Stem cell research: the battle against Parkinson’s

The press coverage of US clinical trials in March 2001 to test a new treatment for Parkinson’s – a degenerative brain condition that affects the elderly and causes shaking and muscular stiffness – focused on the failures. It focused on the five patients who began to show worse symptoms after the treatment than before.

But were these trials really a ‘catastrophe’?

I asked Robin Lovell-Badge, head of the Division of Developmental Genetics at the National Institute for Medical Research, about the Parkinson’s trials and how they relate to the MRC’s stem cell research.

‘If you read the paper that was published in the New England Journal of Medicine’, says Lovell-Badge, ‘it actually reads as if this was a good study and that they showed overall that there was some benefit – they don’t make a big point of these five patients who had problems’.

According to Lovell-Badge, the failure side of the story hit the newspapers because ‘the second author of the paper [Dr Paul Greene] became very upset at what they had done to these five individuals’. But there were benefits to the trials: ‘Some of the patients were worse off, but there was some evidence that they knew why they were worse off – they obviously had too many cells put back in, or they had put the cells back in the wrong place.’

The techniques being tested are not new. Previous studies have produced what Lovell-Badge calls ‘quite good evidence that it could work’, with patients who were no longer responding to the standard drug treatment for Parkinson’s, L-Dopa. So what was the thinking behind the latest study?

‘The researchers for this particular paper decided to set up a more formal clinical trial to assess whether the treatment was really valid’, says Lovell-Badge, ‘because a lot of the evidence was sort of anecdotal’.

The key to the treatment is to try to get the patient’s brain to produce enough of its own dopamine – the chemical which Parkinson’s sufferers lack – by introducing cells from aborted fetuses. ‘The idea is that some of the fetal cells have to incorporate into the region of the brain around the substantia nigra, and not just be there, but make connections, and make dopamine. So you’re asking a lot of the transplanted material. In fact, from the studies done initially in Sweden and then subsequently in the States, it was a surprise to many people that it worked at all. But it did work – for some people.’

All the participants in the trials were volunteers, and all suffered from severe Parkinson’s. Both the control group and the experimental group had holes drilled in the front of their skulls, but only the experimental group actually had fetal brain material introduced into their brains.

‘The patients didn’t know whether or not they had received cells’, says Lovell-Badge. ‘The medics who were doing the treatment obviously knew, but once the operation had been done the follow-up was done in a double-blind way – separate teams, so that nobody knew who had received what until the code was broken a year later. They saw, in some of the patients, a significant recovery (1), but this was only up to a year. In many of the studies done by others the real benefit was often seen after a year, so one of the criticisms about the study was that they didn’t keep it going long enough to really assess whether it had worked.’

The results were not miraculous – but they did suggest that the treatment tended to benefit at least the younger patients.

‘After a year, it looked as if some of the patients under 60 had benefited, some hadn’t, and those who were over 60 didn’t benefit at all. So when you looked at the overall frequency, clearly the group that had received cells were better off than the group that hadn’t received cells. At that point they’d broken the code, and they told people whether or not they’d received cells. Some of the individuals who had not received cells then chose to have transplants.’

So the study seems to show that the treatment does have its benefits – but there was a complication. ‘This team changed the protocol that’s been used by other people’, says Lovell-Badge. ‘The standard procedure had been to take material from six embryos – at around 10 weeks – and to transplant it very soon after. One thing this new study did was to culture it for up to four weeks – and it’s not clear from the paper why they did this. So there may have been too many dopamine-producing cell precursors, compared with the previous studies, and that may be one reason why they had these five patients who had excessive dopamine being produced.’

It was only after the original period of the study was over that some of the volunteers began to develop problems. ‘A few of the original group started developing signs of producing too much dopamine – they had extreme jerky movements, uncontrolled movements, which are classic symptoms of being hypersensitive to dopamine or having too much dopamine.’

The rest of the control group was then advised not to receive grafts, as the researchers were worried about the consequences. The odds for these patients had changed: now they not only risked having no improvement to their symptoms – they risked the symptoms getting worse. But Lovell-Badge is not so sure the experiment should be labelled a ‘catastrophe’, although it was clearly bad for the five patients who suffered.

‘You can’t expect every experiment to work first time, you never can. What they did demonstrate was that it can work – you can get dopamine-producing cells by transplanting fetal brains into Parkinson’s patients. The problem is to get the dose right, or the number of cells right. This is a typical result in a biological experiment, because you will get some patients showing no effects, some showing a little bit, some maybe showing a good amount, and then some showing too much. That’s the sort of range you’d expect to have, so that’s what they got.’

‘The study was not particularly well-designed as they had changed protocols, but it was a fairly typical experiment’, he says. ‘It’s just unfortunate it was done on humans.’

So would it be wise in the future to use animals instead? ‘You can make animal models of Parkinson’s, and they’re pretty good, so studies have been done using embryonic stem cell derived cells to cure “Parkinson’s mice” of Parkinson’s, and it works, to some extent. From that, you would say probably that we should do a lot more research on animals.’

So why didn’t the team spend more time refining their technique on mice before putting human subjects at risk? The problem, says Lovell-Badge, is the appalling nature of the disease itself. ‘Many Parkinson’s patients are desperate to have a cure, a treatment that works, so you are going to get Parkinson’s patients volunteering to undergo this procedure because they’re desperate. And probably even now with this study, there will still be people who just decide that they would risk it.

