TEDWomen 2010

Deborah Rhodes: A test that finds 3x more breast tumors, and why it's not available to you

Filmed:

Working with a team of physicists, Dr. Deborah Rhodes developed a new tool for tumor detection that's 3 times as effective as traditional mammograms for women with dense breast tissue. The life-saving implications are stunning. So why haven't we heard of it? Rhodes shares the story behind the tool's creation, and the web of politics and economics that keep it from mainstream use.

- Physician, cancer researcher
Deborah Rhodes is an expert at managing breast-cancer risk. The director of the Mayo Clinic’s Executive Health Program is now testing a gamma camera that can see tumors that get missed by mammography. Full bio

There are two groups of women
00:15
when it comes to screening mammography --
00:18
women in whom mammography works very well
00:20
and has saved thousands of lives
00:23
and women in whom it doesn't work well at all.
00:25
Do you know which group you're in?
00:29
If you don't, you're not alone.
00:32
Because the breast has become
00:34
are very political organ.
00:36
The truth has become lost
00:39
in all the rhetoric
00:41
coming from the press, politicians,
00:43
radiologists
00:45
and medical imaging companies.
00:47
I will do my best this morning
00:49
to tell you what I think is the truth.
00:51
But first, my disclosures.
00:54
I am not a breast cancer survivor.
00:57
I'm not a radiologist.
00:59
I don't have any patents,
01:01
and I've never received any money from a medical imaging company,
01:03
and I am not seeking your vote.
01:06
(Laughter)
01:09
What I am
01:11
is a doctor of internal medicine
01:13
who became passionately interested in this topic
01:15
about 10 years ago
01:17
when a patient asked me a question.
01:19
She came to see me
01:23
after discovering a breast lump.
01:25
Her sister had been diagnosed with breast cancer
01:27
in her 40s.
01:30
She and I were both very pregnant at that time,
01:32
and my heart just ached for her,
01:35
imagining how afraid she must be.
01:37
Fortunately, her lump proved to be benign.
01:40
But she asked me a question:
01:43
how confident was I
01:45
that I would find a tumor early on her mammogram
01:47
if she developed one?
01:49
So I studied her mammogram,
01:51
and I reviewed the radiology literature,
01:53
and I was shocked to discover
01:55
that, in her case,
01:57
our chances of finding a tumor early on the mammogram
01:59
were less than the toss of a coin.
02:01
You may recall a year ago
02:04
when a firestorm erupted
02:06
after the United States Preventive Services Task Force
02:08
reviewed the world's mammography screening literature
02:12
and issued a guideline
02:15
recommending against screening mammograms
02:17
in women in their 40s.
02:20
Now everybody rushed to criticize the Task Force,
02:22
even though most of them weren't in anyway familiar
02:25
with the mammography studies.
02:27
It took the Senate just 17 days
02:30
to ban the use of the guidelines
02:32
in determining insurance coverage.
02:34
Radiologists were outraged
02:38
by the guidelines.
02:40
The pre-eminent mammographer in the United States
02:42
issued the following quote
02:44
to the Washington Post.
02:46
The radiologists were, in turn, criticized
02:49
for protecting their own financial self-interest.
02:52
But in my view,
02:55
the radiologists are heroes.
02:57
There's a shortage of radiologists
02:59
qualified to read mammograms,
03:01
and that's because mammograms are one of the most complex
03:03
of all radiology studies to interpret,
03:06
and because radiologists
03:09
are sued more often
03:11
over missed breast cancer
03:13
than any other cause.
03:15
But that very fact is telling.
03:17
Where there is this much legal smoke,
03:20
there is likely to be some fire.
03:23
The factor most responsible for that fire
03:26
is breast density.
03:29
Breast density refers to the relative amount of fat --
03:33
pictured here in yellow --
03:35
versus connective and epithelial tissues --
03:37
pictured in pink.
03:39
And that proportion
03:41
is primarily genetically determined.
03:43
Two-thirds of women in their 40s
03:45
have dense breast tissue,
03:47
which is why mammography doesn't work as well in them.
03:49
And although breast density generally declines with age,
03:52
up to a third of women
03:55
retain dense breast tissue
03:57
for years after menopause.
03:59
So how do you know if your breasts are dense?
04:03
