Thesis

Developing spectroscopic serum diagnostics towards clinical translation : the detection and stratification of brain tumours

Downloadable Content

Download PDF
Creator
Rights statement
Awarding institution
  • University of Strathclyde
Date of award
  • 2020
Thesis identifier
  • T15711
Person Identifier (Local)
  • 201793299
Qualification Level
Qualification Name
Department, School or Faculty
Abstract
  • The diagnostic pathway for brain tumour patients is currently ineffective. As there are no methods in place for the early detection of brain cancer, the affected patients’ average life expectancy is reduced by 20 years, which is the highest of all cancer types. Thus, the development of rapid, low-cost platforms in primary care to triage patients for medical imaging may reduce diagnostic delay, whilst potentially providing cost-effective infrastructures for health care providers. Attenuated total reflection Fourier transform infrared (ATR-FTIR) spectroscopy has demonstrated the ability to diagnose a wide range of pathologies with high accuracies, but the technique is yet to make it into a clinical setting as a regulated spectroscopic test. Recently, innovative high-throughput accessories have been developed which could accelerate clinical translation. The research described in this thesis focuses on the development of the technique by examining the diagnostic ability for the detection and stratification of brain tumours. Initially, the novel high-throughput ATR-FTIR technology was validated in one of the largest spectroscopic studies to date, by separating brain cancer and non-cancer patients with balanced accuracies of 90%, which is comparable to traditional fixed diamond crystal methodology. Distinguishing brain tumour types with serum spectroscopy would be useful for neurologists, as some are difficult to discriminate through medical imaging alone. For example, the highly aggressive glioblastoma and primary cerebral lymphoma (PCNSL) can appear similar on magnetic resonance imaging scans. The differentiation between glioblastoma and PCNSL patients achieved a sensitivity and specificity of 90.1% and 86.3%, respectively. Several other types of brain lesions were then distinguished with balanced accuracies >80%. A reliable blood serum test capable of stratifying brain tumours may avoid the need for surgery in some cases, and could speed up time to definitive treatment. Rapid determination of a glioma patient’s IDH1 status facilitates vital neurosurgical decisions, such as pursuing with resection or opting for alternative therapeutics. Synchrotron-based infrared light has been utilised to probe brain tumour tissue microarrays and differentiate between IDH1-mutated and IDH1-wildtype glioma, at a sensitivity and specificity of 82.4% and 83.4%, respectively. Additionally, centrifugal filtration of patient serum was examined, with the aim of detecting the global epigenetic and metabolic changes associated with mutations in the IDH1 enzyme. The filtration step ultimately improved the classification performance, by delivering a balanced accuracy of 69.1%. Finally, a health economic evaluation was carried out to examine the associated costs and benefits of the blood serum test in clinical practice. Based on recent prospective clinical data, it was found that test costs up to ￡100 would likely be considered cost-effective, whilst primary care tests set at ￡75 would be cost saving to the health services. When comparing the additional costs required for implementing a brain tumour subtype test, the cost-consequence analysis reported an estimated saving of ~￡138,075 per 10,000 patients, equating to a potential saving of ~￡568 per individual cancer case. Furthermore, this could prevent up to 8 unnecessary surgeries, per 100 patients. Therefore, a brain cancer diagnostic test that can also stratify tumour type would have a profound impact for patients, as well as the health services.
Advisor / supervisor
  • Baker, Matthew
Resource Type
Note
  • This thesis was previously held under moratorium from 28th October 2020 until 28th October 2025.
DOI
Date Created
  • 2020
Former identifier
  • 9912922587102996

Relations

Items

Thumbnail Title Date Uploaded Visibility Actions
file details PDF of thesis T15711 2021-07-02 Public Download