The aim of this study is to establish the reference intervals (RIs) for serum gastrointestinal tumor markers (AFP, CEA, CA199 and CA724) in healthy adults of Han nationality in Chongqing by Roche ECLIA system. According to CLSI EP28-A3, the appropriate statistical analysis method was selected and the RIs were determined using the indirect method based on the analysis results. All the test data of the four research projects showed a skewness distribution (p < .05). Gender and age are two important factors influencing the level of detection value. According to the analysis results, the data of the four research projects were grouped by gender and age, and the non-parametric percentile method (95%, double-sided) was used to establish the RIs of each group. The established RIs were validated using the method of Reference Change Value (RCV). The relative deviations between the upper limit value of reference interval in each group of AFP and CEA and that provided by the manufacturer is all smaller than the RCV of eath, so the RIs established by this study is proved to be reliable. While, the relative deviation of the group CEA (male > 40 years old, female > 60 years old) and CA724 is higher than each RCV, which indicates that the RIs provided by the manufacturer is not applicable to the test population and test system of the laboratory. We established RIs of serum gastrointestinal tumor markers (AFP, CEA, CA199 and CA724) in healthy adults of Han nationality in Chongqing by Roche ECLIA system. Furthermore, our study suggests that it is necessary to establish the age-specific and sex-specific RIs for gastrointestinal tumor markers in different detection systems.
Keywords: AFP; CA199; CA724; CEA; Reference interval; tumor marker.