Introduction to Clinical Data Talks
00:00:14
Speaker
Welcome to Clinical Data Talks, a podcast brought to you by CRS-Cube.
00:00:18
Speaker
I'm your host, Sylvain Bertelot.
00:00:21
Speaker
Join me and industry experts as we discuss the latest trends impacting the world of clinical data.
Passion for Patient Involvement in Clinical Trials
00:00:31
Speaker
So today, my guest and I share the same passion about increasing patients' involvement in the definition of clinical trials.
00:00:44
Speaker
He dedicated many years in increasing the efficiency of the clinical supply chain, and he recently joined or created even the company One2Treat, where he is the CEO, driving a change in how we analyze the effect of treatments in clinical trials.
00:01:06
Speaker
I'm very pleased to introduce Sébastien Koep.
00:01:10
Speaker
Hi Sébastien, how are you doing?
00:01:12
Speaker
Hello Sylvain, I'm great.
00:01:14
Speaker
How are you doing?
00:01:16
Speaker
I am doing great, thank you.
Limitations of Traditional Clinical Endpoints
00:01:18
Speaker
So today we're going to talk about something that I know you're very passionate about, which is how to consider clinical endpoints differently, which should lead to more meaningful clinical trial outcomes.
00:01:35
Speaker
And I want to start with the problem essentially that you're trying to solve.
00:01:41
Speaker
So are traditional clinical endpoints limiting
00:01:46
Speaker
um how are they limiting how we assess treatment benefits at the moment so as you know sylvain the vast majority of chemical trials that are running today only focuses on one single outcome so one single what we call primary endpoint that will
00:02:08
Speaker
tell whether the clinical trial is a success or not at the end of the clinical trial in a pre-specified way.
00:02:14
Speaker
That's the goal of the primary endpoint.
00:02:16
Speaker
Define whether this drug ultimately deserve to get to market or not.
00:02:22
Speaker
And if you step back right now and you try to ask patients around you, like, does that make sense?
Patient-Centric Clinical Trials
00:02:29
Speaker
Most of the patients may be surprised that you only or mainly rely on one single criteria
00:02:36
Speaker
to tell whether a new treatment deserves to get to market, whether a new treatment is good for them or not.
00:02:43
Speaker
So they will probably tell you that there are a lot of other outcomes, a lot of different facets of their own individual stories that matter to them that, I mean, they would expect a new treatment to help them or at least, you know, get data access and so on.
00:02:57
Speaker
So I do believe that's where we may help, I mean, answer those kind of naive
00:03:04
Speaker
needs from patients is more about we do collect a huge amount of data points in clinical trials and how can we make sure that the primary endpoint, so again, I mean, this kind of pre-specified question that you need to ask before the trial starts and that at the end of the day, you will analyze the clinical trial success based on this primary endpoint.
00:03:24
Speaker
How do we make sure that the primary endpoint better reflect what matters most to patients?
00:03:31
Speaker
Yeah, and that's a very interesting topic.
One2Treat's Innovative Approach
00:03:34
Speaker
As you know, I have a patient advocacy podcast, and that's something that patients who have had a bit of involvement with the clinical industry in general, when they realize that thinking about Parkinson's, for example, there's only a very limited amount of endpoints that are
00:03:59
Speaker
deemed significant from a clinical point of view, they're quite shocked actually.
00:04:06
Speaker
And in some ways it limits the, it can limit the trials that are started if we focus just on one end point, but I'm sure we're gonna get to the benefits.
00:04:20
Speaker
So how are you redefining then the way clinical end points are designed at One2Treat?
Introduction to Generalized Pairwise Comparison (GPC)
00:04:28
Speaker
So that's basically the idea of OneTreat founder, Mark Beuys.
00:04:32
Speaker
You may know Mark Beuys because he also funded the company's IDDI, but also ClouPonts based on a slightly different kind of reliable biostatistical technique that opened the door to this kind of era of risk-based quality management.
00:04:52
Speaker
So that's basically what Mark Beuys has been doing for the last 30 years is having
00:04:57
Speaker
I would say creative ideas from a biostatistical methodological standpoint.
