Professor Rohit Bakshi on iron deposition in the brain.(Transcription)Ferrara. September 23, 2010.

Why a transcript? Because Bakshi's lecture is very interesting for people who want to learn more and feel the need to deal with the hard "scientific stuff". The transcript was the basis for a German translation we wanted to provide for that matter.

The sources are 3 videos from Fondazione Hilarescere. There is also a video on youtube, which joins these 3 videos together. (The time-stamps in brackets refer to this one.)The transcript has been reviewed by Prof. Bakshi - thank you for your corrections.

First Videohttp://www.fondazionehilarescere.org/video/Bakshi%20_CCSVI%20_01.avi

[Prof. Paolo Zamboni welcomes the audience and introduces Prof. Rohit Bakshi.]

3:04

Good afternoon, everyone. Prof Zamboni, thank you for that kind introduction. It was very nice to have you visiting us in Boston about a year ago and start to exchange ideas together. Your work is known throughout the world, we know a lot about your work in Boston. It is a subject of lot of discussion, it's really provoked our thinking in MS. I want to congratulate you on bringing these ideas to the forefront, and I am very much excited about the opportunity to be in this beautiful city. Thank you so much for inviting me and giving me such nice hospitality, and I'm looking forward to our discussion today because I think we are going to see a common theme in these discussions: that we are starting to think about MS differently than we did, say, 15 or 20 years ago. And that's very exciting for all of us.

3:48

My focus is going to be on iron deposition in the brain. And I'm going to try to give you an overview, first, of how we image iron using MRI techniques regardless of what disease we are looking at. If we are looking at normal aging, if we are looking at MS, or if we are looking at other neurologic diseases. I'm going to talk about multiple sclerosis for most of my talk, and we are going to hone in on what iron looks like in MS, how we image it with MRI, and what is the clinical relevance of the iron we see in the brain. So how does it relate for example to disability that patients suffer.

Towards the end of my talk I'm going to try to clarify what our current thinking is on the pathophysiologic role of iron. So, does iron contribute to damage in the brain in MS patients, or is iron simply a result of the disease, and serves as a biomarker but not a mediator of neurotoxicity? We'll debate that towards the end. And of course, if iron has a role in contributing to damage in MS, then that would open up potential new therapies that we could combine with our immune-mediated therapies to try to help MS patients. So, this is all very exciting.

4:59

So, let's start with the basics. The normal brain contains metallic ions. And iron is very important for normal neuronal metabolism. So, all of our brains have iron present, especially in gray matter structures. And iron can be picked up by MRI scanning because it has magnetic properties. So it shows up very easily on MRI scans. Most of the iron that's present in the brain is in a mineralized form which we call ferrihydrite, in ferritin or inside hemosiderin. MRI is a very robust and sensitive tool to be able to assess iron in the brain.1 This is especially true when you use ultrahigh field imaging like 3 Tesla, 4 Tesla, and 7 Tesla. But you can also pick up iron deposits at 1.5 Tesla MRI field strength.

5:53

This is a nice correlation that you see typically between the histology and autopsy identified iron, and MRI defined iron. So on the left is a Perls' stain2 of a brain, and you can see the substantia nigra and also the red nucleus3 contain high levels of iron. This is a normal finding with aging. And on the right you see the T2 hypointensity present at 4 Tesla. In the same areas we have iron deposition. So, iron manifests on T2-weighted images as causing shortening of the relaxation time which leads to hypointensity. It shows up even better when you use a T2* technique, which is a gradient echo form of T2 imaging. This is much more sensitive to picking up iron.

There are many ways you can quantify and assess iron. This ranges from simple techniques that are readily available on pretty much all MRI scanners to more complex techniques that require special sequences. So, you can simply quantify iron from single spin echo studies by looking at the normalized intensity on T2-weighted images, you can go to a dual spin echo, and you can generate relaxation times and relaxometry curves. You can use specialized sequences to pick up relaxometry like the CPMG sequence. Then you can go to T2* to get more sensitivity to pick up iron deposits, and then finally there are other newer pulse sequences that are hybrids that allow you to measure T2 and T2*.

7:22

On the right I'm showing you some of our work from Brigham and Women's Hospital in Boston. This is a technique that allows you to do 96 echoes with a T2* technique, and this is performed at 3 T. And I'm showing you examples of relaxometry curves that we generate in gray matter structures of interest, the cortical gray and the basal ganglia. You can see how nicely we can map the relaxation times, and therefore assess iron in these tissues.

As far as quantifying iron and getting actual data from scans, the simplest way to do this is by measuring intensity from T2-weighted images. You only need one echo to be able to do this. You could also measure intensity from T2* studies. But a better way to do this, a more sensitive, more accurate way is by doing formal relaxometry studies. And there are several relaxometry values you can measure from MRI. The first is the basic R2, which is from a spin echo based study. Then we have R2*, which comes from gradient echo studies, and finally we have R2', which is the difference between R2* and R2, which is thought to represent the iron dependent signal that you see in the R2* image.4

8:37

And remember: R2 is just the inverse of T2 relaxation time. R2 is a rate, and T2 is a time. But they are interrelated completely. Other advanced techniques are listed here in the slide. I won't go into these in a lot of detail, I just want to mention at the very bottom: SWI. Susceptibility weighted imaging is a newer technique we have available, which is especially helpful at 3 Tesla and higher field strengths – I'll come to that later in my talk. This is not yet a standard sequence. But if it's available it's very helpful to measure iron.

Now, before we talk about diseases that cause increased levels of brain iron we have to understand what we see with normal aging. What's the normal iron? That we should see first. Well, first of all, at birth there is little or no iron in the brain. But it gradually starts to increase even in childhood. So, by the time a person is, say, 15 or 20 years old there are very high levels of iron in the globus pallidus, in the red nucleus, and other structures such as the substantia nigra.5

And then as aging goes forward into the seventh decade, in the middle of adulthood, we see iron in other structures. These structures accumulate normal levels of iron. That includes the putamen, the caudate, and yes, even the cerebral cortex has normal iron present as people age. There are structures that still remain low even very late in life, and don't contain a lot of iron with normal processes. That includes the thalamus. And you should remember that as we go forward, because in my talk today you're going to see that the thalamus is one of the structures that accumulates a lot of iron in MS patients. So that is always going to be abnormal if you see thalamic iron.

10:17

Here is an example of what happens to the MRI scan as we go through normal aging. Starting first with a person who is 43 years old, we are now going up to a 60-year-old, and then an 84-year-old, and notice the gradual increase in T2 hypointensity, which I already mentioned to you is the marker of iron deposition. So, for example I mentioned to you that the putamen starts to accumulate iron in the fifth, sixth, and seventh decades of life, and notice that early on we don't see much T2 hypointensity. It starts to increase and gets very dramatic as a person gets into her ninth decade of life. Nonetheless, the thalamus remains free of iron deposition relatively, even in a person in her ninth decade of life. It remains relatively iso-intense, it doesn't have a dramatic T2 hypointensity.6

11:06

Now, let's consider some neurologic diseases where we know that iron is increased. And there is a wide variety of these diseases. I'll start first with PKAN, this was formerly called HSD.7 This is a movement disorder – shown here in the MRI scan that is pointed to in the slide – which gives us a characteristic eye-of-the-tiger sign on MRI scans. The idea being that these look like two eyes looking at you. They are in the globus pallidus. They are centrally bright on T2, and the reason for that is we have gliosis present, we have necrosis of the globus pallidus, which raises the T2 relaxation time. This is the opposite of what we see with iron, which is now seen in the periphery of these lesions. This is T2 hypointensity surrounding the centrally bright areas, and this is correlated with iron deposition in this disorder.

