{introduction music}
{introduction music}

Well thank you Tom Denny and Mike Datto for
Well thank you Tom Denny and Mike Datto for

joining us today. Tom is the chief operating officer of
joining us today. Tom is the chief operating officer of

the vaccine institute and Mike Datto is in charge of
the vaccine institute and Mike Datto is in charge of

director of the clinical labs for the
director of the clinical labs for the

health system and associate vice president in the
health system and associate vice president in the

health system. So today we really want to talk a little
health system. So today we really want to talk a little

bit about testing; the kinds of tests, their efficacy,
bit about testing; the kinds of tests, their efficacy,

their turnaround time. So maybe we can start, Mike can you
their turnaround time. So maybe we can start, Mike can you

tell us a little bit about you know we hear about uh deep nasal
tell us a little bit about you know we hear about uh deep nasal

swabs at home nasal swabs uh etc can you tell us a little bit
swabs at home nasal swabs uh etc can you tell us a little bit

about uh what you're actually using for tests in the clinic
about uh what you're actually using for tests in the clinic

and what what if any alternatives there are?
and what what if any alternatives there are?

Sure I'd be happy to and thank you Dr. Kornbluth for the opportunity to meet
Sure I'd be happy to and thank you Dr. Kornbluth for the opportunity to meet

and talk today. So uh in our health system
and talk today. So uh in our health system

laboratories we are doing predominantly
laboratories we are doing predominantly

what's being referred to as deep nasal swabs, or nasopharyngeal swabs. They
what's being referred to as deep nasal swabs, or nasopharyngeal swabs. They

are uh one of if not the most sensitive
are uh one of if not the most sensitive

sampling methods to detect coronavirus.
sampling methods to detect coronavirus.

There are other sampling approaches for example
There are other sampling approaches for example

sampling of the anterior nerves or front of the nose,
sampling of the anterior nerves or front of the nose,

There are saliva tests that can be performed
There are saliva tests that can be performed

for saliva and nasal testing these can be
for saliva and nasal testing these can be

either administered by a healthcare professional or self-collected.
either administered by a healthcare professional or self-collected.

In terms of sensitivity for viral collection, like I said, nasopharyngeal
In terms of sensitivity for viral collection, like I said, nasopharyngeal

is the highest sensitivity, followed by
is the highest sensitivity, followed by

a provider collected nasal swab, followed by self-collection of saliva or
a provider collected nasal swab, followed by self-collection of saliva or

a nasal swab. So in the health system where we're trying to get the
a nasal swab. So in the health system where we're trying to get the

highest probability of getting the detection of virus,
highest probability of getting the detection of virus,

we stick to what is arguably the least comfortable
we stick to what is arguably the least comfortable

sampling approach which is the nasopharyngeal swap.
sampling approach which is the nasopharyngeal swap.

So Tom am I correct that you've done a little bit of calibration on the
So Tom am I correct that you've done a little bit of calibration on the

saliva testing and I have concerns about that being a
saliva testing and I have concerns about that being a

viable root forward at this moment? Yeah
viable root forward at this moment? Yeah

and thanks for having me here too Dr. Kornbluth. We
and thanks for having me here too Dr. Kornbluth. We

were hoping to develop an easy sample collection that would
were hoping to develop an easy sample collection that would

make this less intrusive for for everyone
make this less intrusive for for everyone

and we've been sampling saliva for a number of months now.
and we've been sampling saliva for a number of months now.

The best that we're seeing is about an eighty percent
The best that we're seeing is about an eighty percent

to eighty-three eighty-four percent agreement with the
to eighty-three eighty-four percent agreement with the

as the suave that um dr dotto uh compared to the deep
as the suave that um dr dotto uh compared to the deep

nasal uh swab. It seems that it's the
nasal uh swab. It seems that it's the

again the most accurate, if you will, or concordance between the swab
again the most accurate, if you will, or concordance between the swab

in the in early infection. So what you would call acute infection
in the in early infection. So what you would call acute infection

maybe the first 10 days or so after 10 to maybe 15 days out
maybe the first 10 days or so after 10 to maybe 15 days out

you start to see a wider spread of discordance between
you start to see a wider spread of discordance between

saliva and and a nasopharyngeal slop, in our experience.
saliva and and a nasopharyngeal slop, in our experience.

And I think I've talked to a few colleagues around
And I think I've talked to a few colleagues around

the country in Seattle and Pittsburgh and others and they're seeing
the country in Seattle and Pittsburgh and others and they're seeing

pretty much the same rates as we are. So when, Mike,
pretty much the same rates as we are. So when, Mike,

when there's a a positive test in the hospital
when there's a a positive test in the hospital

obviously you're dealing with possibly false positives, you get
obviously you're dealing with possibly false positives, you get

negatives you may do false get false negatives. Do
negatives you may do false get false negatives. Do

you have to confirm tests over and over again or no?
you have to confirm tests over and over again or no?

