Over the last five years, a confluence of trends have combined to reshape the reality of clinical research. COVID-19 pushed decentralised trials into the mainstream. Global conversations about clinical trial diversity and representation came to a head. And the increasing availability of AI and large, historic data sets has opened up the door for synthetic control arms and digital twins.
As chief medical officer and head of biopharma development solutions at UCB, Dr Iris Loew-Friedrich has been in the thick of these seismic changes since 2008, leading the Belgium-based global pharma company into its own clinical trial future. pharmaphorum sat down with Loew-Friedrich last year to get her thoughts on the ongoing evolution of the clinical trial. The interview has been edited for length.
Dr Iris Loew-Friedrich: I’m a physician by training. I spent the first seven years of my professional career practising medicine in academic hospitals and then moved to industry, and ever since have been working in drug development. I think it has become my passion and my way of serving patients as a physician. I’m very passionate about providing innovative and differentiated medicines to patients.
My role at UCB is twofold. I’m the chief medical officer of the company and I lead an organisation called Development Solutions. We are covering the key drug development aspects, including all of clinical development, regulatory affairs, quality, patient safety and pharmacovigilance, real world evidence, statistics, and also the data office for the company.
At UCB, we had started to work on decentralised clinical trials already in 2016. We have evolved into our preference towards a hybrid model – having brick and mortar sites and adding digital elements into the clinical studies, like telemedicine visits, remote monitoring, and other elements. And I think that’s also what I’m seeing across the industry.
I think COVID has given decentralised clinical trials a big boost because it was the only way to keep trials going without major disruption, without major loss of data, without taking undue impact on patients. It pushed the industry to move in that direction and I think it’s become a lasting change. There are different degrees of adoption, different levels of passion around the topic, but the whole industry is moving towards digitalisation of the clinical studies.
For me, there are two watchpoints. First, you have to rely heavily on technology, and technology needs to work, right? And you have to ensure that, for patients, there’s always stable internet access. You have to have technology that is easy to use, that’s intuitive. The more specific training you have to offer, the more complex and the more difficult it’s getting for patients. So, simplicity and ease of use and intuitive use is kind of the theme of the day.
The other watchpoint is that we are all human beings, and thrive on individual interactions. Trust is an important topic in the patient-physician relationship, and we will have to make sure, dependent on individual patient needs, that there is enough face time with the physician, the study coordinators, and nurses at the clinical sites. It’s the balance that we need to strike between technology and human interaction.
Digitalisation, for me, has three dimensions. There’s the technology, there’s data, and then it’s also the change in culture and change in mindset that’s required. At UCB, we are using vast amounts of data from very diverse sources, and we have quite sophisticated advanced analytics and artificial intelligence to look for patterns in data.
We use them in the beginning to get a better understanding of patient populations because we truly want to understand the unmet need and how the unmet need connects to the disease biology. Then we are using data to understand where patients are living. Where do we see clusters of patients? Where should we put our clinical sites? And, of course, we are using data to model clinical studies.
What is very real and what we are using quite regularly, particularly in the rare disease space, is synthetic or historical control arms. When you have only very few patients with very severe diseases, you cannot do a randomised controlled trial. You have to find a historical control and that’s very often structured real-world data from the literature or from registries. I should also say we are running natural course of the disease studies by using available real-world data.
Yeah, we are using wearables – or I should maybe say digital health technology – for a number of topics. Of course we are collecting data, blood pressure, and heart frequency – standard things. If we have a patient population that’s not able to use digital devices, we will provide a paper version, so we really try to tailor it to patient needs. But the standard for this data collection is now through apps and devices.
Of course, the big advantage of the clinical outcomes assessments via technology is that what was previously done at a visit and then at the next visit, maybe four weeks later, can now be collected on a continuous basis. So, you have many, many more data points. And the more data points you have, the more you limit your variability. This way of collecting data also helps us to reduce the sample size in certain clinical studies.
