The cervicovaginal microbiome in human papillomavirus-associated cervical carcinogenesis: potential value for clinical practice

Background: Infection with high-risk human papillomavirus (hrHPV) is a necessary, but not sufficient, cause of cervical cancer and its precancerous lesion, cervical intraepithelial neoplasia (CIN).Most human papillomavirus (HPV) infections are transient; a small proportion persist and lead to cervical cancer.The paradigm for detection of cervical abnormalities is cytology, with HPV testing recently introduced for screening and risk prediction.Relevant to the latter is the role of the cervicovaginal microbiome (CVM).Evidence suggests that the CVM is implicated in HPV and carcinogenesis.However, research has been limited by small sample size studies and low taxonomic resolution.Objectives: This thesis investigated the relationship between the CVM, HPV and CIN.The objectives were to: 1) conduct a review on the CVM in cervical cancer, 2) assess CVM composition, and 3) compare the diagnostic accuracy of the CVM, cytology, and HPV for CIN and hrHPV detection. Methods:In manuscript 1, 3 databases were searched until July 27th, 2022.Eligible research articles discussed the CVM in HPV-associated cervical cancer, characterized the CVM via metagenomics and included a measure of association.Statistics, study design, population, and methodology were extracted and summarized.Manuscript 2 included 186 women [54 normal, 50 CIN1, 40 CIN2, 42 CIN3] referred for colposcopy following abnormal cytology.Samples were genotyped for hrHPV with the Roche cobas 4800 assay.The CVM was characterized with 16S rRNA gene sequencing of two regions (V3V4, V5V6) and bioinformatic processing via the highresolution ANCHOR pipeline.Logistic regression models were constructed with 1) CVM species, 2) hrHPV, 3) cytology, and 4) CVM species and hrHPV as predictors and CIN2+ as the outcome.The coefficients were used to construct linear scores on CVM species, cytology, HPV, and CVM species/HPV.Species were selected via logistic regression with stepwise forward selection.Receiver operating characteristic curves were plotted, and the area under the curve (AUC) and 95% confidence intervals (CI) were compared to assess clinical performance (reported as AUC;95%CI). Results:In manuscript 1, high CVM diversity and Lactobacillus depletion appear to increase and decrease the risk of adverse outcomes in HPV-associated cervical cancer, respectively.In manuscript 2, 77 species were identified; 8 unique to V3V4, 48 V5V6 and 21 shared.For CIN2+ LIST OF APPENDICESTables and figures prefaced with "S" refer to supplementary tables and figures. S-Table 3-1.Search strategies to examine the epidemiological and clinical role of the CVM in HPV-associated cervical carcinogenesis.S-Table 3-2.Observational studies on the association between the CVM and HPV prevalence, acquisition, persistence, clearance and/or cytology interpretations or biopsy confirmed CIN and cervical cancer.S-Figure 4-0.Overall methodology for the empirical research manuscript.S-Table 4-1.Distribution of bacterial species by histology and descriptive statistics of their raw abundance based on V3-V4 primer set.S-Table 4-2.Distribution of bacterial species by histology and descriptive statistics of their raw abundance based on V5-V6 primer set.S-Figure 4-1.Correlation between bacterial species in normal samples.S-Table 4-3.Stepwise logistic regression coefficients of cytology-, HPV-, and microbiome-(species presence/absence) based scores used to construct linear scores, comparing CIN1+ to normal histology.S-Table 4-4.Stepwise logistic regression coefficients of cytology-, HPV-, and microbiome-(species presence/absence) based scores used to construct linear scores, comparing CIN2+ to normal and CIN1 histology.S-Table 4-5.Stepwise logistic regression coefficients of cytology-, HPV-, and microbiome-(species presence/absence) based scores used to construct linear scores, comparing any high-risk HPV positive to negative.S-Table 4-6.Stepwise logistic regression coefficients of cytology-, HPV-, and microbiome-(species presence/absence) based scores used to construct linear scores, comparing CIN2+ to normal and CIN1 histology among women who tested positive for high-risk HPV.S-Table 4-7.Stepwise logistic regression coefficients of cytology-, HPV-, and microbiome-(species raw abundance) based scores used to construct linear scores, comparing CIN1+ to normal histology.

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.