12:01

But there are a wide variety of other neurologic conditions, especially chronic nerve degenerative disorders where we get accumulation of brain iron. That includes Alzheimer's disease, Parkinson's disease, and Huntington's disease, and even ALS – motor neuron disease.

Just some quick examples of this. This is work from Albany, New York, where they use a 3 T scanner, and they measured actual relaxation times, which I had mentioned to you is the preferred method for accurately quantifying iron.

And the point of the slide is when they look at the hippocampus of patients with Alzheimer's disease they see a shift in the relaxation curve, which is shown in blue here in Alzheimer's patients, and pink in normal controls. And you notice that the T2 relaxation time has shifted towards shortening of T2 in Alzheimer's disease. The peak height is less, and there's a shift to the left. So this is substantial evidence that there is iron deposition occurring in Alzheimer's disease. And Alzheimer's is generally limited to a few areas of the brain, it's not widespread like we see in MS for example. The hippocampus is the most prominently involved.

13:09

Here is an example of a Parkinson's plus syndrome. This is cortical basal ganglionic degeneration, which presents like Parkinson's disease but is not idiopathic Parkinson's disease. And notice the markedly abnormal hypointensity in this person who is only 45 years old. I've already shown you that you don't see this much hypointensity at that age, and if this person was 90 years old that would be normal to have that much iron in your putamen and in your globus pallidus. But this is markedly abnormal.

This is a substance-abuse condition called toluene abuse, which is caused by sniffing glues or inhalers, or paint thinners for example. This causes intoxication, and when it's done chronically over several years this causes severe damage to the brain. So this is a person who was abusing toluene, and has marked T2 hypointensity of the thalamus and also the basal ganglia. This is most likely iron deposition, but it turns out that toluene itself also causes T2 shortenings. So, both processes may be causing this.

Notice in multiple sclerosis in this example there is T2 hypointensity of the bilateral thalamus, and I have mentioned to you that that is always abnormal if you see that. And that's shown right here. There is also hypointensity of the globus pallidus and the posterior putamen, which is more than you would see in this … if you compare this to an age-matched normal control to the left. You see the dramatic difference between these two cases.

14:34

So now let's turn our attention to multiple sclerosis. Where do we see T2 hypointensity? What is the time course for when this occurs? What is its relationship to disability and cognitive impairment? Well, as a starting point several of us have been studying this for probably 10 or 15 years. The first word came from Burton Drayer and colleagues.8 They made the first identification of this finding in a study of 47 MS patients, 25 of whom had darkening on T2 of the thalamus and putamen.

The group in the Queen's Square was able to identify T2 hypointensity in the thalamus.9 And then we started our work about 10 years ago. And in our first study of 114 patients10we showed darkening of the basal ganglia, the thalamus, and the Rolandic cortex on T2-weighted images, which was related to clinical findings. This was associated with the level of disability, this was associated with the stage of MS. The more severe MS cases had more evidence for T2 hypointensity. This is also been shown throughout the world in other cohorts. If you look at Japanese patients with MS or Mexican patients, they also have T2 hypointensity. And since these initial observations there are probably more than 10 groups now that have verified these findings, including work done right here in the EU.

15:50

So, that is the first set of observations: simply looking at images and identifying T2 hypointensity. But – that is not a robust or sensitive way to monitor and accurately quantifying the level of iron. So we quickly moved into quantitative studies. And this was our first quantitative study in 2002, where we showed T2 hypointensity in MS patients, and I'm showing you 3 MS cases on the top.

Each set of pair images is a different patient. And on the bottom you're seeing an age-matched normal control. These patients are in their early 50s. And you can see the T2 hypointensity involving the structures I've already mentioned. The midbrain is also involved in MS. We see T2 hypointensity of the red nucleus for example. And sure enough, when we did quantification using normalized intensity from these images – and these were standard clinical images by the way that you could do routinely on all scanners – we were able to show anywhere from a 4% to 11% decrease in intensity – or T2 shortening – in these tissues, which were significant between MS and controls.

We then were are able to link the T2 hypointensity with cognitive, and ambulatory, and physical impairment in MS patients. This is a follow-up study where we looked at T2 hypointensity in the cerebellum, and the patient's ambulatory status. So, a normal control has very slight amounts of iron deposition in the sixth decade of life in the dentate nucleus. You can see this is not very hypointense on T2.

17:29

Here is a patient with MS with mild ambulatory impairment. The normal walking should be about 5 seconds, and this takes the patient 8 seconds to walk 25 feet. And you see the slight T2 darkening compared to a normal control. On the far right you see an extreme example of a lot of T2 hypointensity, presumably this is high levels of iron. And this patient is a very slow walker. This patient takes about three minutes to walk just 25 feet.

Importantly in this study and throughout the studies that we performed, the T2 hypointensity was the best predictor of the ambulatory impairment, only put it side-by-side with established conventional MRI markers of MS.11 And let me take you through that. For example, I didn't go into this but I'm sure you're all familiar with MS causing bright lesions in the white matter, which we pick up by FLAIR or T2. These are called plaques, and we quantified that in these MS patients, and the data is shown here. We also quantified T1 black holes, which are dark areas on T1-weighted images, which show areas of axonal loss.

We also quantified whole brain atrophy, that's called BPF12. This is a measure of total atrophy in the brain, which is an important part of MS. And finally we measured centrally predominant atrophy by looking at third ventricular width. This gives us a good assessment of atrophy around the ventricles and in the basal ganglia. And the take-home message from this data set is that the best predictor showing the lowest P-value13 and the highest correlation with ambulatory impairment was the T2 hypointensity in the dentate.

19:06

You can see the R-value14 here was .46, which is fairly moderate to strong. This is the actual simple correlation we saw in that sample. The time 25 foot walk is plotted to the left, and the T2 intensity of the dentate is plotted on the x-axis. You can see the negative correlation. The darker the T2 is in the dentate, the worse off the patient is on walking. And again, this was the best predictor of the walking ability when we compared all the other MRI variables.

A similar kind of study15 showed that T2 hypointensity is related to cognitive impairment in MS patients. Cognitive impairment is a very important part of MS. It's present in probably half or 60% of MS patients. So again, just like the previous slide I showed you… normal control, this is a 42-year-old woman, does not have much in the way of T2 intensity changes, and we measured that quantitatively down here.

Here is a patient who has mild cognitive impairment. The neuropsychological score should be 100, and is dropped to about 88, and this patient has very slight T2 hypointensity, which you really don't see very well in the actual images. But with the quantified approach we can see the drops compared to the normal control, particularly in the thalamus and putamen. And then finally, this is a patient with relatively severe cognitive impairment. The neuropsychological score has dropped about three or four standard deviations compared to normal, down to 64. This is a patient who would have obvious cognitive impairment, and you'd expect that will be affecting her ability to be employed, her ability to interact with her caregivers and her family members. This would have a significant impact on her life.

20:44

And in this case you can see dramatic T2 hypointensity that you don't really need to quantify to be able to detect it. So I just point out to you, in the thalamus there is T2 hypointensity in this patient, and also throughout the globus pallidus and the posterior putamen. And sure enough, just like in the previous study we compared side-by-side the T2 hypointensity variables to the established conventional MRI measures of disease severity. So we look at T1 black holes, T2 hypo-intense lesions, or whole-brain atrophy. Again, you can see the T2 hypointensity variables are the winners here. They show the best correlation, the highest correlation with cognitive impairment. One of the most striking correlations we saw was with the globus pallidus, P-value .001, this is a moderate to strong correlation.

