The Development and Optimisation of a Urinary Volatile Organic Compound Analytical Platform Using Gas Sensor Arrays...

Area of Science:

• Biochemistry
• Medical Diagnostics
• Oncology

Background:

• Volatile Organic Compounds (VOCs) profiles in urine offer potential for non-invasive cancer screening.
• Urine analysis is patient-acceptable for disease detection compared to other methods.
• Early detection of colorectal cancer (CRC) and precancerous polyps is crucial for improved patient outcomes.

Purpose of the Study:

• To evaluate the efficacy of urine VOC analysis using an electronic nose for early cancer detection.
• To develop an enhanced data-processing strategy for improved accuracy in identifying significant bowel diseases.
• To assess the potential of urine VOCs as a rapid, low-cost biomarker for CRC and high-risk polyps.

Main Methods:

• Urine samples analyzed via an electronic nose with metal oxide gas sensors and solid-phase microextraction.
• A radial basis function neural network was trained using a calibration dataset of CRC-positive patients and healthy controls.
• An enhanced data-processing strategy was implemented to categorize samples by significant bowel diseases, including CRC and high-risk polyps.

Main Results:

• Initial blinded analysis showed limitations in accurately detecting CRC.
• Retraining the neural network with the enhanced strategy achieved an area under the ROC curve of 0.88 for distinguishing CRC from non-significant bowel disease.
• The improved method demonstrated effectiveness in categorizing samples related to significant bowel pathologies.

Conclusions:

• Urine VOC analysis, when combined with an enhanced data-processing strategy and appropriate training sets, shows potential for early cancer detection.
• This method could serve as a rapid, low-cost approach for identifying individuals with precancerous colorectal polyps and CRC.
• Further validation is needed to establish urine VOC profiling as a reliable tool in clinical diagnostics for colorectal conditions.