Clinical Trial Arenas: Direct Communication to Patients

In this episode of "In the Interim…", Dr. Scott Berry recounts a presentation he gave to the Speak Foundation, a patient organization focused on limb-girdle muscular dystrophy and makes the case for master protocols directly to the stakeholder trials exist to serve. Opening with the deliberately absurd premise of a sports league that constructs a four-billion-dollar stadium for every single game and demolishes it at the final whistle, Scott reframes the conventional one-off trial as a two-year construction project dismantled the moment the contest ends. The episode distinguishes platform trials and basket trials as standing "arenas," explains the modular role of the master protocol described in Woodcock and LaVange's 2017 New England Journal of Medicine paper, and draws on worked examples including I-SPY 2, the HEALEY ALS Platform Trial, GBM AGILE, Precision Promise, and the ROAR basket trial. Scott takes each stakeholder in turn from the treatment arm sponsor, the site, the regulator, and the patient. Scott lays out how three-to-one randomization lets an arm fund 133 patients while reading out inferences with more than 200, how 667 patients in a single platform can do the work of 1,000 across five separate trials, and why far fewer patients are assigned to placebo. He closes on the genuine bottleneck: the first arena is the heavy lift, and in rare disease it is frequently patients themselves who make it happen.

Key Highlights

• The one-game stadium analogy of two years to build a trial, one contest, then demolition.
• Platform trials and basket trials as standing arenas for multiple arms and multiple subtypes.
• Master protocols as modular documents, with arms plugging in and out as appendices.
• Worked examples from I-SPY 2, HEALEY ALS, GBM AGILE, Precision Promise, and the ROAR basket trial.
• A stakeholder-by-stakeholder view: arm sponsor, site, regulator, and patient.
• 667 patients in one platform versus 1,000 across five separate trials, with far fewer on placebo.
• Bayesian borrowing across 39 limb-girdle subtypes, and the rare-disease shots never otherwise taken.

For more, visit us at https://www.berryconsultants.com/

Creators and Guests

Host

Scott Berry

President and a Senior Statistical Scientist at Berry Consultants, LLC

What is In the Interim...?

A podcast on statistical science and clinical trials.

Explore the intricacies of Bayesian statistics and adaptive clinical trials. Uncover methods that push beyond conventional paradigms, ushering in data-driven insights that enhance trial outcomes while ensuring safety and efficacy. Join us as we dive into complex medical challenges and regulatory landscapes, offering innovative solutions tailored for pharma pioneers. Featuring expertise from industry leaders, each episode is crafted to provide clarity, foster debate, and challenge mainstream perspectives, ensuring you remain at the forefront of clinical trial excellence.

Judith: Welcome to Berry's In the
Interim podcast, where we explore the

cutting edge of innovative clinical
trial design for the pharmaceutical and

medical industries, and so much more.

Let's dive in.

Welcome everybody back to In The Interim.

I'm your host Scott Berry, and today
I'm gonna, uh, try a new topic.

Uh, but by the way, just broadly,
uh, In The Interim is a podcast,

uh, hosted by Berry Consultants.

We are statisticians, clinicians,
uh, mathematicians working

on clinical trial designs,
innovative clinical trial designs.

In The Interim is a place that we
explore different topics in science,

different topics in clinical trials,
medical decision-making, typically

a quantitative, uh, focus, uh, on
this, a statistical focus on it.

And today is kind of a-- I,
I thought I'd talk about an

interesting experience I had.

So I was invited to give a presentation
to a patient organization And the patient

organization is the SPEAK Foundation,
and they are a patient organization

focused on limb-girdle muscular dystrophy.

And, um, a wonderful organization
founded by patients, run by patients

for the benefit of patients.

And they…

Uh, it's, it's an interesting disease from
the perspective of clinical trial science.

They're…

This is a disease that, uh, um, muscle
atrophy largely not starting distally,

but starting centrally in the body,
and really from the, the trunk,

the large muscles start to atrophy.

And the, the atrophy generally leads to,
um, uh, inability to walk, and, uh, within

that many of the patients that were at
this, this meeting were in wheelchairs.

And, uh, it's, it's a relative, of course,
of Duchenne muscular dystrophy, myotonic

muscular-- myotonic, uh, dystrophy.

Uh, there are a number of, of
dystrophies, uh, within this.

Limb-girdle is a specific
rare disease, but one of the

interesting aspects of it…

And by the way, go to the SPEAK Foundation
website to read much more about this.

