Abstract
BACKGROUND: Previously, it was demonstrated that serum levels of tumor markers, CEA and CA19-9, correlated with chemotherapy. Consequently, it has been hypothesized that dynamic monitoring of changes in these markers may predict the shrinkage or growth of colorectal cancers. To test this hypothesis, we analyzed CEA and CA19-9 serum levels in patients with advanced colorectal cancer who received cetuximab in combination with chemotherapy. These levels were evaluated at various time points to identify their potential to serve as early efficacy predictors during treatment and early predictors of disease progression. PATIENTS AND METHODS: Measurements of tumor markers, CEA and CA 19-9, in patients with metastatic colorectal cancer (n = 73) who received cetuximab plus folinic acid, fluorouracil, and oxaliplatin or irinotecan (FOLFOX4/FOLFIRI) as a first-line treatment at our center were retrospectively analyzed. These levels were also compared with objective responses according to the World Health Organization criteria. Initially, 65 patients had elevated CEA levels (>5 ng/ml), and 59 patients had elevated levels of CA19-9 (>37 U/ml). A total of 172 cycles and 165 cycles of computed tomography/magnetic resonance imaging observations were available for review from these two patient groups. RESULTS: After completing three cycles of treatment, the best diagnosis of cetuximab resistance was achieved when CEA increased by 35% (efficacy, 83.33%; sensitivity, 75.41%) and when CA19-9 increased by 28% (efficacy, 80.00%; sensitivity, 84.31%). Next, the efficacy of cetuximab at the time of diagnosis (at the first imaging examination/after three cycles of treatment) was evaluated after the first cycle of chemotherapy. When CEA decreased by 60% from its baseline level, the best effective rate and sensitivity were observed (63.64% and 80.95%, respectively). Similarly, when CA19-9 was 45% lower than its baseline level, the best effective rate and sensitivity were observed (84.21% and 93.18%, respectively). To evaluate progression-free survival (PFS), levels of both CEA and CA19-9 were evaluated after the third cycle of chemotherapy. Increases of 35% and 28%, respectively, resulted in a shorter PFS period compared with the other patients (3.15 months vs. 9.10 months, respectively; P < 0.0001). Conversely, when the evaluation was performed after the first cycle of chemotherapy, patients exhibiting a 60% decrease in CEA and a 45% decrease in CA19-9 had a longer PFS period (11.13 months vs. 8.10 months, respectively; P = 0.0395). CONCLUSIONS: CEA and CA19-9 are useful indicators of therapeutic curative effect from cetuximab combined with first-line chemotherapy. These markers also helped assess cetuximab resistance and served as early predictors of initial treatment effectiveness. Furthermore, a simultaneous increase or decrease in the levels of both indicators was consistent with the observed differences in PFS.