Circulation October 30, 2018 Issue

Dr Carolyn
Lam:               
Welcome to Circulation on the Run, your weekly podcast summary
and backstage pass to the journal and its editors. I'm Dr Carolyn
Lam, associate editor from the National Heart Center and Duke
National University of Singapore. This week's issue provides much
long awaited healthcare resource utilization and cost
implications in the MOMENTUM 3 randomized controlled trial of a
magnetically levitated cardiac pump in advanced heart failure.
All of this coming right up after these summaries.


                                               
The first original paper this week provides important mammalian
data on the acute effects of phosphodiesterase type 1 inhibition
on the heart. Now phosphodiesterase type 1, or PDE1, is known to
hydrolyze cyclic AMP and cyclic GMP in the heart. However, what's
important to understand is that data from rodents may not be
applicable to humans because rodents express mostly the cyclic
GMP favoring PDE1A isoform, whereas human hearts predominantly
express PDE1C isoform which has a balanced selectivity for cyclic
AMP and cyclic GMP.


                                               
In today's paper, first author Dr Hashimoto, corresponding author
Dr Kass from Johns Hopkins University School of Medicine and
colleagues, determined the acute effects of PDE1 inhibition on
PDE1C expressing mammals, dogs and rabbits, in normal and failing
hearts. They found that selective inhibition of PDE1 with ITI-214
induced positive inotropic, lusitropic, chronotropic, and
arterial vasodilatory effects in dogs and rabbits. These effects
occurred via cyclic AMP modulation and were observed in failing
hearts. ITI-214 contractile increase was insensitive to beta
adrenergic blockade or heart rate increase, but inhibited in vivo
by adenosine receptor inhibition. Furthermore, isolated myocytes
revealed differences between PDE1 and PDE3 inhibition. Wherein
PDE3 inhibition, augmented beta receptor agonism and calcium
transients, whereas PDE1 inhibition enhanced function without
calcium increase. These findings have important clinical
implications for ITI-214 which has completed phase 1 trials and
may provide a novel therapy for heart failure.


                                               
We know that macrophages are involved in foam cell formation in
atherosclerotic plaques, but our next paper tells us we may now
have a way to therapeutically modify this. Co-corresponding
authors Dr Wei and Schober from Ludwig Maximilian's University
Munich elucidated the role of microRNA generating enzyme Dicer in
macrophage activation during atherosclerosis. They showed that
Dicer deletion in macrophages accelerated atherosclerosis in
mice, along with enhanced inflammatory response and increased
lipid accumulation in lesional macrophages. In vitro, alternative
activation was limited, whereas lipid filled foam cell formation
was exacerbated in Dicer deficient macrophages due to impaired
mitochondrial fatty acid oxidative metabolism. MicroRNA
biogenesis promoted the degradation of fatty acids by
mitochondrial respiration in macrophages, which in turn reduced
intracellular lipid storage and limited atherosclerosis. Thus,
reducing foam cell formation in atherosclerotic arteries by
enhancing energy metabolism through microRNA mediated fatty acid
oxidation may be a promising approach for the treatment of
atherosclerosis.


                                               
The next study evaluates how aortic stiffening relates to resting
cerebral blood flow and cerebral vascular reactivity in older
adults. First and corresponding author Dr Jefferson from
Vanderbilt Memory and Alzheimer's Center and her colleagues
studied participants free of clinical dementia, stroke, or heart
failure, including 155 older adults with normal cognition and 115
mild cognitive impairment. They found that greater thoracic
aortic stiffening quantified by cardiac magnetic resonance was
associated with lower cerebral blood flow in cognitively normal
older adults. Aortic stiffening was associated with reduced
resting cerebral blood flow in the presence of preserved
reactivity and associated vasodilatory capacity, particularly
among participants without hypertension. ApoE4, a well-known
genetic susceptibility risk factor for Alzheimer's disease,
modified the results with stronger effects among carriers in the
temporal lobes, where Alzheimer's disease pathology is known to
first evolve. In summary, greater aortic stiffening related to
lower regional cerebral blood flow and higher cerebral vascular
reactivity in cognitively normal older adults, especially among
individuals with increased genetic predisposition for Alzheimer's
disease. Understanding the association between higher aortic
stiffness and compromised brain health, including cerebral
hemodynamics, may allow for earlier detection and targeted
interventions to prevent or mitigate the onset of more serious
cerebral vascular damage associated with greater aortic
stiffening.