No so so our false negative rate is quite low.
No so so our false negative rate is quite low.

Our providers tend to be very cautious and
Our providers tend to be very cautious and

on occasion disbelieve negative results so we have a lot of data on
on occasion disbelieve negative results so we have a lot of data on

repeat testing of people where the clinical suspicion is a bit higher.
repeat testing of people where the clinical suspicion is a bit higher.

And they do a second test and it's a very small percentage of times,
And they do a second test and it's a very small percentage of times,

maybe two to three percent of the times, where we get a discordant from a
maybe two to three percent of the times, where we get a discordant from a

negative to a positive test. Now that isn't to say that the
negative to a positive test. Now that isn't to say that the

sensitivity of our testing is 97 to 98 percent of 98 percent. Some patients for
sensitivity of our testing is 97 to 98 percent of 98 percent. Some patients for

whatever reason don't shed a lot of virus in their upper
whatever reason don't shed a lot of virus in their upper

respiratory tract and can be infected but have negative uh results by
respiratory tract and can be infected but have negative uh results by

our testing. So false negatives do occur.
our testing. So false negatives do occur.

All of the lab based platforms that we use have a sensitivity
All of the lab based platforms that we use have a sensitivity

above 90 percent. The ones that we use at uh highest capacity, the avid m2000 and the
above 90 percent. The ones that we use at uh highest capacity, the avid m2000 and the

avid allenity have a sensitivity probably around 97 percent.
avid allenity have a sensitivity probably around 97 percent.

And the point of care testing that we use has the lowest sensitivity between
And the point of care testing that we use has the lowest sensitivity between

85 and 90 percent. But that has the benefit of having
85 and 90 percent. But that has the benefit of having

immediate results. Right, right. So are we doing any
immediate results. Right, right. So are we doing any

antibody testing at all? I know, are we doing any in the clinic from
antibody testing at all? I know, are we doing any in the clinic from

Mike and then Tom I know that you've done a little bit of also experimental
Mike and then Tom I know that you've done a little bit of also experimental

work with antibody testing. So are we using it
work with antibody testing. So are we using it

in any way at all at this point? Great question uh our laboratories
in any way at all at this point? Great question uh our laboratories

have done a full validation of an antibody test which involved testing
have done a full validation of an antibody test which involved testing

several hundred uh clinical specimens. The performance of the assay is actually
several hundred uh clinical specimens. The performance of the assay is actually

pretty good in terms of uh sensitivity and
pretty good in terms of uh sensitivity and

specificity. It's not as good as our
specificity. It's not as good as our

RNA based testing for detection of bars. But we have not activated that test
RNA based testing for detection of bars. But we have not activated that test

because clinical demand to this point has been pretty low.
because clinical demand to this point has been pretty low.

It's not a good , the test itself uh can't be used
It's not a good , the test itself uh can't be used

as an indicator for resistance to re-infection because of
as an indicator for resistance to re-infection because of

false positive rate. And the test is negative early in
false positive rate. And the test is negative early in

infection. So it can't really be used as a
infection. So it can't really be used as a

diagnostic tool either. But Dr. Denny's also done a lot of
diagnostic tool either. But Dr. Denny's also done a lot of

work on this and probably has some other thoughts as well.
work on this and probably has some other thoughts as well.

Yeah Tom? Yeah so we're doing a lot of clinical research
Yeah Tom? Yeah so we're doing a lot of clinical research

uh studies with the antibody assays that we have
uh studies with the antibody assays that we have

and I'll talk about the one of the most exciting ones I think for me
and I'll talk about the one of the most exciting ones I think for me

recently that we've done in collaboration with Mike and
recently that we've done in collaboration with Mike and

the clinical lab. So um in April we were able to do
the clinical lab. So um in April we were able to do

a rapid shear prevalence study of of everybody that showed up at the ER that
a rapid shear prevalence study of of everybody that showed up at the ER that

there was a sample blood drawn and it went to Mike's
there was a sample blood drawn and it went to Mike's

shop. Couple days after that you're able to use that as a residual sample.
shop. Couple days after that you're able to use that as a residual sample.

