Establishment of cut-off value of serum pro-gastrin-releasing peptide for diagnosis of small cell lung cancer and evaluation on the clinical diagnosis efficiency

Abstract

Objective: To determine critical reference value (cut-off value) of serum pro-gastrin-releasing peptide (ProGRP) and neuron specific enolase(NSE) in the diagnosis of small cell lung cancer(SCLC). To evaluate the clinical significance of serum levels of ProGRP and NSE in diagnosis and differential diagnosis in SCLC. Methods: Three hundred and fifty-two SCLC patients, 163 non small cell lung cancer(NSCLC)patients , 193 benign pulmonary disease patients and 140 healthy people visiting in National Cancer Hospital were analyzed retrospectively from January 2014 to July 2017.The levels of serum ProGRP and NSE of people were determined using electrochemiluminescent immunoassay respectively . Reference value ranges of the makers were determined by using the method of ROC curves. Results: In NSCLC group, benign lung disease group, healthy control group and mixed group (NSCLC+ lung benign diseases+ healthy control group) as a reference, the cut-off values were 58.3, 62.3, 57.8, 61.3 ng/L. In the diagnosis and differential diagnosis of SCLC and NSCLC, benign lung diseases, healthy controls and mixed group, AUC of ProGRP was 0.940 (0.919-0.961), 0.941 (0.921-0.960), 0.959 (0.944-0.975), 0.946 (0.928-0.963) respectively. The sensitivities of ProGRP were 86.4%, 84.9%, 86.4% and 84.7% respectively. The specificities of ProGRP were 95.7%, 96.9%, 99.3%, 98% respectively. In all groups the Youden's index of ProGRP and NSE were 0.821 vs 0.612, 0.818 vs 0.674, 0.857 vs 0.810, 0.827 vs 0.674. In healthy controls, no statistically significant difference was found between ProGRP and NSE (P>0.05) in the diagnosis of AUC. However, in the remaining 3 groups, the ProGRP diagnosis of AUC was significantly greater than that of NSE (P<0.01). Compared with single marker detection, the sensitivity of combined detection of ProGRP and NSE in diagnosis of SCLC increased to 95.5%, 94%, 96.6% and 94% in each group. There was no significant difference between ProGRP and ProGRP+ NSE in the diagnosis of AUC when compared with the NSCLC group and the mixed group (P>0.05). However, when combined with a healthy control group and a benign lung disease group, the ProGRP+ NSE combination was the highest for AUC diagnosis, compared with ProGRP and NSE (P<0.01). In the SCLC ED group serum ProGRP and NSE levels[776.33(3 103.4)ng/L, 52.14(60.59)μg/L]were higher than those in the SCLC LD group[295.59(799.65)ng/L, 23.36(22.97)μg/L], respectively (all P<0.001). The serum ProGRP levels of N0, N1, N2 and N3 in TNM staging were 113.0(343.65), 167.04(724.56), 427.42(1 388.62), 735.99(1 709.95)ng/L respectively (all P<0.001). Serum ProGRP and NSE levels were not statistically different between the sex groups and the age groups (all P>0.05). Conclusion: To establish the cut-off value of serum ProGRP is helpful for the diagnosis and differential diagnosis of SCLC.