Every cancer patient deserves their own equation: Kristin Swanson at TEDxUChicago 2014
Summary
TLDRIn this talk, a mathematician shares a personal story about losing several family members, including their father, to cancer. This personal experience inspired them to apply mathematics to cancer treatment, specifically focusing on improving clinical decisions for individual patients. The speaker highlights how mathematical models can help tailor treatments to unique patients, rather than relying on generalized clinical trial results. They illustrate the potential of these models to better predict tumor growth, treatment success, and improve patient outcomes, advocating for more personalized and precise cancer care.
Takeaways
- 🧑⚕️ The speaker shares personal experiences of losing family members, including their father, to cancer, highlighting the widespread impact of the disease.
- 📊 The speaker emphasizes that cancer is a significant global issue, and everyone can likely relate to knowing someone affected by it.
- ➕ The speaker believes math can contribute to finding solutions for cancer, despite the complexity of the disease.
- 👨👧👦 The speaker's love for math began at a young age, influenced by their father, who played math games with them during long road trips.
- 💔 The speaker's father was diagnosed with lung cancer at age 62, which deeply impacted the speaker and led them to focus on cancer research.
- 📚 The speaker pursued a PhD in applied mathematics, studying under Jim Murray, a pioneer in mathematical biology, to explore the intersection of math and biology.
- 👩🔬 Attending tumor boards, the speaker learned how physicians make complex treatment decisions based on clinical trials and patient data.
- 🏥 Clinical trials typically focus on the 'median patient,' but the speaker argues that every patient is unique and should not be treated as a statistic.
- 🔬 The speaker's research focuses on glioblastoma, an aggressive brain cancer, and how math models can help predict tumor behavior and improve treatment plans.
- 💡 The speaker advocates for using mathematical models to better individualize cancer treatments, potentially improving patient outcomes and avoiding unnecessary suffering.
Q & A
What personal experience did the speaker have with cancer?
-The speaker’s father was diagnosed with lung cancer when the speaker was 20 years old. The speaker was heavily involved in his care during the last seven months of his life, which motivated the speaker to study the intersection of math and biology to contribute to cancer research.
How did the speaker’s background in mathematics influence their approach to cancer research?
-The speaker’s background in mathematics, cultivated by their father during childhood, gave them the tools to approach cancer as a complex system. This mathematical approach became the foundation for developing models to understand and improve cancer treatments.
What is the role of mathematical biology in cancer research, according to the speaker?
-Mathematical biology helps understand the complexity of cancer as a system. It allows researchers to model tumor growth, cell behavior, and treatment responses, providing more personalized and effective strategies for cancer treatment.
What frustrations did the speaker experience while dealing with their father's cancer treatment?
-The speaker felt powerless and frustrated by their inability to contribute to medical decisions and understand the medical conversations. Despite their mathematical background, they struggled to grasp the complexity of cancer treatment decision-making at the time.
Why does the speaker believe the solution to cancer involves math?
-The speaker believes math is essential because it allows for better modeling of the complexity of cancer. Mathematical tools can simulate how tumors grow, how they respond to treatments, and predict outcomes, moving beyond generalized clinical trial results to more personalized patient care.
What is the concept of ‘treating to the median’ in cancer treatment, and what is its limitation?
-‘Treating to the median’ refers to basing treatment decisions on the average outcomes from clinical trials, which may not reflect the individual patient's response. The limitation is that every patient is different, and treating them based on a median approach can overlook those who fall outside the typical response curve.
How does the speaker suggest cancer treatment should be individualized?
-The speaker suggests using mathematical models to predict how a specific patient's tumor will behave and respond to treatment. This would allow doctors to tailor treatment plans more precisely to the individual patient rather than relying solely on generalized data from clinical trials.
What example does the speaker give of a case where traditional clinical trials might mislead treatment decisions?
-The speaker gives an example of a 66-year-old patient with a brain tumor whose tumor grew during treatment, which would typically be considered a failure. However, the patient lived for five years, much longer than expected, suggesting that despite the tumor growth, the treatment was effective in derailing the tumor's course.
What is the role of MRI in cancer treatment decision-making, according to the speaker?
-MRI scans are used to provide detailed images of tumors before and after treatment. Physicians use these images to assess the tumor's size and shape, plan surgeries, and tailor radiation doses. However, the speaker argues that MRI alone may not tell the full story of a tumor's behavior and response to therapy.
What drives the speaker’s commitment to improving cancer treatment through mathematics?
