Latest / Cardio Pod: Heart of the Matter / DSMB Training Academy: Preparing the Next Generation of Members (Part 1)

Transcript

DSMB Training Academy: Preparing the Next Generation of Members (Part 1) · Cardio Pod: Heart of the Matter

0:00
Good afternoon, everyone.
0:05
I'm Chris O'Connor.
0:06
I'm principal investigator at the Heart Filler Collaboratory, and we're here with our cardio
0:10
podcast Heart of the Matter.
0:12
And today we're discussing data safety monitoring boards in a recent data safety mining board
0:17
training academy.
0:19
And I'm here with my distinguished colleagues, and I'm going to ask to go around the table
0:23
and introduce themselves.
0:25
And I'll start with Dr. Janet Wittes.
0:27
Hi, I'm Janet Wittes.
0:29
I'm a statistician.
0:31
I've been involved in clinical trials of many kinds and have been on DSMBs now for over
0:37
40 years.
0:38
So it's time to roll off and have a new generation start.
0:42
Fantastic.
0:43
Thank you, Janet.
0:44
Dr. Mona Fusa.
0:45
Hi, I'm Mona Fusa.
0:46
I am a Francis Byte training and co-PI of the Heart Filler Collaboratory and helped to
0:55
organize the first pilot training academy at the FDA.
1:00
Dr. Maria Rosa-Constanzo.
1:03
I'm a heart failure cardiologist and a very proud member of the Heart Failure Collaboratory.
1:12
I was fortunate to participate in the training session at the FDA.
1:18
Also, I'm currently on a DSMB that Dr. Greenberg invited me to be on.
1:27
And I used to be the humble servant of Dr. Mike Bristow when as a fellow, I used to collect
1:36
biopsy samples to evaluate function of beta receptors in the myocardium.
1:44
That must have been a challenge, Maria Rosa.
1:46
We'll talk about that after the podcast.
1:49
Dr. Anna Barak.
1:51
I'm Anna Barak.
1:52
I'm a cardiologist with the Nova Heart and Vascular Institute.
1:55
And I was grateful to be a participant and trainee on this spectacular session.
2:02
My area is cardiology, which I think is very relevant for the Heart Failure Collaboratory.
2:08
So I learned a lot and I look forward to continuing to learn with this initiative.
2:15
Thank you.
2:16
Barry Greenberg.
2:17
I'm Barry Greenberg.
2:18
I'm a heart failure cardiologist at the University of California in San Diego.
2:24
I've worked in clinical trials ever since the SALV study, which started back in the
2:29
1980s.
2:31
And I've been involved in a number of data safety monitoring boards.
2:35
I was also fortunate enough to be on the faculty for the Collaboratory meeting about data safety
2:43
monitoring boards and perhaps not so much to my amazement, but to my enjoyment.
2:48
I learned a lot at that meeting and I think everybody else there did too.
2:53
Thank you, Barry.
2:54
Dr. Vanessa Bloomer.
2:55
Hi, everyone.
2:56
I am Vanessa Bloomer.
2:57
I'm an advanced heart failure and transplant cardiology fellow for another week.
3:03
I'll be finishing this Thursday.
3:06
And I was also very honored to be part of this first training experience, which I agree
3:11
with Dr. Brock was spectacular and hope it continues in the future.
3:15
Also like Dr. Constancio, a proud member of Heart Failure Collaboratory and very eager
3:20
to be part of this discussion today.
3:23
And last but not most senior, Dr. Michael Bristo.
3:26
Thank you.
3:27
Yes, it's a pleasure to be here.
3:28
I'm Michael Bristo.
3:29
I'm a cardiologist and pharmacologist at the University of Colorado.
3:34
Long standing, meaning 40-ish type years in clinical trials, including all aspects, which
3:40
includes fairly extensive DSMB work.
3:43
And I'm delighted that this group has gotten together to discuss this important topic.
3:48
Well, this Academy was truly wonderful because of the people that were there and the fact
3:54
that we could be in person and we were at the FDA White Oak campus and we had a guest
4:01
visitor from Dr. Rob Calath.