                                               
Aortic valve replacement for aortic stenosis is usually timed
according to the development of symptoms, but could the timing be
too late once irreversible myocardial scar has developed?
Co-first authors Drs Musa and Treibel, corresponding author Dr
Greenwood from University of Leeds and their colleagues found
that in patients with severe aortic stenosis, focal myocardial
fibrosis determined by cardiac magnetic resonance imaging was
present in over 50% of patients and was associated with a
two-fold higher late mortality. Focal scar was independently
associated with all cause and cardiovascular mortality, after
both surgical and transcatheter aortic valve replacement. In
severe aortic stenosis, late gadolinium enhancement appears to be
a useful biomarker of left ventricular remodeling, and its
presence is associated with worse long-term outcomes following
aortic valve intervention. Thus, in severe aortic stenosis, late
gadolinium enhancement may be a useful biomarker of left
ventricular remodeling, and its presence may be associated with
worse long-term outcomes following aortic valve intervention.


                                               
The next study suggests that endogenous factor Xa activity may be
irrelevant pharmacodynamic marker to guide Edoxaban dosing in
future. First author Dr Yin, corresponding author Dr Giugliano
from TIMI Study Group, Brigham and Women's Hospital in Boston,
and their colleagues, describe the value of endogenous factor Xa
activity as a pharmacodynamic marker, linking Edoxaban
concentrations and clinical outcomes in the ENGAGE AF-TIMI 48
trial. They showed that the extent of inhibition of endogenous
factor Xa activity was influenced by Edoxaban dosing and clinical
characteristics, and was associated with both antithrombotic
benefit and risk of bleeding. The implications are that this
approach of linking endogenous factor Xa activity to clinical
outcomes may be used to guide dose selection in future clinical
trials, to monitor patients in certain clinical scenarios, or to
define the doses of oral factor Xa inhibitors in patients who
require precise anticoagulation therapy.


                                               
The next paper describes a novel multi-protein complex that plays
a critical role in regulating cardiomyocyte survival. First
author Dr Zhang, corresponding author Dr Yan from University of
Rochester School of Medicine and Dentistry and colleagues, showed
that phosphodiesterase 1C is activated by transient receptor
potential canonical channel-3 derived calcium, thereby
antagonizing adenosine A2 receptor cyclic GMP signaling and
promoting cardiomyocyte death or apoptosis. Targeting these
molecules individually, or in combination, may represent a
compelling therapeutic strategy for potentiating cardiomyocyte
survival.


                                               
The final paper demonstrates a molecular link between two
well-recognized biomarkers of fibrosis, Galectin-3 and
Osteopontin. First author Dr Shirakawa, corresponding author Dr
Sano from Keio University School of Medicine and their
colleagues, showed that Osteopontin was almost exclusively
produced by Galectin-3 high CD206 positive macrophages, which
specifically appear in the infarct myocardium after a myocardial
infarct. The interleukin-10-STAT3 Galectin-3 axis was essential
for Osteopontin producing reparative macrophage polarization
after myocardial infarction, and these macrophages contributed to
tissue repair by promoting fibrosis and clearance of apoptotic
cells. These results therefore suggest that Galectin-3 may
contribute to reparative fibrosis in the infarct myocardium by
controlling Osteopontin levels. And that brings us to the end of
this week's summaries, now for a feature discussion.


                                               
Left ventricular assist devices have truly revolutionized our
management of advanced heart failure. In fact, these devices have
allowed us to keep patients not just as a bridge to
transplantation, but as destination therapy. The devices get
better and better but also more and more expensive, and the
problem is, that places a lot of strain on our healthcare
systems. A lot of us are crying out for information on the cost
effectiveness of these newer devices, and guess what? We have
answers this week with our featured paper.


                                               
I am delighted to have with us the first and corresponding author
Dr Mandeep Mehra from Brigham and Women's Hospital in Boston,
Massachusetts, as well as our senior editor Dr Biykem Bozkurt
from Baylor College of Medicine in Houston, Texas. Hello, Mandeep
and Biykem! I am so pleased to be talking about a subject really
close to all our hearts. Mandeep, could you start by maybe
sketching out the actual issue, and maybe reminding our audience
what's the difference between the different types of left
ventricular assist systems that you compared.


Dr Mandeep Mehra:       The era of
left ventricular assist devices took a major therapeutic shift
when we recognized that we could usher in continuous flow
devices. These are devices that generate no peripheral pulse,
they do not have systole and diastole. And these devices are
small in profile, have very few moving parts, and there are
several commercially available devices, two in the United States
and up to three worldwide, that bear these characteristics.


                                               
The HeartMate II device, which is a continuous flow device that
flows blood in an axial format. The HeartWare, or HVAD device,
which is a centrifugal flow pump, where the blood comes in and
then is ejected at a 90 degree angle. The Jarvik 2000 pump that
is still used in some areas, in many regions experimentally, and
then the new kid on the block, the HeartMate 3 device, which is a
centrifugal flow pump with some very unique technological
characteristics.


