“Pulmonary Hypertension Assessment – Bench to Bedside” by Dr. Jeffrey Fineman for OPENPediatrics


Welcome to World Shared Practice Forum. I’m Dr. Jeff Burns, Chief of Critical Care
at Boston Children’s Hospital and Harvard Medical School. We’re very pleased to have with us today
Dr. Jeffrey Fineman. Dr. Fineman is Professor of Pediatrics at
the University of California, San Francisco, and Benioff Children’s Hospital. He is also the Vice Chair of Pediatrics, the
Director of the Critical Care Program, and the Director of the Pulmonary Hypertension
Service at Benioff Children’s Hospital in San Francisco. He is also a faculty member at the Cardiovascular
Research Institute at the University of California, San Francisco, where he has had a lab that’s
been continuously funded over the last several decades by the NIH (National Institutes of
Health)—where he has studied pulmonary vascular resistance in the infant and pediatric population. Jeff, welcome. Pleasure to be here. Thank you. You are an expert in pulmonary hypertension. Could you update us on the current working
definition of pulmonary hypertension—and, of course, in particular, in the pediatric
population—, and what we know about the epidemiology of pediatric pulmonary hypertension? I’d be happy to. First, just with the nomenclature: Pulmonary
hypertension, as a disease, is really defined with a hemodynamic parameter. So, it’s defined as having a mean pulmonary
arterial blood pressure of greater than 25 millimeters of mercury at rest, and greater
than 30 millimeters of mercury at exercise. And there is a lot of discussion now, actually,
about redefining that to as low as 20 millimeters of mercury. And the calculated pulmonary vascular resistance
is greater than 3 Wood units. Having said that: to call the
disease “pulmonary hypertension,” particularly in the pediatric population, I think is problematic, because there are clinically relevant pulmonary vascular disorders within pediatrics—and
single ventricle physiology, for example, is one example—where you may not reach that
pulmonary hypertension criteria yet. You have clinically relevant pulmonary vascular
disease. So many of us are starting to think of using
the nomenclature– instead of “pulmonary hypertension,” more either “pediatric
pulmonary hypertensive vascular disease” or just “pulmonary vascular disease in general.” The other part of the problem with talking
about pulmonary hypertension is: it’s a broad spectrum of different diseases that–
ultimately you end up with elevated pressure and resistance, but there’s a variety of
different etiologies and probably a variety of different underlying pathobiologies. If you look at the last updated classification
of pulmonary hypertension—it was in 2014 at the World Congress—, there are five groups. Group 1 is “Pulmonary arterial hypertension,”
which is the most common form in adult pulmonary vascular disease. So, within primary arterial hypertension,
you have the idiopathic pulmonary hypertension. You have the familial forms of pulmonary hypertension. You also have congenital heart disease, which,
as you know, is a major part of pediatric pulmonary vascular disease. And you also have drug-induced or toxin-induced,
et cetera. Group 2 is “Pulmonary hypertension due to
left heart disease,” which is very much more common in the adult world. Group 3 is “Pulmonary hypertension due to
lung diseases and/or hypoxia.” Altitude comes into play there– sleep-disordered
breathing, for example. And within pediatrics, you have a lot of your
lung hypoplasias, chronic lung disease, interstitial lung disease, developmental lung disorders. And Group 4 is more of a hematologic problem,
“Chronic thromboembolic pulmonary hypertension”– not nearly as common in pediatrics. And then Group 5 is what we call “Unclear
multifactorial mechanisms:” variety of hematologic disorders, systemic disorders, metabolic disorders. So, clearly there are significant differences
in not only the epidemiology but the pathobiology of pediatric versus adult pulmonary vascular
disorders. Obviously, as you can see on this slide, there
are a lot more chromosomal and genetic aberrations associated with pediatric pulmonary vascular
disease. You really have to think about developmental
biology and developmental disorders associated with the lung parenchyma and lung vasculature
and then pathological insults on a growing premature lung and vasculature. And then maturational issues related to the
right and left side of the heart are very important. Obviously, this becomes not just a disease
of the pulmonary vasculature, but it becomes a disease of right heart strain and failure. So there are some fundamental pathobiologic
differences between adult and pediatric pulmonary vascular disease. And this next slide gets to the differences
in epidemiology. This is taken from the Pediatric Pulmonary
Hypertension Network registry, where we’ve put a database with 1,500 pediatric pulmonary
hypertension patients in North America. Here, you can see that, actually, the