So we looked to see what percent of that cohort had antibody and it was about 3.5
So we looked to see what percent of that cohort had antibody and it was about 3.5

percent. We just recently repeated that study and
percent. We just recently repeated that study and

um literally within the last couple weeks and the data is just coming out
um literally within the last couple weeks and the data is just coming out

the last couple days and we're about 10.2 percent
the last couple days and we're about 10.2 percent

wow uh for uh those now this time through we did it um
wow uh for uh those now this time through we did it um

a bit differently. We did this as an honest broker approach so that we can go
a bit differently. We did this as an honest broker approach so that we can go

back and not us, but someone else that
back and not us, but someone else that

will go back and look into the medical records
will go back and look into the medical records

and try to understand if these were individuals that were
and try to understand if these were individuals that were

acute infection or if there's no evidence of acute infection and perhaps
acute infection or if there's no evidence of acute infection and perhaps

that represents, you know, an older infection or maybe even an
that represents, you know, an older infection or maybe even an

asymptomatic individual that converted. The other exciting part of it
asymptomatic individual that converted. The other exciting part of it

is we looked at everyone that made IDT antibody in this cohort
is we looked at everyone that made IDT antibody in this cohort

of and looked to see if they were in agreement with a signal of an assay that
of and looked to see if they were in agreement with a signal of an assay that

was developed in the Duke Singapore campus that
was developed in the Duke Singapore campus that

correlates with neutralization and what we found was a very small number of the
correlates with neutralization and what we found was a very small number of the

overall group expressed that neutralizing
overall group expressed that neutralizing

antibody signal. Hmm. So we're taking it except further
antibody signal. Hmm. So we're taking it except further

with David Montefiori who's really one of the experts in the world of doing
with David Montefiori who's really one of the experts in the world of doing

neutralization assays is looking at all those IgG positive antibody cases
neutralization assays is looking at all those IgG positive antibody cases

and looking to see if there is a functional neutralization antibody and
and looking to see if there is a functional neutralization antibody and

we hope to know that another few days. So we're trying to map out just when does
we hope to know that another few days. So we're trying to map out just when does

an antibody response come up, an infection, and then when do you start
an antibody response come up, an infection, and then when do you start

to see neutralization. And then the long-term part of some of
to see neutralization. And then the long-term part of some of

our work is how long-lasting is that? In other words if
our work is how long-lasting is that? In other words if

you start to look at people three months out, six months out, you know,
you start to look at people three months out, six months out, you know,

year out, will they have neutralization and that's very
year out, will they have neutralization and that's very

relevant for developing vaccines and whether or not people are protected for
relevant for developing vaccines and whether or not people are protected for

another round of infection. Right right. Very interesting.
another round of infection. Right right. Very interesting.

So going back to the current clinical testing Mike,
So going back to the current clinical testing Mike,

what can people expect if they, if they do have a test what's your sort of
what can people expect if they, if they do have a test what's your sort of

turnaround time looking like what's your capacity uh looking like, etc.
turnaround time looking like what's your capacity uh looking like, etc.

Sure so currently we can perform
Sure so currently we can perform

about 1400 to 1500 tests a day in our clinical laboratories.
about 1400 to 1500 tests a day in our clinical laboratories.

Just for our perspective that's a very large volume.
Just for our perspective that's a very large volume.

Our total volume of testing in health system laboratories is about 30,000
Our total volume of testing in health system laboratories is about 30,000

samples a day. So when we're talking 1500 that's
samples a day. So when we're talking 1500 that's

that's a big change from a couple months ago.
that's a big change from a couple months ago.

If you're a hospital patients a hospital patient, meaning
If you're a hospital patients a hospital patient, meaning

admitted in the hospital, the turnaround time is one to two hours.
admitted in the hospital, the turnaround time is one to two hours.

If you are being seen as an outpatient our median turnaround time is
If you are being seen as an outpatient our median turnaround time is

about 12 hours. Of course sometimes it's a little bit
about 12 hours. Of course sometimes it's a little bit

longer depending on the number of samples we get. We are
longer depending on the number of samples we get. We are

fortunate to have that turnaround time for our Duke patients.
fortunate to have that turnaround time for our Duke patients.

As many of the people listening probably know
As many of the people listening probably know

there are supply shortages for testing and many of the
there are supply shortages for testing and many of the

commercial laboratories have turnaround times as high as seven days right now.
commercial laboratories have turnaround times as high as seven days right now.

Right. So we maintain the ability to do very rapid testing for our Duke
Right. So we maintain the ability to do very rapid testing for our Duke

laboratories. No that's that's obviously important.
laboratories. No that's that's obviously important.

So this is a little maybe speculative far
So this is a little maybe speculative far

afield but from you know you I'm sure you're
afield but from you know you I'm sure you're

following sort of the vaccination literature. What are you all thinking
following sort of the vaccination literature. What are you all thinking

when, when might people start to see a
when, when might people start to see a

vaccine, you know?
vaccine, you know?