Well, you need to read the details
04:06
of your mammography report.
04:08
Radiologists classify breast density
04:10
into four categories
04:12
based on the appearance of the tissue on a mammogram.
04:14
If the breast is less than 25 percent dense,
04:17
that's called fatty-replaced.
04:20
The next category
04:22
is scattered fibroglandular densities,
04:24
followed by heterogeneously dense
04:26
and extremely dense.
04:28
And breasts that fall into these two categories
04:30
are considered dense.
04:32
The problem with breast density
04:34
is that it's truly the wolf in sheep's clothing.
04:36
Both tumors and dense breast tissue
04:39
appear white on a mammogram,
04:41
and the X-ray often can't distinguish between the two.
04:44
So it's easy to see this tumor
04:47
in the upper part of this fatty breast.
04:49
But imagine how difficult it would be
04:52
to find that tumor in this dense breast.
04:54
That's why mammograms find
04:57
over 80 percent of tumors in fatty breasts,
04:59
but as few as 40 percent in extremely dense breasts.
05:02
Now it's bad enough that breast density
05:05
makes it hard to find a cancer,
05:07
but it turns out
05:09
that it's also a powerful predictor
05:11
of your risk for breast cancer.
05:13
It's a stronger risk factor
05:15
than having a mother or a sister with breast cancer.
05:17
At the time my patient posed this question to me,
05:20
breast density was an obscure topic
05:23
in the radiology literature,
05:25
and very few women having mammograms,
05:27
or the physicians ordering them,
05:29
knew about this.
05:31
But what else could I offer her?
05:33
Mammograms have been around since the 1960's,
05:36
and it's changed very little.
05:39
There have been surprisingly few innovations,
05:41
until digital mammography was approved
05:44
in 2000.
05:46
Digital mammography is still an X-ray of the breast,
05:48
but the images
05:51
can be stored and manipulated digitally,
05:53
just like we can with a digital camera.
05:55
The U.S. has invested
05:58
four billion dollars
06:00
converting to digital mammography equipment,
06:02
and what have we gained from that investment?
06:05
In a study funded by over 25 million taxpayer dollars,
06:08
digital mammography was found
06:12
to be no better over all
06:14
than traditional mammography,
06:16
and in fact, it was worse in older women.
06:18
But it was better in one group,
06:21
and that was women under 50
06:24
who were pre-menopausal and had dense breasts,
06:26
and in those women,
06:29
digital mammography found twice as many cancers,
06:31
but it still only found 60 percent.
06:34
So digital mammography
06:37
has been a giant leap forward
06:39
for manufacturers
06:41
of digital mammography equipment,
06:43
but it's been a very small step forward for
06:45
womankind.
06:47
What about ultrasound?
06:50
Ultrasound generates more biopsies
06:52
that are unnecessary relative to other technologies,
06:54
so it's not widely used.
06:56
And MRI is exquisitely sensitive for finding tumors,
06:59
but it's also very expensive.
07:02
If we think about disruptive technology,
07:06
we see an almost ubiquitous pattern
07:09
of the technology getting smaller and less expensive.
07:11
Think about iPods compared to stereos.
07:14
But it's the exact opposite in health care.
07:17
The machines get ever bigger
07:20
and ever more expensive.
07:22
Screening the average young woman with an MRI
07:26
is kind of like driving to the grocery store in a Hummer.
07:29
It's just way too much equipment.
07:33
One MRI scan
07:35
costs 10 times what a digital mammogram costs.
07:37
And sooner or later, we're going to have to accept the fact
07:40
that health care innovation
07:42
can't always come at a much higher price.
07:44
Malcolm Gladwell wrote an article in the New Yorker
07:49
on innovation,
07:51
and he made the case that scientific discoveries
07:53
are rarely the product of one individual's genius.
07:56
Rather, big ideas can be orchestrated,
08:00
if you can simply gather
08:03
people with different perspectives in a room
08:05
and get them to talk about things
08:07
that they don't ordinarily talk about.
08:09
It's like the essence of TED.
08:11
He quotes one innovator who says,
08:14
"The only time a physician and a physicist get together
08:16
is when the physicist gets sick."
08:20
(Laughter)
08:22
This makes no sense,
08:24
because physicians have all kinds of problems
08:26
that they don't realize have solutions.
08:28
