Application Of Indirect Method To Validate And Establish Biological Reference Intervals For Serum Tumor Markers Of Lung Cancer

ObjectiveEstablish the biological reference intervals of serum Carcinoembryonic antigen(CEA),Neuron-specific enolase(NSE),Progastrin-releasing peptide(proGRP),Cytokeratin 19 fragments(CYFRA21-1)by indirect method,and compare them with the current reference intervals.MethodsAll results of CEA,NSE,proGRP,and CYFRA21-1 that were stored in our laboratory information system were collected from January 2014 to December 2017.After excluding the outliers,220 538 cases of CEA,22338 cases of NES,2 971 cases of proGRP,and 39 677 cases of CYFRA21-1 were included to establish the reference intervals.Groups were divided according to gender and age.The rank-sum test was applied in comparing serum tumor marker levels of each two adjacent groups.Adjacent age groups with no statistically significant difference(P>0.05)were combined.All groups showed non-normal distribution afterKolmogorov-Smirnov test,and the non-parametric method was used to establish the biological reference intervals of tumor markers.ResultsThe biological reference intervals of serum CEA,NSE,proGRP,and CYFRA21-1 in gender and age groups were established by indirect method.The upper limits of the reference intervals of the 4 projects are related to age and gradually increased with age.The differences between different genders of CEA,NSE and CYFRA21-1 were statistically significant(P<0.05).The relative deviation of NSE,proGRP,CYFRA21-1 are less than the corresponding reference change value(RCV)(<37.13%,16.26%,62.47%),suggesting that there are no significantly difference between the self-established reference intervals and the current reference intervals.The relative deviation of CEA is significantly greater than its corresponding RCV(>36.31%),indicating that the difference between the self-established reference intervals and the current reference intervals are statistically significant.ConclusionConditional laboratories should use big data for validation of biological reference intervals.The indirect method is a simple and feasible way to verify and establish biological reference intervals by using big data,which has a good clinical application value.