I, um, I, I, I may be
getting parts of this wrong.

But, uh, there are identified 39 different
subtypes, uh, genetic mutations leading

to limb-girdle muscular dystrophy.

Yes, there's different levels of, of
rates of decline based on these different

levels of prevalence wi-within these.

Uh, but it's a, it's a really interesting
scientific disease from that perspective.

And so they were interested, the,
the SPEAK Foundation, the patients

were interested in multiple
things in clinical trial design.

They were interested in basket trials.

As you can imagine Uh, going in and
running a trial within one of these

39 subsets, uh, is, is perhaps a
negative to the other 38 groups,

or if you go in within a subset.

Could a trial be set up where
there are multiple subtypes being

explored or multiple treatments?

And so they, they were interested
in hearing about this innovation

of basket trials, platform trials.

So I was asked to speak on
this, and being a statistician,

I, I give a number of talks to
quantitative people, statisticians.

S- uh, speaking to statisticians,
many times there are formulas,

very specific details of that, but
this is presenting to patients.

Now, it turns out a number of these
patients are quite sophisticated.

They just may not have the same sort
of training, uh, of a statistician.

They're very familiar
with clinical trials.

They are in clinical trials, uh, i- in
there, but they were interested in hearing

about this, but a, but a very broad range.

So I thought on today's episode, I
would do a little bit about what did

I present to this and, and the, the,
the aspect of presenting to patients.

It's…

And, you know, I…

It's, uh, it's obvious to say that
these are, you know, th- this is an

incredibly important stakeholder.

They are the reason we do clinical trials.

And the, the, the impact that they
should have in trials, i- if you can't

explain impacts of clinical trials to
patients, that's prob- that's a problem.

So I, I took this as a
particular challenge, as a,

as a, as a, as an opportunity.

Very much an opportunity, and I've
always thought of platform trials and

basket trials as, wow, this would be…

If I were a patient within the, the
syndromes or diseases being explored,

I'd wanna be in one of these.

And so this was a, an,
an opportunity for that.

So I thought I'd take today's
episode and go through a little

bit of, uh, of this meeting.

Nothing-- I, I'm not gonna
give any confidential

information out based on this.

Not sure there was confidential
information about this.

But this was, uh, last week, a, a
presentation that I did Okay, so I'm gonna

go probably in character a little bit,
uh, and, and, and in podcast a little

bit as, uh, how I presented this to the
patients, what, what that sounds like.

But then backing out a little
bit, talking to the podcast, maybe

even some questions that came up
during the, the, the course of it.

Uh, the other part of this that's
always, that's always a, a challenge,

but a wonderful challenge, is I
don't use slides in the podcast, and

there, there's actually functionality
that I could have slides there.

Many of you consume this purely audio.

You're, you're driving to work in
the morning, driving home in the

afternoon, you're going for a jog,
you're listening to it, and, uh,

you're, you're listening to the audio.

You don't wanna be looking at slides.

It's, it's a, it's a wonderful challenge
to have to explain these concepts

without the crutch that is a slide.

It, it also, it, it, it accentuates
presenting to patients as well the

role of slides or not, uh, within that.

Largely the slides that I'm presenting
to them are a way to keep me on track.

But I'm thinking of this as I'm
speaking to them almost like a podcast.

The slides can help me, but it's,
it's, uh, really in, in what I'm

going to say, it keeps me on track
and it keeps me for-- not to forget

that I wanted to say some part of it.

Uh, but I don't want them reading my
slides or trying to gain information

from them, specifically patients.

This is a particular challenge.

Okay, so the name of my presentation
is Clinical Trial Arenas, and I'll

kinda go in character a little bit.

And by the way, my first slide
is a picture of, of the n-

the new planned arena, largely
football arena, in Washington, DC.

They're, they are planning and building
a new arena for the Commanders, the,

the NFL team in Washington, DC, and it's
supposed to be done in 2030, I think.

So I have a big picture, an
artist rendering of the new one.

And my, my first slide is that the
NFL has decided on a new approach.

For every game played in the NFL season,
they're going to build a brand-new

stadium They're gonna create, uh, a new
security for that stadium, new processes

around tickets, about, about concessions.

Uh, hi- they're gonna hire all new staff.

They're gonna train the staff
on how to, to run a, a game day.

And an NFL football game is
typically about three hours, but

the, the fans are gonna go there
and they're gonna tailgate before.