And physicists have all kinds of solutions for things
08:31
that they don't realize are problems.
08:33
Now, take a look at this cartoon
08:36
that accompanied Gladwell's article,
08:39
and tell me if you see something disturbing
08:41
about this depiction of innovative thinkers.
08:43
(Laughter)
08:46
So if you will allow me a little creative license,
08:48
I will tell you the story
08:52
of the serendipitous collision
08:54
of my patient's problem
08:56
with a physicist's solution.
08:58
Shortly after her visit,
09:00
I was introduced to a nuclear physicist
09:02
at Mayo
09:04
named Michael O'Conner,
09:06
who was a specialist in cardiac imaging,
09:08
something I had nothing to do with.
09:10
And he happened to tell me
09:12
about a conference he'd just returned from in Israel,
09:14
where they were talking about a new type of gamma detector.
09:17
Now gamma imaging has been around for a long time
09:20
to image the heart,
09:22
and it had even been tried to image the breast.
09:24
But the problem was
09:27
that the gamma detectors
09:29
were these huge, bulky tubes,
09:31
and they were filled with these scintillating crystals,
09:33
and you just couldn't get them close enough around the breast
09:35
to find small tumors.
09:38
But the potential advantage was
09:40
that gamma rays, unlike X-rays,
09:42
are not influenced by breast density.
09:44
But this technology
09:47
could not find tumors when they're small,
09:49
and finding a small tumor is critical for survival.
09:51
If you can find a tumor
09:54
when it's less than a centimeter,
09:56
survival exceeds 90 percent,
09:58
but drops off rapidly
10:00
as tumor size increases.
10:02
But Michael told me about
10:05
a new type of gamma detector that he'd seen,
10:07
and this is it.
10:09
It's made
10:11
not of a bulky tube,
10:13
but of a thin layer of a semiconductor material
10:15
that serves as the gamma detector.
10:18
And I started talking to him
10:20
about this problem with breast density,
10:22
and we realized that we might be able to get this detector
10:24
close enough around the breast
10:27
to actually find small tumors.
10:29
So after putting together
10:31
a grid of these cubes with tape --
10:33
(Laughter)
10:36
-- Michael hacked off the X-ray plate
10:39
of a mammography machine
10:42
that was about to be thrown out,
10:44
and we attached the new detector,
10:46
and we decided to call this machine
10:49
Molecular Breast Imaging, or MBI.
10:51
This is an image from our first patient.
10:55
And you can see, using the old gamma technology,
10:57
that it just looked like noise.
10:59
But using our new detector,
11:01
we could begin to see the outline of a tumor.
11:03
So here we were, a nuclear physicist,
11:06
an internist,
11:08
soon joined by Carrie Hruska, a biomedical engineer,
11:10
and two radiologists,
11:13
and we were trying to take on
11:15
the entrenched world of mammography
11:17
with a machine that was held together by duct tape.
11:19
To say that we faced
11:23
high doses of skepticism
11:25
in those early years
11:27
is just a huge understatement,
11:29
but we were so convinced that we might be able to make this work
11:31
that we chipped away with incremental modifications
11:34
to this system.
11:37
This is our current detector.
11:39
And you can see that it looks a lot different.
11:41
The duct tape is gone,
11:43
and we added a second detector on top of the breast,
11:47
which has further improved our tumor detection.
11:49
So how does this work?
11:52
The patient receives an injection of a radio tracer
11:54
that's taken up by rapidly proliferating tumor cells,
11:57
but not by normal cells,
12:00
and this is the key difference from mammography.
12:02
Mammography relies on differences
12:05
in the appearance of the tumor from the background tissue,
12:07
and we've seen that those differences
12:10
can be obscured in a dense breast.
12:12
But MBI exploits
12:15
the different molecular behavior of tumors,
12:17
and therefore, it's impervious to breast density.
12:20
After the injection,
12:24
the patient's breast is placed between the detectors.
12:26
And if you've ever had a mammogram --
12:28
if you're old enough to have had a mammogram --
12:30
you know what comes next:
12:32
pain.
12:34
You may be surprised to know
12:36
that mammography is the only radiologic study
12:38
that's regulated by federal law,
12:41
and the law requires
12:43
that the equivalent of a 40-pound car battery
12:45
come down on your breast during this study.