00:05:03
Speaker
And then later on, I mean, find some kind of companies that can also develop some softwares that will facilitate the engagement from the pharma industry, leveraging those kind of innovative statistical methodologies.
00:05:19
Speaker
What he created 15 years ago is this idea of generalized pairwise comparison, so GPC.
00:05:25
Speaker
The idea is to incorporate multiple clinically relevant outcomes within a single assessment.
00:05:30
Speaker
And this single assessment may become the primary endpoint.
00:05:33
Speaker
And what you will try to measure is what we call the net treatment benefit.
00:05:37
Speaker
That's the most important aspect
00:05:39
Speaker
for non statisticians, I would say to remind out of this podcast is the net treatment benefit is basically the global picture of what you want to assess when you speak about treatment effects.
00:05:51
Speaker
And basically you may ask people, I mean, to prioritize the outcomes of interest.
00:05:57
Speaker
Like what's the most important outcome of interest for your disease?
00:06:01
Speaker
Imagine for, I mean, if I'm suffering from cancer, maybe
00:06:05
Speaker
overall survival is the most important outcome, but not at any cost.
00:06:09
Speaker
What does it mean?
00:06:10
Speaker
It means that if I have to choose between two treatments and they are more or less similar in terms of expected overall survival, maybe I want to look on a second outcome.
00:06:19
Speaker
That's where I mean comes the second outcome in my list of prioritized outcomes.
00:06:23
Speaker
is if I do not benefit at least by six months on average, six months benefit in overall survival, then I do not care too much about overall survival in my choice between choosing between two treatments.
00:06:35
Speaker
Then I would focus on, I don't know, another criteria that may be my quality of life.
00:06:40
Speaker
And again, imagine that two treatments would offer me similar quality of life.
00:06:44
Speaker
I would focus on the third.
00:06:46
Speaker
on the third outcome of interest that may be some kind of toxicities or, you know, grade three or grade four adverse events or something like that.
00:06:56
Speaker
So again, you may have different clinically relevant patient focused outcomes that one may need to prioritize and this prioritized outcome will be the
00:07:09
Speaker
the primary endpoint.
00:07:10
Speaker
And at the end of the day, you will assess what we call the net treatment benefits based on those kind of list of priority type outcomes.
00:07:16
Speaker
So that's very simple idea that Mark basically suggested to the scientific community 15 years ago.
Adoption and Benefits of GPC in Pharma
00:07:23
Speaker
So the first scientific paper
00:07:24
Speaker
peer-reviewed paper was published in 2010 and in the meantime there have been i don't know probably 60 70 scientific papers published by a lot of different labs a bit everywhere in the world and more and more pharma companies and regulators have been starting to leverage the gpc and net treatment benefits methodology
00:07:44
Speaker
But still, it's a bit slow.
00:07:45
Speaker
So I mean, to give you an idea, there are just a few drugs that have been approved so far by regulators, FDA, EMA, getting to market, especially in cardiovascular disease.
00:07:57
Speaker
And so that's really the idea of Mark Boyce.
00:07:59
Speaker
It's creating one to treat
00:08:01
Speaker
company two years ago in order to really speed up this i would say implementation of the net treatment benefit for all clinical trial designs or most of them at least those that are relevant meaning that one single outcome one single endpoint is maybe not sufficient to
00:08:20
Speaker
better reflect, correctly reflect what matters to patients.
00:08:23
Speaker
So that's really the idea is make sure that we package this kind of net treatment benefit approach in easy to be used software solutions so that we can help pharma sponsors and regulators to run those kind of innovative statistical methodologies in a much more easy way, but also transparent way because we believe, and that's also the point of Mark Beuys, is we believe that
00:08:48
Speaker
Transparency is something that is really important for regulators, for Vama sponsors, but also for patients.
00:08:55
Speaker
And transparency is also part of understanding why a drug was getting to market, what are the positive effects, what are the effects of one treatment versus the other.
