The value of FDG PET/CT parameters, hematological parameters and tumor markers in predicting KRAS oncogene mutation in colorectal cancer.

OBJECTIVE

In this study we investigated the predictive value of maximum standardized uptake value (SUVmax), metabolic tumor volume (MTV), total lesion glycolysis (TLG), neutrophils/lymphocytes ratio (NLR), platelets/lymphocytes ratio (PLR), carcinoembryonic antigen (CEA) and carbohydrate antigen (CA 19-9) in the prediction of KRAS gene mutation which plays an important role in the choice of treatment in colorectal cancer patients.

SUBJECTS AND METHODS

A total of 55 cases with untreated colorectal cancer who had undergone both PET/CT examinations for initial staging and also mutation analysis of KRAS oncogene were studied. Fluorine-18-FDG PET/CT parameters (SUVmax, MTV, TLG), hematological parameters (NLR, PLR), and tumor markers (CEA, CA 19-9) were recorded and the relationship between these parameters and KRAS oncogene mutation was evaluated using receiver operating characteristics (ROC) analysis and multiple logistic regression analysis.

RESULTS

In 20 cases mutations in the KRAS gene were detected, while in 35 cases mutations were not observed (wild-type). ROC analysis revealed that SUVmax, MTV, TLG, NLR, PLR, and CA 19-9 could not predict mutations in KRAS oncogene (P=0.600, 0.263, 0.214, 0.057, 0.104, 0.189, respectively) although CEA value showed signi..cant difference (P=0.031) but without high value of the area under the curve (0.676). Multivariate logistic regression analysis also did not show significant association between KRAS gene mutations and SUVmax, MTV, TLG, NLR, PLR, CEA, CA 19-9 values.

CONCLUSION

We observed that in patients with colorectal cancers, we cannot predict KRAS gene mutations using PET/CT parameters (SUVmax, MTV, TLG), hematological parameters (NLR, PLR) or tumor marker CA 19-9. We detected a significant but not very strong association only between CEA and KRAS mutations.