Lessons learned delivering virtual trials during Covid-19

Digital health expert Dr Jordan Brayanov from Takeda Pharmaceuticals shares his lessons learned delivering virtual trials during Covid-19 and predictions on what’s next for the industry.

Covid-19 has accelerated the adoption of decentralized trials

Covid-19 has caused the pharma industry to look for new ways of running clinical trials. In a conventional trial, patients are expected to go to a specific hospital or clinic (usually a large academic center), often multiple times during a trial. Decentralized trials break the mold by adopting new ways to engage with patients, in particular:

Telemedicine
Move from large academic centers to local hospitals and clinics
Collect data at home using digital devices

What are the key challenges with decentralized trials?

Lack of trained users

In the hospital, a nurse or a doctor are familiar with a specific measurement instrument and can confirm that a measurement is done correctly. If there is a problem with the equipment they are likely to notice, for example a BP cuff leaking air or an ECG lead has fallen off.

This is hard to do at home without an expert user. Patients and caregivers record measurements without the ability to scrutinize their validity. Thus, when we get poor quality data, we often don’t know if this was because the device didn’t work or it was not used correctly.

Access to devices

Devices can be expensive and not everyone can afford them. Furthermore they often require internet access to collect or transmit data, which is either not available or prohibitively expensive in many places around the world.

Lack of scientific rigor behind consumer devices

Consumer devices often make claims about what they measure without any data provided to the public to support these assertions.

Furthermore, there is no common standard or requirements around many of these measurements. Without such standards, it is hard to evaluate the performance of a particular device and we often struggle to compare it to existing clinical measurements or other similar devices.

Studies span continents and last years

In clinical trials, compatibility and comparability are very important. A device used in Europe, the US, and China needs to be able to produce the same measurement in each location. However, a company may decide to make one device for the US and a different device for China. We need to ensure the two devices are indeed equivalent before we can use them. The same goes for various product generations. An improved product must provide the same data as the old product, otherwise we may not be able to use them in a study.

Clinical devices require data provenance

Any measurement that is used to support the development of a medicine needs to be “traceable” back to the “raw data” used to generate it. Highly specific information, like the site where the measurement was made (patient’s chest, finger, wrist), what the patient was doing at this time (sitting, sleeping, walking), and how long did this measurement take (instantaneous reading or average over time) may need to be recorded and provided with a regulatory filing.

Clinical outcomes or end-points are strictly defined

Regulatory agencies often specify exactly what they need to see for a specific measurement. For example a BP measurement may be defined as “The average of three measurements, collected from the upper arm, over 15 minutes with a patient sitting quietly on a chair” and convincing them to accept a different measurement could be a challenge.

So how can we solve these challenges?

We can’t do this alone, in fact no one can, so as a pharmaceutical company we have started working with specialized vendors like Cambridge Cognition who can establish the use cases of devices and help us solve some of the operational challenges, the data access, the training, figure out the clinical methods and work on the measurement standards and the requirements for specific diseases and populations.

Collaboration is key. Pharma has to support the efforts in developing new technologies, devices, and measurements as part of patient care. We can’t sit back and wait for this to happen on its own.

We also have to encourage companies maintaining existing infrastructures, such as large CROs, to open those up to new opportunities and allow new data to flow through existing data pipelines. This will substantially reduce the overhead, especially for small companies, in bringing new tech to market.