SCIENCE AT THE SHINE DOME canberra 4 - 6 may 2005

Symposium: Recent advances in stem cell science and therapies

Friday, 6 May 2005

Professor Bob Graham
Executive Director, Victor Chang Cardiac Research Institute, Sydney

Bob Graham is the Des Renford Professor of Medicine at the University of New South Wales and Executive Director of the Victor Chang Cardiac Research Institute in Sydney. His research for many years has focused on molecular cardiology, with an emphasis on circulatory control mechanisms, receptor signalling and cardiac hypertrophy. He is the author of over 200 peer-reviewed scientific papers. Currently, he is Chairman of the Scientific Committee of the Cardiac Society of Australia and New Zealand. He is also a member of the NSW Ministerial Advisory Council on Medical and Health Research, and of the Research Committee of the National Heart Foundation (NSW Division). He is also serving on the Boards of EngeneIC Ltd, MirACL Therapeutics Pty Ltd, St Vincent's Research and Biotechnology Precinct Ltd, the Rock Eisteddfod Foundation, the Scientific Advisory Boards of the Institute of Molecular Biosciences at the University of Queensland, Mesoblast Ltd, and Zensun Ltd.

The heart – potential use of stem cells to treat coronary heart disease

The issue that I want to address today is a very major issue in our community, ischaemic heart disease: 50 per cent of people over 65 years have some degree of coronary artery disease, and it leads on to infarction and then on to heart failure, which is really an emerging epidemic in our society. In the United States alone, 5 million people are afflicted with heart failure, and there are some 400,000 to 600,000 new cases a year. That rolls off the tongue, but it is a staggering figure and it is very expensive, because it requires recurrent hospitalisations. It is by far the single biggest factor in the blow-out of expenses in our health care. And it is going to increase as our population ages. It has been, in fact, estimated that over the next 10 years there will be a two- to three-fold increase in the number of people with heart failure, unless we find some new ways to address the problem.

The issue we face is, firstly, that the myocardium is largely incapable of self-repair, whether following ischaemic injury or reperfusion injury. And, shortly after birth, cardiomyocytes – much like brain cells – irreversibly exit the cell cycle. There is some glimmer of hope that there may be some cells that are able to continue to proliferate, but by and large the great mass of heart muscle cells are post-mitotic and can't proliferate.

Myocardial injury heals by scar formation, with very little evidence of regeneration. Histologically, you rarely see mitotic figures. And that leads to impaired myocardial function and to enhanced apoptosis. Heart failure, therefore, can be viewed as a muscle cell deficiency disorder.

(Click on image for a larger version)

This failure to regenerate is very different from what you see in other organisms. For example, shown here is a work of Aeschylus, showing Prometheus, who was damned to be chained to Mount Caucasus for 30,000 years. Every day an eagle would come down and pick his liver, and the next day it would have regenerated. So the Greeks have known from ancient times that the liver can regenerate. Some organs have a tremendous regenerative capacity.

(Click on image for a larger version)

And certain species, such as the newt and the fish, have a tremendous capacity to regenerate. The newt, when you amputate a limb, undergoes a process of wound healing involving de-differentiation, proliferation of the cells, and then a redifferentiation process to completely regenerate the limb that has been removed.

(Click on image for a larger version)

Similarly in the zebra fish, as shown recently by Mark Keating and colleagues, if you amputate the heart, taking 20 per cent of the tip of the heart off, you get a big clot formed but over about 60 days you completely regenerate the heart, without getting any scar formation at all.

(Click on image for a larger version)

You can prevent this regeneration from occurring by using, for example, a mitotic checkpoint kinase mutant, as Mark Keating did, which blocks cell proliferation. Here you see the mutant as compared with the wild-type. With the latter, you see some collagen staining of the fibrotic scar that is formed early on, which disappears as the heart regenerates. But if you prevent regeneration then you go on to get massive amounts of scar formation, as seen in the regeneration-incompetent mutant zebra fish. So regeneration is important to prevent scar formation from occurring.

We don't understand these processes. In some tissues we do, such as the brain – we have a good understanding as to why lower species such as fish can regenerate central neurons, whereas we can't – but in the heart we really are just starting to grapple with these processes.

So what are the potential solutions? Certainly there are a number, and obviously I am not going to be able to cover all of them. Gene therapy has been looked at but so far it has not gone very far; there is a variety of cell therapies, which I am going to focus on today; and there are combinations of gene and cell therapy.

