In the age of “AI can do everything,” many people outside the pharmaceutical industry think scientists are applying AI carte blanche to drug design, specifically the burgeoning field of RNA therapeutics.
But more is not better in drug design. An infinite number of lead candidates does not mean an infinite number of drugs, let alone good drugs. Let’s explore how simple economic levers — specialization and resource allocation — explain why all leads are not created equally.
Specialization: Not all lead candidates are created equally.
Most pharmaceutical and biotech companies pursue specific therapeutic areas or modalities within drug design. This focused specialization is a calculated choice, informed by market constraints (market value, market space, patent boundaries, etc.) and production constraints (productive capacity, capabilities, expertise, etc.).
And every pharma and biotech is looking towards the quickly approaching “patent cliff” in 2030, where we’ll see a number of drug patents expire. In order to protect their intellectual property (IP), pharma and biotech companies must bolster their drug pipelines by working on new products that strategically carve out their IP space.
So just cranking out AI-generated lead candidates, even within the same modality, can spread your focus so broadly that there are no gains in specialization: Not in knowledge, not in IP, and not for patients.
This is especially true if you’re utilizing black-box algorithms for drug design, which is like throwing darts at a target in a pitch-dark room. It doesn’t matter how many darts you throw, you’ll still have no idea if you’re hitting the target (much less the bullseye). And you’ll have no opportunity to adjust your aim, as you cannot learn from your previous throws.
Resource allocation: Navigating tradeoffs in drug design.
Is the misconception of “more is better” driven by AI’s vast computational power? Its ability to crunch massive amounts of data? Its ability to work 24/7? Even AI has to navigate tradeoffs, whether that’s societal or environmental.
But within the pharmaceutical industry, resources are stretched thin. Increasing the number of lead candidates can stretch those resources (human, time, and money) even thinner, impacting the quality of research and development. Focusing on candidates with the highest likelihood of success based on thoughtful, early-stage data is the best way to balance capacity and infrastructure constraints.
Therefore, a lean approach – not just in drug design, but in data – is the best way to increase market impact for bottom lines and patients. The best way to be lean is to understand the “why” within the cycle of innovation and optimization: If you can understand, confirm, and learn while you design, optimize, and experiment, then you have the insights and confidence to focus on the candidates with the highest likelihood of success.
Drug design requires quality over quantity.
Although these can seem like purely commercial concerns, it’s also simply true that focused scientific approaches to lead candidates improves outcomes for patients. Prioritizing depth over breadth accelerates the path to regulatory approval, and ensures that patients will receive a safer, more personalized treatment that significantly improves their quality of life.
At Abzu, we recommend a strategic partnership that limits the number of lead candidates and balances competencies. Although this might surprise some, this approach focuses all teams on areas of mutual interest and gain.
And, in the end, results in developing novel and innovative therapeutics for patients in need.
Lykke is the Chief Pharma Officer of Abzu. She was awarded her PhD in Biophysics from the University of Copenhagen and has worked in RNA therapeutics for over 10 years.
More by Lykke:
