The Health of Women of Color: A Critical Intersection at the Corner of Sex/Gender and Race/Ethnicity

>>Jennifer Plank: Thank
you for attending this symposium on the health of
women of color, a critical intersection at the
corner of sex/gender and race/ethnicity;
and today you will see presentations by six
fantastic researchers. And right now, I would
like to take the opportunity to introduce
you to the co-chairs of this symposium, Janine
Clayton and Marie Bernard to give you some
opening remarks.>>Janine Clayton:
Thank you Dr. Plank. You didn’t introduce
yourself; this is Dr. Jennifer Plank. And we’re delighted
to have you all here. And I’m particularly
excited that we are able to talk about this
really important subject at the NIH
Research Festival. So you can see our little
graphic here that we’re very concerned about the
health of women of color; and one of the ways
we’re thinking about this as a critical
intersection at the corner of sex/gender
and race/ethnicity. And we are so — we do
think is so important that we actually spend extra
effort to try to get data disaggregated by
sex/gender and race/ethnicity,
simultaneously, so that we can look at what
is occurring in various populations of
women of color. And the Office of Research
on Women’s Health will be releasing the Women of
Color Health Data Book. This is the
fourth edition. You should have a little
card in your materials there. It’ll be released
as an e-book. And of course the — one
of the statistics I want to share with you in terms
of setting the stage is the fact that by 2043
the United States will become a majority-minority
nation. And I don’t know if that
means we should just stop using those terms since
that’s where we’re going, and really prefer to use
descriptive terms to talk about different
populations of women; because I don’t think that
it’s helpful to define one group of individuals in the
context of another group. By 2050, women of color will
be 53 percent of the total U.S. female
population. And so the Women of Color
Health Data Book is going to have information that
I think will be helpful to scientists and
clinicians alike. We know that the leading
cause of death in women is heart disease. In general, however, if
you look at populations of women of color, you can
see there, that it actual varies and that cancer is
the leading cause of death over heart disease in
Asian/Pacific Islander women, Hispanic women, and
American Indian/Alaskan Native women. So that gives you an idea
of when you start to look at populations of women of
color and really carefully parse that out, what you
start finding differences. And so we’ll be
talking about this critical intersection. And I wanted to just share
with you in the context of this session,
where NIH and the Office of Research on
Women’s Health is looking at this issue beyond inclusion,
beyond looking at including women in minorities in
clinical research, we’re thinking about it as
sex and gender influences. So we’re thinking about
this as an important issue that needs to be infused
and integrated throughout the entire research
continuum, from the basic to the clinical
to the translational. And that sex and gender,
considerations of sex and gender, need to be
incorporated at the pre-clinical level. So when you’re doing
those animal studies, looking at male and female
animals, when you’re doing cell culture, understanding
whether you’re working with male or female cells,
because that can make a difference in your
pre-clinical work. Clearly, in toxicology
we know that this is important, and different
compounds and chemicals interact and affect men
and women differently; and that’s already considered
throughout every phase of the clinical
trial’s continuum. In terms of reporting
sex-specific data, so that that data gets out in the
literature, and you can parse the data different
ways, sex-specific analyses and reporting,
and then education and health policy, and
ultimately informing health care — because
clearly why we are doing this here at NIH is to
inform and improve health. And if — I recently added
my face for block there. So at this point I want
to turn the podium over to my esteemed colleague,
Dr. Marie Bernard, the Deputy Director of
the Aging Institute.>>Marie Bernard: So
I’d like to join Janine in welcoming you all and
thank you for being here. It’s nice to see the
room packed and the need to bring in
more chairs. This is a symposium
that was proposed by the Women of Color
Committee of the NIH Working Group on Women
in Biomedical Careers. And you see listed here
the members of that committee. Dr. Belinda Seto,
the deputy of NEI, co-chairs this
committee with me. And I particularly would
like to acknowledge — everyone on this committee
has been really busy helping to make
things move forward, but I’d particularly like
to acknowledge Debbie Cohen from ORWH for support of
things; Cerise Elliot from NIA who has
been the backbone to the Women of Color
Research Network. Let me go down my
list and make sure I’m getting
everybody. Wendy Lee and Kate Nagy
who’ve been doing the Spectrum Blog. Roland Owens who
always has a lot of sage advice
for us. Jennifer Plank who really
has made this symposium go forward. Golda Phillip and Sara
Williams, who’ve developed some regional
networks for us. Let me tell you what I
mean when I talk about those things that
I’ve acknowledged. The Women of Color
Committee was established to address the unique
needs of women of color in the sciences, including
recruitment, retention, and career advancement. The work that was reported
in Science in the summer of 2011, led by Donna
Ginther, analyzing racial and ethnic representation
of NIH grants has led to us to have some
subsequent discussions with Donna; where she did
further analyses of that intersection between race,
ethnicity, and gender, and has led us to be
reinvigorated in our attempts to facilitate
things for women of color in science. So we have the Women of
Color Research Network, a social media site
that was established by Dr. J. Taylor Harden
and Cerise Elliot. I guess that was
2007 or 2008. Is that right? 2009, so it’s
very young. From that has grown the
Spectrum Blog, which is posted once a week with
lots of good information. If you are a woman of
color in the sciences, if you are interested in
issues for women of color, it’s a great resource for
you to take a look at — the WoCRn site as
well as the blog. We’ve been trying to
systematically nominate women of color for
lectureships and awards, and this symposium is part
of that process of our recommending women to
increase their visibility. And from the Spectrum Blog
and the Women of Color Research Network, we’ve
now been able to get some local chapters of women
in color committees to — established in Indiana
and North Carolina. And [email protected], if
you’re not familiar with it already, I would
encourage you to take a look at it. Again, you don’t have
to be a woman of color to benefit from this. There’s lots of
information about ongoing events, how to go about
getting grants, what to do when you are thinking
about that to do after your post-doc. There’s more than 1,400
members of it currently. It’s an opportunity to
network and get mentoring, and so I would encourage
you to take a look at it. And before I turn
the podium back over to Jennifer, I would just
like to point out that you have handout materials
from ORWH as well as NIA. And my NIA colleagues
would kill me if I didn’t point out that in the back
you have what is your aging IQ, a workout to
go, what’s on your plate in terms of nutrition and aging,
and things about menopause. Take a look at the
materials in the back as well as the
things on your desk. Jennifer, it’s
all yours.>>Jennifer Plank: Thank you
Dr. Clayton and Dr. Bernard. And so, with that, I would
like to introduce our first presenter, who
is Dr. Lauren Wood. She is a staff clinician
at the Vaccine Branch at the Center for
Cancer Research. And today she will be
presenting on the advances and challenges in the
clinical translation of therapeutic
cancer vaccines.>>Lauren Wood: Well,
I want to thank the organizers for the
opportunity to present at this symposium. My research is not
specifically going to be addressing issues
of gender and ethnicity in terms of the conduct
of clinical trials. But I hope as an
individual who focuses on first-in-human trials
and clinical translation of novel, immune-based
therapies and therapeutic vaccines for cancer, which
clearly as Dr. Clayton highlighted is a major
issue for women of all ethnicities, that I can
highlight some of the issues that we face as we
translate clinical vaccines. Within the Vaccine Branch,
the focus of our research is novel vaccine platforms
that are developed by laboratory PIs
within our branch into first-in-human
studies. And actually, I’m going to
be talking about a peptide vaccine that was developed
for men with prostate cancer, but actually
has relevance in future clinical development,
we hope, for women with breast cancer as
potentially a novel therapeutic target. So some of the inherent
challenges that we face when we are dealing with
the immune system has got to do with the complexity
and the redundancy of the immune system; the
fact that tumor antigens are self-antigens that the
patient may actually have become tolerant to;
that immune dysfunction and evasion by these
cancers allow them to avoid eradication, especially in
patients who have advanced metastatic solid tumors,
which are the majority of patients. Most patients present
because of symptoms, and at the time of their
diagnosis they have advanced disease; and
this is particularly true of racial and ethnic
minorities who present, especially with colon,
prostate, and other types of cancer, as well
as breast cancer. The anti-tumor effect
of cancer vaccines is really
indirect. Unlike chemotherapy, where
we can basically drop the bomb, know that it
will hit the target, also probably cause some
innocent bystander damage to normal healthy tissues,
vaccine therapies are indirect by nature. We have to first stimulate
the immune system to make an appropriate anti-tumor
immune response, and that immune response then has
to ultimately end up eradicating tumors. What I’m going to focus
on, because of the nature of the peptide vaccine
that I’m going to be discussing, I’m going
to focus on two issues: tumor antigens that are
self-antigens, as well as the issue and challenges
of the anti-tumor effects and the fact that
they’re indirect. So the key function of the
immune system basically is to distinguish self
from non-self, but more importantly, it’s also to
distinguish danger from things that are
not dangerous. So it’s a combination of
the self/non-self and the danger/non-danger that
basically correlates how an immune response
is directed. Now, usually, self is not
dangerous, and typically and classically anything
that is non-self is considered dangerous. The main problems that we
see, characteristically in immunity, is that either
we have too little immunity — so we have —
it means to compromise — that’s associated with
malnutrition, cancer, HIV infection, other immune
deficiency diseases very, very common in this age of
transplantation of organs. We can have too much
immunity, where the immune system is basically
over-activated. We know clearly from
pathophysiology and preclinical animal models
that chronic inflammation is associated with
the development of malignancies, independent
of whether or not there’s also associated chronic
persistent viral infection. Self is seen as dangerous. This is a problem
when we have problems of autoimmunity, which
disproportionately affects African American women,
who are affected by lupus. Other examples of
autoimmunity include inflammatory bowel disease,
rheumatoid arthritis. Again, this element of
dysregulation is not a good thing. And it’s also a problem
when non-self is not seen as dangerous. As what we see when
we have established infections that are
chronic and persistent, such as with human
papillomavirus, hepatitis B, Epstein-Barr
virus, hepatitis C, which ultimately can
lead to inflammation. So here’s my quadrant of
the good and the bad, of self versus non-self,
danger versus non-danger. And up across the X-axis,
we have self versus non-self; and along
the Y-axis, we’ve got danger and
non-danger. When there is a danger
signal, and self is seen as dangerous,
it’s bad. That’s what you get,
autoimmune disease. We don’t
want that. When non-self is seen as
dangerous, that’s good. We want to reject things
that are non-self. When it comes to self,
we want it to be seen as non-dangerous, because
basically that’s good. Self is not attacking
self, there’s health — there’s immune system —
is in balance with cell’s tissues and they’re
healthy in balance, and there’s no
autoimmune disease. If non-self, which is
foreign, is not seen as dangerous, then
the non-self becomes tolerated; and that’s
what we see with chronic persistent infections, such
as EBV, hepatitis B, that leads to chronic activation,
chronic inflammation, and ultimately to malignant
transformation. So how I am charged as an
immunologist with trying to harness the immune
system to attack self tumor antigens is that I
have to deal with the fact the immune system
accomplishes its functions because it always presents
everything in the context of self. Self is described by the human
leukocyte antigen system. It’s the major tissue
histocompatibility that’s done when everybody gets
transplants, and they constitute our tissue
type, and it’s the main reason why donated
organs or tissues are either accepted or rejected,
because they need to be matched. So this is my handprint
analogy of what the immune
system sees. Just like everyone here
has a unique fingerprint, so your HLA tissue type is
your unique fingerprint. This is what the
immune system sees. And when it is rejecting
things that are non-self or foreign, it is actually
seeing this in the context of yourself. So the orange triangle,
the green octagon, the blue triangle, the
yellow sun, those represent non-self foreign agents
that are presented in the context
of self. Well, they’re easy to get
rid of and easy to reject because they’re
non-self. The challenge that I
have as an immunologist in cancer is — that we have
our self HLA tissue type, but then what we’ve got
that’s being presented to the immune system is
— we’ve got awkward self gone bad. This is clearly a jacked
up hand, but it really doesn’t look that much
different from the other one. Okay? And what I’ve got to do
is take this jacked up self that’s gone bad, and
I’ve got to somehow enhance the danger signal and
train the immune system to recognize it as dangerous
and to be rejected. So again, cancer is
self gone bad, but it’s still self. And that’s probably the
number one challenge in immunology in terms
of translating this. So now I’ve modified our
quadrant box so that bad self, which is cancer,
is not rejected. That’s the problem. It’s part of self,
but it’s not seen as dangerous, so it’s
not rejected and it gets established. So the take-home
message from that is — is that for our therapeutic
vaccines, we have to develop effective
vaccines, not only for cancer, but for chronic
persistent infections that have better antigens,
that have better danger signals, and that
overcome mechanisms of immune suppression
and immune resistance. Example for that is
blocking negative regulation, that’s
a very hot area in cancer
immunotherapy. A new negative checkpoint
regulator was approved just ten days
ago by Merck. I want to finish up by
highlighting the fact that the anti-tumor
effect of cancer vaccines is indirect and give
you an example of that in a first-in-human clinical
translational trial that we have. One of the issues that we
know is is that clinical responses to immune-based
therapies take time. So unlike oncologists
who are able to deliver chemotherapy and drop
the bomb, per se, as we deliver therapeutic
cancer vaccines, I’ve actually got to train the special
ops forces to shoot right, to arm them with bullets,
to make sure that they hit the targets, and that they
make sure the targets are ultimately
killed. And that does take a
little bit of time and training. Oftentimes we see with
immune-based therapies that there can actually
be very, very important and highly statistically
significant improvements in the clinical endpoint
of overall survival; but oftentimes, we do
not see any differences in progression-free
survival. And part of that has got
to do with the fact that actually patients who get
immune-based therapies oftentimes — sometimes
demonstrate progression before they demonstrate
regression. And this is because of
the fact that if you do harness an anti-tumor
immune response, and you start to see destruction
and necrosis of local tumor tissue, you get local
inflammation at sites of tumor, but they all look
worse on CT imaging scans. And oftentimes,
micrometastatic disease that was previously
invisible, once there is a focal anti-tumor
inflammatory response, actually becomes visible,
and it looks like there are new lesions or
more lesions there. Generally, our
immune-based therapies are really more effective in
patients who have minimal burdens of disease,
because there’s tremendous dysregulation when
patients have advanced solid tumors. And reliable validated
surrogate markers for clinical responses are
often lacking; and that’s actually what I’m going
to demonstrate for you. TARP is a 58 amino acid
protein; it was actually discovered by Ira Pastan and
colleagues here in the NCI. It is expressed in
about 90 to 95 percent of prostate cancers, which
is why we started in men first, but it’s also
overexpressed in about 50 percent of
breast cancers. It uses a different
open reading frame from the normal TCR
receptor gamma. In their initial
description, Dr. Pastan and colleagues identified
the fact that even though TARP was expressed
on normal prostate epithelium, it
was actually due to the epithelial
cells, it’s expressed at the mitochondrial
level, but it’s not due to infiltrating
tumor cells. So it’s actually unique
to the prostate organ. It’s also expressed
in breast cancer. And importantly, work by
others has shown that it’s highly expressed
in primary as well as metastatic prostate
cancer in patients who have a range of aggressiveness
in terms of histopathology. Gleason score is
associated with disease aggressiveness in
prostate cancer. And it’s important that
it’s also expressed in hormone-sensitive as
well as castrate-resistant prostate cancer, making
it ideal candidate for clinical
translation. So I’m going to present
five quick slides regarding a first-in-human
trial that we did of a peptide
vaccine. I can’t go into all the
pre-clinical animal models and testing that
were done that led to the human clinical
translation. But we did the study in
men who have D0 prostate cancer, which means
they’ve been treated for their prostate cancer,
but they have evidence of PSA biochemical
recurrence. Their PSA levels have
come up after radical prostatectomy or radiation
therapy, but they have no evidence of bone or
visceral metastatic disease. We used two TARP peptides;
the scientific principle incorporated in our
platform was that one of these peptides was the
wild-type form and another was an epitope-enhanced
peptide. Basically, we made
a single amino acid substitution to the
peptide and could document that we could enhance
the immunogenicity, and the immune responses,
and the cytolytic responses to the peptide that
was epitope-enhanced. We delivered the peptide
vaccine, either with Montanide, which is an
adjuvant in GM-CSF, or we custom made an autologous
dendritic cell vaccine and delivered it to
these gentlemen — the ultimate in personalized
cancer therapy. Our primary outcomes
were safety, toxicity, and immunogenicity, but
we also wanted to see if we could slow the rate
at which the PSA was going up in these men, because it
is [unintelligible] with clinical outcome, overall
disease progression, and survival. The one thing that I want
to show you is — is that as the PSA goes up, we
can calculate the slope. PSA doubling time is the
really, very clinically relevant outcome that most
clinicians are interested in. The bottom line is you
don’t want a PSA doubling time less than three
months, because you’re at 100 percent risk for
progressing and dying from your prostate cancer; and
if your doubling time is greater than 15 months,
the issue is — is that it could be five, 10,
even 15 years before your prostate cancer is
a problem for you. So our goal with our TARP
vaccination was basically to slow the rate of rise, how
fast the PSA was going up. And ultimately, we look
at the slope rather than doubling time. Because actually when you
totally slow the rate at which the PSA is going,
the doubling time actually goes to infinity; and when
you actually lower the PSA, the PSA doubling
time becomes negative. So the takeaway from our
first trial was that, in this group of men, we
had a highly statistically significant slowing
in PSA doubling time. The thing that I want you
to take away is is that the majority of men are
on the right hand side. This represents a change
in their slope from pre-vaccination to
post-vaccination, and it shows that we’re slowing
the rate at which the PSA is going up. It’s highly statistically
significant, and it’s seen actually in a
majority of patients. This is at 24 weeks. And we saw the same
thing at 48 weeks. I’ll get the last two
slides in in two minutes. The other thing that
we saw was that — we met our primary
immunogenicity endpoint — that the vaccine was
indeed immunogenic, and it increased
TARP-specific reactivity. Here we have at week
zero, no evidence of TARP-specific reactivity;
and again, no evidence of TARP-specific
reactivity. And at weeks 12, 18, 24,
it did not matter whether we tested it to the TARP
wild-type peptide, which is in light blue, the
epitope-enhanced peptide, which is in the medium
blue, or the native confirmation of the
non-vaccine wild-type 29-37 peptide; we saw an increase
in immunogenicity. Highly statistically
significant, it was in the majority of men, and
wonderfully, which kind of rarely happens in clinical
trials; it completely recapitulated what we
saw in the pre-clinical animal models. But the bottom line is —
is that the people who are responders, and had
a slowing in their PSA velocity, had the same
kind of TARP reactivity as non-responders. So the bottom line is —
is that even though we have a very
clinically-relevant surrogate outcome, we
have immunogenicity, we have the issue of
— I did something to the immune system. The immune system had a
clinically-relevant effect on a surrogate of outcome. It was immunogenic, but my
immune response doesn’t correlate with my clinical
outcome of interest; and that is one of
the major challenges of therapeutic
vaccines. So I wanted to highlight
that as an issue of the challenges
that we deal with in clinical
translation. And thank you very
much for your time and attention. [applause]>>Jennifer Plank:
Thank you very much. Does anybody have any quick
questions for Dr. Wood?>>Lauren Wood: I’m going
to run, go get vaccine, but I’ll answer the
questions real quick. And then I’m going
to come back. What were the
questions?>>Female Speaker: Oh,
I’m [inaudible] — that was a really good talk. Over time do you see any
specific [inaudible] that are
TARP-specific?>>Lauren Wood:
We absolutely –>>Female Speaker:
[inaudible] immune suppression
over time –>>Lauren Wood: So we
do see TARP-specific cell responses. I didn’t have time to
go into details of this particular assessment
of immunogenicity. But it was by
[unintelligible], so all of these are T-cell
responses that were documented that
are TARP-specific. We tried to look at — to
try — because there was no correlation here during
gamma [unintelligible] activity; we looked to see
if there was correlation with TARP-specific
reactivity as measured by Granzyme, or perforin
A, or Granzyme A, Granzyme B. And we still have
not found anything. We are continuing our
phishing expedition, because the
[unintelligible] uses is infinite; and we’re
looking at aggravated, TARP-specific CD8 cells
to see maybe that might be a surrogate of whether
or not individuals had that slowing in
PSA velocity. It is not only
TARP-specific cells, but are they activated
TARP-specific cells. We’re also looking to see
whether or not we see TARP-specific CD8 cells
that have evidence of highly functional
cytokine expression, because that’s shown to
correlate with actual, functional
anti-[inaudible].>>Female Speaker: I was
wondering about the new [inaudible] regulator –>>Lauren Wood: Yes.>>Female Speaker: —
[unintelligible] –>>Lauren Wood: So the
first one to be approved was for a malignant
melanoma two years ago, it’s anti-CTLA-4,
also known as the [unintelligible] or
[unintelligible]. And the most recent one is a
Merck anti-PD-1 inhibitor. Okay? I am really excited
as an immunologist, because literally four years
ago a program was approved, the oncology community
was like that’s voodoo; it doesn’t work; the
immune system — ah — not important; we
don’t care about it. But now, with the
activity and the success and the clinical response
that individuals will see, two of the checkpoint
inhibitors — where, basically, tumors are able
to express and secrete immune system proteins
that put the brakes on the immune system. So the immune system
is trying to attack [inaudible] —
that’s gone bad. But the tumors are able
to overexpress these negative regulators that
then put the breaks on the immune system. So the hunt is out
in multiple areas of biopharma to develop
[unintelligible] antibodies to these
checkpoint inhibitors that, in essence, take
the brakes off the immune system and allow
multiple planners and arms of the immune system to have
at it — at the tumors, and actually result in
regression of tumors and clinically very, very
relevant [inaudible] –>>Jennifer Plank: Okay. We’ll wrap up here and
we’ll continue questions later in the
session. [applause]>>Wendy Lee:
Good afternoon. My name is
Wendy Lee. And I am a
biologist at the National Institute of Allergy
and Infectious Diseases. And I’m also a proud
member of the Committee on Women of Color in
Biomedical Careers that Dr. Bernard
previously spoke about. It is my pleasure to introduce
to you Dr. Nakela L. Cook, who is the Chief of Staff
