From the Howard Hughes Medical Institute, the 2003 Holiday Lectures on Science.
This year's lectures, Learning from Patients, the Science of Medicine, will be given by
Dr. Burt Vogelstein, Howard Hughes Medical Institute Investigator at Johns Hopkins University
School of Medicine, and Dr. Huda Zoghbi, Howard Hughes Medical Institute Investigator at Baylor
College of Medicine.
The first lecture is titled, Research Mechanics, Putting the Breaks on Cancer.
And now, to introduce our program, the President of the Howard Hughes Medical Institute, Dr.
Thomas Chaik.
Welcome to the Howard Hughes Medical Institute and to our 11th Annual Holiday Lectures on
Science.
This program is being webcast live, and in addition, we have almost 200 students in the
audience from the greater Washington, D.C. area high schools.
This series is going to be available on DVD starting in the spring of 2004, and you can
order these DVDs through our website.
Now, the website is also a great place to learn about all of the educational programs
supported by the Howard Hughes Medical Institute, and in addition, about the research activities
of our 323 Howard Hughes Medical Institute investigators who run laboratories across
the United States.
This particular series is called, Learning from Patients, the Science of Medicine.
And this is really a two-way street.
We're going to talk about the way that biomedical scientists derive the inspiration for their
research from the patients that they see in the clinic, but then in the other direction,
after figuring out the basic details of the disease process, they try to turn around and
find ways that they can use this understanding to improve diagnosis and hopefully eventually
treatment of these diseases.
Now, for this exploration, we couldn't have better guides than our two speakers for this
series.
They are both MDs.
They both continue to have contact with patients while they're doing their cutting-edge research.
Buddha Zoghbi is interested in patients that have problems with their nervous system.
You'll hear about that tomorrow.
Today, we have Bert Vogelstein, who's going to be talking about genes that are involved
in the process of cancer, and also particular interest in translational research.
This is a term that is used to describe trying to translate or apply the information that's
learned in the research laboratory back into the clinic.
Now, Bert's going to talk for 40 minutes.
He will ask for questions halfway through the lecture and then again at the end, so
you can already be thinking as he's talking about what you might ask.
The title of his talk is Research Mechanics, Putting the Breaks on Cancer.
I did a residency in pediatrics.
That's when I first met my first cancer patients.
That made a large impact on me.
At that point, no one had any real idea what cancer was.
It just seemed like some mysterious disease that affected people.
Could come from outer space, as far as we know.
It was really very hard to be hopeful with the parents or with the kids.
So at that point, I think my goal was solidified to doing research on cancer, and I did my
postdoctoral fellowship in a cancer research laboratory, and I've been doing cancer research
ever since that time.
There's been a huge change in terms of understanding of cancer.
A real revolution.
There's still lots of details to be learned and some major things still to be learned,
but a general outline of how cancer's form has been generated, and you can kind of summarize
that revolution in a single sentence.
The sentence would be, cancer is, in essence, a genetic disease.
So from pessimism, or at least a feeling of hopelessness in the 70s, has turned to real
optimism in the new century, because once you understand a disease, history shows that
it's only a matter of time before you can do something about the disease.
And what's critical for the future is to get young people with creative ideas to work on
this problem and to solve it in their lifetime, and I think that's really possible now.
I hope my lectures will accomplish a couple of things to educate the students about what
cancer is.
I want to convey to them some of the excitement that's now widely felt in the scientific community
about cancer, and I want to stimulate those of them who are thinking about science as
a career to think harder about it and to understand that this is a field where they really can
make an impact, a difference on millions of people worldwide.
It's just a wonderful way to spend your life.
Thank you.
I'd also like to welcome you here to the Howard Hughes Medical Institute.
You are an essential part of this program.
There will be hundreds of thousands of high school students who view these lectures in
the coming years, but you have the advantage in that you're the only ones allowed to ask
questions to us, because the rest will only see the CDs.
So you represent all of the other high school students, and please think about what questions
all high school students may have, including those that you might have.
And as Dennis said before, we have these wonderful lights.
Just hold up a light when you want to ask a question.
We're going to reward some questioners with T-shirts.
So look out.
I'm also going to ask you some questions as we go along.
I'm going to try to make this an interactive program.
I don't know many of your names, but I'll just pick random names and hope that strikes
a bell.
Any of you named Jim or James?
That's good.
How about Ann or Anna?
Anybody named Mauricio?
Got it.
Niorica?
Okay.
I think we got all the common names covered then.
So let's start with a patient that was mentioned in that tape, Melissa.
She was my first patient.
This was really tragic, not only because Melissa was such a beautiful little girl and because
she developed cancer, but also because at that time, which was in the 70s, we really
knew very little about this disease.
It seemed just to be something which came from outer space.
And it was so hard to be optimistic that we would ever be able to do anything for kids
like Melissa or older patients with cancer until we understood something.
Now 30 years later, we do understand something.
There's been a real revolution in cancer research.
And I'm going to try to introduce you to that revolution in these two lectures.
There will be four chapters.
The first chapter will be the nature of cancer.
What is cancer?
The second will be an explanation of the genes that cause cancer, what they are and what
they do.
The third will be focused on how we use that information to prevent cancer in the future.
And the fourth will be how we might be able to use that information to treat cancers again
in the future.
And let's begin with the nature of cancer.
There are lots of different types of tumors.
And I'll tell you the difference between cancer and tumors in a second.
But cancers are a form of tumor.
This is another form of a tumor.
But it's benign, not malignant as cancers are.
Benign tumors are not life threatening.
They're not particularly dangerous.
You can see that small tumor on this young woman's forehead.
You might call it a birthmark.
Probably all of you have one or two of these scientifically.
They're called nevi or nevus.
And they're really not very important.
Some people even think they're beauty marks.
Another kind of tumor is shown here on the vocal cords.
There's two of them.
These are also benign.
They're not dangerous.
They probably wouldn't cause these patients any problems unless the patient was Britney
Spears or Michael Jackson.
But those people have more important problems to worry about than their voice at present.
Tumors are not, benign tumors are not defined by their size.
I showed you ones that are small, but here's a slightly bigger one.
Growing out of the salivary gland of this gentleman's face, this gentleman was a Native
American, lived on a reservation, and in fact wasn't able to get to a hospital for a long
time.
It took about 30 years to grow a tumor this size.
But it's still benign.
So when he eventually reached the hospital, the surgeon was able to cut out the tumor
and the patient went home and lived another 30 years.
So it's not size that defines the difference between a benign tumor and a malignant tumor.
Another name for a malignant tumor is a cancer.
It's not size.
What is it?
It's something called invasiveness.
On your left is another birthmark, a nevus.
It's a tumor of pigment cells in your skin that determine this color of your skin.
And the benign tumor is round, it's circumscribed, it has smooth borders.
It can be removed easily by a dermatologist or a surgeon.
But the one on the right is malignant.
And the difference is not so much the size, but the fact that it's not smooth any longer.
It's invading.
It's invading through the skin laterally.
And more importantly, it's also invading in three dimensions.
So it's invading underneath the skin.
It will eventually hit a blood vessel or a lymphatic.
And then tumor cells can go inside the vasculature, that is the vessels, the blood vessels, and
travel to other parts of the body.
That is called metastasis.
When a malignant tumor reaches a blood vessel or lymphatic, travels throughout the blood
stream, reaches other organs like the lungs or the liver, and starts to form a new tumor,
a kind of colony of the original tumor in the new place.
And that's why people die, because of metastasis.
The primary tumors can almost always be removed surgically.
But if the tumors aren't detected until after they've metastasized, then it's too late.
The Greeks recognized this invasiveness of cancer very early on, and that's why they
called it cancer, which means a crab.
They saw the invasive arms of the cancer, such as in the melanoma I just showed you,
and that reminded them of the claws of the crab.
And this animation shows what happens in a tumor.
There's cells which are dividing abnormally.
We'll see how in a few minutes.
They keep dividing at the expense of the normal cells that surround them.
And in three dimensions, once they divide enough, they start to pop out.
You can begin to see them with your naked eye, not just under a microscope.
And then they recruit blood vessels.
They need to recruit blood vessels in order to grow to a size that's greater than just
a tenth of an inch in diameter.
The blood vessels supply oxygen and nutrients.
And the last stage is the metastatic stage.
That blue cell in the middle represents a metastatic cell that has come from the primary
tumor, invaded into the blood vessel, and now can also get out of the blood vessel,
invade another tissue, and form new cells there.
The end process of the metastatic lesion, the metastatic process, is shown here.
On your left is a normal lung, an X-ray, and on the right is a picture of a patient who
originally had a breast tumor that had metastasized to the lung.
So the tumor cells from the breast had invaded a blood vessel and traveled to the lung where
they have formed nodules composed of breast cancer cells.
Another example is provided by colon cancers.
Colon cancers start in the colon, of course, and they invade a different kind of vessel.
It's called a portal vein, and it connects the intestines to the liver.
The cells which invade wind up in the liver.