They're gonna y- you know,
enjoy in the parking lot.

They're gonna barbecue,
um, a- and enjoy it.

Then they're gonna go to the
game and probably afterwards.

So this is a full day.

But they're gonna create a brand-new
stadium for every one of these games.

And then when the game is over,
they take the stadium down.

They dismantle it, and then they're gonna
build a brand-new stadium for game number

two And r- now, of course, this is sort
of what we do for Olympic Games, and

we start 16 years ahead of time where
they're building this, and they spend

billions of dollars setting up what turns
out to be three weeks of Olympic Games.

Yes, they use these for other
things afterwards sometimes.

Sometimes these, these stadiums
lay relatively vacant depending

on the, the, the place.

But we're gonna do this in the
NFL, and, uh, of course, what,

what's the ramification to the
NFL that we're gonna do this?

By the way, the NFL is the
National Football League.

It's American football, and
it's, it's newly started here

in the fall, uh, uh, within it.

The ramification, of course, is that we're
only gonna play games every three years.

That's the time it takes us to build
these stadiums, and the cost is going

to be massive because we have to w- w-
in order for the NFL to be financially

productive, they have to charge prices
for this one game in this one stadium

to pay for what, you know, s- probably
$3 to $4 billion now to build a stadium.

That's about what this new
Washington, DC stadium is.

Okay, so that, that was largely
in character that I'm speaking

to the patients, uh, of this
new approach to continue.

Now, clearly this is, uh,
this, this is not real.

The NFL is not going to do this, but
this is what we do in clinical trials.

We build a large stadium that is a
clinical trial, typically two years

to build that trial from the creation
of the initial synopses to the initial

protocols, to going to regulators, to
getting a CRO that's going to run the

trial, to setting up the database,
creating the rules for the trial, and the

behavior of the people, the security, and
everybody who's going to run the trial.

What happens at a visit?

What data are we collecting?

The processes around this.

The whole thing takes two years
for us to build the stadium.

Then we run the contest.

We run the clinical trial.

By the way, we don't always let
people watch the result of the game.

We don't always even say what
the score of the game was.

Usually, we generally say who won the
game, w- but the, the, the, the, the video

of the game, the, the details of the game
rarely are that, that, uh, disclosed.

Now, sometimes we write up
an article about the game.

It comes out nine months
after the game was over.

You can read about what happened
to it in a particular journal.

Uh, and then we take it all down.

We take down the process, and then
another one is built somewhere else.

Again, a two-year
lead-in time to run this.

This is what we do.

This, this, this description, this clearly
fallacy of what the new NFL is going

to do is what we do in clinical trials.

And when you, when you look at it
in this light, it- it's no, it's no

mystery as to why clinical trials
are so unbelievably expensive.

We have to be able to pay the
costs of building a new stadium and

taking it down for one game Okay.

So what might a clinical
trial arena look like?

And of course, I'm using this analogy
of the arena now that I've given

this description of, of a large
sports league and how absurd it

would be to behave this way in it.

Um, uh, there, there are really two
general arenas that we talk about.

A platform trial is a standing arena where
multiple treatments regimens come in, and

they get the opportunity to play the game.

They're inserted into an
existing trial, and the data are

collected, the results are there.

There's other synergies that we'll,
we'll talk about, about that, but

that's really a platform trial.

Multiple different teams
come in and use the arena.

Multiple arms, regimens are, are
enter into this standing platform.

A basket trial is similarly an
arena where there are multiple

distinct patient groups.

So for example, in this limb-girdle
muscular dystrophy, there's these

39 known mutation subsets, subtypes.

Uh, within that, you could run a clinical
trial in subtype one, in subtype seven, or

in subtype one and subtype seven, and you
could write up a clinical trial in that.

A, an arena here is one where we're
going to enroll many of these subtypes,

perhaps all of them, and the subtypes get
the opportunity to play in this arena.

Th-- And they come into this, uh,
where patients from the different

subtypes have the opportunity to
be assigned to d- to a treatment.

It might be a single treatment that's
exploring the multiple subtypes,

or it could be a combination of
multiple arms and multiple subtypes.

So the basket trial is one where the
arena exists for multiple subtypes to, to

play in the game, and it might be that at
some point you stop enrolling a subtype

where other subtypes continue to enroll.

So these are the two general types,
uh, of arenas that are out there,

and of course, it can be both.

