Streamlining clinical data management - part 1: Collecting data

Ahead of his session at Pharma IQ Live: Clinical Data Management Excellence, Fausto Artico, Global R&D Tech Head and Director of Innovation and Data Science at GSK, shares his insights on the challenges of handling clinical trial data. 

Pharma IQ: Clinical trial phases can easily require five to seven years of the standard 10 to 15 years required in bringing a new large-scale drug to market, because the effects of a new drug must be studied for at a minimum length of time. What are the challenges that make it difficult to reduce the time necessary to execute clinical trials?

Fausto Artico: Every single clinical trial phase is a very complex process, unique I would say. Furthermore, for each single large-scale drug, the final phase of its clinical trial involves hundreds or thousands of participants and not all patients start to take the drug at the same time. The participants are generally geographically distributed and each phase requires a considerable amount of coordination between all the parts involved (e.g. regulatory organizations, doctors, patients, data brokers, etc).

In addition, different countries have different regulations and many clinical trial activities require the creation of legal, ad hoc documents using procedures that are very difficult to automate or replicate across clinical trials.

Finally, today there is no agreed standard or IT infrastructure that can be used to greatly reduce the time necessary to gather data from patients and enable secure access to the data for scientists and regulatory organizations. The average lead time for just collecting data ranges from weeks to months, and worse, after gathering the data, it is necessary to clean, link, contextualize and harmonize it as well as enable secure access to it through procedures that can easily require additional weeks just to receive approval.

Pharma IQ: What is the time necessary today to make the data of just a single clinical trial phase available to scientists? Could focusing on optimizing a single clinical trial phase first allow one to accelerate the whole clinical trial pipeline?

FA: Independently of the phase, months can pass by from when the data for a patient is generated to when the data is finally available to scientists for data engineering processing – that is, to be prepared for data science activities and the report generation activities necessary to disclose results to regulatory organizations.

It is true that later phases require bigger samples (such as the number of patients), but all the phases are subject to the same kinds of problems. The long lead times of each phase are due in part to the great fragmentation present in the clinical trial market and to the fact that many required activities are executed by many different suppliers across the clinical trial supply chain, no matter which phase it is.

Pharma IQ: What does the typical clinical trial workflow look like for a large-scale drug from the data point of view?

FA: At the abstract level, no matter whether we are talking about large-scale drugs or drugs for rare diseases, a simplified and typical example of the workflow that must be executed to make data available to scientists and analysts is the following:

1. Doctors manually gather a limited amount of data related to the effects of the drug given to the patients under the direct supervision of the organizations who decided to participate in the trial.
2. Doctors manually input the results into digital systems made available by vendors.
3. The data is transferred from the input systems and aggregated through different tiers of data brokers.
4. The data is made accessible to the pharma company that designed and is executing the clinical trial with the doctors.
5. The pharma company imports the data into its internal systems for data processing.
6. Care needs to be taken about who can see the data and what they can see.
7. Data engineering activities are executed on the data.
8. Data is finally made available to data scientists and other analysts who need to prepare the reports to submit to regulatory organizations.

Pharma IQ: Is there a way to reduce the time doctors need to invest in these activities to help their patients?

FA: Doctors spend a lot of time just collecting the data and inputting it into digital systems. However, there now exist many portable (i.e., wearable or under-the-skin) devices that can greatly help them in their activities. These devices are becoming cost effective, are not invasive, take samples with much higher frequency than a doctor (e.g. every second or minute), scale easily with the number of patients compared to the number of necessary doctors and provide much more data variety compared to what a doctor could collect.

These devices have the potential to become doctors’ and pharma companies’ “best friends” because they reduce the need for tedious and error-prone activities. However, doctors will of course remain fundamental in helping monitor the trial and executing tests, which are as yet impossible with said devices.

• Read part two of this interview where Artico advises how to enable data sharing through automation and other streamlined procedures.

Clinical Data Management Excellence featured experts from Janssen, Pentavere, GSK and Rgenix Inc discussing how to enhance clinical trial outcomes through efficient integration, standardization and compliance of data. You can watch the event on-demand by visiting Clinical Data Management Excellence