The Correlation Between Serum Tumor Markers With EGFR Mutation And The Efficacy Treated By TKIs In Lung Adenocarcinoma

Objective:1. To explore the correlation between initial serum tumor markers and EGFR mutation in the lung adenocarcinoma patients.2. To explore the prediction significance of initial serum tumor markers to efficacy and survival who treated with EGFR-TKIs in lung adenocarcinoma patients acquiring EGFR mutation.The first partThe correlation between serum tumor markers and EGFR mutation in lung adenocarcinoma patients.Methods:1. The 131 cases of lung adenocarcinoma patients which confirmed by histopathology were collected from the First Hospital Affiliated to Dalian Medical University since March 2012 to March 2014. Recorded clinical data of all patients, including the patients' gender, age, smoking history, ECOG score, specimens source, degree of differentiation, clinical staging, the results of EGFR gene mutation and initial serum tumor markers.2. EGFR gene was detected by new Ion Torrent genetic sequencing technology. Serum tumor markers in cases who had not yet undergone surgery, radiotherapy, chemotherapy, targeted therapy and other anti-tumor treatment were assessed by electrochemical luminescence.3. Using SPSS 22.0 statistical software for data processing. Count data described as a rate, and clinical factors were analyzed by single factor x2test or Fisher exact test and multivariate Logistic regression. The ROC curve was drawn to analysis the value of tumor markers predicting EGFR mutations. All significance levels were used two-sided test, and P<0.05 was considered statistically significant.Results:EGFR mutation rate is 49.6% in lung adenocarcinoma. Mutations are mainly concentrated in the region of the 18-21 exon.19 exon and 21 exon mutation are the most common, the mutation rate are 20.6% and 25.2% respectively. EGFR mutation rate of high-medium differentiation is higher than low differentiation (57.1%vs32.5%, P=0.009). EGFR mutation rate of abnormal CEA levels is higher than normal CEA level (63.2%vs39.2%, P=0.007). The results of set groups showed that the 19 exon mutation rate in the group of abnormal CEA levels is higher than the group of normal CEA level (29.8%vsl3.5%, P=0.016), but there is no statistically significant difference of 21 exon mutation rate (29.8%vs21.6%, P=0.358). In both men and women, EGFR mutation rate of the abnormal CEA levels group is higher than the group of normal CEA level (60.0%vs35.3%, P=0.060; 65.6% vs42.5%, P=0.051).Furthermore, Multiariable Logistic regression analysis shows that the serum CEA level is an independent predictor of EGFR mutation and 19 exons mutation. Draw the ROC curve. When predicting EGFR mutation, the areas of CEA under the curve is 0.691, the sensitivity Se=50.9%, specificity Sp=63.1%. When predicting 19 exon mutation, the areas of CEA under the curve is 0.762, the sensitivity Se=61.0%, specificity Sp=69.5%.The second partThe correlation between serum tumor markers with the efficacy and prognosis of lung adenocarcinoma acquiring mutation treated by TKIsMethods:1. Followed up the 40 patients with EGFR mutation and acceptting EGFR TKIs treatment (gefitinib, erlotinib or icotinib) as the first line in the first part by outpatient or telephone. Recorded the efficacy, survival situation and analysised clinical factors impacting on the prognosis of patients.2. EGFR gene was detected by new Ion Torrent genetic sequencing technology. Serum tumor markers in cases who had not yet undergone surgery, radiotherapy, chemotherapy, targeted therapy and other anti-tumor treatment were assessed by electrochemical luminescence.3. Used SPSS 22.0 statistical software for data processing. Clinical factors were analyzed by single factor ?2 test or Fisher exact test. Used Kaplan and Meier method to calculate survival rate and draw the survival curve. Used log-rank test to compare the different groups. Used Cox proportional hazards regression for multivariate analysis. All significance levels were used two-sided test, and P<0.05 was considered statistically significant. Because there were only 2 cases in the group of initial serum NSE>24 ng/ml and 1 cases in the group of initial serum SCC>2.5 ng/ml, so we didn't analysis the influence of NSE and SCC levels.Results:The objective response rate (RR) and disease control rate (DCR) are 70% and 95%, respectively. The median PFS and median OS are 9.4 months (95% CI: 7.08?11.05 months) and 19.7 months (95% CI:15.56?24.15 months), respectively. The RR in the group of CEA<5 ng/ml and CEA?5 ng/ml are 78.6% and 70.8% (P= 0.601). There is no statistically significant difference of median PFS betweet the two groups (10.0 months vs9.0, P=0.361). The median OS in the group of CEA<5 ng/ml is longer than the group of CEA?5ng/ml (20.9 months vs 14.4 months, P=0.013). And there is no significant correlation between serum CA19-9 levels, serum CYFRA21-1 with RR, PFS and OS. COX multi-factor analysis shows that smoking status, ECOG score and mutation type are the independent factors of PFS, and smoking status, ECOG score, mutation type, differentiation degree and the serum CEA level are the independent factors of OS.Conclusion:1. EGFR mutation rate (especially 19 exon mutation rate) in the group of abnormal serum level is significantly higher than the group of normal CEA level in lung adenocarcinoma patients. Therefore serum CEA level to a certain extent can further determine the advantage groups of EGFR mutations.2. The CEA levels don't affect the EGFR TKIs curative effect in the lung adenocarcinoma patients with EGFR mutation who treat by EGFR TKIs in the first line, but the patients of CEA<5 ng/ml have longer OS.