An Answer to Cancer? Using the immune system to fight cancer -- Longwood Seminar

Good evening.


I'm Gina Vild.

I'm the Associate Dean for Communications and External

Relations from Harvard Medical School.

And welcome to our grand finale of the 2017 Longwood Seminars.

As you know, we had a snow date for the first one.

So we're thrilled that all of you decided to come back.

May I ask if there is anyone here who has attended all four

of the Longwood Seminars Wow, fabulous.


Before we actually begin our program tonight,

I'd like to ask Belinda Davis to come forward.

Belinda, this is her first year organizing the Longwood

Seminars And I believe she's done an extraordinary job.

And it's an awful lot of work that goes into this.

And so just please join me in thanking her.


And it's her birthday.

So if you see her on the way out--

Thank you.


The Longwood Seminars, we initiated them

about 16 years ago.

And the concept behind them was to share

the knowledge of Harvard Medical School

and our extraordinary faculty with the Boston neighborhoods.

But we soon grew far beyond that.

And in the last few years, we've been Livestreaming.

And tonight, again, we are Livestreaming around the world.

And I want to welcome all of our visitors

who are joining us from afar.

In fact, I thought you'd be interested to know

that, in the first three seminars,

they were viewed in 93 countries and watched

by more than 638,000 people.

So those in the room tonight represent just a small number

of who are enjoying these seminars around the world.

Some of the top viewing countries

are India, Taiwan, South Korea, Mexico, Singapore, and the UK.

In addition, we stream live on Facebook.

And the first three seminars were

viewed by 33,400 on Facebook, and another 5,000 on YouTube.

So to all of you, welcome.

And thank you for being a part of this very special program.

Before we get started with our program, An Answer to Cancer,

I just have a few announcements.

If you've earned a certificate of completion

or professional development points,

you can pick up your certificate in the lobby following the Q&A.

If you've earned the certificate and are not here tonight,

you can contact us via email.

We also invite you to complete an evaluation form.

We have those in the lobby.

And your ideas are really an important way

of how we can infuse some really good, new ideas

into these programs so that we keep

them vital and interesting.

So please take some time and do that.

We also will be collecting your questions.

You all know this because you're veterans.

And we will have people strolling up and down

the aisles, so please turn your questions over to them.

If you're part of our audience watching through Livestreaming,

you, too, can send us questions.

Join us and send your questions by using the hashtag


So this evening, we have four of the world's leading


And they truly are four of the world's leading authorities,

both scientists and clinicians, here to talk with you.

They'll share with you the strategy

of using the body's own defenses to combat cancer.

First, we have our moderator, Dr. Arlene Sharpe.

Dr. Sharpe is the George Fabyan Professor

of Comparative Pathology.

She is the Interim Co-Chair of the Department

of Microbiology and Immunology at Harvard Medical School.

And she's a member of the Department of Pathology

at Brigham and Women's Hospital, an Associate Member

of the Broad Institute of Harvard and MIT,

a leader of the Cancer Immunology

Program at the Dana-Farber Harvard Cancer Center,

and Co-Director of the Evergrande

Center for Immunologic Diseases at HMS and Brigham

and Women's Hospital.

Dr. Sharpe's lab focuses on new therapies

for autoimmune diseases and cancer.

And she'll be our moderator.

Dr. David McDermott is the Director of the Melanoma Skin

Cancer Program and the Biologic Therapy Program at Beth Israel

Deaconess Medical Center.

As a medical oncologist and clinical investigator,

he has a particular interest in therapies that enhance

the immune response to cancer.

Dr. McDermott is an Associate Professor of Medicine at HMS

and a member of the editorial board for the Journal

of Clinical Oncology.

Dr. Jerry Ritz is the Executive Director

of the Connell-O'Reilly Cell Manipulation Core Facility

at the Dana-Farber Harvard Cancer Center and an HMS

professor of medicine.

He's a recipient of the prestigious Stahlman Scholar

Award given by the Leukemia and Lymphoma Society of America.

Dr. Ritz is an expert whose lab focuses

on hematologic malignancies, bone marrow transplantation,

and cancer immunology.

Dr. Catherine Wu is an HMS Associate professor of medicine

and a staff physician at Dana-Farber and Brigham

and Women's Hospital.

She's a member of the American Society of Clinical

Investigation, a recipient of a 2011

Stand Up to Cancer Innovative Research Grant

from the American Association of Cancer Research,

and a scholar of the Leukemia and Lymphoma Society

of America.

Her laboratory focuses on dissecting the underlying

mechanisms of pathobiology of chronic lymphocytic leukemia,

which is a common adult leukemia,

as a means to generate more effective therapies.

Particularly immune-based treatments.

This is truly an illustrious panel

that we bring to our final Longwood Seminars.

So thank you, Dr. Sharpe.


Well, thank you very much for that very kind introduction.

In my talk this evening, I'm going

to be introducing the topic of cancer immunology and cancer


We all think of the immune system as a way that

defends us against infections.

For example, the immune system exists to defend us against

the diversity of the microbial world,

viruses which we cannot see with the naked eye.

This is an electron micrograph of influenza virus,

which causes the flu, as well as bacteria, shown here.

Cancer immunotherapy is designed to boost

the body's own immune defenses.

In this case, to fight cancer.

Now, white blood cells in the body

are cells that are important in the immune system.

And a particular type of white blood cell,

called a T cell is able to kill microbes that cause infection

and also cancer cells.

We have-- whoops.

Is it possible to go back one slide?

We have a very limited number of cells

that can recognize a particular microbe or cancer.

But once these cells are activated, they can expand.

And one can get large numbers of these cells that

then can function to help us fight infection or cancer.

Now, a key function of the immune system

is to distinguish normal cells in the body from foreign cells,

such as infected cells or cancer cells.

And the immune system uses proteins

that are on the surface of, for example, T cells

shown here, that need to be either activated

to promote an immune response or inhibited

to prevent an immune response at certain checkpoints.

These checkpoints are important for proper functioning

of the immune response.

And one needs to distinguish infected cells, for example,

or cancer cells from normal cells.

And these immune checkpoints lead

to responses against infected cells,

but not to our own bodies or self.

Now, cancer cells are very clever.

And they've figured out many different ways

to block immune responses.

And one of the ways that cancers do this

are to use these inhibitory checkpoints that

normally are used to prevent responses

against our own bodies.

In this case, we have an immune response,

and the tumor has put up a blockade

against the immune system.

And this system, called checkpoint blockade,

that you've heard about, essentially

blocks pathways that the tumors are using

to inhibit anti-tumor immunity.

So the tumor prevents an effective immune response.

And one type of immunotherapy called checkpoint blockade

blocks pathways that the tumors are using

to inhibit the immune response.

Now, the tumor cells themselves can

make a variety of substances that

can prevent immune responses.

And essentially, they're able to use

many of these inhibitory checkpoints

to their own devices.

Two of these checkpoints that are important and have

been translated into therapy are cell surface receptors

called PD-1 and CTLA-4.

These cell surface receptors can be

expressed on T-lymphocytes, these T-cells,

as well as some other cell types.

And when they interact with their ligands,

or binding partners, these cells then

can deliver inhibitory signals into the T-cells

and prevent their responses.

Cancer cells themselves can send signals through PD-1

into a T-cell, essentially tricking the immune system

into seeing these cells as healthy normal cells.

And checkpoint inhibition blocks these pathways

that then can expose these cells to immune attack.

This is a picture showing you two different types of tumors,

A kidney tumor on the left and the non-small cell lung

cancer on the right.

And we've used an antibody to PD-L1, one of the molecules

that combine to PD-1.

And this brown staining here is an antibody

that can visualize PD-L1 on the surface of the tumor cells,

and showing you in this brown stain

here that tumor cells can express

high levels of the ligand for PD-1.