A quick side note to this: as you may wonder why with the globus pallidus would have an impact on cognition, but you probably can recall from your experience in Parkinson's disease when patients get pallidotomies done for movement disorders like Parkinson's disease and other movement disorders, one of the side effects of that procedure is cognitive impairment. So clearly the globus pallidus does have a role in cognition.

21:54

How about T2 hypointensity predicting future changes in a person's disability? Up until now I've only shown you cross-sectional data comparisons, but it becomes very important to try to understand whether iron deposition actually predicts and precedes neurologic changes in MS patients. In order to embark on this type of study we got launch to new data from our collaborators in Cleveland, Ohio, and also from Denver, Colorado. This is a collaboration with Drs. Rudick and Fisher from Cleveland, and Dr. Jack Simon also helped us with this study.16

The simple question we asked here was: in patients with very early-stage MS who had relapsing-remitting MS for a disease duration of only six years, is there any way we can predict how much brain atrophy they are going to develop in the next two years? And we did a regression modeling statistical approach, and we showed that T2 hypointensity was the best predictor of the subsequent two-year rate of brain atrophy, after adjusting for the effects of conventional MRI measures. So T2 lesions, T1 lesions, Gd-enhancement at baseline, and also whole-brain atrophy: none of those measures significantly predicted the change in subsequent atrophy to the extent that T2 hypointensity could. And you can see that the final model shows that T2 hypointensity predicts about 33% of the variants in subsequent whole-brain atrophy. So it clearly has a link to atrophy.

23:27

Here is a representative case, this is a patient who does not have any T2 hypointensity. The normalized intensity looks good, there are no changes in the scan, and two years later the BPF is stable. There is no atrophy of the brain. This patient has baseline T2 hypointensity of the thalamus and globus pallidus, and then two years later this patient loses about 5% of her brain volume.

Another way we can get at the question is to whether iron deposition actually contributes to damage, besides just looking longitudinally, is we can look at patients with very early-stage MS to see if they have it. Because if its only a result of the disease we are not going to see much of it early on, it's only going to come in later stages of MS. We just completed a collaboration with Milan, Italy, Dr. Filippi and Dr. Ceccarelli, and we looked at patients with clinically isolated syndromes, which means they are having their very first attack of demyelination. They don't have full-blown MS at this stage. And we have showed in this study17 that T2 hypointensity was present even in these early-stage patients. We were able to detect T2 hypointensity in the head of the caudate in patients with clinically isolated syndrome. You can see .53 in controls, and .51 in the CIS patients. So, T2 hypointensity is very, very subtle in the earlier stage of MS. But it is present, and it can be detected.

24:53

However, I don't want to give you the impression that iron deposition explains all of the clinical disability and all of the severity that we see in MS. There are probably many other factors. For example, this is a study that surprised us, so we recently completed it with our collaboration in Milan. Which is that patients who had benign MS, which means they have had the disease for a very long time like 15 years, yet, they don't accumulate significant disability. These patients also have T2 hypointensity. So here is a normal control, here is a patient with totally benign MS who has had the disease for a very long time. This patient has substantial T2 hypointensity in the globus pallidus.

So, what I take away from this is that T2 hypointensity and iron deposition is only one factor that has to interact with many other factors before a patient will get dysfunction. I think we need to keep that in mind as we go forward. Now, there are several other groups that have picked up on this work and have taken iron imaging really to a new level in MS patients, and have verified the early work that has been done, but have expended on that work. This is work from the group Dr. Zhang and colleagues working in Calgary in Canada. And they did a very beautiful study where they took MS patients and scanned them at 1.5 T, they also scanned the same patients at 3 T. The idea being that 3 T… the higher field strength is more sensitive to iron deposition and should give us potentially better clinical MRI correlations. And that's exactly what they found. So, at 1.5 T the correlation with EDSS score, which is a measure of disability, was a R2 of about .2, where as you can see they get a doubling – more than a doubling – of the R2. They get a much better correlation when they look at T2 hypointensity at 3 T.

26:40

So, this would go along with, number one, the notion that the T2 hypointensity is related to iron because it's field dependent. At higher field strengths we see more of it. And this would also corroborate the fact that this is clinically relevant, that this is a marker or potentially a mediator of additional damage in the brain. These are some of the correlations. I won't go into these in great detail, but the bottom line is that at 3 T you see better correlations. This is work done right here in the EU. This was done in Austria, recently completed by Dr. Fazekas and colleagues18, and I was very pleased to see that the work that they have been able to expand on was using R2* imaging. And I mentioned earlier that this is more sensitive to iron, and it works very well especially at 3 T. So here is an MS patient showing you the anatomic T1 weighted image, which was overlaid with the R2* image. And you can see in this MS patient there are dramatic R2* changes that are consistent with iron deposition in the basal ganglia.

27:39

And they were able to show that in the study that patients with relapsing-remitting MS have more significant R2* changes than patients with clinically isolated syndromes who have earlier stage and more benign MS. You can see the significant P-values are indicating that higher levels of iron are associated with more advanced stages of disease. We are continuing this work at our center, this is our work in progress on our 3 T magnet, and we are going to be completing the study very soon, which is to look at R2* and T2* imaging to generate new types of images in MS patients that we hadn't seen several years ago. Images like this where you see an R2* map, and you see all these dramatic changes in the gray matter, which are probably iron related in an MS patient compared to a normal control. Now, I want to spend a few minutes on susceptibility weighted imaging because this is going to be the future of iron deposition, this and also moving to 7 T to quantify iron in a more sensitive way.

28:45

And I want to thank Drs. Haacke from Detroit, Michigan, and Dr. Zivadinov from Buffalo for providing these slides. Susceptibility weighted imaging is also called SWI. It uses tissue magnetic susceptibility differences to generate contrast, which is different from our standard T1, T2, and T2* images. That's a totally unique way to image the brain. It uses a combination of magnitude and phase images from 3D images that are ideally taken at 3 Tesla or higher field strengths. And I'm not going to go into all the physics behind this, I would just ask you to read about this on your own. But the bottom line here is that we are able to take a combined magnitude image plus a phase image and generate an miP image, which gives you your susceptibility weighted image. That's on the far right of the scan. And you can see one of the major advantages of SWI, it's very sensitive to iron. So it picks up blood in veins quite beautifully, and gives you a very nice depiction of the venous vasculature.

29:48

It's incredibly detailed, noninvasively with this technique. Here is an even more dramatic example. This is a courtesy of Dr. Saloman. This is done at 4 Tesla, and again you are just struck by the beautiful vasculature that comes out, that we have not been able to pick up in the past with standard imaging techniques. But SWI is not only sensitive to vascular contained iron, but also deposits of iron in the parenchyma. And of course it's much better if you use higher field strengths. So, here are 4 T SWI source images, both magnitude and phase, compared to 1.5 T, and you can see at 1.5 T you really can't do high quality SWI scanning. You really need to go to 4 T to get the signal-to-noise that you require. And very nicely you can pick up iron deposition in the red nucleus and the substantia nigra with this technique. Now, SWI has shown us that iron deposition in MS is not just limited to the gray matter. It had been suspected in the past that the white matter was also involved, but SWI is showing us this in a very robust way that we hadn't seen in the past. So here are some examples from the group in Buffalo where they showed three patterns of iron deposition in the white matter of MS patients. There is a ring-like pattern, and notice again, the T2 hypointensity that we saw with standard images now translates to hypointensity on these SWI scans. This is iron deposition. We see a nodular pattern inside MS plaques, we can also see a scattered amount of iron, which looks more patchy.

