The Story of Berry Consultants

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.

Brandon Giella: Hello and welcome
to Barry's in the interim podcast.

This is episode number one, the very
first inaugural episode of the podcast.

I'm so excited to talk to you guys.

I have with me Dr.

Don Barry and Dr.

Scott Barry, statisticians, experts,
thought leaders, and founders of Barry

consultants, and I want to get into all
of that, but we are kicking off this

show because This is the 25th anniversary
coming up for Berry Consultants and it

is such a momentous occasion And I'm
so glad to have the founders and uh and

experts in this arena because you guys
really are the trendsetters in Statistical

trials, clinical trials, adaptive trials.

I want to get into all that We're
celebrating this year with so many

new projects and just a lot of
excitement, um, coming up in 2025.

So this podcast is a part of that.

Um, and so as such, I would love to hear
the story of Barry consultants and I

have, uh, the best people to do that.

Um, and so I would love to hear and
share with our audience and our network.

What.

How did Barry begin?

What was the motivations behind the firm?

Um, who impacted you in
those early beginning stages?

What kind of conversations do you have?

What are the stories, what
challenges did you encounter?

And then toward the end of the show,
I would love to hear what's next.

You know, where are we going from here?

And so Don, you, you are the founder,
you are the, the man of the hour.

I would love to hear.

What was it like when you started Barry?

How did that come about?

And why, why are we here today?

Don Berry: So I was a graduate
student at Yale University,

Scott Berry: Heh, heh,

Don Berry: Frank Anscombe, a very
famous statistician, and my wife

was pregnant at the time, and, um,

Frank asked me, he said, so how
are you going to pay for this?

And I said, I have no idea.

Uh, and he said, well, When your
wife is in the hospital and you

get the bill, bring it to me.

I brought it to him, and he had
a grant that paid for the, uh,

the hospital and everything.

I don't know how he did it, but he did it.

Um, and I count that as Scott's
very first grant, because

Scott was the, The, uh, baby.

And of course, you see him
now, he's no longer a baby.

Um, but he still, is, uh, thinking in very
naive thoughts and very creative thoughts.

And that's really the
beginning of Berry Consultants.

Just a word about, uh, the interim.

In the interim between that and
when we founded Berry, Berry.

wasn't a particular time
that Barry came into being.

It was, uh, I was known as a radical
reactionary, crazy guy, loosey goosey,

some people called me, trying to change
the way we looked at medical research.

And so it was my touring around, Brandon,
I use the term, Johnny Appleseed, around

the country and pharmaceutical companies
and wherever, whoever would listen, I

would, give them a talk and, spur them on.

Uh, just an anecdote about that,
along the way, I was in Washington, D.

C.,

on the metro, and, uh, coming back
from one of my talks, And a guy came

up to me, it turned out to be he was a
statistician at the FDA, and he said, Dr.

Berry, uh, I'm a fan of yours,
every time I hear you talk, I

become a Bayesian for five minutes.

Can you work on a
sustained release version?

So, uh, I did.

But, uh, when did Barry start?

It started like 2001.

Scott and I happened to be in Texas.

and I had been doing consulting
and, Scott, take it from here.

Scott Berry: So, 33 years after Frank
Anscombe's grant, uh, with that,

so, I was on faculty at Texas A&M
Statistics, and, doing research, very

much enjoying it, teaching, and a number
of projects came through Don, who is

at MD Anderson Cancer Center, building
clinical trial designs, and so I was

helping with those, and I loved them.

Everyone turned out to be a puzzle.

I love board games.

chess, bridge, settlers of Catan, and they
all have, what's the problem to solve?

And all of these had a problem to solve.

We were using all of The techniques,
the statistics, the science to solve

these and everyone I worked on, I loved.

And so after five years, I
wanted to do more of that.

so we talked about what
would this look like?

Suppose we started a business to,
to, design better, smarter trials.

And so in 2000, June of 2000, we
decided I would leave academia and

we would start Berry Consultants.

I was the full time person.

Don was at MD Anderson Cancer Center.

And so June 1st, 2000, we started
Berry Consultants with, really No

idea what was going to happen, but
we had shared interest, desires

to design smarter, better clinical
trials and really enjoyed it.