So simply put, then, in a tumor microenvironment,

we can have T-cells that express high levels of PD-1,

shown here.

And these cells then can interact with PD-L1

on a tumor cell.

When this happens, the T-cell cannot function and kill

the tumor.

But the drug that's used in this case

is an antibody either to PD-1 or PD-L1.

And when this is given, this prevents PD-L1

from interacting with PD-1 and blocks this inhibitory signal.

So now the T-cell can function and is

able to kill tumor cells, and make certain protein

hormones called cytokines that also can work together

to boost anti-tumor immunity.

There are now a number of checkpoint inhibitors

that are approved by the FDA that

are targeting these receptors.

One is against CTLA-4, a molecule called Yervoy.

And there are several antibodies to PD-1, one called Opdivo,

and the other is Keytruda.

And then antibodies to PD-L1 as well.

And there are, in addition, about 20 agents

that are present in ongoing clinical trials.

These antibodies have been approved by the FDA

for several tumors, shown in green here.

And one of the remarkable things about the PD-1

pathway is that it's been used over and over again

by a variety of types of tumors to inhibit anti-tumor immunity.

And so antibodies that inhibit the PD-1 pathway

have been FDA-approved for melanoma, for bladder cancer,

non-small cell lung cancer, head and neck tumors, kidney tumors,

Hodgkin's disease, and Merkel cell tumors.

And what this is showing you is overall response rates, showing

you the overall response rate ranges from 20% to 30%

in many tumors.

Up to 87% in Hodgkin's disease.

Now, why is there such enthusiasm for immunotherapy?

This slide illustrates one of the major reasons.

If we first look at this graph on the left, what we're

looking at are patients whose tumors

have been treated with blockers of CTLA-4 or PD-1.

The experience is longer with blockers of CTLA-4.

And what you can see is that there are about 20%

of the patients who respond.

But the benefit is durable, lasting for at least a decade.

The experience with PD-1 is more recent.

But in addition, in this situation,

blockers of the PD-1 pathway, also

can lead to durable benefit.

So the enthusiasm for immunotherapy

is that it can lead to a long-term benefit,

but a moderate number of patients respond.

And I'll come back to that point.

In contrast, patients who receive chemotherapy,

or other types of therapies such as kinase blockade,

there is a higher percentage of patients who initially respond.

But, unfortunately, the response is short-lived.

So the great enthusiasm for immunotherapy

is the durability of response that's seen.

Now, what is it the immune system

sees in a tumor to attack?

Tumors have a variety of changes in protein coding regions.

And these are [INAUDIBLE] mutations.

And as a result of this, these changes

that occur in the tumor cell can be recognized

by the immune system.

And these protein coding changes are called tumor neoantigens.

And so, in cancer, there are two types of evolutionary processes

that are driving the tumor.

The first are the mutations that are occurring.

And there are two types of mutations.

There are so-called driver mutations

that initiate the tumor-forming process.

And then there are these protein coding changes

called neoantigens that the immune system can recognize.

In addition, there are a number of immune evasion pathways

that we're learning more and more about.

The PD-1 pathway is one, and there are many others as well.

And one of the reasons why the immune system is

the perfect approach to use against the tumor

is that, as the tumor continues to evolve,

so can the immune system.

So the immune system can keep up with the tumor.

We're just beginning to understand

how to identify those groups of patients

who benefit from PD-1 therapy.

Understanding of immunology, as well as cancer genetics,

has identified several groups who are responders.

One are groups of people who have highly mutated tumors,

where there's defects in DNA repair.

These patients have a high response rate to PD-1.

In addition, there are certain tumors,

such as Hodgkin's disease, where the tumors have

many, many copies of the ligands for PD-1.

And so when you then block this pathway,

there is a very high response rate

to PD-1 pathway inhibitors.

In addition, certain tumors of viral origin,

such as those caused by papilloma viruses,

this can lead to head neck tumors, for example,

or Merkel cell tumors, one can see in those settings

that there also is a high response rate.

But we're really just beginning to understand these responses,

and further work is needed to identify biomarkers

that we can correlate broadly with responses

to PD-1 immunotherapy.

But this is really just the beginning.

And really, we're at the beginning

of a revolution in cancer therapy

because there are many more cancer immunotherapy targets

than PD-1 and CTLA-4.

This is a diagram showing you here that these T-lymphocytes

can express many of these inhibitory receptors.

And all of these receptors have become druggable targets

for cancer immunotherapy.

And so while we now have therapies

to CTLA-4 and the PD-1 pathway, many of these other targets

are also now, molecules are being made for therapies.

And there are clinical trials that are ongoing.

And so the PD-1 pathway is really

becoming a foundational building block

to combine with other types of therapies.

And right now, there is great interest

in combination therapy.

How to increase the benefit to PD-1 pathway blockade.

And there are a number of approaches

that are being undertaken.

The first is to combine PD-1 blockade

with blocking other inhibitory pathways to release the brakes

on the immune system.

The other is to block PD-1, and then stimulate

the immune system in some way.

Another is to block PD-1, and then give chemotherapies

or certain types of kinase inhibitors or radiation


Or give PD-1 blockade, along with a cancer vaccine,

a cellular or engineered T-cell.

And you'll be hearing much more about these

in the subsequent talks.

And so hopefully over the next decade,

we're going to have tremendous progress because there's

so much work ongoing now.

So the future of cancer therapy and decisions

may look something like this.

One will have sequencing of the genome.

So one can identify what genes are causing the tumor,

and these can be drug targets.

In addition, one can identify those protein coding mutations

and identify neoantigens, which also can

service targets for therapy.

In addition, one can develop a so-called tumor immunoevasion

score, identify which ways the tumor is

using what pathways to evade the immune system,

and then target these.

So this would lead to choosing the best immunotherapy, and how

to combine immunotherapy with the best

targeted therapy or vaccine.

So to summarize, then, the success of checkpoint blockade

therapy has led to incredible change

in the strategy for cancer therapy.

We're beginning to appreciate that tumor immunity is

limited by natural regulatory mechanisms

that the immune system is using.

And by boosting the body's immune defenses,

we can fight cancer.

The future is in combination therapy

to extend the benefit of cancer immunotherapy,

to build on the success we have with checkpoint blockade

and combine it with a variety of other approaches.

And in the next talks, you'll be hearing about some

of these other approaches, taking basic discoveries

in immunology and translating them to therapy.

The appreciation that there are many pathways that

inhibit tumor immunity has translated

to checkpoint blockade.

And you're going to hear from Cathy Wu about approaches that

are leading to t-cell vaccines, from Dr. Jerry

Ritz about T-cell engineering and stem cell approaches,

and from Dr. David McDermott about clinical trials

in combination therapies.

Thank you very much for your attention.


And I'll turn it over to Dr. Wu.


So great, happy to be here.

And just to follow Arlene's talk, as she said,

we have a very complex number of different types of cells that

comprise the immune system.

There are specialized organs.

They have the ability to travel and go and survey

the whole entire body to make sure that we're safe.

And just to remind ourselves that as hardwiring,

the immune system is hardwired to recognize

these external pathogens, so protect us

from these external evaders.

They're really a mixed and motley group of different types

of organisms.

Bacteria, fungi, viruses and parasites.

But also to reiterate that, as a system,

the immunity responses, the function

to survey all the interfaces that our body interfaces

with the environment, detect whether or not

these pathogens are present.

If so, mount an immune response.

For example, traffic to the lymph

node, where robust responses can be generated.

And there's another very important function

that happens which, is memory is developed.

So there may be, first, the initial encounter

with a pathogen. And once an immune response

is developed, if there is a secondary encounter,

there is a very swift and high amplitude response.

And again, this serves to protect us in the long term.

Now, the Danish physiologist Peter Panum, back in the 1800s,

was called to the Faroe Islands to investigate

a measles outbreak, and was able to study this phenomena

of immunologic memory.