00:09:07
Speaker
And so all this ability to compare in a transparent way,
00:09:11
Speaker
We believe it's really, really important and that's the main outcome from all the softwares from One2Treat, leveraging this net treatment benefits approach.
00:09:21
Speaker
Yeah, so you've touched on a lot of different points here that personally I find fascinating.
00:09:27
Speaker
So that's why I was very excited about having this discussion with
Redefining Primary Endpoints with Patient Input
00:09:32
Speaker
First one, I love the fact that you bring statistics to something very, very useful for clinical trials and for patients.
00:09:44
Speaker
But then you've mentioned patients multiple times and you've hinted to the fact that patients essentially are in a way involved in the definition of what is beneficial for them.
00:10:00
Speaker
And maybe you've already answered that question, but why do you think this new way of defining primary endpoints is more patient focused from your point of view?
00:10:19
Speaker
For many different indications,
00:10:22
Speaker
Again, one single outcome may not always reflect easily what matters most to patients.
00:10:27
Speaker
So let me take some simple example in rare disease where there've been treatments, new treatments that have been commercialized based on a biomarker for instance.
00:10:40
Speaker
Is a biomarker really something that
00:10:43
Speaker
that is understood by patients, that is really bringing them a lot of added value of trust in the new treatment.
00:10:50
Speaker
So you can wonder this.
00:10:51
Speaker
And again, I'm not challenging whether or not those treatments are bringing a strong added value to patients.
00:10:58
Speaker
That's not my point.
00:10:59
Speaker
My point is maybe there are different ways to also
00:11:04
Speaker
incorporates multiple outcomes that are more patient focused and maybe something more meaningful to patients not only to patients maybe also to clinicians that know what is good or not i mean for patients and think about oncology i mean you can ask the feedback from clinicians from key opinion leaders from site investigators for instance
00:11:26
Speaker
about what would be the most important, the most clinically relevant outcomes for your patients.
00:11:32
Speaker
So again, when we speak about being patient focused, it's not only just speaking to patients, it's also, I would say, giving more importance to clinicians that are working really closely to patients and asking them to,
00:11:48
Speaker
to define this kind of multi-dimensional way to describe what is good or not for patient.
00:11:55
Speaker
We do believe that's the most important way.
00:11:57
Speaker
We feel really, really excited about this opportunity to just engage a lot of people and at the end of the day, yeah, define better, more efficient primary input.
00:12:10
Speaker
Yeah, and you're right.
00:12:11
Speaker
That's something that, like in rare disease, it's very difficult.
Defining Relevance of Multiple Outcomes
00:12:19
Speaker
And often patients know more than clinicians themselves what is significant to them.
00:12:27
Speaker
And symptoms can vary so much between patients as well.
00:12:30
Speaker
It's quite difficult to find something, one primary endpoint that will have a significance for everyone.
00:12:38
Speaker
So you mentioned engaging with sites, engaging patients in the definition of what is this composite primary endpoint.
00:12:50
Speaker
But one step further, how do you define the relevance then of all those criteria that compose the endpoint?
00:13:03
Speaker
So typically what we would suggest is not being too much creative in defining new outcomes.
00:13:10
Speaker
So what we would do typically is suggest to take multiple outcomes that have been already used in past kinder trials as either primary or secondary or expiratory inputs,
00:13:23
Speaker
but making sure that you measure them and you incorporate them within this composite primary endpoint.
00:13:30
Speaker
That's what we typically suggest to our pharma sponsors.
00:13:34
Speaker
Then something really important is if you define this primary endpoint being composed of, let's say, four different prioritized outcomes, what you will get at the end of the clinical trial is also an individual contribution
00:13:51
Speaker
of each of those four outcomes to the net treatment benefits, the total net treatment benefit.
00:13:56
Speaker
So imagine that at the end of the clinical trial, you show that your new treatment is bringing a 20% net treatment benefit.
00:14:05
Speaker
You will also be able to tell whether this 20% net treatment benefit is mainly coming from the first outcome
00:14:13
Speaker
or the second outcome, or is there one out of the four outcomes that is bringing some kind of small detrimental effect to patients?
00:14:24
Speaker
So that's also something positive.