most common etiology right now of pediatric pulmonary vascular diseases is Group 3: that
associated with lung and hypoxia. And then Group 1—pulmonary arterial hypertension—is
the second most, and then Groups 2, 4, and 5 are very uncommon. That’s compared to the adult epidemiology,
where Group 1 is clearly the most common, followed by Group 2 and then Group 3– so,
clear differences in the epidemiology and the pathobiology. Unfortunately, we are stuck with the current
adult classification, but I think we need to make modifications as it relates to the
pediatric pathobiology. Jeff, thank you for that overview. The definition, as you noted, encompasses
so many different etiologies, and yet you’ve been studying this in the lab, at the bench,
for several decades now. How did you go about thinking about: How are
you going to isolate this down to a model that you can study? Well, I made a decision long ago to focus
on congenital heart disease. I think, even though we call it all pulmonary
hypertension, I think it’s very clear that these are all different diseases. So, I focus on congenital heart disease for
several reasons. I’m a clinician-scientist, and I work in
the cardiac ICU. And as you know, perioperative pulmonary hypertension
is a major source of morbidity and mortality. And so, clinically, I was very interested
in trying to understand this disease and help come up with therapies. Also, there are some aspects of the congenital
heart disease subpopulation that are rather unique to pulmonary hypertension. The overwhelming majority of patients that
present with pulmonary hypertension are presenting with symptoms of right heart failure. So, they already have very advanced disease. With congenital heart disease– and we don’t
really understand when it started, how it started, why it started– With the subpopulation
of congenital heart disease, we understand that it’s the holes in the heart that create
these aberrant mechanical forces on the pulmonary vasculature that lead to the development of
pulmonary hypertension. And there are some very, very nice, natural
history studies showing over time, these changes become irreversible. But we also know that if we fix these congenital heart defects at a young age, that the vascular phenotype actually reverses. So, there are several rather unique aspects
of congenital heart disease. One, we have some sense of what the underlying
initial insult is. Two, we have a good sense of the natural history. Three, we know that it’s one of the few
diseases that, if you take away these abnormal mechanical forces, it can completely reverse, where other forms are never completely reversible. And four, there seems to be a time where it
is no longer reversible. So, even if you fix the defect, it continues
to progress. And so, if we can understand those mechanisms,
we can really learn a lot about this particular pathobiology and, hopefully lead to better therapies. Here’s a slide showing what we know about
the natural history of pulmonary vascular disease with congenital heart defects. So, these are if the repair is not corrected. We know that those lesions that result in
increased pulmonary blood flow, like a ventricular septal defect– that the lesions that give
you not only a lot of flow to the pulmonary vasculature but a direct pressure head to
the pulmonary vasculature, like a truncus arteriosus– if you don’t fix them,
100% of them go on to have irreversible pulmonary vascular disease. And it occurs very early in life—clearly
within the first two years of life, if not the first year of life. On the other end of the spectrum, the ASD,
or atrial septal defect: that is a pre-tricuspid defect where there’s no direct pressure
head to the pulmonary vasculature, but just increased flow– so, flow-alone. And that– The incidence of developing irreversible
pulmonary vascular disease without surgery is only about 10% to 20%. And for that defect– it takes, actually,
decades for you to develop irreversible pulmonary vascular disease. So, there’s some interesting natural history
and lessons here that I think we can start to focus on and figure out potential mechanisms
of disease. So, it seems like the high-pressure, high-flow
lesions are much more susceptible to pulmonary vascular disease versus the flow-alone lesions. The other big subcategory of congenital heart
disease that can lead to pulmonary vascular disease is those that result in increased
pulmonary venous pressure, like left heart failure– very common in the adults in Group
2. But mitral stenosis, obstructed veins, cor
triatriatum– there we know much less about the natural history of the disease, and that’s
why, on this table, it’s very variable. So, the next slide is representative of the
vascular changes associated with a very common congenital heart defect: a ventricular septal
defect. So, first, you see the blood going from the
left side of the heart to the right side of the heart, resulting in increased pulmonary
blood flow. So, initially, the normal vascular phenotype—noted
by the red there—is that there’s a thin muscle layer, and it’s very proximal. It doesn’t extend down to the periphery. The initial change that you see over time
with increased flow—and, in this case, pressure—is that the muscle layer gets thick. So, it’s called medial hypertrophy. The proximal layer becomes more muscular. With time, the muscle layer then extends abnormally
down to the periphery where it, in the normal form, is not present. And then ultimately, with irreversible disease,
you actually lose your distal arterials, so-called “pruning of the vessels.” The anatomic morphology was classically described
by Heath and Edwards in 1958, and these were mostly autopsy specimens. And a big subset, actually, of these patients
had unrepaired congenital heart disease. And it’s very nice to categorize the spectrum
of advancing pathological lesions. But Marlene Rabinovitch actually took lung
biopsy samples of patients undergoing repair of their congenital heart defects right here
in Boston Children’s Hospital and looked at the morphologic abnormalities and related
them to the perioperative physiology. And she characterized them as three grades. Grade A is where you have some abnormal extension
of the muscle down to the periphery with mild medial hypertrophy. Those patients tend to have altered reactivity
of the pulmonary vascular bed postoperatively, but easy to manage, usually normal resistance,
and clearly very reversible phenotype once you close the repair. Grade B had some abnormal extension of the
muscle, but moderate to severe medial hypertrophy. And then grade C is where you actually started
to see the decreased arterial size and number. And those patients tended to be irreversible
patients. And even if you fix them, they’d
go on to have advanced pulmonary vascular disease. But there’s a big gray area in the middle
where it’s not clear whether they’re reversible or irreversible. And then just to touch on the right heart
because, as we talked about, there’s a lot of focus on the pulmonary vasculature, appropriately,
but many of the signs and symptoms that you show are related to the right heart failing. So, here’s just a cartoon. As we can see, different grades of pulmonary
vascular change are on top. The pulmonary artery pressure continues to
rise. And then at some point, when you get right
heart dysfunction, your cardiac output falls. So, the PA pressure actually can go down a
little bit, but your calculated resistance continues to go up. But the point here is that this becomes a
disease of right heart failure. And what’s interesting is: there are different
right heart phenotypes. We all know that the adult thin right ventricle
really does not tolerate an acute afterload very well. That’s where an acute pulmonary embolus
has such high morbidity and mortality. But there are some patients with pulmonary
hypertension that, actually, the right ventricle becomes quite adaptive, and it actually hypertrophies
and can maintain a good cardiac output despite a very high resistance. And then there are other patients where, instead
of hypertrophy, the right heart actually dilates and fails rather quickly. And when we see those things, that dictates
how aggressive we are with therapy. But I also think there are a lot of lessons
to be learned there– and: Why are some right hearts adaptive and others are not? And are there potential therapeutic targets
to be learned from those lessons? And that’s a whole area that really needs
to be investigated. As the pulmonary vascular disease progresses
in an unrepaired congenital heart defect, ultimately, instead of the blood going from
the left side of the heart to the right side of the heart, a portion or all of it begins
to go from the right side of the heart to the left side of the heart where you get,
now, desaturated blood going to the systemic circulation. And as you know, that’s referred to as the
Eisenmenger’s syndrome, and this is the report in 1958 by Paul Wood, which is a tremendous,
insightful report of these patients. Interestingly, if you look at the survival
of patients with congenital heart disease, they tend to do much better than other forms
of Group 1 pulmonary arterial hypertension, most notably idiopathic. This is a lovely, long study by Manes in his
clinic where he followed them– in this case, I think it was 14 years– 278 patients. And he showed that survival of those with
congenital heart disease was overall 85% at 10 years, 77% at 20 years, compared to their
group– same clinic, same treatments, the idiopaths– 46% survival at 10 years and a
38% survival at 15 years– much improved with congenital heart disease. Why that is is very interesting. And again, I think further evaluation of that
will lead to new therapeutic targets. But if you break up the congenital heart disease
into, kind of, the different subtypes– Here is an updated classification of congenital
heart disease. There’s the Eisenmenger’s syndrome. There are the ones with the large left-to-right
shunts—what I consider, really, pre-Eisenmenger. They just haven’t started going right to
left yet, but the pathobiology is probably the same. The third group is an interesting group: the
