The people behind healthcare innovation: Interview with Luca Naef, CTO at VantAI

As part of my work, I have the privilege to speak with many inspiring innovators. However, the business community usually focuses more on companies, pitches, and valuations, and less on the innovator. I think it would be interesting to learn a bit more about the people behind healthcare innovation. Therefore, I’m sharing some of my conversations with innovators in a condensed format: six questions and six answers about their experience, their opinions, and their learnings.

The latest conversation is with Luca Naef, CTO of VantAI, an AI-first biotech.

Tobias: What’s your story, and how did you become an innovator in healthcare? 

Luca: I come from an engineering family: my grandfather was an engineer and my father, a car mechanic who switched to IT, is a real-life MacGyver [1]. I have always thought engineering is the closest thing to magic. My first exposure to computer science was through building computers from spare parts, later working as a software engineer. Naturally, I wanted to understand the world’s most complex machines and focused my studies on molecular biology at ETH Zurich. Around then, deep learning had its landmark moments with AlexNet in 2012 which opened my eyes to its potential.

Doing research in Stanford, building hardware and software for a cell screening device was my first practical exposure to computer vision and led me to focus on deep learning for my Master’s degree. I really wanted to see deep learning’s potential at industry scale and so I joined McKinsey Digital. There I met Zach Carpenter who later started VantAI as a Roivant spinout in 2019. I joined him very shortly after and we built VantAI, an AI-first biotech, from scratch to be the leading machine-learning company in a new field called “induced proximity” [2] and are working with the world’s largest pharmaceutical companies as well as early pioneers in the field.

Tobias: What do you see as the most exciting areas of drug discovery in the next ten years? 

Luca: Two areas excite me strongly. The first is the field of induced proximity. Traditionally, there are two types of drugs: 1) small, chemically synthesized molecules and 2) biologics, often produced by cells and extracted. The small molecules mostly work like keys, fitting into the “lock” of a disease-causing protein. This works great if your problem is a jammed, or overactive protein. If you can’t drive your car because of a locked door, a key helps. With a flat tire, it won’t. Biologics are more like a replacement tire, replacing a broken part. But they are more expensive and larger, making it hard for them to reach disease proteins.

Most cars have replacement tires, so what if we could just use a screwdriver to mount a replacement tire that’s already there? Proximity-inducing molecules are these “screwdrivers”; they use proteins that are already inside the cell to fix the disease-causing protein, by pulling both into proximity. Thus, they combine the best aspects of biologics and small molecules in one. The potential of this approach is quite mind boggling: we only had keys, and now we can use any of the cell’s own 20,000 proteins to fix diseases.

The second area is artificial intelligence. AI has already led to major scientific breakthroughs like Google’s AlphaFold2, which can predict protein structures. As it turns out, proximity-inducing molecules are uniquely suited for this. They mimic the natural interactions of proteins, from which deep-learning models can learn, something we cannot do with classical small molecules. These are VantAI’s two core focus areas.

Tobias: Looking more broadly, what are the biggest opportunities and obstacles you see for innovation in the healthcare environment?

Luca: Low success rates for developing cures for biologically complex diseases such as cancer and Alzheimer’s clearly show that our biggest obstacle is a lack of understanding of the underlying biology causing the disease. Therefore, we often go after the wrong causes. However, I see one of the biggest opportunities in this area. With the advent of personal genomics, which then expanded into other omics areas, like proteomics and metabolomics, we are for the first time ever able to create molecular data to understand disease drivers across a large population. Already today, drugs with genomic rationale tend to be more successful. 

Tobias: When you look at the health system as a whole (providers, payers, doctors, patients) who do you see driving innovation the most?

Luca: I believe that we’re in the middle of a fundamental shift in who is driving innovation. I worked with large pharmaceutical companies in the past, and today I’m building a biotech company that partners with pharmaceutical companies. But I’m also part of a highly tech- and health-savvy generation increasingly in charge of their own health and fascinated with technology-first, decentralized systems. This is a transition from past decades, when academia and large public and private organizations have traditionally driven most innovation.

Today’s generation is much more involved in their health and research. From activity trackers to consumer genome sequencing, people increasingly use devices to generate data to understand their health. Today, billionaires and everyday people have the same supercomputers in their pockets, more powerful than institutional supercomputers a short time ago. Similarly, the scale and accuracy of personal health devices will surpass current institutional standards and become ubiquitous. The unsolved challenge is to orchestrate people, devices, and data to deliver scientifically valid studies. There are exciting early efforts: Wearables let you participate in trials from your wrist, and in the field of longevity, large community-science studies are now partially conducted or funded by people directly.

Tobias: In your opinion, what is the single most important change that will better enable digital transformation in health systems?

Luca: Non-standardized data formats and exchange protocols, especially around experimental data, patient data, and the academic literature are creating massive amounts of friction and stifling innovation. It’s like trying to organize a world cup but everybody plays a different sport and keeps score differently, so how would you choose a winner? Protocols might sound boring but are crucial. Internet communication protocols were the driving force behind the internet revolution because they enabled infrastructure platforms, like cloud computing. With the cloud, startups don’t need to buy server racks for millions of dollars and this allows them to generate technology for consumers at unprecedented scale. We need strong, possibly decentralized institutions to drive the same in the health sector.

Tobias: What do you know now that you wish you had known when you were starting out as an innovator? 

Luca: Drug discovery is one of the most interdisciplinary team sports and requires experts from various fields. Applying artificial intelligence intensifies that. I learned that my biggest lever as CTO isn’t working on the technology but bringing together a world-class team. I underestimated that in the early days. So, for readers who work in a similar space, please reach out. We love to collaborate and have open positions across all areas.

[1] Angus “Mac” MacGyver is the main character in the eponymously titled TV series, which features the adventures of a non-violent problem solver.

[2] Induced proximity is a novel approach to tackle difficult-to-drug targets and diseases through the development of medicines that act like a molecular matchmaker by bringing two proteins together.

For more information, see Luca Naef and VantAI.

Disclaimer: The views and opinions expressed in this article are solely those of the author and his guest contributor and do not reflect the views of McKinsey & Company.