Purpose: To assess prognostic/predictive value of carcinoembryonic antigen (CEA), transthyretin (TRT), αenolase (NNE), β2-microglobulin (β2-micro), B-cell activating factor (BAFF) and circulating tumor cells (CTCs) in metastatic colorectal cancer (mCRC) patients treated with chemotherapy with or without bevacizumab.
Methods: 72 histologically confirmed mCRC patients treated at Oncology Institute Cluj were included. Biomarker levels were measured through validated methods. A manual method was used for CTCs, involving hemolysis, cytospin centrifugation and immunocytochemical staining for pan-cytokeratin. Statistical endpoints were response, progression- free survival (PFS) and overall survival (OS).
Results: Initial chemotherapy was fluoropyrimidine/oxaliplatin-based in 93.1%; bevacizumab was added in 58.3% of the patients. Median PFS and OS were 16.4 and 24.4 months. Two-year OS for CR & PR vs SD vs PD were 90% vs 48% vs 12%, respectively (p<0.01). Two-year OS for chemo/ bevacizumab vs chemotherapy: 65% vs 42% (p=0.09). Baseline CEA ≥5 ng/ml had a negative prognostic impact on OS and PFS (p<0.01). High baseline CEA was predictive of improved OS when adding bevacizumab (2-year OS chemo/bevacizumab vs chemo: 60% vs 17%, p<0.01); adding bevacizumab in patients with normal CEA did not improve OS (p=0.29). Higher than cut-off values for TRT had a positive OS prognostic value (p<0.01); higher levels for NNE, β2-microglobulin and BAFF had a negative impact (p<0.01). Two-year OS for baseline <1 CTC/ml vs ≥1 CTC/ ml was 74% vs 64% respectively (p=0.15).
Conclusions: The evaluated biomarkers could be useful prognostic factors for survival. Baseline CEA also has predictive value, suggesting that patients with low levels do not benefit from bevacizumab. A non-statistically significant correlation was observed between the number of CTCs and outcome.