2018 has seen a surge in interest in artificial intelligence (AI) and its use in the pharmaceutical industry. When considering the reasons for this, the answer is not that AI is new. ‘AI’ is an umbrella term covering a spectrum of technologies, from smart algorithms to machine learning. In recent years, advances in AI have accelerated. Yet, for some time now, AI has been used across a wide variety of industries. This includes the pharma industry, where humans have been assisted by computer programs capable of modelling compounds or simulating reactions, for example.
Arguably what is garnering the interest and imagination of not only the pharma industry, but also interested parties such as healthcare providers and regulators, is the growing number of cases where AI is being used in drug discovery successfully. This advance is already disrupting long-standing ecosystems in the pharma industry, and prompting important debates about how AI-derived innovations should be regulated.
In drug discovery AI is promising to cut the time and costs of generating a hit to candidate, from around five years to one or less. Specifically, AI is being used to identify and plan the synthesis of new molecules or known molecules for new uses, by analysing vast amounts of public and proprietary data, including scientific literature, the results of clinical trials and compound libraries.
So far, the results have been promising. Earlier this year, it was reported that UK-based start-up, BenevolentAI, having used AI to isolate five lead compounds for the treatment of amyotrophic lateral sclerosis (ALS), plans to take several of these compounds forwards for clinical trials. Other companies, such as Exscientia and Healx (also UK-based), have been reported to have made similar advances for different therapeutic uses.
The use of AI in drug discovery is already having a tangible effect on the industry’s long-standing ecosystems. Companies wanting to take advantage of the opportunities offered by AI are having to seek expertise from, and divest some of their R&D to third parties, including novel players in the field such as tech start-ups.
Examples of drug discovery collaborations reported this year include GSK, which entered into a $43M collaboration with Exscientia, and Pfizer, which entered into a collaboration with IBM Watson.
AI is also prompting the rise of much smaller – typically tech – companies. In April, BenevolentAI generated $115M in funding, and is reported to have, besides its ALS programme, over 20 other drug discovery programmes in progress. What’s more, these ‘smaller players’ are looking to bring the entire AI and R&D process in-house – a feat which could now be possible as AI promises to slash the costs of R&D.
The emerging use of AI in drug discovery is prompting important questions about how AI, and AI-derived innovations should be regulated. Earlier this year, companies including BenevolentAI, GE Healthcare and Insilico Medicine formed the Alliance for Artificial Intelligence in Healthcare (AAIH). The AAIH plans to become an industry advocate for public policy, regulation and market access for AI-developed products. Another, related concern increasingly aired this year is the patentability of AI and AI-derived innovations. The pharma industry traditionally relies on patents to help protect innovation and fund future R&D, yet the use of AI could pose challenges for meeting the inventive step requirement for obtaining a patent.
This was among the topics discussed at the European Patent Office’s (EPO) first ever conference on AI in May. The EPO has since published a draft of its updated guidelines on patenting, which include a new section devoted to AI. Whilst there are no specific provisions on AI-derived innovations, what’s clear is that, for the time being at least, the usual rules on patenting algorithms will apply to the AI in the life sciences space.
Imogen Ireland is a life sciences IP associate at Hogan Lovells