31:25

All of these deposits we see in white matter usually are associated with plaques, the traditional MS demyelinating plaques, which have edema inflammation and demyelination, but they also contain iron deposition. And as you would predict SWI can also show us what we already suspected at 1.5 T, which is the gray matter iron deposition I've shown up until now. This is work from the group in Buffalo. A normal control is shown on the left. Now, these are color-coded SWI maps, which gives your eye a better way to detect these changes. You can do this fully quantitatively with post processing. And I want to draw your attention to the fact that the patient shown here on the right with MS – we have four sets of images – you can see these dramatic bright areas, which indicate high levels of iron on this color scale compared to the normal control. This is throughout the globus pallidus, this is throughout the caudate, this is also creeping into the putamen in these cases.

32:27

Taking it a step further, we can also do iron quantification at 7 Tesla. And you may have some 7 T magnets here in Italy. We have several throughout the US. There is one in Boston. This work I'm showing here is from San Francisco.19 And Dr. Pelletier and colleagues have done some beautiful work using SWI at 7 T, and they are getting even more robust depiction of this iron deposition. So, here is an MS patient again. This is color-coded, the blue on this color scale indicates high levels of iron. And the normal control only has minute amounts of iron. And again, you get iron with normal aging, so this is not a surprise that there is iron present in controls. But there is much more in the MS cases, and here are the P-values that show that. You typically don't get iron deposition in normal appearing white matter. So here is the corpus callosum, which is free of lesions in these cases, and it doesn't show any significant iron deposition. But when Dr. Pelletier and colleagues looked at lesions in the white matter with 7 T they again were able to show what you saw at 3 T, which is you are able to see iron deposition associated with MS plaques.

33:33

So, here is another example. This is T2 hypointensity in the rim of a MS lesion. This is a central vein, so this is intravenous iron, this is not parenchymal iron. And then we have also some diffuse hypointensity at the edge of this T2 bright lesion in an MS patient. Now, when we take this information and we go back to conventional images we are reminded by the fact that sometimes we see evidence for iron deposition in MS white matter. Here is an example of T1 shortening that you can see with your conventional T1 weighted scans. This bright rim of hyperintensity on non-contrast T1 weighted images is seen fairly commonly in MS patients. You might see on average one of these lesions in every one or two of your MS patients.

This is associated with iron deposition most likely, it corresponds nicely to what I showed you with the phase imaging work from UCSF. And when we then quantify these T1 hyperintense lesions we see they are more common in secondary progressive MS patients than in relapsing-remitting, and the secondary progressive patients have more severe MS. They have more disability, they don't respond well towards standard therapies. And sure enough, they have on average more of these T1 hyperintense lesions. So, this is additional data telling us that iron deposition is associated with more severe clinical disease.

34:55

So do we have any histologic proof that all these changes that I've shown you on MRI actually indicate iron deposition in MS? Well, there now is mounting data that's being accumulated both in humans and also in animal studies. This is work that we did in collaboration with Case Western in Cleveland and also the University of Maryland in Baltimore. And we showed in MS patients who came to autopsy, who died from MS or other causes, that the T2 hypointensity we saw in life in these patients - like here in the thalamus in this person - was associated with iron deposition we could pick up by advanced Perls' stains in the autopsied thalamus itself. So these are iron deposits in the thalamus. They correspond to T2 hypointensity. I don't think any of this is really surprising, because we know from all these MRI techniques one after another, they all are pointing in the direction this is going to be iron.

So what are the implications of these findings? Is this a marker of more severe disease, or does iron contribute to the pathophysiology of MS? Well, this all becomes very important as we think about MS from a new perspective that we didn't have about this disease, say, 15 years ago. Which is that MS has a dramatic effect on the gray matter, it's not simply confined to the white matter. And here is an example of how the brain gets affected metabolically by the MS disease process. This is a glucose PET scan based on a study we did in the 1990s.20 And you can see that normal people have very high metabolism

Second Videohttp://www.fondazionehilarescere.org/video/Bakshi%20_CCSVI_02.avi

(36:26)

in gray matter. They take up a lot of glucose. They have high metabolism in gray matter structures. MS patients have hypometabolism, decreased metabolism of their gray matter. This is commonly seen in MS. You can see all the widespread loss of a normal color in these gray matter structures, even the visual cortex is not as bright as it should be, the deep gray nuclei like the caudates are not as bright, not as metabolic.

So we have a disease process where we are getting structural changes in the gray matter which we pick up like T2 hypointensity, we also get hypometabolism in the gray matter suggesting that it's dysfunctional. The gray matter is not working like it should. And this has major clinical implications. For example, fatigue is a major symptom in MS. 70 to 80 percent of patients with MS have fatigue, and it doesn't correlate well with our standard MRI measures. But yet, Dr. Filippi has shown that it correlates with hypo-activation of the thalamus.21 This is highly correlated with the greater likelihood of having fatigue. So this shows a clinical relevance of this gray matter damage.

1:05 (37:32)

Also the gray matter is a major target of the atrophy that we see in MS. So I've already mentioned to you that brain atrophy occurs in MS, it's really common, it occurs early in the disease course, but atrophy of the brain is driven disproportionally and selectively by dramatic gray matter volume loss. Here is an example of a reconstruction study22we did a few years ago where we looked at the caudates, which are hypometabolic as I showed you already, which contain iron deposition as well. I've showed you several examples of that – and the caudates also undergo a substantial amount of atrophy.

So here are five pairs of MS patients, shown in green. They are co-registered to normal volunteers while age-matched to those patients and you can see there is about a 20% loss of the caudate volume in these MS patients. Their whole brain is only about 6% smaller. So they have threefold or fourfold disproportionate atrophy of these deep gray structures. The thalamus is not spared as well, the thalamus undergoes substantial atrophy in MS. It's another site where we see hypometabolism and iron deposition. And in this study we show the thalamic atrophy was highly correlated with cognitive impairment in MS patients. You can see the R-values here are fairly strong.

2:25 (38:52)

And this is further proof that the gray matter is likely to be involved early in MS. This is some of our work in progress we are going to show in Sweden, coming up at the MS meeting.23 These are patients with early-stage MS imaged at 3T, and we did a voxel-by-voxel comparison of atrophy, and in red you see the areas of the brain that are most susceptible to atrophy, even in this early stage of MS. The thalamus undergoes a lot of atrophy, the caudates, and also the putamen. So again, this raises a red flag if you think about a place where there is a lot of damage occurring early in MS, it's affecting the patient's functioning like their cognitive function, maybe also their fatigue, and we also see structural changes like iron deposition. And we also know that the current way we target MS treatments is by targeting the immune system.

So this is a study, a very up-to-date recently done study comparing interferon therapy with glatiramer acetate.24These are two of the mainstays of treating MS as a first-line. And even with the best available therapies patients continue to have brain atrophy. So, just focus your attention on the second year of this trial. Both of these drugs have a partial effect on reducing atrophy, but the patients have ongoing, continuing brain atrophy. So the point is that our current treatments are not complete, and are not as effective as we would like in stopping the neurodegeneration in the patients' brains.