-The speaker is driven by personal loss, including the death of their father and other family members to cancer. Their work is further inspired by the colleagues and friends in their lab who have faced cancer. They are motivated to ensure that every patient receives treatment tailored to their unique condition.
Outlines
👨👩👧👦 Family Ties and Personal Loss to Cancer
The speaker introduces their family members who passed away from cancer, emphasizing that many in the audience likely share similar experiences. They explain that their personal connection to cancer began with their father, who was a Vietnam War veteran and engineer, diagnosed with lung cancer at age 62. The speaker recounts the emotional and logistical challenges of caring for their father while balancing university studies in math and physics, ultimately feeling powerless in the face of medical decisions they couldn’t fully comprehend.
📊 The Role of Clinical Trials in Cancer Treatment Decisions
This section delves into how clinical trials guide treatment decisions in tumor boards. Physicians rely on clinical trials that focus on median outcomes to determine the best therapy for cancer patients. However, the speaker highlights the complexity of individual patient responses, stressing that clinical trials provide general trends (median survival rates), but not all patients respond in the same way. The speaker uses their own family’s experiences to illustrate the diversity of cancer outcomes, where one brother lived longer than expected while another died earlier.
🧠 Studying Brain Tumors: A Deeper Look into Glioblastoma
The speaker shifts focus to their research on glioblastoma, a highly aggressive and uniformly fatal type of brain cancer. They explain how tumor boards discuss individual patient cases, with medical teams trying to find personalized treatment plans. MRI scans play a critical role in this process, giving detailed images of the tumor's location and response to treatment. The speaker emphasizes that even though median survival is low, each patient is unique, which makes cancer treatment a combination of scientific precision and the art of medicine.
📉 The Complexities of Cancer Growth and Treatment Response
The speaker discusses the limitations of judging a treatment's success based on tumor size alone. In one case, although a patient’s tumor grew during treatment (usually indicating failure), the patient lived much longer than expected, defying median statistics. This suggests that understanding cancer is more complex than just tracking tumor size, as factors like individual cell behavior and treatment impact on tumor growth trajectories are crucial. The speaker calls for a more nuanced approach to treatment decisions, focusing on how therapies deflect tumor growth rather than simply stopping it.
Mindmap
Keywords
💡Cancer
💡Mathematics
💡Clinical trials
💡Tumor board
💡Median patient
💡Tumor growth
💡Mathematical biology
💡MRI
💡Personalized medicine
💡Complex systems
Highlights
The speaker's family has been significantly impacted by cancer, with multiple members losing their lives to the disease.
Cancer is a widespread problem, and many people can relate to knowing family, friends, or colleagues who have battled the disease.
The speaker emphasizes that while the problem of cancer is well known, the solution to it might involve mathematics.
The speaker's early love for mathematics was sparked by solving math puzzles with their father, who was an engineer and a pilot.
After their father was diagnosed with lung cancer, the speaker was deeply involved in his care, commuting back and forth while finishing a degree in mathematics and physics.
The speaker expresses frustration with being unable to contribute to medical decisions and not understanding the discussions around their father's cancer treatment.
This frustration led the speaker to pursue a PhD in applied mathematics, focusing on the intersection of mathematics and cancer biology.
The speaker worked with Jim Murray, a leading figure in mathematical biology, to bridge the gap between math and clinical decision-making.
Tumor boards, where physicians from different specialties collaborate on patient care, are a key part of cancer treatment, but their decisions are often based on median results from clinical trials.
The speaker notes that clinical trials often group patients with the same diagnosis, but there is a broad spectrum of outcomes, and treating based on the median can be misleading.
The speaker's research focuses on modeling tumor growth and response to therapy, using mathematical models to predict individual patient outcomes rather than relying on average results.
An example is given of a patient with a brain tumor (glioblastoma) who, despite tumor growth, lived five years due to the therapy keeping the tumor off its natural growth course.
The speaker highlights the potential of using mathematical models to guide personalized treatment decisions, extending or adjusting therapies based on predicted individual outcomes.
Mathematical tools can help physicians simulate different treatment options, offering a more tailored approach to cancer care rather than relying solely on clinical trial averages.
The speaker concludes by emphasizing the uniqueness of each patient and the need for individualized equations to optimize cancer treatment.
