VOLUME AND DYNAMIC BLOOD FLOW OF THE CHOROID PLEXUS IN NEUROINFLAMMATORY DISEASE

PV048a / #612 Poster Topic: AS05 - CNS Lupus Background/Purpose Systemic lupus erythematosus (SLE) and multiple sclerosis (MS) are autoimmune neuroinflammatory diseases that lead to cerebrovascular alterations. Compromise of cerebral microvasculature is a plausible route by which immune actors infiltrate the brain parenchyma, resulting in damage. The blood-brain-barrier (BBB), comprised of the brain capillary endothelial cell layer with tight junctions, is the most studied regulator of blood-parenchyma exchange. We have demonstrated regionally increased permeability of the BBB in SLE compared to healthy controls.[1] A recent murine study of SLE found CSF immune infiltrates in CSF even with the BBB intact, suggesting alternative mechanisms of brain tissue infiltration at interfaces between blood and CSF.[2] The Blood-CSF barriers (BCSFB) are in the choroid plexus (CP) within ventricles and at the meningeal barrier that surrounds the brain. In the CP, fenestrated capillaries exchange with a stromal layer at the basal side of an epithelium thereby regulating exchange with ventricular CSF. Once an infiltrate enters the ventricules, it can exchange with parenchyma through the ependymal layer at the ventricular surface. Imaging studies provide evidence for a BCSFB route to MS pathology by observation of periventricular gradients of tissue abnormalities. Our ongoing study is investigating alteration of the BCSFB at the CP in SLE and MS including changes in volume, blood flow, and periventricular gradients in properties of normal-appearing white matter. We report preliminary findings for volume and dynamic blood flow of the CP in SLE and MS patients vs healthy controls. Methods Preliminary analysis includes 5 SLE (4F, 15.8(2.9)yrs), 5 MS (5F, 18.6(3.2)yrs), and 15 healthy controls (HC, 8F, 21.7(3.1)yrs) who completed MRI to assess CP volume (3D T1-weighted) and perfusion dynamics (pseudocontinuous arterial spin labeling at 15 label+delay times from 900 to 4000ms). The protocol also included diffusion, relaxometry, and quantitative susceptibility imaging to assess periventricular white matter. ASCHOPLEX[3] was used to segment the CP and determine volume on the T1-weighted images. Dynamic perfusion signal was fitted to a model to determine arrival time and blood flow at the CP. Pairwise group comparisons of volume, arrival time, and blood flow were expressed as effect size (Cohen’s d) with statistically significant results reported at p<0.05. Results Total CP volume was found to be significantly lower for MS compared to either HC (p=0.012) or SLE (p=0.032) groups (Figure 1). HC and SLE CP volumes were similar (effect size d < 0.2). Among all three groups, arrival time was not found to differ pairwise with remarkable effect size (d < 0.2). Pairwise blood flow differences did not reach statistical significance at p<0.05, but did attain robust effect sizes (Figure 2). CP blood flow was greater in both MS and SLE groups compared to HC with effect sizes of d=0.84 and 1.11, respectively. SLE blood flow exceeded MS with a moderate effect size of d=0.33. Figure 1: CP volume comparisons. Error bars reflect standard error. Figure 2: CP blood flow comparisons. Error bars reflect standard error. Conclusions Volumetric and blood flow changes in the CP were observed in neuroinflammatory disease compared to controls. Interestingly, reduced CP volume was evident in MS but not SLE, suggesting pathophysiological differences. Increased blood flow to the CP in SLE and MS was unexpected assuming vascular compromise but might be explained as compensatory blood flow increases associated with some inflammatory processes. If evidence for a CP route of brain infiltration is found with collection of more data, it could realign ways to phenotype SLE and MS and develop targeted therapies. References: [1.] Gulati G Arthritis Care Res (Hoboken) 2017;69(2): 299-305. [2.] Gelb S. J Autoimmun 2018;91:34-44. [3.] Visani V. Computers in Biology and Medicine 2024;182:109164.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.