Detection of Lung Cancer via Blood Plasma and 1H-NMR Metabolomics: Validation by a Semi-Targeted and Quantitative Approach Using a Protein-Binding Competitor

Abstract

Metabolite profiling of blood plasma, by proton nuclear magnetic resonance (1H-NMR) spectroscopy, offers great potential for early cancer diagnosis and unraveling disruptions in cancer metabolism. Despite the essential attempts to standardize pre-analytical and external conditions, such as pH or temperature, the donor-intrinsic plasma protein concentration is highly overlooked. However, this is of utmost importance, since several metabolites bind to these proteins, resulting in an underestimation of signal intensities. This paper describes a novel 1H-NMR approach to avoid metabolite binding by adding 4 mM trimethylsilyl-2,2,3,3-tetradeuteropropionic acid (TSP) as a strong binding competitor. In addition, it is demonstrated, for the first time, that maleic acid is a reliable internal standard to quantify the human plasma metabolites without the need for protein precipitation. Metabolite spiking is further used to identify the peaks of 62 plasma metabolites and to divide the 1H-NMR spectrum into 237 well-defined integration regions, representing these 62 metabolites. A supervised multivariate classification model, trained using the intensities of these integration regions (areas under the peaks), was able to differentiate between lung cancer patients and healthy controls in a large patient cohort (n = 160), with a specificity, sensitivity, and area under the curve of 93%, 85%, and 0.95, respectively. The robustness of the classification model is shown by validation in an independent patient cohort (n = 72).

Unraveling the Rewired Metabolism in Lung Cancer Using Quantitative NMR Metabolomics

“…The application of metabolomic platforms in the search for cancer biomarkers has increased exponentially over the past decade [ 7 , 41 , 42 , 43 , 44 , 45 , 46 , 47 ]. A frequently used technique to analyze low-molecular-weight molecules in biofluids, such as plasma, is proton nuclear magnetic resonance ( 1 H-NMR)-based [ 48 , 49 ]. Each hydrogen nucleus in a different chemical environment has a different, specific resonance frequency and therefore results in a signal at a different position in the NMR spectrum.…”

“…In addition, the chemical shift values (peak positions) depend on conditions such as pH, concentration, temperature, and ionic strength, resulting in less accurate identification and quantification if not exactly replicated. A way to accurately determine the chemical shifts of the plasma metabolite signals under the experimental conditions used is by spiking plasma samples from a large plasma pool with different, known metabolites [ 49 ]. Hereto, a small quantity of a known metabolite is added to a sample aliquot of the plasma pool and this is repeated for all metabolites of interest.…”

“…In a next step, the spectrum is divided into multiple integration regions of which the integration values can be used as variables in multivariate statistical models. As described by Derveaux et al, metabolite spiking of biofluid samples results in well-defined integration regions representing a single metabolite, or a combination of several metabolites [ 49 ]. The primary goal in metabolomics is to extract discriminating information from complex large datasets.…”

“…As this matrix contains the effect of the intervention on each patient, the variation between (predictive component) and within (orthogonal components) subjects is separated, and intrinsic differences in treatment effect between individuals can be identified. Although the application of 1 H-NMR to lung cancer research is an emerging field, several research groups have established a metabolic profile for lung cancer [ 7 , 8 , 41 , 45 , 49 ]. However, defining prognostic metabolic profiles remains challenging.…”

“…However, TSP binds to human serum albumin which causes fluctuations in the chemical shift and intensity of the TSP signal. Derveaux et al described the methyl signal of alanine as an optimal standard to calibrate the chemical shift ppm scale, and a known concentration of maleate as an ideal internal standard to quantify the metabolites [ 49 ]. The lung cancer versus control classification model obtained by Derveaux et al resulted in better model diagnostics (specificity of 93%, a sensitivity of 85%, and an area under the curve of 0.95) as compared to the model of Louis et al (specificity of 92%, a sensitivity of 78%, and an AUC of 0.88) [ 7 , 49 ].…”