And on this next slide, you can see a normal liver on your left, and on the right is a
liver that has been invaded by colon cancer cells, metastatic colon cancer cells.
Each of those nodules represents an individual cancer cell which has made its way to the
liver and started to multiply.
Each of those nodules now contains several hundred million cells, and the liver will
eventually be totally replaced by cancer.
A person can't live without his liver, so that's why this person will die.
Now there are lots of kinds of cancers.
Cancer is often thought of as a single disease in the newspaper, but it's really not.
It's hundreds of different diseases.
For every person in this room, there's at least one form of cancer.
For every cell type in the body, there's at least one form of cancer.
And they're different.
For example, a breast cancer is different than a melanoma or a colon cancer.
On your left is a normal breast, a mammogram, the kind of test that women are expected to
get so that breast tumors can be detected before they're metastatic.
And on your right is an example of a breast tumor, a cancer that has been detected by
mammography.
Another kind of cancer that particularly affects young people is leukemias.
On your left is a normal blood smear, that is, a drop of blood is taken from a patient
and smeared on a slide, really just smeared and then stained with dyes.
You can see all the red cells.
Most of the cells in the blood are, of course, red cells.
But there are also a few white cells.
Those are normally there, and they help prevent infections, guard against infections.
But in a blood smear from a leukemia patient, there's still a few normal-looking white
cells.
But most of the white cells are these large and abnormal forms which represent the leukemia
cells.
And because these cells are already in the blood, they're a cancer of blood cells.
They spread immediately throughout the body, and they have to be treated immediately for
the patient to survive.
Now, despite the fact that all of these tumors are different, in some ways, they also have
something in common.
And what they have in common is that there is an abnormal ratio between cell birth and
cell death.
So, in a normal adult tissue, say the skin or the intestines, cells are always dividing
in the skin and the intestines, there's always a very specific ratio of cell birth and cell
death.
Can anybody tell me what that ratio is?
Just raise your hand, or your lighter, one, exactly, has to be exactly one.
For every cell that's born, one cell dies.
If too many cells die, more cells die than are born, the tissue will shrink, it will
atrophy.
On the other hand, if more cells are born than die, that's a tumor.
That's the physiologic definition of a tumor.
And it doesn't require much of an increase of that ratio in order to get a big tumor
over time.
You can think about this like interest.
Any of you working in a lab this summer?
Okay.
So if I were to ask you, if I were to offer you a million dollars to work in my lab over
the summer, you would probably take it, right?
Okay.
I'm not going to offer you that much.
Suppose I offer you a penny, but I tell you that every day you'll get 25% interest on
that penny.
So after one day, you'll have 1.26 cents, and after two days, you'll have 1.6 cents,
and three days, 1.9 cents.
Would you rather have the penny with interest or the million dollars?
Everybody.
Who would rather have the penny?
100 days, one summer.
And who would rather have the million dollars?
Okay.
Good.
So most of you chose wisely.
A few of you chose poorly.
Those of you who took the penny would end up with $49 million.
Those of you who took the million, you wouldn't do so bad, but you'd still only have a million
after the end of the summer.
So now, this same kind of thing happens in tumors.
It only requires a slight difference in the ratio of cell birth to cell death to develop
a very large tumor over time.
And since most tumors take 20 to 30 years to develop, it really only requires a difference
of 1 percent in that ratio to get a huge tumor over decades.
And that is the common underlying theme of tumor genesis.
Just to go over that again in graphic form, when a normal cell divides every normal cell,
two cells are, of course, formed, but one of those cells will die.
So cell division, you get two cells, one of those cells eventually dies, leaving the net
the same as the beginning, one cell.
But in a tumor, it's quite different, as again shown on this movie.
The tumor cells, which are mutant, we'll see in a few minutes why they're mutant.
The cell divides, but one of the daughter cells doesn't die.
They'll keep being more cells, more cells, more cells over time.
And that starts a benign tumor.
Again, to review the difference among tumors, there are three kinds, all called tumors.
The first is benign.
It's not dangerous, it's not life-threatening, and it's not invasive.
The second type is malignant.
This is dangerous.
It's locally invasive.
That means it can invade surrounding tissues.
Eventually, the malignant tumor may become metastatic, in which case it can travel to
other parts of the body, and that's when the real problem occurs.
In the next chapter, we'll discuss the molecular basis for this process, but now we'll quit
and ask for questions.
How much time do we have for questions?
Five minutes.
Good.
Just raise your hand or light.
Yes?
How is it possible to tell if a benign tumor is going to travel the internet, say?
Can you all hear that question?
The question was, how do you tell if a benign tumor is going to become malignant?
The only person who can tell is the oracle.
You can't.
It's just stochastic, which means it's random.
You can look at a benign tumor, you can estimate the probability that it may turn into a malignant
tumor, but you can't tell for sure.
In the next chapter of this lecture, I think you'll understand why.
Other questions?
Yes?
You said the cancer cells, they travel through the blood vessels.
Are they selective in the tissue they choose to?
Are the cancer cells selective for the tissues they choose?
Yes, they are.
Sometimes investigators call that the seed in the soil.
Certain kinds of cancers only grow in certain tissues in metastatic form.
For example, prostate cancers generally go to the bone marrow, whereas colon cancers
generally go to the liver.
Do we know the reason for that?
The reason for that, no.
We don't know the reason for that.
In fact, we don't understand the basis for metastasis in general either, and that's what
some of you are going to help us find out over the next 10 years.
Yes?
My name is Crystal Shavers, and I attend Margaret Murray Washington Career Senior High School.
Is it more difficult to control a massive malignant tumor or a metastatic tumor?
The question about controlling a massive malignant tumor or a metastatic tumor.
Massive malignant tumors, first of all, they usually don't become really massive as long
as people have access to medical care, because people notice them if they're really big.
But there are some tumors that have been taken out that have weighed 250 pounds.
These are morbidly obese people, and they occur inside the abdomen, and they thought
they were just getting fat, but they weren't.
They had a tumor growing inside, and that can be removed surgically.
It's really not a problem until metastasis, because if the metastatic cells invade the
liver or the lung, you can't take out somebody's liver or lung, so then the person is in trouble.
That's really the problem.
We have time for one more question.
Yes?
Hi.
Good morning.
My name is Tiffany Fields.
I attend Spangon Senior High School, and the question I wanted to ask was, when you stated
earlier that the tumors are caused when the cell keeps multiplying, is there any way that
we can paralyze that cell that's multiplying so that it won't be so many?
Right.
Good question.
Can we paralyze cells so that they'll stop multiplying?
In fact, that's what cancer therapy is, is an attempt to try and stop cancer cells from
developing.
There's also something called preventive drugs, which try to prevent even early tumors from
multiplying rather than shrinking tumors, which are large.
All cancer drugs do that now to some extent.
Unfortunately, they also stop normal cells from growing, so they're toxic.
If any of you have ever seen patients who are treated for leukemia or cancer, you know
they lose their hair.
They feel bad.
They're nauseous.
That's because the drugs so far work on both normal cells and cancer cells.
They're not selective.
Probably the next generation of drugs will be more selective.
Okay.
Thank you.
I'm not going to throw them, but can you hand them back there to those questioners?
Thank you.
All right.
Let's go on to part two.
Now, when I was treating Melissa back in the 70s for her leukemia, we again had no idea
of what caused her cancer.
There were lots of theories.
Theories are cheap.
Some people thought that cancers were caused by viruses.
Others thought that cancers were caused by bacteria or other infectious agents.
Still others thought there was a breakdown in immunity.
Most forms of cancer don't occur in people until they're middle-aged or older.
It's also known that immunity tends to decrease with age.
So is a reasonable theory that perhaps when immunity breaks down, cancers pop up.
Still others thought that there were mutations in specific genes that caused tumors.
Now, it turns out in general that this last theory is the right one.
That's the basis of the cancer revolution.
Mutations in specific cancer genes cause the disease.
So cancer is, in essence, a genetic disease.
But it's different than most other genetic diseases with which you're familiar.
For example, let's look at some common genetic diseases.
DMD is a gene which when mutated causes muscular dystrophy.
I assume you all know what muscular dystrophy is.
It occurs in children, causes them muscle problems.
It's a very severe disease.
Now, some of you probably live in rural communities.
In Bowie or Gaithersburg, anybody live there?
Anyone live in an urban setting, Washington, D.C., city?
Now, the air is probably cleaner in the rural environment.
You don't have all the car fumes.
Environment may be better.
If you had the same mutation in the DMD gene, would that alter the disease that you would
get?
Would the muscular dystrophy be better or worse depending on where you lived?
Anybody?
Who thinks it would be better if you lived in Gaithersburg?
Who thinks anything would be better if you lived in Gaithersburg?
The truth of the matter is, of course, it doesn't matter where you live.
If you have a mutation in the DMD gene, you're going to get exactly the same disease.
It's totally independent of environment.
The disease is only dependent on the specific mutation.
A somewhat more complex case is caused by mutations in a gene called LDLR, which predisposes
to heart disease.