in the Immediate Office of the Director at the
National Heart, Lung, and Blood institute. And she’ll be speaking
to us today about cardiovascular
risk and treatment at the intersection of
race, ethnicity, and sex. Dr. Cook?>>Nakela Cook: Well, it’s
a pleasure to be here this afternoon and to be able
to speak to you about something that’s very
important to us at the institute, as well
as to me personally. I’m particularly looking
at the risk and treatment of women with
cardiovascular disease as — still volume —
the risk and treatment of women with
cardiovascular disease. But specifically, I
wanted to take an example of heart failure and some of
the conversation and talk to you about some
of the risk there that’s particularly of interest
around women of color. So many would say, why
is this even important? And I think it’s actually
important to think about this in a way
that bridges a gap. Because we know that
women represent half of the patients with
cardiovascular disease and half of the patients
with heart failure in this country. And as we saw from an
overview that Dr. Clayton gave us, is that overall,
it’s the leading cause of death for women; although
when you start to break it down by race and
ethnicity, we know that there are certain
ethnicities where this is a second leading cause
of death behind cancer. And there are known
differences that are both reported in the
epidemiology as well as the care and treatment
and clinical presentation for women with
heart disease. And the clinical outcomes
among women have improved over time, but not as
dramatically as it has for men. And thus, we know that
there is an attention that needs to be given to
the health care disparities that we see for women
as compared to men. And we know that health
care disparities exist in evidence-based treatments
that are not being given as regularly for women;
and specifically, when you look at that
intersection of race, ethnicity, and gender, that we
actually see lower rates of evidence-based care. And specifically, symptoms
are under-recognized and women have been historically
underrepresented in the trials that actually provide
that evidence base for the therapeutics that we use
on a regular basis. And not just that it’s a
socially important and socially just thing to
do, but it’s also politically
important. And that we know that
there are many members of our congressional body
that are also interested in us improving the care
of women with heart disease. So maybe women are from
Venus and men are from Mars. We know that there are
differences in biology and pathobiology
between the sexes. And just as an example
of how this is manifest, is cardiovascular disease
affects women at an older age than it does men. And why is
that the case? And is there some hormonal
interactions that actually protect women at an earlier
stage — earlier age? And we do know that amongst
a certain condition in heart disease that’s
called takotsubo cardiomyopathy, or stress
related cardiomyopathy, that 90 percent of those
affected are women. So why is that
the case? And what’s the biology and the
pathophysiology behind that? At NHLBI we sponsored
a study called the Women’s Ischemia
Syndrome Evaluation, which basically evaluated what
exactly happens in a woman that has a cardiac
event but doesn’t have an obstructive lesion in a
coronary artery that can account for
that event. And the whole syndrome of
microvascular disease, the entity itself was
defined in the study, which seems to be something
that predominantly affects women. So, again, a different
pathophysiology that seems to be present at time
between the sexes. And we know there are
documented differences, as I mentioned, in the burden
of cardiovascular disease amongst the sexes. And if we look at our
remaining lifetime risk of cardiovascular disease
at age 40, we see that one in two women are likely to
develop cardiovascular disease after the age 40
in terms of their lifetime risk; and when you look at
men, it’s two in three. And specifically,
if we narrow in on cardiovascular disease
deaths from coronary disease, stroke, or heart failure, you
actually see these differences start to occur where there’s
a slight bit of balance between what we see in terms
of deaths in overall cardiovascular/congenital
disease. But when you delve
a little deeper for coronary disease, you see
that the deaths but for women are less than that of men —
and hospital discharges less than that of men. But when you start to look
at stroke, you see higher rates of stroke and deaths
amongst women as compared to men. And heart failure itself,
while there’s a lower prevalence of heart
failure among women in this overall population
survey, when you look at actual hospital
discharges, there are a larger number of hospital
discharges amongst women as compared to men. So some imbalances there
and documented differences that we know in the
burden of disease. But there are also
documented differences in treatment. And this is actually, I
think, an area where we have lots of opportunity
for improvement. And when we have
looked over the years, we’ve proven through many
studies that women are less likely to get early
medical interventions when they present
with heart problems. They’re less likely to get
invasive cardiovascular procedures for
treatment and therapy. They have shorter or — I
should say longer times for reperfusion, which
is actually opening that blocked artery when they
come in with a heart attack. And in hospital, their death
rates are higher than men. And so we definitely know
that disparities in the treatment can contribute
to poorer outcomes; and this is an area where
we can actually make a difference
in care. And if I drill down
more specifically, and I’m using the case
example of an implantable cardioverter
defibrillator, or one of those therapeutic
treatments that usually we use in patients where
— that have a reduced ejection fraction
with heart failure. What we’re actually trying
to allow the heart the opportunity to be shocked
back into a normal rhythm if a life threatening
rhythm persists. When we look at
the data from the American Heart Association, Get
with the Guidelines registry, we actually find that there
are differences in the rates of receipt of this
life-saving intervention amongst women
and men. But what I think is
particularly striking is that when you look at
that intersection of race and ethnicity as well as sex,
you see that black women are the least likely
to have an implantable cardioverter defibrillator
in this data set. Now that data is back from
2008, so I was interested in looking at some trends
over time in order to understand if we’ve
seen any differences since some of these
early reports. And the graph that’s there
on the right for you, or I guess on your left,
basically shows that overall we’ve seen some
improvement with — in terms of the
race/ethnicity receipt of implantable cardioverter
defibrillators, but still those in the dotted lines,
the female patients, still lack under the
male patients in terms of receiving this
device therapy. And this data has shown us
some trends over the last four or five years. So we know that early
reports with differences by race and
sex do exist. And ICD use [phonetic sp]
and Medicare recipients between years 1999 and
2005 show — overall women were 75 percent lower
— had lower rates of implantation as
compared to men. And that recent reports do
show this narrowing in the race/ethnicity gap, but
do not show that narrowing for
sex differences. Drilling a little bit
deeper on heart failure, because I think it’s an
interesting case study. If we look over time
with mortality and heart failure, we see
that there’s been improvements
in general. You see the trend actually
declining in terms of deaths per 100,000
of the population. But again, if you start
to look at the rates of black women as compared to
men, black women on average are falling right along
that line of white men, which is a much higher
risk and much higher rate of death as compared to
that of white women. And this graph actually,
I think, shows it a little bit better, because
you can directly compare those two green bars; and you
see that, essentially, when you start to look at
where black women fall in terms of heart
failure/mortality, they’re more equivalent to
that purple bar, over amongst men, so much more
equivalent to the rate of death as white men as
compared to their white female counterparts. If you delve a little bit
differently in terms of incident heart
failures, we were looking at mortality and death
rates before, if we look at incident — we
actually sponsored a very interesting study at
the Heart, Lung, and Blood Institute a while
back when we had this cohort called the CARDIA
cohort that many of you may be familiar with. It actually looks at young
adults and follows young adults from the ages of 18
to 30 over the next 20, 30 years, and actually
found that incident heart failure that occurred at
an age less than 50 years of age was substantially
more common amongst blacks as compared
to whites. And if you look at the
orange line, that’s the black women, and
you actually see that that line crosses over the
black men and continues to rise in
this graph. And we’re interested in
seeing where the follow-up is going to be years later
in this cohort, but we’re concerned about this trend
where 26 out of the 27 cases of incident heart
failure occurred in blacks and the majority of
them in black women. And when you start to look
at other cohorts that we sponsor, such as the
ARIC cohort, we also looked at heart failure incidents
there amongst our — it was a middle-aged cohort,
so a little bit older cohort than the CARDIA
patients or participants. And what we found there
is that blacks had the overall highest incidents
of heart failure in that cohort as well. With black women, if you
categorize them as less than 60 or over the age of
60, black women over the age of 60 with the highest
rates of any other group. And this graph that’s
overlaid here just shows you now after incidents —
if we’d start to look at survival after incident
heart failure in that same cohort, that survival
after incident heart failure hospitalization
was significantly less for black men and women as
compared to the white men and women in
the cohort. We have seen some
narrowing of gaps. We talked about the
implantable cardioverter defibrillator narrowing in
the race/ethnicity gap, but the sex gap
persisting; and if we look at hospital discharges
for heart failure, we’re starting to see some
narrowing in the sex gaps. Over time we are seeing
some level of improvement. And we also are
seeing that overall, cardiovascular disease
mortality for women in that red line is now
approaching that closer to where men have been, and
the decline is starting to catch up, but there’s still
lots of room for improvement. Specifically, there
are lots of challenges in trying to
address these gaps. And I want to take a
couple of examples from the literature to show you
some of these challenges. But one is — is that
we have to gather the necessary information
amongst women in order to really understand if there
are therapeutic treatments that need to differ from
men as compared to women. And I highlight this
one, the glycoprotein IIB/IIIA story is
an interesting one. This is actually —
basically, a blood thinner that’s used in the
setting of a heart attack and, particularly, when
someone’s having an intervention that’s
going to open the artery that’s blocked. And early on in the usage
of glycoprotein IIB/IIIA inhibitors, there was some
concern that the early trial data didn’t show
as much of an effect for women as
for men. And some actually thought
that this was because there weren’t enough women
included in the studies to actually prove
that benefit. But the challenge
clinically was that people were raising the
question and not using the evidence-based care
for women because of this concern that that wasn’t
as effective for them. So several meta-analyses
later, and years later, there was a proven
efficacy of GIIB/IIIA inhibitors for women. And one of the then
raised issues became — well women tend to
bleed more than men, so are we sure
it’s safe? And additional studies
had to be done in order to understand that
yes, indeed, there was an increased rate of
bleeding amongst women as compared to men. But actually that increased
rate of bleeding, almost 46 percent of that, was
accounted for by overdosing. And so understanding
the appropriate dose and targeted dose strategies
actually made this a very effective strategy
for women and educated the clinical
community. So gathering the necessary
information was key in this situation. We also saw in a very
similar fashion, as we start to progress
with more sophisticated cardiovascular devices,
that questions were raised about whether
women should even receive left ventricular
assist devices. And these are devices
that actually support the pumping function of
the heart, and the failing heart, and can often be
used to bridge towards a cardiac transplantation
or even as final therapy for someone with
heart failure. And one of the registries
that is sponsored by the NHLBI decided to take this
