Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
Applying real world evidence to clinical trials with Cristina Chang
6 Plays9 months ago
Applying real world evidence to clinical trials with Cristina Chang
In the final episode of season 1, Sylvain Berthelot welcomes Dr. Cristina Chang, Chief Medical Officer at HiRO, to unpack the growing importance of real-world evidence (RWE) in clinical development.
With decades of experience across global pharmaceutical companies, Cristina shares her deep insights into how real-world data is being applied to clinical trials to enhance patient outcomes, regulatory submissions, and post-marketing strategies. She discusses the practical considerations for using RWE, from data collection challenges to ensuring scientific rigor, as well as how the regulatory landscape is evolving to support these efforts.
Cristina also reflects on cultural and organizational changes needed to integrate RWE into clinical development successfully, and how emerging technologies and cross-functional collaboration can drive this transformation.
Join us as we explore how real-world evidence is shifting from a buzzword to a critical component of evidence generation in modern clinical research.
Recommended
Transcript
Introduction and Host Welcome
00:00:14
Speaker
Welcome to Clinical Data Talks, a podcast brought to you by CRS-Cube.
00:00:19
Speaker
I'm your host, Sylvain Berthelot.
00:00:21
Speaker
Join me and industry experts as we discuss the latest trends impacting the world of clinical data.
Understanding Real-World Evidence
00:00:30
Speaker
Our topic today, I'm sure you're aware of, even if you've been only a few months in the industry, as we're going to talk about, real-world evidence.
00:00:41
Speaker
But what does it
Expert Guest Introduction: Dr. Christina Chang
00:00:43
Speaker
really mean?
00:00:43
Speaker
I'm very pleased to have an expert on the topic with me, joining the podcast, Dr. Christina Chang.
00:00:52
Speaker
She's been in the industry for many years and has been working in many of the top pharma companies.
00:01:00
Speaker
She's now chief medical officer at HERO and she's joining us all the way from Taiwan.
00:01:08
Speaker
So I've been looking forward to this discussion because like many in the industry, I've heard talks about real world evidence, but I don't know much about the topic.
00:01:21
Speaker
So I'm sure you will enlighten me.
00:01:26
Speaker
But before we
00:01:29
Speaker
start learning about your experience, would you be able to tell us what, in your words, real world evidence means?
Defining Real-World Evidence
00:01:41
Speaker
Thank you for the question.
00:01:43
Speaker
So, Silvia, I think that real world evidence is the true usage of the data that is collected to provide
00:01:57
Speaker
benefit of the patient either, for example, to be part of a clinical trial as part of the design or do we use the clinical data that is collected, for example, from claims, right, or electronic health record and we will build the profile of this patient and we will
00:02:26
Speaker
of this registry we will learn a lot right about the patient about the disease and in those time uh these uh real world evidence will help us understand better the disease so in those time it helped us design better future studies so that was uh how we utilize them in those time and then the second type
00:02:55
Speaker
of usage were where we will use these data to support the post-marketing, right?
00:03:04
Speaker
Meaning that the drug will go through the clinical stage, will be approved, but we call it conditional approval, right?
Case Study: Pfizer and Male Breast Cancer
00:03:15
Speaker
So the regulatory agency will just give us the approval and that these real-world data will be the supportive data that
00:03:26
Speaker
we will collect from the product after the product is approved.
00:03:31
Speaker
So let's say that a product will be approved for a certain indication, but then after it's launched, we have a study design that is submitted and also agreed with the US FDA and we continue collecting the data.
00:03:51
Speaker
From day one that the product is on the market,
00:03:55
Speaker
until depending on how we agreed, right?
00:03:58
Speaker
For example, we collected data for one year or three years, and then we analyzed this data and we submit to the U.S. FDA and saying, okay, I had done this clinical trial and this is the real world evidence and is quite consistent with what we had found in the phase one or phase two, or even phase three in some time.
00:04:23
Speaker
So that was in the past.
00:04:25
Speaker
What we had seen evolving in the last few years is that we are evolving to utilize more of these data as part of the clinical trial.
Ensuring Data Validity: FDA Standards
00:04:39
Speaker
And also one of the very successful even first cases that came up in the market is the experience from Pfizer, where they already approved the product, iBrands for
00:04:54
Speaker
breast cancer, but for female, but they utilize the real world data to support the approval of the same product, but for male breast cancer.
