Vibrational spectroscopy and multivariate analysis techniques in the clinical immunology laboratory: a review of current applications and requirements for diagnostic use

Laboratory tests are essential for clinicians to reach an accurate diagnosis and informing appropriate treatments. The expansion in the use of immunotherapies has highlighted the gap between the knowledge of molecular pathways and targeted therapies with availability of laboratory tests. The translation of vibrational spectroscopic techniques such as Fourier-transform infrared (FTIR) spectroscopy and Raman spectroscopy into clinical practice offer rapid-, noninvasive and inexpensive methods to obtain information on the molecular composition of biological samples. Advances in instrumentation, data analysis and machine learning techniques are key developments that have permitted the availability of results to clinicians in an appropriate timescale. Immunological disorders are complex, often demonstrating interaction across multiple molecular pathways which results in delayed diagnosis. Vibrational spectroscopy is being applied in many fields and here we present a review of its potential use in clinical immunology. This review addresses the potential use of spectroscopy in clinical immunology. Potential benefits that these novel techniques offer, including enhanced definition of molecular process and its use in disease diagnosis, monitoring and treatment response is discussed. Whilst not covered extensively, an overview of the method principle, quality control processes, and the requirements for multivariate data analysis is included to provide the reader with sufficient understanding of its application in the clinical setting.

Automated summary

Laboratory tests are crucial for accurate diagnosis and treatment decisions in clinical practice. The translation of vibrational spectroscopic techniques into clinical use offers rapid, noninvasive, and cost-effective methods for obtaining molecular information of biological samples. Advancements in instrumentation, data analysis, and machine learning have facilitated timely availability of results to clinicians.