Because certainly what he found was

there had been already a measles outbreak many years before,

some several decades before.

No measles for 65 years.

And when there was this new outbreak that came out,

those that were infected earlier were not affected again, back

in 1846.

So this really demonstrated this long-lived immune protection

that exposure to this infectious agent had caused.

Now, one question you might ask, all of us

are well-acquainted with the flu.

So why, for example, do we get flu pandemics?

And it gets back to the idea of antigen and antigen exposure.

Because influenza, as a wily organism,

it can avoid immunologic memory by changing its molecules

every so often.

So mechanistically, this is how we

can understand why these outcroppings of flu pandemics

come from time to time.

Now, what about cancer?

We all know that cancer arises because of DNA mutations that

can lead to altered proteins.

And there is quite a bit of different types of mutations

per different cancer.

Many of you know that there's been a technological revolution

that's happened over the last couple of years,

with the availability of broad sequencing

that allows us to systematically sequence

the coding genomes of cancers.

And many of these large-scale studies

have had the goal of understanding the mutation

spectrum of tumor cells, putting that all together

with other layers of information so

that we can come to understand what

are prognostic markers related to that cancer.

And also, can we think better how

to develop novel therapeutics?

And certainly in the disease that I study, but certainly,

this could have been any cancer, we've

come a long way ever since these sequencing technologies

have been upon us, which has been a mere seven or so years.

So at least in chronic lymphocytic leukemia,

where a long-standing question has been,

can we understand why certain patients progress more rapidly?

Why do some patients have a more indolent course?

Can we understand what are the clinical or the protein


Well, I would say that since the advent

of next-generation sequencing, there's

been this explosive increase in our knowledge of the genetics,

the transcriptomics, and the epigenomics that

allow us to understand that.

And just to give you a flavor of what we've been able to find,

through the mining of all this high-dimensional data,

is that we certainly have been able to find,

for example, again, this is the example

of chronic lymphocytic leukemia.

But it certainly could have been true for any

of the many other tumors that have been characterized

in a similar fashion.

We can identify those type of molecular events

that statistically appear to be present

at a higher-than-expected frequency that suggests

that there was positive selection,

and was involved in the genesis of these tumors or leukemias.

Each column here is an individual patient.

And what I want to point out to you

is that, although there are some recurrent events that

are common amongst patients who have this type of leukemia,

for the most, part each patient has

a different genetic profile.

And that really speaks to not only the immense heterogeneity

in the genetics of individuals that have the same disease,

but also the differing evolutionary trajectories

that each patient takes as all of these alterations

are acquired.

And altogether, again, in the example

of chronic lymphocytic leukemia, but I

think this is extendable across the different cancers,

we have come to a picture such as this.

Which is we understand each tumor as not monolithically

all the same type of cells that are one block of population,

but rather many sub-populations within each cancer.

So there's maybe one clone that has three mutations that

define this sub-clone.

There's another group that has another five

mutations, and this one with another two mutations,

and so on and so forth.

So within individual patients' cells,

there may be different sub-populations.

What that means therapeutically is

that, as we try to start to treat patients

with different types of therapies,

be it chemotherapy or targeted therapy,

if not all the cells are eradicated, because there

is such diversity in the population already,

there are evolutionary pressures that

lead small populations then to expand

and be the cause of subsequent relapses.

So too often in cancer, we see remissions that are short,

that come back.

And we go through successive rounds of therapy.

So this really provides us with a tremendous challenge

as we think about how to eradicate cancer.

And so I would say that if we acknowledge that not only is

each individual's cancer different from the next

individual, but also within each individual's cancer,

there are many different sub-populations,

and that each of those as in aggregate, that

means that per individual, there's

a unique set of antigens per tumor per patient.

Then perhaps the solution to this conundrum

is we need to have an army of T-cells.

Because the immune system is actually,

again, hardwired to be able to have a capacity

to generate diverse T-cells that can recognize diverse antigens

and eradicate them.

So in that respect, again, along the same lines

of what Arlene mentioned, cancer vaccines

actually have perhaps a very complementary role to play.

So Arlene already talked about how

blocking immune checkpoints that have

the brakes on the immune system can now be overcome

with some therapeutics.

Perhaps we can use cancer vaccines

to help steer that response, focus the response,

raise T-cells that have the right specificity,

and guide the immune response in the anti-tumor direction.

So that together with strong adjuvants, which

are going to add strength and speed to the process,

and if we put this together with checkpoint blockade,

perhaps this is a nice, effective way

to generate highly-specific anti-tumor immunity

with fewer side effects.

So how do we generate cancer vaccine?

So I've tried to bring it down to the essential ingredients.

And I think that we would all agree

that, for a vaccine to work, we need at least

these four complements.

You need antigen, so something to target against.

You need adjuvants, so something to kickstart the system.

You need to figure out how to deliver the vaccine.

And of course, again, as Arlene mentioned, decrease that shield

around the tumor that's immunosuppressive

and inhibits the activity of cytotoxic T-lymphocytes.

So we need to block that suppression.

Do we have any evidence already that vaccines can work?

Well, it turns out this is not a new idea.

People have been kicking around the idea

of vaccines for a long time.

This is from back in the late 1890s.

There was a very renowned New York

surgeon who created what he called, his name was Dr. Coley,

he created Coley's toxin.

And he had this one patient who had

this awful tumor on his face.

And he created this toxin, which was really

mushed up bacterial components, and injected it in the site.

And this gentleman received at least [INAUDIBLE] or so

rounds of these injections.

And it did lead to regression, and this man

continued to live for several more years

than would have expected for this type of tumor.

And I think we now recognize that this was probably

the first adjuvant.

So with all the biological and molecular understanding

that we've gained, a very effective way

to jumpstart the immune system is

to remind the immune system of those external pathogens.

So these were microbial components

that were mushed up and injected onsite.

So perhaps he was able to awaken those T-cells that

happened to already be there and get them

to work, at least for this person, effectively

for a number of years.

Since that time, we've come up with a list

of so many different ways to think

about antigens or adjuvants and formulations and deliveries.

And it can be quite bewildering.

But I will say that, in between all of this mixing and matching

of complements, there have been some notable responses

that we've seen in patients.

So I will remind you that, for early stage bladder cancer,

we are using a microbial complement to treat that.

This has been in use for decades now.

There have been, in another virally-mediated early type

of carcinoma, there have been peptide vaccines

that have been able to generate clinical responses

in this setting.

And some of you will know about the first FDA-approved

immunotherapy, which is Provenge,

a vaccine for prostate cancer.

So there are a few examples of where vaccines have worked.

But what's especially exciting to me right now

is that we have an ever-growing toolkit that

seems quite clinically efficacious,

that makes the possibility of vaccines seem like a reality


So one is, in terms of adjuvants,

I'll just mention as an example.

This Poly ICLC is, again, a microbial component

that has been distilled down to its active ingredient

and has been already tested in clinical trials.

There's been a number of different studies

that have used long peptides.

So this is peptides that can be synthesized

and given in a syringe, and used as a way to deliver antigens.

And then Arlene already mentioned the antibodies

that we have to block immunosuppression.

So what we need is actually a good target.

So what are we going to go after?

And there, I would say that we are in a time where

we can identify neoantigens.

I already talked about how each tumor has its own specific set

of mutations.

Well, at least some of those mutations

will generate peptides that have altered amino acid.

So it can be translated.

It can generate these altered peptides.

Some of them have the potential to bind

to the patient's own HLA molecules.

And it is this coupling that allows these peptides to be

displayed on a tumor surface, where t-cells

can come and recognize them.

So neoantigens are those antigens

that are born out of mutation and are

present on the tumor specifically, but not

on normal cells.

And so this has always been conceptualized

to be a really ideal type of antigen.

