ABSTRACT
BACKGROUND: We previously reported that CEA kinetics are a marker of progressive disease (PD) in metastatic colorectal cancer (mCRC). This study was specifically designed to confirm CEA kinetics for predicting PD and to evaluate CA19-9, cell-free DNA (cfDNA), circulating tumour DNA (ctDNA) and circulating tumour cell (CTC) kinetics. METHODS: Patients starting a chemotherapy (CT) with pre-treatment CEA > 5 ng/mL and/or CA19.9 > 30 UI/mL were prospectively included. Samples were collected from baseline to cycle 4 for CEA and CA19-9 and at baseline and the sixth week for other markers. CEA kinetics were calculated from the first to the third or fourth CT cycle. RESULTS: A total of 192 mCRC patients were included. CEA kinetics based on the previously identified >0.05 threshold was significantly associated with PD (p < 0.0001). By dichotomising by the median value, cfDNA, ctDNA and CA19-9 were associated with PD, PFS and OS in multivariate analysis. A circulating scoring system (CSS) combining CEA kinetics and baseline CA19-9 and cfDNA values classified patients based on high (n = 58) and low risk (n = 113) of PD and was independently associated with PD (ORa = 4.6, p < 0.0001), PFS (HRa = 2.07, p < 0.0001) and OS (HRa = 2.55, p < 0.0001). CONCLUSIONS: CEA kinetics alone or combined with baseline CA19-9 and cfDNA are clinically relevant for predicting outcomes in mCRC. TRIAL REGISTRATION NUMBER: NCT01212510.
Subject(s)
Antigens, Tumor-Associated, Carbohydrate/metabolism , Carcinoembryonic Antigen/metabolism , Circulating Tumor DNA/genetics , Colorectal Neoplasms/drug therapy , Neoplastic Cells, Circulating/metabolism , Adult , Aged , Aged, 80 and over , Colorectal Neoplasms/genetics , Colorectal Neoplasms/pathology , Female , Gene Expression Regulation, Neoplastic , Humans , Male , Middle Aged , Neoplasm Metastasis , Neoplastic Cells, Circulating/drug effects , Prospective Studies , Survival Analysis , Up-RegulationABSTRACT
BACKGROUND: Although carbohydrate antigen 19.9 (CA19.9) is widely used in pancreatic adenocarcinoma (PA), no consensual cut-off value of CA19.9 decrease has been established for treatment monitoring. METHODS: This was a retrospective study including patients with a baseline CA19.9 ≥ 37 UI/ml and with locally advanced or metastatic PA from two French centers. CA19.9 measurements were performed at baseline and first CT-scan evaluation. The aim was to use a training set to determine the best cut-off of CA19.9 decrease for predicting progressive disease (PD) and to analyze its performance in an independent validation cohort. RESULTS: A total of 95 and 93 patients were included in the training and validation sets, respectively. A ≤15% CA19.9 decrease was the best cut-off for predicting PD with a sensitivity (Se) = 68% and a specificity (Sp) = 90%. In the validation set, this threshold was associated with Se = 76% and Sp = 83%. A >15% CA19.9 decrease was significantly associated with improved PFS (median 8.3 versus 3.1 months, p < 0.0001) and OS (median 14 versus 7.2 months, p < 0.0001). A >15% CA19.9 decrease was also identified as a factor independently associated with OS (HRa = 0.25, 95% CI:0.14-0.44). CONCLUSIONS: A CA 19.9 decrease >15% is a favourable predictor of outcome in patients treated for advanced PA.