And so we started it, the two of us.

Brandon Giella: And
how's that been for you

Scott Berry: Uh, fabulous.

Still, still, still going on 25 years and
still very much enjoying every day of it.

Brandon Giella: I love that.

I love that.

Well, I'm curious, what was, I
mean, starting your, your first, uh,

starting your, your company like this.

And, and I know you, you had, you
know, clients that were needing

your expertise, very complicated,
sophisticated trials, you know, big,

big money involved in some of the,
um, you know, pharmaceutical companies

or medical devices, things like that.

Um, what were some of the challenges
that you encountered early on?

Thanks, Tom.

And how did that shape Barry
now into to who it's become?

Scott Berry: Yeah, it's, uh, we've been.

Part of what we're trying to do is, is
through change, it's always changing

the status quo when you're trying to
do something different and just to set

this up, the standard of a clinical
trial is, um, the standard thing is

let's enroll 200 patients, enroll the
trial for four years, gather the data

and look at the data after four years
and see if we've answered the question.

Invariably, you look at the
data and you say, Oh, Shoot.

I wish I wish we'd have
done something different.

I wish we to change the patients, change
the treatment gone longer gone shorter.

Uh, we could have done a better
trial, but the sort of not looking

at the data for four years.

So Let's design smarter trials.

We were, we were doing
something different.

Uh, and early on, what was great is early
on Don's going around as Johnny Appleseed

was saying, we need to do this different,
uh, and saying we need to do it different

and publishing about doing it different.

And now we were going to do it different.

We were creating a company that was
going to overcome all of the challenges.

The regulatory challenges, the
operational challenges, the people

that say you can't do that, the people
who don't do this as normal things.

So, any time you're saying you're going
to do it different, you have to sort

of tear down all of those barriers.

And we had a number of
people who believed in this.

Believed we are doing good
things, believe this was better.

So we had those, um, you know, the,
the, the people who saw differently,

that were a huge part of, uh,
thinking that what we were going to

do was, was the right way to do it.

Don Berry: So the believers,
yes, um, there were, uh,

people that were on our side.

People that were on our side
didn't know they were on our

side until we talked with them.

And Brandon, you said, um, that
companies needed our expertise.

They didn't know that.

Brandon Giella: hmm.

Don Berry: Uh, they had to be persuaded
that there was something better.

Um, and some of them took, uh,
uh, uh, you know, first steps.

Um, I mentioned to you a stroke
trial that I designed for Pfizer.

Uh, and there was one very key
person, uh, in the audience for a,

uh, short course that I was giving,
it was Mike Krams, who was an MD.

Um, this was in, in the United
States, but he was, he was based

in Europe and England, in fact.

It's still part of Europe, uh, and, um,
he asked me if I could design a trial for

him, um, and I did, uh, and just to talk
about the, uh, the various companies.

After, this was, he was for, working
for Pfizer, uh, after Pfizer, Uh,

ran the trial, they decided that it
was a terrific thing and the, the

upper management decided it was a, a
terrific, uh, trial that answered the

question as efficiently as possible.

Turned out to be a very negative
question, which was good because it

got rid of the drug and it's, you
know, no longer using, uh, resources

that there are many, many drugs in
the world that deserve them better.

Um, and the company, Pfizer, wanted to
patent the algorithm that ran the trial.

And, um, the reason they wanted to do
that is not necessarily they wanted

to do additional trials like it, but
they didn't want anybody else to be.

They wanted to own it.

Luckily for me Uh, our
contract said that I own it,

Brandon Giella: Hmm,

Don Berry: them.

So they weren't able to patent it.

Um, and it's, it's that kind of thing
that we were really providing the service

that they didn't know they needed, but
we knew the world needed, needed it

because we were going to change the world.

And it was that kind of.

dedication, and almost a religion,
um, that we have the answer.

And the point was to
convince other people.

And slowly, but surely, we've done it.

So we now, I don't think Scott
mentioned, but now we have like 35, uh,

people, uh, mostly PhD statisticians.

who are carrying the ball for us.