4:03
But Janet, maybe let me open up with you and say, you know, the urgency of the problem.
4:10
There's so many trials now.
4:12
There's so many new therapies.
4:14
We're excited about new devices, new molecules, new therapeutics, new biologics.
4:20
And yet we know we need data safety monitoring boards on many of these trials to be involved
4:26
at the very beginning.
4:27
And the number of people that are really trained to run complex data safety monitoring boards
4:34
has become limited.
4:36
And the talent pool is moving up in age and some are considering retirement.
4:42
Tell us more about the urgency.
4:44
Well, I agree.
4:45
It is very urgent.
4:47
The, partly there are many more trials now.
4:52
Some of them are very complicated.
4:55
The interventions are new.
4:57
We have interventions that we've never seen before.
5:01
Cell therapy, gene therapy, and all kinds of combinations.
5:05
And so it's really important that to preserve and make sure that participants in these trials
5:13
are safe, that we will learn from the trials, that people be watching the data as they come
5:20
in and making sound judgments about what to do and what you recommend to trials to do.
5:29
And I think the only way you can learn how to do that is by participating in these meetings,
5:37
but not being the only one, not being the only cardiologist, but not being the only
5:42
statistician.
5:43
There's got to be some people, some of whom you can learn and discuss.
5:48
And we have to have lots of people, but very well trained and very sensual as those of
5:56
us who've been doing it for many years fade out.
6:00
So that's the urgency.
6:01
The urgency is the needs for a lot of people and the people who have been there are getting
6:07
old.
6:08
And as you said, the world is getting so much more complex in these trials.
6:13
They're international, there's multiple different new techniques to evaluate information.
6:20
Dr. Fuzat, the Collaboratory embarked on this with a conversation with Dr. David Demetz.
6:27
Why was this important for the Collaboratory?
6:29
Sure, yes.
6:30
And to acknowledge Dr. David Demetz, who's not with us today, this really was an idea
6:35
of his.
6:37
And so we hope to expand on that perspective in a future podcast.
6:41
But basically, during COVID, he had already created these amazing videos and had already
6:48
engaged this whole statistical team to really do a lot of didactic, but it didn't have much
6:55
air time.
6:56
And he came to us at the Collaboratory and said he could really use some help with a
7:00
platform to make this known.
7:03
And we thought it was very well aligned with the initiatives of the Collaboratory in training
7:09
the next generation and diversifying the next generation who are involved in clinical trials.
7:16
And this is such an important part of the clinical trial fabric that is very rarely
7:22
receiving notoriety or even maybe much knowledge, but a crucial part of a clinical trial.
7:28
And so we thought this was right up our alley to help take this initiative to the next stage.
7:34
Well, and it was truly the leadership of David and Janet and Tom Fleming in putting together
7:42
a really unique day in which there were some didactic sessions.
7:48
We had former chairs at DSMBs like Dr. Pristo and Dr. Greenberg give their reflections on
7:54
lessons learned.
7:55
And then we went into True Practicum, where we set up mock DSMCs and we had the young
8:02
trainees co-chair those.
8:05
We had sort of the mid-senior career clinical investigators be part of that DMC and the
8:12
statisticians walked us through relevant data that emerged as if it was an ongoing live
8:20
clinical trial.
8:21
And I'm going to go around the room and just ask the team, my colleagues, what were the
8:26
top takeaways from the Practicum experience?
8:30
And I'll start with Dr. Anna Barak, who you were a trainee.
8:35
Yes.
8:36
And I felt like the best days of medical school when you sit in the room and the history unfolds
8:42
in front of you and say, oh my, I need to apply all of this.
8:46
So it provided not only the experience of how critical DSMB role it plays in assuring,
8:56
yes, safety of the patients and clinical trials, but how a visionary DSMB can help in the design
9:04
in assuring the integrity of a clinical trial, also assuring that the right endpoints are
9:10
being collected accurately and frequently as we learn really decide or drive the fate
9:18
of the clinical trial.