Transcripts
this is my
brother my aunt my mother-in-law my
other brother and my
father these are some of my family
members who have lost their life with
cancer now the sad fact is that the
numbers tell us that each one of you in
this audience also can generate a
similar list of people around you family
friends colleagues who have battled
cancer I don't have to convince you that
the problem of cancer is a big big one
but I might have to convince you of my
conviction that the solution involves
math so my
introduction
to my introduction to cancer involved my
father he was an engineer and a pilot he
served three tours in Vietnam before
retiring he retired shortly after I was
born so what that meant for me is in his
retirement he became an antique dealer
so that meant I spent many many moons
many hours on the road trans traveling
from antique show to antique show we uh
I by the time I was in I think third
grade I had traveled to 48 different
states so on those long Journeys we had
to do something and one of the things we
often did was play play games
specifically Math Games my father would
teach me well if x + y is 5 and x * Y is
6 what are X and Y I was like 6 years
old so uh needless to say that interest
and enthusiasm for mathematics started
early unfortunately when my father was
62 when I was 20 uh my father was
diagnosed with lung
cancer uh I was very involved with my
father's care um unfortunately he died 7
months after after he was
diagnosed but being involved in his care
um meant that I was commuting back and
forth from tane University where I was
finishing my my bachelor's degree in
math and
physics so in traveling back and forth I
would spend all this time with my father
and his dying months and I would spend a
lot of time with his Physicians trying
to figure out what the next best step
was for my father now at the same time I
also happened to be doing research at
tan with some lovely mathematical
biologists who were doing doing some
amazing work on sperm motility in tubes
but it is mathematics meeting biology I
don't think you can dispute that right
so those two things brought together an
opportunity for me said well my father
is sitting here I could watch him over
those sad seven months watch him die and
I watched all those decisions being made
by The Physician about what is the next
best choice for my dad should we do this
therapy should we do that therapy is he
failing this therapy should we move on
to another
therapy and I remember being so
powerless and frustrated in the fact
that I couldn't contribute to those
conversations and worse yet I couldn't
understand what the conversation was
happening it was like a man with a
bubble above his head with an equation
and it just didn't add up to me and I
was a mathematician or a virging one at
that so while I was doing this research
at tane on sperm motility and tubes I
could see that maybe there was an
opportunity for me to bring math the
mathematics my father taught me to the
biology of cancer and contribute to the
world that to improve the life of my
father and others like him so yes I'm a
mathematician and I study
cancer so a month after my father was
died um I uh was admitted to a PhD
program and Applied Mathematics I chose
this specific program because it had the
world leader what I would argue as the
father of the father of the field of
mathematical biology his name was Jim
Murray he had just moved a few years
earlier from the center for Center for
mathematical biology at Oxford
um and so he was the perfect person to
introduce me and to tutor me in the in
the field of the interface between
mathematics and
biology at the same time I was very
interested at the interface between
mathematics and the clinic and how
Physicians were interacting with
patients and how I could create this
opportunity for adding math to our
understanding of cancer so I began
attending tumor board so tumor boards
are totally interesting unique
environment right so each they're
basically often a large table group of
Physicians around this table and they're
all from different areas so there's
Radiologists there's physi physicist um
radiation
biologist there's radiologist there's a
oncologist there's surgeons uh there's
chemotherapist you name it they're all
around the table and they're all making
the best decision possible they're all
collaborating to figure out what the
best decision possible is for that
individual patient but what's also
interesting kind of like the bubble
above the head of the Physicians I was
telling you earlier that couldn't decode
what I learned was that the the Baseline
decisions that were being made were a
function of a simple answer clinical
trials so each physician has in their
back pocket some knowledge of the result
of how how a patient um how patient
should do give if I give them this
therapy ver versus the next therapy and
the format of those clinical
trials are something like this you group
get a group of homogeneous patients
together homogeneous in the sense that
we call them the same diagnosis and I
use air quotes liberally in this
particular context so they have the same
diagnosis they're extremely similar in
some way and we want to introduce a new
therapy to them and we want to compare
that to some standard approach what we
call the standard of care so these
physician these patients are then given
some therapy and the net result is a
survival curve so at the end of the day
the key measure of this of this outcome
of this trial is this curve so the curve
has little steps on it each step is
actually a patient dying and so you get
a a sense of the degree to which this
therapy is working summarized by one
thing this dot in the middle the Green
Dot that's the median patient that's the
that's the average result that is the
result that's in the pockets of the
physicians at the clinical Tri at the
room in the tumor board so they couldn't
possibly carry this whole curve around
them the reality is you look at this and