And we're really, uh, you know,
fast forward, changing the world

because of the, some of the trials
that we, uh, have designed and

that have been very successful.

Scott Berry: So that was the Aston stroke
trial and that was actually pre Barry.

That was probably one of the
things that helped get us started.

Um, and another interesting person
involved in that was Tom Park.

So Don had designed this amazing adaptive
trial with many, many looks in the trial.

Uh, uh, 16 doses and to
implement that was brand new.

Don Berry: Daily, daily randomization
changes in the algorithm

that was picking the doses.

There were 15 positive doses.

Scott Berry: So even today that would
be considered a very innovative trial

design and Tom Park was at Tasella at
the time and helped us implement that.

He's now at Berry and he runs
our software division and

helps us overcome the barrier.

So we've run into a number of
people like this that have a huge, a

huge, uh, uh, impact, uh, in Berry.

Brandon Giella: Well, I want to pick up
on a thread that you were talking about,

Scott, that you you're, you know, it's
always about change and kind of disrupting

the status quo, if I could call it that.

And so you guys have touched on the
different ways that you do trials

and, and, and how that's come about.

But I'm curious, both of you, how would
you describe succinctly or briefly or how,

yeah, just in a, in a, in a short way, how
would you think, how would you describe

the difference That Barry is compared to
the traditional way to do Either clinical

trial designs or the way that you run your
firm or their culture Like if i'm let's

say i'm at pfizer And i'm looking across
the the sea of you know statisticians

that I could interact with across a
different Um, you know range of firms and

different people that are doing that and
what is it that makes barry so different?

Scott Berry: Oh, it's a great question.

I think Traditionally, there's not a ton
of statisticians who have a huge impact

on the trial design, really focusing
on what's the question being asked.

And so we dive in and thoroughly
investigate that and show through

clinical trial simulation, which
is one of the things we've sort

of helped developed over 25 years.

What the trial is going to do.

It's almost like, um, uh, you use
simulation to design an airplane,

to design the most efficient cars.

We now do it with clinical trials.

We're able to show them
what it looks like.

Here's alternatives.

You could shape this this way.

You could do it that way.

And the wonderful thing about it is
if you're trying to get somebody to do

something different, you can't tell them.

Do it differently.

We, we, we're named consultants,
but if you go in and say, I think

you should do this, it doesn't work.

What they want to see is they want
to see and compare and you show them,

if you do it this way, you get this.

If you do it this way,
you get this or this way.

And then, Oh, I want to do that.

And, and I want to do that.

And I would do that and they get
to this place and it was them.

They're making the decisions,
they're driving it.

They're the, they're the agent of change.

We help show them that, you know,
we don't tell them what to do.

We show them the possible ways
and the scenarios, and it allows

them to be the biggest advocate
for the direction they're going.

Brandon Giella: So

Don Berry: So, Brandon, um,

Somewhere in this conversation, we want
to talk about the FDA because it has been

essential for our existence, essential for
what we do, back in 2010, the FDA issued

a guidance for adaptive design that, uh,
that said, simulation for controlling

type 1 error rate is little understood.

And, um, the Bayesian approach,
which we take in statistics, Uh,

is heavily dependent on that very
thing, about simulating a trial.

Simulating a trial, what we
mean is, we make an assumption

about what the truth is.

Is the drug good or not?

Uh, and if it's good, how good is it?

And we simulate a trial to show
that regardless of what the truth

is, we're going to find the truth.

But We have complicated trials.

The ASTIN trial is updating daily.

The results coming in
by fax in those days.

Um, and, um, the, the simulations,
when we do 100, 000 of them, sort of

guarantee that we've got the answer.

That it's, it's doing, you know, it
has a 5 percent error rate, but we

demonstrate that from the simulations.

And that was a big block for
us, the fact that the FDA said

they didn't understand it.

And they weren't convinced that
it was, uh, the right thing to do.

Or that it was giving
you the correct answer.

Um, and so we were lucky.

Another person to mention in this
is at the FDA was Lissa LaVange,

who took over, uh, shortly
after that guidance was issued.

Um, and was head of the statistics
group at the CDER, the Center

for Devices, Center for Drugs.