I show a picture of the
Woodcock, Lavange paper.

This is, uh, 2017 New England
Journal of Medicine: Master

protocols to study multiple
therapies, multiple diseases, or both

Now, master protocol.

So then I, the, I have a slide
that describes master protocols.

So master protocol is a document.

It is a protocol, and it generally
is attached to one of these arenas.

And it's a very modular document.

So in the example of a platform trial,
it describes the rules of the game,

but it doesn't list any treatment
arms within it because those are gonna

come and go, and it's modularly set
up that you can plug in a treatment

into the trial in a modular document,
and then you can pull that out and you

haven't changed the master protocol.

It's built to do this.

So you could start with three treatments
at the beginning of the trial, beginning

of the existence of this arena.

One of them leaves, it stops enrolling,
patients are followed up, but eventually

this, this, uh, appendix, uh, the FDA
guides refers to this as a sub-study.

Don't like the term to that, but, uh,
by the way, that was out of character.

I wouldn't have said that to the
patient organization, uh, within it.

But the, this document, this,
this appendix comes and goes.

And then you might add in a
fourth one, and you plug it in,

and it's plug and play within it.

That master protocol is a,
a document that allows that.

Same thing in a basket trial.

The, the, the master protocol
describes that there will be multiple

subtypes, and they will, they will
be there as part of an appendix.

And those could come and go.

So if you have a single treatment arm in
a basket trial, you might start enrolling

in seven subtypes, and at some point you
stop enrolling the subtypes, and they're

pulled out of, of the master protocol And,
uh, again, to describe a trial as a master

protocol, I think is, is vague because
it could be multiple of something or

both within that basket trial or platform
trial is more descriptive of that.

But-- And the master protocol
is a document that drives the,

the, the rules of the arena.

Okay.

I give an example of one of these,
and it's a brilliant example.

The I-SPY 2 trial, neoadjuvant
breast cancer started enrolling

in two thousand and ten.

And, uh, I-SPY 2 there, there
are subsequent e-evolutions of

this where the, the trial has
largely become a smart trial.

I'm unfamiliar with changes in the
last couple years, but was involved

in this trial for the first thirteen
years, I believe, of it, and I believe

twenty-seven or twenty-eight different
arms entered into this single arena.

And on the I-SPY 2
website, ispytrials.org,

you can see a picture of this with
each arm coming in during the course

of it and the overlap of the arms.

And it's brilliant because you can take
any particular time, two thousand and

fourteen, and you can see the arms that
were enrolling patients at that time, and

then two thousand and seventeen, the arms
that were enrolling patients at that time.

And you always have this overlap of
the different arms during the course

of it, and you can see h- what this
looks like for different arms to

enter into this arena And I show
publications that came out of this.

So this is a brilliant example
of a, of an arena, a platform

trial, uh, the I-SPY2 trials.

I show them an example of this.

Then I show an example, a high-level f-
view of multiple arms coming in, and I

try to give a picture of the important
stakeholders in this trial and what's

different about a trial like this.

So what is the view for a treatment arm?

So just thinking the, the view in
this, this, uh, arena of an arm

coming in, and what does it look like?

An arm enters.

I do talk about how much faster it
is to time to first patient in here.

Uh, but an arm enters and it enrolls,
and the arm may enroll where, where the

desire is, say, to get 100 patients on the
treatment and 100 on a control or placebo.

That arm might enroll 100 against 33,
three-to-one randomization, because

there are two other arms that are also
enrolling three to one in the trial.

So this arm three that we're looking at
the viewpoint of arm three enters into the

trial and starts enrolling three to one.

The patients now are randomized
between that arm three or the other

two arms that are currently enrolling.

And then if randomized to
arm three, they carry out and

they're assigned to arm three.

They take arm treatment
three or three's placebo

And you can then show,
well, okay, what happens…

Now we'll come back to patients.

I, I started to go into pa- a patient
experience there a little bit, and, uh,

uh, my intent was to talk about that arm.

That arm may enroll 100 patients
on its, i- its treatment, active

treatment, and 33 on its placebo.

Then when it does the analysis of was
it effective, safe and effective, it

could be using 100 on its arm, 33 on its
placebo, and then another 67 on other

placebos that were eligible for that
arm but were assigned to other placebos.

That arm gets 100 patients on its
active, 100 placebos for determining

whether it was safe and effective.

Its, its analysis reads out.