you say okay well we're talking the
median patient that's where the
terminology treating to the median comes
from from treating to the mean but now
we all know we all know in this room
that everybody's different my father was
here he died very early compared to when
he should have by the median my brother
was here and my other brother was way
out on the table tail he lived three
years when he would really shouldn't
have so we know there's a diversity we
know there's Shades of Gray everybody is
different they're all part of the
rainbow and the reality is how do you
match those things together how do you
use information about individual
patients and combine that with median
understanding of a clinical trial and so
that's where our tools sort of interface
so my research began into a specific
type of cancer known as Goma
specifically the most aggressive of
those
gastas uh I was working with this
amazing neuropathologist Buster alard um
he was a great Menor mentor and friend
in this
area so my research into this disease uh
began began um at this tumor board where
I would learn all about what the
clinical decisions were being made but
what's interesting and what's horrible
about this disease is it's a horrible
disease it's a very bad disease patients
typically live 15
months so the median survival for those
patients it's 15 months and it's also
considered uniformly fatal the term is
used in the first sentence of several
any article you would pull up on this
topic but we all know and the Physicians
all know that it's still just part of a
rainbow that each patient is part of a
rainbow this rainbow so now let's fast
forward to a tumor board for the
specific case of a of primary brain
tumors gomas gly
blastoma so each of these patients are
being discussed by The Physician they're
trying to figure out what the best plan
is for that particular patient knowing
in their back pocket what the green
patient's going to do what that median
patient should be expected to
do so the Physicians are often trying to
F to fight for that individual patient
right they're often coming to the table
going hey we know this patient is part
of a rainbow we just don't know where in
the rainbow it is where in the rainbow
this patient it will will lie and so
there's lots of indicators for figuring
that out and so this is kind of where
the science of biology meets the art of
medicine it's figuring out where these
this interface might be so one of the
tools that Physicians use in the case of
T brain tumors is Mr so here is an MRI
three-dimensional reconstruction of an
MRI we're going to scroll from the
bottom of the brain all the way up
you'll see it outlined in red the tumor
and you're going to scroll back down and
rebuild the threedimensional region that
is the tumor now this is exquisitely
detailed right it's amazingly beautiful
right it's all math and physics
ironically
underneath but what's interesting is
Physicians have to use these images to
figure out how to tailor their therapy
Choice given their knowledge of the
clinical trial that tells me the median
patient is expected to live 15 months
right these images are used by surgeons
to figure out how to approach the tumor
there are radiation therapists that
figure out how to um sculpt their dose
for radiation therapy they're
also also uh used by uh Physicians to
understand response to therapy and so
that's the case I would like you to
think about first so in this case you
see the tumor and there's a brain
there's a tumor you can at least see the
tumor so this is good okay so the tumor
um prior to treatment treatment is
introduced post treatment there is
another there is a a post- treatment
response another another image so what's
interesting about this is that if you
have um a physician you're sitting here
going well look prior to treatment this
tumor was this size after treatment was
another size at the intervening time
there was some sort of treatment in the
decision is at the second time point
should we change therapy the clinical
trial says we should this patient should
have an improved survival of whatever
amount by because of the median patient
outcome but not necessarily for um this
particular patient so do we do we make
the next what's the next choice for this
particular patient so clinically it is
determined that this would be considered
a
failure the tumor grew it got bigger as
a function of therapy so it's a failure
but I'm going to challenge you to
reconsider that choice reconsider the
possibility that this may not be a
failure I'm going to give you a little
insight this particular patient was a
66y old male who ended up living five
years with a
Goma so if that's a failure I'm not sure
I'm not sure what is so now let's go
back to our our MRI and think about the
Physicians are sitting there talking
about these two images the image the
tumor grew grew through therapy
therefore we should change treatment but
also we have these Exquisite detail of
this image and we have an Exquisite
understanding of what's underlying this
underlying the image that we see
specifically in this case we know if we
were to zoom
in we could if we had the tool for every
patient if we could zoom in we would
find individual cells migrating
individual cells proliferating within
this tumor you would see them um growing
and migrating and responding potentially
to the therapy that's being applied to
them so what's interesting is although
we have this Exquisite image that can't
quite get down to the Single Cell detail
what you understand about the tumor is
that it's a complex interacting system
this complex interacting system of
agents this complex system is start of
at the basis of a lot of what we do
scientifically in the world right the
weather in the evening news is all about
a complex system where data is
integrated about the current wind speeds
right are understanding the financial
markets there are people all populating
Wall Street that are all about
quantifying and integrating complex