00:05:09
Speaker
And why that was accepted is because male breast cancer is more rare, right?
00:05:19
Speaker
So the execution of this clinical trial
00:05:23
Speaker
in a way would be quite challenging, right?
00:05:26
Speaker
So the engagement, right, that the Pfizer team has with the FDA concluded with designing this collection of the LWE to support approval in that specific indication.
00:05:44
Speaker
Yeah.
00:05:45
Speaker
That's amazing that because
00:05:48
Speaker
without the real world evidence, it would have taken years or even maybe there wouldn't have been a justification to start a trial.
00:05:56
Speaker
So the fact that they could expand the use to mail is amazing.
00:06:03
Speaker
And I didn't realize that in the past, real world evidence was pretty much used in real world, but now it's being used in K-Equal trials.
00:06:14
Speaker
So yeah, thank you for sharing that.
00:06:16
Speaker
and also for regulatory submissions.
00:06:18
Speaker
Yes.
00:06:19
Speaker
Yeah.
00:06:20
Speaker
So talking about that, actually, how do you use real world data and what kind of data can you use as part of a clinical trial?
Regulatory Protocols and Data Points
00:06:34
Speaker
It's an excellent question.
00:06:36
Speaker
I think that the data should fulfill some
00:06:42
Speaker
criteria, some conditions, right?
00:06:45
Speaker
It's not any data, right?
00:06:47
Speaker
So first, actually, to start being able to use the data, they should have in place a solid and rigorous database in place, right?
00:07:03
Speaker
For example, one of the very famous database that is well recognized
00:07:10
Speaker
is the data come from the team from Flatiron in the US.
00:07:17
Speaker
So they had been able to build these very, very solid, right?
00:07:25
Speaker
Data around any kind of diseases.
00:07:29
Speaker
And the condition actually is that you need to discuss this design and the purpose of using this data
00:07:41
Speaker
with the FDA or any agency, right?
00:07:44
Speaker
Let's say for example, the EMA or the PME MDA in Japan, right?
00:07:52
Speaker
And then the agency should review the design, right?
00:07:56
Speaker
And agree with you, the company utilizing this RWE, right?
00:08:04
Speaker
And with that also, they will review
00:08:07
Speaker
the methodology that you are using, right, to avoid bias, right, and also ensure that the data doesn't have confounding, right, factor.
00:08:20
Speaker
So the sensitivity should be very well tested.
00:08:25
Speaker
And with that, then the regulatory agency is able to approve that and also
00:08:37
Speaker
any claim,
Planning and Data Collection Methods
00:08:38
Speaker
right?
00:08:38
Speaker
Because you do this RWE for a purpose.
00:08:43
Speaker
And the key purpose is that we want to use it to claim, right?
00:08:49
Speaker
Something, right?
00:08:50
Speaker
After we get the report from this study.
00:08:54
Speaker
So all this should be pre-specified in the protocol, right?
00:09:00
Speaker
It's not that you do the study and then you realize that, oh, this is quite interesting, right?
00:09:05
Speaker
And then you say, oh, I do this ad hoc analysis.
00:09:09
Speaker
And then you submit to the FDA and then you say, oh, I would like to utilize this data.
00:09:13
Speaker
No, you need to precipice if I, how you're going to use this data and in those conditions.
00:09:21
Speaker
So if I have to summarize is that the data should be qualified.
00:09:25
Speaker
You should have a base that is rigorously, uh, validated, right?
00:09:33
Speaker
Uh, controlled.
00:09:35
Speaker
And you should discuss with the FDA about the design.
00:09:41
Speaker
And also you should ensure that any of the endpoints are pre-specified in the problem.
00:09:48
Speaker
Yeah.
00:09:48
Speaker
Okay.
00:09:50
Speaker
And I assume that then... So you do all this analysis of the data to make sure it's good enough, but then having the disapproval from the regulatory agency
00:10:04
Speaker
then confirms that you can use it on your trial.
00:10:07
Speaker
Yes.
00:10:08
Speaker
Yeah.
00:10:09
Speaker
And do protocols have to include exactly which data points will be used from real world evidence?
00:10:19
Speaker
Yes.
00:10:20
Speaker
Wow.
00:10:20
Speaker
Okay.
00:10:21
Speaker
So it's very... But also you can... Yes.