But I'm just curious. This is this is about ten times a day, this question.
But I'm just curious. This is this is about ten times a day, this question.

Yeah I'm the cautious optimist i guess.
Yeah I'm the cautious optimist i guess.

I think, I think first quarter. I don't see how we get to having
I think, I think first quarter. I don't see how we get to having

something that we feel we're ready to move forward
something that we feel we're ready to move forward

with before first quarter of 21. And the reason I say that is
with before first quarter of 21. And the reason I say that is

we're we're starting Chip Walter of the institute is starting
we're we're starting Chip Walter of the institute is starting

trials very soon. We're going to enroll a thousand people
trials very soon. We're going to enroll a thousand people

here at Duke. They have to get a prime and a boost. So
here at Duke. They have to get a prime and a boost. So

if they get enrolled in August, a month later they'll get a booster
if they get enrolled in August, a month later they'll get a booster

vaccination and then you have to follow the minimum of six months from that
vaccination and then you have to follow the minimum of six months from that

point to see if they have antibody responses and how,
point to see if they have antibody responses and how,

strong the antibody responses are before you can even think about
strong the antibody responses are before you can even think about

going to an EUA authorization of some sort. So I
going to an EUA authorization of some sort. So I

don't see this until early uh sometime in first quarter. And then
don't see this until early uh sometime in first quarter. And then

the real challenge is, um although if it's if it's some of the
the real challenge is, um although if it's if it's some of the

big players right now that are going ahead
big players right now that are going ahead

and manufacturing vaccine at risk, so they're making doses,
and manufacturing vaccine at risk, so they're making doses,

you still have a supply chain issue of vialing those vaccines,
you still have a supply chain issue of vialing those vaccines,

getting them out around the world around the country and getting them
getting them out around the world around the country and getting them

administered. So I'm cautious and I said you know
administered. So I'm cautious and I said you know

but with some optimism I think next spring and summer, we're
but with some optimism I think next spring and summer, we're

looking at dosing people and administering them if
looking at dosing people and administering them if

we get an EUA approval on something. So I hope it's sooner
we get an EUA approval on something. So I hope it's sooner

but I will be very surprised if we can roll it out faster than that.
but I will be very surprised if we can roll it out faster than that.

So for our listeners can you define EUA approval?
So for our listeners can you define EUA approval?

Yeah so EUA, so traditionally uh approvals take a very long
Yeah so EUA, so traditionally uh approvals take a very long

process. When you're in in a pandemic such as this
process. When you're in in a pandemic such as this

in the field for diagnostics or for vaccines
in the field for diagnostics or for vaccines

or for drugs that are useful, you can apply to the FDA and it takes a
or for drugs that are useful, you can apply to the FDA and it takes a

sort of a different route and it's called Emergency Use Authorization
sort of a different route and it's called Emergency Use Authorization

and that then gives a licensure that that diagnostic or that vaccine or that
and that then gives a licensure that that diagnostic or that vaccine or that

medication can be used. Eventually when the emergency is over
medication can be used. Eventually when the emergency is over

you have to go back and seek a permanent authorization or approval process but
you have to go back and seek a permanent authorization or approval process but

there are these are fast-track ways to get things out to help when you're
there are these are fast-track ways to get things out to help when you're

in an emergency situation. So finally I just want to
in an emergency situation. So finally I just want to

say you know we've heard a lot of concerns about undergraduates coming in,
say you know we've heard a lot of concerns about undergraduates coming in,

we have a lot of graduate and professional students and you know we
we have a lot of graduate and professional students and you know we

had originally settled on a plan, excuse me, where we would test all the
had originally settled on a plan, excuse me, where we would test all the

undergraduates coming in and i know we're starting to work on
undergraduates coming in and i know we're starting to work on

a larger sort of surveillance, screening that Tom you're playing a role
a larger sort of surveillance, screening that Tom you're playing a role

in and then Mike we would confirm anything in the clinic for, you know,
in and then Mike we would confirm anything in the clinic for, you know,

delivery of results etc. I'm just wondering if you
delivery of results etc. I'm just wondering if you

could each just comment on that that briefly, no promises yet but I
could each just comment on that that briefly, no promises yet but I

know this is where we're working towards. Well from from our standpoint, we're
know this is where we're working towards. Well from from our standpoint, we're

working very hard to bring this online in early August
working very hard to bring this online in early August

and we're doing a lot of different things. So we've we've started securing
and we're doing a lot of different things. So we've we've started securing

supply chain of reagents that we wanted a 90-day
supply chain of reagents that we wanted a 90-day

supply in-house to support some of the assays that we're
supply in-house to support some of the assays that we're

going to use for this. We're also going down the strategy of
going to use for this. We're also going down the strategy of

so-called pooling of of samples and that's where you take
so-called pooling of of samples and that's where you take

in our example we're going to take five
in our example we're going to take five

samples pooled together, run that and then if you get a
samples pooled together, run that and then if you get a

positive within that pool you do a process of deconvoluting the pool
positive within that pool you do a process of deconvoluting the pool

or splitting it and running those individuals to confirm who's positive.
or splitting it and running those individuals to confirm who's positive.