And, um, she arranged for bringing
in the statistical community over the

next 10 years to help them understand.

And in 2019 they issued another
guidance for adaptive clinical trial

design That said, simulation is okay.

And that was a really big thing for us,
because we have this software that you,

that Scott mentioned that Tom Parks is in
charge of, um, it's FACTS, it's called,

uh, uh, Fixed and Adaptive Clinical Trial
Simulator, and it can be used to simulate

any trial, not just the ones we design.

But, it's out in the world and
people are using it to design

these trials and now, I've said
too much, uh, Scott, back to you,

Scott Berry: No, it's, it's, it's
one of the list has been a huge,

a huge benefit to the scientific
community, medical community, for sure.

Brandon Giella: Well, I'm really glad
for people like you because when I

was getting my MBA in finance, I had
a statistics class in my curriculum.

Never taken statistics before, and
I'm proud to announce that I got a C.

So I'm really glad.

So I'm glad people like you exist.

I can't do it.

Don Berry: no, no, you, if you, we
hate it when somebody says, uh, you

know, I talk about clinical trial
design, I'm helping advise a company.

And there's this old geezer, maybe
even older than me, um, who says, I've

designed clinical trials for 30 years.

I know how to design a trial.

Immediately, you know that
his head is blocked off.

He's not going to listen
to anything you say.

We don't like people who think they
know the answer because they don't.

Uh, and they don't know what the
best way is to do a clinical trial.

That's what we deliver to them.

is you're going to take this
trial, the one Scott mentioned,

you know, 200 patients, four years.

We're going to give you the
answer in six months, maybe.

We don't know.

Maybe we have to go longer, but we
can give you, in this Aston trial,

as I said, it stopped the trial, um,
as soon as it was allowed to do so.

And it was because it was looking,
it was like artificial intelligence,

it was looking across the spectrum of
the doses, and it started out looking

at the low doses and comparing to
placebo, decided that wasn't good

enough, and it moved up the chain.

It finally started to focus on the
highest doses compared to placebo,

uh, decided nothing was there.

And the good news is for patients
that it started to go to placebo.

Um, namely it wasn't any good.

And that thing saved Pfizer millions
in terms of the patients, in terms

of the patient resources, but also an
uncountable, you know, a priceless.

benefit, that it got them off that thing.

This was a, a particular, uh, uh,
mechanism, mechanism of action, um, that,

uh, the Pfizer and the rest of the stroke
community decided was not the way to go.

Uh, and so it saved, we, we really
don't know how to quantify it, but

it saved, uh, I mean it's almost
priceless what the benefit was.

Scott Berry: Yeah, it's always
hard to quantify time in it.

So the, the, the story of
brain, and so we started off.

So it was Don and me, and we, we didn't
know whether this was going to work.

Um, and we, we were doing a
lot of medical device trials.

There's been a long history of, of
innovative Bayesian things at the center

for devices, Greg Campbell was a huge, uh,
uh, a huge innovator at the FDA for that.

And all of a sudden we had
people that were interested.

in looking at adaptive trials,
looking at flexible trials.

full time employee other than myself.

Jason Connor joined
several years after that.

Um, and we kept getting
more and more interest.

And as Don described, 2010, it was a
little bit where was drugs going to go?

Center for Drugs were
doing a lot of devices.

And the Center for Drugs became
more and more interested.

Pharma became more interested.

And we have now been
steadily growing since 2010.

Uh, within that setting up to now,
as Don described, our 35 scientists

here at Berry Consultants, uh, still,
still growing and still innovating.

Brandon Giella: Yeah, it's, it's such an
amazing story and it's to Don point the

way that you guys do your, your trials
and the way you run your business, it is

so incredibly helpful, not just for the
firms that hire you, your clients and how,

you know, you can save them money and time
and all of that, but also for the patients

themselves and to be, to have the, the,
the specificity, um, that you have for

these trials is, is really amazing.

So how I guess following that point scott,
um, and we'll take maybe the last 10

minutes or so of the show Uh, how would
you describe the current state of barry?

You know, what what is what do you got?

What are maybe projects
you're you're working on?