00:10:24
Speaker
So, but that is the majority of it.
00:10:28
Speaker
But the good thing about science is that you can say there will be few things
00:10:34
Speaker
is that I will find during the execution of the clinical trial, right?
00:10:40
Speaker
But still you have to mention it.
00:10:41
Speaker
You have to say during this study, I'm going to collect some specific biomarkers, right?
00:10:50
Speaker
And I'm going to validate if these biomarkers are good enough, right?
00:10:56
Speaker
To validate and compare to a standard of care.
00:11:03
Speaker
either testing or treatment, right?
00:11:06
Speaker
And again, you have to press specify in the protocol that you want to do that.
00:11:11
Speaker
Yeah,
Real-World Evidence in Rare Diseases
00:11:12
Speaker
yeah.
00:11:12
Speaker
Okay.
00:11:14
Speaker
Wow.
00:11:14
Speaker
So a lot of planning upfront.
00:11:16
Speaker
It's not just, oh, we get this data and we can do whatever we want.
00:11:21
Speaker
You can collect those data, but again, you need to design a protocol and say, I'm planning to collect this data where, in which countries,
00:11:34
Speaker
which hospitals, right?
00:11:36
Speaker
Which specific targeted patient, right?
00:11:40
Speaker
And you design that protocol and then you say, I'm going to collect all this data and I wanted to discover what I had to have a question, right?
00:11:51
Speaker
And after collecting those, it teach you, right?
00:11:56
Speaker
It's a learning process, right?
00:11:58
Speaker
Where at the end, when you analyze the data, you have
00:12:03
Speaker
is a report, right?
00:12:05
Speaker
Where you will say, oh, that was my finding.
00:12:08
Speaker
And that again, help you design the next either study or the next AWE design.
00:12:18
Speaker
Okay.
00:12:20
Speaker
And in your experience, are there any specific clinical trials or programs that are more suitable for real world evidence use?
00:12:30
Speaker
Excellent question.
00:12:32
Speaker
I think that one of the example that really real world data can help a lot is rare diseases, right?
00:12:44
Speaker
Because as we know, let's say there are these diseases where the incidence is just 0. something, right?
00:12:55
Speaker
0.1.
00:12:56
Speaker
In the whole world, maybe you just have
00:13:00
Speaker
a few hundred or few thousand of those patients, right?
00:13:04
Speaker
In that occasion, rare diseases can really take advantage of AWE.
00:13:13
Speaker
Yeah.
00:13:15
Speaker
Is it because there's more data available, already available from patients than what you can collect in clinical trials?
00:13:26
Speaker
Yeah, because as I said,
00:13:30
Speaker
the patient are already right there and the researcher are trying or had been collecting already data from this patient for many, many years.
00:13:44
Speaker
It's just they didn't design a structure way to analyze the data.
00:13:52
Speaker
The data is already there in the hospital into every patient medical record.
00:13:59
Speaker
Right.
00:14:00
Speaker
So the good thing of RWE nowadays is that let's say, for example, the FDA or the EMA has very well established guidance where they define how you should think about utilizing this data.
00:14:24
Speaker
Right.
00:14:25
Speaker
So.
00:14:27
Speaker
companies, right, or people in the hospital as a researcher, they will design this protocol to say, okay, this data had been seated here, right, for so many years, that we needed to see how we can utilize them, right, for a good purpose, right.
00:14:47
Speaker
So actually, one of the definition of the FDA is saying, a WE should be fit for purpose,
00:14:57
Speaker
for a purpose, right?
00:14:58
Speaker
Yeah.
00:15:02
Speaker
So as a follow-up question on that topic, as a company interested in developing a treatment for a specific rare disease, would I be able to access real-world evidence to build my strategy for a trial, for a treatment?
00:15:26
Speaker
Yes.
00:15:27
Speaker
Or, but can I do that outside of a clinical protocol or do I have to have a clinical protocol first?
00:15:36
Speaker
No, you should have because the data will be seated there.
00:15:41
Speaker
But as I said earlier, the data might be incomplete.
00:15:46
Speaker
Yeah.
00:15:46
Speaker
Okay.
00:15:47
Speaker
Right.
00:15:48
Speaker
So you need to have a protocol to define very clearly and say, and tell the hospitals
00:15:56
Speaker
or the centers, right?
