Curing Hair Loss With Sean McClain

Key Concepts

Generative AI in Drug Discovery: Utilizing AI to design novel molecules, specifically antibodies, for challenging drug targets.
Undruggable Biology: Targeting biological mechanisms previously considered too difficult to address with traditional drug development.
Novel Biology: Discovering and targeting new biological pathways or mechanisms.
AI-Driven Drug Development Timelines: Significantly reducing the time from drug discovery to clinical proof-of-concept.
Capital Efficiency in Drug Development: Lowering the investment required to bring a drug to market.
Regenerative Biology: Focusing on reversing damage and restoring biological function, rather than just treating symptoms.
Precision Medicine: Developing highly targeted therapies with minimal off-target effects.
AI and Chip Manufacturers: The strategic importance of companies like Nvidia and AMD in advancing AI-driven life sciences.
Multimodal Models: AI models trained on diverse data types (structure, sequence, affinity, developability).
Translational Data: The need for new data to improve the predictability of AI models in human outcomes.
AI-Centric Organizations: Companies increasingly integrating AI into their core operations and strategies.
Longevity and Regenerative Medicine: Developing therapies to extend lifespan and improve healthspan.
Consumer-Driven Healthcare: The potential for direct-to-consumer marketing and payment for advanced therapies.
Principal-Agent Problem in Insurance: Misaligned incentives between insurers and patients regarding long-term health investments.
Preventative Medicine vs. Sick Care: Shifting focus from treating illness to preventing it.
Life Years: A metric for quantifying the value of extending human life and healthspan.

Generative AI Revolutionizing Drug Discovery and Development

This episode of FYI, the 4-year innovation podcast, features Sean Mclain, co-founder and CEO of Absi, discussing the transformative impact of generative AI on drug discovery and development. Absi leverages generative AI to design novel molecules, particularly antibodies, targeting previously "undruggable" biological mechanisms. This approach not only accelerates the drug development process and reduces costs but also opens up entirely new avenues for treating diseases.

Unlocking Novel Biology and Addressing Unmet Needs

Sean Mclain highlights Absi's success in tackling complex biological targets that have eluded traditional methods.

HIV Drug Discovery: Absi partnered with Caltech to design an antibody targeting the highly conserved "caldera region" of the HIV virus. This region is crucial because it remains unchanged across different HIV clades, offering a potential pathway to a universally effective neutralizing antibody and a vaccine. Previous attempts to drug this deep crevice were unsuccessful due to its complex structure.
Dermatological Ion Channels: In collaboration with Almo, Absi designed an antibody to block a specific ion channel implicated in dermatological diseases, a target that had been known for over 30 years without a successful therapeutic intervention.
Pain Management (Potassium Ion Channels): Mclain uses the example of potassium ion channels KV 1.7 and 1.8, which are known to regulate pain. Genetic data shows individuals with these channels knocked out do not feel pain, suggesting that blocking them could lead to non-addictive, non-opioid pain medication. This represents a significant unmet need in pain management.

Accelerating Drug Development Timelines and Reducing Costs

A key theme is the dramatic acceleration of the drug development pipeline through AI.

