Emmanuelle Quilès, President of Janssen France, a subsidiary of Johnson & Johnson, is in charge of the digital transformation of the company and improving the standard of care offered to patients through innovative partnerships and solutions. Quilès, a biotechnological engineer, began her career as a clinical research associate first at Rhône-Poulenc, then at Pierre Fabre and then at Genetics Institute until Wyeth Pharmaceuticals bought the company. She held marketing positions in the hemophilia field before being appointed head of Wyeth’s rheumatology and dermatology business unit. In November 2007, she became President of Wyeth France and after Pfizer’s acquisition, she was confirmed as President of Pfizer France. In December 2012, she left Pfizer to set up Harmonium, a start-up specialized in diabetology. Quilès was appointed President of Janssen France in January 2015 and is also President of the Janssen Horizon Endowment Fund, which facilitates public-private partnerships in the field of translational research. She is a scheduled speaker at the September 5 AI For Health conference in Paris, which Janssen co-founded. Quilès recently spoke to The Innovator about the impact of artificial intelligence on health.
Q: You are a speaker at a Sept. 5 Paris conference on AI For Health. What do you plan to talk about?
EQ: I am participating in a round table on predictive medicine and personalized care. We will talk about the disruption of our business model. At Johnson & Johnson, we are convinced that it is time to look at healthcare in a different way. Instead of looking at people as patients for whom we are trying to find a cure we need to start looking at people who are healthy and try to intercept or stop disease progression. That is the purpose of our company. Our tagline is ‘ creating a future where disease is a thing of the past.’ This is a completely new era. We have to do it or pharma companies will not have a future. In 2015 we created a new team called the Disease Interception Accelerator. It is a research unit focused on disease interception strategies that include addressing the root causes of disease; intervening earlier than today’s clinically accepted point of diagnosis; and seeking solutions that stop, reverse or inhibit progression to that disease. The group is working to integrate science, medicine, diagnostics and new business models to create totally novel solutions to intervene earlier and deter disease onset. We are concentrating our efforts on disease areas of high unmet need and where there is exciting potential for interception solutions, including gestational diabetes mellitus, chronic obstructive pulmonary disease, and perinatal depression, just to name a few. To date, we have assembled a committed and driven group of Janssen scientists and external collaborators who believe, like we do, in the possibility that through cutting-edge science and innovative methods, we can together shift the focus toward the preservation of health, as opposed to the treatment of disease.
Q: Today there are an estimated 9,000 untreated diseases and 300 million people suffer from rare diseases with little hope of effective or life-saving treatment. Even in an age of unprecedented technological advancement, developing a new drug costs big drug companies an average of €2.5 billion, takes 10 to 15 years, and more than 95% of efforts fail. How specifically is Janssen leveraging AI to improve the process?
EQ: We are using AI in three ways. First, it is helping enhance our knowledge of what happens during the course of a disease. We know what happens at the end stage of the disease. AI will allow us to better understand what causes diseases and how to prevent them. The second thing AI is helping us with is to tailor molecules to the disease mechanism in the early stages to stop diseases from progressing. And, AI is helping us implement these things into the usual health care process.
Q: Can you tell us more about your recent agreement with Iktos Pharma, a French company that applies AI to drug design? What do you hope this collaboration will achieve?
EQ: Iktos Pharma will help us with our global ambition of improving drug discovery. It uses in silico drug design, a term that means ‘computer-aided molecular design’, or in other words the rational design or discovery of drugs using a wide variety of computational methods. Especially in the early phases of drug discovery projects, in silico tools based on the knowledge of the target receptor structure (protein-based) or on the chemical structure of active small molecules (ligand-based) are routinely used for discovering and optimizing hit or lead compounds. It is a new way of designing drugs. In the past, drugs were designed using three dimensional models that are very complex. In silico does a better job predicting how effective a molecule is going to be, improving speed and efficiency. The amount of money needed for drug development is huge and also very long. If you can reduce the time needed then indeed you can reduce the cost of development.
Q: Is Janssen working with other private companies or startups on the application of AI to drug discovery?