00:15:57
Speaker
Where that data is seated to say, I need you to collect this, right?
00:16:05
Speaker
You had to define the criteria that you want to collect from those patients because there are many criteria.
00:16:11
Speaker
And also actually you needed to do a pre-assessment actually of the data that is available in the hospital first to be able to design the protocol.
00:16:25
Speaker
Yeah.
00:16:25
Speaker
Right.
00:16:26
Speaker
Yes.
00:16:27
Speaker
So there is a pre preparation, right?
00:16:30
Speaker
That you need to do a check first on what is the data that is available in the hospital, right?
00:16:37
Speaker
That is one way.
00:16:38
Speaker
And the other way, as you said very clearly, very well before is that there are companies, right?
00:16:46
Speaker
That design actually the collection of these data already because they found this opportunity already, right?
00:16:55
Speaker
So many years ago.
00:16:56
Speaker
So the design already a very solid platform, right?
00:17:02
Speaker
In place to say, I'm collecting these data, right?
00:17:06
Speaker
And then these data is collected every day for the last, let's say five to 10 years, right?
00:17:13
Speaker
So the key about data is that you should do it in a structured way.
00:17:21
Speaker
Yeah.
00:17:23
Speaker
Okay.
00:17:25
Speaker
And can you then submit a protocol or submit for review in any country or are there specific countries that accept trials with real world evidence and others that don't?
00:17:44
Speaker
Yeah, I think that the country's regulatory platform has been evolving.
00:17:55
Speaker
a lot for the last few years.
00:17:57
Speaker
But I think that the agencies that had more experience, right, doing this already for the last few years had been the US FDA, the EMA, in some cases also Japan, Japan and BMDA also had been using the national database for some indication
00:18:23
Speaker
and growing, you know, for example, the KFDA, right, in Korea, and for example, China, the China NMPA also is trying to utilize that at the benefit of patients and also pharmaceutical companies.
00:18:44
Speaker
So it's quite widespread then, it's good.
00:18:46
Speaker
Yes, it's quite widespread, but I think that
00:18:51
Speaker
It is evolving, right?
00:18:53
Speaker
But I think that some agencies have more experience using it and some less, right?
00:19:01
Speaker
They're more trying to do it in a case by case.
00:19:06
Speaker
Yeah, yeah.
00:19:09
Speaker
Are there, in your experience, are there any trials for which you wouldn't use real-world evidence, where it wouldn't be suitable?
Challenges with First-in-Class Drugs
00:19:18
Speaker
Yes.
00:19:19
Speaker
So I think that for the first in class product, right, that has not yet proven, right, yet how specifically the drug is going to behave, right?
00:19:37
Speaker
Those are the cases that I think that wouldn't fit, right, to use, to utilize in a WWE.
00:19:47
Speaker
The order is also, uh, we are not quite, uh, able, right.
00:19:55
Speaker
To leverage there from one country to the other.
00:20:00
Speaker
What you can do is a hybrid, a combination of having local data from one country that plus, uh, an MRCT, right.
00:20:12
Speaker
Plus it, it, it, a, uh, a, um, a study.
00:20:15
Speaker
right?
00:20:16
Speaker
But you cannot, for example, collect a WE from one country and say, I want to leverage that to another country, right?
00:20:27
Speaker
What is the rationale behind that?
00:20:29
Speaker
Yes.
00:20:30
Speaker
It's very hard to do that because a WE mean that you are collecting the data from one specific country, right?
00:20:39
Speaker
How the drug had been utilized, right?
00:20:43
Speaker
How the patient had
00:20:45
Speaker
been treated for the last five years in one specific country in one specific indication, right?
00:20:54
Speaker
So mostly the, that data collected in that country will be quite different, right?
00:21:03
Speaker
From the other country because the practice would be different.
00:21:07
Speaker
But I'm not saying that 100% impossible, right?
00:21:11
Speaker
If you are very lucky,
00:21:12
Speaker
that after the analysis, you do see that, oh, the clinical practice in country A is very similar to country B. You can utilize that to be supportive, but you cannot say, I want to use the data to be leveraged or I want this data to be used as waiver.
00:21:35
Speaker
Okay.
Strategies for Regulatory Success
00:21:38
Speaker
Wow, that's very, very specific.
00:21:41
Speaker
I didn't realize that that's all that.