9:20
So living it through the Practicum experiences and getting a live discussion between the
9:28
chair and the statistician was just invaluable.
9:32
Thank you.
9:33
And I got to ask Dr. Blumer, who was also a young trainee who was asked to share the
9:39
DSMB that Dr. Wittes was the statistician and Dr. Costanzo was the clinical investigator
9:47
part of that committee.
9:48
Dr. Blumer, what did you take away?
9:51
First and foremost, I want to congratulate you all because this was honestly an incredible
9:56
experience that I learned a lot.
9:58
It's hard really to summarize just in one statement.
10:01
I think first, one thing that I learned is how much is available to DSMB members.
10:06
And I think it's important for DSMB members to know what they can ask for and what information
10:11
can be available to them.
10:14
So I think that in itself is a learning experience because I think had I not done this training,
10:19
I would not know what I can ask for my trial and what information can be available to me.
10:24
When can I unblind a trial and that I should ask for unblinded information.
10:29
So I think that is one.
10:30
I think the other one is the impact that DSMB members can have in a trial.
10:36
And I think that's another discussion that we can have moving forward.
10:38
But the impact that it can have, for example, in diversity of a trial and in terms of outcomes
10:43
of a trial.
10:45
And then I think a large thing that came up in our discussion and in our table is how
10:50
do you balance keeping patients safe and patient safety while also maintaining trial integrity
10:57
and when is it okay to stop a trial for futility versus safety.
11:04
And I think that also comes down to having enough statistical support or statistical knowledge
11:08
to know when is it early enough for you to actually stop a trial or considering stopping
11:14
a trial for futility.
11:15
In our table, for example, we always kept going with a trial and that was always the
11:21
answer.
11:22
But is there a point where you should actually consider just stopping the trial for futility
11:27
or should you feel comfortable enough doing that?
11:30
And then when is that point?
11:32
When do you feel when should you actually stop the trial for futility or safety?
11:36
I'll leave it at that for now.
11:38
Well I'm going to ask Dr. Costanzo to comment because she was on that committee and then
11:43
Dr. Wittes to reflect as the statistician whether this committee got it right.
11:50
So Maria Rosa?
11:52
Well first of all, I think that I learned something that was not completely clear to
12:00
me and that is the ability to obtain unblinded data and how knowing unblinded data can help
12:11
you make the right decision for the patients.
12:15
And also I think that all the statistician presence with one voice said that you should
12:25
request it and you have the right to see it as a DSMB member.
12:31
That was very reassuring to me.
12:36
And the other thing I learned and probably because of the way things are presented when
12:42
you're asked to be on a DSMB, you think that a DSMB charter is this rigid series of rules
12:52
that you have to always abide by and instead the statisticians present at the workshop
13:02
said that the charter is actually a series of principles to guide the proceedings of
13:11
the DSMB instead of being a series of rigid requirements.
13:18
And the other thing that they pointed out that was very useful to me was that you have
13:26
to make a decision by consensus, not by rigid voting.
13:32
That was very important and I think we did a good job at our table in reaching a consensus
13:41
rather than being forced to a vote.
13:45
Well Janet, you were the teacher.
13:48
Did table one get an A based on those two comments?
13:51
You guys got an A plus.
13:54
And part of the A plus was how engaged you were and how the committee was able to really
14:03
argue with each other and think about different alternatives.
14:08
One of the things I feel very strongly about is that committees should not be recorded.
14:14
There are committees that are recorded and the nature of the discussions, remember, at
14:20
a DSMB you're playing around with decisions that have very serious consequences for the
14:28
safety of patients in the trial and for the wider world.
14:34
And so it's reasonable that people on the committee will disagree and they have to feel
14:40
very comfortable and they may even disagree with themselves, disagree with the others.
14:46
You have to feel very comfortable expressing your views and your uncertainties and your
14:51
certainties without feeling that there's a big brother over you listening.
14:57
So over the years I've changed my mind a lot.
15:00
I used to think we should stay blind.
15:02
I think that's ridiculous.