3:58 (40:25)

So perhaps there are other mechanisms of damage that are not being targeted by our current therapies. And one of those mechanisms may be iron mediated toxicity, which is a… probably a secondary result to the disease process initially, but then becomes a contributor and causes ongoing damage. And what is some of the evidence that iron may have a toxic role? Well, first of all we know that cell membranes are very rich in polyunsaturated fatty acids. And these are unfortunately susceptible to lipid peroxidation. And iron may be a source of free radical oxidations, especially when they are present in high enough levels, which are then going to induce lipid peroxidation through a mechanism known as the Fenton reaction.

And here is the Fenton reaction.25 The idea is that ferrous iron - Fe2+ - combines with hydrogen peroxide and generates ferric iron - Fe3+ - plus the end product of hydroxyl free radicals. These free radicals now go into further cascades and cause more free radicals to generate. And this leads to a vicious cycle where there is more and more free radically generated damage as you get more higher levels of iron, and as the brain gets dysregulated and hypometabolic, it's not able to clear the iron properly, so iron continues to build up in that tissue.

5:28 (41:55)

Is there any evidence that iron has this toxic role in MS? Or is this all theoretical? Well yes, there are several studies now that are pointing in that direction. First of all, iron deposition has now been shown to occur in animal models of MS. So these are the classic models like EAE26 or the Theiler virus model27 for MS. We can see changes in those mice that suggest early ongoing iron deposition. I already showed you evidence that iron is occurring in MS. There are lots of MRI and pathologic data for that. There is an interesting study showing that iron depleted mice, who are nutritionally depleted of iron, are more resistant to developing the full MS disease process.28 When you try to induce them, then mice who have normal iron in their diet.

So iron may be a co-factor for the pathogenesis of these animal models. And we also see evidence that oxidative stress has been detected in MS gray matter. That's been done with MR spectroscopy. So let's go through some of the details of this interesting data that seems to show a pathophysiologic role for iron. First of all, T2 hypointensity has now been co-localized with iron deposition in a mice model of MS. This is from the University of Kansas. This work is still in progress, but you can see these mice were induced with MS, and they have these T2 hypointense areas, which nicely correlate with Perls' staining at autopsy. When you zoom in you see that these are clearly are areas of iron deposition.

So it parallels what we see in the humans. We see iron deposits also in the EAE model, which are shown here. There are examples of this from work done at the University of Kansas. We can also now see with 7 Tesla MRI that there is evidence for oxidative stress in MS patients. This is particularly true in the gray matter. And you can see for example decreased glutathione picked up by these MRI spectroscopy techniques. And this is the finding right here showing the decreased glutathione peak. This is a marker of oxidative stress and would seem to link potentially to the iron mediated damage that I proposed.

7:35 (44:02)

How about treatment, how does that affect the MS disease process? Well, this is data from the group in Kansas where they show that iron chelation therapy – this happen to be desferriox, I mean, an older iron chelator – improve the outcome of MS given to animals.29 This is EAE, and you can see the patients who got treatment… This is the desferrioxamine on the bottom… They have a more rapid recovery with EAE, they don't have as bad of damage with EAE, they have a more limited form of EAE than patients who don't get the desferrioxamine.

There is evidence to suggest that there are therapies available that may impact on iron deposition in other neurologic disorders. And maybe we can learn some lessons from this in MS. This is worked on in Friedreich's ataxia. This is a human study in children. These are nine children who got an oral chelator which is newly available called deferiprone, and I'm showing you that their iron measured by R2* shows evidence for decreasing over the next several months. Not all the patients responded, but there were several patients who clearly had reduced iron, which could be measured by MRI on an ongoing basis - these are examples I'm pointing to here - after they were given this iron chelating therapy. This was safe to administer for six months to children, and their clinical scores also got better.

So they had ambulatory benefits from this type of therapy targeting iron. And they showed removal of iron by imaging techniques. So, if iron has a pathophysiologic role and it's not just a marker of neurodegeneration there will be a wide range of new therapies we could consider in MS patients to combine with our partially effective available immunotherapies. This would include metal chelators, free radical scavengers, and antioxidant therapies.

9:27 (45:54)

So, to tie the thing together I would like to summarize what I've tried to say to you. So MRI is a very powerful tool to assess brain iron. We have a wide range of techniques available, ranging from simple ones you can do at clinical scanners with your T2, your T2* methods, ranging all the way to more sophisticated methods like SWI and R2' imaging, which are specially effective at ultrahigh field strengths. Brain iron deposition, well, it is a part of normal aging. It is associated with a host of neurodegenerative disorders like Alzheimer's, and Parkinson's, and multiple sclerosis.

In MS patients we see iron deposition occurring early in the disease course. It predicts, and is associated with, disease's severity measured by brain atrophy and measured by physical and cognitive impairment. It doesn't explain the whole picture because benign patients also have iron depositions suggesting it may be in part time-dependent. Iron has been linked to a pathophysiology of MS through animal studies showing iron deposition in those animals, and also showing that you can treat the disease by either depleting them nutritionally of iron or chelating the iron from animals.

10:41 (47:08)

So the take-home message is that potentially iron may initially result from the disease. It's probably not the first event that triggers MS, but it is a result of the hypometabolism in the brain, the demyelination in the brain, the dysregulation of the brain that occurs in MS early, leading to dysfunction of the gray matter. But then iron takes on a life of its own as it accumulates in the gray matter and causes ongoing secondary damage resulting in gray matter atrophy.

Let me take this opportunity to again thank Prof. Zamboni for inviting me here, it's really a thrill to be able to be with all you today in this absolutely beautiful city. Thank you for such a nice turnout today. I hope you found it to be useful to exchange ideas about new ways to think about MS, and I'd be happy to take your questions, and I thank you all very much. Thank you, thank you… Thank you very much, thank you.

11:35 (48:05)

Zamboni: Thank you, Rohit, for your excellent lecture. You gave us an overview of the value of iron in multiple sclerosis' complex parthenogenesis, and the value of iron in prognosis and the possible significance of iron chelation in the treatment of multiple sclerosis. The lecture is of course open for discussion, so if you have questions you may also… [sets forth in Italian].

14:29 (50:54)

Prof. Zamboni, thank you for your very kind comments and your question. First we should keep in mind that the EAE models of MS are not entirely replicative of the human condition. Unfortunately, I wish we had perfect animal models, but we don't. For example, EAE is usually and predominantly a spinal cord disorder. There is very little involvement of the brain. The slides I showed you, that were the black-and-white iron deposition slides, were all of the spinal cord. And that was an older study from about 10 or 15 years ago. The most recent data I showed you from Dr. LeVine and colleagues, where you saw T2 hypointensity on MRI and iron deposition in the gray matter: that is a new animal model that Dr. LeVine himself has pioneered, which is similar to EAE, but hopefully will get us closer to these more direct one-to-one correlations that we would see in the human condition.

So, the starting point for my answer is: unfortunately those two diseases are different. EAE for example unfolds over several months and usually runs its course by the end of a year, whereas MS as you know is a chronic relapsing-remitting disorder that unfolds over the lifetime of a patient. EAE involves primarily the spinal cord, and MS involves primarily the brain, but also the spinal cord.

15:46 (52:13)

Turning to another important consideration in trying to understand your question: Unfortunately there has been relatively little study of iron deposition in both the mice and also in humans in MS. Most of the iron related work has come from imaging studies, as you know. There really has not been a lot of effort. Certainly before, say, the last 10 years there has been very little effort to try to even look at gray matter for anything in MS, whether it be plaques, or whether it be iron deposition.