00:14:25
Speaker
You do not always expect new treatments to bring only positive advantages or benefits to patients.
00:14:32
Speaker
You just also want to be aware in a transparent way as a clinician, as a patient that, I don't know, the quality of my sleep, for instance, may be slightly affected by this new treatment.
00:14:42
Speaker
But compared to another treatment, I will, I don't know, do not lose my care.
00:14:47
Speaker
Maybe I don't care.
00:14:49
Speaker
So again, I mean, having this kind of transparent understanding at the end of the clinical trial about what's the individual contribution of each outcome is also something very, very engaging.
Engaging Regulators and Innovative Methodologies
00:15:00
Speaker
Talking about the regulators,
00:15:02
Speaker
So do you yourself engage with regulators directly or is it something that you ask or recommend that your clients do when they're thinking about implementing such a model?
00:15:23
Speaker
So we typically try to work in parallel on both streams.
00:15:27
Speaker
So definitely, or I mean,
00:15:29
Speaker
pharma sponsors will interact with regulators as they do with a more standard single outcome, single endpoint approach.
00:15:39
Speaker
But in parallel, I mean, we definitely have a lot of discussions ongoing discussions with regulators so that they're aware about this, I would say more innovative approach, knowing as well that
00:15:52
Speaker
For instance, let's say the example about the FDA, there are different divisions, for instance, cardiovascular disease that are really aware of the methodology.
00:16:02
Speaker
And I've been using again and again, this methodology for many different clinical trial designs.
00:16:07
Speaker
But also, as I was mentioning, for some drugs that are approved right now that are on the market, one drug for instance, being Daphamidis from Pfizer.
00:16:16
Speaker
So this drug has been approved by FDA EMA based on the composite
00:16:22
Speaker
approach and generalized comparison approach.
00:16:25
Speaker
So again, I mean, that's just the beginning.
00:16:28
Speaker
And right now that's the goal is really making it popular, not only for a few specific therapeutic areas, but more broadly for all indications or therapeutic areas that may benefit from a multi-facet approach in order to better reflect what matters to patients.
00:16:48
Speaker
It's great to hear that it's already been used and actually approved.
00:16:54
Speaker
I'm going to play devil's advocate here.
00:16:59
Speaker
I mean, I'm behind everything you've said.
00:17:02
Speaker
I think it's amazing progress for our industry.
00:17:06
Speaker
However, maybe some of our listeners will think, well, that sounds like a lot of work.
00:17:14
Speaker
upfront before you even start your trial.
00:17:17
Speaker
So what benefits do you expect them to gain from applying it to their trial?
Benefits of GPC: Sample Size and Statistical Power
00:17:26
Speaker
So I probably already touched a few of those benefits.
00:17:29
Speaker
I mean, for me, the first one is, um, being able to engage with different clinicians.
00:17:35
Speaker
I mentioned site investigators or key opinion leaders within, you know, scientific advisory reports before, you know, defining the protocol of the trial.
00:17:44
Speaker
Um, long story short, because you leverage more data points, you have more
00:17:50
Speaker
data and more statistical power.
00:17:52
Speaker
What does it mean?
00:17:53
Speaker
It means that the consequence and the direct consequence is you can expect a significant reduction in sample size.
00:17:59
Speaker
So typically what we see is for rare disease, we have done it a few times recently, instead of having, I don't know, 150 patients to be recruited,
00:18:09
Speaker
you may run your clinical trial with the same power with only 90 or 100 patients instead of 150.
00:18:15
Speaker
So you really reduce your sample size by 40%.
00:18:18
Speaker
And again, 40%, I mean, it's not always happening.
00:18:23
Speaker
From time to time, it's less.
00:18:23
Speaker
From time to time, it's a bit more.
00:18:25
Speaker
But let's say between 20 to 50% sample size reduction.
00:18:29
Speaker
You bring transparent analysis of your treatment effects, not focusing on single primary endpoint and then individual secondary endpoints.
00:18:39
Speaker
that you will always find someone like you were playing the advocate, you will always find someone that will play later on at the end of the trial, the devil advocate, just pick one single primary endpoint and just say, the results is not that good for that specific.