question on and really look at whether or not
women should be receiving this device. Again, this is a scenario
where only one device was originally tested
in the early trials. The device was too large
to actually be implanted in many women. So the evidence wasn’t
there in order to be able to really provide
the clinical impetus to use the device. And actually what we’ve
found as they really kind of delved into this issue,
is that in fact there was no difference; and that
women had the same degree of benefit from implanting
this device after the smaller devices were
developed and were able to be used — as showing
that actually gathering that necessary information
basically staved off the unintended consequences
of the low enrollment and inclusion
in the trials. One last example is that
there had also been concerns about whether or
not women had increased adverse events from a left
ventricular assist device. And there was concern that
there was more bleeding; there was more infection;
there was more neurologic dysfunction and device
malfunction in women as compared to men. Again, the investigators
took this on and did find, actually, that there was
one important difference. There were no significant
sex differences or mortality, bleeding, or
infection in this study, but they did find that
women had a shorter time to have first
neurologic event. And that’s something that
actually deserves more clinical attention, as
well as more research attention, and has
prompted more studies in this area in order
to understand what the difference is here in
terms of the implantation of this device in women
as compared to men. And the last example
I’ll show you is around sex-specific differences
for re-sequenization [phonetic sp] therapy. And so this is also
used in heart failure. And there’s a Class 1
indication, which means the highest
recommendation, of using this device in patients
that have a certain finding on their ECG. And one of the questions
that was raised is whether or not actually this would
be beneficial for women at lower ranges of this
duration of their QRS interval in the ECG. Because there were some
early hints at this, but there wasn’t a real
indication from the guidelines perspective to
implant these devices in women. So actually what we
found is that women had a 76 percent relative reduction
in heart failure or death when they actually used
this device at a different interval than where men
actually have this. So the range of
possibility for when a woman would benefit
from this device was much greater than
that for a man. And the findings persisted
in adjusted analyses as well as in a
replication of this study. So again, it’s getting
the necessary information in women and really doing
the sex-related analyses and reporting
that’s important. I will pass over
this last example. It’s again about
implantable cardioverter defibrillators; again,
a question raised as to whether they’re
even effective in women. And similar types of
analyses have showed as equal effectiveness
for men as for women. And that guideline
directed ICD therapy would be associated with reduced
mortality for women and should be implanted
regardless of gender. So there are lots of
barriers to including women in clinical trials,
and also unintended consequences, as we
talked about, of the underrepresentation
of women in our cardiovascular studies. There really has to be
an act of consideration of differences in etiology
and pathophysiology when we talk
about inclusion. And inclusion/exclusion
criteria can sometimes unintentionally
exclude women. Women tend to be of
older age, and have more comorbidities, and
be frailer — frailty is an issue when they
come to presentation for cardiovascular disease, so
enrollment and studies can be more
complicated. But there can also be
an underdiagnoses and under-referral, which
we’ve talked about, pregnancy concerns and
family responsibilities can be of concern
for inclusion. But what we find is that
if you have criteria that focus on symptoms etiology
or pathophysiology that are more common in men or
that were developed based upon symptomatology
studied in men, that women may unintentionally
be excluded. That risk scores that
underestimate risk in women could actually
unintentionally exclude women. And criteria for medical
devices that are based on size can be a problem,
if there’s only one device and the device is
too large in order to be implanted
in women. And upper age limits can
sometimes exclude women who present, for example,
with heart failures, as we talked about,
at an older age. So there’s still a lot
more work to be done to achieve sex equity
and cardiovascular care and outcomes. And we talked about
the need to increase enrollment in
clinical trials. We talked about the need
to sex-stratified — sex-specific analyses and
sex-specific research when questions arise where
evidence-based therapy may not be given to women based
on some of the challenges. And we also highlighted
this intersection of race/ethnicity and sex
where research focused on special populations
with greater risk and disproportionate burden
and poor outcomes is extremely
important. And lastly, I want to
mention that we have to ongoing-ly monitor
our health care system differences in order to
even know if there are differences in treatment
and outcome that persist. So these — the trend data
beyond the initial data is very helpful in
understanding where we need to delve
more deeply. With that I want to
thank you for the time, and I hope that this
was informative. [applause]>>Jennifer Plank:
Thank you Dr. Cook. Our next speaker
is Dr. Gina Brown. She is the Coordinator
of Women and Girls and Microbicides Research in the
NIH Office of AIDS Research. Dr. Brown.>>Gina Brown: I just
want to thank you for inviting me to
come and do this. I think it’s really
important that these issues that are being
discussed today are actually something that’s
brought to the forefront. And, in particular, my
area of work is around issues of women
and girls and HIV. But often we get lost —
meaning women get lost in this HIV discussion,
particularly in this country, when in fact it
is a considerable issue. And so what I’d like to
talk to you about today is an effort to take a look
at one particular issue around the risks of HIV
in women, and give you an example of how NIH is
really looking at what’s the research that
needs to be done. And so what I’m
going to talk about is sexual violence
and HIV risk. I’m going to spend a
little bit of time trying to give you some
definitions, so you know exactly what it is we’re
talking about, and then get to kind of one of the
research efforts that are being put forth and
being led by the National Institutes
of Health. So one in three women
— there’s the — Centers for Disease Control
does something called the National Intimate Partner
and Sexual Violence Survey about every two years,
and then they’ll report it for the two years. So the data that we
have is from 2010, got reported in 2012; coming
up over probably by the end of this year, we’ll have
the data from 2012 that will be reported
in 2014. So this is the most
recent that we have. In the year proceeding the
survey that they’ve done — and they do this as
a telephone interview, mailing out questions
for people to answer, and then having someone sit
down and go through — and it’s considered to be
— this work is considered to be one of the most
representative samples that you can possible have
across the United States. So in the year preceding
the survey, 1.3 million women admitted to being
raped; one in five women admit that — will have
been raped in their lifetime, as compared
to 1 in 71 men. Granted, all of this
is probably quite underreported, but at that
— we can probably all agree that there’s a
considerable imbalance here in terms of who’s
more likely to be raped. One in five black
women and one in five white women — and I think
is important, because if you look at the news, it
looks as though it’s largely a black or African
American women’s issue, and that in fact
is not true. And about one in seven
Hispanic women admit to being raped. If you ask the question
in a given community, you’ll get an answer for
that given community. So if you ask the question
in the African American community, you’ll say, oh
my God, there’s an awful lot of this going on, but
it has to be considered as compared to what. One in four women will
experience some severe enough physical violence
in their lifetime by an intimate partner,
as compared to one in seven men. And one in two women in
their lifetime will have experienced some form
of sexual violence. The rates of intimate
partner violence, sexual violence, childhood
sexual abuse in HIV-infected women is twice the
national rate. And so that’s why we’re
starting to pay attention to this issue with respect
to HIV-infected women. An absolute definition of
what is intimate partner violence, you can get it off
the Centers for Disease Control website, and this is how they
make sure that they’re asking exactly the same
questions of all women. This is pretty much the
same definition that the World Health Organization
also uses, so that we can look at international
data as well and consider it to be quite similar. And what I want to point
out is that it’s — sexual violence isn’t the only
intimate partner violence, it’s also physical
violence. But the other thing that’s
achieved some level of notoriety in the news and
other places has been this issue of control of
reproductive or sexual health, or control of
finances, which also counts as some amount of
intimate partner violence. I am an obstetrician
gynecologist, and I would have women who could not
come to see me on their given appointment days
because their partner wouldn’t allow it. And so it’s something that
affects more than just, you know, the physical
health as a result of the violence, but it also
affects a woman’s physical health as a result of her
ability to get care when necessary — and
it’s, particularly, care when she’s
been attacked. Even the emotional
threats of physical or sexual violence
will have impact. So I always thought, why
are we using this term “intimate partner
violence”? I mean, what does this
mean — particularly, when you’re talking
about sexual violence. And the thing that really
kind of threw me was that when you look at rape,
51 percent of the time, it’s by an intimate partner;
40.8 percent of the time it’s by an acquaintance,
someone this person knows. So almost 92 percent of
the time when a woman is raped, it’s by someone
that she’s known — that she knows. It’s not the stranger as you
walk down the dark corridor in the street quite as
much as it is the stranger who may appear in your
home, in your bedroom, in your social circle,
that is the person who’s the rapist. So there are absolute
definitions of what is rape, and what we’re going
to talk about specifically around these research
issues is what’s considered to be a completed rape
where it’s — penetration. But you can see the long
range which could lead a number of people to not
even quite be sure: the sexual coercion, being
pressured, feeling like you’ve got to go along in
order to either not be — have some other method of
violence, or just kind of trying to make good,
kind of the position that women often find
themselves in. And then there’s the
non-contact sort of things that people may
come across sort of every single day in their daily
work lives that they may or may not admit that’s
been part of what’s considered to be
sexual violence. But, in particular, on
these research issues, I’ll focus in on the
issues of penetration. So the other thing I want
to point out is — again, this is something else that
kind of dropped my jaw. And I had a colleague look
at my slides, you know, just to make sure that
I’ve got no typos and, you know, the sort of thing
— you know what you want to say and you think
that’s what the slide says and you — it may not
exactly say that. So she — we were looking
and she said, “Oh my God, I can’t believe this”
— which is 42 percent of the women who are raped
in the United States — it’s happened by the time
they’re 18 years of age. And 80 percent by the time
they’re 25 years of age, and so it’s a real issue
amongst young women. However, you can’t
discount that if you look at the pie chart, that
if you look at yellow, women over the age of 45 years,
2 percent admit to having been raped — or to being
raped, rather, and then at the time of
their first rape. And 35 to 44 percent,
another large — 5 percent, and then 25
to 34 percent of — at 14 percent. So it’s not something
that doesn’t come across the age range, but it’s
certainly something that young women have
to contend with. So typically the way we’ve
looked at this issue, particularly in the
research world, is we think of it in terms of
— there’s sexual violence and it’s either
direct transmission, so in the course of being
sexually assaulted, a woman will get — is at risk
for getting HIV infection. We’ve built an entire
care system around that. In the emergency rooms
pretty much across the United States, and I have
to say, probably started in New York, is that when