ABS 2011 (Hair Regrowth): Absi is developing an antibody (ABS 2011) targeting the prolactin receptor for hair regrowth. This mechanism involves shunting hair follicles from a dormant, apoptotic state back into an active growth phase, potentially leading to hair regrowth and repigmentation.
- Timeline: The program, initiated 2-2.5 years ago, is projected to reach a Phase 2 proof-of-concept readout in humans within approximately 3.5 years from the program's start. This is significantly faster than the typical 3.5-5 years to get a drug into the clinic, followed by years for proof-of-concept.
- Cost Efficiency: The development to Phase 2 proof-of-concept is estimated to cost around $15 million, a fraction of the $50-100 million typically required to enter the clinic, and billions for full development.
Regenerative Biology: ABS 2011 exemplifies Absi's focus on regenerative biology, aiming to reverse age-related damage. The mechanism involves blocking the prolactin receptor to counteract prolactin buildup on the scalp, which drives follicles into a dormant state. By blocking this receptor, stem cell growth is restimulated, promoting hair growth and pigmentation.
Serendipitous Discovery and AI's Role: The prolactin receptor target was initially discovered by Bayer while researching endometriosis. Animal studies showed hair regrowth in mice treated with a drug targeting this receptor. Bayer did not pursue it commercially due to the need for frequent injections (24 over six months) for efficacy, making it commercially unviable. Absi's Chief Innovation Officer, Andreas Busch, brought this knowledge to Absi, where their generative design platform was used to create a more efficient antibody, enabling less frequent dosing (2-3 times over six months) and commercial viability.
Combined Phase 1/2 Trials: For ABS 2011, Absi is conducting a combined Phase 1/2 trial to assess both safety and hair regrowth. This is feasible because individuals with androgenetic alopecia (AGA) are generally healthy.
Projected Market Entry: If the readout is positive next year, Absi plans to file a BLA with the FDA, seeking approval around 2030, representing a rapid turnaround from discovery to market.
AI's Impact on Timelines and Success Rates: Kathy Wood of ARK Invest notes that their research projects AI can shrink drug development time to market from an average of 13 years to 7-8 years. Absi's ABS 2011 program aligns with this projection.
Cost Reduction: The total investment to bring ABS 2011 to market is projected to be sub-$100 to $150 million, a significant reduction from the average $2.4 billion cost (including failures) for traditional drug development. ARK's cost estimate, including failures, is $600 million per drug.

Precision Medicine: Engineering the "Key" for the "Lock"

The conversation emphasizes the shift from searching for solutions to engineering them.

From Searching to Creating: Mclain likens the process to moving from searching for a needle in a haystack to creating the needle itself. Absi can now engineer specific features into antibodies, such as binding in tumor microenvironments at acidic pH but not in healthy tissues, or activating/blocking specific pathways.
Faster Failure and Learning Loops: This precision allows for faster hypothesis testing, enabling quicker identification of viable drug candidates and more efficient failure of unpromising ones, dramatically shortening the learning loop in drug discovery.
Multimodal AI Models: Absi utilizes multimodal models trained on protein structures, antibody sequences, affinity to targets, and developability parameters (manufacturing feasibility, high titers, solubility, low cross-reactivity, low immunogenicity).
Prompting the Inference Engine: The "prompt" for the inference engine involves providing the target structure (e.g., the caldera region of HIV) and specifying the desired outcome (an antibody that binds to a particular epitope). The AI then designs the antibody's binding "fingers" while considering all developability parameters. This is akin to providing the "lock" and having the AI design the precise "key."

Strategic Partnerships with Chip Manufacturers

The involvement of major chip manufacturers like Nvidia and AMD underscores the critical role of computing power in AI-driven drug discovery.

AMD's Investment in Absi: AMD invested in Absi, attracted by their commitment to life sciences and the specific capabilities of AMD's chips.
Higher Memory for Enhanced Training: AMD's chips offer higher memory, allowing Absi to train models on entire protein structures rather than cropped inputs. This leads to higher fidelity and more accurate models by incorporating more comprehensive information.
Nvidia and Recursion: Nvidia has invested in Recursion Pharmaceuticals, another AI drug discovery company, highlighting a broader trend of chipmakers supporting this sector.
Healthcare as the Next Big Unlock: Both AMD and Nvidia view healthcare as a significant growth area, anticipating astronomical compute utilization due to the complexity of biological systems and their suitability for AI and deep learning.
Training vs. Inference: Absi, as an R&D-focused company, primarily uses chips for training. However, they are seeing increasing use for inference, where AMD chips are noted for their efficiency, which is crucial as inference costs rise.
Public Company Investments: Absi and Recursion are noted as the only public companies that Nvidia and AMD have invested in, positioning them uniquely.