Um, you know different things that you
guys are thinking through and innovating

And then where do you see barry going
over the next year five years ten years?

Where where are you pointing
your ship if I can ask that?

Don Berry: So, uh, let
me try to address that.

Uh, there's, back around this 2010, um,
I was designing clinical trials for the

cancer leukemia group B in breast cancer.

And, um, uh, a friend of mine that I,
uh, uh, uh, developed a relationship

with was Laura Esserman, who's the
surgeon at, um, at UCSF, University

of California, San Francisco.

Um, and she was, you know, I was
radical, but she was off the deep end.

Um, and, uh, wanted to do things.

That, similar to what I wanted to
do, and we built a trial together.

We couldn't do it through the
National Cancer Institute, which

sponsored these CLGB, uh, trials.

So we went to the Foundation for the NIH,
and we worked with them to fund the trial.

This is a, a public private partnership
with, uh, NIH leaders and industry.

And so we built a trial called iSpy2,
um, that was, had lots of innovations,

had Scott's, um, uh, innovation for using
historical controls that he had designed

when he wrote, uh, uh, an article that
won the, uh, Journal of the American

Statistical Association award for best.

article that year in Applied Statistics,
uh, that used the, in baseball,

Babe Ruth versus, um, uh, who,

Scott Berry: Mark McGuire.

Don Berry: Mark McGuire, um, uh, and
who was better, adjusting for the time.

And so he built this thing that
used things in the past, Well,

we wanted to use controls.

We were randomizing, but we wanted to
keep those controls, uh, randomization,

uh, patients in the trial that we
could continue to use as a, as a basis.

That was just one of the many innovations.

Another innovation, and speaking
to your point about patients,

was we wanted to treat these
women as effectively as possible.

So we built in adaptive randomization.

So, if a drug was doing poorly in
a subset of women, it got lower

probability of being assigned and
soon was zeroed out because there were

other drugs that we were looking at.

So, we ended up over the course of the
next 10 years, um, evaluating 23 drugs, 9

of them went on to bigger things, um, and,
um, the, the, the, the, the, the, the,

Scott Berry: So like, so, so Don,
let me jump into that part since,

since you stole my question.

Let me, let me steal this
from you and take the ball.

So, uh, so the, the, imagine
how different this is.

The, the historical way this
is done is you have a drug a

And you build a whole trial.

It's like it would be like the NFL
deciding every time they're going to do

a game, they build a whole new stadium,
they play a game and they take it down

and then they build a whole nother one.

This is what we do.

So this I-SPY 2 trial was building a
single stadium, a single protocol and

move 25 drugs through the protocol.

Using the same stadium, comparing
them, using common controls.

If you have four trials, all enrolling
50 percent controls and you move

them all together and have one arm of
control, you've now saved patients.

You've saved time.

So this was a landmark trial in
2010 in breast cancer, and it's

called a platform trial and it was,
it was as it was off the deep end.

Fast forward to COVID.

The pandemic hits and we're trying
to figure out how to treat COVID and

you can't wait two years to build
a clinical trial to investigate

a single drug, tear it down.

So platform trials globally that copied
I-SPY 2 were used to treat COVID.

Everything we learned how to treat
therapeutically COVID came from platform

trials that really the grandfather of
that was this crazy idea of I-SPY 2.

So what is that?

Now we're doing, we're doing trials
of that in ALS, in glioblastoma, uh,

in, in a number of rare diseases.

And it was all born out of these crazy
ideas of Laura, Laura and Don in 2010.

So that's where we're going.

But we don't know what that innovation is.

We have all kinds of ideas of
how to make this more efficient.

And we have to, it's too expensive.

We've got it.

We there's so many amazing medical
therapies out there to be using 50 year

old trial designs doesn't really work.

So we want to be on that cutting edge.

We want the next eye spy, the next thing
to, to do better science in this for

patients for, for better treatments.

We don't know what it's going to be, but
we want to be on that sort of cutting

edge and, and, and, and doing this better.

Don Berry: So, one important, important
note to that, um, in the glioblastoma

study, uh, it's a phase three trial.