That's because we all got put into, I would call it, a pigeonhole way of thinking. It's almost like we got too focused on white matter damage early on and when MS was first described, and how it was taught to us in the 1950s, 1960s, 1970s. We were taught and what was emphasized was that MS is a white matter disorder, and that's where we need to look. We need to look there to find the answers to MS.

16:45 (53:11)

So the pathologists, the histologists really didn't pay attention much to gray matter involvement. But now in the last five or ten years there's been an explosion of data showing us the importance of gray matter involvement, and how common gray matter involvement is, whether it be hypometabolism, whether it be atrophy of the gray matter or… on functional MRI studies you saw hypoactivation of the gray matter, and there are finally iron deposits. So I think you're going to see in the future a better delineation of what exactly is the nature of the iron deposition in MS, in the gray matter and in the white matter. I think that in the white matter from the very few studies that we have done ourselves on histology which we never published, we see iron in the perivenular areas. We also see it scattered throughout the parenchyma in what I would call crenations of iron. They are not intracellular, they are not next to vessels, they are just spilled out into the parenchyma. We also see iron inside glial cells, which are not associated with vessels. That might be in microglia, that might be in macrophages. And we see a similar pattern for iron deposition in the gray matter. It's in many different forms, it's not just perivenular.

17:58 (54:25)

So I don't really have an answer for why iron is there. I don't know what the mechanism is. I think that your proposal that it could be on the basis in part of venous occlusions is a viable mechanism that still needs to be better understood and proven. For example, if it could be shown that as you increase and improve the patient's venous flow you start to actually clear iron from their brain, that would be a very, very important observation, if that could be shown. So, I probably gave you a longer answer than you wanted, but… but it's a very provocative area.

18:36 (55:02)

Zamboni: Thank you...

[Introduces questioner. Question from a neuroradiologist.]

20:50 (57:17)

Thank you for your kind comments. It's terrific to have a neuroradiology audience side-by-side with a neurology audience. We always benefit from these collaborations. And I would like to try to answer your question in several ways. First of all, I would love to come back another time and give you a general imaging talk on MS. Of course, today I really wanted to focus on iron deposition and gray matter damage. But you raised a lot of issues, that I didn't have time to go into today. First of all, what is the role of doing routine 3T imaging instead of 1.5 T. Do you want to eventually get away from 1.5 T? What about routine clinical imaging? What pulse sequences should we use? When should we image the cord versus the brain, or cord and brain?

21:35 (58:02)

How shall we report an image in a MS patient clinically, what shall we be putting into the body of a report, what are the key findings that we need to report? So, let me try to answer some of those questions, and I'll start by saying that I would strongly recommend - if you're not already doing so - that you incorporate cord plus brain for your routine imaging exams, both at the time of diagnosis of MS, and for routine follow-up of patients. The minimum would be brain and cervical, if you can also get thoracic that's very beneficial. There are several benefits to adding cord imaging as part of your routine imaging: number one, you can meet the international panel McDonald criteria for diagnosing MS much more sensitively if you do cord plus brain at the time of diagnosis. You can raise the diagnostic yield to, say, 85% compared to, say, 65% with brain only imaging, which has major implications for therapy, and for monitoring patients, and for their prognosis.

22:31 (58:58)

Secondly, for the differential diagnosis cord imaging is extremely valuable, even in patients who already have MS, because MS patients get other diseases in their cord like disc disease. And we don't want to miss that. If somebody comes in with weakness, don't let us already assume that that is from MS. It could be that they are having ambulatory problems, they may have fallen down, they may have osteoporosis from all the IV steroids we are giving them. So they actually have a disc herniation and have spinal stenosis. There are also diseases in the cord that can look like an MS clinically like transverse myelitis, post vaccination myelitis, sarcoid associated myelitis, and this looks very different on MRI than typical MS. You typically have a longitudinal lesion with a swollen cord over multiple levels. I would include Devic's disease in that category. It looks very different from MS.

23:17 (59:44)

Another major benefit to our field is going to be focusing more on gray matter pathology on a routine basis. There are several efforts going on currently which I think are going to bear fruit for us clinically and going to give us new clinical tools. The first is double inversion recovery … DIR. I don't know if you're familiar with that technique … DIR?

I see, you've heard of that technique. Now, double inversion recovery was pioneered and brought to the forefront by the group in Amsterdam, and that technique is even better than a FLAIR for looking at gray matter disease, because it has two inversion recovery pulses. The first suppresses the CSF30like the FLAIR does, but the second inversion recovery pulse also suppresses the white matter. So it's a very nice imaging tool to see gray matter lesions in MS, and gray matter pathology. Unfortunately, DIR is not readily available, it's not a standard sequence shed on most machines. But I think in the future, in the next few years it's going to become a standard tool that we are going to do in MS patients. And that will help you when you're doing routine imaging, because the gray matter is also affected by direct demyelination and direct plaque formation, which I didn't have time to cover today.

24:32 (1:00:59)

SWI is another provocative tool. SWI is, as you have already seen, is very sensitive, but it really requires a more ultrahigh field magnet. So, once we are able to switch most centers in the world from 1.5 T to 3 T, then it would make sense to add the SWI as a routine tool. But even the SWI, unfortunately, is very challenging to use clinically at this point. It's not readily available. It takes a physicist to really get it to work. The post-processing is very labor-intensive. But I think in the future that will get better. And you also… what also did you ask about? I think there's some other potential advances coming in the near future.

I want to talk about subtraction imaging for a minute. I didn't again have time to go into this, but subtraction imaging is the idea of taking a patient with MS, a baseline scan, and a follow-up scan. And instead of just looking at them side-by-side and to say okay, there is may be a new lesion here, this lesion is smaller, this lesion is bigger – you have a very tough time doing that on a clinical basis, because of the patient's position in the magnet, because of the slice orientation, because of the signal intensity differences, there are a lot of problems with doing that.

We try our best, but sometimes we are not accurate. But with subtraction imaging you can develop a pipeline, which co-registers those data sets, re-slices them, normalizes the signal intensity and does a homogeneity correction, and now on a voxel-by-voxel basis you can compare the images perfectly, and you can see if there's new lesions, you can see if lesions are resolving. And I would hope in the future that will become a tool that's built-in with scanning platforms and scanning software, so that you can actually get that as a clinician on the fly while you're doing your routine imaging.

So I guess my quick answer to your question is: you're on the right track. I completely agree, we have to do a better job, and I think the tools are there for us. We just have to roll them out, and we have to validate them.

26:32 (1:02:59)

[Zamboni introduces questioner. Question.]

27:07 (1:03:34)

You are asking some excellent questions, actually nobody has done any studies in primary progressive MS to look at the iron deposition pattern. As you probably know, primary progressive MS is a rare form of MS, probably only 10% of patients with MS will have primary progressive MS. So they are hard a group to identify. And they tend to be more disabled than are relapsing-remitting patients and CIS patients. So it's harder to get them into studies to do research on these patients, because they have mobility problems, and they may have lifestyle issues that makes it very challenging. My own clinical viewpoint on primary progressive MS is that it's not going to look much different from the other phenotypes. You're going to see a widespread iron deposition as the disease advances involving deep gray, cortical gray, and posterior fossa - gray matter structures.

