Alzheimer’s Disease (AD) is a progressive brain disorder that debilitates memory, learning, and decision-making. Early-stage AD represents the initial phase where individuals are still able to function independently, but with increasing age, their condition steadily progresses to dementia and loss of independence. Because a significant number of the aging population is affected by AD, understanding the neuroinflammatory processes would help develop more effective strategies for treatment. Examining markers such as MCP-1 and TNF-alpha, known to be associated with inflammatory response, will help identify the modulatory processes that lead to mild cognitive impairment associated with early-stage AD. Subsequently, higher levels of inflammation markers within the brain leads to mild cognitive impairment. This research study involved 40 C57BL/6 mice, 20 males and 20 females (21 months old), retro-orbitally infected with 80 µL of neurotrophic AAV-AD vector or AAV-Sham for a duration of 2 months before humane euthanasia. Brains were collected, and specific regions were examined by immunohistochemistry (IHC) and digital imaging to assess the expression levels and distribution of the inflammation markers. Preliminary observations showed that hippocampal regions of the brain from mice with early-stage AD had higher staining intensity for MCP-1and TNF-alpha compared to respective areas in Sham mice, suggesting increased inflammation is a very early lesion that develops in the presence of AD pathogenic components that might be controlled by anti-inflammatory drugs. The preliminary data suggests that the characteristics of AD manifest in part due to the neuroinflammatory response of brain factors that change with onset AD.

Merkel cell carcinoma (MCC), a rare and aggressive skin cancer, shows approximately 60% response to immune-checkpoint inhibitors (ICIs) in immunocompetent patients. The efficacy of ICIs in immunosuppressed patients, who generally have poorer MCC prognoses, is less clear. This retrospective study assesses ICI outcomes in both immunocompetent and immunosuppressed MCC patients, and across various immunosuppression types. In this project, I determined the cohort of 183 patients with advanced MCC who were treated with first-line ICIs from a Seattle-based data registry. I collected the following data from analyzing patients’ medical records: treatment response, immunosuppressive status, disease-specific and overall survival. I coordinated with the statisticians for Kaplan-Meier analyses of the data, and am now working on the manuscript. The results show that initial response rates to ICIs were comparable between immunosuppressed (50%) and immunocompetent (61.5%) patients (p=0.17). After starting ICI treatment, immunosuppressed patients experienced a 65% increased rate of disease progression compared to immunocompetent patients (Hazard Ratio [HR]=1.65, p=0.04), with their median time to disease progression being 11.2 months, versus 32.9 months in the immunocompetent group. Outcomes also varied by immunosuppression subtype, with chronic lymphocytic leukemia patients having the lowest chance of response (20%, 2/10) and highest progression risk. In our cohort, immunotherapy often elicited an initial response in immunosuppressed MCC patients; however, the duration of this response was significantly shorter. Despite limited duration in immunosuppressed patients, ICIs still offer a high response rate for patients with advanced MCC, regardless of immunosuppression type. It is crucial to have a comprehensive prognosis discussion with patients before initiating treatment.

Heschl’s gyrus (HG) is a region of the brain containing the primary auditory cortex. The extent of folding within the HG shows high morphological variability across individuals. Interestingly, increased HG folding is more likely to be found in expert than amateur musicians, suggesting a possible role of auditory experience in shaping HG gyrification. In my research, I examined HG folding in blind individuals—another population with extensive auditory experience. I hypothesized that, if experience alters HG structure, then individuals with early-onset blindness might have increased HG gyrification compared to those with late-onset blindness or those who are sighted. I analyzed T1-weighted images collected from previous MRI studies at the University of Washington, University of Pennsylvania, and Oxford University. The combined dataset included 6 anophthalmia (individuals born without eyes), 48 early blind, 18 late blind, and 28 sighted control participants. I created hand-drawn HG regions of interest for each participant in both hemispheres and measured HG gyrification in two ways: 1) by visually categorizing the extent of HG folding (single, partial, or complete duplication), and 2) by obtaining continuous metrics (gyrification index and curvedness index) using FreeSurfer. A chi-squared test revealed that the degree of HG folding was not different across the four groups. A linear mixed-effects model (controlling for the effects of age, hemisphere, and scan location), similarly showed no effects of group on the gyrification index or the curvedness index. To conclude, my findings show that blindness does not affect HG gyrification. The results challenge the idea that auditory experience alters HG structure and offer important insights into previous findings in professional musicians. Our results suggest that the prevalence of duplicated HG in musicians may be the result of individuals with larger processing capacity within the auditory cortex being more likely to take up music as a profession.

