by Giuliana Miglierini
Generative AI is perhaps the more advanced form of artificial intelligence available today, as it is able to create new contents (texts, images, audio, video, objects, etc) based on data used to train it. Applications of generative AI are not limited to, for example, the famous ChatGPT chatbot used to write complex texts, or to algorithms producing incredible images.
Generative AI is becoming a new paradigm in drug discovery, as it promises to greatly reduce both time and costs to develop new molecules, or to repurpose already existing ones for new indications. A fundamental goal for pharmaceutical companies, given that the average cost of developing a new medicines is estimated at $2.6 billion.
Algorithms can be trained on chemical-physical characteristics and 3D shapes of molecules in order to generate completely new molecules of interest for a certain application, and/or to predict their behaviour in the biological context (e.g. binding to a specific receptor). We resume the current status of implementation of generative AI in the field of drug development.
Quintillions of data
It seems ages since the first full sequencing of the human genome was completed in year 2000. Since then, vast amounts of genomic and other biological data have rapidly accumulated. To give an idea, the National Human Genome Research Institute estimates between 2 and 40 exabytes (i.e. quintillions) of data available within the next decade. The number becomes even more larger when considering other domains relevant to drug development, including chemical structures and properties, complex biochemical pathways, 3D protein structures and receptors, data on the efficacy and toxicity profile of already approved medicines and candidates in the pipelines, etc.
No matter to say, the parallel growing interest in artificial intelligence that characterised the last twenty years has turned fundamental for the availability of new technologies able to digest, extract and analyse these extremely large datasets.
Machine learning and deep learning algorithms represented just the first step towards this goal. Generative AI came as a consequence, its birth is attributed to a paper by Ian Goodfellow et al., published in 2014.
Opportunities and challenges of generative AI for drug discovery
The implementation of generative AI in the pharmaceutical and medtech sectors may lead to the an estimated economic value of $60-110 billion/year, says the report by McKinsey and Co. “Generative AI in the pharmaceutical industry: Moving from hype to reality”.
More specifically, McKinsey analysed 63 generative AI use cases in life sciences, calculating the potential economic impact for different domains. The higher values ($18-30 bln) are expected for the commercial domain, followed by research and early discovery ($15-28 bln) and clinical development ($13-25 bln). Less impacted appear enterprise ($8-16 bln), operations ($4-7 bln) and medical affairs ($ 3-5 bln).
Implementation of generative AI may prove not a so easy exercise for pharma companies, as it has to fit within an already complex organisation and with the strict regulatory requirements typical of the pharmaceutical lifecycle. An important message comes from the analysis from McKinsey: it is of paramount importance to exit the hype climate surrounding generative AI and understand exactly what it can and cannot be done.
The question is highly complex to be solved, and it requires multiple skills (data scientists, researchers, medical affairs, legal, risk and business functions) jointly working to set up the solution more suited to each company. The availability of a proper data infrastructure is just the first step, the chosen generative AI model has to be adapted to the complexity of the specific case of use, focusing on key applications to avoid disruption of the business.
According to an analysis by Boston Consulting Group, generative AI may prove useful to include also unstructured data among those used as data sources by the pharmaceutical industry. Possibly a challenging goal to achieve, as data access and management must fulfil regulatory requirements, for example in relation to the possibility to use data generated in clinical trials to support regulatory approval.
Governance of generative AI must also reflect the key principles established in the EU for AI systems, i.e. they “must be ‘safe, transparent, traceable, non-discriminatory and environmentally friendly,’ as well as ‘overseen by people, rather than by automation, to prevent harmful outcomes’.”
The need to integrate generative AI with human activities would probably call companies to redesign core processes. To this instance, selection of the more suited AI infrastructure and platform may turn critical for success of the initiative. Integration with already existing AI tools and flexibility are among other main features to be kept in mind. Not less important is the choice of the right partners, that should fit with the strategic business goals.
Many algorithms already available
The first AI applications based on deep learning algorithms were used, for example, to predict the sequence and structure of complex biological molecules. It was the case of the AlphaFold Protein Structure Database, which contains over 200 million protein structure predictions freely available to the scientific community. Other algorithms of this kind are ESMFold (Evolutionary Scale Modeling) and and Microsoft’s MoLeR, specifically targeted to drug design.