00:18:58
Speaker
That's happening when you just analyze and investigate endpoints one at a time.
00:19:03
Speaker
If you just put them all together, then again, I mean, you can discuss, I mean, the prioritization of those outcomes, but you can rerun the analysis.
00:19:12
Speaker
And that's again about the transparency game that I was thinking about, that regulators, they love it.
00:19:18
Speaker
That's basically what you will be asked for HTA discussions with, you know, payers with, you know, to get reimbursement and so on.
00:19:26
Speaker
So that's after R&D, I would say step, but that's really allowing you to prepare all those discussions by showing again, I mean, the totality of the evidence, all the data points, all the outcomes that you are, have been measuring and
Impact on Clinical Data Capture
00:19:39
Speaker
You can change a bit of prioritization and, you know, answer a lot of questions transparently without having to hide things and so on, but more, I mean, discuss the prioritization and more the clinical relevance of this kind of list.
00:19:54
Speaker
how do you expect this methodology to impact the clinical data capture or at least the type of data that's captured?
00:20:04
Speaker
So that's a good question.
00:20:05
Speaker
So what you've seen over past clinical trials is from time to time there are secondary endpoints or exploratory endpoints that are captured and, you know, tracked, but not always with
00:20:20
Speaker
I would say perfectly from the side.
00:20:22
Speaker
So you may have missing data, you have for some patients and some stuff.
00:20:28
Speaker
It's not always easy later on at the end of the clinical trial to analyze, for instance, patient reported outcomes for such a reason.
00:20:35
Speaker
Just taking an example.
00:20:37
Speaker
Imagine right now that the same patient reported outcomes is incorporated
00:20:43
Speaker
as part of this list of prioritized outcome, probably not as the top one or two outcomes, but maybe outcome number four or five.
00:20:52
Speaker
You may have quality of life or scales.
00:20:53
Speaker
You may have patient reported outcomes.
00:20:57
Speaker
And then all in a sudden it's changing a lot because then you're putting much more visibility, transparency on those outcomes.
00:21:03
Speaker
So the way you will have to measure to track, I mean, those kinds of outcome will be
00:21:08
Speaker
globally the same, but you will put much more, I would say, importance on those.
00:21:12
Speaker
And again, I mean, everything related to data management will be much more important because it will be part of the primary input and you want to make sure you can bring visibility on that.
00:21:23
Speaker
Make sure even if it's not part, coming back to this example of patient reported outcomes, even if it's not part of the primary endpoint, coming back to the discussions later on about the payers and discussion with HCM, maybe at some point of time, you just want to
00:21:38
Speaker
offer players the opportunity to define their own list of prioritized outcomes.
00:21:43
Speaker
So again, you need those data points to be really, I would say, even in a better way, captured tracks without having too much missing data.
00:21:52
Speaker
And so I do believe it's just a methodology that will put even more importance and strategic visibility of spotlight on the way data management is handled throughout kindergarten.
00:22:06
Speaker
And the point you're making about patient reported outcomes, like I can see now the impact on not necessarily the data capture, because I imagine that you would still have that as part of your trial.
Quality of Life in Data Management
00:22:20
Speaker
But then you mentioned risk-based quality management.
00:22:25
Speaker
That could then become like your quality of life questionnaire could become one of your areas that you pay more attention to if you know that it will impact your data analysis at the end.
00:22:37
Speaker
Unfortunately, that's all we've got time for today, except for one last question.
Advice on Aligning Industry Efforts with Patient Benefits
00:22:43
Speaker
What's the best piece of advice you've received that you apply to your daily work?
00:22:52
Speaker
That's a good one.
00:22:54
Speaker
I would probably say, let's make sure you truly listen to the end user, what we may call in many different industries,
00:23:04
Speaker
truly listen and truly try to understand your end customer.
00:23:08
Speaker
And again, I understand working in clinical trials that customer is not a word that we like to use.
