Predicting the Unpredictable Off-Target Activities of Drugs

Episode Summary

Executive Summary: This episode of The Bio Report features David Younger, CEO of AAlpha Bio, discussing how the company’s AlphaSeq platform uses engineered yeast and next-generation sequencing to measure thousands of protein-protein interactions at once. The goal is to identify drug off-target effects earlier, reduce late-stage failures, and provide pharmaceutical partners with richer data on specificity and safety.

Main Topics: Why drug development is so expensive (Priority: 5/5): The discussion centers on biological complexity and late-stage attrition: many drugs appear promising in preclinical work but fail in humans due to unintended protein interactions and side effects. AAlpha Bio’s AlphaSeq platform (Priority: 5/5): AlphaSeq is presented as a high-throughput system that screens thousands of protein interactions simultaneously and quantifies interaction strength, enabling broad assessment of drug effects. Yeast-based synthetic biology as the technical foundation (Priority: 4/5): The company repurposes yeast agglutination and surface display to swap native mating proteins for human or engineered proteins, turning cell sticking into a readout of protein binding. Benchmarking on a known disease pathway (Priority: 4/5): AAlpha is validating the platform using the BCL2 apoptosis pathway and known drugs such as Navitoclax and Venetoclax to confirm it detects expected on-target and off-target effects. Commercial strategy and partnerships (Priority: 4/5): Rather than developing drugs immediately, AAlpha plans to partner with pharma and CROs to support drug development programs, then potentially move into internal drug discovery later. Startup formation and funding path (Priority: 3/5): Younger describes the transition from academic science to entrepreneurship and notes support from UW, entrepreneurship resources, and NSF SBIR funding to extend runway.

Key Arguments: Most drug failures are driven by biology’s complexity, especially unpredictable protein interactions, so broader early-stage data is needed to reduce late-stage attrition. Off-target effects are a major reason drugs fail in the clinic, and screening thousands of protein interactions early could save substantial development costs. Existing accurate protein-interaction assays are too low-throughput to handle thousands or millions of interactions, creating a bottleneck that AlphaSeq is designed to solve. Measuring interactions on yeast cell surfaces isolates proteins from intracellular noise while still enabling controlled extracellular perturbations such as drug addition. Benchmarking on well-studied pathways and drugs is essential to prove that the platform can detect both intended effects and unintended off-target binding. A partnership-first business model is more realistic than immediate internal drug discovery because the company is still defining the highest-value applications of the technology. Non-dilutive funding is allowing AAlpha to continue technical development while postponing equity fundraising until the platform’s most important use cases are clearer.

Data Points: Clinical trial failure rate: 9 out of 10 drugs - Younger says most drugs entering clinical trials ultimately fail. Interaction scale: thousands of protein interactions simultaneously - AlphaSeq is described as screening large protein libraries in one tube. Funding runway from SBIR: 11 months - Younger says the recent NSF SBIR grant will sustain current operations for this period. Pathway example: BCL2 apoptosis regulatory pathway - Used as the initial benchmarking system for validation. Company origin: spin-out from the University of Washington - AAlpha Bio emerged from the Institute for Protein Design and Center for Synthetic Biology.

Pivotal Quotes: "nine out of ten drugs that enter clinical trials will end up failing." — David Younger: Explaining the scale of clinical attrition and why early screening matters. "we have developed a technology that allows us to screen the effect of a drug on thousands of protein interactions simultaneously" — David Younger: Describing the core function of AlphaSeq. "the infrastructure and time and cost associated with purifying all of those proteins and individually screening a drug against all of those protein interactions is just not tractable" — David Younger: Describing why current specificity-testing approaches do not scale.

Implications: If AlphaSeq performs as claimed, it could help pharma detect efficacy and toxicity risks earlier, reduce expensive late-stage failures, and improve decision-making in programs centered on protein targets and protein-protein interactions.