But not everybody who has a mutation in LDLR will get heart disease.
It's not like muscular dystrophy, where everyone who has a mutation in DMD will get muscular
dystrophy.
It's only if a person with an LDLR mutation is exposed to specific environmental influences,
in particular diet, fats in the diet, that they will get heart disease.
And the severity of the disease is then determined both by the genetic component, this mutation
that they inherit, other genes, and their diet.
So it's more complex.
And cancer is even more complex than that, although it's related.
Some patients inherit a mutation in a gene which predisposes them to cancer.
They won't necessarily get cancer, but they are more likely to get cancer than a person
who doesn't have an inheritable mutation in this gene.
And there are lots of genes.
This is just an example.
One of them, MLH1.
In order for them to actually develop a cancer, they need more stuff.
And in this case, what they need is more mutations.
No single mutation actually causes cancer.
It's an accumulation of mutations in the same cell until the proverbial straw that breaks
the camel's back.
And enough mutations are accumulated, then cancer results.
And you can think about this process in terms of waves of clonal expansion.
Pretend each of these white balls represents a normal cell.
Suppose one of those cells becomes mutant.
Then it will proliferate.
Its ratio of cell birth to cell death will increase to greater than one.
And it will proliferate over and above that of the normal cells.
That is called a clone.
It's different than a clone like Dolly the Sheep.
That's an organism clone.
This is just a cell clone.
But it's the same idea.
A cell with the same genetic characteristics is proliferating and making identical copies
of itself.
But that's not the end of the story, because once this mutant cell forms a benign tumor,
it can also get more mutations.
Or a cell from the benign tumor can get another mutation.
And through a second wave of clonal expansion, expand to become malignant, invasive.
And of course, the malignant cell, or one of the malignant cells, can subsequently accumulate
another mutation and become metastatic.
Any of you see that movie called Multiplicity with Michael Keene?
Well it's a bit like that.
Remember Michael Keene wanted to clone himself because he didn't have enough time.
Turned out not to be such a great idea.
But he cloned himself.
The first clone was OK.
But the next clone was worse.
And with each successive clone, there were more mistakes that were made.
That's like mutations.
Until eventually you got a very stupid Michael Keene.
It's the same kind of thing that happens in tumors.
Successive waves of clonal mutations until you get a very bad one.
In this case, not stupid, but evil.
Now it's important to recognize that these mutations, like mutations that cause muscular
dystrophy and heart disease, are sometimes inherited.
But usually they're not.
Cancer in general is not a hereditary disease.
There are cases of cancer which are hereditary, as we'll discuss in a few minutes.
But generally the mutations occur after birth.
Now the only mutations that are hereditary are those that occur in egg or sperm cells.
They can be transmitted from father or mother to son or daughter.
Other mutations that occur in any of the other cells of the body obviously cannot be transmitted
to a son or daughter.
The germ cells, the egg cells and the stem cells are only a very small proportion of
the total cells, the total trillion cells in the body, 50 trillion cells.
There are only a few million germ cells at any one time.
So it's just a really small fraction of germ cells.
All the other mutations that occur in other cells aren't important.
They're not hereditary, but they can cause cancers.
What are these genes that get these mutations, either inherited or after birth?
There are only three kinds.
One kind is called an oncogene.
These genes normally, because they're all normal until they're mutated, everybody has
them, everybody has oncogenes, but they're normal and they normally stimulate growth
or what you need to grow to live.
But if an oncogene becomes mutated, it's like having an accelerator in a car that's stuck
to the floor because there's a heavy weight on the foot.
The car continues to go even if the driver wants to stop it by lifting her foot off the
accelerator pedal.
That's just what a mutation in an oncogene does in a cell.
It makes the cell continue to grow, whereas normally without the mutation, the cell would
stop growing because it's being controlled properly.
This analogy works, too, for the second kind of gene, which is called suppressor gene.
These are the cell's breaks.
These normally inhibit growth.
A mutation in a suppressor gene is much like having a dysfunctional break.
Just as cars have more than one break, they have a foot break, they have a hand break
or emergency break, and if all else fails, you can pull the key out of the ignition.
Cells have multiple breaks, too.
It's only when several breaks plus an accelerator or two all become dysfunctional in an automobile
that the car spins out of control, and it's the same in a cell.
It's only when several of these genes become mutated, don't work properly, that the cell
spins out of control and becomes a metastatic cancer.
There's a third kind of gene, which doesn't directly cause tumor genesis when mutant,
but only indirectly does it, and these are repair genes.
Having a faulty repair gene is like having an inept mechanic.
Can't fix the mistakes that are made, and cells are always making mistakes, always.
Now, you'd think that with five billion years of evolution, you'd have evolved a system
that's not so prone to mistakes, so we wouldn't have cancers through these mutations.
Can anybody suggest why evolution hasn't gone that far and made DNA polymerases, et cetera,
that are perfect?
Yes.
That would eliminate the variety in the genetic coding of the species.
Mistakes are part of evolution.
In fact, you can look at cancer as a side effect of evolution.
If mistakes weren't made, we'd all be amoebas.
We wouldn't be here in this auditorium.
Unfortunately, cells are still making mistakes, and they can cause disease, both hereditary
diseases and cancer.
Now, are our cells still making mistakes now?
Krista, somebody named Krista in the audience?
Krista McAllister?
Okay.
Look behind you.
There's a guy named Dan.
Is he a mutant?
Look carefully.
Look.
You're sure?
All right.
Well, you're wrong.
He's a mutant.
We're all mutants, right?
Because the egg and sperm that gave rise to us, there was a mutation, at least one, probably
a few.
Now, those mutations didn't occur in genes which caused disease, hopefully, but we all
are mutants in that sense.
Now, let's look for a minute at these genes in more detail, and we'll start with suppressor
genes.
Suppressor genes are important because they cause many forms of familial cancers, inherited
cancers.
In the United States this year, about 135,000 people will develop colon cancer.
In the world, about a million people this year.
Most of them, 85 to 90 percent, they're non-familial.
That is, they'll be the only people in their family to develop colon cancer.
No one else has.
Their parents, their grandparents, their siblings, their cousins.
So they're non-familial.
But a small fraction, roughly 10 to 15 percent, occur in a familial pattern.
And there are two major forms of familial colon cancer.
One's called FAP for polyposis, and the other one's called HNPCC.
The operative letters are NP for non-polyposis.
A picture of a colon from a patient with polyposis is shown here.
You can see all of those bumps are polyps.
Polyps are just benign tumors.
Sometimes they're called adenomas.
That's why FAP actually stands for familial adenomas polyposis, but we'll just call it
polyposis for short.
And this patient has lots of these polyps.
They're benign tumors, but there are lots of them.
In fact, there are roughly 5,000 such benign tumors in the colon of each patient with polyposis.
Some of those tumors eventually are going to progress to cancer.
So patients, unless they're treated for this disease, will generally get colon cancer by
the time they're in their 40s.
They usually start to develop the polyps when they're in their teens.
This syndrome is caused by a classic tumor suppressor gene called APC.
It's just a cellular break that goes awry, can be inherited in a mutant form, and cause
this disease.
The other form of hereditary colon cancer is caused by repair gene defects.
That form is called HNPCC non-polyposis.
And the difference between non-polyposis and polyposis is just what its name suggests.
There are lots of polyps and polyposis, but in HNPCC, the non-polyposis, there's usually
only a single tumor.
A single tumor shown here, a large tumor, in fact a cancer.
Not thousands of polyps, just a single tumor.
And the genes that cause this disease are called mismatch repair genes.
This animation shows how they work.
There's a polymerase, DNA polymerase, which copies both strands of the DNA, the top strand
and the bottom strand.
Sometimes those two strands are called Watson and Crick strand.
But they're not perfect, as we just mentioned.
Sometimes they make mistakes.
And the kind of mistake they might make is, you'll see in a second, to incorporate the
wrong nucleotide.
Normally, there's going to be an A opposite a T and a C opposite a G. But suppose it makes
a mistake and copies a T where a C should be.
That should be GC, but now there's a T. So that's a mistake, a potential mutation.
Unfortunately, cells have repair systems that can erase those mutations.
And those repair proteins indicated here called MSH2, MSH6, MLH1, PMS2, the names don't matter.
What's important is that they recruit another enzyme called exo-1, exonuclease, which chops
off the mutant strand.
And then it allows a DNA polymerase to come by and synthesize the correct strand, thereby
fixing up the DNA and making it normal.
Now if a patient doesn't have a proper mismatch repair system, then the cell with that defect
will accumulate mutations much more quickly than a normal cell.
And that's why these patients with HNPCC develop cancer.
They have defects in this repair system.
And one of the things that's important about this defect is it accelerates, telescopes
the whole process.
In polyposis, the whole scheme of mutations, the multiple mutations in addition to the
inherited one that are required to get to a malignant or metastatic cancer generally
takes 20 to 30 years.
So patients don't get tumors and don't get malignant cancers or die from the disease
until they're in their 40s.
But in non-polyposis, this whole process only takes a few years.