Then there are various regenerative procedures, obviously very much in the experimental stage. They include the evolutionary paradigms I mentioned, where you get a blastema formed, as in the newt, that is the de-differentiation of cells and then repair that I mentioned; the use of telomerase, for example, to inhibit pro-death pathways (overexpresses of telomerase have been shown to produce continued proliferation of heart muscle cells in animal models); and overriding cell-cycle checkpoint constraints, such as overexpression of cyclin D, which have also been shown to be of benefit in animal models of coronary artery disease or ischaemia.

Today I am just going to focus on stem cells, particularly skeletal myoblasts and other stem cells, and only talk about adult cells, by and large – which is what all the clinical trials to date have utilised.

Adult progenitor cells have been used in a number of trials, as we have heard already. All have used autologous cells to avoid problems with rejection. And three different cell types have been evaluated.

Skeletal myoblasts, which are undifferentiated, proliferation-competent cells, are being utilised in proof-of-principle studies. They have been isolated from skeletal muscle biopsies and then expanded ex vivo to generate a sufficient number of cells; and they have been directly injected into the ventricle or the heart of patients at the time of coronary artery bypass grafting.

Bone marrow cells are also being utilised, either unfractionated or fractionated, to get crude or enriched mononuclear preparation respectively. As many of you may know, bone marrow is complex. It contains hemopoietic stem cells and it also contains side population cells, which are probably responsible for most of the long-term self-renewal. I will tell you a little bit more about those in a moment.

Bone marrow also contains mesenchymal or stromal cells, and then a subset of mesenchymal cells called multipotential adult progenitor cells. So bone marrow is complex, and in most of the trials to date people have used crude preparations so we don't know, when they do see benefits, which cell type is responsible and why.

In addition, peripheral blood-derived cells, particularly endothelial progenitor cells – or what we think are endothelial progenitor cells – have been utilised, either directly isolated from the bone marrow or after their mobilisation from marrow by the administration of cytokines.

(Click on image for a larger version)

Let me mention a little bit about side population cells. If you do fluorescence-activated cell sorting on cells from various different organs, you will see a population of cells – here in panels B and C shown on the side, hence the name – which do not show staining with a particular dye. The reason for that is that they overexpress or they express in high amounts an efflux pump of the ABC cassette transporter family. This seems to be a very important marker of self-renewal of these cells. If you did this same analysis with hemopoietic stem cells and you took just a few or even 20 of these cells, you could completely regenerate all the hemopoietic lineages in a mouse that had been lethally irradiated, and the animal thus survives. So they are clearly extremely important. They are present even in cardiac cells, as shown here in work from Dan Garry at Southwestern Medical School, and you can block the efflux pump – it is an ATP-dependent pump – by inhibiting ATP production or you can block it by drugs that are also pumped out by the same pump.

So these side population cells are of great interest, and I think you will see more of this in the future. They haven't yet got into the realm of cardiovascular stem cell therapy, as we have only recently realised that they are also present in the heart, and I think we will see more work on that in the future.

Let me move now to skeletal myoblasts. As I mentioned, these have been used for the delivery of undifferentiated muscle cells to damaged myocardium. They can be expanded ex vivo, they retain certain degrees of proliferative capacity, they can colonise a scar and differentiate into contractile cells, and they improve function, possibly directly or by improving compliance.

Endothelial cells have also been used. Presumed endothelial precursor cells have been injected into the peri-infarct region. They have been shown to form new blood vessels, so-called vasculogenesis; to improve myocardial perfusion and/or function; and to limit remodelling. Remodelling is a very important issue for us in cardiovascular work. What happens following an infarct is that a certain amount of tissue dies and is replaced by a scar which can't contract. So the remaining myocardium has to work harder, and the cells in the peri-infarct region undergo hypertrophy. That places increased demands; there is a deleterious effect on the Laplace relationships in terms of the wall tension, and therefore that increases oxygen demand by that hypertrophied tissue, which eventually fails. So one of the Holy Grails in cardiovascular medicine is to try and prevent remodelling. What one would like to do, of course, is to produce blood vessels, which allow perfusion of that peri-infarct region properly to prevent expansion of the infarcted area and remodelling.

