Precision Healthcare - A novel diagnostic handheld platform for the detection of cancer biomarkers in urine

Lead Research Organisation: King's College London
Department Name: Chemistry

Abstract

Cancer is the leading cause of death in developed countries and there is a major desire to pivot towards preventative rather than curative based medicine. Currently, effective treatment heavily relies on early stage detection and an accurate diagnosis of the cancer through molecular profiling. Liver cancer is the third most common cause of death due to cancer and has a global incidence of 1 million new cases annually. The prognosis for patients is poor and even worse in resource-poor settings such as sub-Saharan Africa, and Central and Far-East Asia. For example, liver cancer, linked to hepatitis B infection, currently kills nearly four times as many people as HIV/AIDS in Africa, however early detection could have a significant impact on survival rates. In both the developed and developing world, there is a critical need for new tools and technology for the routine detection and diagnosis of cancer and diseases in general.

The goal of this project is to develop a handheld device that can detect biomarkers in urine that will be able to diagnose liver cancer at the point-of-care. It will be assessed using validated patient urine samples. The technology upon which this is based is high performance liquid chromatography (HPLC). Like how a glass prism separates white light into its component colours, HPLC separates a liquid into its component analytes. HPLC is a gold standard analytical technique crucial to many industries worldwide in its ability to separate and identify chemicals in a complex mixture. HPLC is ideally suited to detecting and quantifying biomarkers in urine; however, it is not currently portable or suited to point-of-care analyses due to its size, cost and complexity. As part of this project, we will miniaturise the technology to a handheld device. Point-of-care or on-site HPLC analysis would provide results that could be acted on within minutes that otherwise would take weeks.

Due to the crisis in healthcare provision, such technology would ideally be suited to monitoring any individual, not only patients, in the home in order to realise the vision of next generation precision healthcare. Such a device has the potential to monitor us on a daily basis and act as an early warning system for doctors. Such person-specific molecular data may be used to detect or even predict the onset of disease."

Planned Impact

A diagnostic urine test for liver cancer represents a major step forward in cancer screening. However, detection of these biomarkers requires instrumentation currently restricted to centralised laboratories due to their size, complexity and cost. This fellowship aims to address these obstacles which prevents the widespread and routine adoption of such precision diagnostics. We aim to develop and validate a handheld device for the detection of liver cancer biomarkers in urine, based on high performance liquid chromatography (HPLC). A diagnostic urine test for liver cancer at the point-of-care would provide a practical and cost effective solution to detect this disease early and improve survival.

Benefit to Society
Cancer is the one of the leading causes of death worldwide, particularly in developing countries. One in three women and one in two men will be diagnosed with cancer in their lifetime. Through public awareness and screening programmes, significant improvements have been made over the last few decades. More advanced diagnostic techniques hold significant potential in diagnosing disease early and saving lives. The project coordinates the expertise of clinicians, analytical technology development and industry to focus efforts toward ultimately benefiting patients. The project has the potential to offer in-field cancer screening in low and middle income countries and detect liver cancer at an early stage when the tumours are still treatable. The development of lightweight, field-portable quantitative detection methods will allow the introduction of a range of screening programmes and disease monitoring in areas which have previously not had access to such schemes, providing rapid and precise information to medics in both routine treatment and emergency relief work.

Benefit to the economy.
The potential for personalised medicine and molecular diagnostics to transform clinical practice has been well documented. The global personalised medicine market is expected to reach $233 billion by 2025 and is growing over 9% annually. The growth of new technologies and the rising use of biomarkers for diagnosis are major trends in the market. The outputs of this award will be the achievement of a handheld diagnostic for urinary cancer biomarkers supported by experimental data and a physical prototype. This is a crucial step to achieving our aim of improving access to healthcare and will place the project in an advantageous position for commercialisation. The broad applicability of HPLC in analytical chemistry means that the physical outputs of this project will have many applications in a broad range of environmental monitoring and agricultural applications, from identifying and responding to chemical spills and industrial contamination to optimising fertiliser and pesticide application and crop residues in agriculture. It is therefore important to initiate discussions with key industrial partners in these sectors.

Benefit to People
The award of the fellowship will allow me to continue my programme of research and the supervision of Ph.D., MRes and MSci/BSc students as well as provide training and skill development to the PDRA. The technical and analytical advances that will arise from this proposal will be of interest to the analytical and healthcare scientific communities and will be disseminated through publications and conference presentations. I have significant experience in public engagement and the topics here will form part of any future outreach. A Twitter account is already in place to disseminate my results and publicise events relevant for a broad audience.
 