12:48
But with MBI,
12:52
we use just light, pain-free compression.
12:54
(Applause)
12:57
And the detector
13:03
then transmits the image to the computer.
13:05
So here's an example.
13:07
You can see, on the right, a mammogram
13:09
showing a faint tumor,
13:11
the edges of which are blurred by the dense tissue.
13:13
But the MBI image shows that tumor much more clearly,
13:15
as well as a second tumor,
13:18
which profoundly influence that patient's surgical options.
13:20
In this example, although the mammogram found one tumor,
13:24
we were able to demonstrate three discrete tumors --
13:27
one is small as three millimeters.
13:30
Our big break came in 2004.
13:33
After we had demonstrated that we could find small tumors,
13:38
we used these images
13:41
to submit a grant to the Susan G. Komen Foundation.
13:43
And we were elated when they took a chance
13:46
on a team of completely unknown investigators
13:49
and funded us to study
13:51
1,000 women with dense breasts,
13:53
comparing a screening mammogram to an MBI.
13:55
Of the tumors that we found,
13:58
mammography found
14:00
only 25 percent of those tumors.
14:02
MBI found 83 percent.
14:05
Here's an example from that screening study.
14:08
The digital mammogram was read as normal
14:11
and shows lots of dense tissue,
14:13
but the MBI shows an area of intense uptake,
14:15
which correlated with a two-centimeter tumor.
14:18
In this case, a one-centimeter tumor.
14:21
And in this case,
14:24
a 45-year-old medical secretary at Mayo,
14:26
who had lost her mother to breast cancer when she was very young,
14:29
wanted to enroll in our study.
14:32
And her mammogram showed an area of very dense tissue,
14:34
but her MBI showed an area
14:37
of worrisome uptake,
14:39
which we can also see on a color image.
14:41
And this corresponded
14:44
to a tumor the size of a golf ball.
14:46
But fortunately it was removed
14:48
before it had spread to her lymph nodes.
14:50
So now that we knew that this technology
14:54
could find three times more tumors in a dense breast,
14:56
we had to solve one very important problem.
14:59
We had to figure out how to lower the radiation dose,
15:02
and we have spent the last three years
15:05
making modifications to every aspect of the imaging system
15:08
to allow this.
15:11
And I'm very happy to report that we're now using a dose of radiation
15:13
that is equivalent to the effective dose
15:16
from one digital mammogram.
15:18
And at this low dose, we're continuing this screening study,
15:20
and this image from three weeks ago
15:23
in a 67-year-old woman
15:26
shows a normal digital mammogram,
15:28
but an MBI image
15:30
showing an uptake that proved to be a large cancer.
15:32
So this is not just young women that it's benefiting.
15:35
It's also older women with dense tissue.
15:38
And we're now routinely using one-fifth the radiation dose
15:41
that's used in any other type of gamma technology.
15:44
MBI generates four images per breast.
15:48
MRI generates over a thousand.
15:51
It takes a radiologist
15:54
years of specialty training
15:56
to become expert in differentiating
15:58
the normal anatomic detail
16:00
from the worrisome finding.
16:02
But I suspect even the non-radiologists in the room
16:04
can find the tumor on the MBI image.
16:07
But this is why MBI
16:10
is so potentially disruptive --
16:12
it's as accurate as MRI,
16:14
it's far less complex to interpret,
16:16
and it's a fraction of the cost.
16:19
But you can understand why there may be
16:21
forces in the breast-imaging world
16:23
who prefer the status quo.
16:25
After achieving what we felt were remarkable results,
16:29
our manuscript was rejected
16:32
by four journals.
16:35
After the fourth rejection,
16:37
we requested reconsideration of the manuscript,
16:39
because we strongly suspected
16:41
that one of the reviewers who had rejected it
16:43
had a financial conflict of interest
16:45
in a competing technology.
16:47
Our manuscript was then accepted
16:50
and will be published later this month
16:52
in the journal Radiology.
16:55
(Applause)
16:57
We still need to complete the screening study using the low dose,
17:05
and then our findings will need to be replicated
17:08
at other institutions,
17:10
and this could take five or more years.
17:12
If this technology is widely adopted,
17:15
I will not benefit financially in any way,
17:18
and that is very important to me,
17:21
because it allows me to continue to tell you the truth.