That's really the only way that we can get through large numbers in short
That's really the only way that we can get through large numbers in short

periods of time. And so in my view to make a surveillance
periods of time. And so in my view to make a surveillance

system work you have to be able to turn around results quickly.
system work you have to be able to turn around results quickly.

So we're going to be employing the use of some robotic
So we're going to be employing the use of some robotic

devices to make those pools and to be able to hopefully
devices to make those pools and to be able to hopefully

get a a very rapid turnaround time of the result. The other key part
get a a very rapid turnaround time of the result. The other key part

of this is you know the the team that's looking
of this is you know the the team that's looking

at modeling and just who are you going to sample
at modeling and just who are you going to sample

on campus? That's different than the testing part. We're trying to
on campus? That's different than the testing part. We're trying to

ramp up the testing capabilities and capacity and then we'll have to work
ramp up the testing capabilities and capacity and then we'll have to work

with the modelers to see which which subsets or which groups of
with the modelers to see which which subsets or which groups of

students get tested and how frequently. So then
students get tested and how frequently. So then

let's say you start finding positives then. Mike they're going to come over to
let's say you start finding positives then. Mike they're going to come over to

you, they're going to get a confirmatory test
you, they're going to get a confirmatory test

I think to Tom's point, one of the most critical things is
I think to Tom's point, one of the most critical things is

who do you test and how frequently? And and the devil's sort of in the details
who do you test and how frequently? And and the devil's sort of in the details

of that with what do you do with the results?
of that with what do you do with the results?

And I think uh the university has done a fantastic job actually over the
And I think uh the university has done a fantastic job actually over the

past couple weeks trying to sort through these very complex questions
past couple weeks trying to sort through these very complex questions

because there is no answer that anyone can point to and say
because there is no answer that anyone can point to and say

that's the right answer. It's the best you can do is that's the best
that's the right answer. It's the best you can do is that's the best

guess. But I think we're going to make some really good guesses on what a good
guess. But I think we're going to make some really good guesses on what a good

surveillance program looks like. Our clinical laboratories will continue
surveillance program looks like. Our clinical laboratories will continue

to build out capacity. By mid-fall I would expect our capacity
to build out capacity. By mid-fall I would expect our capacity

to be up around 3,000 tests a day. With that capacity we
to be up around 3,000 tests a day. With that capacity we

could easily help support the vaccine institute
could easily help support the vaccine institute

with screening programs. I know Tom is anxious to get back to his exciting
with screening programs. I know Tom is anxious to get back to his exciting

research and we can help shoulder back some of
research and we can help shoulder back some of

the screening capacity. In the short term, the vaccine
the screening capacity. In the short term, the vaccine

institute is doing a great service uh to the university by providing some of the
institute is doing a great service uh to the university by providing some of the

throughput through Tom's clear accredited laboratory to
throughput through Tom's clear accredited laboratory to

to pick up this screening. So but yes, when we find positives they'll come
to pick up this screening. So but yes, when we find positives they'll come

over go through one of our drive-through tents,
over go through one of our drive-through tents,

roll down the window, get a swab, that's a bit of an exaggeration but it's a good
roll down the window, get a swab, that's a bit of an exaggeration but it's a good

process, and get confirmed. Okay well
process, and get confirmed. Okay well

you know I want to thank you both uh not only for this conversation but for
you know I want to thank you both uh not only for this conversation but for

everything you're doing. I know it's been kind of round the clock. I mean
everything you're doing. I know it's been kind of round the clock. I mean

you know, I see emails from people at all hours
you know, I see emails from people at all hours

and I hear about tom was talking about machines that got set up at 3am and I
and I hear about tom was talking about machines that got set up at 3am and I

know that clinical volume is just you know unfortunately kept going up. But I
know that clinical volume is just you know unfortunately kept going up. But I

but I think that the testing piece is incredibly important for
but I think that the testing piece is incredibly important for

uh for both the Duke community and the larger community so I want to thank
uh for both the Duke community and the larger community so I want to thank

you both very much and uh thanks for this interview.
you both very much and uh thanks for this interview.