you — if you’ve been sexually assaulted, you go
to the emergency room, and in addition to
getting a pregnancy test and emergency contraception,
you’ll also get an HIV test. And if that rapid test
is negative, you’ll get post-exposure prophylaxis,
or a month’s worth of antiretrovirals to take
to prevent you from getting HIV infections. We’ve built an entire
system pretty much across the United States to kind
of deal with this issue for women who
do seek care. The other issue though we
talk about when we look at research is if you
get this history of sexual violence and kind of look
at women who ultimately have HIV infection, it’s
these issues of being so — put in such a position
of lack of power that your negotiating skills around
sex are compromised, even with your given
consensual partner. Being able to negotiate
condom use is much more difficult in someone
who’s had sexual violence than someone
who’s not. And that’s one of the
things that stands out in terms of looking
at studies. There’s a long-term
follow-up study called the Women’s Interagency HIV
Study that’s done in the U.S. And one of the things
they’ve been able to point out is the women
who are HIV-infected, and it compares
women at risk versus women who become infected, is
their inability to negotiate being able to use condoms or
other methods of prevention, or negotiate when and if
they actually can have sex. And then the third thing that
people tend to look at and have paid attention to — is the
other imbalance — is this issue of HIV-risk behaviors, so that
women who are HIV-infected who’ve — and women who’ve had
sexual violence have a much greater rate of HIV-risk
related behaviors. So, for example, you’ll see
that a woman who’s had sexual violence is — works at this
from a much lower self-esteem, but they also have a larger
number of sexual partners. They’ll engage in more risky
sexual acts, anal sex, sex without protection. And there’s a real lack of a
balance in the power dynamic in the negotiation of sex,
and they tend to be women who also end up at risk
for repeated instances of sexual violence. So when we look at this,
and we started to think about — so what’s
actually missing here? So one of the things, you know,
you can talk about this from the point of view of —
oh, she had too many partners; oh, she couldn’t
negotiate a condom. But the other thing is, growing
along with this, in parallel, has been a greater level
of understanding of what’s happening in the female
genital tract and what the actual physiology is of
the female genital track. And we’ve been trying
to benefit and, sort of, intersect this kind
of information. So one other things
that is missing is, how does having had
sexual violence affect your biological risk for
HIV infection? And we’ve over the past few
years been able to talk about biological risk for HIV from
a much more knowledgeable point of view; so it’s
not just sex, HIV risk, but it’s all the other
things that are going on. There have been a number
of projects that have come on; it’s almost
like a tipping point. So in our Office
of AIDS Research, we sit in the Office
of the Director, we have an advisory council,
and we did a project or did one of our advisory
councils around women’s biological risk, taking
it from the very basic science
point of view. And I’ll talk a bit about some
of that information in a second. Around the same time, the White
House actually did their Women and Girls HIV/AIDS Awareness
Day on sexual violence and HIV in women and girls
— without there being a conversation between
the different groups. We, OAR, with the Social
Science Research Council and UNAIDS, and this
was brought to us by the Social Science
Research Council, started talking about — well, how
do we learn more about what these possible
biological risks are? And so we sponsored a meeting
that brought in people from around the world to talk about
the biology of women’s risk; and it was a meeting that
was held at Green Tree, which is kind of the
Camp David of the UN, to be able to sit and get
people in a think tank for a couple of days to
kind of hammer out — and we had everybody from
mathematical modelers to people who do genital
tract immunology, general immunologists, people
who actually see women in clinical care, and just sort
of the wide range of folks who are very often not in
the same room together, and the social
scientists as well. And what they did was outline
and modeled what you need to do about sexual violence
and biomedical HIV risk; and then came up with —
defining a research agenda, and that was published in
the American Journal of Reproductive Immunology
in a special issue from November of 2012, rather. In 2014, we do the Trans-NIH
plan for HIV-Related Research, which lays out what the plan
is for doing research and what the priorities are
that will taken up by both NIH, but also is used by
external researchers, to determine kind of what’s
available for funding and how funding will be matched. And without us mentioning
it, it was listed as one of the priorities, when
we bring internal and external advisors,
starting in 2014. It was also brought up as a
priority for 2015; and in 2016, it’s actually been fleshed out
to not just looking at sexual violence itself, but also
looking at trauma-informed issues for trying to do
HIV-related research in women. And then there’s a White House
task force on violence against women and HIV, which has been
contributed to by all of the major HHS agencies
running the range from the research agencies, to
the care agencies like HRSA, et cetera, to
justice, education. And within NIH it’s
actually been a very multi-collaborative group
with representation from the Office of Research
on Women’s Health, National Institutes of
Allergy and Infectious Diseases, Child Health
and Human Development, people from National Institutes
of Alcohol and Alcoholism, Institute on Drug Abuse and
— I’m trying to think who else am I leaving
out — Mental Health. And so all of us have come
together to talk about the projects that we’ve been doing
and the projects that are being put forth around these issues
of research on violence against women and HIV for
this White House task force. So I’m going to give you
a quick anatomy lesson. If I cut you in half this
way, if you’re female, this is what female genital
organs would look like. I guess for that — is it
— bottom — so — vagina, you can see the
opening of the cervix, which is the lower
portion of the uterus, and it’s all contiguous. I don’t know if you can see
how sort of the peat goes all the way up through
the lining of the uterus, and then it goes out to
the lining of the tubes which sit over the ovaries. That’s one contiguous path. The beauty of that one
contiguous path is, all of these have their
own immune function, and the design of that immune
function is to keep — this was Lauren talked about, sort of
protection from external things, and one of those things that
it protects you from is HIV. There have been folks
who’ve really devoted their entire
careers to this. Charles Wira from Dartmouth,
and he’s actually a physiologist, does human
genital tract immunology. And what this shows is, as
you go up the genital tract, and this is looking at the
difference in cells from up through the vagina, the cervix,
inside the cervix, uterus, tubes, the cells may
look somewhat different, and then also these, you
have CD4 cells, macrophages, dendritic cells, a variety
of different cells that actually have immune function
and immune protection, that exist along
the genital tract. And these folks have been
studying this and looking at how you can infect
those cells with HIV. This is important, because a
number of other factors have come along, that
it’s not just the cells, but sex and semen also may play
a role in whether the risk — what the genital tract
immune function is like. So it can increase — the
presence of semen can increase inflammation, which puts
you at greater risk for HIV, it may disrupt the barrier,
the epithelial barrier, the cellular barrier
in the genital tract. You can transport viral
particles to the sub mucosal area, as enhanced by both
the presence of semen but also the presence of sperm,
and it also can alter the microbiology of the
female genital tract, which puts you at
greater risk — which can put you at
greater risk for HIV. So this normal female
genital tract microbiota is actually protective,
and when that’s altered, sexually transmitted
infections, the presence of semen, the activity
of sex, it can also put you at greater risk. And the other things people
have talked about are whether or not hormones may —
which change over age, change with the menstrual
cycle, change during pregnancy, some of Wira’s works has shown
that during the menstrual cycle, around the time of ovulation,
a period of time when you don’t want someone to
recognize sperm as foreign, there’s actually a dampening
of the genital tract immune function that allows
sperm to get in, get up through the cervix,
and fertilize an egg. And they can call
this period of time the window of vulnerability. Pregnancy is thought to also
have a similar window of vulnerability; otherwise, you’d
treat the fetus as though it’s a foreign object. Of late there’s been a
discussion about whether or not exogenous hormones,
for example, Depo-Provera, depot medroxyprogesterone
acetate, injectable once
every three months, really very commonly used
method of contraception, particularly in young women,
because it works for three months, whether or not
that can increase the risk. And so this is — these are
some of the issues that are being discussed about what
are the variety of things that can influence the
female genital tract. So our effort is to really
better understand female genital tract biology,
what increases or protects it from HIV infection or
increases the risk. Is — if some has
had sexual assault, is it that current event that
puts you at greater risk, or is there a change in the —
from the tissue damage that may occur with sexual assault that
alters the genital tract immune function, and how does that
affect your future HIV risk, even if you’re having sex
with somebody where it’s consensual, but that
person may be infected? And then also, what we’re also
working on his how we analyze this integrated behavioral
and biological risk for HIV in someone who’s had sexual —
experienced sexual violence. So there have been a number
of things that have done — I’m almost done. One is, the Centers for
AIDS Research are these multidisciplinary centers that
are funded through the National Institutes of Health that are
external centers across the country, and they also do some
of their work internationally. It’s an opportunity to take
advantage of NIH — or, HIV funding that occurs,
and also the wide range of researchers, from basic
science, to clinical, to behavioral scientists. So we funded three supplements,
National Institutes of Allergy and Infectious Disease
funded another supplement, to really start to give us
some pilot data around this data, and the four
supplements are listed. The immune effects of sexual
violence and associated HIV risk; this is important,
because this study takes women in a sexual violence clinic,
looks at what their immune factors are at the time
of presentation, and then follows them over a number
of months to see how these immune factors may
in fact change. It also collects samples of
tissue and cells so that you can actually look at working at
infecting those cells with HIV and see what that
difference is over time. And then you can see the
others that are listed there as well. One of these studies is
being funded in Africa. There’s also some studies
looking at just the general mucosal immune environment
and seeing how that’s altered and what that can mean
at the cellular level for the ability to
infect with HIV. And then there are a number
of studies — I mean, a number of meetings
that are being confirmed. One is, that will
be done by NIH, is to look specifically about
the ethics in doing research in people who are at risk,
because it really tells us, we have to look at looking
at women under the age of 18 to really understand
these risks well, and there are
tremendous ethics with being able to do
research in this younger-than-age-18 population. And then, most of the research
studies that are being conducted around HIV, HIV, and women’s
risks are now including questions that will ask
about sexual violence. And so if you want
additional information about this, there are a
number of places to go. On the NIH website, you
can get the trans-NIH plan, which will flesh out HIV
and sexual violence-related research and what
the issues are. You can look for the Centers
for AIDS Research supplements, and then there are a number
of RFAs that will come up over the next couple of
years to talk about some of these things. And then the Centers for
Disease Control has a tremendous amount of
demographic information about sexual violence,
and then the intersection of sexual violence and partner
violence and HIV risk. Thank you. [applause]>>Cerise Elliott:
Good afternoon, my name is Cerise
Elliott, and I’m a member of the Women of Color
committee, and I’m here to introduce Dr.