pulmonary arterial hypertension with coincidental congenital heart disease. This is a subset of the atrial septal defects, Where they have very high resistance,
a very small defect, a very small shunt. They are probably idiopathic, and they happen
to have a small atrial shunt. They probably don’t have real congenital
heart disease-induced pulmonary hypertension. And then the patients where you made a decision
that you think they’re reversible with surgical repair: you operate, you get them through
the surgery, and then they reemerge— a year, five years, 10 years—with advancing
pulmonary hypertension: that’s a group that the clinical phenotype is very, very aggressive. And that’s why it’s such a vital decision,
before you operate, as to whether you think they’re reversible or irreversible, because
if you look at the survival: the top two curves are the Eisenmenger’s—the Group 1 and
the Group 2, the Eisenmenger’s and the large left-to-right shunts; the pre-Eisenmenger’s:
they have the best survival compared to the Group 3, which have the small defects and
which are really similar to idiopath. And then the last group where they’re corrected:
their survival at 20 years was 36%– so much, much lower than the Eisenmenger’s or the
pre-Eisenmenger’s. So, it’s really vital to make the right
decision in terms of surgical correction. Now, why Eisenmenger’s do better
is fascinating to me. I think: clearly, they have the ability to
pop off. So, when their pulmonary vascular resistance
goes really high, instead of the right heart acutely failing, they just shunt more blood
to the left side. So, they can get very blue, but they can maintain
their left-sided output. There’s no question that that attenuates
the incidence of syncope and sudden death. But, also, there are some elegant studies
showing that, anatomically, the right ventricle kind of maintains its fetal dominance. The right ventricle, as you know, is dominant
in the fetus. And there are pictures of 77-year-olds with
Eisenmenger’s where the right ventricle looks identical to the fetal right ventricle;
that there’s something intrinsic about the right ventricle never getting to
remodel as pulmonary vascular resistance falls because it doesn’t fall normally. And this right ventricle for, let’s say,
a large unrestricted VSD is always exposed to increased flow and pressure. So, we think that there’s something related
to maintaining that fetal phenotype and perhaps genotype that makes it protective. And again, the theme here is: If you could,
with those insights, look at what those mechanisms are, there may be some nice therapeutic targets
for the right ventricle that can help these patients as well as other patients with right
ventricular dysfunction. Jeff, that was a fascinating overview. So, you took congenital heart disease not
as a convenient sample, but, as you said, it provided you with the kind of perfect model
to study because you had the natural history from fetus to outcome of unrepaired infants
and children. And from that work, you’ve been able to
distill these distinctions between flow and flow with pressure. Now, what have we learned about therapies? Because we’re upstairs
in the ICU using these therapies right now. Take us through: How did you and others learn
about how to target the mechanisms that are causing these issues? Sure. So, as you know, all the therapies currently
being utilized in adults and in children are really based on endothelial biology. And there’s a clear picture that, early
on, with congenital heart disease, that there’s endothelial cell dysfunction. And it’s interesting because, in the old
days, people would look at the blood vessels and knew that all the action was at the smooth
muscle cell layer. The constriction, the relaxation was mediated
by the smooth muscle cell layer. And the endothelial cell was kind of just
seen as a barrier cell layer between the blood and where all the action was, the smooth muscle
cell. But it wasn’t until the mid-1980s, with
a landmark study by Bob Furchgott in Brooklyn, New York, Downstate, where he
did a very elegant, simple study, but he basically showed that the endothelial cells were making
something that caused the blood vessel to relax. The smooth muscle cell then relaxed based
on what the endothelial cell was making, showing this interaction between the two layers and
that the endothelial cell was actually functional. And he first coined that an “endothelium-derived
relaxing factor,” or EDRF, and there were years where we called it EDRF and went to
EDRF meetings. And then, as you know, it was subsequently
shown to be nitric oxide. And so, all the subsequent therapies have
really been based on this. So, I’ve had a real interest in early endothelial
dysfunction as a major player in the pathobiology of congenital heart disease. It would make sense because these abnormal
mechanical forces are first being seen by the endothelial cell layer. And so, there are actually some nice, really
eloquent early observations from Boston Children’s/Harvard initially that, I think, really laid the groundwork
