Hello, it's a pleasure to be here today to talk to you about our work with Rathke's cleft cysts using MR spectroscopy. I have no disclosures with regard to the subject matter of this presentation and I'm very grateful to the AANS and the Integra Foundation for presenting us with the Integra Award for this work. So when you look at an MRI scan, you're actually looking at water and you're looking at the differences between the tissue you're studying and water, and this produces the image that we all use every day. And so we're all familiar with the T1 weighted image, which is dark, the T2 weighted image, which turns water white. We use DWI, which looks at the anisotropy of water. But water is the main focus of standard MR imaging. MR spectroscopy is quite different. MR spectroscopy allows you to look at different metabolites, each which have a unique frequency. The spectrum is obtained by a Fourier transformation of the obtained signal. And the spectrum that you get can be modified a little bit by pulse sequences and advanced techniques can be used to detect specific chemical signatures. So this is a much more powerful technique in many ways. This is what you usually see on an MR spectrogram, and you can see the variety of compounds that we can measure just using the standard usual sequences for this type of study. So MR spectroscopy both can magnify an area. You can look at one voxel or a series of voxels, and it can interrogate that area and say, okay, what's going on in this region? It's very different than an MRI, which just shows you structure with intensity related to water. So here's an example of a patient with MS, and you can see the T2 flare on the left where you see areas of demyelination. But if you then put a cursor on a particular area in the corpus callosum, you can tell a whole lot about a variety of chemical compounds and what's happening with them within that region. So it really shows you different things, and in a way, the spectroscopy is much more powerful. Here's an NFL football player with chronic traumatic encephalopathy. His MRI scan on the left is normal, but on the right with MRS, you can see a decrease in NAA, which is associated with neuronal loss, which is what was the case here. So we all know about Rathke's cysts. They're benign epithelial cysts arising from remnants of the Rathke's pouch in the pituitary gland. The majority of these cysts are asymptomatic, but if they're large enough, they can press on surrounding structures. So you can get optic apparatus compression and visual loss. You can get headaches. You can get pituitary hypofunction from damage to the pituitary, and occasionally, they can hear. Their histology is very bland. They're basically large regions of hydrophobic proteins, which stain red with eosin, as opposed to pituitary adenomas that are very packed and cellular. An MRI scan can suggest a diagnosis of a Rathke's cyst, but sometimes it can be very difficult to distinguish them from tumors. The signal intensity can vary. Sometimes they can be bright on T1, sometimes they're bright on T2, and this really depends on the intratumor cystic protein content. So if the protein's high, you get a different signal than if the protein's low, and there can also be microhemorrhages, which further complicates this. So if there was a new diagnostic method which would differentiate Rathke's cyst from pituitary adenomas, that would really help us for surgical plans. There is no such method right now, and sometimes we explore patients' pituitaries because we're concerned about a tumor, only to find that it is a Rathke's cyst, and in the operating room, there can be delay waiting for a frozen section to make the differentiation. So if there was some way to confirm preoperatively that we really had a Rathke's cyst, this would be helpful for treatment planning, both whether or not we should operate at all, and to smooth and hasten intraoperative flow. So since there is no MRI technique alone, we looked at MR spectroscopy with the idea that we might be able to find something, since it does provide biochemical information about the tissue, and should provide a metabolic characterization. So as a first step for this study, we actually took tissue out, processed it, and then performed MR spectroscopy on an intensely powerful machine. We took five patients with Rathke's cysts and five patients with non-functional adenomas. This was an IRB approved protocol, informed consent was obtained. We collected these specimens during surgery, and we analyzed them with NMR spectroscopy after extraction in methanol and chloroform. So here is our methodology, you can see the samples on the left, they are extracted in methanol and chloroform, we homogenate the tissue, we evaporate the liquid layer, and then we collect a sample. That sample is then put in a tube that is put in a very powerful machine, this is an NMR spectroscopy machine, that is a 14.1 Tesla field strength. Really very high powered, great resolution, great separation of spectra. And here's our results. On the bottom is a pituitary adenoma, on the top is a Rathke's cyst, and you can see there is a tremendous difference between the two. The normal spectra that you see in a tumor is just gone in the cyst, and instead you see two main peaks, you see a peak of uronic and amino sugars, and you see an acetyl amino sugars. So there's two separate little peaks, which really aren't present at all in tumors. If you do superimposition of the spectra, and you do some fancy analysis, you see that what these compounds are, are glycosaminoglycans. These are what's unique to Rathke's cyst, free glucose is seen in both conditions. So here's another way of looking at it. You can see that the red is Rathke's cleft cyst, that there are some significant differences in lactate alanine and taurine, but the striking overall big time difference is in these glycosaminoglycans. They're not present at all in pituitary tumors, and they're very high concentrations in Rathke's cyst. So we feel that these can be used as metabolic markers to make the diagnosis. So our major findings are that Rathke's cyst contain high concentrations of glycosaminoglycans, which are basically high molecular weight sugar polymers with repeating units of a disaccharide made up of uronic and acetyl sugars. They're completely absent in pituitary adenomas, and therefore are really a true metabolic marker. For the future, we have lots of plans. We believe we can take this methodology that's ex vivo and translate it into either three or 70 clinical MRI scanners and add it to our present protocol for pituitary adenoma assessment. This will allow us to detect these glycosaminoglycans non-invasively and help make the diagnosis preoperative. We have a pilot study underway now to do this, and we're using a 7-Tesla scanner to develop the proper sequences, and we're excited about it. In conclusion, we've characterized the metabolic composition of surgically resected Rathke's cleft cysts and pituitary adenomas using an MR spectroscopy. The Rathke's cysts contain high levels of glycosaminoglycans. They're not present at all in pituitary adenomas. It's a huge signal, and therefore we believe that they can serve as a metabolic marker for the detection of Rathke's cleft cysts. Using MRS with a pituitary MRI protocol will provide non-invasive metabolic imaging of the pituitary gland and help us distinguish pituitary adenomas from Rathke's cysts preoperatively in every case. I'd like to acknowledge the people who worked on this with me. I have a brilliant research team with three PhDs, all who perform very important roles in making this study possible. I certainly want to thank all of our donors, both our private foundations, individuals, and public funding from NIH, NSF, and the state of Texas. And last but not least, we're very grateful to our many patients and families who have participated in so many of our studies and are dedicated to join us in the battle against pituitary tumors and brain cancer. Thank you very much.

MR spectroscopy

Rathke's cleft cysts

metabolites

glycosaminoglycans

preoperative diagnosis