Tamara Harris. She is the Chief of the
Interdisciplinary Studies of Aging section in the
Laboratory of Epidemiology and Population Science
at the National Institute on Aging, and she’s
going to talk about disability outcomes, data
from health and aging and body composition.>>Tamara Harris: Thank you
very much to the organizers for inviting me to speak, and
for that nice introduction. So, I’m going to spend a
little bit of time today — let’s see if I can get this
so that it actually picks up my voice — I’m going
to spend a little bit of time talking about
disability, and then I’m going to show some
data from a study that I’ve designed, but
that our group has carried out for the past 15 years,
and that is available for other people to
collaborate with us, and also to
analyze the data. So please get in touch
with me if you have any interest at all in analyzing
data from this study. So, let’s see if I can
figure out how to do this. Top button?>>Female Speaker:
No, left-right.>>Tamara Harris: Left-right? Okay. Okay, there we go. Now, people tend to think of
older people as one large group of individuals, but in fact,
from the point of view of the National Institute
on Aging, we see the older population as very
heterogeneous, and at least dividing into three
different groups, depending on health trajectories
over the lifetime. So that is, that in terms
of things that happen, the genes, health behaviors,
education, and life events, tend to separate the cohorts,
and this is the period of time that was discussed earlier
from the ARIC study and the Cardus
[phonetic sp] study, so that by the time
that people get into old age, they are either
frail, they’re usual aging, or they might be
considered healthy agers. And the thing that’s
interesting about this, is that people contribute
from these groups differentially in terms
of health outcomes. The study that I’m going to
discuss in terms of this paradigm is the health, aging,
and body composition study, and this is a hypothetical
trajectory that we think many people follow in relationship
to weight, illness, and risk of disability. That is, that people are going
along in their usual state, and then they’ll have some
sort of catastrophic incident, health incident, like a
hip fracture, pneumonia, congestive heart
failure, and basically, most people experience an
incomplete recovery from that episode that leads them then to
have a lower level of energy, a lower level of function, and
eventually they pass through the threshold of what we call
functional limitation — and I’ll define that in
just a minute — until they come through the
threshold for disability. This diagram is originally
created for people who had AIDS in the period when
we had no treatment for AIDS, and in fact has
also been adapted for malnutrition as well. Now, the study itself was
considered — we started this study because — with the idea
that change in body composition is a common pathway by which
weight-related health conditions contribute to risk of
disease and disability, and we started this study
15 years ago with a group of 3,075 men and women
who were initially age 70 to 79, and we had
four race/ethnicity groups. The study was 42 percent
African-American, and we did that because we
wanted to get four separate estimates of what those
changes look like in terms of the subpopulations. So we began the study
in 1997-’98 in Memphis, Tennessee and Pittsburgh,
Pennsylvania, and at baseline, everyone who came into the study
had to tell us at least twice that they were able to walk
a quarter of a mile without difficulty and up 10
steps without difficulty, and that was the primary
outcome of the study, functional limitation, which
was incident difficulty walking a quarter of a mile
or up steps persisting for six months. This is the schematic
that we used, and we felt that we wanted to
include a lot of information about past history. Most of that information we
got through health history, looking at socio-demographic
status, psychological factors, and weight history. And then we had a core
examination where we brought people to a clinic, and that
core examination consisted of body composition studies,
walking endurance studies, tests of strength, physical
performance where people would do things like have timed
walks, and drawing of blood. But the most important
thing for this study was really to have information
about weight-related health conditions, and we
tried to get information about — not only about
clinical conditions, but about subclinical
conditions as well, and then we had
the outcomes of the study. These are some of the
characteristics of the study, and you can see that among
the body mass index, the men, black men and white men had
about the same range of BMI, whereas the African-American
women were much heavier, a BMI of about 30
as the mean. When we looked at
percent fat, though, you can see that the percent
fat was about equal because of the different distribution
in terms of subcutaneous versus visceral fat in
the subpopulations. Lean mass was slightly higher
among African-Americans. Days of hospitalization were
higher among African-Americans. Even though everyone told us
that they were at the same functioning level, we
could see signs that the African-Americans
were at increased risk. Pack-years were lower in the men
and about equal in the women, and the proportion that were
hypertensive was much higher in the black population. I’m going to skip this
in the interest of time, and I’ll be happy
to come back to it. So over the 13
years of the study, if we look at the percent
who have incident functional limitation — these figures
were really astounding when I ran them last
night — the white men, 67 percent of the white men
had persistent disability, problem walking a quarter
of a mile or up 10 steps, black men, 73 percent. White women, 72 percent, and
the black women, 80 percent. And I think that this really —
it is very consistent with all of the data that’s been
shown at this meeting for the differences by
race and gender. So what about risk factors
from functional limitation? These are just the risk
factors for the black women, and what I thought was
important to point out here, I’m going to go through this
and sort of read through this, it’s using the same paradigm
that we used for the study. So first, in the
area of past history, measures from the
core examination, from the mediators of disease,
and then from the outcome. So, the first thing is
that for past history, history is very important. So, education or family
income, all these risk factors were significant at the
point — less than .001 level when looked at
in the univariate sense, and all of them maintained
significance when put in a multivariate model. So number of hospitalizations
prior to coming into this study was a risk factor
for disability over the next 13 years,
pack-years of smoking, and self-rated health. In the area of core
examination, body mass index was a risk factor for
mobility disability. The penetration of fat
into the muscle of the leg, which I’m not going to have time
to talk about but is really a new risk factor that’s
emerged from this study, was a risk factor for
mobility disability, as was lower grip strength. But if we look at the diseases
that were present at baseline that predicted the future onset
of disability, depression, scoring high on a screening
test for symptoms of depression, both
osteoarthritis of the knees, having been told by a doctor or
other health provider of having osteoarthritis, and then
actually saying that, in addition, that
people had pain due to osteoarthritis of the
knees, both were predictive of future mobility
disability. Lower extremity
arterial disease, measured by blood pressure
in the arm and the feet, coronary artery disease history,
and then three other factors which we had thought were
mediators that fed back to body composition, which
were exercise, cystatin, which is a marker of renal
disease, and fasting glucose, a marker of diabetes. The fasting glucose actually
extends beyond diabetes, because diabetes itself was not
a risk factor for disability, whereas fasting glucose was
a factor for disability. Now, what happens if you
include functional limitation? In aging, there’s a lot of
controversy about whether diseases are actually the thing
that predicts what’s going to happen to you in the
future, or whether it’s the presence of disability,
and it doesn’t really matter what diseases you have. So, I went on to look at
walking speed and reported ease of walking
distance, and you can see that several risk
factors dropped out, but the number of
hospitalizations prior to coming into the study,
pack-years of smoking, lower extremity arterial
disease, renal disease, a heavy body mass index,
osteoarthritis of the knees, and symptoms of depression
all continued to contribute to risk of mobility
disability. And the thing that’s
really striking, and it’s striking in
terms of the entire study, is the fact that past
history contributes so strongly to what
happens in old age. That is, that we’re not
really discovering for the most part new risk
factors, but what we’re discovering is that
people have the product of what’s happened,
accumulated over their lifetime, so if we could
make interventions earlier in the lifetime on the diseases
that are common in old age, but much earlier, we have a
chance to actually reverse the mobility disability and to
break the cycle of lack of independence in old age. So, I just wanted to close by
thanking you very much for your attention and to welcome any
questions you might have. [applause]>>Tamara Harris: My
email is in the — among the common directory, and we
have lots and lots of data, and if anybody’s interested,
I really would like you to get in touch with me.>>Cerise Elliott: Thank you. Does anyone have a
question for Dr. Harris?>>Male Speaker: Just one
quick question about the model, did you include
caregiving in the –>>Tamara Harris: Well, it
turns out that caregiving was a factor, but it wasn’t
— it was of moderate significance, but it wasn’t
of major significance. Actually, what was
important was working. It turns out that people
who are able to continue to work in old age, and we
found that if you look at the proportion of
African-Americans who are working, it’s
substantially higher, which we think is really
related to income. Yeah.>>Cerise Elliott:
Thank you, Dr. Harris. Our next presenter is
Dr. Tiffany Powell-Wiley. She’s Assistant Clinical
Investigator for the Social Determinants to
Obesity and Cardiovascular Risk at the National Heart,
Lung, and Blood Institute, and she will be
discussing neighborhood environment as a social
determinant of obesity and cardiovascular risk.>>Tiffany Powell-Wiley: So,
I’ll be talking about obesity as a cardiovascular risk
factor in particular, and highlight some data that
suggests it’s particularly pertinent for women of color,
but also talk about neighborhood environment
in particular as it relates to the
development of obesity. So, as most of us know,
obesity has become an epidemic in this country, but what
isn’t discussed as much are the disparities in
obesity prevalence in the U.S. population
in particular. Data from the National Health
and Nutrition Examination Survey shows that women of
color, particularly African-American women and
Mexican-American women, have the highest rates of
obesity as defined by a body mass index at or above 30. And the data shows that while
the obesity prevalence is leveling off, or appears to
be leveling off in the U.S. population, most recent
estimates show that almost 60 percent of
African-American women in particular have a BMI
at or above 30. So, there’s definitely
controversy as to what constitutes cardiovascular
risk, particularly in relation to a BMI
between 30 and 35, but what we do know is that
grade III obesity, or the highest level of
obesity as measured with a body mass index at or above
40, is most prevalent amongst women of color, particularly
African-American women. Data from NHANE shows
us that 16 percent of African-American women have
a BMI over 40, as compared to 8 percent of all race
and ethnic groups. And more and more studies are
showing us that the highest rates of BMI in particular
put individuals at risk for cardiovascular events, but even
at risk for all-cause mortality. And so data from the Black
Women’s Health Study, which is a population-based
cohort of women across the United States, shows that when
we look at the relationship between body mass index
and all-cause mortality, we see a gradient, particularly
for those BMIs above 35 and well into — well above
40, where those highest BMIs are associated with higher
rates of mortality. And when we look — when those
individuals look specifically at cardiovascular death in
relation to body mass index, when compared to
normal-weight individuals, women who had BMIs above
30 and as well as above 35 to 40 were at highest risk
of cardiovascular death. Data from the National
Cardiovascular Registry, or ACTION Get with the
Guidelines Registry, also suggests a relationship
between class III obesity or grade III obesity and
mortality, particularly 30-day mortality after myocardial infarction. When we look specifically at
those with class III obesity as compared to those
with class I obesity, those with class III obesity
had the highest risk of 30-day mortality, despite the
fact that they presented at a younger age with acute
myocardial infarction or heart attack, as well as the
fact that they presented with a lower burden of
cardiovascular disease. And so the question
becomes, what really leads to this higher risk
of death at 30 days, despite these discrepancies
in disease burden? And so we understand that
obesity is a cardiovascular risk factor and contributes not only
to morbidity related to the weight itself, but also in
increased lifetime risk of diabetes and subsequent
cardiovascular morbidity and mortality, and it
contributes to increasing costs related to
cardiovascular disease, and the disproportionate burden
of cardiovascular disease, particularly among
African-American women. And this relationship between
obesity and cardiovascular risk suggests a need to identify
the potential causes that lead to higher rates of obesity
amongst African-American women, and in the effort
that we can understand how these causes may help
us develop effective interventions for a
specific population. And so, we’ve looked
at — in our review, we’ve looked at what we deem
to be the potential reasons or causes behind the genesis
of obesity in African-American women, and we’ve divided
those up into specific — three specific areas,
the first being biology, the second being behavior,
and the third being, in particular, what we term
social determinants of health, either psychosocial or
environmental factors that may play a role in the
development of obesity, particularly for
African-American women. And what I’d like to focus on
now is just really some of the data that we have developed or
that we’ve analyzed in relation to neighborhood environment and
how that relates to obesity. And while we haven’t been
able to look specifically at African-American women in
the cohorts that we have looked at thus far, we do — we have
been able to really tease out a relationship between
neighborhood environment and weight change over time in
a population-based cohort. And so one way of thinking about
neighborhood environment as — one way of measuring
neighborhood environment as a social determinant
of health is measuring a neighborhood’s
environment as it relates to socioeconomic status on
the neighborhood level, and this can be thought
of as an objective measure of environment, and
really as a proxy of the types of resources that are available
in that neighborhood, whether it be food sources,
areas for physical activity, or potentially areas
for safety within that neighborhood
environment. One method that has been used
and what we used in measuring neighborhood socioeconomic
status is the development of what’s called the
neighborhood deprivation index, and this is based
on U.S. Census-level data, particularly at the
block group level, which is the smallest
neighborhood unit provided by the U.S. Census,
that — and these data allow us to
aggregate specific measures from the U.S.