A more recent generation of generative AI are IBM’s MoLFormers UI, a family of foundation models trained on chemicals which can deduce the structure of molecules from simple representations. MoLFormer-XL screening algorithm, for example, was trained on more than 1.1 billion unlabelled molecules from the PubChem and ZINC datasets, each represented according to the SMILES notation system (Simplified Molecular Input Line Entry System). As reported by IBM, MoLFormer-XL is able to predict many different physical, biophysical and physiological properties (e.g. the capacity to pass the blood-brain barrier), and even quantum properties.
Mutual Information Machine (MIM) learning is the approach used by NVIDIA to built its MolMIM algorithms, a probabilistic auto-encoder for small molecule drug discovery. The NVIDIA BioNeMo cloud service uses these models to deploy a generative AI platform to create molecules that, according to the company, should fulfil all properties and features required to exert the desired pharmacological activity.
Not only big players: many new companies were born specifically to support the creation of generative (often end-to-end) AI platforms for drug discovery. Among the main ones, Insilico Medicine’s Pharma.AI platform is being used to build a fully self-generated pipeline comprehensive of 31 programs and 29 targets. The more advance product under development targets the rare disease idiopathic pulmonary fibrosis and is currently in Phase 2 in the US and China. The company’s inClinico AI data-driven multimodal platform to calculate the probability of success of single clinical trials proved useful to predict outcomes of Phase 2 to Phase 3 trials and to recognise weak points in study design.
UK’s based Exscientia, founded in 2012, is an AI-driven precision medicine company. Among its main achievements is the creation of the first functional precision oncology platform to successfully guide treatment selection and improve patient outcomes. The more advanced product in its pipeline is GTAEXS617, an oncology product targeting CDK7 in advanced solid tumors.
These are just few main examples, you can learn more on companies focused on AI for drug discovery in these articles published on Forbes and Pharmaceutical Technologies.
Reactions to the proposed ban of PFAS
by Giuliana Miglierini
A proposal to ban around 10,000 per- and polyfluoroalkyl substances (PFAS) was submitted in January 2023 to the European Chemicals Agency (ECHA) by authorities of Germany, Denmark, the Netherlands, Norway, and Sweden. The proposal was published on ECHA website on 7 February 2023.
The focus is the so-called “forever chemicals”, i.e. very high persistence PFAS typically characterised by bioaccumulation (also in plants), great mobility and a long range transport potential, and potential endocrine activity.
“This landmark proposal by the five authorities supports the ambitions of the EU’s Chemicals Strategy and the Zero Pollution action plan. While the evaluation of such a broad proposal with thousands of substances, and many uses, will be challenging, we are ready.”, said Peter van der Zandt, ECHA’s Director for Risk Assessment.
The proposal was open to public consultation on 22 March 2023, giving rise to the collection of 5,600 comments. Opinions will be issued by ECHA’s scientific committees for Risk Assessment (RAC) and for Socio-Economic Analysis (SEAC), to be then forwarded to the EU Commission for final decision.
The current role of PFAS
PFAS are characterised by the presence of alkyl groups in which many or all the hydrogen atoms have been replaced with fluorine. The main carbon chain of these substances may have different lengths, from small molecules to long chain PFAS and polymers, and may carry a very wide variety of other functional groups. The strength of the carbon-fluorine bond is the root cause of PFAS persistence, leading to these substances remaining in the environment for decades to centuries.
Per- and polyfluoroalkyl substances are currently used in many different industrial sectors, thanks to their useful technical properties. Among others, PFAS can be used to repel water, oil and dirt from surfaces, and is characterised by a high durability under extreme conditions of temperature, pressure, radiation, and chemicals. PFAS also present electrical and thermal insulation properties.
The main features of the restriction proposal
According to the authorities that submitted the proposal, around 4.4 million tons of PFAS would end up in the environment over the next 30 years in the case of no action. Ban would refer to manufacture, placing on the market and use as such, as constituent in other substances or in mixture as well as in articles.