You also asked about the spinal cord, and that's a brilliant question. Nobody has done any spinal cord imaging to look for iron in MS. The reason for that has been that at 1.5 T the spinal cord by itself is hard to visualize for it is so small, even just seeing lesions is not necessarily very easy. When you now try to do T2* imaging or susceptibility weighted imaging it becomes extremely challenging at 1.5 T. So with 3 T becoming available, I think, that we will be able to do studies in the spinal cord. My own guess would be that we are going to see iron deposition. We are going to see it associated with plaques in the spinal cord, and we also going to see it associated with the gray matter in a more diffuse and neurodegenerative way. But that has never been studied, that has never been shown. Thank you for the question.

28: 34 (1:05:11)

[Zamboni introduces questioner. Question from a molecular biologist.]

30:22 (1:06:49)

Just like it's good to have a neuroradiologist in the room it's good to have a molecular biologist.

[Zamboni]

I think it's wonderful. I had not thought about that, your question is very provocative. And if you look at, for example, the benign MS patients that I showed you in that study we completed, these are patients who have had the disease for 15 or 20 years. They have a lot of iron in their brain, but yet, they are benign clinically. They don't have the cognitive impairment or physical disability that typical MS patients have. So what is missing in those patients? Why don't they go directly from high levels of iron to neurotoxicity? What is it about them? Is it genetics like you suggested? Could it be the patients have a very high brain reserve capacity? What I mean by that are they able to repair their brain much better than average people?

It may be that they have formed better connections in their brain before they got the disease, so as they get damage they are able to recruit new pathways and adapt to injury that's occurring. That may be genetically determined, but it also may be based on their IQ or the way their cognition works, the way their brain is actually wired from their education level. We don't really have good answers to that. I think that the genetics of MS is so complex right now, we are still trying to understand even if any genes are associated with MS by itself, separate from whether they have iron present or not. So I think we are probably a long way away from being able to try to target specific genes.

31:53 (1:08:20)

But, I think it's very important to conclude - and just emphasize what you already have mentioned - which is I don't think iron is the sole factor contributing to damage in all patients. I think it's a very complex disease. Iron may be one of the co-factors that interacts with a complex set of events including demyelination, including gliosis, including axonal loss, neuronal loss, neurotoxicity. There are so many events that are occurring. Nitrous oxide may play a role, excitatory amino acids may also play a role, depletion of growth factors may be important. There are so many mechanisms. Dr. Zamboni's hypothesis is also there that there may be vascular origin or vascular contributing. So I guess we'll have to get back to work and try to figure this out. Thank you.

32:45 (1:09:11)

[Zamboni introduces Dr. Salvi. Some joking around. Salvi starts asking questions.]

Zamboni: Stop, Salvi... because you have million questions. The last one...

[Salvi goes on.]

35:16 (1:11:42)

That was only 500,000 questions, that wasn't a million. Thank you Dr. Salvi, it was really nice to host you in Boston about a year ago. I'm glad to be able to come for the visit. Nice to see you again. I'm going to try to remember your questions. Is this a cognitive test for me if I can remember your questions?

I'm going to start with your first question, which is what is the relationship between gray matter damage that we see with T2 hypointensity and white matter damage which we see with plaques for example. Several studies have looked at that and there is a relatively weak association. So, we did a study in 2000 I believe where we looked at T2 hypointensity in various structures and we showed that it has a very weak correlation with white matter lesion burden. It was a significant correlation, but the R-value was something like .2 or .3. It's a very weak association. So you can't make a direct connection

Third Videohttp://www.fondazionehilarescere.org/video/Bakshi%20_CCSVI_03.avi

between white matter lesions, at least the obvious lesions, and gray matter pathology. The same is also true, by the way, for gray matter atrophy. If you look at gray matter atrophy and white matter lesions the correlations are pretty poor. I think one of the major reasons for that is that in the past we have been blind to cortical lesions that we can see with DIR for example. So there is gray matter pathology that is directly a result of the disease that we miss if we just measure white matter lesions. The gray matter is really a direct site of damage. That includes, by the way, meningeal inflammation.

That's recently been described by the Mayo Clinic group and Hans Lassmann and colleagues. So, overlying the gray matter there is an abundance of inflammatory cells, both T cells and B cells, which are thought to enter into the nervous system through the choroid plexus from the blood and go directly into the CSF and bathe the whole brain with this inflammatory milieu which then extends down from the pia into the gray matter. So, this may be a case where we have been looking in the wrong place to try to get our answers, and may be the gray matter is going to tell us the real story for why the iron is there.

Can you remind me the other questions?

1:12 (1:13:48)

Zamboni: Your opinion between radiologic isolated syndrome and… iron.

Oh yes, right. I don't have any personal experience with looking at patients with neuroimaging isolated MS, clinically silent, and then doing iron specific imaging on those patients. Unfortunately, as the neuroradiology colleague pointed out, standard clinical imaging is not effective in detecting iron deposition, even if you do a T2* at 1.5 T it's not sensitive enough to actually see iron deposits unless you get pretty high levels of iron. You really need SWI to see the iron, you need to do quantitative relaxometry preferably at 3 Tesla. So, this issue of RIS31 has not been answered yet and will require specialized studies.

My own opinion though is that the RIS is going to show us something similar to what we see in CIS which is there is not much iron there, maybe very little. And if you have iron present in RIS, you probably had a higher risk for developing problems later on, based on the data I've showed already. T2 hypointensity predicts brain atrophy. T2 hypointensity also predicts disability, which I didn't show a slide about that. But we have data showing ongoing T2 hypointensity over time predicts physical disability. So, RIS is going to be pretty rare but it may be a marker for that subgroup that's not going to do well.

Zamboni: The other question was immunomodulation and iron.

2:40 (1:15:13)

So there has not been a good study done yet to answer the question. Once you put patients on disease modifying therapy, do you get actually clearing of iron, does iron decrease or does it remain stable, or does it continue to increase? We don't know the answer to that. The reason for that is again, we don't have the routine ability to quantify iron in a sensitive way from clinical scans, even if you have a dual echo, a T2 conventional spin echo dual echo.

We have tried to do that kind of quantification, and you just get too much variability in your data and you can't follow iron over time in a two-year timeline. My own feeling is that you probably will see some reduction of iron levels with disease modifying therapy, because iron is in part a result of the neurodegeneration that's occurring. The less atrophy that you get, the less neurodegeneration, the more opportunity the brain has to clear the iron through normal homeostasis by getting the iron out of the gray matter through axonal transport and dumping it into the CSF and getting rid of it. And we know disease modifying therapies have some impact on limiting neurodegeneration. They have, say, 30 or 40 or 50 percent effect on limiting ongoing brain atrophy. So at the same time they probably are going to limit and reduce the accumulation of iron.

4:00 (1:16:38)

[Zamboni and another question from the auditorium.]

4:32 (1:17:10)

We have not done perfusion imaging at our center, but we have done PET scanning. I've showed an example of glucose PET, and the group at NYU, Inglese and colleagues, have done the most work in perfusion imaging in MS. This is the New York University in Manhattan, New York. You see similar findings whether you do metabolic imaging with glucose or perfusion imaging with MRI or you do blood-flow imaging with SPECT32, which we have done some of as well. The general finding is that the brain is hypometabolic, and the reason it's hypometabolic I believe is not because the blood-flow is decreased. I think the hypometabolism starts first because of demyelination, because of axonal loss, because of inflammation in the brain, because of iron deposition. And when the brain is hypometabolic it now has a lower demand for arterial blood flow.