That means that if a drug does
well in, in, in, uh, GBM Agile,

we call it, um, we take it to the
FDA and the FDA will approve it.

We worked with the FDA.

Uh, really intimately, uh,
interacting with them and building

the trial that they would okay.

So, and they sent us a letter saying,
if you build the, if you have an arm

that comes through, a treatment arm
that comes through the trial, uh, that

graduates and is confirmed within GBM
Agile, within this, uh, uh, adaptive

platform trial, we'll approve it.

So, and that was a sea change.

iSpy 2 trial was great.

It's known throughout the world
as being an innovative trial that

people want to, uh, uh, uh, imitate.

But it's a phase 2 trial.

It evaluates these drugs and then
moves them, decides which drugs

can go into a phase 3 trial.

GBM Agile does the phase
3 trial within itself.

Moving seamlessly from one part of
the drug development spectrum, namely,

uh, is it, is the drug working,
to the, uh, confirmatory phase.

And this we owe to the FDA, who has
just been enormously helpful to us

and to, you know, uh, medical research
and drug development generally.

Brandon Giella: I love these stories.

I love hearing that something that
you designed and brought into the

world, You know 15 years ago is still
impacting the world in huge ways

like covid and some of these other
major Major diseases and issues.

It's such an amazing Legacy
if you want to call it that

Don Berry: Just to mention Alzheimer's and

Scott Berry: heh

Don Berry: Uh, and just to mention
diabetes type 2, and just to mention

the new craze, the weight loss
craze with the GLP1 agonists, um,

we designed the first trial for, uh,
uh, Eli Lilly's, uh, dulagletide,

trulicity, uh, that became their best
selling drug and, uh, type 2 diabetes.

But in that trial We also saw, um, you
know, it's a 5 billion drug, has been, uh,

and, and was the original approved, uh,
GLP 1 agonist, and it showed in the trial

we designed a benefit on weight loss that
was essentially the same benefit that they

eventually saw when they got Trulicity
approved for weight loss, and now the,

Well, you know the rest of the story.

The world is getting skinnier
all the time, in part because

of these, uh, Glyphon agonists.

Brandon Giella: quickly on its axis
uh, no, I So if if I could summarize

all of this or if I guess if you could
summarize all of this what is one thing?

in one minute That you two are very
excited about in this coming year

the 25th anniversary 2025 Barry
is going through a lot of change,

very exciting things happening.

What is one thing that you want
listeners to take away from, or one

thing that you wish that they knew
about Barry in this upcoming year?

Scott Berry: Oh boy.

Uh, that's a great question.

I'm most excited about all
of the new statisticians and

scientists we have at Berry.

they're brilliant.

They've, they're coming into
this with all new ideas.

They're also their
communication is, fantastic.

so I'm excited to see what they do.

it was incredible, over these 25 years,
but, it's, they're much smarter than I am.

and so I'm really excited to see what
they're going to do with clinical trials.

Brandon Giella: There are at
least B students for sure.

Scott Berry: Yeah.

Yep.

Don Berry: Jason Connor, who, Scott
mentioned was our first employee,

went off on his own, partway along.

I told Jason That my
loyalties are with people.

My loyalty is with you.

and Jason Kept, has his own,
company and he consults.

He's part of what I call the Berry
Diaspora, who is spreading the word.

we can't change the word,
the world, in Austin, Texas.

we want to change the world in,

Peoria, in Australia, in China, in
Europe, and we're partway along with that.

GBM Agile, for example, is accruing
patients in Australia and in, Europe,

in addition to the United States.

that's, that's the exciting thing, is to
change the world, one person at a time.

Brandon Giella: I love that.

Doctors Barry.

Don and Scott, thank you so much
for the work that you're doing.

And I'm very excited about the 25th
anniversary this year, about next year and

all the things that you guys are doing.

And I'm excited for the show.

I think it's gonna be a lot of fun
and I can't wait to have other guests.

I know we've got 30 something
episodes that we're kind of

planning out and working on.

So I'm really excited to keep this going
and we'll see you again on episode two.

Scott Berry: Appreciate it.

Thanks, Brandon.

Don Berry: thanks, Brendan.

Thanks, Scott.