Show simple item record

dc.contributor.advisorWiner, Rachel L
dc.contributor.authorVasavada, Anjali
dc.date.accessioned2022-07-14T22:11:12Z
dc.date.available2022-07-14T22:11:12Z
dc.date.submitted2022
dc.identifier.otherVasavada_washington_0250O_24359.pdf
dc.identifier.urihttp://hdl.handle.net/1773/48997
dc.descriptionThesis (Master's)--University of Washington, 2022
dc.description.abstractBackground: HIV-positive individuals remain at risk for anal cancer and will likely benefit from research that prioritizes early detection of anal cancer and optimized screening practices. We assessed host deoxyribonucleic acid (DNA) methylation markers for detecting high-grade anal intraepithelial neoplasia (AIN) versus low-grade AIN or normal samples in HIV-positive individuals in Seattle, Washington. Methods: HIV-positive individuals ≥18 years of age and with assigned male sex at birth were recruited at anal cancer screening, diagnosis, or treatment visits at the Harborview Medical Center Madison Clinic in Seattle, Washington between 2015 and 2016. Anal brush samples were collected by an anoscopist for HPV genotyping and pyrosequencing methylation analysis. Demographic and clinical data, as well as anal cancer screening, diagnosis, and treatment results, were collected through chart abstraction and review of electronic medical record (EMR) data. The exposures were the mean percent of DNA methylation for genes ASCL1, PAX1, FMN2 and ATP10A. The primary outcome was AIN grade, dichotomized as normal/low gradeAIN and high-grade AIN, determined by clinically available histology and cytology results from the same visit. Generalized estimating equation (GEE) logistic regression was used to estimate odds ratios (ORs) and 95% confidence intervals (CI) to assess the associations between methylation and high-grade disease, adjusted for age, HIV viral load and CD4 count. The performance of a marker panel was assessed by measuring sensitivity and specificity of a model at the Youden’s index threshold using genes that were individually associated with highgrade AIN detection. A sensitivity analysis measuring high-grade disease extent was conducted using ordinal GEE logistic regression. Results: 125 samples were analyzed from 85 participants. The mean age of participants was 50.3 (SD 10.2) years, their mean CD4 count was 568 (SD 348) cells/µl and 94.2% had an undetectable HIV viral load. Methylation of ASCL1 (aOR per 1 unit increase in mean percent methylation 1.07, 95% CI 1.01, 1.13) and FMN2 (aOR per 1 unit increase in mean percent methylation 1.12, 95% CI 1.06, 1.19) were significantly associated with high-grade AIN versus low-grade AIN/normal. Mean percent methylation of PAX1 and ATP10A did not show significant associations with high-grade disease. A marker panel combining ASCL1 and FMN2 had a sensitivity of 90.4% for high-grade AIN and a specificity of 45.6%. Methylation of ASCL1 (OR 1.06, 95% CI 1.01, 1.11) and FMN2 (OR 1.07, 95% CI 1.02, 1.10) were each positively associated with increasing extent of high-grade disease, indicating a dose-response effect. Discussion: Increasing levels of DNA methylation of ASCL1 and FMN2 were positively associated with high-grade AIN detection. With further validation studies, self-sampling host gene methylation testing has promise to be an effective screening and triage tool for anal cancer.
dc.format.mimetypeapplication/pdf
dc.language.isoen_US
dc.rightsCC BY
dc.subjectanal
dc.subjectbiomarkers
dc.subjectcancer
dc.subjectcytology
dc.subjecthistology
dc.subjectmethylation
dc.subjectEpidemiology
dc.subjectStatistics
dc.subjectMedicine
dc.subject.otherEpidemiology
dc.titleMethylation markers for anal cancer screening: a repeated cross-sectional analysis of HIV-positive individuals, 2015-2016
dc.typeThesis
dc.embargo.termsOpen Access


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record