But what was missing for so long was the ability

to systematically identify them.

What we have now is new technologies

that allow us to use DNA and RNA sequencing

so that we can readily identify these new antigens.

We can confirm that their expression

is present on the tumor cells.

We also have very well-vetted out prediction algorithms that

allow us to rapidly identify whether those peptides

that we're identifying from our DNA

and RNA analysis would bind to that person's own HLA


And these are what we call candidate neoantigens.

And really, conceptualized as hitting the sweet spot.

Because by virtue of arising from tumor mutations,

they have exquisite specificity in terms of expression

on tumor cells, and akin to pathogens,

they are highly immunogenic.

Because we do not automatically have

a way of deleting out T-cells that might react to them.

And this is really different than where we've historically

been, in terms of the type of antigens

that we have tried to use for vaccines or immunotherapy.

And these are what we call the shared

overexpressed self-antigens that are variable

in terms of their specificity for expression on the tumor,

and variable in the degree to which they're immunogenic.

Now, if you've been following this, what's

been very gratifying is that, aside from conception and idea

that these would be ideal antigens,

there is now lots of exciting data

that has cropped up in the last three

to four years that have really supported

the idea that neoantigens truly appear to be effective tumor

rejection antigens.

Some of that data comes from studies

that have looked at hundreds and thousands of tumors

and try to identify those antigens,

and look at clinical outcome and compare whether those tumor

samples that had higher numbers of neoantigen

seem to do better in terms of survival,

compared to those that had lower neoantigen load.

These types of studies have also tried

to look at specific diseases or specific treatment settings.

Other studies have asked, well, if neoantigens

are such great antigens, are they actually expanded?

Can I see whether or not they're expanded in those settings

where patients demonstrably have clinical responses?

And the answer is yes.

After stem cell transplant, after cellular therapy,

after checkpoint blockade therapy,

in fact, there have been a number of interesting studies

that have shown that these T-cells seem

to expand out and have specificity for neoantigens.

And finally, both in model systems and in patients,

it's been demonstrated that neoantigen-specific T-cells can

directly kill tumor cells.

So really, with all this data, it's

been very supportive of the idea that these are

in fact immunogenic antigens.

And then maybe the next logical step

is, well, if that's the case, can we

not generate individualized personalized vaccines?

If we know what the mutations are in the individual's tumors,

can we not actually generate a vaccine

that would be specific for that individual person?

And so the concept that we had already quite some time ago

was, could we not take a tumor, have normal cells sequence

the DNA from these in parallel?

And then use the HLA typing information

and predict what are those peptides that

are personal for that specific individual,

come up with a synthesis strategy

that we could make peptides, for example, that were reactive,

that would be predicted to be individualized

for that individual's tumor, and get it in a syringe

where we could give it as a vaccine?

And already, at the Dana-Farber, we've

conducted and recently completed two proof of concept studies.

One in high-risk melanoma, as well as

previously untreated glioblastoma, that have in fact

demonstrated safety, feasibility,

and immunologic activity.

And so I think we're really here now at a time of a paradigm

shift in terms of, at least in the field of vaccinology,

where we have the ability to find these neoantigens

and conceive of ways of targeting not one,

but multiple antigens that are specific for a person's

individual tumor in a tumor-specific

in immunogenic fashion.

And so I had to put this up, because I often think about,

well, I like Star Wars.

But also, I think one of the opportunities

that we've been afforded by checkpoint blocking antibodies

is that we do have a Obi-Wan Kenobi that

has been able to bring down the shields from the Death Star.

And so I think part of our challenge

is how to use that opportunity and put this opportunity

together, so that we can actually get in there

and actually ultimately destroy the enemy.

And so I would say that, again, coming together

with what Arlene said, combinations.

I think that's really where our future is.

She's talked about overcoming the immunosuppression.

But can we also add another step,

which is potent/specific, anti-tumor,


That is certainly where vaccines come in.

But also, cellular therapy also has a really important role.

So I'm going to pass the podium to my colleague, Dr. Jerome

Ritz, who will tell us more about that.

Thank you.


Thank you, Cathy.

So I'm here to talk about cellular therapy.

So we've heard a lot about the immune system.

And the immune system works, in large part,

through the generation of cells that

are part of this immune system, that can then actually

do the work.

They either make the antibodies that fight infections,

or they actually attack these invaders themselves.

And the question is, can we use cells as the treatment itself?

And this gets us to the idea of thinking

about cells as a living drug.

Now, that's not a totally new concept

because we've been involved in stem cell transplantation

for 40 or 50 years.

So if we think about a stem cell transplant,

this is for a patient, often with leukemia or lymphoma,

or for a patient who has an abnormal immune system that

is not able to make their own immune system,

not make their own blood cells.

Well, we know that if we can find an HLA-matched brother,

sister, or an unrelated donor, and if we transfer

those stem cells into the individual,

that we can actually get rid of the leukemia.

And that can last for a long time.

There's a single transplant, and that single transplant

of a relatively small number of stem cells

actually lasts for the lifetime of the recipient

and continues to make blood for the rest of their time.

Well, it turns out that we can do that with immune cells,

and with T-cells, too.

T-cells are long-lived cells.

We can generate those T-cells.

It's been done primarily here in the past

for virus-specific T-cells.

Or it's been done from taking T-cells from actual tumors,

growing them up, activating them, and giving them back

as a T-cell therapy.

So these are actually long-lived, if you will,

living drugs.

Unlike an antibody or a chemical that

is going to be great degraded over time,

these cells can live.

They can multiply in the recipient.

And I'm not going to talk about it today,

but I'll mention it at the very end,

there's actually a new way of generating cells,

long-lived cells, through reprogramming.

And that's the ability to start with a mature cell

and reprogram that cell so that it may become an immature stem


And then once you have that immature stem cell,

that cell can then be programmed to make other mature cells

in the body, often the same individual.

So that's an exciting, very, very new technology

for cell therapy that is yet in the future.

But I think it's going to be coming.

So another way of thinking about this

is, if you think about the evolution of cell therapies,

what I mentioned to you initially about stem cell


Well, we've been doing stem cell transplants for over 40 years.

It's a common procedure.

There are over 20,000 done in the United States,

over 50,000 worldwide, in each year.

But what we're doing is we're taking

a mixed cellular product, we're taking where we can get,

and infusing that into the recipient.

And that's where the whole field started.

We're now at a point where we can

begin to identify specific cells within a larger context.

We can purify those cells.

We can activate them.

We can expand them.

And we can then give them some more defined therapy.

What's coming now in the future, and I'm

going to talk a lot about chimeric antigen receptor, CAR,

T-cells, we can now actually start with a cell,

and we can genetically engineer it

to do what we want it to do for that individual.

So all of this is really the evolution of cells,

starting with a mixed cellular product 30, 40 years ago,

then being able to identify the cells that we really want.

And now actually being able to engineer the cells that we

want for individual patients.

So how is that done?

So here's an example that has had

an awful lot of interest and excitement

over the last few years.

It's called-- can we go back a step?

It's the ability to generate chimeric antigen receptor


So you heard from Arlene and from Cathy

that T-cells have specific receptors that can be

used to identify cancer cells.

Well, in this approach, we start with a patient.

This would be a patient with leukemia or lymphoma.

We do a leukapheresis procedure that

allows us to get out large numbers of the patient's

own T-cells.

Those T-cells are then activated in the laboratory.

And then they're genetically engineered

through various viral vectors to now express

a new receptor, a receptor that is now

going to be able to allow those T-cells to actually recognize

the patient's own cancer cells.

Those cells are generated in the laboratory,

and they're expanded for a week or two weeks or three weeks

until we get enough of those cells.

And then they're infused back into the patient, where

they now become a living drug.

Those cells, now our immune cells,

derive from that patient.

