Next up will be Dr. Rudy, talking about genome-wide analysis of penumbral infarction in ischemic stroke. Good afternoon. My name is Rob Rudy. I'm a rising fourth-year medical student at Harvard Medical School, and I'll be presenting today research done in the lab of Dr. Rose Du at Brigham and Women's Hospital, specifically looking at a genome-wide analysis of the volume of the ischemic penumbra in mice after middle-spherical artery occlusion. Thanks very much for the opportunity to present this afternoon. So no disclosures from anybody in our study group. So as you all know, ischemic cerebrovascular disease is a leading cause of death around the world, estimated at about 6.5 million deaths per year, and that number is only expected to go up in the coming century. After the occlusion of an intracerebral artery, a core area of tissue infarcts within a few minutes, but there's a much larger area of tissue surrounding that core that persists in this at-risk state for a variable period of time. Without intervention, this tissue will eventually infarct, completing the stroke. Getting a better sense of what drives the rate of this infarction might help inform patient outcomes or, sorry, patient selection for interventional procedures or even identify new therapies. So when we think about studying cerebrovascular disease, mice present a really good option. They've been studied extensively in the past. There's differences in the completed ischemic infarct volume in different strains of mice, and because we can control their environments, we can really hone in on the genetic differences between these strains of mice. So in this study, we hypothesized that the volume of the ischemic penumbra would vary in different strains of mice and that we could associate that with single nucleotide polymorphisms. So we used a middle cerebral artery occlusion model. Mice had their MCA occluded and then were sacrificed at either 6 or 24 hours after the initial occlusion, and the infarct was identified with TTC staining. We estimated the penumbra by dividing the rate, the infarct volume at 6 hours by the mean infarct volume across that strain at 24 hours. So if you look at this, this would be the complete infarct in the bottom right corner, and then you can imagine in two extremes having a very small infarct at 6 hours and thus a larger penumbra or a large infarct at 6 hours and thus a smaller penumbra. So we included 445 mice from 33 different strains, 215 of which were sacrificed at 6 hours. The box plot on the right shows the distribution of the penumbra volume across these strains, and there was, in fact, significant variation across strains. The next thing we did was look at the circle of Willis in these mice. So variation in the circle of Willis can affect the final infarct volume, and mice tend to have incomplete circle of Willis, often having missing P1 segments. And so we wanted to know if this would affect the volume of the penumbra, so we summed the number of P1 segments per strain and then tried to correlate that or correlated that with the penumbra ratio and did not find a significant correlation. So that's sort of, with that in mind, sort of thinking that that wouldn't confound our results, we went ahead and did our GWAS. We used 130,000 SNPs from the Broad Institute and filtered by minor allele frequency and missing data. And we used an efficient mixed model to account for strain interrelatedness with penumbra ratio as the outcome. So these are our results. We found 18 genome-wide significant SNPs, and by that I mean they had an adjusted p-value of less than 0.05. And there were six protein-coding genes that contained significant SNPs, and for the sake of time I won't get into each gene here, but they might be involved from a literature search in function of the blood-brain barrier as well as a hardiness to ischemia, suggesting that targeting these might help in treating patients with stroke. So this is an ongoing study. The next things we're doing are validating these genes. And then the other thing we'd like to do is test these using MR perfusion rather than the estimated penumbra ratio we had, probably in a smaller number of mice to make it a little bit more tenable. And then with that in mind, there are some limitations to discuss here. The ratio we used is really an estimate. Given the number of mice, we felt that it would be a little bit easier, a little bit more easy to do. But in the future, using a more focused MR-based perfusion technique would allow you to actually look at the penumbra a little bit better. And then variations in single nucleotide polymorphisms really give you candidate genes that doesn't necessarily mean there's any functional difference in that protein or quantitative difference in that protein. So this really is pending the validation studies at this point. So in conclusion, penumbra volume varied across mouse strains and did not depend on a complete circle of Willis. We identified six candidate protein coding genes involved with blood-brain barrier function and neuronal hardiness to ischemia, but we are working to validate these as we speak. And with that, I'd like to thank my lab, especially my mentor, Dr. Du, and the rest of the members of our lab and our colleagues at Harvard and the University of Pittsburgh. Thank you.

genome-wide analysis

penumbral infarction

ischemic stroke

mouse model

blood-brain barrier function