From a small benign tumor to a malignant tumor, only three to five years.
The mismatch repair system in essence functions like a spell checker on your word processing
program.
Now let's put all this together.
And in colorectal tumors, starting from normal cells, the first mutation might cause a very
small microscopic small tumor, which then could grow over time to a large benign tumor.
Both these are benign.
Another mutation might cause a wave of clonal evolution, causing a malignant tumor and a
metastatic cancer.
The genes that drive this process have been identified.
APC, the gene that causes polyposis, is the gene that starts it all off.
It initiates the process.
It's a cellular break.
And then tumors enlarge when they acquire other mutations in oncogenes like KRAS.
They become malignant when they acquire still other genes, mutations in other genes like
SMAD4 and P53, and still mutations in other genes, an oncogene called PL3, can help drive
it to metastases.
All these genes are mutated in most colon cancers.
Now one of my favorite genes is P53.
P53 is mutated like all of the others, in part because there's some sort of faulty repair.
But it's my favorite gene because a graduate student in our lab named Susie Baker discovered
its role about 15 years ago, and she was not much older than you guys, actually.
She discovered that P53 mutations occurred in just about all colon cancers.
And then another graduate student named Janice Nigra worked with her to find that P53 was
mutant in virtually every cancer that occurs throughout the world.
It's hard to be a cancer without getting a P53 mutation.
So all of these different tumor types have mutations in P53.
It's kind of a common denominator for cancer.
And knowing that, we can do some interesting kinds of experiments to try and relate the
initial insult to the mutation.
And here are some examples.
Sunlight, such as you get when you're on the beach, everybody likes the beach.
Anybody use a tanning salon?
Raise your hand if you do.
Okay.
Well, you not only will get a good tan.
You may get a mutation at codon two, at one of the residues of P53, changes a C to a T.
Now, this is a characteristic mutation caused by sunlight.
If you radiate with ultraviolet light sunlight bacteria, they get this kind of mutation,
a C to a T. And that mutation occurs in the P53 gene in skin cancers.
Skin cancers are, of course, induced by ultraviolet light.
You'd find them much more rarely in other forms of tumors, which, of course, are not
exposed to light.
Another example is aflatoxin.
It's a fungal toxin found in parts of Asia and Africa.
And it causes a different mutation.
At residue 249 of P53, it causes liver cancer.
So it's found only in the liver cancers of people who live in that area and are exposed
to that toxin, not found in other tumor types.
And of course, one of the worst forms of carcinogens are found in cigarette smoke.
Cigarette smoke, we now know, causes cancers by, in part, inducing mutations in P53 at
this position.
And they, of course, occur in lung tumors, but not in other tumors.
And in colon cancers, it's a more complex scenario.
We know specific mutations in P53 that occur, but we don't know exactly what the carcinogen
is.
That's something that remains in the future.
Now, how does P53 work?
P53 works as shown in this animation.
P53 molecule protein binds to specific sequences adjacent to genes which it controls.
There's a P53 molecule binding to its site, its binding site.
It recruits an RNA polymerase, not a DNA polymerase, that then makes RNA from these genes that
it controls.
And the RNA is translated into two proteins, P53 primarily regulates two proteins called,
in the next slide, WAF and PUMA.
Now, WAF is a cellular break, PUMA is a different kind of gene.
WAF, one of the nice things about discovering genes is you can name them anything you want.
WAF was named for Wafik Aldiri, who was a student in our lab who discovered it.
And PUMA was named by Jen Yu, because she was a runner and she liked her shoes.
It's true.
My favorite names for genes, there's one in flower called Superman, another one called
Clark Kent.
So one of the good things about being a scientist.
At any rate, what do these genes do?
The P21 works as a break.
Look at this movie, here's a typical break, that didn't work.
If you don't have P53, you don't have the regulation of P21, a break.
The PUMA works differently, it's a kind of a hairy carry gene.
If all else fails, what the cell does to control itself is to commit hairy carry, really, it's
suicide.
It's not accidental death, it's very purposeful.
Scientific name is apoptosis, but it's really a form of suicide.
And it protects the organism from the development of cancer, obviously kills the cell, but it's
good for the organism.
And the last kind of gene, which I'll just briefly touch because we'll discuss this further
in the second lecture, is oncogenes.
Oncogenes, again, are the accelerators in a car, and oncogenes get activated often by
chromosome translocations.
And in the second lecture, we'll talk about one specific translocation that causes leukemia,
a translocation between chromosomes 9 and 22.
And the interesting thing about this translocation is that its discovery has led to a novel therapy
for that form of leukemia.
And that's where I'm going to stop and ask for questions.
Yes?
How do mutagens produce such specificity in the mutations produced?
How do mutagens produce such specificity?
Mutagens are chemicals, and chemicals bind to specific sequences within DNA and the proteins
that DNA is packaged by.
So because they're chemicals, like all chemicals, they have certain charges, certain distributions
of forces, which direct them to specific molecules.
Now, they'll react with lots of DNA throughout the cell.
They only cause a problem if they mutate an oncogene or suppressor gene.
If they mutate a gene for hemoglobin, it doesn't matter.
That's not going to cause a tumor.
But in an individual cell with a mutation of an oncogene or tumor suppressor gene, that
could represent a step towards cancer.
Yes?
My name is Greg Swider.
I'm representing Fairfax High School.
In your chart that had P53 with all the different cancers, all the base pairs were changed to
thymine.
Is there a reason for that, or is that just coincidence?
I'll have to go back, but I think it was coincidence.
Yes?
I'm in the University of the Potomac School.
If someone had a familiar cancer, would there be any way to fix the mutated gene so that
it couldn't be passed on?
The question is, can you fix a mutated gene in an hereditary tumor type?
The answer is, in theory, yes.
In practice, no.
That's really not going to happen because of the serious ethical implications.
What people can do is, with in vitro fertilization techniques, you can select an embryo cell
that doesn't have a mutation to implant.
If a couple is known to have a hereditary mutation, you can select an embryonic cell
through in vitro fertilization that is devoid of the mutation, thereby ensuring that particular
child that develops from that embryo will not have the disease.
But fixing it is not on the table.
Yes?
I was just wondering, as a potential patient, I have tumors all over my body, right?
Some of them are bigger than others, but how would I distinguish between a benign and malignant
and metastatic, and when should I consult a doctor?
That's a good question.
The tumors you'll generally see are the ones in your skin.
If you feel anything inside, you should immediately consult your doctor because you can't tell
whether a tumor is benign or malignant from feeling it inside you.
If you're talking about tumors on your skin, the general rule is if it's round and circumscribed,
it's benign and you don't have to worry about it, especially if it's small.
If it starts to grow, if you see a change or you see it start to spread or it turns
color, then you should immediately see a physician.
But everybody should see their physician once a year anyway and just ask him or her to look
at any suspicious lesions.
You shouldn't have to guess.
That's what doctors are for, to save you the trouble of making those kinds of decisions.
Other questions?
Yes?
Why is it cancer more prevalent amongst children who are left rapidly divided in cells due
to growth?
That's a good question.
Cancer is not simply a function of dividing cells.
For example, cells of your small intestine divide much more frequently, in fact, every
day than the cells in your brain.
Brain cells hardly ever divide.
Yet brain tumors are pretty common and small intestinal cancers never occur.
So it's something more than just cell division and we don't understand exactly what that
is.
Yes?
Lily Young, Thomas Jefferson High School.
In cancer cells, are there usually multiple mutations in genes coding for apoptosis factors
or is the mutation in P53 enough to stop the process?
In every cancer that forms, there are several mutations, at least four or five, certainly
in every metastatic cancer.
So it's P53 in colon cancers plus APC plus RAS plus other mutations, all of them occur.
Each one inactivates a different growth controlling pathway.
Again, it's very much like a car.
You can control the car until you lose all your brakes and your accelerator.
Losing just one, you can still control the car.
And that's a pretty good analogy for what happens in a cell.
That's it.
We have a break.
See you in a minute.
Thanks for the great talk, Bert.
And now we're going to also thank the students for those fantastic questions.
We're going to now take a 30-minute break.
When we return, Bert will continue his discussion of the genetics of cancer and how that's critical
to both prevention and treatment.
From the Howard Hughes Medical Institute, the 2003 Holiday Lectures on Science.
This year's lectures, Learning from Patients, the Science of Medicine, will be given by
Dr. Bert Vogelstein, Howard Hughes Medical Institute Investigator at Johns Hopkins University
School of Medicine, and Dr. Huda Zoghbi, Howard Hughes Medical Institute Investigator at Baylor
College of Medicine.
The second lecture is titled, Chaos to Cure, Bringing Basic Research to Patients.
And now to introduce our program, the Vice President for Grants and Special Programs
of the Howard Hughes Medical Institute, Dr. Peter Bronze.
Welcome back to Holiday Lectures 2003.
These lectures are a product of the grants program at Hughes.
And that program actually focuses on science education across the educational spectrum
and across the nation.