00:21:43
Speaker
Yeah.
00:21:44
Speaker
But I completely understand that, yeah, the practice of medicine varies from country to country.
00:21:51
Speaker
So it perfectly makes sense.
00:21:55
Speaker
Whoa, very, very interesting indeed.
00:21:59
Speaker
And I've learned so much from you today.
00:22:02
Speaker
Thank you.
00:22:03
Speaker
I've got one last question for you, Christina, if you don't mind.
00:22:07
Speaker
What's the best piece of advice that you've received that you apply in your daily work?
00:22:18
Speaker
I think that the best advice that I got first, even in my past experience, but also that I will share with the audience is that engage as early as possible, right?
00:22:33
Speaker
Engage with the
00:22:35
Speaker
regulatory agencies, right?
00:22:37
Speaker
About what is your plan, right?
00:22:40
Speaker
What do you plan to do with IWE, right?
00:22:45
Speaker
But also engage with other stakeholders, right?
00:22:49
Speaker
You remember that we say engage with companies, right?
00:22:53
Speaker
That are collecting this data, right?
00:22:55
Speaker
For sure, they had learned a lot through the years, right?
00:23:00
Speaker
So talk to people, right, that are working in IWE.
00:23:05
Speaker
share information as we do today, right?
00:23:08
Speaker
People will have best practices, right?
00:23:12
Speaker
They went through the journey, they learned things, right?
00:23:16
Speaker
The good and the bad, right?
00:23:19
Speaker
So one of the things, so in summary is that engage early, talk to people that has experience, ensure that when you are designing a clinical trial, try to set up
00:23:34
Speaker
the end points that are in compliance with the regulatory agency that you are submitted.
00:23:43
Speaker
You are planning to submit the data to, right?
00:23:47
Speaker
That is very, very important.
00:23:49
Speaker
Also, uh, ensure to bring, uh, as part of the team, people that understand, uh, how to make the data rigorous, right?
00:24:03
Speaker
are people that have experience in statistics, in methodology on how they're going to avoid the bias.
00:24:12
Speaker
Right.
00:24:14
Speaker
And lately also referred to the latest guideline, right?
00:24:20
Speaker
Because that guideline also are evolving, right?
00:24:24
Speaker
Are being updated, right?
00:24:27
Speaker
In different countries and in different agencies.
00:24:32
Speaker
So always refer to the latest, right?
00:24:35
Speaker
Guideline and even confirm, right?
00:24:37
Speaker
With people around you that have experience to say, oh, what piece of advice, right?
00:24:46
Speaker
Can you give me?
00:24:47
Speaker
Do you think that I should use the reference from the EMA or to the FDA, right?
00:24:54
Speaker
Because every case is different, right?
00:24:57
Speaker
And then lastly, I think that also,
00:25:01
Speaker
Talk to the agency about a more comprehensive plan, meaning that it's not a one-time, right?
00:25:10
Speaker
A one-time submission on, oh, this is just what I want to do today, right?
00:25:18
Speaker
Think on a plan that is more comprehensive and talk to the agency about what do you want to achieve through that data?
Collaborative Drug Development
00:25:28
Speaker
I think that with those advisors,
00:25:31
Speaker
I believe that the success will be quite high.
00:25:36
Speaker
Nice.
00:25:37
Speaker
I love how you're advising people to collaborate essentially, which is great.
00:25:44
Speaker
Cross-functional collaboration, collaboration with agencies is crucial for success early on.
00:25:51
Speaker
So yeah.
00:25:52
Speaker
Well, thank you very much, Christina.
00:25:54
Speaker
It's been an absolute pleasure talking to you today.
00:25:57
Speaker
And I'm sure everyone who's listening will be very thankful for you sharing your expertise and your experience.
00:26:05
Speaker
Thank you.
00:26:07
Speaker
No, it's really my pleasure.
00:26:08
Speaker
And thank you, Sylvain, for giving me this opportunity.
00:26:12
Speaker
I think that learning from the journey of drug development is really a pleasure and honor.
00:26:22
Speaker
So for me, it's really...
00:26:25
Speaker
It's really a pleasure for me to really share with the audience.
00:26:29
Speaker
Well, I appreciate that.
00:26:31
Speaker
And thanks everyone for listening to us.
00:26:33
Speaker
You can find more episodes of Clinical Data Talks on our