15:04
I used to think you should vote.
15:06
I think that's ridiculous.
15:08
But it's been a learning experience as different committees, as something happens and you realize,
15:15
no, we have to think of ourselves as one body that figures out what to do.
15:23
And there's also, I mean, in the particular examples that we dealt with, it turned out
15:29
that we were right, that the decision to continue the trial was right in the sense that by the
15:36
end of the trial we showed benefit of the drug.
15:40
It could have been that the decision was wrong.
15:44
It could have been that nothing happened and that it would have been a waste of money to
15:49
continue when there was no evidence of death.
15:53
But you only know that at the end.
15:57
Really well said.
15:58
And I want to go to two of our colleagues who combined may have more experience sharing
16:05
the SMBs and start with Dr. Barry Greenberg.
16:08
You gave some really sound advice from the lessons learned from your experiences, Barry.
16:14
Maybe you could share a couple of those with the audience today.
16:18
Yeah, I'd be glad to.
16:21
Some of the lessons that I've learned over the years and were discussed and affirmed at
16:28
the meeting included the need to look at unblinded data when we're evaluating a clinical trial
16:36
to really understand what the data means in the context of the strategy or the treatment
16:44
that's being tested and to be able to understand the potential benefits and also the potential
16:53
side effects of that treatment.
16:57
And then also the need to develop that consensus amongst the group and to do it in a way that's
17:05
unfettered.
17:06
And I think that I agree with what Janet said very strongly about not having the meetings
17:13
recorded so that people can really freely express their views on a clinical trial.
17:21
What's particularly interesting to me about the meeting, and I think it served a tremendous
17:26
purpose, is the fact that what was discussed at the Collaboratory was really the canon
17:34
of knowledge that's developed over the last 40 years and how it's evolved and what we've
17:41
learned and the need to be able to pass this on to a new generation is absolutely critical
17:50
for us to be able to do.
17:53
So I was really pleased at the meeting, the fact that we had younger people there and
17:59
we could share some of these lessons.
18:02
The other aspect of this is that clinical trials, and I think others have alluded to
18:07
this, clinical trials are changing.
18:10
They're not the same as they were 40 years ago.
18:13
They're not the same as they were five years ago.
18:17
And there are a lot of new ways and new methods that are being incorporated in clinical trials.
18:24
And this is going to require flexibility on the part of data safety monitoring boards
18:30
as they view these clinical trials, as they evolve and as they unroll.
18:38
And one of the lessons, going back to the question you asked me, Chris, that I learned
18:43
and was reaffirmed at the meeting was the strong support that people who had served
18:49
on data safety monitoring boards offered for viewing the protocol before the initiation
18:56
of the trial so that the data safety monitoring board could make their comments about the
19:04
design of a clinical trial and offer a very independent and constructive critique of that
19:12
trial because after all, they are going to be monitoring that trial and making critical
19:18
decisions about its continuance over a period of time.
19:23
Really well said, Barry.
19:25
Thank you for sharing that.
19:26
Mike Bristow, you, Dr. Fusat, myself worked on an important paper looking at regional
19:31
differences in outcomes by the beta blocker trials.
19:35
And then fast forward from that paper, you were the chair of the data safety monitoring
19:41
board for Topcat, which faced some geographic challenges.
19:45
How much, how important is it for a chairperson to look at geographic variation by outcome
19:53
in an ongoing clinical trial?
19:55
Yeah, sure.
19:57
Geographic influences, I think, can be manifested a number of ways.
20:02
It's certainly an important consideration when you're monitoring the progress of a trial
20:07
to pay attention to that and you need data that's subdivided by geographic region.
20:12
And then if trends emerge that look like they're causing some sort of effect on outcome or
20:19
enrollment or whatever, you need to do a deeper dive on that as the trial goes on.
20:25
It's kind of a controversial topic because you can have, you know, random differences
20:30
going on, various things in a trial and you're dealing with incomplete data when you analyze
20:36
the data.
20:37
So it's a balance between sort of a statistical paradigm and then common sense of what could
20:43
be happening in one place versus another.
