“Doctors pour drugs of which they know little, to cure diseases of which they know less into patients of whom they know nothing,” claimed Molière the French playwright and notorious hypochondriac. Thankfully this is no longer the case.
We now know the structure of our drugs, so we can accurately design them. We know the physiology of diseases and pathogens, so we can identify targets to aim for when designing drugs. We know more about our genetic make-up than ever before, so we can design a therapy for an individual. There is no longer a ‘one size fits all’ approach to modern medicine.
We need new drugs for unmet medical needs. These drugs need to be invented. But the process is complex.
What is drug discovery?
Tackling some of these issues, Yogeeta Manglani, Strategy Lead at top 10 global pharmaceutical company MSD, kicked off the event by focusing on the problems being faced in bringing drugs to market. Key amongst these is the fact that it can take up to 15 years from finding a suitable candidate molecule before a patient can benefit. The process is also astronomically expensive as anything up to $2bn is required. “It is cheaper to put a man on the moon than it is to develop new drugs,” lamented the erstwhile chief of GSK Jean-Pierre Garnier as far back as 2006.
Despite all this investment though, there is a very low probability of success. Only 1 in 5,000 medicines in development actually make it to market. For key diseases like cancer it will be at least a decade before new approaches emerge. But what about underfunded rare diseases?
“Essentially there is more demand than supply” said Yogeeta. The process of drug discovery is full of inefficiencies and needs to be expedited and costs reduced. The main battle ground is in the clinical trial process itself. This is where the testing of new treatments to determine how they work, how well they work and how safe they are happening, before we can launch them to the wider public.
What are the key issues in clinical trials?
Natalia Balko, Head of QuintileIMS Analytics Centre of Excellence, highlighted some depressing statistics. Patient enrolment is the biggest problem. Balko gave an example: “In a trial for a rare neurological disease, only 17% of the required number of patients were enrolled into the study”. On average 48% of trial sites don’t meet their patient targets required to even begin a trial.
As a result, up to 80% of trials are delayed. This is where the costs dramatically escalate, resulting in losses of between $600,000 – $8 million per day.
Unfortunately, patients are simply not taking part. Only 5% of actual cancer patients are involved in trials and even for those who do get involved 1 out of every 5 patients will drop out. Either patients don’t know about the trails taking place or they are not engaged in the process.
Clearly there is a need for change, for finding innovative ways to disrupt the current approach.
Disruption, disruption, disruption
Disruption has revolutionised other industries. There are technologies now that can have clear impacts in the clinical trials space. For example, social media or big data. There are many other underutilised examples: wearable technology can monitor patients remotely and unobtrusively. Virtual reality can help patients understand and overcome their fears - an informed patient is a more engaged patient.
We need to deconstruct and reconstruct the process. As Balko pointed out: “There is no silver bullet to disruption, but it has to come across the continuum, end to end, impacting all stakeholders - changing the business model, acting at scale and globally”.
Data takes centre stage
As part of the Data Science Festival, data and analytics took centre stage, not only when discussing the role it can play in helping to understand where potential participants for trials can be found, but also in enabling greater patient-centricity. There is gold in the data ‘exhaust fumes’ – after all, we are able to use all of an organisation’s data, even if it wasn’t collected for the purposes that we intend to use it. The previously mentioned neurological disease trial is an example that showed 80% of potential patients were actually in 4% of the hospitals. But the recruiters where looking in the wrong places. They would have improved their chances and reduced their effort by examining hospital data ahead of recruiting.
Yet even here, there are barriers to overcome. Just accessing the data is one problem, never mind privacy or the refinement that is required to make the data mineable.
“Getting the data out is the hardest challenge,” said Balko. “But then there is so much more we can do such as applying machine learning to analyse the big data sets and understand predictive modelling.”
The regulators are another problem. There is a need to convince them that new innovative practices can be used. This is not new for technology. Think about the music industry for example. Copyright laws came to exist once technology enabled artists to record their music and sell it. Copyright is a function of technology. The advance of streaming and sharing music means that to make money, musicians have to go and play live, which is what they did before the recording era.
The same can be true for trials. Being able to go to the public to recruit rather than being limited by gatekeepers such as doctors is infinitely more useful.
Companies disrupting clinical trials
Rebecca Morris-Chippendale, Business Development Manager of NorthWest EHealth, showed us the world’s first digitally enhanced Randomised Controlled Trial (RCT). You will rarely hear the words “Salford is the hub of innovation” but the Salford Lung Study showed this to be the case. Using the unique integrated health records in the Salford and the Greater Manchester area, researchers recruited and tracked the progress of patients in the real world to examine the safety and effectiveness for a new treatment for respiratory disease.
Some 2,800 patients were recruited and supported through 80 GP practices and 130 pharmacies. The linked database allowed monitoring of the patients in real time and with minimal disruption to their daily lives. Furthermore, the people of Salford were very proud that their community was part of this trial.
Getting patients to understand their illness is difficult. This is made infinitely more difficult in the case of rare diseases. Sarah Venugopal, patient advocacy and engagement specialist at RareMark, explained if you have a rare disease it’s nearly impossible to do your own research. Some 350 million people worldwide with rare diseases are frequently asked to consult Dr Google but as Venugopal pointed out, “How do you know about a clinical trial if you don’t know what your illness is?”
This is a problem that Venugopal is helping to solve. As a community lead manager, she is engaging with the rare disease community and patient groups to help them understand and be partners in the drug development process. Ultimately this is about building a patient powered research network and getting patients upskilled in participating in research and trials.
Bringing all of together requires access to all of this disparate knowledge. Magdalena Zwierzyna, Biomedical Data Scientist at Benevolent AI, showed us how her company is working to create a knowledge map for trials using AI to accelerate and enhance discovery of new treatments. “By essentially mining the data, we can learn from past trials before planning another one - mapping and analysing from genes to drugs”.
New technologies are going to change this space. Data is most important. But this event proved that shaking up the regulators, better engagement with patients, and tapping into the skills of some talented and committed women are going to be key factors in successfully disrupting clinical trials.