Two options for restriction have been considered, a full ban or specific derogations for certain industries, based on the analyses of alternatives, efforts put in place for switching to them, and socio-economic considerations. The ban would be effective above a set concentration limit; a transition period of 18 months should occur between final adoption and entry into force. Use-specific, time-limited derogation might refer, for example, to a 5-year period in the case of food contact materials for industrial food and feed production (as alternatives are already under development, but are not yet available to entry into force), or to a 12 years for implantable medical devices (for which identification, development and certification of alternatives is still needed).
During the public consultation phase, comments were received from more than 4,400 organisations, companies and individuals, to be reviewed by both the RAC and SEAC committees and the five proposing countries. Sweden, Germany and Japan are the countries that contributed the higher number of comments, well in advance of Belgium, China, Italy and the US. Companies provided more than the half of the comments (58,7%), followed by individuals (27,3%), and industrial or trade associations (9,8%). The full list of entities participating to the consultation is available at the consultation webpage.
EFPIA response to ECHA’s consultation
The European Federation of Pharmaceutical Industries and Associations (EFPIA) contributed to the consultation with a detailed document. Another joint ISPE-EFPIA document particularly addressed the use of fluoropolymers and fluoroelastomers in medicinal product manufacturing facilities.
“While we support the need to restrict certain PFAS, we need to find the right approach to ensure the continued manufacturing and availability of medicines in Europe. A total ban would see medicines’ manufacturing in the EU grind to a halt in under three years. It would also jeopardise the production of all pharmaceutical substances in Europe and would conflict with the EU’s strategy of reducing dependency on nations outside of the EEA in the event of shortages or pandemics.”, said EFPIA’s director general, Nathalie Moll.
EFPIA’s consultation documents highlights the many different uses of PFAS in the pharmaceutical industry, ranging from active pharmaceutical ingredients (API) falling within the definition of PFAS used in the proposal, to building blocks and raw materials used within chemical synthesis of PFAS and non-PFAS medicines. Other reagents and equipment might also fall within the scope of the ban, as well as packaging materials or combination products such as pre-filled syringes. The ban would also affect the manufacturing process, where PFAS materials are used in a wide variety of applications.
It might thus result in the disappearance from the market of a large number of important medicines, warns EFPIA, due to the unavailability of replacement materials, and the time required to obtain regulatory re-approval of alternatives. The supply chain of pharmaceuticals would be also impacted at many stages, thus possibly exacerbating shortages.
In its analysis, EFPIA highlights how some PFAS are considered of low concern by the OECD, and in particular “those used in actual medicines have no or low identified risk through medicines risk benefit or environmental risk assessments”.
A patient access impact analysis was also jointly developed by the involved industrial associations (AESGP, EFCG, EFPIA, Medicines for Europe and Vaccines Europe), showing that the current proposal would lead to at least 47,677 global marketing authorisations being affected by the ban. More than 600 medicines from the WHO Essential Medicines List would be at risk; restrictions would greatly impact also the European Member State’s “Critical Medicines lists”.
EFPIA submitted also a socio-economic assessment of the proposal, according to which a broad restriction of PFAS used in the production of human medicines would have disproportionate negative impacts on the European economy and society. “Without additional derogations, the entire pharmaceutical industry would no longer be able to manufacture active pharmaceutical ingredients (APIs) (whether classified as PFAS or non-PFAS APIs) or associated medicinal products in the EEA”, writes EFPIA, resulting in APIs production to necessarily move out of the European Economic Area.
The position of the medical device sector
MedTech Europe also published a position paper on the PFAS restriction proposal and called for “a realistic transition pathway to non-PFAS alternatives that are both reliable and feasible for medical technologies (including their manufacturing and supply chain) to avoid shortages of medical technologies for patients and practitioners”.
The position paper presents many PFAS use cases in the field of medical devices, together with the criticalities posed by the proposed transition. In particular, broad derogations should be considered to allow sufficient time to first “identify all PFAS uses in medical technologies and to subsequently move to alternatives where these are proven to be technically viable, available besides in conformity with the sector-specific MD and IVD Regulations so as fit for the intended purpose”. In this case too, the need to manage complex supply chains would require a realistic timeline in order to address dependencies, and long development timelines and steps to ensure compliance with the sectorial legislation.