Adult stem cells, as I mentioned, can be induced to proliferate and to differentiate, or to home to the myocardium. Or they can be isolated from bone marrow directly, or with cytokines, and they can be administered by various routes – I will show you some of these – either directly into the myocardium, intravenously or via the intracoronary route. There are no ethical issues versus embryonic stem cells, and of course no immunological problems because they are autologous. (So far all the trials have used autologous cells.)

The rationale for these studies comes from preclinical studies in animals, which have shown that if you use satellite cells, which are muscle myoblast cells, you get survival for 12 to 18 weeks, you get improved function after injection into a cryo-scar injury in a rabbit – so, rather than an ischaemic injury, you have a cryo-scar injury – and they show nice improvements in function, probably because you improved compliance. Rather than a stiff scar being there, you have got a scar now that gives a little bit when the ventricle contracts so that less work needs to be done, and the heart likes that.

They certainly do show some switch to fatigue-resistant, slow-twitch type fibres. But the important thing about the skeletal muscle cells, unfortunately, is that they do not form electrical junctions with the endogenous cardiovascular cells. This is a problem, as I will show you in a moment – so far the trials have shown some benefit but they have also shown the development of arrhythmias, which in some cases have been malignant ventricular arrhythmias requiring, in some instances, the placement of implantable defibrillators to stabilise the patients.

(Click on image for a larger version)

In terms of bone marrow mobilisation, cytokines have been used, initially by Orlic and colleagues. In their very exciting study, published in 2001 in the Proceedings of the National Academy of Sciences (USA), they showed that if you give these animals, after an infarct, G-CSF (granulocyte colony stimulating factor) plus stem cell factor-1 (kit ligand) you in fact get nice regeneration of the myocardium in the treated animals as compared with the untreated animals, and you reduce scar formation. You can see that survival is markedly improved in the treated animals as compared with the untreated. So this is sort of proof-of-principle that the mobilisation of bone marrow cells may be beneficial.

(Click on image for a larger version)

I won't go into the details here, but I think you can see that, compared with the sham-operated animals, which have a nice apposition between the anterior part of the myocardium and the posterior, there is a big dilated heart in the controls, which is much less dilated in the animals receiving the mobilising cytokines.

(Click on image for a larger version)

In addition, in a subsequent study from the same group, they isolated bone marrow cells and injected them into the peri-infarct region. These are lineage-negative, c-kit⁺ cells. You can see the ischaemic tissue, the endocardium, the inside and the outside of the heart and the epicardial region. In panel b you can see staining for muscle tissue. The green staining is for the cells that have been injected, and in panel c you can see the overlay: there are stem cells all the way through the myocardium. Associated with this was an improvement in cardiac function in this study.

(Click on image for a larger version)

The other cell that has been utilised is a presumed endothelial precursor cell – these are studies by Silviu Itescu and colleagues at Columbia University, NY. The interesting thing about this work is that if you use increasing numbers of these presumed endothelial precursor cells you get larger blood vessels formed – they actually showed blood vessel formation in these animals that have had a ligation of their coronary artery to produce infarction – and, very provocatively, they seemed to see an increased number of cardiomyocytes which are now undergoing proliferation. So the suggestion is that maybe one of the reasons that myocardial cells can't proliferate effectively is that they don't get enough blood supply, and if you could enhance the blood supply to them and the oxygenation of the cells, then you might stimulate regeneration of myocardial cells.

(Click on image for a larger version)

As proof-of-principle, as I mentioned, a group in France was the first to inject skeletal myoblasts. You can actually see the cells here that have grown and taken. One of the patients died some 17 months afterwards, of unrelated causes, and they were able to do an autopsy and actually show engraftment of the skeletal myoblasts.

As I mentioned before, four of their patients developed sustained ventricular tachycardia, which is a malignant tachycardia, and some of them required an implantable defibrillator.

In terms of bone marrow administration, at least eight studies have now looked at this. The routes of administration have been via intracoronary infusion, intramyocardial injection at the time of bypass, or NOGA-guided intracmyocardial injection. (I will show you in a moment what that is.) The patients have been studied either acutely after an infarct or as a more chronic group.

To summarise, from the data to date, almost all of the studies in which patients received cells after an infarct have shown some benefit. How they get the benefit, and the fact that these studies have not been properly controlled, are important caveats that we will need to address in the future.

(Click on image for a larger version)

So these cells have been given directly into the coronary artery, using a catheter.