Joubert syndrome (JS) is a rare, congenital neurodevelopmental condition diagnosed by the appearance of the “molar tooth sign” on axial brain magnetic imaging (MRI). Patients typically display hypotonia, abnormal eye movements, and ataxia. Substantial progress has been made on identifying the genetic causes of JS, which typically displays recessive inheritance. Nonetheless, the cause still cannot be identified in ~27% of our cohort of JS-affected families and the contribution of causal variants that impact RNA splicing remains unknown. Canonical splice variants impact RNA splicing by disrupting the splice site directly, whereas noncanonical splice variants affect it through alternative mechanisms, which must be validated by RNA analysis. Our goal is to evaluate the role of noncanonical splice variants in the pathogenesis of JS. We previously identified genetic causes in 582 of 714 families with JS. To identify additional causes, we used SpliceAI, a deep learning-based tool to identify variants with predicted splicing effects (SpliceAI score >0.5) for functional validation. We extracted RNA from patient cell lines then used polymerase chain reaction (PCR) and evaluated amplicons using gel electrophoresis and Sanger sequencing. Five families with candidate noncanonical splice variants that may confirm their genetic cause provided skin biopsies for further splice analysis. We confirmed the pathogenicity of three of these variants by demonstrating abnormal splicing in JS genes, AHI1 and MKS1, bringing the total contribution of aberrant splicing up to 77/714 families. The final two variants did not render conclusive results after 10+ attempts of PCR-based assays and may therefore be candidates for other methods of investigation such as mini-gene assays or long-read sequencing. By extrapolation from our data in JS, splice variants may contribute ≥11% to the genetic causes of conditions. A precise genetic diagnosis informs prognosis, avoids unnecessary work-up, guides monitoring for associated complications, and opens the door to gene-specific treatments.

Alzheimer’s disease (AD) is the most common cause of dementia in the aging population, characterized pathologically by the presence of amyloid plaques and tau neurofibrillary tangles in the brain. However, AD often coexists with other pathologies contributing to dementia, such as hyperphosphorylated aggregates of the protein TDP-43. TDP-43 induces a dementia syndrome similar to AD and the combination of AD and TDP-43 is associated with accelerated cognitive decline, greater brain atrophy, and increased AD pathologic burden. AD and TDP-43 pathology are definitively diagnosed post-mortem upon neuropathologic examination but there is a great need to be able to identify these pathologies in living patients using biomarkers. Currently there are accepted biomarkers for AD, including measures of amyloid beta and hyperphosphorylated tau proteins in cerebrospinal fluid (CSF), but there are no biomarkers for TDP-43. Leveraging the reliability of CSF in detecting pathologic proteins, we hypothesize that measurable hallmarks of underlying TDP-43 pathology also exist in CSF. We tested four groups of brain donors (n=36 per group) defined by presence or absence of AD and TDP-43 pathology at autopsy: healthy controls, AD only (amyloid plaques and tau tangles), TDP-43 only, and AD+TDP-43. Post-mortem CSF samples are analyzed for TDP-43, hyperphosphorylated tau (pTau-181), and the brain injury marker glial fibrillary acidic protein (GFAP) using the Quanterix SR-XTM Biomarker Detection System. Because these assays are intended for ante-mortem samples, the first aim of the study was to determine optimal sample preparation for post-mortem samples, followed by the second aim to determine if there are concentration differences between proteins in CSF across groups. Successful identification of reliable TDP-43 biomarkers in living patients would improve neurodegenerative disease diagnostics, enabling accurate underlying pathology diagnosis and facilitating tracking disease progression and treatment response as therapies for AD, TDP-43, and other causes of dementia emerge.