Description CAMS Co-Funded Lectureship Award
Amount £200,000 (GBP)
Funding ID 600322/02 
Organisation Analytical Chemistry Trust Fund of the Royal Society of Chemistry 
Sector Charity/Non Profit
Country United Kingdom
Start 05/2019 
End 05/2024
 
Description KCL EPSRC IAA
Amount £40,000 (GBP)
Funding ID EP/R511559/1 
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Sector Public
Country United Kingdom
Start 02/2019 
End 01/2020
 
Title Automatic extraction of spectra by data-mining literature 
Description In the field, unsupervised real-time results offer the ability for users to make decisions instantly. Compared to using retention times exclusively to identify species, using spectral fingerprinting is a more robust and discriminating approach, suited to complex or 'messy' samples, where matrix effects can unexpectedly interfere with the elution of species of interest. We investigated the ability of the full spectrum system in conjunction with machine learning techniques to enable the real-time classification of PAHs. In order to classify species, it was necessary to curate a database of reference spectra. To achieve this, we data-mined spectra firstly from exclusively online sources, including literature published spectra and public databases. In addition, using automatic extraction techniques we were also able to process graphical figures of spectra in the literature to obtain the underlying numerical data. The motivation to this approach was to explore how our prototype portable LC system could cope in classifying species without the pre-curation of a bespoke or specialised spectral database. This would allow the platform to be more adaptable to different applications and help minimise the time and cost for rapid deployment. The source of all spectra, and extraction/processing procedures used, here can be found in 10.1038/s42004-021-00457-7. 
Type Of Material Data analysis technique 
Year Produced 2021 
Provided To Others? Yes  
Impact This has been disseminated as part of the publication 10.1038/s42004-021-00457-7 for chromatographers, spectroscopists and research related to polycyclic aromatic hydrocarbons. 
URL https://github.com/AnalyticalSystemsResearch/
 
Title Identification of chemical species by HPLC using Machine Learning Techniques 
Description Computation code and data analysis method to exploit machine learning for the autonomous identification of chemical species eluted from a liquid chromatography device using spectral information. Machine learning techniques such s random forests and convolution neural networks established by deep learning were validated and performance tested against human and conventional statistical methods (PCA, LDA) and shown to have superior performance. 
Type Of Material Data analysis technique 
Year Produced 2020 
Provided To Others? No  
Impact The model has enabled the autonomous identification of species in a hand-portable liquid chromatography prototype. 
 
Title Spectral Identification of Species 
Description Code to analyse multispectral chromatographic data and identify with a confidence score species using public or non-public reference spectra. 
Type Of Material Data analysis technique 
Year Produced 2020 
Provided To Others? No  
Impact The technique will underpin manuscripts and de-restricts research avenues available to the research group. 
 
Description Agilent Technologies UK - Portable HPLC 
Organisation Agilent Technologies
Country United States 
Sector Private 
PI Contribution Development of knowledge and expertise pertaining to portable liquid chromatography systems.
Collaborator Contribution Access to equipment and methodological expertise as pertaining to liquid chromatography.
Impact The development of knowledge and expertise pertaining to commercial liquid chromatography systems and their comparative analytical performance with portable systems.
Start Year 2018
 
Description Heraeus Noblelight Ltd 
Organisation Heraeus
Department Heraeus Noblelight Ltd
Country United Kingdom 
Sector Academic/University 
PI Contribution Expertise and knowledge in detection strategies of chemical species in the field.
Collaborator Contribution Expertise and production of broadband, low electrical power light sources for field chemistry.
Impact No outputs at the time of submission.
Start Year 2020
 
Description Novartis - Counterfeit Pharmaceuticals in the Field 
Organisation Novartis
Country Global 
Sector Private 
PI Contribution Expertise and knowledge pertaining to field based analytical methods using portable liquid chromatographic systems.
Collaborator Contribution Expert, knowledge and reagents pertaining to anti-counterfeiting measures and drug authenticity in the field.
Impact Methodology and techniques for the detection of drug counterfeits in the field.
Start Year 2018
 
Title Portable HPLC Software Control 
Description Code to operate, control and read data from a portable liquid chromatography instrument. System utilise open source code to communicate to instruments from various manufacturers and communication protocols. 
Type Of Technology Systems, Materials & Instrumental Engineering 
Year Produced 2019 
Impact Manuscripts.