20:47
It's just one of the many duties that you have in monitoring a trial.
20:52
I also like to kind of return to some general principles that have been mentioned already
20:58
here just for the sake of people who might be listening to this, who we don't really
21:02
understand what a DSMB is.
21:05
I think safety of the patient is paramount, but also integrity of the trial.
21:09
These have both been mentioned and you have to balance those two.
21:13
But the really nuts and bolts of what you're supposed to be doing or need to be doing is
21:18
dealing with problems that arise.
21:21
And so you can just expect the unexpected in a clinical trial.
21:26
So it's incredibly important to have people on that committee who are very experienced
21:31
for working and multidisciplinary.
21:33
In other words, you need statisticians, you need clinicians, you may need a pharmacologist
21:38
in some cases, you may need various people.
21:42
It's not a matter of a one or two person show typically in a large phase three trial.
21:48
And all kinds of things happen in these trials, which makes this work very interesting and
21:54
challenging.
21:55
Common sense needs to be the common denominator here when this data, these data come in that
22:01
you need to evaluate.
22:03
So I think, and I've said this many times in the context of the Academy and other places
22:09
as we were planning, having our planning sessions, I think of all the things that you do in a
22:14
clinical trial, I think DSMB work is actually the most interesting thing you can do.
22:20
Just because there are these challenges that arise that need to be dealt with and there
22:24
may not be any information, any background press that we're for dealing with.
22:30
So if you like to not be bored in the work you do, this is one place that you can operate
22:37
and be entertained if nothing else.
22:41
Well said Mike, but not only interesting and knowledge acquisition, but so important to
22:48
the patients and to, like you said, the integrity.
22:51
I mean, there's just a lot of activity ongoing in a trial that the DNC has a lot of response
22:59
for.
23:00
So thank you for sharing that.
23:03
I'm going to pose to the panel a challenging question.
23:07
And I don't know if I should start with Janet or not, because every once I asked Janet a
23:11
question it's like a leading answer and everybody says, oh yeah, I agree with Janet.
23:16
So you're on it, you're chairing a DMC, a data safety monitoring board, and data is
23:22
coming in and it's clear that you don't have enough information in the special populations,
23:31
whether it's in the women or it's in the underrepresented minorities, but it's deemed
23:37
important.
23:38
But you recognize that the sample size in those groups are small and yet there's emerging
23:45
trends in the overall trial.
23:48
Do you keep going to get a more definitive answer in the population of women or underrepresented
23:57
minorities when there is not enough evidence in those sets, but there may be enough evidence
24:04
overall?
24:05
And I'm going to start with Maria Rosa because she's a good friend of mine and this is a
24:09
hard question.
24:12
From what I understood from the proceedings at our table, I think you can request that
24:21
the investigators redouble their efforts to include underrepresented groups.
24:31
And in fact, I think this is one of the areas where I learned that the SMB can intervene
24:39
that I didn't think was part of the charter of the DSMB.
24:45
Dr. Blumer?
24:48
I will agree with Dr. Constanzo and I'll agree in advance with whatever Janet says.
24:54
But yeah, I think this is one of the lessons that I also learned from the training and
24:57
it's the impact that DSMB can have in enhancing diversity.
25:02
And I think that's what I would do.
25:03
I think I would make a call on the investigators to try to, to some degree, increase recruitment
25:08
and retention of underrepresented groups and women if that's the case in this research
25:14
or in this trial.
25:15
But yeah, I'd be intrigued to see what Janet says or what her approach is going to be.
25:20
Well, Janet, like EF Hutton, we listen when you speak.
25:24
I wish my family would say the same thing.
25:28
I thought you were asking a very specific question and that's probably because that's
25:33
what I was thinking.
25:34
You're in a trial, there are men and women, that there's a clear benefit in the men and
25:41
there's no benefit in the women.
25:44
What do you do?
25:45
And I was involved in a trial like that.
25:48
I'm comfortable saying what it was, but I looked at the data and I thought to myself,
25:53
oh my goodness, I'm going to want this trial to continue because it's going to be so hard
26:00
to interpret.