The Future of Drug Discovery: Data Generation and Translational Insights

The conversation touches upon the evolving landscape of data in AI drug discovery.

Need for New Data: While existing data has been valuable, the field is reaching a point where new data generation is necessary to train advanced AI models.
Translational Data Gap: A critical gap exists in "translational data" – data that accurately predicts how a molecule will behave in humans. This data is difficult to generate, requiring both animal and human studies.
FDA's Shift Away from Animal Testing: The FDA's guidance encouraging the replacement of animal models with AI models is a significant development. Absi is working towards creating AI models that can predict immunogenicity, safety, toxicity, and efficacy, aligning with the FDA's direction. This shift is expected to streamline drug discovery and development and reduce costs.
Precision Medicine and Individualized Treatments: By increasing the probability of success and decreasing costs, AI can enable the pursuit of treatments for smaller patient populations, moving towards individualized medicine.

Longevity and Regenerative Medicine: A New Frontier

The discussion expands to the burgeoning field of longevity and regenerative medicine.

The Scale of Longevity: The longevity market is estimated to be a massive $1.3 quadrillion opportunity, considering lost life years due to early death and declining health.
Sarcopenia and Muscle Wasting: Absi is exploring targets for sarcopenia (age-related muscle wasting), which can lead to significant strength degradation and increased mortality risk from falls.
Extending Lifespan and Healthspan: The goal is to not only extend lifespan but also to maintain vitality and function throughout life, reversing the damage of aging.
Consumer-Driven Demand: Therapies for conditions like hair loss and sarcopenia may be driven by consumer demand and out-of-pocket spending, especially as individuals seek to maintain their identity and quality of life.
GLP-1 Side Effects and Antidotes: The conversation notes that GLP-1 drugs, while beneficial for weight loss, can accelerate muscle degradation and cause hair loss. Absi's work on regenerative medicine could potentially offer solutions to these side effects.
Preventative Medicine and Insurance: The potential for preventative therapies to reduce long-term healthcare costs is significant. However, the principal-agent problem in insurance, where insurers have short-term horizons, may hinder upfront investment in preventative treatments.
Regenerative Longevity: Beyond preventing disease, the focus is on regenerating biological components (like hearts or lungs) to enable individuals to live beyond current estimated maximum lifespans (around 120 years).
ABS 2011 Pricing and Business Model: For ABS 2011, pricing is envisioned between Botox and hair transplants ($5,000-$10,000). The therapy is expected to be pulse-based, with a durable effect lasting 2-6 years, potentially leading to subscription-like models.

Market Perceptions and Future Outlook

The conversation addresses the current market sentiment towards AI drug discovery.

Biotech "Nuclear Winter": The biotech sector has experienced a challenging period ("nuclear winter"), disproportionately affecting earlier-stage AI drug discovery companies.
Investor Skepticism and Validation: Investors and analysts are seeking proof of efficacy in humans, translating preclinical data into clinical success. Upcoming Phase 2 readouts from companies like Insilico, Recursion, and Absi are expected to provide strong validation.
Market Turnaround: A potential market turnaround, driven by falling interest rates and increasing investor interest, is anticipated.
Value of AI Platforms: The commercialization of AI-discovered drugs will not only validate individual assets but also the underlying AI platforms, creating a competitive moat for companies like Absi.
Long-Term Potential: Despite the inherent uncertainties and unexpected failures in drug development, the long-term promise of AI in drug discovery is substantial, with the potential for high returns on R&D and a vastly expanded target set.

Conclusion

The discussion with Sean Mclain and Kathy Wood highlights a paradigm shift in drug discovery driven by generative AI. Absi's work exemplifies how AI can unlock novel biology, accelerate development timelines, reduce costs, and enable the creation of highly precise, regenerative therapies. The strategic investments from chip manufacturers and the evolving regulatory landscape further underscore the transformative potential of AI in healthcare, paving the way for a future of more effective, personalized, and preventative medicine, with significant implications for human longevity and well-being.