for this unifying hypothesis of early endothelial dysfunction in the pathobiology of pulmonary
hypertension related to congenital heart disease. So, this next slide, I think, really, for
me, was fundamental in pointing me into this world of endothelial cell biology. Marlene Rabinovitch, as you know, when she was at Boston, took these biopsy samples of children undergoing
complete repair of their congenital heart defects. And these children were young, and they all
had very reversible disease. And she did scanning electron microscopy looking at the endothelial cell anatomy. Then she went on to show that there were alterations on von Willebrand factor production, suggesting
not only anatomic aberrations in the endothelial cell—again, early on, before they have any
significant disease—, but also functional aberrations of the endothelial cell. So, this slide represents that landmark study
by Celermajer. This was performed in the cardiac catheterization
laboratory. They studied three groups of children. These squares that are black are called controls. Those are actually children that had normal
pulmonary vasculature, no increased flow or pressure. The open squares were young children that
had increased pulmonary blood flow but a normal calculated pulmonary vascular resistance,
and they were generally all infants. And the triangles, or pulmonary
vascular disease: those were older children, and they had advanced disease with resistances
calculated greater than 6 Wood units. So, the y-axis is flow velocity, and that
was determined by a catheter placed in the pulmonary artery. And then the x-axis is different conditions. C1 and C2 are control or baseline conditions. There were three increasing doses of acetylcholine
given, and then sodium nitroprusside—or NP—was given at the end. What you can see is the control patients dilated
nicely in a dose-dependent fashion to acetylcholine– no surprise. And then they also dilated nicely to sodium
nitroprusside– again, not surprisingly. If you look at the triangles, the patients
with advanced disease– so, they have a lot of structural remodeling of the pulmonary
vasculature. They don’t dilate well to acetylcholine
or sodium nitroprusside. The fascinating group to us was the open squares,
the ones with increased flow. They’re young, clearly reversible disease,
normal calculated resistance. They dilate normally to nitroprusside, but
they dilate just as poorly to acetylcholine as the group with advanced disease. Now, nitroprusside is what we call an NO donor,
or an endothelium-independent vasodilator. It does not require the endothelial cell to
make nitric oxide in order to dilate. It just donates nitric oxide by itself. Acetylcholine, on the other hand, is a clear
endothelium-dependent vasodilator. Acetylcholine dilates by binding to a receptor
on the endothelial cell and forcing the endothelial cell to make nitric oxide in order to dilate
the smooth muscle cell layer. So, what Celermajer and colleagues were showing
were that patients that were young, had increased flow, and normal resistance–
functionally had endothelial cells that were not capable of making nitric oxide to the
same extent as normal children. To me, this study was really classic evidence
that even young patients with early disease had significant endothelial dysfunction as
underlying pathobiology. Well, Dr. Fineman, thank you for walking us
through that study. This is very interesting. How does that translate, or does it translate,
to what we know about perioperative morbidity in these patients with increased pulmonary
blood flow and, in particular, increased pulmonary blood flow with pressure? Right. Well, that’s a great question. And actually, the very same year of Celermajer’s
study, there was what I think was a landmark study from this institution, with Frank Hanley
and Aldo Castañeda looking at the timing of repair of truncus arteriosus. And it really was the combination of those
studies that really stayed with me in terms of endothelial dysfunction. So, back in those days, truncus arteriosus
was really a high-risk surgery because of significant morbidity and mortality related
to pulmonary hypertension. And because of that, the approach that we
all took was to try to hold off until the patient was a little bit older and showed
signs of the pulmonary vascular resistance falling, which meant that they had signs of
increased pulmonary blood flow. So, we’d often try to get them to go out for six weeks
of age before repairing them, thinking that, if their pulmonary vascular resistance has
fallen, they’d have fewer pulmonary hypertension episodes perioperatively. What doctors Hanley and Castañeda showed
was actually the opposite. They showed that the patients that were operated
on sooner did better in terms of outcomes, and they came out of the operating room with
lower pulmonary artery pressures than the ones that were done later, and they
had far fewer pulmonary hypertension episodes if they were done earlier. So, they concluded—and I completely agree,