26:01
And I thought to myself, nobody else is going to think the same thing.
26:07
And I cheered it so I didn't have to start.
26:09
It started with the first person on my left and he said, look, none of you are going to
26:14
agree with me, but this trial has to go on.
26:18
We can't have no effect in the women and a big effect in the men.
26:23
And if we stop the trial for the men, it'll be just too confusing.
26:26
And everybody breathed a sigh of relief and it turned out that they all want, that they
26:31
had to continue and that nobody else would agree.
26:34
But we were very nervous.
26:36
We were very nervous that at the end of the day, we would be criticized.
26:41
And I do think that's an issue that you need to know about when you join a DMC, that there
26:47
are decisions and recommendations that you make that other people may not have and that
26:53
you may be criticized and you have to be able to accept that.
26:57
I do want to say one more thing that's germane to this, but not necessarily.
27:03
One of the wonderful things about being on a DSMB is the interaction among the various
27:11
disciplines where each discipline has to teach the other discipline so that you can all talk
27:18
and so that you can reach the consensus.
27:20
And that's really been a wonderful, for me, one of the best things about DMCs.
27:25
Janet, can I ask you a question in terms of when you don't see a benefit in women, isn't
27:35
often the case that the benefit is not seen because women are underrepresented in the
27:43
trial?
27:44
I'm referring specifically to experiences I had at the end of this trial where people
27:53
were going around saying that ICD did not benefit women.
27:59
And then when you looked at the composition of the population, women were grossly underrepresented.
28:06
Yes.
28:08
So I think there's so many.
28:10
This question to us is so broad and there are many situations.
28:15
Yes, if you have an underrepresented group, it's very hard to interpret the data.
28:20
I'm thinking of a group where it wasn't underrepresented and yet there was a huge difference.
28:26
And the question that you kind of have to pose is at the end of the trial, how will
28:33
people interpret these data?
28:35
And that's why it's so useful to have different disciplines talking about it.
28:40
Really highly informative.
28:42
Wow.
28:43
We've learned a lot today and I'm going to let Dr. Fusat give the last word.
28:48
We set up an academy.
28:49
I think we trained two dozen or three dozen people.
28:54
David Demets and Janet said we're about several thousand people down.
28:59
We need the SMB Academy to occur every week to catch up.
29:05
What are the collaboratory's plans to train the necessary amount of workforce that the
29:11
urgency that Janet has prescribed?
29:13
Yeah, I mean, we didn't set up the Academy.
29:17
The statisticians did all the work.
29:19
We just brought them all together and we really need to give them the credit for all of the
29:24
work that they've done.
29:25
And then the participation of all of our senior leaders or people with experience that came
29:32
to the table.
29:33
It was really a phenomenal experience.
29:35
And we only just touched the surface.
29:37
We're only talking about training investigators that would be members of the DSMB, but there's
29:43
a whole other group of people that we didn't even get to, which is the industry.
29:49
The industry members that need a lot of training probably as well that have never been through
29:54
it.
29:55
Janet just touched on things that they should include in reports.
29:59
And I think there's still so much opportunity.
30:03
Unfortunately, there's just not enough time and resources to train everyone, but having
30:09
the free videos that the statisticians created is just such a gem.
30:14
And we hope to get the word out.
30:16
FDA members, regulators, even NIH members and industry could really benefit from this.
30:23
So I hope that we continue to just make it known that the videos are there and to continue
30:29
reaching the community.
30:33
So really, Janet, and we'll obviously relay this to David and Tom, you all thought that
30:38
maybe a year or two and you would retire, but I'm looking at about 10 to 15 years of
30:43
an annual academy going and a lot of work that we need to do.
30:47
So I just want to thank this panel for an excellent discussion today.
30:51
Thank you so much for sharing your thoughts on what's so important to the ecosystem of
30:56
development, the data safety monitoring boards, and we look forward to further discussions
31:01
on this topic on the Heart of the Matter podcast.
31:05
Thank you.