Census into a measure of deprivation or a
measure of socioeconomic status at the
neighborhood level. And what we have done is used
statistical analyses to tease out a measure of deprivation
based on six different areas, particularly education at
the neighborhood level, the level of income and poverty,
residential stability or whether people — how long people have
lived in that neighborhood, housing conditions
of that neighborhood, the racial composition
of the neighborhood, and employment and occupation
status of those within the neighborhood environment. And while there have been
several studies that suggest a cross-sectional relationship
between neighborhood environment and obesity, or
prevalent obesity, there has been really limited
data on how neighborhood environment relates to weight
shape — weight change over time, with mixed results
in prior studies looking at this issue. And so to utilize longitudinal
data available on environment and cardiovascular risk,
we’ve looked particularly at a cohort called the
Dallas Heart Study, which is a cohort based
in Dallas County, Texas. It was designed as a cohort
to look at specific social and biologic factors that
relate to the development of cardiovascular disease. This was a probability-based
sample originally of the Dallas County
population, and at baseline, individuals were
adults aged 18 to 65. The cohort was designed
to be ethnically diverse, where half were
African-American, and so there were — it
was designed to look at race-specific outcomes;
however, race-sex-specific outcomes were limited
due to the power. And sample weights were
calculated for this cohort so that inferences could be
made to the population of Dallas County
as a whole. And so one of our
first projects looking at the relationship between
neighborhood deprivation or neighborhood
socioeconomic status and weight change was
looking at individuals who had been living in their
— the same environment, the same neighborhood, over
the entire study portion, so over a seven-year period. And what we first determined
was that there was actually differences based on how long
individuals lived in their neighborhood, there was an
interaction between time in neighborhood and the
neighborhood deprivation index. So there were differences
depending on how long somebody lived in their neighborhood,
there were differences in the relationship between
deprivation and weight change. And so what we looked at were
several different models, but as you can see, we have the
different model types on the X-axis, and the change in
weight over the seven-year period for every one-unit
change in deprivation on the Y-axis. And for those who had lived in
their neighborhood the longest, meaning they had lived in
their neighborhood over the median time of 11 years,
they had significant weight change or a significant
increase in weight relative to neighborhood
deprivation. So the higher the
neighborhood deprivation, or the lower the
socioeconomic status, the more weight they
gained over time. And this was independent —
when we looked at adjusted models, first we adjusted
for confounders such as age, sex, race, socioeconomic
status, and smoking, but we also looked at behaviors
like physical activity as potential mediators, and
still saw this relationship, as evidenced
by model number 2. We then looked at the
relationship when we adjusted for what we termed perceptions
of one’s neighborhood environment, meaning how they
perceived the resources — what they perceived to be
available resources in their neighborhood, such as resources
for physical activity, resources for healthy foods, and
again we saw that there was — there continued to be this
relationship between higher deprivation — or lower
socioeconomic status, and higher weight gain over
the seven-year period for participants in this study. And so this led us to look more
closely at this relationship to really tease out if
— for instance, if we saw a change in the
neighborhood environment over time, for instance, if
somebody moved over time, would that change in
neighborhood environment also affect weight change? And our hope was to look at both
that change in neighborhood environment in relation to
weight change and incident diabetes in a protocol
here at NHLBI. And so, just to give a sense
of what we were looking at conceptually, we were trying
to not only look at this relationship independent of
confounders, but also to, again, tease out whether these
potential psychosocial mediators played a role in
this relationship, or could there be some
explanatory — could some of the relationship
between change in neighborhood
environment due to moving and weight change, could
that be explained by these psychosocial factors related
to perceptions of one’s neighborhood environment? And we also wanted to, again,
see whether length of time in neighborhood moderated or
modified the relationship between change in neighborhood
socioeconomic environment and weight change, again,
providing a little more evidence of a potential
relationship between socioeconomic
environment and weight. And so when we think of looking
at the change in neighborhood socioeconomic environment
related to moving, we understood that because
we were looking at a natural experiment, what people do over
a specific amount of time, over a seven-year period,
we couldn’t — we weren’t necessarily randomizing people
to moving versus not moving, and so we had to in some
way control for the fact that moving in this case
was not a random event, that there had to be some way
to correct for the fact that moving had to do with
other factors besides just the need to move for
those individuals. And so, we used what’s called a
Heckman’s correction factor to account for this non-random
chance of moving over time, and specifically, the non-random
chance of moving to an area of higher deprivation or
lower socioeconomic status. And this correction factor
took into account specific characteristics of the
participants that may dictate why they would move: for
instance, age, sex, race, their education level
or income level, whether they owned a home
or whether they rented a home, and also whether
they were married or single. And these have been looked at,
particularly in econometrics literature, but had not been
used as much in the health literature, but if one
is familiar with a propensity score in the
cardiovascular literature, for instance, or clinical
trial literature, this Heckman’s correction
factor can be thought of in a similar manner. And so again, we wanted to see
this relationship between change in neighborhood deprivation,
this time related to moving to an area of higher deprivation,
and weight change over time. And again, the X-axis shows
our — the various models that we looked at, first
adjusting for the confounders, and then adjusting
for perceptions of neighborhood environment
as potential mediators. And what we saw for the
overall population is that for those who move to an
area of higher deprivation, or who moved to an area of
lower socioeconomic status, as shown under
“overall” on the graph, those who moved to
an area of greater deprivation had greater
weight change over time. And again, wanting to tease out
this relationship about length of residence, to see if that
modified the relationship, we stratified by median time
in neighborhood for those who had, again, moved to
a new neighborhood at that time period. And for those who had lived
in their neighborhood the longest period of time,
we saw — we continued to see the relationship where
higher — moving to an area of greater deprivation was
associated with greater weight gain over the
seven-year time period. And so these data really
pointed to neighborhood socioeconomic status as
a specific risk factor or — particularly
related to weight change in this population. And so our hope is that
we’ll continue to tease out what may explain
this relationship. We’re looking at alternative
measures of neighborhood socioeconomic status,
particularly measures such as home value, which can be
measured more easily than census-level data. It’s measured more frequently,
and so you can look at a more dynamic change over time,
and we would like to look at that in relation to weight
change in the population. But we also plan to look at
built environment measures in this — in the context
of how they may explain some of this relationship,
particularly things such as neighborhood
walkability and access to food stores in these
neighborhoods. We’re also looking at
additional outcomes, particularly how neighborhood
socioeconomic environment relates to the development
of diabetes in this cohort, as well as the development of
other cardio metabolic risk factors such as hypertension
and hyperlipidemia. But more importantly in relation
to how we might be able to improve health in communities,
we hope to incorporate this type of neighborhood environment
data into the development of community-based interventions
focused on improving physical activity and dietary intake,
and reducing obesity and cardiovascular risk. And so, with our
current studies, we at least have some
evidence that cumulative exposure to neighborhoods
may impact weight change, and right now we have looked
at psychosocial or behavioral factors as potential
mediators, but have not found one that really changes
the relationship thus far. But these data suggest that
these at-risk neighborhoods really may serve as targets for
community-level interventions for reducing obesity and
improving cardiovascular risk. So, I’d like to acknowledge
those who worked here and elsewhere on these
projects, and I’d be happy to take questions. [applause]>>Cerise Elliott: Dr. Wood?>>Lauren Wood: Were you able
to also assess whether or not participants in the study had
immediate family members or close friends that
lived within the same neighborhood and potentially
account for that? Because there have been some
other studies that clearly, you know, if you hang
out with your friends –>>Tiffany Powell-Wiley:
[affirmative]>>Lauren Wood: — and your
friends have certain types of eating patterns or
caloric intakes, then you eventually
adapt to that, so one of the things that I
was kind of curious about is whether or not there was
— or if it’s even able to do, an analysis of, are
there other family members or friends that are living
within the same neighborhood, if you see a difference
maybe along that factor as well as [unintelligible]
neighborhood or [unintelligible]?>>Tiffany Powell-Wiley:
No, that’s a great point. I think you’re
alluding to some of the data regarding
social networks in relation to the
development of obesity. We — that data was not
available at baseline on who was living in the household,
or data on spouses, or that type of thing. We do have data on
spouses at follow-up, so that may be something
that we could look at. But in our multi-level
modeling, we do attempt to control for clustering
of individuals in neighborhoods, so
we do try to take that into account.>>Male Speaker: I was
thinking along the same lines as Lauren. Time in neighborhood
and increased number of invitations to parties –>>Tiffany Powell-Wiley:
Exactly.>>Male Speaker: —
and things like that.>>Tiffany Powell-Wiley: Right. No, exactly.