Dr Carolyn
Lam:               
Nice! And now drumroll, please tell us what you found in your
brilliant study this week.


Dr Mandeep Mehra:       First, I'd
like to remind the audience that the MOMENTUM 3 trial which
randomized patients to the HeartMate II versus the HeartMate 3
device, was called MOMENTUM 3 and was a two-year study. We
presented the pivotal two year trials results in 366 randomized
patients earlier this year in The New England Journal of
Medicine, and this study showed that the HeartMate 3 was superior
on the primary endpoint when compared to the HeartMate II. The
primary endpoint was survival, free of a disabling stroke, or the
need to replace the pump surgically for a pump malfunction. And
much of that, Carolyn, was driven by the need for replacement of
the pump because the HeartMate 3 pump has some unique features
that reduce its proclivity for pump thrombosis.


                                               
The HeartMate 3 pump is a frictionless pump. It's completely,
magnetically, dynamically, born in the rotor. It has wider blood
flow paths, so we don't see hemolysis with this pump. And this
pump also has an artificial intrinsic pulse that has been
created, that pulsates the pump in a 40 beats per minute
configuration. So this was the primary trial result, and one of
the lucky foresights that we had when we designed the trial was
to embed, prospectively, economic analysis within this trial. We
recognized that the cost effectiveness related issues and cost
configurations with these devices would become very, very
important as we scale into today's day and age of healthcare
transformation. And the paper that is being presented in
Circulation this week, really speaks to the health resource
utilization and cost outcomes between the two devices.


                                               
We found that the HeartMate 3 pump is actually a cost
minimization device, and what that means, Carolyn, is that we
have become very used to thinking of new technology as providing
incremental costs. So we think that, "Oh, well, what incremental
costs should society bear for the benefits as we allocate new
technology?" And in this particular trial, what we found is that
while the costs of the pump itself, the HeartMate II and the
HeartMate 3, were kept the same, which means its operational
implant costs were the same, pretty much. We found that the
HeartMate 3 pump was associated with a reduction in healthcare
resource utilization over two years and with a marked decrease in
cost. And in fact, our estimate of cost reduction was in the
range of about 65 thousand dollars less, compared to the
HeartMate II, in favor of the HeartMate 3.


Dr Carolyn
Lam:               
Wow, Mandeep, first of all, congratulations on these remarkable
findings. Biykem, I really have to bring you in here. What do you
think of the implications of this?


Dr Biykem
Bozkurt:         First, I
would like to congratulate the authors for a very innovative
approach. As Mandeep has stated, they prospectively collected
very challenging billing data from the hospitals, and then also
did a very complex analysis including the VRG, as well as looking
at payer reimbursements for public versus private. And did a
variety of subgroup analysis, which I thought was quite helpful
in sorting out that perhaps the cost effectiveness was concurrent
both from the Medicare, the public, as well as the private, or
regardless of the intent for destination versus bridge to
transplant.


                                               
Probably the most important concept when you look at these close
analysis is incremental cost effectiveness ratio, per quality of
adjusted life year gained. Now, I do realize the current analysis
doesn't allow us to infer the ICER benefit or the incremental
cost effectiveness, which I think the investigators are planning
to do with a thousand and more patients over a course of two
years, which is going to be probably the more definitive. But as
it currently stands, with what is provided by Dr Mehra and his
colleagues is, we're probably reaching that sweet spot of what is
construed as the cost effectiveness ratio of a cost.


                                               
Let's say 100 thousand dollars over the course of a year, then I
would like to ask Mandeep whether on the prediction will reach
that threshold of less than 100 thousand dollars. Because the
former studies, looking at the ICER ratios, or incremental cost
effectiveness ratios for the DT destination therapies, usually we
select somewhere around 200 thousand dollars. And I know that
usually that is seen as a prohibited cost, and there was a
discussion whether we would be able to reduce the cost by about
half, either doing index admission and add subsequent
hospitalizations. With the data Dr Mehra and his colleagues have
shown, it looks like the re-hospitalization cost is about,
approximately half, or reduced by 50%. Mandeep, any thoughts on
that, on that sweet spot?


Dr Mandeep Mehra:       Yeah. I
think, Biykem, you have articulated this extraordinarily well.
And for the audience, since it's worldwide, I'd like to place a
few things in perspective on how to think of economic modeling.
First of all, the point I would make is that this is the first
prospectively collected data that we have in the field, and as
you pointed out, it was very, very difficult to pull this data
together and is still very complex. But let's just think about
what ICER really is. It all starts with what we consider to be
health utility.