17:24
But I recognize --
17:28
(Applause)
17:30
I recognize that the adoption of this technology
17:34
will depend as much on economic
17:37
and political forces
17:39
as it will on the soundness of the science.
17:41
The MBI unit has now been FDA approved,
17:44
but it's not yet widely available.
17:47
So until something is available
17:50
for women with dense breasts,
17:52
there are things that you should know
17:54
to protect yourself.
17:56
First, know your density.
17:58
Ninety percent of women don't,
18:00
and 95 percent of women don't know
18:02
that it increases your breast cancer risk.
18:04
The State of Connecticut became the first and only state
18:07
to mandate that women receive notification
18:10
of their breast density
18:12
after a mammogram.
18:14
I was at a conference of 60,000 people in breast-imaging
18:17
last week in Chicago,
18:20
and I was stunned that there was a heated debate
18:22
as to whether we should be telling women
18:25
what their breast density is.
18:27
Of course we should.
18:29
And if you don't know, please ask your doctor
18:31
or read the details of your mammography report.
18:34
Second, if you're pre-menopausal,
18:37
try to schedule your mammogram
18:39
in the first two weeks of your menstrual cycle,
18:41
when breast density is relatively lower.
18:43
Third, if you notice a persistent change in your breast,
18:46
insist on additional imaging.
18:49
And fourth and most important,
18:52
the mammography debate will rage on,
18:54
but I do believe that all women 40 and older
18:57
should have an annual mammogram.
19:00
Mammography isn't perfect,
19:02
but it's the only test that's been proven
19:04
to reduce mortality from breast cancer.
19:06
But this mortality banner
19:09
is the very sword
19:11
which mammography's most ardent advocates use
19:13
to deter innovation.
19:16
Some women who develop breast cancer
19:18
die from it many years later,
19:21
and most women, thankfully, survive.
19:23
So it takes 10 or more years
19:25
for any screening method
19:27
to demonstrate a reduction
19:29
in mortality from breast cancer.
19:31
Mammography's the only one that's been around long enough
19:33
to have a chance of making that claim.
19:35
It is time for us to accept
19:38
both the extraordinary successes of mammography
19:41
and the limitations.
19:43
We need to individualize screening
19:45
based on density.
19:47
For women without dense breasts,
19:49
mammography is the best choice.
19:51
But for women with dense breasts;
19:54
we shouldn't abandon screening altogether,
19:56
we need to offer them something better.
19:58
The babies that we were carrying
20:03
when my patient first asked me this question
20:05
are now both in middle school,
20:08
and the answer has been so slow to come.
20:11
She's given me her blessing
20:16
to share this story with you.
20:18
After undergoing biopsies
20:21
that further increased her risk for cancer
20:23
and losing her sister to cancer,
20:26
she made the difficult decision
20:28
to have a prophylactic mastectomy.
20:30
We can and must do better,
20:34
not just in time for her granddaughters
20:37
and my daughters,
20:40
but in time for you.
20:42
Thank you.
20:44
(Applause)
20:46

▲Back to top

About the Speaker:

Deborah Rhodes - Physician, cancer researcher
Deborah Rhodes is an expert at managing breast-cancer risk. The director of the Mayo Clinic’s Executive Health Program is now testing a gamma camera that can see tumors that get missed by mammography.

Why you should listen

For all of the lives it saves, mammography still cannot detect the early onset of breast cancer in as many as one of every four women ages 40 to 49. And women with dense breast tissue are four to six times more likely to develop cancer than others. Deborah Rhodes and her colleagues at the Mayo Clinic in Minnesota think they’ve found an effective way to screen these high-risk patients: molecular imaging.

Rhodes, who specializes in evaluating and managing breast cancer, is collaborating with a nuclear physicist and various radiologists on a dual-head “gamma camera” that can capture the tiny tumors in dense tissue. The new technique, which would complement (not replace) mammography, is sensitive enough to pick up a mass two-fifths of an inch in diameter. Molecular breast imaging requires patients to be injected with a radioactive drug, but it is much more comfortable than the vise-grip mammogram and is expected to cost only slightly more.