[affirmative]>>Male Speaker:
So, same comment. You know, opportunities
for social mobility, information about
employment, you know, ways to kind of get out,
if that’s the intent, [unintelligible] networks. I really like your measure of
the neighborhood environment. Incorporating some part of
the network of it, you know, may provide — explain
some of the variance.>>Tiffany Powell-Wiley: Right. And we do — in some ways, we
try to get at that measure of social cohesion with our
psychosocial measure, because it asks about the
cohesion of the environment, but we didn’t see any
variability when we adjusted for that, so that — but
that, in some ways, gets at how close-knit
individuals feel within that — within their environment.>>Cerise Elliott: Our final
speaker is Salman Tajuddin. He’s a visiting post-doc fellow
in the health disparities research section at the
National Institute on Aging, and he’ll be talking about
genome-wide analysis of carotid intima-media thickness
among African-Americans. Thank you very much.>>Salman Tajuddin:
Thank you so much, and I would like to thank first
the organizers for giving me the opportunity to
discuss what we are doing at the health disparities
research section up in the Baltimore NIA. My work focuses on genomics of
the [unintelligible] study, and today I would like to
talk about the carotid intima-media thickness and
its genomic determinants. Yeah, we have seen
this slide before when Dr. Cook was presenting. I just would like to reiterate
and emphasize that there is a disparity in terms of
cardiovascular mortality, mainly heart disease, which
contributes more than half of cardiovascular disease
mortality, and as you can see, both black men and women
experience high mortality and if you see, African-American
women, they have even a greater risk of mortalities
than Hispanic men. And also, mortality from
cardiovascular disease is improving over time, this
disparity is still persistent. So, here is the data
from 2008, and in 1995, the mortality between
African-Americans — the ratio between African-Americans and
whites was 1.5 times higher, and this mortality was the
same when you look back to data to the 1950s, so
over most of the century, still there is a
gap in mortality, and what we believe is genomics
may have a contribution in addressing and reducing
the gap in this one. Another data I’d like
to emphasize is also, there is this gap, it increases
while individuals are aging, and that’s true for both
black men and women. As it has been
discussed previously, cardiovascular diseases, it’s
multifactorial in origin, there are multiple factors. The main one is aging, it
contributes [unintelligible] correlative increase to
cardiovascular disease, and there are also
socioeconomic status factors, cigarette smoking,
high cholesterol, elevated blood
pressure, obesity, as has been discussed earlier,
diet, physical inactivity, and diabetes and
high glucose level. And the common denominator here
is that these risk factors are common among African-Americans. For example, in our study,
more than half of the study participants are
current smokers. Physical inactivity and
obesity, as has been said, is common among
African-Americans. And another common factor
for this is all these factors contribute to
cardiovascular disease through arteriosclerosis picked
up via endothelial damage and increased inflammation. So, looking at the
pathophysiology of arteriosclerosis is
that inflammation is prominent and it’s
participating in all stages of arteriosclerosis buildup. Looking at acute phase
reactants such as CRP and phospholipase-A2, they are
highly associated with arteriosclerosis development
and plaque buildup. And looking at data
among women only, they experience a
higher level of CRP, and that data is — it’s
more significant, actually, among African-American women,
and our previous data from our previous study, we have
shown that CRP is highly associated with oxidative
stress in both white and African-American women. So, what is the best
indicator, then, for arteriosclerosis
development? We know that
intima-media thickness, which is an arterial wall
thickness through lipid buildup in the arterial system, it
results in wall thickening and blood flow obstruction,
and this arterial-wall thickening, it occurs in
the whole vascular system. What we know about carotid
intimia-media thickness is it is a measure of
subclinical arteriosclerosis. And there has been previous
studies which showed that IMT actually strongly predicts
future cardiovascular events. And studies among
African-American population from family studies have
shown that IMT is heritable, and that there are genetic
risk factors which contribute for IMT development. And in this particular
study, which is from 2002, nearly 40 percent of variation
in intima-media thickness among African-Americans could be
explained by genetic factors. So our aim was to identify
genetic susceptibility factors that could be associated
with carotid IMT among African-Americans, and we hoped
that by identifying these genetic susceptibility factors,
we would be able to fine-tune the risk prediction for
cardiovascular disease by adding genetic markers,
so that we would improve early detection of
coronary heart disease. So this — the data which
comes from the Healthy Aging in Neighborhoods of Diversity
across the Life Span study, it is an interdisciplinary
community-based prospective epidemiologic study
conducted in Baltimore, in the different
neighborhoods of Baltimore. It examines the influence of
race and socioeconomic status on age-related
health disparities. Initially, it recruited 7,700
whites and African-Americans in the Baltimore area. What is unique about this
study is that by going out to the neighborhoods and
sending out these mobile medical research vehicles,
would like to enable recruitment and retention of
non-traditional research participants into age-related
clinical research. My focus, and the data
that I’m going to show you, is on the genetic
component of the study. Also, the study has behavioral
and social domains. And we focused on only
genotyping African-Americans for the genetic
component of the study, and genotyping was done using
the Illumina 1 million SNP genotyping platform, and
we increased the density of the SNPs by imputing using
the 1,000 genomes project, the multi-ethnic panel. And for this particular study,
we had around 600 individuals with carotid IMP data and
carotid IMP was measured using ultrasound on the
left carotid artery, the far wall, and we
took five data points and averaged them. This is a summary of the
association analysis and quality control we
applied, so we would like to estimate the
point effect for the SNP, and we’d like to adjust
for potential confounders such as age and sex,
and as you know, the African ancestry
genome is diverse, so — and there is known
population substructure, and we’d like to adjust for
population substructure by including the first
intrinsic components. And we applied the standard
quality-control criteria and defined genome-wide
significance level like that. And what you see here is a
quantile/quantile plot where we would like to see whether
the P values from our result, whether it was inflated or not. And under normal circumstances
and under normally distributed data, what you’d like to
see is the expected P value [unintelligible] should
be on the red line. And what you see here is at the
tail of this curve is that you see a departure from the normal
distribution of where there is a strong evidence of
association with carotid intima-media thickness in
our study population. So this, over all the study, is
a Manhattan plot where the P values are plotted for each
chromosome colored differently on the X-axis, and the
log P value on the Y-axis. Each bar represents each
autosomal chromosomes, and this red line is the
genomic significance P value at 5 x 10 to the minus 8,
while the dark line represents the suggestible
affirmative association. As you can see, there are
some hits on this plot, one in chromosome 1,
another one on chromosome 2, and a third one on chromosome 4,
and I would like to zoom in and show you the local
individual structure in the genes that I tagged
by these [inaudible] The first one is the locus
on MMAC for 51 locus. Here, what you see on the X-axis
is the genes that are located on that particular region
of Chromosome X, and the Y-axis is similar to
the previous slide I showed you, where it shows you
that P values in log. And the most significant P
value is the one in purple, and the different colors shows
you the local individual structure, and the SNPs that
are closely related to the lead SNP, which
is shown in purple. The red one means that these
three SNPs are closely related to the lead SNP, the
yellow — I mean, the orange one between point 6
and point 8, the green ones, it goes on like that. So what we see here is that this
SNP is tagging that MMA1 gene, which is a transporter
gene which is involved in the transportation of
cobalamin from the cytosome to the mitochondria, and
we know that cobalamin is important in various
metabolisms, including methionine and
[unintelligible], and this is the gene that
is — we found a hit, and what it does is it
transports cobalamin into the mitochondrion end. So if this SNP results in a
dysfunction on this gene, what it results is there
will be lower production of methylcobalamin; that results
in an increased buildup of homocysteine in the cell,
and we know that increased homocysteine levels is
related to endothelial damage and arteriogenesis. So these lack a support
from the literature. Another hit we found is in
the [unintelligible] locus, where the SNP is located
in the intergenic region, and it is strongly
associated with IMT. The third region, what we
have found is this SNP, it’s also intergenic, and it’s
located in the CAMTA1 gene, which is a transcription factor,
and previous studies in a European ancestry
population have shown that SNPs in this gene are associated
with plasma fatty acid levels. And we’re hypothesizing
that it may be related to arteriogenesis and
arterial development by altering plasma
fatty acid levels, but that needs
functional validation. And I would like to show you
a stratified analysis by sex and talk [unintelligible]
is the results for women and the lower one for men. And as you can see here, there
are some SNPs in chromosome 4 and chromosome 9 which are
strongly significantly associated with carotid
IMT in the women, and these SNPs are quite
different from the one that you see in Chromosome
— for men, which are located in
Chromosome 5 and 4. So this suggests that maybe men
and women have a different risk factor for carotid
intima-media thickness, and maybe subsequent
cardiovascular disease development. So, in conclusion, so these
are preliminary studies, just one study. We need to replicate it and
validate it and, you know, as you know, [unintelligible]
in one study. But the preliminary
conclusion for this study is that MMA1 and MMA and
CAMTA1 SNPs could be a potential
susceptibility marker for arteriosclerosis development
in African-Americans and may provide new
mechanics to give insight to subclinical
arteriosclerosis. There are ongoing
works for this project. We approached other
African-American cohorts from the COGENT [phonetic sp]
and CHARGE consortia, so we’re not pulling data
from this cohort into genomic meta-analysis, and also we have
genotyped our cohort SNPs that are located on the exomes
of — from the genome, and would like to identify
an association with this predictively — functional
variance and validate their functional effect on IMT. We’d also like to explore
general environment interaction, particularly age, smoking,
and socioeconomic status, and we have another ongoing
project where we would like to see [unintelligible]
binding variance and hypertension
among women. I would like to thank
my boss, Dr. Evans, for giving me this
opportunity, and also my lab mates and collaborators here
in the research program, and my work has benefited
discussion with them. Thank you so much, and if
you have any questions, I’m happy to answer. [applause]>>Male Speaker: How far
along are you with this gene environment analysis?>>Salman Tajuddin: Yeah.>>Male Speaker:
[unintelligible] the other environmental –>>Salman Tajuddin: Yeah,
that’s a really intense thing, and actually we would really
like to tease out that association, but we would like
to first identify, validate, and replicate the SNPs before
proceeding to the gene environment analysis. So, once we have the
meta-analysis from the genome-wide studies, the
next thing we’d like to do is the gene
environment interaction.>>Female Speaker: So, you said
you were [inaudible]. How about finding
[unintelligible]?>>Salman Tajuddin: Yeah,
that’s a good thing, yeah. We contacted them, the
[unintelligible] services part and the COGENT consortia, so I
contacted, actually, Dr. Harris, and they don’t have IMT
data, unfortunately. Yeah. There aren’t really a
lot of studies, actually, I would like to say that,
among African-Americans and so far, we have contacted,
I think, more than 10-12, but like five of them
had only IMT data, and only immediate numbers,
so that really affects sort of the power of the study,
I should stay that.>>Female Speaker: Did you
see any difference in the association of the SNPs and
did you differentiate by age, specifically targeting the lower
end of individuals who if, you know, if you postulate that
this is possibly a genetic risk factor, to see whether or not
there was different expression in certain genes in a younger
age group, because then, those are individuals where we
can somehow target some kind of therapeutic or [unintelligible]
tremendous bang for your buck in terms of delaying that
co-morbidity in the — you know [unintelligible]
working years, stuff like that. Did you see any
[unintelligible] you know, if you had intima-media
thickness at a certain age, and then differentiated
that out, maybe at the far
ends of the spectrum.>>Salman Tajuddin:
Yeah, looking at the extreme phenotypes. That is an
interesting point. We haven’t yet,
mainly because yeah, it’s true that arteriosclerotic
development is a lifetime process, so looking at the
different extremes of age, it may give you a good picture
of the gene age association, actually, and you make get
some interesting hits. But our sample
size is smaller, so stratifying may not
give you a good power, so I think — and that goes
with that gene environment interaction analysis, so
what we’d like to do is get a bigger sample size,
and have a good power, and then do a stratified
analysis with age, sex, and socioeconomic
status, actually. Thank you. [applause]

Leave a Reply

Your email address will not be published. Required fields are marked *