They now can go because they now have a new receptor.

They can now go target and find the patient's own tumor cells.

This has been done most successfully

using CD19 as a tumor antigen. And this

is an antigen that is expressed by normal B-cells,

but also all of the cancers that come from the B-cell lineage.

So that would include acute lymphocytic leukemia,

chronic lymphocytic leukemia, and B-cell lymphoma.

So these are actually a relatively large,

different kinds of cancers that now can be

treated using CD19 CAR T-cells.

That's just the first target.

There have now recently been several papers that

have shown that another antigen called BCMA, stands

for B-cell maturation antigen, it's

another antigen and that is expressed by myeloma tumor


And that also can be generated into a CAR T cell,

again, specifically for patients with myeloma.

So just to show you how this field has evolved,

the whole genetic engineering field,

starting over 20 years ago, people

tried to generate these chimeric receptors,

that they could put them into cells.

And they started out by taking a signaling domain.

This is a receptor that is now synthetically put

into a cell that has one binding domain.

It has one signaling domain.

People use different kinds of binding domains,

actually, to take an antibody.

The binding domain, the specificity of an antibody,

hook it up to a signaling domain.

And these actually worked in some animal models,

but actually were not very effective in patients.

The trick was to enhance the signaling capacity

of these receptors.

And when you begin to add signaling domains

and costimulatory domains, that Arlene Sharpe talked about,

you can begin to add those together into a receptor.

You get a much more powerful signal that is transmitted.

And these are now the receptors, the CAR receptors, that

are now being very effective.

Of course, you can go even further.

You can add multiple different signaling domains.

These are called second-generation CARs.

We now have third-generation CARs.

We have armored CARs.

We have all kinds of different CARs

that people are beginning to use in clinical trials.

This actually is a diagram, that I borrowed from a paper

from Michel Sadelain, that actually shows the CD19 CAR

T-cells, the vectors, that are actually

being used in clinical trials.

And they all have many features in common.

They're all a little bit different.

But the ones that are effective have two signaling domains.

These have CD28 as a signaling domain, in addition to

CD3 zeta.

These have 4-1BB, in addition to CD3 zeta.

They all have very similar receptors,

all against the specific CD19 antigen.

And these are the ones that are currently

being tested in clinical trials that are showing really

very dramatic efficacy.

Now, Cathy, in her talk, talked about having specific T-cells,

with T-cell receptors to generate

those cells expressing receptors for specific tumor antigens.

Well, those can also be genetically engineered

and put into cells that may not express them already.

So you can generate both an antibody

CAR that's called a surface antigen CAR,

or a transgenic TCR CAR.

So both of these are now being developed for patients

for different treatments.

The advantage to this is that it really

attacks a target, an engine that is

expressed on the surface of the tumor cell, of the cancer cell.

The advantage of this approach is, as Cathy identified,

these are actually specific for internal proteins,

internal mutation, so that you can

see different kinds of tumor targets

using these two different approaches.

So CD19 CAR T-cells have really hit the hype, as it were.

And the reason is because they've

been very effective in patients with

relapsed B-cell malignancies that have actually

failed all other treatments.

And in that setting, anywhere from 50% to 90%

of patients with these various diseases,

pre-B-cell ALL, CLL, B-cell lymphoma, Hodgkins disease,

actually can now respond to these chimeric antigen receptor


And as I said, these are effective in patients,

both pediatric and adult patients, that

have failed other treatments.

And everyone is expecting that the FDA will actually

approve these as living drug therapies sometime later

this year or next year.

However, there is a price to pay,

in that there are some significant toxicities

with these treatments.

First of all, these antigens may not be expressed only

on the tumor cells.

Certainly the ones that are expressed

on the surface of the tumor cell,

but also some normal cells.

So those cells can also be eliminated.

When these T-cells get activated,

and actually, they can expand 1,000 fold within a week,

you get the release of various growth factors and cytokines

that can be very toxic in itself.

And then there's also very poorly-explained neurotoxicity.

And then, even though these patients

have very dramatic responses, some of those patients

actually do relapse, even after they've had this treatment.

This is data from the University of Pennsylvania.

Really, one of the leaders in developing

this type of treatment.

This is for patients with relapsed ALL.

Their response rate is very high.

And about a year afterwards, still, 50% of the patients

are doing very well.

And these are patients where you would expect very few, if any,

of these patients to survive this long after treatment.

So it's really very dramatic therapy

in patients who have failed other treatments.

So this is now an area of intensive investigation.

This, again, from another review paper

that came out earlier this year, just looking

at the various types of trials that have either

been completed or ongoing.

And there are 20 different CAR T-cell

targets that are being evaluated in clinical trials.

And for each one of these targets,

there may be five or six different clinical trials.

These are the trials that are actually looking at the TCR

targets that I mentioned.

There's a little bit less activity for these.

But these also are, I think, going to be very promising.

So this is an area of very, very active investigation.

So as I said, I think the challenges in this treatment

is that there are some toxicities.

And so people are really looking at how

to manage these toxicities.

And as more people get treated, as people more become better

at understanding what's working and what's not working,

these toxicities are being managed.

There is a relapse.

And sometimes, these patients relapse because the cells

actually stop persisting.

And so sometimes, you may need to actually give

a second dose of those cells.

But sometimes, the tumors actually figure out a way

of deleting that target antigen. So even though it's

something that is expressed very commonly, and at high level,

cancers are very smart and they figure out

how to protect themselves.

And then, I think solid tumors, so we've

talked about leukemias and lymphomas and myelomas.

The big challenge is, how do you get these CAR T-cells to really

infiltrate a solid tumor where you

have problems of what is the right antigen, what

is the right target?

How do you get the T-cells to target into the tumor


And once the T-cells get there, even though they're genetically

engineered to get there, how are they going to protect

themselves from the immunosuppressive environment

that the cancer has generated?

And so again, combination therapies,

trying to figure out how to pull all this together,

they're going to be very important.

And then, in the last few minutes

that I have, I want to spend a few minutes talking

about what's actually also needed in this type of therapy,

is you actually need a place to make these cells.

And that's what we call the cell manufacturing

for the early phase trials.

And because these are very complex therapies,

require a lot of work to generate one dose of cells,

the cost of these is, I think, going to be very high.

So one of my hats at the Dana-Farber

is as the head of what we call the Cell Manipulation Core


So this is a laboratory that we've

developed to actually do this under controlled situations.

And the mission of our laboratory

is to provide safe and effective cellular products for use

in clinical trials.

These are for patients who are getting stem cell transplant,

but also for patients with cancer

who are getting these innovative cell therapies.

Unlike a drug or an antibody that you

can get from a pharmaceutical company,

or you can actually have made for a lot

of different patients, these individual therapies,

we're talking about making one cell

product that's actually going to go for only one patient.

That's the ultimate personalized therapy.

So this is the people that work in our laboratory,

just to show that these are all clean room environments.

To actually get into the facility, you have to gown up.

There's constant monitoring of the environment

and the procedures that are ongoing to make sure

that there is no contamination, that these products are

safe and sterile.

Because of the increased interest in this,

we're actually building a new facility.

It's going to be located on the 12th floor of the Smith

Building at the Dana-Farber Cancer Institute.

It's going to be completed in about a year from now.

And this is just a design, looking at this facility.

All of these light blue rooms here,

that are shown in magnification here,

are all clean room suites designed

to make products for individual patients.

But you need to have the place to do it.

You also need to have the well-trained people that can

they can manufacture the cells.

But then, you need to have quality control systems.

You need to have quality assurance.

You have to deal with all the regulatory issues.

You have to keep track of everything

in the data that's generated in the facility.

And then you have to have an inventory system for all

the different bits and pieces that go into actually

making these things.

So it's a complex environment, and we already

have about 50 people who are working full time

in this facility.