The sponsor of that gets the
data on the 100 patients.

They get the data on the 100 controls,
probably more controls than that,

some that came in before that arm
within that, and gets to carry out

its drug development within that.

And so the arm gets the same data
that they would if they were to

go out and run their own trial.

Perhaps some on a different placebo, and
that's a, that's a scientific discussion

of does the mode of administration
of placebo, is that something we're

concerned about, uh, in the disease?

I think in limb-girdle muscular
dystrophy, it's rarely something

we would ever be concerned about.

Uh, and, and the drug
development decision can go on.

So what's the view to a sponsor?

Now, the important thing is, is this
a better place for a sponsor to be?

So you describe they, they might
fund 133 patients, but they get

the inferential strength of more
than 200 patients within that

They could build their own trial.

They could build their own arena.

It might take two years before
they're enrolling patients.

In the I-SPY2 trial, it was four months.

Yes, we'd like to get in the trial.

Their appendix was written.

It was put in IRB approval, uh, of
just that appendix because the master

protocol's already been approved.

It, it plugs in.

The database exists.

Uh, everything about this
trial already exists.

You're plugging in, and it's ready to go
for, for that arm to enter into the trial.

So it's faster, it's smaller.

We're already enrolling at the different
sites, so you don't have this long

sites being entered into the trial.

Uh, and then, of course, you're
not taking this down at the

end, and so it enrolls in there.

So you're gonna roll faster, you're gonna
get in sooner, it's going to be smaller.

Inference is going to be
as good, if not lar…

uh, better, uh, within this.

So can you make that argument
to a particular sponsor?

Presumably the cost, you're sharing the
cost of the CRO, of the database, of the

contracts at sites, all of these parts.

It's likely cheaper for the sponsor.

So you can lay out what does it
look like to a sponsor Then, uh,

in the process of this, you can lay
out what does it look for a site?

And I'll kinda go fast on the podcast
here of, uh, imagine a site now is a site

in this arena, and every arm that comes
in, they get to play a role in that arm.

And what's gonna happen to their patients?

A single contract for
multiple arms to come through.

You don't have to…

Site doesn't have to be going
to look, should we wanna go

in this trial or that trial?

The long time it takes to
get this approved at it.

It's already up and running.

It's perpetually, and
now new arms come in.

So this me- much of this can be
a description of is this a good

place for sites to be within that?

The, the other really important
stakeholder is the regulatory.

Is regulatory going to
accept the data from this?

This is an interesting
discussion, of course, over time

as to the acceptance of this.

And, uh, you know, COVID had a huge
impact on the regulatory understanding.

We now have guidance documents.

Regulatory, this is a positive
thing, and they're generally

very positive, of course.

Now, it depends on what you're doing
in the trial as it does with any trial.

Are we using non-concurrent controls?

Are we using historical controls?

What's the endpoint we're using?

Um, w- are we using different modes
of administration of a control,

and does that matter aspect of it?

Then, of course, the
stakeholder of the patient.

What's the view of a patient?

So describe if a patient is
interested in entering into a

platform trial, you would have…

You would know th- maybe the two
or three different arms that are

currently enrolling in the trial Now
the patient is generally randomized

to the cohort for one, two, or three.

The patient can't say, "Oh, I
want the opportunity for three."

And these platform trials, it's generally,
at least I'll refer to that as a, as a,

an umbrella trial where there are multiple
trials and you get to pick which one

you go in, but there's a infrastructure
to them, as opposed to a platform trial

where the trial assigns you to the cohort.

And the h- And the benefit of that now is
that you've been randomized across these.

It enables each arm to use the
placebo from the other arms.

Direct comparability to the,
the placebos on the other arm.

So a patient comes in and
there's master protocol level

inclusion/exclusion, and a patient is,
uh, i- is, is the patient eligible?

And then there's a randomization to,
uh, a consent is gotten to randomize

across the arms that they're eligible
for, and then they're randomized

to a cohort, meaning that they're
randomized to, say, cohort three.

They don't know whether they're gonna
get the active for three or placebo

for three, but they're not gonna
be blinded between two and three.

Uh, it's an interesting exercise for
the podcast people to think of what

would it take to, to blind for that,
and it would be extraordinary and almost

impossible because arms come and go.

And now y- you, you can't blind
somebody when an arm is gone

within that particular one.

So the-- generally, generally being
I think everywhere, a patient is not

blinded that they might be getting three,
but they don't know three are active.

