Biotech's $100M gamble: How AlpBio de-risks drug failures pre-clinical

Apr 23, 2026 · Predictable Revenue Podcast
🎧 PodShort 22 min squeezed to 2 Tab HopperTH Sales Tech New
Episode artwork
Puneet
Co-founder and Chief Business Officer at AlpBio
Predictable Revenue Podcast
22 min squeezed to 2
Full episode from Predictable Revenue Podcast
Quotable Moments

It was this comfort creeping in. It was like, okay, I know the trajectory here, I know it's going to go. It's super cool, like you're going to get your promotion, you're going to get more money, you know, you can get to the higher level, but you cannot really change big things.

The problem with academia is it doesn't generate the same drive. So, to get real traction anywhere, it feels like you have to work within the system and it needs to be commercialized at some point.

We're trying to be a bit slower in everything, but presenting very credible evidence and data. And that's where we're trying to build also our moat or our business identity, which is different than going in software where you want to be fast and you can fail and it's fine to fail and then you can rebuild. In pharma, that failure rate should be lower because credibility has a strong value attribute in this space.

Key Insights
  • The idea for AlpBio came from a co-founder's PhD research at Caltech, where he explored using machine learning to create new types of proteins, specifically antibodies, and how to explore that space.
  • AlpBio's core concept involves using patient-derived tonsils to test drugs for immunogenicity, allowing them to identify and re-engineer potential drugs to avoid adverse immune reactions before clinical trials.
  • The transition from academia to a commercial setup was driven by the need for real traction and the understanding that commercialization is necessary to bring scientific breakthroughs to market.
  • The biotech industry, particularly drug development, is a highly regulated space, and it's crucial to identify where a new technology fits within that process, specifically in the pre-clinical safety assessment phase.
  • Biotech companies are highly attractive partners for AlpBio because they have a high-risk development model, where a single drug failure can be catastrophic, making de-risking solutions like AlpBio's very valuable.
  • A drug failure in phase one or two clinical trials can cost between $5 million to $100 million, and a phase three failure can cost hundreds of millions, highlighting the immense financial risk involved.
  • AlpBio's pricing model aims to align with the existing budgets of biotech companies for immunogenicity testing, with an additional single-digit percentage payment on future milestones if their re-engineered drugs succeed.
  • The key to AlpBio's re-engineering capability is their ability to generate human-relevant immune data from tonsils, which current modalities lack, allowing them to computationally predict and reduce immunogenicity.
Metrics Mentioned
  • 300,000 operations per year (The number of tonsil removal operations in the US annually, providing a source for AlpBio's research.)
  • Up to 2,000 experiments per tonsil (The number of experiments that can be derived from a single tonsil, indicating the efficiency of AlpBio's method.)
  • $5 million to $100 million (The cost of a drug failure in phase one or two clinical trials.)
  • $750 million (The R&D cost incurred by Pfizer for a phase three failure of their drug 'Bococizumab'.)
  • Half a billion in revenue or more (The potential revenue opportunity lost due to the phase three failure of 'Bococizumab'.)
  • Single-digit percentage range (The percentage of future milestone payments AlpBio aims to receive for successful re-engineered drugs.)

RevBots.ai View:

  • Tab Hopper alert: Pharma's high-stakes environment demands manual credibility-building vs. scalable playbooks.
  • Risk-sharing pricing models (like milestone payments) are rare outside capital-intensive industries.
  • AI Sprinkler potential: ML for protein engineering could scale if integrated with clinical workflows.