine. The 'Why Now' was compelling: AlphaFold had just revolutionized protein structure prediction in 2020, China was aggressively investing in biotech self-sufficiency post-COVID, and Baidu's AI infrastructure provided a strategic moat. Base likely aimed to accelerate drug candidate identification using machine learning models trained on Chinese patient data—a massive untapped dataset advantage. However, the company collapsed in 2025 after just 4 years, a critical juncture when most biotech startups are entering Phase I/II trials. The failure highlights the brutal reality that computational predictions don't translate to clinical efficacy without deep wet-lab validation, regulatory navigation, and patient recruitment infrastructure.
Discover the reason behind the shutdown and the market before & today
Market Analysis
The AI drug discovery market has matured significantly since Base's 2021 founding, with clear winners and losers emerging. Survivors like Insilico Medicine (Hong Kong,...
Startup Learnings
Biotech requires clinical co-founders, not just ML engineers. Base likely had a team of computational experts but lacked a Chief Medical Officer or VP...
Market Potential
The global AI drug discovery market is projected to reach $4B by 2027 (CAGR 28%), and China represents 30%+ of that TAM due to...
Difficulty
Biotech remains the hardest category to rebuild even with modern AI tools. While AlphaFold3, ESMFold, and RFdiffusion have democratized protein structure prediction, and platforms...
Scalability
Biotech has fundamentally non-scalable unit economics in early stages. Each drug candidate requires bespoke wet-lab validation, animal studies, and multi-phase human trials—costs that don't...
Rebuild & monetization strategy: Resurrect the company
Execution Plan
Phase 2
Step 2 - AI-Driven Molecule Design and Wet-Lab Validation (Months 6-12): Use AlphaFold3 to model the target protein and existing drug binding site. Use RFdiffusion to generate 50 novel molecules with improved binding affinity and predicted lower off-target effects. Outsource synthesis of top 10 candidates to GenScript (cost: $50K total). Run binding assays and cellular toxicity screens via WuXi (cost: $200K). Iterate based on results—if all 10 fail, pivot to a different target within the partnership. Goal: Identify 1 lead candidate with 10x better selectivity and 50% reduced toxicity vs. existing drug. This is your IND-ready molecule.
Phase 3
Step 3 - Preclinical Package and NMPA IND Filing (Months 12-18): Outsource GLP toxicology studies (rat, dog) and PK/PD profiling to WuXi (cost: $1.5M, funded by pharma partner). Engage a Chinese regulatory consultant (cost: $100K) to prepare the IND application for NMPA. File IND for rare disease fast-track designation (6-month review vs. 12 months standard). Simultaneously, publish a preprint on bioRxiv showing your AI design process and preclinical data—this attracts VC attention and validates your platform for future targets. Goal: NMPA IND approval by Month 18, enabling Phase I trial start.
Phase 4
Step 4 - Phase I Trial and Series A Fundraising (Months 18-30): Pharma partner funds Phase I trial in China (cost: $3M, 20-40 patients). Your role: Monitor data, prepare publications, and use Phase I safety data to raise a $20M Series A from biotech VCs (8VC, Lux Capital, or Chinese firms like Qiming Venture Partners). Pitch: We have a de-risked clinical asset in humans, a validated AI platform, and 2 more partnered programs in preclinical. Series A funds internal pipeline expansion—pick 2 additional targets where you retain 100% economics (no pharma partner). Goal: Phase I data showing safety and early efficacy signals, plus $20M in the bank to become a standalone drug company.
Monetization Strategy
Hybrid model combining pharma partnerships and internal pipeline. Phase 1 (Years 1-3): Partner with 3 Chinese pharma companies at $3-5M per target for preclinical development. Retain 50% profit share on approved drugs. This generates $9-15M in non-dilutive capital and proves the platform. Phase 2 (Years 3-5): Use Series A ($20M) to develop 2 internal programs where we retain 100% economics. Target rare diseases with small patient populations (5,000-20,000 globally) where orphan drug pricing ($100K-300K per patient per year) enables profitability with modest market share. Phase 3 (Years 5-7): License internal programs to global pharma (Roche, Novartis) for $50M upfront + $200M in milestones + 10-15% royalties. This is the exit for early investors. Long-term (Years 7-10): Build a portfolio of 5-7 approved rare disease drugs generating $500M+ annual revenue. Maintain the AI platform as a competitive moat, but the business model is a biotech royalty company, not a software company. Key insight: By focusing on fast-follower rare diseases and pharma partnerships, we avoid the $100M+ capital requirement of platform plays and reach profitability faster.
Related Startup Failures
BIOTECH
30
Fermata
\Israel
$5.0M
THE VALUE PROPOSITION
Precision fermentation tech producing sustainable, animal-free proteins to replace dairy and meat ingredients with lower environmental impact.
CAUSE OF DEATH
Failed to achieve commercial-scale production economics; burned through capital before reaching cost parity with traditional proteins.
Rise
2019
Fall
2024
BIOTECH
23
Spotlight Bio
\USA
$40.0M
THE VALUE PROPOSITION
AI-powered platform to identify and validate novel drug targets using genomics and computational biology, aiming to reduce pharma R&D failure rates.
CAUSE OF DEATH
Failed to bridge the gap between computational predictions and wet-lab validation; long timelines and capital intensity exceeded runway.
Rise
2018
Fall
2025
BIOTECH
52
Sichuan Kelun-Biotech
\China
$400.0M
THE VALUE PROPOSITION
Chinese ADC biotech aiming to develop next-gen cancer therapies leveraging lower costs and faster trials to compete globally.
CAUSE OF DEATH
Burned $400M without approvals in overcrowded ADC market; clinical failures and regulatory delays killed runway.
Rise
2016
Fall
2024
Disclaimer: This entry is an AI-assisted summary and analysis derived from publicly available sources only (news, founder statements, funding data, etc.). It represents patterns, opinions, and interpretations for educational purposes—not verified facts, accusations, or professional advice. AI can contain errors or ‘hallucinations’; all content is human-reviewed but provided ‘as is’ with no warranties of accuracy, completeness, or reliability. We disclaim all liability for reliance on or use of this information. If you are a representative of this company and believe any information is inaccurate or wish to request a correction, please click the Disclaimer button to submit a request.