And in the process of doing that, we've created a network of educators of all sorts around
the country who help us in a lot of different ways.
For example, Tom mentioned earlier that these lectures end up on DVDs.
And to make sure that the DVD really is useful for teachers, we actually activate our network
of teachers and send them the DVDs and ask them what they would do with them to field
test the lectures and to create other supplemental materials that they can use educationally.
Those materials end up actually on another website that we maintain called biointeractive.org.
And you can go look there to see not only previous lectures, but the other things that
this network of teachers have added to it.
You might go take a look on that website.
This is a commercial plug.
On that website, you'll find 3D animations.
You can test your skills in interactive quizzes that we have on there.
And you can even do an experiment on the virtual labs that we have there.
So there are a lot of other things that we do here educationally that come from these
lectures.
So now we're ready for the next lecture by Dr. Vogelstein.
The second lecture is called Chaos to Cure, Bringing Basic Research to Patients.
And we'll have a short video to further introduce Dr. Vogelstein.
People in our lab come from all over the world.
We're very lucky to have applicants from all over.
We've had, I guess, over a hundred people who have trained in our lab over the past
twenty-five years or so.
The lab is run kind of like a family.
You know, we consider everybody a lab family member.
People socialize a lot while they're here.
They work together as groups, which helps them both scientifically and professionally.
We have meetings a couple times a week where everyone gets together.
We have some recreational activities.
So people generally remember it as a good time in their lives when they learned a lot
and matured scientifically, but also had a lot of fun.
There are lots of different ways to be a scientist, lots of different ways to enjoy science.
The key thing is, I guess underlying all of that, is not being satisfied with a status
quo.
If you want to be a research scientist, then you have to want to make things better.
If you're a scientist, your homework is never done.
There's always more to learn.
You're never out of school.
In addition to doing your own experiments, you have to read about others, and there's
lots to read.
You have to be a bit skeptical about almost everything, that is it really the way it seems?
Is this explanation really the way it works?
And one of the things that we try and teach our trainees is that sort of skepticism.
The most fun part of what I do is seeing some young person make a discovery.
That's by far the best part.
The first thing that they themselves have created, have learned, that no one else in
the world right at that moment knows.
It's an incredible feeling to have, and it's a great thing to be able to watch.
In that moment of discovery, any young person is a real joy, both to witness, but as well
as to help them get there.
Welcome back.
In this lecture, we'll discuss some of the practical implications of the cancer revolution,
and you'll recognize this slide, which is a diagram of the genes that are responsible
when mutated for causing colon cancer.
What the challenge is in the years ahead is to figure out how to use this information
to actually help people, and there are three ways that we envision this could be done.
First, with risk assessment, trying to test patients who have hereditary predispositions
to see who does and who doesn't, early detection, and treatment.
All of these are at the experimental phase because most of the genes on this slide have
just been discovered in the last few years.
This is something that we hope some of you are going to help with in the upcoming years
because we need good creative minds to work on these challenging problems.
Let's start with risk assessment.
You'll remember that most colorectal cancers, like most cancers in general, are non-familial,
but there are a small fraction which are familial, and the two syndromes we discuss.
Just as a reminder, the pedigree symbols are a square for male, a circle for female.
When the circle and the square get together, if they're fruit flies, they call it mating.
The technical term that we scientists use is hooking up.
If they're blocked in, they're affected, and I think you know the rest.
Let's start with a young woman named Jennifer who presented to our clinic a few years ago
with pain and blood in her stool and her feces.
She turned out to have hundreds of polyps in her large intestine.
It turned out that her father had died in his 40s of colon cancer, and in fact, his
mother, Jennifer's grandmother, had also died from colon cancer.
So it was very clear that she had polyposis, familial adenomatous polyposis, a familial
form of cancer.
The real question was, what about her four siblings, her two brothers and two sisters?
It's important to point out that these are not just symbols.
These are real people, and this sort of information is really important for them to manage their
lives.
Polyps in this disease, these benign tumors, don't generally form until patients are in
their teens or even 20s.
So it was impossible to tell whether her younger siblings, she was the oldest in the family,
actually were going to get this disease.
In the past, what physicians would have to do would be simply to wait and see whether
patients got them and then advise them accordingly.
But now that we know the gene that causes this disease, we can do a simple blood test
in families like this.
And when we did that blood test on Jennifer and sequenced her APC gene, we knew the APC
gene is the only gene that causes this syndrome, we found a mutation.
And you can all see where that mutation is.
This is just a standard sequencing result that we do in our lab every day.
You can read off the bases, A-G-A-A-A-T-A, et cetera, and you can see there's one position
that's mutant.
It's G normally, and Jennifer had both a T energy.
The T, the mutant base, she inherited from her father, who had polyposis, and the G,
she inherited from her mother, who was normal.
Now it's kind of awesome to think about the fact that there's six billion of these building
blocks, these A's, C's, T's, and G's in every one of our cells, and a mutation of only one
of them, that is, this specific mutation in the APC gene, just one out of the six billion,
that mutation alone is sufficient to guarantee 100% certainty that a patient will get polyposis,
thousands of polyps, and will eventually get cancer.
Now how do we use this information in this family?
Well we can then do the blood test on the four siblings who were too young to have polyps.
And when we did that, we found that two of them, Barbara and Matt actually, had the syndrome,
and the other two didn't, had the mutation, and were going to get the syndrome, whereas
the other two didn't.
And that was very helpful because in the two siblings that didn't, the brother and sister,
they didn't have to worry.
Their mother didn't have to worry, and that was a great relief to them.
And Barbara and Matt, we'll discuss in a minute what could be done for them.
But this raises some other issues, the fact that we can do simple blood tests and predict
what's going to happen to patients in terms of diseases like this.
We can clearly do it, the technology for doing it is available, it's done in labs every day.
But should we do it?
And it raises a host of ethical issues.
For example, should Jennifer's test result, or those of her brothers and sisters, be given
to insurance companies?
She's going to be applying for jobs in a couple of years, should they be given to her employer?
And what do you think?
Who thinks that it's permissible on ethical grounds to give genetic information like this
to life insurance companies, which would allow them to either give or not give insurance
or raise the price?
Who thinks it is permissible and reasonable to do that?
Only one person, whose father probably works for an insurance company, who thinks it's
not?
Everybody else.
That's the response we usually get.
But think about it, everyone who gets a life insurance exam gets a test, lots of tests,
but including a test for blood pressure.
Now, this is a low tech way of predicting who's going to get heart disease.
I'm not saying it's right that blood pressure results should be given to insurance companies,
but if that's right, then genetic testing really isn't that much different.
They're either both right or both wrong.
They're just tests to predict disease.
So questions like these have no simple answers.
Here's another one.
Suppose, should these tests be given to every family member?
Suppose Krista finds that Daniel's mutations aren't so bad and decides to marry him.
But Daniel, unbeknownst to Krista, actually has polyposis.
Does he have to tell Krista, or should he tell, on ethical grounds, his fiancee that
he is going to have polyposis and that his kids, at least half of them, are likely to
have it?
Who thinks Daniel should tell Krista?
And who thinks Daniel shouldn't tell Krista?
Nobody.
All right, well, we've all seen movies of what happens if you don't tell your fiancee
the truth.
But, you know, if I asked you whether if Daniel had an infection with HIV, would he have to
tell Krista, everybody, without doubt, would say yes.
Nobody would say no.
And for the same kind of reason, I want to show you one pedigree of a gentleman in Baltimore
who married six wives and had well over 100 grandchildren, half of whom had polyposis,
a one-man epidemic of polyposis.
So both from a public health perspective, not only from an ethical perspective, who
should get results about genetic testing, distant family members, et cetera, is not
an easy one to answer.
And we'll discuss other ethical issues in the bioethics session this afternoon at 1.30
or 2.
Now, let's go back to our family with Jennifer and Matt and Barbara.
What can we do for Matt and Barbara?
And Jennifer, well, we can take out their colon.
That will prevent them from getting colon cancer, for sure.
And you can do that.
You don't need your colon.
You need your small intestine, but you don't need your colon.
All the colon does is dehydrate the stool.
So it becomes solid.
People can live without their colons.
It's not pleasant, but it can be done.
On the other hand, many investigators, including us, are trying to develop drugs that will
prevent the onset of polyposis in such patients.
And to test such drugs, we use animal models.
We obviously can't and don't want to use people to test experimental drugs.
And through genetic engineering, we can create a mouse with polyposis.
In fact, we create a mouse that has exactly the same mutation that we found in Jennifer.
This mouse gets polyps, which you can see there highlighted in red.
And they look under the microscope the same as an apoliposis patient.
We find these mice have lots of polyps indicated by those red dots.
And if we treat them with a drug called Solendac, this is a pretty harmless drug.
It's the same chemical class as Advil.
I'm sure many of you have taken Advil for aches and pains.
But Advil actually can help with colon polyps.
And people who take Advil or aspirin actually have less polyps and colon cancer than people
who don't.
At any rate, this reduces polyps in the mice by about 50%.