‘It’s very hard to do clinical trials on humans in this sort of area. Take multiple sclerosis – as far as I know there’s no good animal model for multiple sclerosis. So where do you start? My argument would be that you have to start making the model, because otherwise everything you try to do will be experimental on humans.’

I ask Lovell-Badge if he fears a backlash against his work from the reaction to the American Parkinson’s trials. ‘No. If anything, it’s an argument for why we should be doing stem cell research. It has the potential to be a lot more precise. Although that trial relied a little bit on stem cells, the sorts of things they were doing were – not to be rude about it – were crude in comparison with what we’re trying to do. They were using a mixed population of cells, they had no idea what most of them were, they were taking them from the mid-brain of an early embryo, a 10-week embryo. And the brain, even at 10 weeks, is quite complex – so they’re putting back a complex mixture of cells and not really knowing exactly what they’re going to do.’

The MRC’s work on stem cells is starting from the other end of the process – looking at producing exactly the right kind of cell to be used in the treatment of Parkinson’s and other diseases. ‘The sort of things we’re trying to do are much more general, so I hope we can have a benefit for Parkinson’s, but that would just be one part of it.

‘In these studies they’re taking cells from the mid-brain, because that’s the region where you expect to get cells that produce dopamine. But it’s possible that you could take stem cells from elsewhere in the nervous system and they would also, in the right treatment, give cells that make dopamine. It’s known that if you take embryonic stem cells (2), these cells that come from a blastocyst stage early embryo, that several ways have been established to allow these to differentiate, to specialise, into becoming first of all stem cells of the nervous system, and then mature cell types. And you can show quite readily that they will make cells that make dopamine (3). Good studies have been done in mice in several labs.’

Lovell-Badge hopes his work will single out the most useful cells for using in the treatment of disease. ‘This is not using aborted embryos, this is done by cell lines. We are studying that sort of approach to try to understand what you can use embryonic stem cells to do, and we’ve developed some ways of selecting out these precursor stem cells for the nervous system as they are forming.

‘So rather than having 50 percent of cells in a population which do that, and 50 percent which do other things, make muscle or whatever, you can end up with 100 percent of cells that just specialise in making the nervous system. One idea is to find the conditions for again selecting out a pure population of cells not just to make dopamine-producing neurons, but to make all sorts of cells that might be useful for spinal cord injury.’

‘The aim of a lot of the stem-cell research is to have defined cell types that you use to cure individuals’, says Lovell-Badge, ‘where you can take a specific set of cells of a known, recognisable type, a known number of them, and you know what they can do, and then you put them back.’

This offers the real possibility of administering a specific dose of cells. ‘Yes – much more rigorous. If you can control the number of cells you’re putting back and you know what they are, then you’re much more likely to get a balanced therapy for the individual. You can even say that for an individual who has bad Parkinson’s you put back more cells, and for one who has less bad Parkinson’s you put back fewer cells.’

So in the future, the problem of transplanted cells over-producing dopamine could be avoided. But the potential of such artificially produced cell cultures offers many more possibilities than simply tidying up the weaknesses of existing techniques.

‘You could take it even further because, particularly if you’re starting off with embryonic stem cells, we know how to manipulate those, in terms of manipulating their genes. So you could, in theory, engineer the cells so that you have control over the amount of dopamine they actually make. It could be a crude level of control, so if you ended up with a patient who was making too much dopamine, you could then give them a drug which would turn it off. Or you could have more subtle control, where you could actually, by giving another drug, modulate the level very easily.’

Basically engineering the cells to respond to a third-party drug? ‘Exactly. A better way than just giving L-Dopa. It would be much more specific. It would be like the choice between curing your car with a sledgehammer or curing it with a set of precision tools.’

Robin Lovell-Badge is speaking at the Maverick Club on 8 May 2001. For more information email Jan.Macvarish@MaverickClub.com.

Read on:

Let stem cell research begin, by Toby Andrew

Animal research: a scientist’s defence, by Dr Stuart Derbyshire

(1) There were various ways of measuring whether the treatment was reducing the severity of the Parkinson’s symptoms, but the trials were complicated by the fact that the patients were still receiving L-Dopa treatments – so any benefits from the transplants were potentially hidden by this. The researchers found that the best time to assess the patients for effects of the cell transplantation was early in the morning, before the patients had taken their morning drugs, and 12 hours after they had taken their evening drugs.

(2) Cell lines (taking blastocyst stage embryos, which in the mouse is at three days old and in humans at five days old) are made when the embryo consists of about 100 cells in total. There are just two cell types in the blastocyst at that stage – the outer cells (which give rise to part of the placenta) and the inner cells (which give everything else, including the whole embryo plus other cells which contribute to the placenta). Then embryonic stem cell lines can be established from these inner cells. These are abbreviated in the literature to ES cells – embryonic stem cells. They are different from stem cells taken from embryos.

(3) ES cells have the ability to form any cell type in the body – even, in theory, ‘in the dish’, in culture. Researchers can change the way the cells grow, and around 50 percent of the cells will start going along the pathway to give nervous system cells. They can cultured in different ways, with different factors added, and they will make heart muscle or blood cells. If they are treated in a way that favours their specialisation into cells of the nervous system, researchers can then take some of these cells and change the conditions again and get different types of nerve cells. In a recent study, researchers found a new set of conditions which essentially allowed them to get from embryonic stem cells to dopamine-producing cells with one treatment over a couple of weeks.