In April 2016 Manchester eScholar was replaced by the University of Manchester’s new Research Information Management System, Pure. In the autumn the University’s research outputs will be available to search and browse via a new Research Portal. Until then the University’s full publication record can be accessed via a temporary portal and the old eScholar content is available to search and browse via this archive.

Prognostic Markers in Head and Neck Cancer

Dr Catriona M Douglas

[Thesis].School of Medicine;2011.

Access to files

Abstract

Purpose: The management of head and neck squamous cell carcinoma (HNSCC) is complex and often involves multimodality treatment. Currently, most management decisions are based on clinical parameters with little appreciation of patient differences in underlying tumour biology. The identification of biomarkers that predict response to radiotherapy would be clinically useful in determining optimal management. The purpose of the thesis was to investigate potential biomarkers that might predict radiotherapy outcome in patients with HNSCC. Aims: 1) To investigate the hypoxia-associated biomarkers carbonic anhydrase 9 (CA9) and hypoxia-inducible factor -1α (HIF-1α) in patients with early glottis cancer who underwent radiotherapy as their primary mode of treatment, furthermore to investigate the role of accelerated hypofractionated radiotherapy in the management of T2 glottic cancer. 2) To investigate markers of hypoxia (CA9 and HIF-1α) and viral infection in oropharyngeal cancer, and in particular to test for an association between hypoxia markers and viral infection. 3) To investigate HIF-1 and CA9 in a series of patients undergoing surgery as their primary mode of treatment to explore whether they are associated specifically with radioresponsiveness or a general poor prognosis.Results: 1) Adverse prognostic factors for locoregional control were low pre-treatment haemoglobin (Hb; p = 0.010), advancing T stage (p = 0.001) and high CA9 expression (p = 0.032). Low Hb and high CA9 expression were independent factors on multivariate analysis; and combined predicted locoregional recurrence with an odds ratio of 8.0 (95% CI: 2.7-23.9), or either/or with an odds ratio of 3.3 (95% CI 1.5-7.1). In the subset of T2 patients, five-year locoregional control following radiotherapy was 82% and cancer specific survival was 90%. Serious morbidity occurred in 1.8% of patients. T stage subdivided by vocal cord movement was significant for local control. 2) Features associated with a poor locoregional control were older age (p=0.002), tongue base subsite (p=0.002), heavy alcohol use (p=0.004), heavy smoker (p=0.0002), low Hb level (p=0.001), advancing T (p=<0.0001), N (p=0.001) and AJC (p=0.001) stage, high CA9 expression (p=0.020) and high HIF-1α expression (p<0.0001). In multivariate analysis T stage (p=0.003) and high HIF-1α expression (p=0.001) remained significant. 3) Extracapsular spread was significantly associated with poor cancer specific survival (p=0.022). No other patient variables were associated with outcome. HIF-1α expression was significantly associated with poorly differentiated tumour (p=0.019) and the tumour having a cohesive front (p=0.026). Conclusion: 1) Hb and CA9 have potential to be used together as a biomarker to identify glottis cancer patients with a high probability of a poor outcome following radiotherapy, furthermore, vocal cord movement should be taken into consideration when managing glottis cancer. 2) As it does not appear to be influenced by HPV status, HIF-1α warrants further investigation as a biomarker in oropharyngeal patients treated with primary radiotherapy. 3) As HIF-1α and CA9 had no prognostic significance in patients undergoing surgery, they should be explored further as markers to help guide management decisions in patients with HNSCC.