Another drug, an experimental drug not yet released to the public called EKI, also can
reduce them by 50%.
When they're both used together, they're dramatically reduced.
The mice hardly get any polyps, even if they're just treated for a couple of months.
They don't get polyps throughout their entire lives.
A clinical trial of these drugs will probably be instituted in a couple of years in polyposis
patients, and hopefully will at least delay the time when they need to get their colons
removed if not precluding the need for colon removal at all.
Risk assessment can only be done in familial cases, because then we're looking for changes
say in the blood with blood tests that are present in every cell in the body.
Only those heritable mutations are the subject of risk assessment.
Most cancers are not heritable, not familial.
What can we do for those patients?
What we can do is try to detect their tumors early.
And the best way to do that now for colon tumors is through colonoscopy.
And this video shows Katie Couric, who's the host of the Today Show, and her colonoscopy,
at least part of it.
Hi, everybody.
Here we are in my kitchen.
It's about 18 hours plus before I get my first colonoscopy.
And here's my cherry flavored New Lightly.
Looking forward to drinking this in the next several hours.
Bottoms up, so to speak.
The next morning, it was business as usual, except my stomach was empty and my colon was
ready.
After the show, I went directly to Columbia Presbyterian Hospital for my date with Dr.
Ford.
I heard you're a little light on the sedation.
I'd like to request plenty of sedation.
I'll give you what you need.
Okay, good.
I'm a little nervous.
Sure.
Is that normal?
Normal.
Have a pretty little colon.
So Katie has a pretty little colon and also a normal colon.
She's a wonderful lady whose husband actually died from colon cancer in his 40s.
She's been a great champion of colon cancer research.
If she had had a polyp, it could be removed through the colonoscope like this.
The colonoscope is a fiber optic tube that's inserted through the anus and kind of snaked
up through the colon, the large intestine, allowing the physician to view the inside
of the colon.
And if they see a polyp like that, this is how they remove it.
You show the video.
There's a little snare coming from the end of the colonoscope, kind of catches the polyp,
like a little scissors snips it off, and then takes it out.
And this procedure works, the colonoscopy works, but most Americans don't get it.
In part, they don't get it because it's uncomfortable, it's a bit painful, and that has established
a need for non-invasive techniques, techniques where you don't have to insert things into
people and where there's no dangers associated with the procedure.
And we and others are trying to develop non-invasive techniques based on the knowledge of the genes
that actually cause colon cancer.
Cron cancer takes 30 to 40 years to develop, and that gives us a large window of opportunity
to intervene in the process, to try and uncover or identify mutations in patients who have
cancers.
Now, let me show you this movie, which I hope will be informative.
Early in the season, dung is collected strictly for food.
Later, when the hunger is satisfied, these males gather even larger balls as gifts for
females.
It's the perfect engagement present.
She rides on top while he searches for a patch of soft ground.
So I would not advise Daniel to give that gift to Krista for their engagement.
What does this have to do with cancer research?
Well, people think stool is not very valuable, but some people think it's very valuable.
Some insects do, like dung beetles, but also people, weird scientists like me, I guess,
and the members of my lab, here's members of my lab.
You might think we're weird, but I think we're just well dressed.
But at any rate, the reason we think stool is valuable is because we can find mutations
in it in patients who have colon cancer.
These cells, cancer cells, are secreted into the stool, and we can simply examine the stool
looking for mutations in genes which we know cause the disease.
This slide illustrates a typical example of that.
Everybody see the mutation?
What is it?
Just yell it out.
Right.
Two bases missing, deletion, which is the kind of mutation that will change the reading
frame of the protein and cause it to be abnormal.
We can currently detect about 80% of cancers simply by doing a stool test, which is obviously
not a dangerous procedure, and hopefully we'll get more people to undergo screening.
In summary, the knowledge about the genes that cause colon cancer have already led to
advances in patient management, particularly in risk assessment, but hopefully in the future
through early detection using the technique that I just showed you.
Prevention is always the best way to go for managing diseases.
We've learned that throughout history, and I hope in the future this will become more
and more of a reality.
Now it's time to break for questions.
Yes?
How do you determine what drug to give, like the mice with, like you said, the polyp, whatever
it's called?
How do you determine what drug to give them to make them better?
Yes.
How do we determine what drugs to give them?
The one drug was discovered serendipitously by a surgeon many years ago.
He was treating a patient with polyposis with Solendak, a drug like Solendak, because the
patient had tennis elbow.
That's what it's used for.
Notice that when he performed colonoscopy on that patient about six months later, all
of the polyps had disappeared.
He wrote a paper about this in the scientific literature.
I believe them, because it just seemed so extraordinary.
It seemed like a mistake.
When we see a dramatic result like that in the lab, we think maybe you mixed up samples.
Maybe people thought he mixed up colons or something.
I don't know.
I also want to-
Wait a minute.
I'm not finished.
Anyway, that's how that drug started.
The other one was developed as an inhibitor of a growth factor.
It was developed through a knowledge-based approach.
Other questions?
Yes.
Yes.
I'm Jamie Worley from St. Albans School.
How do you explain the effects of Solendak and EKI?
More importantly, can we harness these effects in maybe a more effective way?
Solendak is an inhibitor of an enzyme called cyclooxygenase, which is known to produce
certain small chemicals that stimulate the growth of cancer cells.
That's the best guess for how it works.
EKI is an inhibitor of kinase.
We'll talk about kinase inhibitors later.
A receptor kinase that's important for the growth of cancer cells.
Lots of people throughout the world are working on doing just what you said to improving these
drugs, making them work better and more specifically.
Yes, in the back.
I remember, my name is Sandra Katz from Winston Churchill High School.
I remember the beginning of this presentation began with a little girl.
My question is relating to the age of onset of cancers.
Are we fighting different cancers depending on what age a person is getting them or is
this more of a reflection of the different varying immune systems from one person to
another?
In general, it has nothing to do with the immune system.
Most cancers appear in people who are middle-aged or older and the reason for that is because
it takes many years to accumulate all of the mutations that are required to get to a cancer.
Mutations are not common events and even cancer cells have lots of repair systems which are
still active to guard them against these mutations.
That's why in general, cancer is a disease of older age.
However, some cancers, particularly leukemias, don't require as many mutations as tumors
like colon cancers.
They may require only one or two mutations and those are the kind that develop in childhood.
So conceptually, they're the same but the numbers of mutations that are required are
different.
Yes?
My name is Alexander.
I'm from the Council of High School.
What are the body's defenses against cancer?
Are there any mechanisms the body uses, other cells in the body, to fight against cancer?
Yes.
In fact, the suppressor genes, the breaks, are one sort of protective mechanism.
Remember, there are two breaks in every cell, two genes, one from the mother and one from
the father.
So it's a redundant system.
Just like on the space shuttle, you want to build in redundancy.
Both those breaks have to be altered.
For any suppressor gene, the repair systems are a kind of safeguard to protect against
mutations, carcinogen-induced mutations as well as naturally occurring mutations.
And the fact that you need five or six mutations to get to a metastatic cancer is an excellent
protective mechanism.
It means it's only going to happen rarely to get that many mutations.
So cells are pretty well protected.
Remember, even patients who get cancer, there are 13 billion cells at risk but only one
cancer forms from one cell.
Yes?
Have scientists witnessed any populations that seem to have developed any resistance
to various forms of cancer?
There are definitely geographic differences in incidences of cancer, but largely they're
thought to be environmentally determined.
For example, Japanese historically have had a low incidence of colon cancer compared to
Americans.
But over the past 20 or 30 years, their incidence of colon cancers have gradually gone up.
People think that's due to westernization of their diet, perhaps because there's so
many more McDonald's in Japan now than there were 30 years ago.
Similarly, we're getting more gastric cancers, stomach cancers, than we used to.
Even though that was high in Japan to start with, maybe that's due to sushi.
So geographic differences, but perhaps due to environmental causes, not so much differences
in genes.
Yes?
I'm Ann Carr from Holonimes High School.
It's been said that cancer cells will secrete certain toxins throughout the body.
Is early detection ever dependent on the discovery of these toxins or mainly just the cancer
cells that are found throughout the bloodstream in stool samples?
Well, I think your question was, can cancers be detected before they're metastatic?
Is that basically the question?
Yes.
In fact, it really wouldn't do too much good to be able to detect a cancer that was metastatic,
because by then it's too late.
All of the cancers that I showed you that were detected by analysis of the stool, they
were all curable cancers, early stages prior to metastases.
The reason they could be detected, because those cells are shed into the stool, those
DNA molecules are also shed into the blood.
We can detect some just by a blood test.
That's true in general, not just colon cancers.
All the way in the back.
I have two questions, and I'm Sheila Rajagopal from Montgomery Blair High School.
I was wondering, first off, what places in the body are more likely to develop metastatic
cancers?
Yes.
I think the first question is all we'll have time for, and that's a good question.
The most likely places are liver, because all the cancers that occur in the gastrointestinal
system, their bloodstream is connected to the liver through the portal circulation.