and the field 30 years later completely agrees—that the longer the pulmonary vasculature—and
I would add: the endothelial cell layer—is exposed to these abnormal forces of increased
flow, the shear that’s associated with that, and the pressure and the cyclic stretch that’s
associated with that, the worse it is for the patient. So, by repairing them very early on, you minimize
that time, and, in fact, the endothelial aberrations are less, and that relates directly, I think,
to improved perioperative outcomes and decreased perioperative morbidity related to pulmonary
hypertension. Just as a reminder to the audience: As you
can see in the diagram, there are variations on the theme, but a truncus arteriosus is
basically where the pulmonary arteries are coming off the common aortic trunk. So, it exposes the pulmonary vasculature to
very high pressure and flow. Those observations and studies really
have led to this kind of unifying hypothesis that, particularly with congenital heart disease,
the initial insult is that pressure and flow result in an early endothelial injury, and
that results in the alterations in the vasoactive factors that the endothelial cell makes, such
as nitric oxide, prostacyclin, endothelin-1, alterations in reactive oxygen species generation,
alterations in extracellular matrix, and the combination of these things and probably many
other things that we don’t understand yet results in the two fundamental processes of
this disease where there’s intense vasoconstriction and significant vascular remodeling of the
pulmonary arterials. This is a cartoon of those three endothelial-based
cascades and showing the nitric oxide cascade on the left, the endothelium cascade on
the right, and the prostacyclin cascade in the middle. And there’s a lot of data—both in animal
studies and in humans—suggesting that, with pulmonary vascular disease, the nitric oxide
cascade is downregulated as well as the prostacyclin cascade. Both nitric oxide and prostacyclin promote
vasodilation, and they inhibit smooth muscle cell proliferation. So, they keep the muscle layer nice and thin. Endothelin, on the other hand, which has been
shown to be upregulated in pulmonary vascular disease, is a potent vasoconstrictor, and
it actually causes reactive oxygen species generation and smooth muscle cell proliferation. So, all of the therapies to date– and, as
you know, over the past decade, they’ve increased significantly– they’re all based
on either augmenting the nitric oxide pathway, augmenting the prostacyclin pathway, or blocking
the endothelin pathway. Now, the survival of these patients has improved
dramatically with the advent of these therapies. And we really hope that as new therapies emerge, as our understanding emerges, that, in fact, we’ll ultimately be able to cure some of these patients. But this endothelial biology and understanding
underlying pathobiology has really led to a tremendous emergence of new therapies. Jeff, that’s absolutely fascinating work. Has that translated to an animal model? That is: Is there a model that exists where
you can examine the outcomes for a lesion that’s producing high flow versus a lesion
that’s producing high flow and high pressure? Yeah, that’s a great question. So, that’s exactly what we’ve been investigating
for the last 15, 20 years. We have a large animal model, and we initially
created a model that results in both high flow and high pressure. And we showed that these aberrations, these
endothelial cell aberrations, occur dramatically, and they occur very, very early on. For example, the endothelium cascade, which,
again, promotes vasoconstriction and smooth muscle cell proliferation: that’s upregulated
in our animal model within the first week of life. So, they’re born. Pulmonary vasculature starts getting exposed
to increased flow and pressure, which is abnormal. And within four or five days, there’s a
massive upregulation of the endothelium cascade. The levels are high. The good receptor that mediates nitric oxide
is down, and the bad receptor that mediates vasoconstriction, within a couple weeks later,
is markedly elevated. Conversely, on the nitric oxide side of the
equation, we show that, within the first month of life, that the enzyme that is making nitric
oxide—endothelial nitric oxide synthase, that enzyme which classically gets stimulated
by flow or shear, because there’s flow—is massively unregulated. But instead of making nitric oxide, the enzyme
is what we call “uncoupled” for a variety of reasons, and it actually makes reactive
oxygen species. So, that results in decreased what we call
“bioavailable” nitric oxide. And in addition, it’s causing oxidative
stress, which causes further injury to the vasculature. So, that is the work that we’ve done with
a model of high pressure and high flow. So, to get to your question of kind of translating
the natural history of a flow-alone lesion, like an atrial septal defect, versus a flow
and pressure lesion, like a truncus arteriosus: We just recently created a flow-alone model