They get the investigator
brochure for three.

They get to read that.

They get to decide and they consent.

Are-- Will you consent to be randomized
to active or placebo for arm three?

There may be in some cases unique things
to that arm in terms of data collection

that this inferential machine works really
well when these are minimal, but of course

it, it certainly can and, and, and is
done within these trials that something

slightly different is done for arm three
Now, they, they consent to this, they're

randomized, and then they're carried
out and their experience looks like, uh,

any other trial where their, their visit
schedule, data collection, and there's

nothing, um, untoward or more complicated
within this than a usual clinical trial.

Now, they, they go through the
length of this trial, suppose a

52-week trial, they carry through it.

That, that cohort might even have an open
label extension where all of the actives

or placebos are moved on to it, maybe not.

There may then be the opportunity, if
that arm ceases its experience in the

trial, that the patient, this is a, this
is a, um, this is a disease that these

patients live with for many, many years,
um, and they may have the opportunity to

enter into the trial into another arm.

Uh, we, we, we hope that that
arm is successful, open label

extension, but if not, they have the
opportunity to enter into another one.

So they can have multiple
experiences within the platform

trial into different arms.

Then lay out very clearly, sometimes
the thought of the negative here is you

don't get to choose your cohort, you are
assigned within that, and describe the

patient experience Uh, lay out some of
the pluses and minuses, uh, uh, of this

and thinking about at a macro level,
if five different arms are gonna all

run five separate trials, and just for
an example, suppose they all want a

hundred active and a hundred controls,
then you're gonna enroll a thousand

patients, each one enrolling five hundred.

You're gonna enroll a thousand
patients, five hundred on

active, five hundred on placebo.

The macro level of multiple games
being played are five- five hundred

active, five hundred placebo.

Within the single platform, in, in
conducting this in the three-to-one

randomization, uh, aspect of this, it,
it all at one time or three are at a

time, you might enroll a hundred and
thirty-three for each particular arm.

So at the macro level, you enroll
500 actives, the same as the

individual trials, but a 167 placebos.

So 667 patients are enrolled in the single
platform with each of the arms getting

an inference as strong as separate.

Actually, probably stronger.

There's a number of aspects, uh,
of it where the inferences can even

be stronger with a single model
running for all of these use of

non-concurrent controls, different
aspects of this that could be better.

So each arm enrolls 133 It's better.

So macro level, 667 The time it takes
to enroll those patients will be

shorter, the time for an arm to enter.

At a, at a macro level, we are
approaching a disease, and we're doing

better to, to go at that disease.

We're also enrolling far fewer patients
on placebo, which is, uh, something

patients very much desire within
this Okay, so it laid out at this

level thinking of the stakeholders.

The other part that I think is not,
sometimes not discussed enough are

the disease learnings that happen by
having all of the data in a single

platform, and having those placebos
be community-owned that every one

of the arms gets access to this, the
investigators, the experience of this.

These can go into databases.

In the, in the HEALEY ALS trial,
there's a PROACT database.

These patients go back into the PROACT
database for science to move forward.

So not only is the platform enrolling
fewer patients, but all of this is

going to go into the disease learning
to understand the variation across

patients, to understand the, the behavior
of endpoints, to the disease learnings.

If they're five separate
trials, this may never happen.

Again, we may never get the ability
to watch what happened in the trial.

Here, there's much better
disclosure on this.

Okay.

So these are-- this is the discussion
I, I, I presented to patients.

I went through a number of examples,
and I won't do this here, but

the HEALEY ALS trial, GBMAGILE,
uh, pa- uh, Precision-PROMISE

in pancreatic cancer, and showed
examples of what this looks like.

Then talk about a basket trial, really
important for limb-girdle muscular

dystrophy, given this is really
an inherent thing to this disease,

showed the ROAR trial, if you're
interested in looking at the example.

This is an oncology example where
many basket trials come from.

And describe what this looks
like, uh, at that level.

And then I presented really a platform
trial that has a basket aspect, that

when arm three comes in, it might be
that they're really interested in one

through ten, arm two is interested
in six through fifteen, and so on.

So it might be arms are really targeted
to different groups, or they go into all

the groups and they're interested in pan
limb-girdle muscular dystrophy, and they

want to enroll in all the groups This is
about patients, again, thinking about…

And there are, there are patient
organizations, and they were there

at the talk I gave, that represent
one of these 39, and the patients

that have a particular mutation type.

