Colorectal cancer (CRC) is a growing cause of mortality in developing countries, warranting investigation into its earlier detection for optimal disease management. Global proton nuclear magnetic resonance (1H NMR)-based metabolomics on stool specimens opens up new possibilities for screening new, efficient and high-throughput CRC screening biomarkers[1-5].This study aimed to validate the ability of NMR-based fecal metabolomics fingerprinting as predictors of earlier diagnosis in CRC patients.

Stool samples were collected from 68 CRC patients (stage I/II=20; stage III=25 and stage IV=23) and 32 healthy controls (HC). Samples were extracted with PBS/D20 buffer. A stock solution of TSP/D20 was added to supernatant prior to analysis by 1H NMR spectroscopy. Spectra of the fecal extracts were recorded with a 400MHz Bruker 1D NOESY pulse sequence with the following acquisition parameters: Recycle Delay, RD=1.5s; t1 =3µs; mixing time, tm =100ms; 90o pulse width=7.3us; number of scans, NS=64; number of points, TD=16380; spectral width, SW=5000Hz; acquisition time, AQ=1.47s. Water suppression was achieved by irradiation of the water peak during RD and tm. All spectra were preprocessed and then bucketed with the equal width of 0.002ppm. The region of δ 4.4~ 5.6 was discarded to eliminate the imperfect water suppression. Each bucket was normalized to the total integral of the spectrum prior to OPLS-DA using the SIMCA-P+ program. Representative 1D 1H NMR spectra of fecal extracts obtained from HC and different stages of CRC are shown in Fig 1. The standard one-dimension spectrum gave an overview of all metabolites. Good discrimination between each stage of CRC and HC was achieved by OPLS-DA scores plot generated from 1H NMR spectra of fecal extracts (Fig.2(A)). Model parameters of permutation analysis for different stages were as follows: stage I/II: R2 = 0.949, Q2 = 0.685; stage III: R2 = 0.88, Q2 = 0.574 and stage IV: R2 = 0.789, Q2 = 0.618, which indicated the good fit obtained by the model (Fig 2(B)). The training and testing set evaluations further validated the predictive power of the model, in which the testing set samples were correctly classified as either CRC group or healthy controls (Fig 2(C)). Distinct disturbances to fecal metabolites of CRC patients at various stages were identified, compared with those in cancer free controls, including down-regulation of short chain fatty acids(acetate, butyrate, propionate), glucose, glutamine, and up-regulation of glycerol, proline, alanine, asparagine, valine, glutamate, leucine, isoleucine and lactate (Fig3). There have only been a few reports of fecal metabolic changes associated with CRC to date; previous 1H NMR-based metabolomics studies already suggested fecal metabolic alterations between CRC patients and healthy controls[1-2]. However, none have described early changes to the fecal metabolic profile. Our study was designed to investigate different patterns between stages of CRC patients compared to healthy controls, and to identify patients with early stage (stage I/II). Our results revealed distinct fecal metabolites which were differentially abundant in all stages of disease, including acetate, propionate, butyrate, glucose and glutamine (reduced in CRC), as well as proline, isoleucine, leucine, valine, alanine, glutamate, asparagines, lactate and glycerol (elevated in CRC), relative to healthy controls. The fecal 1H-NMR metabolic profiles of healthy controls were well discriminated from those of even early stage (stage I/II) CRC patients (Fig 2). In addition, the metabolomics of feces at early stage differed markedly from those at later stages (Fig 3). Our findings indicated that the difference in fecal NMR spectral profiles between diseased and non-diseased patients faithfully depicts the pathophysiological changes and metabolic disturbances observed at the different phases of the disease progression, highlighting the benefits of NMR-based fecal metabolomics as a potential noninvasive strategy to identify biomarkers for CRC earlier diagnosis. More work will be required to delineate the contributions of colonic polyps and inflammatory bowel disease (precancerosis) on the fecal metabolic perturbations.