The liver is the common site for stomach cancer, colon cancer, pancreatic cancer, gastrointestinal.
Other forms of cancer say breast cancers and bladder cancers often metastasize to the lung
predominantly because their blood supply and the tumor cells that invade blood vessels
go first to the lung and are trapped there.
They're the two major sites, but no organ is spared.
People who have cancer often die from brain metastases or bone metastases, so it really
can go anywhere.
Once that happens, they're generally incurable.
That is why people are trying to develop systemic therapies that can be injected intravenously
rather to help produce better results than achievable by surgery.
That's the perfect place to begin our last session, which is on treatment.
How do we use this information to develop new treatments?
Certainly one of the most difficult issues in current cancer research.
The conventional forms of treatment include surgery.
That's the mainstay, and that's great when the cancers are confined locally.
That is when they're malignant, but they're not metastatic.
It works.
Patients are cured.
Unfortunately, by the time cancers are detected, often the tumor has metastasized and can't
be removed by surgery, so then patients are irradiated with x-rays or high-energy electrons.
That works to a certain extent, but generally doesn't cure patients.
It just shrinks the tumors and allows them to live for a longer time.
The same with chemotherapy.
These are chemicals that are used.
The problem with conventional chemotherapeutic agents is that they're not really specific
for the tumor cells.
The first chemotherapeutic agents were in fact agents used for chemical warfare in the
First World War.
They were only then later modified to be used for cancer treatments.
They kill all cells, including cancer cells.
They kill cancer cells a little bit better than normal cells for reasons which aren't
really known, but because they kill normal cells, too, they cause lots of side effects.
That's why anyone treated with these drugs that you know probably, they've lost their
hair, they're nauseous, they feel terrible, they're very toxic.
One challenge for the future is how to develop agents which are less toxic and more specific.
After that, we resort to our knowledge of what's wrong with a cancer cell.
Really, there are only three basic things that are wrong.
One is there are altered chromosomes.
Two is there are altered DNA sequences, and three is there is an altered blood supply.
Altered chromosomes are virtually ubiquitously present in cancers.
Here's a normal karyotype.
Each chromosome is colored differently through a technique called chromosome painting.
You can see each member of each pair looks the same.
It's a perfectly normal karyotype.
In cancers, the karyotype often looked like this, wildly abnormal.
For instance, chromosome four, you can see there are three copies instead of two.
Only one of those three copies looks normal.
In chromosomes two and 11, there is one normal copy and one abnormal copy.
Here what's happened is one end of chromosome two has fused with one end of chromosome 11,
containing a fusion chromosome.
Virtually every chromosome, even the sex chromosome, is abnormal in these cases.
This leads to an abnormal DNA sequence, these chromosome translocations.
This slide of the Philadelphia chromosome, which I showed you before, illustrates one
such example.
Here part of chromosome nine has been translocated to part of chromosome 22.
The bottom segments of those two chromosomes have been switched, leading to two new abnormal
chromosomes, a translocated chromosome nine and a small chromosome called the Philadelphia
chromosome.
This chromosome was first seen by investigators in Philadelphia.
That's why it's named the Philadelphia chromosome.
Ten years later, it was identified at the molecular level.
Remember, chromosomes consist of DNA, so when you have a translocated chromosome, it means
you've translocated DNA sequences.
In this case, half of the chromosome nine sequence encoded a gene called ABL, A-B-L,
which is an oncogene, and the chromosome 22 sequence encoded a different protein called
BCR.
The result of this translocation is a fusion gene, half made of BCR, half made of ABL.
The important thing about this particular chromosome is that it has led to a new form
of therapy, which is shown in the following animation.
Normally, the BCR-ABL protein, that fusion protein, functions in a cancer cell to phosphorylate
a substrate or target protein.
The way it does that is it takes a phosphate group from ATP, which binds to it and transfers
that phosphate group to its target substrate protein.
The target substrate protein then changes its shape and goes on to stimulate cell growth
of the leukemia cells.
This abnormal chromosome, Philadelphia chromosome, is found in leukemia cells, a particular form
of leukemia called CML.
Now, Gleevec, a drug which many of you may have heard of, mimics ATP.
It binds to the site within the BCR-ABL that's normally bound by ATP, and it prevents ATP
from binding.
It thereby prevents phosphorylation of the substrate target protein and prevents cell
growth.
If you look in the blood of a patient with leukemia, you'll see cancer cells filling
it up, as I showed you in the first lecture, in this case with chronic myelogenous leukemia.
It's just one of the many forms of leukemia.
Fortunately, this disease can now be treated.
On the X-axis, we've plotted days of treatment with Gleevec.
On the Y-axis, the white blood cell count in the blood, normally, it should be less
than 10,000 cells per microliter, or 10 million cells per cubic centimeter.
In leukemia patients, it's often 30,000 to 80,000 cells.
Treatment with Gleevec results in a rapid decrease of cells to the normal range and
produces remissions of the disease for at least a year.
This is the first example of a specific chemical inhibitor, a new drug based on a very specific
understanding of what's wrong with the cancer cell.
Let me show you another good idea based on another cancer gene, P53.
This is a tumor suppressor gene, unlike ABL, which is an oncogene.
This movie shows a strategy which uses viruses that target cancer cells.
The virus is derived from a typical cold virus called an adenovirus.
It infects or begins to infect all cells.
If the virus infects a normal cell, the DNA from the virus will get out of it and go to
the nucleus, but then the cell has an antidote, P53 in fact, the tumor suppressor gene, which
then prevents that viral DNA from replicating and the cell lives.
On the other hand, if the same virus infects a cancer cell, the DNA will get out from the
virus, travel to the nucleus, and there's no P53 around because the P53 is mutant.
It's defective.
Therefore, the viral DNA will continue to multiply, will eventually burst the cell,
release lots of viruses into the environment.
They will infect both normal and cancer cells, but the normal cells are kind of immune to
the virus, whereas the cancer cells will continue to be killed by it because they have no P53.
This is a very clever way to exploit a defect in a tumor suppressor gene.
It works reasonably well in animals, and it's now being tested in humans.
Let's talk about the third major defect, which is blood vessels.
Blood vessels are different in tumor cells.
This video, which you've seen before, shows how when even a benign tumor arises, it needs
to recruit blood vessels.
If it doesn't recruit blood vessels, it can't get the oxygen and the nutrients that it needs
to grow to a clinically significant size.
It can't even grow more than a tenth of an inch or so.
Blood vessels are needed throughout the process.
The way that the tumors recruit blood vessels is through secreting a hormone called VEGF.
As you'll see in this video, it's a hormone, VEGF stands for vascular endothelial growth
factor.
It's a fancy name for a kind of growth hormone that is a growth factor for blood vessel cells.
Blood vessel cells have receptors for this growth factor on them.
When that happens, the blood vessels are stimulated to grow.
Investigators have recently developed an inhibitor of this growth factor.
This inhibitor can turn a bloody tumor, most tumors are bloody to start with because they
have recruited lots of blood vessels, into a non-bloody tumor shown here.
It's white because there are fewer blood vessels in it.
The question asked me before about Judah Folkman's work, that's this kind of therapy.
It inhibits the process of blood vessel recruitment and growth.
Again, it works well in animals.
The first clinical trial of this form of therapy was just released last month.
It is showing some good effects in patients with colon cancer.
This looks to be perhaps a real gain in therapy.
The last treatment scenario I'll show you is also based on abnormal blood vessels and
tumors.
It's based on the fact that even though there are lots of blood vessels, tumors grow so
fast sometimes outpace the blood vessels.
As a result, tumors are not only composed of the tumor cells themselves, which are shown
here, the cancer cells, but cancers also have these supporting cells, which contain the
blood vessels, which can always keep up and often lead to dead cells, which don't have
enough oxygen and nutrients to grow and often die.
The only place in the body that has no oxygen is in these areas of dead cells within the
middle of tumors that have outgrown their blood supply.
That environment, hostile environment, does not exist in any normal tissue.
This observation has stimulated investigators like us to try to develop new methods that
target that defect.
In particular, one that we're working with now is a strain of anaerobic bacteria.
These bacteria are found ubiquitously in the environment.
If you look at dirt outside, you'll find lots of these bacteria.
They can only grow when there's no oxygen.
If you expose them to oxygen, even for a few seconds, they die.
We've engineered one such strain of bacteria called C-NoV-NT for non-toxic because we've
removed one of its major toxin genes and made spores out of it.
All of you are familiar with spores.
Bacteria often make spores, which allow them to live for even millions of years.
You've heard of it in connection with the anthrax outbreak, but this is a medical use
of a different bacteria.
If we make spores and inject them into a mouse, then the spores germinate only within the
tumors because they can only germinate where there's no oxygen, and they become this hairy
beast.
In fact, when they germinate within the tumor, they actually eat the tumor.
They use it to provide nutrients.
This series of slides shows what happens when you treat a mouse with a tumor with these
bacteria.
This is a nude mouse.
It has no hair.