(Click on image for a larger version)

Or they have been injected directly into the myocardium, using a NOGA-guided catheter which allows measurement of myocardial viability and endocardial wall motion. What they have tried to do is to hit those spots where there is a maximum defect in the myocardium, presumably where the maximum lesion and scarring is, and here you can see improvements in the viability and in the wall motion following injection of these cells.

(Click on image for a larger version)

I won't go through all of the trials that have been done on bone marrow administered cells, because I haven't got time, but, as I mentioned to you, almost all of the ones for acute myocardial infarction have shown some benefit, albeit that it has been fairly small.

There have been three trials of bone marrow mobilisation, and again I am not going to go through the details of them. Suffice it to say that GM-CSF has been used, and also G-CSF, to mobilise cells. One study has reported a high rate of restenosis when they gave this acutely after stenting – a procedure in which you dilate the blood vessel; they saw an increased incidence of restenosis. Further studies since that time have not confirmed this, so it is still a contentious issue as to whether or not restenosis is a problem.

We are currently undertaking a study also looking at G-CSF mobilisation in patients with what we call severe chronic refractory ischaemia, or 'no-option' myocardial ischaemia. You always start in these trials, of course, with patients who are the sickest. These people have all had several bypasses already, they have had every conventional treatment that is available, and yet they still continue to have angina, that is chest pain, due to lack of blood supply to the heart. And they have reversible ischaemia. So there is something there to treat.

What we have elected to do in these patients is to try and look at the safety of giving G-CSF alone in these patients and, if they tolerate that, to repeat the G-CSF administration and then to isolate their endothelial precursor cells from their blood – we are looking at the CD133⁺ cells – the presumed endothelial progenior cells. We then infuse those cells down the coronary arteries to see whether we can show improvements in what are presumably hypoperfused areas of myocardium. If this treatment is successful, hopefully, function will improve in these patients and if nothing else they will, hopefully, get benefits in terms of reduced chest pain.

(Click on image for a larger version)

I can't tell you the results of the study, because it is ongoing and it is blinded, but this summary may give you some idea of the severity of the disease in these patients.

The age at first revascularisation of these patients averages 41, and some patients as early as 26 have required either a bypass or an angioplasty procedure. So they are a very severe group of patients. They have, on average, 0.3 of their native blood vessels patent so they have got bad occlusions, and they take, on average, about eight to nine cardiovascular tablets to control their symptoms. As measured by an index of cardiac function here, namely the ejection fraction, they do maintain fairly good cardiac function, but they are plagued by chronic pain with the slightest bit of exertion.

(Click on image for a larger version)

This shows one of our patients who is receiving leucophoresis to isolate his endothelial precursor cells.

(Click on image for a larger version)

And it is a multidisciplinary trial. These are not easy trials; they involve a lot of people from the basic science arm – we have, as I mentioned, a collaboration with Dr Silviu Itescu, and it involves a number of cardiologists and haematologists, and nuclear medicine people to help us in evaluating the cardiac function – and then of course we have a Data and Safety Monitoring Board to keep us honest in terms of looking at the data and safety. I can tell you so far we have studied 13 patients; we have not had any severe incidents, and certainly no deaths. This is a group where you could predict that over the time of the study at least two or three patients will die, and so far we have not had that. In large part the credit for that, and for the whole trial, goes to a very talented cardiologist, now a graduate student doing a PhD with us, Dr Jason Kovacic – who is actually here today. I am delighted that he will actually be going to the Lindau Conference. He is a wonderful PhD student and this study would not be possible without his work.

The remaining issues that I just want to touch on very briefly are, firstly: hemopoietic stem cells – can they be used to enhance myocardial regeneration by transdifferentiation? I won't say much about that. It is a very contentious issue. While Piero Anversa's group suggested that they can, several other studies now from Irv Weissman and Chuck Murray have shown some very convincing evidence that, if it happens at all, it happens very, very rarely.

In addition, hemopoietic stem cells that have been implanted into the heart don't appear to be able to live very long. They die very quickly. So the promise for them I think is small.

The next question is: what are the mechanisms for improvements of cardiac function that have been observed in stem cell trials? Not a single study has really tried to address this. They have looked at improvements but they don't try and understand why, and we obviously need a lot of work there.

The other issue here is: could it be that some of the improvements are not due to increased contractility provided by myocytes but due to neovascularisation due to a supply of cytokines?

(Click on images for larger versions)