information to understand the complex
system that is the World Market to make
some sort of
predictions so at that so at a base one
could use math to understand this
complexity and even if you're talking
really simply talking about Sumer cells
proliferating invading and responding to
therapy you can write down a really
simple
equation that said the point is is you
look at the words in this case we're
writing down that the cells are
migrating proliferating and respond to
therapy and then you go back to those
Exquisite MRIs that the Physicians were
dealing with in tumor board and you run
this forward in time thankfully you
still see the red tumor here so now
let's look at the same exact case of
that same exact patient but let's
imagine we had the opportunity and the
insight to understand what that tumor
was going to do without
treatment and here is the tumor evolving
in time you can see the untreated course
has not
changed untreated growth compared to the
legion
above now I ask you is this a
response turns out it is so one of the
problems in these clinical trials and
one of the problems with new therapies
that are studied is the fact that if the
tumor grows through therapy The
Physician changes course well what's
interesting is not all patients that
grow through therapy end up doing poorly
just like this patient this patient
lived five years okay but it turns out
in our studies of now hundreds of
patients going on thousands we've looked
at this sort of response the degree to
which the treatment deflects the tumor
off its course is a response that tells
you that that patient is going to live
longer right so this is a completely new
tool to think about when you're a
physician
and you're struggling with the
treatments in the intervening there's
the post treatment tumor grew okay let's
change course well this treatment
derailed this tumor off course for quite
quite significantly to so much so that
this patient actually ended up living
quite long and maybe they should have
continued on this therapy because it was
continuing to keep the tumor off course
keep the tumor derailed right so having
this out this this in your back pocket
is a completely new way to think about
not the median patient but uh but other
other patients but individualizing our
understanding of
patients so now imagine the case where
you've got a tumor you you patient shows
up at tumor board we're all sitting
around this table again and now you have
something equivalent of an iPad
app on that app you have options for
surgery radiation chemotherapy that's
because the surgeons have their have
their plan they have a whole
navigational system where they figure
out how they're going to approach the
tumor same thing thing for the radiation
oncologists they have a plan they're
going to input they're they're going to
they're going to have a plan they're
going to bring it to the table and the
chemotherapist they have a plan they're
going to bring it to the table now a lot
of those plans are based on
understanding of results of clinical
trials but what they don't know is
really how those things marry together
how those things add together right so
now let's imagine the surgical team
brings in their their plan and the
radiation oncology team brings in their
plan and they're going to apply them in
this way so roughly on day 25 we're
going to have the surgery of of a signif
of what whatever degree which I'll show
you in just a moment and then the
radiation oncology plan involves
sculpting the radiation dose so that it
salvages much as much as possible of the
normal normal brain tissue because you
don't want to radiate the normal brain
if you can all avoid it so now let's add
those things
together so in this case we're going to
we see the tumor tumor being reected
there's a large reection from the
surgery there's the radiation therapy
being introduced you you can see the
tumor being derailed sort of off course
during that and then the
recurrence so although this is not the
ideal plan ideal result right this is
what would this patient would actually
receive this is in fact the standard of
care for this particular patient this is
actually what this patient actually
ended up receiving in real life what you
can say is well now you're sitting at
tumor board and you can say well hey
I've got an opportunity here this
patient was significantly deflected off
of their growth curve their tumor was
significantly deflected off their growth
curve during radi ation let's just
extend that out a little bit or approach
it in a slightly different way or what
if we did that first and then did the
surgery later there's a lot of options
that can be played out that are not
practical currently because we focus so
much on the clinical trials of a median
patient this is a way in which you can
think about individualizing our
understanding a given patient's tumor
given patient's response to treatment
and capitalizing on that for that
patient so if my father's Physicians had
had tools like this in their hands
perhaps my dad would not have received
that last round of therapy that was so
painful to his quality of life perhaps
he would have received the next round of
a different therapy that was actually
more successful for him and so that is
the thing that that certainly drives
me this is my
father these are the faces of the folks
in my lab the colleagues of friends and
family of folks in my lab who have all
faced cancer and they are the people
that remind us every day that every
patient is unique and every patient
truly Des deserves their own equation
thank
[Applause]
you
Посмотреть больше похожих видео
Personalized Medicine in Cancer: What does it mean and how is it done?
Digital twins in cancer care
Paclitaxel vs Nab-Paclitaxel in Metastatic TNBC
The key to solving the doctor-patient relationship | Stephen Khachikian | TEDxRapidCity
GUESS WHAAAAAT!?!
Artificial intelligence in healthcare: opportunities and challenges | Navid Toosi Saidy | TEDxQUT
5.0 / 5 (0 votes)