Evolution of clinical trial data calls for evolved response: ICON

By Jenni Spinner

- Last updated on GMT

(bestbrk/iStock via Getty Images Plus)
(bestbrk/iStock via Getty Images Plus)

Related tags Icon Data management CRO Contract research organization Clinical research

In the first of a two-part series, leaders from the CRO discuss how study data has progressed in recent years and how trial teams can keep up effectively.

The clinical trial landscape today is vastly different than what it looked like just a few years ago. One key to keeping on top of the rapidly shifting field is proper data management.

To get an informed view on the topic, Outsourcing-Pharma recently connected with two leaders from ICON, a global contract research organization (CRO):

  • Emily Mitchell, executive director of decentralized trials
  • Kathleen Mandziuk, vice president of project management and real-world solutions

OSP: Could you please share your perspective on how data collection in the clinical trial arena has evolved?

EM: Clinical trial data has started to move closer to the patient and a little bit further from the site being the primary point of collection of that data. The volume of data has also grown partially because to get patient data to be a little more meaningful, you must implement it in a different approach versus the clinician assessment.

Now data includes potentially having a patient response and then correlating it to a sensor or some other modality of assessment to validate that data, which means there is less bias that can be read into it. We are now collecting data from wearables, e-consent, and e-pros, but also the patients’ electronic medical records themselves, rather than sites being the ones that input data from chart to EDC.

Patients can now use secure technology that grants access to their data which allows for collection from a pool of data versus transcription of data. While this is overall a much richer dataset, there is also more noise in the data due to the overall volume. We’ll need a mindset shift to be able to make that data have the same meaning, as we are now working with 20 or 30 data points to connect, where before you had one.

KM: I started my career almost 25 years ago in data management, which were the early days of EDC and studies were still mostly paper, and over time it's really evolved. We are now focused on how we adapt data collection for different types of study designs, which is crucial in real-world research. The way we collect data for a Phase II/III study might look very different for a Phase IV and observational study.

Also, over the past several years we’ve seen greater democratizing of data that exists already, such as claims data. We also now have access to EMR data, genomic data, and the ability to utilize that data in a privacy-compliant way as well as the ability to link that data. And that is where it's really taking evidence generation in drug development to a whole new level.

We’re also seeing an evolution from a regulator standpoint with the 21st Century Cures Act, to support the use of real-world data within submissions. Leveraging existing real-world data can complement the clinical trial data that we're collecting and provide even more insights into the safety and effectiveness of drugs in a real-world setting.

OSP: What about the way the use and application of clinical trial data have evolved?

Emily Mitchell, executive director of decentralized trials, ICON

EM: The pandemic has really set the table for the evolution of being able to use patient or primary source data as an endpoint versus clinician-sourced data as the endpoint. We see more and more movement towards digital endpoints and away from more traditional endpoints like the 6-minute walk test which has been the gold standard. In this example, we can now look at moderate to vigorous activity on a patient and be able to show change of baseline and have it demonstrate the theme of fact.

Traditional clinical trials have very tight and regimented controls, which means we are not really seeing how the drug interacts in a normal life setting. With wearables and health sensors that allow us to monitor the patient remotely outside of the actual clinic, we get a better real-world perspective of how the drug is functioning and how it's impacting the patient’s life.

With regard to applying all this new data, we need to have better analytics with a more powerful engine behind them so that we can tell a better story with the data. In the past, it was relatively easy to sort the data and find an outlier.

Now, with the large volume of data, it’s not as readily apparent. There is a high volume of things that are potential outliers, so you need powerful tools that enable better trend analysis and identification of potential safety risks or areas that there needs to be re-training or focus for the site, the patient, or the investigator.

KM: There are some fast adopters in the industry of new innovations, and it comes down to how we can make research more user-friendly for the end participants. Because we can collect data directly from patients in their homes, it makes it less burdensome for people to participate in research. Many organizations are realizing that this is going to be the new norm of making evidence generation a lot more efficient and effective.

I agree that the pandemic has really shifted the perspective of how we collect data for clinical research. Patients were not able to go into doctors’ offices, but we still had to ensure that patients were being safe and compliant. For an industry that can be reticent to change, it has quickly had to catch up due to this global health crisis necessitating a major change.