So just to end here, I think the points I've tried to make

is that I think it is actually feasible to make these living

drugs not just for a stem cell transplant,

but for patients as a cancer therapy.

It's likely that the CD19 CAR T-cells are actually going

to be FDA-approved sometime later this year or early

next year.

But they have toxicities.

And so it's important to think about that,

and to develop even better generations of these CAR

T-cells that have the same efficacy,

but avoid some of the toxicities.

Now we know this is effective for some blood cancers.

Leukemias and lymphomas, solid tumors, I think,

will also be an important target for these.

And so there's an awful lot that's

going on to be able to engineer these cells and how to do this.

There's a lot of work going on, so stay tuned.

I think there'll be a lot of progress

to be made in this area.

Thank you.



Yes, sir.

There you are.

You're next.

All right.




Thank you.

Thanks to the organizers for inviting me here.

It's really an honor to speak in front of you today.

It's also an honor to follow Jerry Ritz, who's

been a pioneer in the field that I started in 20 years ago,

focusing on bone marrow transplant

and cellular therapies.

I didn't really last very long in that field.

I wasn't very successful.

About five years in, someone pulled me aside

and said something about not earning all of my salary.

And I was given the opportunity to move

from hematologic malignancies, like leukemia and lymphoma,

over to what Jerry was describing.

What we call solid tumors.

Tumors that start in major organs, like the kidney.

Melanoma, lung cancer.

And that's what I've been focusing on

for the last 20 years.

But in the first 15 of those years,

it wasn't exactly a great career move.

And I'll describe a little of my history, the history

of solid tumor immunotherapy, and how

some of the basic science that's been done by some of the folks

here at Harvard has changed not just my career,

but the outcomes for our patients with solid tumors.

So you've heard a lot about different approaches

to immunotherapy.

I'm going to focus on a few of these.

The first one I'm going to talk about

is one that was actually developed over 30 years ago,

the so-called cytokine therapy.

You can think of cytokines as protein messengers

of the immune system.

They communicate between different cell types

and tell them to turn on or turn off.

And some of the most important cytokines,

when it comes to fighting cancer,

are agents like interleukin-2, or interferon.

And they're actually natural substances,

meaning we all have them in our bodies at all times.

We think they're there to help us fight off infections,

like viral infections.

But they can be synthesized in the lab

and made into large volumes, and given back to patients

to essentially act like an accelerator

on the immune system.

To put it into overdrive.

And what impact did that have?

Well, 30 years ago, this is looking at an x-ray.

This is before the advent of CT technology.

You can look at the scan on the right

as a young patient with metastatic melanoma.

You see all those blotches in the lung.

That's all disease.

Those are all individual melanoma metastases,

Stage IV solid tumors.

And after a short course of immune therapy,

you can see the x-ray two years later.

No cancer.

That patient is still in remission.

That patient outlived their cancer.

And that was a truly exciting development

in the field of immunotherapy.

But we can talk a little bit about why

this had a limited application.

One of the reasons why is IL-2 when given in large amounts,

creates significant side effects.

But that wasn't the only reason why

IL-2 had a limited application.

But needless to say, this is one of the pioneers

in the field of IL-2.

This is Dr. Steven Rosenberg, who's still

at the NCI surgery branch.

I remember reading this article that was in Newsweek.

I don't even know if Newsweek still exists.

But at the time, I was in college reading this article.

And after reading it, I decided I

wanted to go to medical school.

It sounded like a really great idea.

And this is some of the early outcomes data

from some of those original trials.

This is a trial that my mentor, Mike Atkins,

presented about 10 years after some of Dr. Rosenberg's

initial discoveries.

And what this essentially shows is, in the patients

who had a major shrinkage of their disease

after getting interleukin-2, if you had a response,

if your cancer got smaller, the benefit could last for years.

You see 120 months here, these patients staying in response.

So what did the IL-2 experience teach us?

It taught us that remission was possible.

That we could take a person with normally a lethal disease,

and put them into remission that could last decades.

And some of these patients are still

coming to see us in clinic.

But what else did we learn during this process?

Well, it turns out your immune system, as you've heard,

is not very well-designed to fight cancer.

It's actually designed to fight infection.

And it's designed to turn on when you're

dealing with a viral infection.

If you've ever had a bad virus, you felt fever, chills, sweats,

those kind of things, that's your body revving up

to try to fight the infection.

Once it's contained, though, the immune system

is designed to shut off.

And it turns out that the brakes on the immune system,

as Dr. Sharpe was talking about, are actually more

potent than the accelerator.

But now, we understand some of these breaks.

And we can actually, while the immune system

is designed to shut it off, it can be reactivated.

And this is just a cartoon of just releasing the brakes

on the immune system.

And one of the most important breaks

on the immune cell, the T-cell, is a substance called CTLA-4.

Now, what happens when you release the brakes,

or you release the pressure on CTLA-4?

Well, this is from one of our first large studies,

Phase III randomized studies in 2010,

that was published in the New England Journal.

These are patients, again, with metastatic melanoma,

who had advanced cancer, Stage IV disease.

They were given a blocking antibody

intravenously that treated their disease

and improved their overall survival.

But more important is this was the first drug

to show an improvement in survival.

But more important, this is what's called a survival curve.

You can see these patients still alive two, three, four years


And their treatment is stopping here.

So once again, this is the ability of your immune system,

you're releasing the brakes on the immune system.

You're not treating the cancer directly.

You're treating the immune cells.

You're creating a benefit in some patients

that's lasting years after the treatment stops.

And this was the first time this was ever

shown in a solid tumor.

And it was outpatient treatment.

So unlike interleukin-2, which has so many side

effects it has to be given in the hospital,

CTLA-4 blockade could be given outpatient.

And patients could come in for treatment, get a benefit,

stop the treatment.

And the benefit, in some patients,

could last for a long time.

But the benefit doesn't come without some cost.

When you rev up the body's immune system,

sometimes the T-cells will actually

go after healthy cells.

And that's what you're seeing here on these slides.

So for example, you hope that when you activate the T-cells,

it's going to fight the cancer.

But sometimes, it can attack the skin.

You can see this patient here with red dermatitis, skin rash.

This is a brain MRI.

This is before CTLA-4 for blockade is given.

And after, you can see, what we're

looking at here in the circle is the pituitary gland.

This is the pituitary gland essentially being

attacked by the immune cells, which

can cause the pituitary gland to not function, which

is not good.

But it can be treated with supplements,

with hormones given.

This is also one of the more serious side effects.

This is looking at a CAT scan.

This white area here is an enlarged colon.

I'll show you a better picture here.

This is a picture of inside a patient's colon, who's

received CTLA-4 blockade.

If you've ever seen one of these before,

if you've ever seen your own colonoscopy,

this is a pretty swollen colon.

And that's because of the effects

of the immune activation.

Now, if detected early, these side effects

can be treated with immune suppression medications.

But they can be serious.

So this story is not just positive.

There are some side effects that we have

to manage on a regular basis.

One of the other concepts, and Arlene

didn't really go into this.

But she is so important in this part of the story, which we'll

talk about in a little bit, which is not only are there

breaks on your immune cells.

The cancer can actually create a braking system of its own.

It can actually create, I like to think of it as barbed wire.

Cathy described it as almost like a force field

in Star Wars.

The tumors can actually create a resistance

to T-cells that have detected them

as a threat, found their way into the tumor,

are about to kill the tumor.

And the tumor puts up this barbed wire,

or this so-called PD-L1, on the surface

of the tumor that takes active t-cell and says,

you're not going to kill me.

I'm going to stop you in your tracks.

Now it turns out, once this once these discoveries were made,

you can actually create antibodies

to block the activity of either PD-1 T-cells,

as Arlene was describing, or this PD-L1 defense that's

often on many tumor cells.

