Serum metabolomic profiles for discriminating early from metastatic disease and predicting relapse in patients with colorectal cancer (CRC): METCOLON

Background: The prognosis of early colorectal cancer differs according to the stage of disease. Presently, TNM staging is the main tool to predict risk of relapse and potential benefit derived from adjuvant treatment. Adjuvant treatment is indicated according to pathological staging and risk stratification, but a significant proportion of patients who receive chemotherapy derive no benefit from treatment. Therefore, biomarkers to improve risk stratification are needed to reduce the number of patients treated unnecessarily. Additionally, it is proven that radical resection of metastatic lesions may give survival benefit in candidate cases; however, metastatic spread is not always timely detected during post-surgical follow up. Biomarkers of early detection of disease relapse would therefore be extremely clinically relevant. We plan to use, in a prospective study performed at the Hospital of Prato, the metabolomic profile of patients with CRC as a biomarker to identify those who are at highest risk of recurrence, and therefore may be more likely to benefit from adjuvant chemotherapy (METCOLON Study). Methods: A population of 41 patients with histologically proven stage I-III CRC, candidate for radical resection were prospectively enrolled in the study. Blood samples were taken from each patient before surgery (T0) and after surgery (T1) and serum samples were analyzed using 1H-NMR spectroscopy to characterize the metabolomic profile. To investigate the differences of metabolomic profiles before and after surgery, OPLS-DA, mPLS) were applied. Results: Using the mPLS, serum samples were correctly classified between pre-operative (T0) and post operative (T1), and showed significant differential clustering, with good separation of the two groups with the all 3 spectra CPMG, NOESY and DIFFUSION Three metabolites (3-Hydroxybutyric_acid, pyruvic acid, acetone) were found to differ significantly (p < 0.05) between the pre-operative and post-operative metabolomic profiles. In particularly pyruvic acid was found high in the post operative samples and the acetone and 3-Hydroxybutyric_acid in the pre-operative samples. Four patients experienced disease relapse in our cohort. For the first ten patients interestingly the pre-operative spectra of two of these relapsed clustered separately from those of disease-free patients suggesting it may be possible to distinguish the metabolomic profile of patients who then presented disease recurrence . Conclusions: Metabolomic analysis seems to correctly classify pre- and post- operative serum samples of patients with CRC. It also seems that metabolomic analysis may be able to correctly identify patients with recurrence from those who have not presented it, even if the events are still numerically few. Although interesting this data are based on a small number of patients with a relatively few relapse events and a short follow up and require further validation.