Bibliographic metadata

Type of resource:
Content type:
Type of thesis:
Author(s) list:
Degree type:
MD
Publication date:
Institution:
Total pages:
218
Table of contents:
Table of Contents 2List of Figures 5List of Tables 6List of Abbreviations 7Abstract 8Declaration 9Copyright Statement 9Dedication 10Acknowledgements 10The Author 101 Introduction 111.1 Head and neck cancer 111.2 Epidemiology 121.2.1 Oral/oropharyngeal cancer 121.2.2 Laryngeal cancer 161.3 Aetiology 191.3.1 Smoking 191.3.2 Alcohol 211.3.3 Diet 221.4 Treatment of HNSCC 231.4.1 Surgery 241.4.2 Radiotherapy 241.4.3 Chemotherapy 261.4.4 Emerging drugs to treat HNSCC 291.4.5 Treatment of oropharyngeal squamous cell carcinoma 331.4.6 Treatment of laryngeal squamous cell carcinoma 341.5 Prognostic factors in head and neck cancer 361.5.1 Staging 361.5.2 Clinico-pathologic factors 381.6 Molecular diagnostics/biomarkers for head and neck cancer 471.6.1 Cancer biomarkers 471.6.2 The need for biomarkers in head and neck cancer 471.6.3 Protein biomarkers 481.6.4 CDKN2A 491.7 Hypoxia 501.7.1 Development of tumour hypoxia 511.7.2 Clinical consequences of tumour hypoxia 521.7.3 Measuring tumour hypoxia 571.7.4 Exogenous markers/ injectable markers 621.7.5 Hypoxia and anaemia 681.7.6 Hypoxia modifying therapy 691.8 Human Papilloma Virus 721.9 Aims of the research 801.9.1 Submission of thesis in alternative format 812 Expression of hypoxia markers in glottic cancer 822.1 Abstract 822.2 Introduction 832.3 Materials and Methods 842.3.1 Patients and Tissue 842.3.2 Immunohistochemistry 852.3.3 Scoring Method 852.3.4 Statistical analysis 862.4 Results – overall glottis series 882.4.1 Patient characteristics 882.4.2 CA9 882.4.3 HIF-1α 882.4.4 Pre-treatment haemoglobin 912.4.5 Multivariate analysis 912.5 Results – T2N0 subseries 932.6 Discussion 953 Prognostic significance of hypoxia markers and HPV status in oropharyngeal cancer 1043.1 Abstract 1043.2 Introduction 1053.3 Materials and Methods 1063.3.1 Patients 1063.3.2 Immunohistochemistry 1073.3.3 HPV genotyping 1083.3.4 Scoring Method 1093.3.5 Statistical analysis 1103.4 Results 1113.4.1 Patient characteristics 1113.4.2 Univariate analysis 1113.4.3 Subsite analysis 1133.4.4 HIF-1α Expression 1203.4.5 CDKN2A 1203.4.6 Multivariate Analysis 1203.5 Discussion 1214 Expression of hypoxia markers in surgically managed HNSCC 1264.1 Abstract 1264.2 Introduction 1274.3 Materials and Methods 1284.3.1 Patients 1284.3.2 Immunohistochemistry 1284.3.3 Scoring method 1294.3.4 Statistical analysis 1304.4 Results 1314.4.1 Patient characteristics 1314.4.2 Univariate and Multivariate analysis 1314.4.3 HIF-1α and CA9 expression 1344.5 Discussion 1355 Overall Conclusion 137References 140Appendix I - Randomised phase III trials of chemoradiotherapy in head and neck cancer patients* 171Appendix II - Prognostic significance of pre-treatment anaemia for radiotherapy outcome in patients with head and neck cancer 173Appendix III - HPV status and prognosis in HNSCC 174Appendix IV - Ethical approval 180Appendix V - Christie Head and Neck Assessment Form 186Appendix VI – Glottic Series 190Appendix VII – Oropharyngeal series 203Appendix VIII – Surgical series 212Appendix IX - Publications and presentations arising from the thesis 217
Abstract:
Purpose: The management of head and neck squamous cell carcinoma (HNSCC) is complex and often involves multimodality treatment. Currently, most management decisions are based on clinical parameters with little appreciation of patient differences in underlying tumour biology. The identification of biomarkers that predict response to radiotherapy would be clinically useful in determining optimal management. The purpose of the thesis was to investigate potential biomarkers that might predict radiotherapy outcome in patients with HNSCC. Aims: 1) To investigate the hypoxia-associated biomarkers carbonic anhydrase 9 (CA9) and hypoxia-inducible factor -1α (HIF-1α) in patients with early glottis cancer who underwent radiotherapy as their primary mode of treatment, furthermore to investigate the role of accelerated hypofractionated radiotherapy in the management of T2 glottic cancer. 2) To investigate markers of hypoxia (CA9 and HIF-1α) and viral infection in oropharyngeal cancer, and in particular to test for an association between hypoxia markers and viral infection. 3) To investigate HIF-1 and CA9 in a series of patients undergoing surgery as their primary mode of treatment to explore whether they are associated specifically with radioresponsiveness or a general poor prognosis.Results: 1) Adverse prognostic factors for locoregional control were low pre-treatment haemoglobin (Hb; p = 0.010), advancing T stage (p = 0.001) and high CA9 expression (p = 0.032). Low Hb and high CA9 expression were independent factors on multivariate analysis; and combined predicted locoregional recurrence with an odds ratio of 8.0 (95% CI: 2.7-23.9), or either/or with an odds ratio of 3.3 (95% CI 1.5-7.1). In the subset of T2 patients, five-year locoregional control following radiotherapy was 82% and cancer specific survival was 90%. Serious morbidity occurred in 1.8% of patients. T stage subdivided by vocal cord movement was significant for local control. 2) Features associated with a poor locoregional control were older age (p=0.002), tongue base subsite (p=0.002), heavy alcohol use (p=0.004), heavy smoker (p=0.0002), low Hb level (p=0.001), advancing T (p=<0.0001), N (p=0.001) and AJC (p=0.001) stage, high CA9 expression (p=0.020) and high HIF-1α expression (p<0.0001). In multivariate analysis T stage (p=0.003) and high HIF-1α expression (p=0.001) remained significant. 3) Extracapsular spread was significantly associated with poor cancer specific survival (p=0.022). No other patient variables were associated with outcome. HIF-1α expression was significantly associated with poorly differentiated tumour (p=0.019) and the tumour having a cohesive front (p=0.026). Conclusion: 1) Hb and CA9 have potential to be used together as a biomarker to identify glottis cancer patients with a high probability of a poor outcome following radiotherapy, furthermore, vocal cord movement should be taken into consideration when managing glottis cancer. 2) As it does not appear to be influenced by HPV status, HIF-1α warrants further investigation as a biomarker in oropharyngeal patients treated with primary radiotherapy. 3) As HIF-1α and CA9 had no prognostic significance in patients undergoing surgery, they should be explored further as markers to help guide management decisions in patients with HNSCC.
Keyword(s):

Institutional metadata

University researcher(s):

Record metadata

Manchester eScholar ID:
uk-ac-man-scw:130436
Created by:
Douglas, Catriona
Created:
12th September, 2011, 20:40:59
Last modified by:
Douglas, Catriona
Last modified:
12th October, 2011, 19:23:31

Can we help?

The library chat service will be available from 11am-3pm Monday to Friday (excluding Bank Holidays). You can also email your enquiry to us.