00:23:13
Speaker
And I do not, I'm not a big fan of this word, but when you compare what we do in clinical developments with a lot of other industries,
00:23:22
Speaker
the best vendors, the most competitive companies are usually the ones that better listen to and better understand the end user, the end customer needs.
00:23:33
Speaker
So we can from time to time, I mean, step back and, and try to wonder whether we truly listen and truly try to understand what are the needs from the end users.
00:23:44
Speaker
And again, I mean, who,
00:23:45
Speaker
are the end users, are there patients, are there payers, are there regulators?
00:23:51
Speaker
That's I do believe where from time to time we struggle as being part of this development industry about defining who is the end customer, the end user, who do we try to convince when we run Kinkam research.
00:24:03
Speaker
But again, if you believe, and I do believe that at the end of the day, what you want to do is help patients in needs,
00:24:12
Speaker
And I would say that's, I mean, that's the goal of everyone working in clinical research.
00:24:16
Speaker
That's really some reason why I love this industry is because everyone is so much passionate about trying to help patients in needs.
00:24:25
Speaker
I do believe we can do a better job in, again, defining what does it mean bringing a good treatment or bringing a better treatment to market for those patients.
00:24:36
Speaker
How do we pre-specify it before trial starts?
00:24:40
Speaker
So that's the primary endpoint objective, but also definitely making sure at the end of the day that you bring more transparency about all those individual outcomes.
00:24:51
Speaker
the advice I would give is let's step back and from time to time challenge ourselves.
00:24:57
Speaker
Like do we truly listen and try to understand what is the most important objective?
00:25:03
Speaker
And there are from time to time, some small changes that we're there to try, there to challenge ourselves, challenge a bit of statical and a comfort zone and trying to redefine a bit the way we do define what is better treatment.
00:25:19
Speaker
and making sure this definition is not just a definition that will please regulators.
00:25:25
Speaker
I don't think that they are the end users.
00:25:28
Speaker
I don't believe that they are very important stakeholders that will allow us something to guarantee.
00:25:34
Speaker
I would say that the end user, the patients in need, they will get exactly what is good for them, meaning that the definition is good.
00:25:41
Speaker
That's the goal of regulators is approving this definition of what is better treatment than the other.
00:25:46
Speaker
But let's make sure that this definition is also something easily
Passion and Purpose in Clinical Trials
00:25:49
Speaker
understandable, visual, transparent, simple to be understood by patients.
00:26:00
Speaker
And interestingly, I fully agree with you.
00:26:04
Speaker
We need to listen to end users.
00:26:07
Speaker
In the case of clinical trials, they are the patients.
00:26:12
Speaker
In my case, the end user may not be directly the patient, but focusing on clinical sites, for example,
00:26:22
Speaker
in a way impacts positively how sites will interact with patients.
00:26:28
Speaker
But yeah, I fully agree with you.
00:26:29
Speaker
Listen to your answers.
00:26:31
Speaker
And I do believe, I mean, when you have a strong purpose, easily explainable and understandable, I mean, then everyone is aligned.
00:26:39
Speaker
As I was mentioning, that's the beauty of this industry, right?
00:26:42
Speaker
Everyone is really so much passionate about doing something good for patients in need.
00:26:46
Speaker
So again, making sure you can remind the reason why you want
00:26:51
Speaker
you know, sides, I mean, maybe to do things slightly differently, you know, regulators and so on.
00:26:57
Speaker
I do believe that's, you know, this kind of idea about dare to challenge, respectfully challenge a bit in the StatuCo and bring smooth innovation that will truly help, you know, better answer this kind of end user needs, patient needs.
Closing Remarks
00:27:14
Speaker
Well, thank you very much, Sébastien.
00:27:15
Speaker
It's been very nice talking to you.
00:27:18
Speaker
I always feel like motivated after a discussion with you.
00:27:23
Speaker
Hopefully everyone who's listened to us will do as well.
00:27:27
Speaker
Thank you for your time.
00:27:29
Speaker
Thank you so much, Sylvie.
00:27:31
Speaker
And thank you everyone for listening to us.
00:27:35
Speaker
And you will be able to find more clinical data talks on the CRS Cube website.