and compared it to what we’re calling the “shunt model”, that I just described. So, the flow-alone model, what we do is we just
ligate the left pulmonary artery. So, the right lung now gets twice as much
pulmonary blood flow, but without the direct pressure head that’s seen with a truncus
arteriosus or an aortopulmonary window. This is a CT angiogram comparing– they’re
all about a month of age, and the controls are twin age-matched controls. And this is just the anatomy of the three models. The shunt on the right side has very large,
engorged pulmonary vasculature compared to the control. And the LPA ligation shows a relatively normal
right side and no left pulmonary artery. What we’ve been able to do is culture the
cells, and they maintain a phenotype, which is very, very helpful. And this is something called RNA sequencing. And so, the control cells, the endothelial
control cells, are in the dark blue, the shunted cells are in the light blue, and the flow-alone, or the LPA cells, are in the red. And what I hope you can appreciate is kind
of this clustering in terms of their genome is that they’re very, very different, these
three groups of animals. And then, this heat map: the things in red
are showing an upregulation of genes. The things in green are a downregulation in
genes. And I hope you can appreciate the marked differences
in the pattern of gene expression between the three models. So, next, we try to look at kind of the functional
differences between the three groups. And so, as you know, classically, patients
with pulmonary hypertension have what we call “increased vascular reactivity.” It could be a stimulus like hypoxia or alpha-adrenergic
activation that causes pulmonary vasoconstriction in all of us, but if you have an abnormal
pulmonary vasculature, it can be very, very intense. So, we challenged these three animal models
with either acute alveolar hypoxia or this drug called U46619, which is a thromboxane
mimic. And, as I think you can appreciate: In the
dark blue, this is a change in pulmonary artery pressure in response to those insults– that
these shunted animals, flow plus pressure, have a marked increase in pulmonary artery
pressure in response to them. The control, or the normal lambs in light
blue have a much attenuated response, and the flow-alone have somewhat of an intermediate
response. So, that was in the intact animal. If we take the pulmonary arteries out and
hang them in a muscle bath, we can do similar studies with different increasing doses of
drugs. And in this case, we study norepinephrine,
the same color-coding. I think you can appreciate that the shunted
animals—pressure and flow—have a marked increase in reactivity to norepinephrine where
the flow-alone are very similar to normal animals, much more attenuated response. If you look at the morphology, we talked about
smooth muscle cell layer being thick in this classic disease. Well, again, these animals are only four weeks
old. They’re very, very young, yet they have
some medial hypertrophy in the pressure-and-flow-mediated group, the shunted group, where the LPA ligation,
or the flow-alone, their blood vessels looked normal. Then, lastly, a couple of things we’ve been
looking at– the ability of the endothelial cells to actually make new blood vessels or
angiogenesis that are budding. And in fact, the pressure and flow stimulus
seems to increase their ability to do that as opposed to the flow-alone, which seems
to be intermediate compared to the normal ones Now, whether that’s adaptive or maladaptive
early on is not clear to us. But the other thing that’s interesting is
their resistance to apoptosis, or natural cell death, the feeling being that an anti-apoptotic
or proliferative state is pathologic. I hope you can appreciate in this slide
that the shunted animals– this is TNF-induced apoptosis. It’s kind of a classic assay. The cells from the shunted animals don’t
apoptose nearly as much as controls. But, again, the flow-alone, the LPA ligation,
they’re somewhat intermediate. So, it seems like the stimulus of pressure
and flow is very different than flow-alone. And then we’ve just started to look at
some of the basic endothelial functional targets that have classically been targeted, endothelin-1
and nitric oxide. And where we’re going with this is: Let’s
start with the drugs that we have, and can we learn about different physiologic insults
within our patient population? What endothelial biology is perturbed with
different insults? And maybe we can start by taking the drugs
that we have and targeting the use of them based on the underlying physiology. It’s a lot of work to go on to try to get
new therapies, but let’s just start with the targets we have. So, here you can see lung endothelin levels. Again, we’ve shown previously and I’ve
spoken about that the shunted animals have marked elevation in endothelin levels. But the flow-alone animals, really it’s
somewhat maybe intermediate, but not very different than control. And then, if you look at the protein levels
of the precursor lung prepro-ET-1, you can see marked elevation of the shunt with pressure
