Dr Nicholas Robert Explains the Benefits, Challenges of Using RWE for Measuring Clinical End Points

Nicholas Robert, MD, medical director of Data, Evidence & Insights, McKesson Life Sciences, discusses the benefits of real-world evidence (RWE) for measuring end points, as well as the challenges posed by variability in clinical practice.

Transcript

How can real-world data be beneficial for measuring endpoints, particularly in rare tumor types?

With the advent of real-world data, which has really been the result of having electronic medical records, we have the ability to identify patients that have been treated in standard ways. So, if we have the opportunity to develop a new treatment, especially if the cancer’s a rare cancer, it’s hard to do the standard randomized clinical trial. So, what we can do is, that new agent can be tested in a group of new patients and we can compare those results with a similar group treated with standard therapy and in this way, we can advance new treatment faster. We recently did this in a very rare tumor called Merkel cell tumor, and the company developed a new agent and identified certain clinical outcomes and we were able to look at those same clinical outcomes with patients receiving standard therapy, and the evidence revealed this was an advance and the drug was subsequently approved.

What challenges are posed by variability in clinical practice that might influence response measurements?

That’s a really key question. Response is really a challenge. In a standard clinical trial, response is defined, there are metrics for that and are followed to see if the drug is really improving a response. When you go to real-world data, the same criteria is not used. Physicians, when they see patients that are being treated for cancer and the patients are doing better, they’re being described as such, and often they’ll include in their assessment results from scans, but not necessarily always. So, we actually recently look at this, we had a poster at ISPOR this year where we looked at the criteria and one of the things we did is we sat down as a group, these so-called playbook rules we developed, where we defined certain terms like response. But, physician assess response. So, I think when you use those terms, you have to be very careful and make sure that people understand you’re not using the same criteria that you would use in a clinical trial.

That being said, we have been able to use the metrics from a clinical trial retrospectively, and that has challenges and has limitations, but there is a way to obtain at least some CT [computerized tomography]-based measurements that you would see in a typical randomized trial. So, I think the message here is to be very transparent in the limitations when you look for evidence of response.