This was the first large trial from just five years

ago in the New England Journal of Medicine,

which looked at one of the first PD-1

blocking antibodies in humans.

The reason this study was so important is

the patients on this study could have had one of five cancers.

They could have had melanoma or kidney cancer, the two cancers

that I treat.

But they also could have had colon cancer, prostate cancer,

or lung cancer.

And one of the exciting parts of this story

is, for the first time, we started

seeing activity outside of kidney cancer and melanoma

with these drugs.

We started seeing activity in lung cancer.

And when we saw that, we knew this was going to be big.

Because if you could get lung cancer to respond,

if you get the immune system to wake up and kill a lung cancer,

there was going to be a large list of cancers that would also

be sensitive to the treatment.

And that story has played out.

What did we see early on?

Well, this is a patient here with metastatic melanoma.

You can see a very large mass here

that I'm circling in the neck.

After treatment, with this outpatient immune therapy,

the mass gets much smaller.

You're seeing a picture of the side

of the neck of this patient.

And what you notice here, I don't

know if you can notice it, it's a little hard,

the woman is actually losing pigment in her skin.

And why is she doing that?

Well, because the T-cells are detecting

the malignant melanocytes, the malignant melanoma,

but also the healthy melanocytes.

So she's losing skin pigment.

So this is an immune reaction, which is dramatic.

You can see it on the microscope as well.

But for the first time, we're seeing this in other diseases,

similarly to this one case.

What else did we see in that initial trial?

Well, this is one of our patients with advanced kidney


You see this very large mass here in the left upper quadrant

of the abdomen.

The mass is so big, it's actually

growing through the abdominal wall here.

With treatment, once again, outpatient [INAUDIBLE]

treatment, you can see the mass getting much, much smaller.

And most importantly, this response

lasted, meaning this patient is still

having benefit from treatment several years

after finishing the therapy.

This is a little bit dramatic.

And I apologize to those people who

may be a little bit squeamish.

But I wanted to show this, because this drives home

the power of PD-1 blockade.

This is a woman who is 78 years old,

who actually had a cancer of the tongue that spread,

unfortunately, to her breast.

You can see the mass here.

This big, bright, angry-looking red thing here.

This person got an outpatient PD-L1 blockade.

You can see, just after two weeks,

you can already see it less angry, less red.

Within three weeks, you can see it's

almost like it's drying up here, and essentially dying.

And almost back to normal within six weeks of treatment.

Why am I showing this sort of graphic story?

Well, it points to several important parts of this story.

The reason why this tumor is so angry

is the immune system is already there.

The immune system is creating an inflammatory reaction

at the tumor that's causing, probably, the redness,

and probably pain and discomfort.

So the T-cells are there.

And what do you do when you give the antibody?

Essentially, those T-cells can now kill the tumor.

And that's why the killing of the tumor is so quick.

Because the T-cells have already detected it as a threat.

And now, the barbed wire around the tumor is blocked,

and they can kill.

And that's one of the reasons why this class of drugs

is not just more active, but is also less toxic.

Because the T-cells you're activating are

sitting, for the most part, at the tumor.

They're already there.

They're ready to do some damage.

But they can't.

You're not activating every T-cell in the body.

With some of these older treatments, like I mentioned

IL-2, for example, you're activating a much larger number

of T-cells.

So you're creating much more side effects.

And the other interesting part of this story,

which I also wanted to mention, is this woman's 78.

These drugs are not more toxic as you give them

to older patients.

In fact, older patients tend to tolerate these treatments

very well and have often dramatic improvements.

This woman had failed multiple rounds of chemotherapy, which

is something we also see, is these treatments can work,

even after chemotherapy fails.

So Arlene showed something similar early on.

This is just a list of cancers where we're seeing activity.

And the list is long and growing all the time, which is good.

And in red are the places where we've

seen FDA approvals, where these treatments are now available.

So far, we've had 18 randomized trials.

And 16 have been positive.

So we're on a roll, although that's probably not

going to continue.

One important question I get a lot is,

do these work in some other cancers?

Well, so far, we're not seeing a lot of activity

in some very common cancers.

Like the most common form of breast cancer,

for example, and a lot of prostate cancer.

And you might ask why that is.

And we're still trying to figure that out.

But it may be because the typical breast cancer does not

generate as dramatic and immune response.

So we don't have the potential of activating those T-cells

at the tumor, like in the last case I showed.

But we're still trying to figure that out.

But I don't want to leave the impression that these

treatments work for most cancers, because they don't.

And they don't work for most patients

with cancers on this list.

So just following up, I was asked

to talk a little bit about combinations.

Oncologists are notorious for thinking,

if one drug is good, well, two drugs has got to be better.

And that's not always the case, because it can sometimes

lead to more side effects.

But there is a long list.

Because PD-1 blockade is relatively tolerable,

you can combine it with Dr. Wu's vaccines or Dr. Ritz's

cellular therapy, or chemotherapy and radiation,

without making those approaches too much more difficult.

And those are being done right now.

There are literally hundreds of clinical trials

ongoing around the world.

I just wanted to focus on one in the last few seconds

that I have.

This is combining the two approaches

that I mentioned, once again in melanoma.

This is combining CTLA-4 blockade,

releasing the brakes on the immune system,

and also adding PD-1 blockade on top.

So covering the barbed wire on melanoma.

And what does that produce?

Well, if you just release the breaks, about 16%

of patients with CTLA-4 blockade will

have a lasting benefit, a lasting

remission of their disease.

But if you do both at the same time

and you look two years later, at least in this trial

that we did, half of the patients hadn't progressed.

Meaning, essentially, their cancer

hadn't grown in over two years.

And some of these patients are in remission of their disease.

So whether we can do better than this, we'll have to see.

But it's truly exciting that, potentially, combinations

might take us a step beyond what we get with just PD-1 alone.

So we're in the era of immunotherapy.

And it certainly made a big splash

in our scientific journals.

Because of the impact across a variety of cancers,

it's also made an impact in the lay press.

But I just wanted to say two important things.

One, this would not have been possible,

this story, without the taxes that you folks pay,

that fund places like the NIH.

A lot of the money early on they'd

funded that led to these discoveries

was actually money that went not to study cancer,

but went to study things like autoimmunity and allergies

and all that sort of stuff.

A lot of that money went to people

like Arlene Sharpe, who is one of the three or four most

important people in this story.

She won't tell you that, because she's very humble.

But a lot of this wouldn't have been possible

without her work and her husband's work, Gordon Freeman.

So keep the support for research.

And hopefully, we can do better work in the next decade.

Thank you very much.


I'd like to invite all of our speakers to sit.

And we have a number of questions from the audience

and from Facebook that have been shared with me.

And so we'll have time to discuss a few of them.

The first question, several people

have asked how will the cost to patients of immune therapy

or vaccines compare with chemotherapy?

And will this be covered by insurance?

Do you want to take that?

So, go ahead.

Well, so I think--

well, David can speak to this as well.

I think the immune therapies that have been approved,

the anti-PD-1 antibodies and CTLA-4.

So these are now FDA-approved drugs.

And they are new treatments for lung cancer, for bladder

cancer, for [INAUDIBLE] cancer.

Those are being covered by insurance.

I'm sorry, so those are being covered by insurers.

They are expensive.

But I think almost any new therapy,

whether it's for autoimmune disease or for asthma

or for cancer, these are now very expensive agents.

The cellular therapies that I've talked about

have not yet been approved.

We do expect that they will be approved sometime this year.

We don't yet know how much they're going to cost.

We think they're going to be very expensive.

And we really don't know how expensive.

We don't know how insurers are going to deal with this.

It is a big challenge for the field.

Because just the cost of one of these drugs

can be tens of thousands of dollars.

Because it can be given month after month.

The good news is it can be given chronically.

The side effects don't build up.

But if you can stay on the drug, the costs continue to rise.