and flow, but the flow stimulus does not nearly generate the upregulation in endothelin-1. If you take these cells and then apply them
in vitro to mechanical forces, if you stretch them, which is, we think, one of the stimuli
associated with pressure, distending and stretching the pulmonary artery, you can see a marked
elevation with cyclic stretch endothelin-1 levels. But if you just expose them to shear, which
is associated with flow, or physiologic shear, not abnormally high shear, endothelin levels
actually come down—again, suggesting that perhaps those lesions with pressure head are
more prone to have an upregulation of endothelin-1. Those may be the patients we want to make
sure we’re blocking that cascade, where the flow-alone patients, maybe we don’t
need to block that cascade. Lastly, with the nitric oxide: We’ve talked
about the enzyme that makes nitric oxide, nitric oxide synthase, classically stimulated
by flow or shear. Makes sense that blood flowing across the
endothelial cell will stimulate nitric oxide production, keeps it nice and relaxed in the
normal state. You can see in the bar on the left—looking
at these enzyme protein levels—that both in the shunt and in the flow-alone, the protein
levels of this enzyme are markedly elevated. They both have flow as a component to their
defect. However, if you look at the end product, are
they making nitric oxide? In fact, as we showed previously, the shunted
animals have this uncoupled enzyme where they don’t make nitric oxide, and, in fact, they
make reactive oxygen species. You can see, in the shunted animal, that the
amount of nitric oxide in the lung is actually significantly lower. But the flow-alone actually is making nitric
oxide, and they have similar bioavailable nitric oxide levels to control. So, even though they’re both upregulated
by the flow, it seems like the pressure associated with it causes further perturbation of the
enzyme and actually uncoupled it where just flow-alone seems to result in nitric oxide
production. So, kind of to summarize where we’re going
with all of this work is we believe that combined mechanical forces of pressure and flow result
in abnormal vascular remodeling, abnormal functional vascular reactivity, endothelial
cell proliferation, angiogenesis, and anti-apoptosis, and marked endothelial dysfunction. Particularly, what we’re starting with is
a marked elevation endothelin-1. Flow-alone, however, seems to be an intermediate
phenotype. And where we want to get with this work–
and as you saw in the heat map, there are a lot of genes to explore here– that we want our
therapy to be informed by the physiologic etiology of the particular pulmonary vascular
disease or the congenital heart disease causing the pulmonary vascular disease. And I think, in general, our pulmonary vascular
disease therapy goal is: To optimize the endothelial-based therapies. We’ve got some great ones. There may be more targets here within the
endothelial cell pathobiology that can be optimized. Right now, we have no targets focused on the
smooth muscle cell. There are clearly a lot of aberrations there. There’s a lot of nice work showing abnormal
metabolic pathways in the smooth muscle cells that are going to be targets. As we talked about with Eisenmenger’s, there
are lessons to be learned there in the adaptive and maladaptive right ventricle. By looking at those mechanisms, perhaps there
are targets to improve RV function in these patients. And then we’re learning more and more about
genetic predispositions to this disease. And as we learn more about these genes and
then their associated signaling abnormalities, perhaps there are targets there that we can
get to. And as you know, Dr. Burns, currently our
therapy is rather crude. We have three types of drugs– again: augmenting
nitric oxide, augmenting prostacyclin, and blocking endothelin. If we have mild disease, we give one of these
drugs. If we have moderate disease, we give two of
these drugs. And if they have really bad disease, we give
all three types of these drugs. It’s not based on any underlying mechanism
or pathobiology. We must get to a place within congenital heart
disease, but within the field overall, where the therapies must be informed by understanding
the underlying pathobiology as opposed to how bad the disease is. Well, Jeff, that’s a wonderful overview
of several decades of work by you and others that has helped advance our thinking. And as you just summarized, right now we’re
still in the era of one size fits all. It’s all “pulmonary hypertension,” and
“Here’s three medications.” But the work that you and others around the
world are doing– I think we can see that we’re entering a more precise era where
we’ll be able to target our therapies. So, on behalf of colleagues around the world,
thank you very much for being with us today and thank you for all the work you’ve been doing. Thank you for having me and thank you for all the work that you do. Really appreciate it.

Leave a Reply

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