The challenge is, in my view, twofold. First of all, as data integrity is of paramount importance, these devices need to be of medical-grade quality and really need to be acceptable by regulators for the purpose we are using them for. Regulators are not always aligned around the globe in their requirements.
The clinical studies that we are doing are typically global studies. So, we have a lot of different regulations to observe when we choose our devices, or we have to choose different devices, or we have patients who want to use their own device, and then we have to find ways to accommodate that. So, there’s much more tailoring to individual needs, which comes with additional effort and additional complexity.
I think the other topic that is important is how many apps, how many devices can you really impose in a clinical study on a patient or on a clinical site? And the more you can measure digitally and the less integrated the individual tools are, the more you add complexity by adding different tools, right?
For example, when you run a Parkinson’s study, and you have an accelerometer, that’s a device in itself. You might still have another device for your patient reported outcomes, and another for heart rate. So, you have to be very mindful of how much you want to impose in terms of complexity on the patient.
But that’s the beauty of it, right? In the past, clinical studies have been kind of a highly artificial environment to show that a medicine is efficacious and safe. And there’s always been criticism that clinical studies do not reflect the real world.
There’s literature that illustrates that, with clinical studies, we only reach a very small fraction of the potentially eligible patient population. And why is that the case? It’s the case because patients might live far away from the centres where we have run our clinical studies. They might live in rural areas. They might not be aware of clinical studies, what they mean, and how they could contribute. The digitalisation of clinical studies allows us to reach those patients in a different way than we could do it before.
That, of course, then also benefits the studies, which become much more reflective of the patient’s reality, of the patient population, and of the way patients live.
It’s a mix of both. We are really thinking about it because it’s so obvious and so much kind of on top of what we need to do. The FDA is demanding diversity plans now, so you cannot avoid thinking about it. We pay particular attention to bringing our medicines to paediatric populations. We have run some of the most extensive paediatric programmes down to neonates. So, really covering the whole age spectrum, we are aware that age is a topic – older age. We live in an ageing society, so you need to understand how our medicines work in the elderly. We are very much invested into women of childbearing age. We’re doing a lot of work in that space. We’re thinking about how we reflect the socioeconomic background of the diversity of our population. It’s a very broad spectrum that we are looking at and that we are mindful of.
And again, digitalisation of the clinical studies provides us the opportunity to go there. When you go outside the world of academic research centres, outside the world of clinical practices who do a lot of clinical studies, you have to start by building trust that the research that we do is ethical, ensuring that it’s understood, why it needs to be done, why it can’t be done differently. Trust-building is very important, and that, of course, is based on education, information transparency, and reliability, so there’s a lot of groundwork to do before you can really broaden your reach.
For example, one of the discussions that is currently very near and dear to our heart is how we involve the community physician, the family doctor, in our clinical trials? Does it always have to be the 200-mile ride to the clinical centre, or do we find ways to engage the family physician next door? It’s easier said than done, but it’s a topic that we are discussing because we believe it’s important and it’s another kind of real-world aspect.
First of all, I believe that, based on the data that we have and with advanced analytics and artificial intelligence, we will be able to modelise clinical trials in a different way than we have done in the past. And if you can modelise trials, then you can improve the design and ensure that you only do the trials that are really helping you advance your medicine or advance the knowledge around medical science.
If you look at the success rates of clinical trials these days, there’s a lot of room for improvement and I think data and advanced analytics will help us improve there. If you can run your trial in silico, that will help a lot. So fewer clinical trials with better design and higher probability of success. The other topic is really making sure that more patients can participate in clinical research.
I’m a big believer that patients own their data. Patients should have their electronic medical records and their clinical trial data and it should all be connected so that they can very seamlessly go from everyday care into a clinical study back to everyday care, and the data collection continue throughout. For me, that’s the big opportunity, that the frontiers between everyday care and clinical trials blur even more, and that it becomes a fully integrated set of data in the end, which will allow us, in a much better way than we can do it today, to monitor long term effectiveness and monitor long term safety.