                                               
For example, Carolyn, Biykem, and me less so, would have a health
utility of 1.0, 1.0 means a perfect health utility number. And I
know, Carolyn, you and Biykem are absolutely perfect so you would
be a 1.0, I probably am not a 1.0. But a patient with advanced
heart failure has a health utility of about .4, so that's only
40% of what is perfect. And when we place ventricular assist
devices, whether you place the HeartMate 3 or the HeartMate II,
the health utility actually jumps up to about .7. So it's not
perfect yet, but it moves all the way up there.


                                               
The incremental cost effectiveness ratios of implanting a device
over time are calculated based on this health utility benefit,
compared to the population of advanced heart failure. And the
best current estimates of the HeartMate II are that ICER is about
200 thousand dollars, per quality adjusted life years gained, and
this has been done by creating what's known as Markov modeling. A
lot of that, by the way, is conjecture, it's not real
information. It is predicted information, so one has to take that
data with a grain of salt.


                                               
Here in this health resource analysis for MOMENTUM 3, we actually
looked at actual data. There are some estimates used in this
analysis as well, where we did not have accurate billing forms
available, but we focused on those things where we had very clear
knowledge of the cost of outcomes. For example, we did not look
at the costs of outpatient follow-up care. We mainly looked at
the cost differences of hospitalizations. And what we essentially
found here is that just looking at hospitalizations and
differences between the two devices, the cost differential,
whether it's Medicare which is public [inaudible 00:20:14], or
whether it's commercial. It ranges somewhere between 50 to 65
thousand dollars of difference between the two devices.


                                               
Now, if you assume that the ICER for the HeartMate II is
accurately at about 200 thousand, and you reduce that ICER by
about 50 to 60 thousand, the ICER would naturally come into the
range of what you would consider to be about 135 thousand to 150
thousand dollars per quality adjusted life years gained for the
HeartMate 3, compared to an advanced heart failure population.
Once we look at it from that perspective, as Biykem pointed out,
we are getting closer and closer to the societal norms.


                                               
At one time-point, society used to think of a quality adjusted
life years gained cost of 50 thousand dollars as something that
would be acceptable to society, and this was seemingly based on
the threshold for what dialysis provides in benefit. And now, we
recognize that we have to really expand that to somewhere around
100 thousand more logically, or between 100 and 150 thousand for
some technologies. The important thing I would say to you is
that, that is society dependent. So what the United States
considers to be a reasonable ICER, say 100 to 130 thousand
dollars per quality adjusted life years gained, may not be the
same that Great Britain would look at, or Sweden would look at,
or another country would look at. And each country actually
creates their own economic value propositions, and this will have
to be taken into account as we think about this data as well.


Dr Carolyn
Lam:               
How cleverly and clearly articulated, thank you so much Mandeep.
Just one last question for both you and Biykem, what do you think
this implies for moving to less and less advanced heart failure
with these left ventricular assist device systems? Biykem?


Dr Biykem
Bozkurt:         It's an
ever-expanding field, and as these devices are becoming smaller,
lower profile with lesser complications and more affordable,
probably the utilization will likely increase as we have been
seeing. As you know, even the percutaneous non-durable device
used, as well as our mechanical circulatory support durable
devices are definitely increasing utilization. And thus, one may
wonder not only the bridge to transplantation, but the
destination therapy portfolio, or bridge to decision portfolio,
may really increase as these devices become safer and more
affordable.


Dr Carolyn
Lam:               
Wow, that's amazing. How about you, Mandeep, what do you think?


Dr Mandeep Mehra:       Carolyn, I
couldn't have said it any better than what Biykem articulated. I
do think that at least in the United States, as we reach the
thresholds of cost effectiveness that we as a society accept, we
will start to see a lot more widespread utilization, particularly
for lifelong therapy or so-called destination therapy. I
completely agree with that. I think that moving the needle to the
less sicker population is still challenging, because there are
complications with these devices that make that slightly
difficult.


                                               
There was a trial called the REVIVE-IT trial that was stopped
midstream largely because of concerns about pump thrombosis, and
that trial was looking at taking these devices to a less sick
NYHA class 3 population and was stopped midstream. Now that the
HeartMate 3 has pretty much resolved the issue of pump
thrombosis, and even show a halfing in stroke rates with this
device over two years, I think that that portfolio of evidence
needs to be reopened. I would caution though, that until we have
confirmatory randomized data in those less sick populations, the
use to that population should still stay restricted.


Dr Carolyn
Lam:               
I don't think anyone could have said it better than both of you.
Thank you so much for this very insightful and balanced
conversation.


                                               
Thank you so much for listening today. You were listening to
Circulation on the Run, and don't forget to tune again next week.