That's why it's called nude, but more importantly, it has no immunity, so you can put human tumor
cells into it.
If you put human tumor cells into a regular mouse, it will reject them as foreign.
At any rate, six hours after injection of the bacteria, a little black spot can be seen
on the tumor, which gradually expands and within a day or two, occupies the whole tumor
mass.
Over the next few days to weeks, the tumor rots out because the bacteria have completely
essentially eaten it, and then the tumor falls off, and many of the mice are cured.
This works well in mice.
Again, it's another thing about whether it will work in patients, but hopefully, clinical
trials in the next couple of years will show results that are similar to what has been
achieved in animals.
It represents another way to exploit the defects in cancer.
There are lots of ways, which we can think of, to use the information gained from the
cancer revolution to help people with cancer, to prevent them from getting cancer, or to
treat cancers once they arise.
Will it be easy?
Well, look at this movie.
I can't imagine doing any of the things in that movie at my age anyway, but if you guys
can do that stuff, you guys can cure cancer, but you just have to try.
Thank you.
So we have some time for some final questions.
Yes?
The viral and bacterial therapy that you mentioned, could that potentially have any harmful side
effects, like causing the virus or the bacteria to develop mutations, which might make it
harmful to the person?
I think it's fair to say that any therapy for a serious disease like cancer is going
to have side effects.
There's no drug known to humans that doesn't have some sort of side effects.
What they are can only be told once people are treated, but it would be unrealistic to
expect that they won't have side effects.
Hopefully the therapeutic effects will be greater than the side effects, but that can't
be told until clinical trials are done.
Yes?
Hello.
My name is Christina Hodkins.
I'm from Northwestern.
I was wondering, since these tumor cells undergo angiogenesis to get more oxygen, is it a way
to give them more oxygen to poison them?
Because if our normal cells get too much oxygen, they die from poisoning.
And since the tumor cells want more oxygen, is there a way to give them an elevated level
of it to the point where they die but not our normal cells?
Yeah.
That's an interesting idea, but the problem, the challenge with that kind of idea is how
to deliver something specifically to tumors.
That's really one of the major challenges for gene therapy, for example, as well as
other forms of therapy, is how you get whatever agent you're using just to go to the tumors
and kill only the tumor cells.
Now the bacteria that I showed you only will germinate in tumor cells, so they're tumor
specific.
The virus that I showed you only will replicate in tumor cells without P53, so that's specific.
In order to implement your idea, you'd have to figure out a way to get oxygen just to
tumor cells.
If you can do that, it would be a great thing, but I don't know how.
Maybe you'll be able to think of a way.
Over there.
I'm Cal McCauley from Episcopal High School.
Considering side effects in extended lifetime, which treatment do you prefer?
Which treatment?
I prefer not to have cancer.
If I do have to have cancer, I prefer detecting it early, so that all I'll need to have is
surgery, but I'm not sure that's what you mean.
You mean which of the experimental therapies?
The experimental therapies really are just that, they're experimental.
At this point, no one would prefer any of them, because you don't know whether they're
really going to work in humans.
Therapies work to a certain degree in mice and are encouraging, but we'll have to wait
until they're really tested in humans to decide whether any human patients deserve to get
them on a routine basis.
Yes?
Jay Levin from Colonel Zeta from Reuter High School.
It sounds like you guys are so close to treating cancer and to detecting it early to possibly
prevent it, but how would you go and make all these treatments available to the mass
public when so many Americans don't have insurance, which may or may not even cover this, all
these revolutionary procedures?
I guess how would you make it available to the public, that way you could go and possibly
eradicate cancer in our lifetime?
Your question is very good, and I'll give you a t-shirt, and others who have asked questions
who I haven't thrown t-shirts to can come up afterwards.
That's obviously a big problem for society, and it's not only with new cancer treatments,
it's treatments for all diseases and how to make sure people who don't have the means
can get them.
Those of you who don't go into science, but I hope some of you will try to answer that
question.
That may even be a harder one than curing cancer.
Yes?
Hi, I'm Miriam Raby from High School of Technology at Edison, and my question was that how can
we use hormone therapy to help us treat cancer?
I've heard of cases that actually stimulates the growth of the cancer.
How's that?
Yes, question about hormone therapy.
Interestingly, hormone therapies are one of the best forms of therapy and the oldest.
For breast cancers, the standard form of therapy is with anti-estrogens, because cancers don't
lose all their normal controlling mechanisms.
They lose enough so that they can grow abnormally and invade, but they still have some remnant.
You can look at them under the microscope and still tell that they're from breast, for
example, and they still are inhibited by anti-estrogens because breast cells normally require estrogenic
hormones, female hormones, to grow.
That's a standard treatment for breast cancers.
For prostate cancers, it's anti-androgens are used.
Most of those treatments don't actually stimulate cell growth when they don't work.
It's just some cancers, most cancers, in fact, eventually become immune to them, so it doesn't
help.
They put people in remission for years, and they're part of conventional therapy for cancers
now.
Yes?
I'm Rachel from Northwest High School, and I was wondering, you said that the tumors
that have outgrown the blood vessels are the ones that form like a dead section inside
of the tumor, so when you insert the Sanove, and it eats away at the dead part, wouldn't
it die as soon as it touches the live part of the tumor, and therefore not eliminate
the entire tumor?
Yes.
That's a very insightful question.
You definitely should get a t-shirt.
In fact, the bacteria, once they eat up the dead part, come to the live part, and importantly,
and delightfully, I guess, around even the live part, there's a small area of hypoxia
where the cells are metabolizing so rapidly that it's anoxic, and the bacteria can grow.
That's one way the bacteria kill the rim.
Another way is, in some forms of this therapy, we combine it with conventional drugs that
kill actually the outside live tumors better than the inside, so the combination actually
generally works better than either agent alone.
It's a very good question.
Yes?
I'm Naveisha Dillon from Northwest High School, and you mentioned adenoviruses with tumor
cells.
I'm familiar with the fact that adenoviruses can be used via genetic engineering to exploit
some weaknesses and cure diseases.
However, once an adenovirus is used, isn't it possible, I mean, it's a fact that that
can't be used again, that you should incorporate as many possible cures into one adenovirus
as possible because an immunity is developed after that?
One of the problems with adenoviruses is that immunity will develop if it isn't already
there, so patients can only be treated once.
In general, there's a wider answer to your question, which is most investigators feel
that no single therapy will ever work to eradicate a metastatic cancer, and that the best hope
for actually treating cancers will be to combine things, combine antiangiogenesis with agents
that target anaerobic parts of bacteria with conventional agents with new agents.
Many of you are probably aware of successes for treating AIDS, and those successes in
general have come from treating AIDS with multiple different agents, not single agents.
Single agents don't work so well, and it's the same for cancer.
The more ways you attack it at once, the more likely you'll be able to handle it.
Yes.
Gentleman with the green light.
You're saying how from the stool sample we can identify most forms of cancer that we
might have?
No, only colon cancer.
Only colon cancer, at least at all.
Other investigators have looked at the blood to find evidence of those mutations because
the cancers not only leak DNA out into the stool, but they also leak DNA into the blood
stream.
So investigators have been able to detect, say, kidney cancers, stomach cancers, some
colon cancers by looking at the blood as well.
Why would we have it as a normal physical, like if you go to the doctor to get a physical,
would they look at your blood to tell you you're going to have cancer or not?
That's my hope.
I mean, my hope in 20 years, remember in Star Trek, we had a little gizmo where you scan
up and down on people and you tell exactly what disease they have.
That's the idea with a drop of blood that you can tell what disease people have simply
by looking at what mutations are, but it's going to require a lot of technical development
to do that.
It's not simple.
Yes.
Anybody.
Good morning.
I'm Dianne Jamerson from Mission Matnamara High School, and I was wondering if multiple
cancers generally don't develop in people.
Because people are so well protected against cancers, most people only develop one cancer
at once.
The only people who develop multiple cancers are those who have a hereditary mutation and
are predisposed.
Like with polyposis, they develop thousands of tumors.
And there, the only way to treat it really is to remove the whole colon.
Last question.
Thank you.
One more or that was it?
Thank you very much for your attention.
Thank you, Bert, for two really exciting lectures.
They were really wonderful.
And unfortunately, there were lots more questions than there were so many good questions.
And we know we can't answer all of them here now.
And certainly the people who are watching this on the web can't ask their questions
here.
And those people who are going to be watching DVDs in the future can't ask questions here.
But we have an answer.
We have a prescription, which is on that other website that we have, www.biointeractive.org,
we have a feature called Ask a Scientist.
And this is useful to know for other parts of science as well.
We have a group of volunteer scientists who will custom answer your questions.
So go to the website and there's a way you can email us your questions.
If you have them, we will answer.
So we hope you join us tomorrow for more stimulating lectures that join basic science with medicine.
Tomorrow, Dr. Huda Zoghbi will take us on a journey from the clinic to the lab and back
to the clinic to deal with disorders of the nervous system.
So thanks for coming.
Good night.
The
The
The
The
The
The
you