At ICON, we recognized the need some years back for more patient-oriented clinical trials, which included the use of technologies to collect data, remote monitoring, and digital platforms. We began to put together hybrid and decentralized trials prior to the pandemic, so once it hit and everyone needed to adapt their trials, we had experience to help sponsors through something that's totally different than anything they've ever done before. The investments that we made really allowed us to provide a level of support through a seismic shift in the industry.

OSP: Please talk about how ICON has worked to help sponsors make better use of the data collected.

EM: ICON has been developing and utilizing powerful analytic tools that allow us to get the most insight from the data collected. We have the experience and the knowledge to be able to look at that large volume of data that is coming in and pull it together in a summary manner that tells sponsors a story on a per patient, per study, and per product basis. This allows sponsors to evaluate data not only in an individual patient or study but over the entire lifespan of the drug.

Kathleen Mandziuk, VP t of project management and real-world solutions, ICON

KM: We have also developed new solutions for drug development companies, such as clinical trial tokenization, utilizing real-world data from ICON’s claims dataset. This gives drug developers the ability, within a compliant privacy framework, to be able to follow groups of clinical trial participants within our secondary data assets which include prescription, medical, and lab claims to enhance the evidence that is generated on patients after they've been exposed to a product. This enables sponsors to follow a cohort after a study and even streamline some of those long-term follow-up programs where there is a requirement to collect data on patients for up to 10 years. This reduces the burden on the sites to collect that data and to get valuable insights from that information which can be particularly challenging with cellular therapies, for example.

Beyond required follow-ups, we're also finding that many sponsors are looking for those types of insights because it really helps us understand the long-term safety and effectiveness profiles that we were never able to really gain before.

OSP: How have demands placed upon sponsors by patients, providers, and other stakeholders impacted the demands they placed on their research partners?

EM: The movement towards primary data, or data collected directly from the patient, requires additional review and knowledge and a data science mindset. A traditional data management mindset, developed based on paper data collection and monitoring and source verifying, looks for everything to match one to one or conversely to look for things that do not match to indicate something is wrong.

The data science mindset required today needs to look for trends and outliers to determine what is the signal or trigger we are looking for and what is that signal detecting? How is the data showing us where there might be reason to pause or reason for intervention from our perspective? There are different triggers from the data collected that we can then alert the sponsors to for management of quality concerns.

We also have seen sponsors looking for new data visualization tools that will allow them to do some ad hoc and aggregate reporting of data and allow them to see in real-time exactly what is going on within the clinical trial. Real-time is the key – sponsors want to know quickly where things stand from a site management perspective, from a clinical data gathering perspective, and from a patient safety perspective.

KM: It's interesting because we’ve seen the demand for data sort of come full circle. Sponsors always wanted to collect more and more data within the studies and as a research partner, we really pushed back and advised them to collect just what was needed. Often, the more data requested maybe would not yield any further insight, but instead was just an additional burden that was put on the patients or the sites or the caregivers of collecting that extra data, which could impact participation or completion of research. So, we pushed to streamline data collection in the traditional study structure.

But by leveraging real-world solutions and real-world evidence we have more data than we did before, so they're getting some of their data back without having to solicit it in the protocol which streamlines the on-site visits. We are at a place now where we have more efficient means of data collection, and in many ways, the data we collect is more robust and we can provide higher quality insights in a less intrusive or burdensome way.

We are also seeing more patient-reported outcomes taken into consideration when a drug is approved. Along with the clinical data, the inclusion of the patients’ experience and voice as part of the data collection process is facilitated by the wearable or sensor technologies that enable the almost passive collection of that data and transmit it digitally directly from the patient.  The value of that data is much higher now in drug development because we are seeing the impact on the end-users of these treatments in near real-time.

In the second of this two-part series (appearing Monday, February 14), Mandziuk and Mitchell will touch upon issues involving site talent, technology tools, and working with partners effectively.

Related news

Show more

Related products

show more

Using Define-XML to build more efficient studies

Using Define-XML to build more efficient studies

Content provided by Formedix | 14-Nov-2023 | White Paper

It is commonly thought that Define-XML is simply a dataset descriptor: a way to document what datasets look like, including the names and labels of datasets...

Overcoming rapid growth challenges with process liquid preparation

Overcoming rapid growth challenges with process liquid preparation

Content provided by Thermo Fisher Scientific - Process Liquid Preparation Services | 01-Nov-2023 | Case Study

A growing contract development manufacturing organization (CDMO) was challenged with the need to quickly expand their process liquid and buffer preparation...

Related suppliers