And it's-- And of course, the whole,
the whole thing of 39 is rare.

Any one of these can
be really, really rare.

And very rarely are these tr- put
into a trial, like maybe even never.

Now this is an opportunity that
these patients get in these trials.

Yes, we might only enroll nine of
these for a treatment, but we can

still make inferences about that.

We can do Bayesian borrowing across
the groups for any particular

drug, and this is better for the
patients within that subtype.

It's better for the arms
to get that opportunity.

An arm might be approved pan limb-girdle
muscular dystrophy as opposed to the three

that they enrolled Within that setting.

Uh, it might be that the, the incredible
opportunity for, uh, maybe one arm

to go into four pretty rare ones,
where for them to start a trial and

create 40 sites to enroll these four
subtypes, they would never do it.

They would never take that

opportunity.

Now the opportunity's there,
and it might be a really small

step to take the opportunity to
enroll those four small subtypes.

And it may be a shot never taken,
which again, thinking about at the

level of the patients, these patients
want to be in clinical trials.

They, they want to for…

Really, I, I am speaking for them.

I, I, I don't need to speak for them.

Um, but the opportunity for
them to get into trials.

But of course these are genetic diseases
where they see their, their children

and their children's children with
these same diseases, and they want the

opportunity to move the learnings on for
their subtypes, and of course everybody

within the limb-girdle, um, uh, patient
organization, limb-girdle muscular

dystrophy, that this is something that
they wanna do for future patients as well.

So, um, uh, laying out these
opportunities, I presented some examples.

Uh, the questions were
wonderful, uh, from the patients.

There was of course questions
about using external controls.

Could you do a platform trial where
you're using external controls?

And absolutely, um, uh, you could do this.

Um, thinking about the
possibility that the endpoints are

slightly different across them.

And the, sorry, the analysis
for one particular arm could

use a different endpoint.

Now, what is ideal is you're
collecting a set of endpoints.

Arm three could analyze endpoint
one, arm four analyzes endpoint two.

That's certainly possible.

You're collecting that across.

It's of course much more challenging
if you're only collecting that in the

cohort, uh, and you don't have it for
the other endpoints, um, uh, within that.

They talked about the
consent part of this.

What does that look like?

I think it was very, very positive.

Uh, the patient organizations
were very positive on this.

The huge challenge in all of this
is how do you get it started?

And it's a bigger jumpstart
to build that first arena.

It's easier to build an arena
for one game because you can

customize it for that one game.

Somebody's paying for that.

It's, it's more work, it's more effort
to build one that's modular for--

that's gonna have multiple games being
played, whether those are patient

groups or treatment arms or both.

That's the huge lift, uh, uh, that,
that tends to hold things back.

The HEALEY ALS trial is an amazing
case where they got this started,

and now it's much easier to add arms.

Now it's-- I don't wanna say it's easy,
but, uh, from that, the, the, the arena's

been built, and each game being played
is a much smaller lift Within that.

So that's kind of the challenge that
you're at here, uh, within these diseases.

Okay, so I, I thought I'd give you
a view of this as this in-incredible

opportunity to speak to patients.

And I, I, I, I speak many times to
statisticians, to pharmaceutical

companies about how these are so much more
efficient, these are better, and thinking

about and speaking about the stakeholders,
to speak directly to the most important

stakeholder in this, the patient, uh,
about these trials is this incredible

opportunity, and of course, keeps me
centered as the reason we're doing this.

Does it really make sense to patients,
to hear from patients, to have the

side discussions afterwards about
their experiences in clinical trials,

having the patients be a part of this?

Going forward, if this arena could be
built, it is going to be built in part by

the patients, as it was for Healey ALS.

And I use that as an example because
such a fantastic, uh, example where, by

the way, the Healey name to that trial
was Sean Healey, a patient who thought

this was such an incredible opportunity
for the disease that afflicted him that

he helped fund it, and he put his name
to it, uh, uh, within that setting.

Uh, similar things being done in
other rare diseases as well, uh,

Parkinson's disease, myotonic
dystrophy, uh, other MPS diseases.

It's, it's, it's a really
awesome, uh, opportunity.

But of course, this was a case to, to
go to the patients and see, do they

think this is an opportunity in that?

Uh, so it was a wonderful day.

I thank them all for that opportunity,
and I thank you all for joining me today.

Until next time, we'll
be here in the interim.