When you talk about adding expensive drugs together,

that will be a major issue.

And as a field, it's up to us to make

sure we're only giving these drugs to the patients most

likely to benefit, as opposed to giving them

to everyone, which is a pharmaceutical approach

to these things.

We also shouldn't be giving them longer than need be,

which is part of the reason why I focus a lot on creating

remissions of the disease.

If you can create a remission of a cancer,

well, then you can stop the treatment.

And also, that means the side effects stop.

That means the cost stops.

So as a field, I think we need to be pushing for treatments

that move us in that direction.

We generate that occasionally with patients.

But not enough yet.

And I think when you go think about that

in terms of the cellular therapies,

they're expensive because you're making a drug, a cell for one

individual patient.

And it's a lot of work.

It could be three weeks or three months

to actually do that work to generate

for that individual patient.

That builds up the expense.

But on the other hand, it is a one-time treatment.

It's a living drug that you really

expect to last for a long time.

And so that is going to be one of the mitigating factors.

I also think it's hard to know, I

think Jerry talked about saying that it's hard to know.

And I think when treatments are in their infancy, when they're

being tested, when they're in clinical trials,

they are expensive.

In the beginning, it is expensive

because there's a lot to test and a lot to optimize

and a lot to figure out.

If they do hit the mainstream, it's

hard to know how those costs will come down over time.

OK, maybe we should move on.

Another question that several people

have asked, how are people looking at genetic differences

in the treatment of cancer?


So, yes.

Very much worth thinking about how

to use the genetic information, on one hand to individualize.

I talked about the new antigen vaccines.

So that is one way to individualize our care.

Another way, another aspect that has been really

touted, and there's a lot of thought going into this,

is there a way that we can genetically test and figure

out, out of all the different menu of options

that we presented, what is the most efficacious way

to proceed?

So for one individual with one type of cancer,

it may be a vaccine.

For another person, it may be cellular therapy.

For another person, it may be a combination, together

with checkpoint blockade.

So I think we're at a time where we're all

thinking about all of that.

So we have, on one hand, lots of genetic data.

On the other hand, we have lots of developments

in immunotherapy.

And I think a major thrust is how to bring that all together.

You hear a lot about generically what are termed biomarkers.

And a lot of the thrust of biomarker research

is to identify a marker of an individual patient that

will tell you what the best treatment would be.

Should it be this drug because the cancer has

this specific genetic mutation?

Should it be an immune therapy because there

are T-cell infiltrates that are already there,

like David talked about?

So I think being able to predict in a better way who

will respond to which treatment will also

help with the cost issue.

Because it is very expensive to give drugs.

It's even more expensive to give drugs that don't work.

Another question that we had is, do you

see CRISPR playing a role in future treatment of cancer

in immune approaches, or other approaches?

So CRISPR gene editing, I think, will play a role

in being able to genetically engineer the cells that we're

going to use for treatment.

So I think that's already being used and being applied.

Gene editing is going to be effective.

But you need to have a cell in which to edit genes,

and to work through.

And you need to have a laboratory,

like the one I showed you that we're building,

that will allow you to do that in a safe way.

So I think gene editing will really

facilitate the engineering of cells

that you really want to target to a specific cancer.



Another question.

Do patients have to fail other therapies before receiving

CAR T-cell therapy?

Right now, yes.

I think right now, it's still an experimental treatment.

And it's really, before you know that something's working,

you really need to know the patient

that you have has exhausted other known treatments that you

know work.

So it's really only in patients who

have failed other treatments that these cells have

been generated.

In the future, I think if they get approved

and if they're shown effective for patients

who fail treatment, then it makes sense to say,

well, is it is it possible to use it earlier

on in the course of disease?

We do a lot of stem cell transplants.

And initially, stem cell transplant was exclusively

for patients with leukemia or lymphoma,

who had failed all other treatments, as a last resort.

It turns out the results are much better if you actually

treat patients before they failed

all the various treatments.

So I think it may well move into an earlier phase of treatment

in the future.

So another question.

Vaccines are used often before illness occurs.

Do you think we can have personalized cancer vaccines

when an individual is a 100% healthy?


I mean, I agree.

In infectious disease, we always think about vaccines

to prevent illness.

In cancer, that's not where we've been.

But along the same lines as what Jerry mentioned,

which is vaccines are going to work the best when

your immune system is more competent and ready to do

some work.

One thing in the near-term that we can think about

is how to bring vaccines earlier in the course of disease,

not way toward the end when it's so hard to mount

any sort of immune response.

In terms of prevention, I think we're still

really far from that time.

But I do know that there's a lot of interest

out there about thinking and conceptualizing

how we can find those premalignant lesions,

for example, and trying to find specific genetic determinants

of the premalignant lesion.

And maybe we could vaccinate against that.

All concepts and things to test.

But certainly, it would be a promising approach.

Cathy did have this on one of her slides,

but HPV vaccine is actually a cancer vaccine.

And probably, it's really not used enough.

Cervical cancer can really be prevented over 90% of the time

by immunizing young people to the virus that

causes that cancer.

Another question.

Have there been extensive studies of healthy people

with no cancer to identify immune system

differences between them and those with a cancer diagnosis?

And how can those findings help develop immune therapies

for those who need them?

You're pointing at me?

I don't know that anyone has done that.

I don't think so.

So I think there's been a lot of work, on the genomic side,

to look at variation in cancer responses

and degree of immune infiltrate over hundreds and thousands

of samples across many, many different cancers

that have been sampled.

In terms of comparison to normals,

I do know that, with respect for example to neoantigens,

it is true that if you take [INAUDIBLE] neoantigens

and you test them in normal donors,

there will be very, very robust responses.

There will also be responses in cancer patients.

But they are less strong and less broad.

But with something like a vaccine,

our hope is that we can put a little fuel into it

and generate those broad responses

that we see in normal donors.

I mean, it is known that patients

who are very immune deficient are more susceptible to cancer.

But it's not necessarily just one type cancer.

And their immune system has a lot of redundancies in it

and overlapping functions.

And so often, someone who might be

immune deficient in one area, other cells can compensate.

So you don't actually notice it very much.

It's really only in patients who have

very profound immune deficiency that you see

this very high risk of cancer.

So we'll have one last question.

How is the polio virus being used to combat cancer?

So I think that there are--

I'm not sure how much of it is polio virus, or only

polio virus.

But there are what are called oncolytic viruses that

are now being used.

So these are genetically-modified polio

virus, or other herpes viruses, that are genetically engineered

to allow them to infect the cancer cells

and kill cancer cells, but not allow them to spread.

And what that does at the cancer itself

is it releases all the other antigens.

And so there's an immune response that's

generated against the virus.

And then also, the immune response

against the dead cancer cells, that

is now one form of a vaccine, if you will.


We're using that approach in melanoma.

There's an actual FDA-approved drug

that's actually a herpes virus that's injected into a tumor

on the skin, or under the skin, to create

that local immune response that hopefully will generate a more

systemic response throughout the body.

Most of the patients who benefit from that injection

have relatively localized disease.

And one of the things folks are trying to do

is to take that modified virus, and combine it

with the checkpoint blockade, in hopes

of expanding the benefit that we see beyond to more

critical organs in the body.

As far as the polio story goes, I

know it's been looked at at Duke with patients

with brain tumors.

And there, they've seen mixed results.

And they made it all the way to 60 Minutes,

helping a couple of people.

But also, several people didn't do so well on that trial.

I don't know whether it's been expanded beyond that.

But the concept is the same idea,

which is to take a tumor that's not being recognized

by the immune system, create a localized immune response that

will then lead to more recognition,

more killing of the tumor by the immune system.

Well, I think, unfortunately, we've run out of time.

So I'd like to thank the audience for their attention

and thoughtful questions, and the speakers once again.