Serum tumor markers for the prediction of concordance between genomic profiles from liquid and tissue biopsy in patients with advanced lung adenocarcinoma.

ABSTRACT

BACKGROUND: The concordance between mutations detected from plasma and tissue is critical for treatment choices of patients with advanced lung adenocarcinoma. METHODS: We prospectively analyzed the association of the serum tumor markers with the concordance between blood and tissue genomic profiles from 185 patients with advanced lung adenocarcinoma. The concordance was defined according to 3 criteria. Class 1 included all targetable driver mutations in 8 genes; class 2 included class 1 mutations plus mutations in KRAS, STK11, and TP53; class 3 included class 2 mutations plus tumor mutation burden (TMB) status. RESULTS: Collectively, 150 out of 185 patients had mutations in both tissue and plasma samples, while one patient was mutation-negative for both, resulting a concordance of 81.6%. The concordance rate for class 1 mutations was 80%, and 65% and 69% for class 2 and class 3, respectively. Carbohydrate antigen 19-9 (CA19-9) or cytokeratin 19 (CYFRA21-1) levels higher than the normal upper limit predicted the concordance of tissue and blood results in class 1 (P=0.005, P=0.011), class 2 (P=0.011, P<0.001), and class 3 (P=0.001, P=0.014). In class 1, the cutoff values of CA19-9 were 30, 36, and 284 U/mL to reach the concordance thresholds of 90%, 95%, and 100%, respectively (P=0.032, P=0.003, P=0.043). For CYFRA21-1, the cutoff values were 6, 18, and 52 µg/L (P=0.005, P=0.051, P=0.354). In class 2, the cutoff values for CYFRA21-1 were 18, 22, and 52 µg/L (P=0.001, P=0.001, P=0.052). In class 3, the cutoff values for CA19-9 were 36, 39, and 85 U/mL (P=0.003, P=0.001, P=0.008). For CYFRA21-1, the cutoff values were 22, 52, and 52 µg/L (P=0.900, P>0.99, P>0.99). When the sum score for 4 serum tumor markers was greater than 35, both class 1, class 2, and class 3 reached a predictive threshold of 90%. CONCLUSIONS: Serum tumor markers can be used as easy and practical clinical predictors of concordance in mutation profiles between blood and tissue samples from patients with advanced lung adenocarcinoma.