These two events are very closely linked physiologically: demand and supply. If an organ shuts down the blood-flow shuts down to the organ. That's a result of the organ shutting down. The brain is relatively decreased in its overall activity in MS, so it demands less of the blood to come in. It also demands less venous outflow for the same reason because there is less inflow from the arterial side, there is less oxygen utilization, so therefore the body will actually reduce the blood-flow. So perfusion studies showed decreased blood flow to the brain, particularly in the gray matter.

My own opinion is that this is probably a result of the disease. And it may be that the venous congestion, the venous anomalies Dr. Zamboni has nicely shown, part of the cause of that is low outflow from the brain because the brain is hypometabolic and the veins are not as rigid as arteries. If you have a low flow system the veins will tend to collapse because they can't maintain the rigidity.

6:28 (1:19:04)

- End of transcription -

1 Magnetic resonance imaging of brain iron. Schenck JF. J Neurol Sci. 2003 Mar 15;207(1-2):99-102. (PubMed)

2 Perls' Prussian Blue (Wikipedia)

3 Nucleus ruber

4 MR imaging of human brain at 3.0 T: preliminary report on transverse relaxation rates and relation to estimated iron content. Gelman N, Gorell JM, Barker PB, Savage RM, Spickler EM, Windham JP, Knight RA. Radiology. 1999 Mar;210(3):759-67. (PubMed) (PDF)

5 MRI of brain iron. Drayer B, Burger P, Darwin R, Riederer S, Herfkens R, Johnson GA. AJR Am J Roentgenol. 1986 Jul;147(1):103-10. (PubMed) (PDF)

6 Magnetic resonance imaging of iron deposition in neurological disorders. Brass SD, Chen NK, Mulkern RV, Bakshi R. Top Magn Reson Imaging. 2006 Feb;17(1):31-40. (PubMed)

7 PKAN / Hallervorden-Spatz-Disease (Wikipedia)

8 Reduced signal intensity on MR images of thalamus and putamen in multiple sclerosis: increased iron content? Drayer B, Burger P, Hurwitz B, Dawson D, Cain J. AJR Am J Roentgenol. 1987 Aug;149(2):357-63. (PubMed) (PDF)

9 Signal intensity on MRI of basal ganglia in multiple sclerosis. J Grimaud, J Millar, J W Thorpe, I F Moseley, W I McDonald, and D H Miller. J Neurol Neurosurg Psychiatry . 1995 September; 59(3): 306-308 .

10 MRI T2 shortening ('black T2') in multiple sclerosis: frequency, location, and clinical correlation. Bakshi R, Shaikh ZA, Janardhan V. Neuroreport. 2000 Jan 17;11(1):15-21. (PubMed)

11 MRI T2 hypointensity of the dentate nucleus is related to ambulatory impairment in multiple sclerosis. Tjoa CW, Benedict RH, Weinstock-Guttman B, Fabiano AJ, Bakshi R.Journal of the Neurological Sciences, 2005 Jul 234 (1-2), 17-24

12 "Brain Parenchymal Fraction". A semiautomated measure of whole-brain atrophy in multiple sclerosis. Bermel RA, Sharma J, Tjoa CW, Puli SR, Bakshi R. J Neurol Sci. 2003 Apr 15;208(1-2):57-65. (PubMed)

13 P-value (Wikipedia)

14 Pearson product-moment correlation coefficient (Wikipedia)

15 Cognitive impairment is associated with subcortical magnetic resonance imaging gray matter T2 hypointensity in multiple sclerosis. Brass SD, Benedict RH, Weinstock-Guttman B, Munschauer F, Bakshi R. Mult Scler. 2006 Aug;12(4):437-44. (PubMed)

16 Prediction of Longitudinal Brain Atrophy in Multiple Sclerosis by Gray Matter Magnetic Resonance Imaging T2 Hypointensity. RA Bermel, SR Puli, RA Rudick, B Weinstock-Guttman, E Fisher, FE Munschauer, R Bakshi. Arch Neurol. 2005;62(9):1371-1376.

17 Deep gray matter T2 hypointensity is present in patients with clinically isolated syndromes suggestive of multiple sclerosis. Ceccarella et al. Mult Scler. 2010 Jan;16(1):39-44DOI: 10.1177/1352458509350310.

18 Quantitative assessment of brain iron by R2* relaxometry in patients with clinically isolated syndrome and relapsing-remitting multiple sclerosis. M Khalil, C Enzinger, C Langkammer, M Tscherner, M Wallner-Blazek, M Jehna, S Ropele, S Fuchs, F Fazekas. Mult Scler. 2009 Sep;15(9):1048-1054. DOI: 10.1177/1352458509106609 (PDF)

19 Quantitative In Vivo Magnetic Resonance Imaging of Multiple Sclerosis at 7 Tesla with Sensitivity to Iron. Hammond KE, Metcalf M, Carvajal L, Okuda DT, Srinivasan R, Vigneron D, Nelson SJ, Pelletier D. Ann Neurol. 2008 Dec;64(6):707-13. (PubMed) (PDF)

20 High-resolution fluorodeoxyglucose positron emission tomography shows both global and regional cerebral hypometabolism in multiple sclerosis. Bakshi R, Miletich RS, Kinkel PR, Emmet ML, Kinkel WR. J Neuroimaging. 1998 Oct;8(4):228-34. (PubMed)

21 Functional Magnetic Resonance Imaging Correlates of Fatigue in Multiple Sclerosis. Filippi M, Rocca MA, Colombo B, Falini A, Codella M, Scotti G, Comi G. Neuroimage. 2002 Mar;15(3):559-67. (PubMed)

22 Selective caudate atrophy in multiple sclerosis: a 3D MRI parcellation study. Bermel RA, Innus MD, Tjoa CW, Bakshi R. Neuroreport. 2003 Mar 3;14(3):335-9. (PubMed)

23 The impact of lesion in-painting and registration methods on voxel-based morphometry in detecting regional cerebral grey matter atrophy in multiple sclerosis. (ECTRIMS)

24 Comparison of subcutaneous interferon beta-1a with glatiramer acetate in patients with relapsing multiple sclerosis (the REbif vs Glatiramer Acetate in Relapsing MS Disease [REGARD] study): a multicentre, randomised, parallel, open-label trial. Mikol DD, Barkhof F, Chang P, Coyle PK, Jeffery DR, Schwid SR, Stubinski B, Uitdehaag BM; REGARD study group. Lancet Neurol. 2008 Oct;7(10):903-14. Epub 2008 Sep 11. (PubMed)

25 Fenton's reagent (Wikipedia)

26 Experimental autoimmune encephalomyelitis (Wikipedia)

27 Theiler's encephalomyelitis virus (Wikipedia)

28 Iron in Chronic Brain Disorders: Imaging and Neurotherapeutic Implications. James Stankiewicz, Scott S Panter, Mohit Neema, Ashish Arora, Courtney Batt, Rohit Bakshi. Neurotherapeutics. 2007 July; 4(3): 371-386. doi: 10.1016/j.nurt.2007.05.006 (PubMed) (PDF)

29 Desferrioxamine suppresses experimental allergic encephalomyelitis induced by MBP in SJL mice. Petchenko TV, LeVine SM. J Neuroimmunol . 1998 Apr 15;84(2):188-97 . (PubMed)

30 Cerebrospinal fluid

31 Radiologically isolated syndrome

32 Single photon emission computed tomography



Acknowledgement