EQ: We have multiple partnerships. We want to support the ambition of the government to position France as a leading country for AI. There are three projects we are working on that use AI to look at real life data. It is a new term in the pharma industry that refers to data collected about patients during normal life rather just during clinical trials. This data is very important. There is not enough money to run control trials because it includes too many patients. AI allows us to efficiently and cost effectively look at how patients are treated in Europe and try to compare best outcomes from different countries, different centers and different doctors. The three studies include MyLord, an epidemiological study using an algorithm to estimate the annual incidence, prevalence and mortality of French patients suffering from multiple myeloma; ORACLE, an epidemiological study using an algorithm analyzing unstructured textual data from electronic clinical records to characterize the disease and describe the care pathways of European patients hospitalized for prostate cancer; and Redress, a study to develop a solution using an algorithm detecting, among a large number of signs and symptoms, early markers of a response or non-response to antidepressant therapy.These studies might help us to adapt the treatments. Clinical trials will always be there. Everyone needs them, but the question is — once a drug is deemed safe and the patient knows how to use it — are there ways we can optimize a treatment or look at other ways of implementing the treatment? We also launched in April of this year a datathon to see how startups can help us address other challenges. Our April datathon winners are Embleema in oncology, for its blockchain solution built around the patient, which allows the connection of patient cohorts in real time; Focus Patient in Immunology, for its project that enhances the design of a broader study by integrating a strengthened patient experience, particularly via a connected watch and the involvement of caregivers; Bioserenity in psychiatry, for its application which allows the collection of subjective and objective (physiological) data based on items on a qualified depression scale and reinforced by data collected by the caregiver; Andaman7, for its shared patient record solution and intelligent module allowing more interactive data collection and better support for patients in the follow-up of their disease. There are lots of startups with great ideas in healthcare, but in most cases they don’t know how to comply with all the regulatory constraints. Janssen knows about market access and knows the pain points of the patient pathway; the start-ups have both the creative ideas and the technology. Together, we are able to solve issues.
Q: Janssen is a member of new consortium of pharmaceutical, technology and academic partners who have come together under the “MELLODDY” (Machine Learning Ledger Orchestration for Drug Discovery) project which aims, for the first time, to use machine learning methods on the chemical libraries of 10 pharma companies and to develop a platform creating more accurate models to predict which compounds could be promising in the later stages of drug discovery and development. Why did Janssen join and what do you hope the project will achieve?
QE: This is a great project. Janssen is in the lead along with nine pharmaceutical companies: Amgen, Astellas, AstraZeneca, Bayer, Boehringer Ingelheim, GSK, Merck KgaA, Novartis, and Institut de Recherches Servier. The project also includes two academic universities: KU Leuven, Budapesti Muszaki es Gazdasagtudomanyi Egyetem, four subject matter experts: Owkin, Substra Foundation, Loodse, Iktos and one large AI computing company: NVIDIA. What is interesting is that the project involves not simply pooling data but using information from others so we don’t repeat something that has already been done while still securing our own intellectual property. This is the first time there has been such a large collaboration that brings all these researchers together.
Q: What are some of the challenges that need to be overcome before big pharma can fully harness the power of AI?
EQ: The biggest challenge is to open up our organization to the external world. Johnson & Johnson has been doing this now for a few years but it takes a while for everyone in the organization to embrace this new mindset and accept that it is essential to embrace the AI ecosystem. Pharmaceutical companies are very traditional. We tend to keep our little secrets but now we need to open our doors not only to startups but also to public research teams. It is a big challenge for all of us. An event like “AI For Health” brings together all the players. It is a new way of collaborating. The second challenge is access to data. This sector is generating a lot of data — both patient data and scientific data. One of the biggest issues is interacting with hospitals. How do we interact? How do we merge databases? And how will it be accepted by the operational teams as we don’t have the same way of storing data? There are also ethical challenges around AI. We have to make sure the data that is being shared — even if it is anonymized — is secure. And who will be making medical decisions? If tomorrow the doctor is not the one what happens if it is the AI and the AI is wrong? How will we collaborate if the general public does not agree and does not understand that there are ethical issues but they are being dealt with? We are at a major turning point. The ambition we have to eradicate disease is not unrealistic at this stage. That is why it is such an exciting time.