White Rose University Consortium logo
University of Leeds logo University of Sheffield logo York University logo

Biomarkers of Lung Cancer Risk and Progression

Taylor, Fiona (2019) Biomarkers of Lung Cancer Risk and Progression. PhD thesis, University of Sheffield.

[img] Text (PhD thesis )
Intro_methods_corrections_FINALFINALversion_04022019.pdf
Restricted until August 2020.

Request a copy

Abstract

Lung cancer causes high mortality because most people present late with advanced disease that is not amenable to curative treatment. Screening high-risk groups with low dose CT imaging of the thorax has been shown to reduce lung cancer mortality by 20%, but at the cost of a high false positive rate. Population stratification with molecular biomarkers could improve the cost-benefit of lung cancer screening programmes and reduce false positives. Tumour cells shed DNA into the blood, enabling tumour-derived genetic alterations to be detected non-invasively by analysing circulating cell-free DNA (cfDNA). The aim of this study was to determine the screening and prognostic potential of total cfDNA levels and two genomic instability scores based on the detection of copy number aberrations in cfDNA samples of lung cancer cases and controls collected in the ReSoLuCENT study (A Resource for the Study of Lung Cancer Epidemiology in North Trent). Controls were identified as low or high risk for the development of lung cancer over five years using the Liverpool Lung Project risk model. CfDNA was extracted from the plasma of 52 untreated lung cancer cases, 32 high risk controls and 10 low risk controls and quantified total cfDNA levels by SYBR green real-time qPCR. Low coverage whole genome sequencing with Illumina HiSeq 2500 was completed for a subset of cases (N=62) and controls (N=40). Two published genomic instability scores were adapted and tested; the plasma genomic abnormality (PGA2) and the copy number aberration (CNA) score. Screening potential was evaluated by performing Receiver Operating Characteristic (ROC) curves to assess the ability of the test to discriminate between lung cancer cases and controls by calculating area under the curve (AUC). Logistic regression was used to further assess the ability of total cfDNA levels and genomic instability scores to predict case or control status. Prognostic value was determined by Kaplan Meir and Cox regression survival analyses. In this preliminary study, there was no difference in total cfDNA levels between early stage lung cancer cases and high risk controls. The PGA2 score was higher in high risk controls compared to lung cancer cases and was not further evaluated. In comparison, the CNA score had good discriminatory ability for high risk controls compared to all lung cancer cases (stage I-IV) with an AUC of 0.74 but poorer discriminatory ability for early stage cases (I-IIIA) with an AUC of 0.60. Although total cfDNA levels and CNA scores above the median value were associated with poor survival, both were statistically significant in univariable but not multivariable cox survival regression analyses. Therefore, total cfDNA levels and the CNA score had limited prognostic value when other factors were taken into account. Total cfDNA levels are not recommended as a screening tool because total levels lack specificity for cancer. The screening performance of the CNA score may be improved by targeting recurrent copy number aberrations and by combining the score with alternative tumour-derived genetic alterations in cfDNA such as point mutations or methylation changes.

Item Type: Thesis (PhD)
Academic Units: The University of Sheffield > Faculty of Medicine, Dentistry and Health (Sheffield) > Medicine (Sheffield)
Depositing User: Dr Fiona Taylor
Date Deposited: 11 Feb 2019 12:22
Last Modified: 11 Feb 2019 12:22
URI: http://etheses.whiterose.ac.uk/id/eprint/22855

Please use the 'Request a copy' link(s) above to request this thesis. This will be sent directly to someone who may authorise access.
You can contact us about this thesis. If you need to make a general enquiry, please see the Contact us page.

Actions (repository staff only: login required)