A portable metabolomics-on-CMOS platform for point-of-care testing

Annese, Valerio Francesco (2021) A portable metabolomics-on-CMOS platform for point-of-care testing. PhD thesis, University of Glasgow.

Abstract

Metabolomics is the study of the metabolites, small molecules produced during the metabolism. Metabolite levels mirror the health status of an individual and therefore have enormous potential in medical point-of-care (POC) applications. POC platforms are miniaturised and portable systems integrating all steps from sample collection to result of a medical test. POC devices offer the possibility to reduce the diagnostic costs, shorten the testing time, and, ultimately, save lives for several applications. The glucose meter, arguably the most successful example of metabolomics POC platform, has already demonstrated the dramatic impact that such platforms can have on the society. Nevertheless, other relevant metabolomic tests are still relegated to centralised laboratories and bulky equipment.
In this work, a metabolomics POC platform for multi-metabolite quantification was developed. The platform aims to untap metabolomics for the general population. As case studies, the platform was designed and evaluated for prostate cancer and ischemic stroke. For prostate cancer, new affordable diagnostic tools to be used in conjunction with the current clinical standard have are needed to reduce the medical costs due to overdiagnosis and increase the survival rate. Thus, a novel potential metabolic test based on L-type amino acids (LAA) profile, glutamate, choline, and sarcosine blood concentrations was developed.
For ischemic stroke, where the portable and rapid test can make a difference between life and death, lactate and creatinine blood levels were chosen as potential biomarkers. All the target metabolites were quantified using an optical method (colorimetry).
The platform is composed of three units: the cartridge, the reader, and the graphical user interface (GUI). The cartridge is the core of the platform. It integrates a CMOS 16x16 array of photodiodes, capillary microfluidics, and biological receptors onto the same ceramic package. To measure multiple metabolites, a novel method involving a combination of replica moulding and injection moulding was developed for the monolithic integration of microfluidics onto integrated chips.
The reader is composed of a custom PCB and a microcontroller board. It is used for addressing, data digitisation and data transfer to the GUI. The GUI - a software running on a portable electronic device - is used for interfacing the system, visualise, acquire, process, and store the data.
The analysis of the microfluidic structures showed successful integration. The selection of the specific chemistry for detecting the analytes of interest was demonstrated to be suitable for the performance of the sensors. Quick and reliably capillary flow of human plasma, serum and blood was demonstrated.
On-chip quantification of the target metabolites was demonstrated in diluted human serum and human plasma. Calibration curves, kinetics parameter and other relevant metrics were determined. For all the metabolites, the limits of detection were lower than the physiological range, demonstrating the capability of the platform to be used in the target applications.
Multi-metabolite testing capability was also demonstrated using commercially and clinically sourced human plasma. For multiplexed assays, reagents were preloaded in the microfluidic channel and lyophilised. Lyophilisation also improved the shelf-life of the reagents. Alternative configurations, involving the use of paper microfluidics, integration of passive blood filter and use of whole blood, were investigated.
The chracterisation of the platform culminated with a clinical evaluation for both the target applications. The same platform with minimal modification of the cartridge was able to provide clinically relevant information for both the distinct applications, highlighting the versatility of the platform for POC determination of metabolic biomarkers.
For prostate cancer, the platform was used for the quantification of the potential metabolic biomarker in 10 healthy samples and 16 patients affected by prostate cancer. LAA, glutamate and choline average concentrations were elevated in the cancer group with respect to the control group and were therefore regarded as metabolic biomarkers in this population. Metabolomic profiles were used to train a classifier algorithm, which improved the performance of the current clinical blood test, for this population.
For ischemic stroke, lactate determination was performed in clinically sourced samples. Clinical evaluation for ischemic stroke was performed using 10 samples from people diagnosed with ischemic stroke. Results showed that the developed platform provided comparable results with an NHS-based gold standard method in this population. This comparison demonstrated the potential of the platform for its on-the-spot use.
The developed platform has the potential to lead the way to a new generation of low-cost and rapid POC devices for the early and improved diagnosis of deadly diseases.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Keywords: point-of-care, metabolomics, CMOS, microfluidics, personalised medicine, enzymatic assays, optical detection.
• Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering
• Colleges/Schools: College of Science and Engineering > School of Engineering > Electronics and Nanoscale Engineering
• Funder's Name: Engineering and Physical Sciences Research Council (EPSRC), Engineering and Physical Sciences Research Council (EPSRC)
• Supervisor's Name: Cumming, Prof. D. R. S.
• Date of Award: 2021
• Depositing User: Valerio Francesco Annese
• Unique ID: glathesis:2021-82001
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 10 Feb 2021 17:04
• Last Modified: 14 Nov 2022 14:31
• Thesis DOI: 10.5525/gla.thesis.82001
• URI: https://theses.gla.ac.uk/id/eprint/82001

Related URLs

• Article DOI
• Article DOI
• Article DOI
• Article DOI
• Article DOI
• Article DOI
• Data DOI