PURPOSE: Carcinoembryonic antigen (CEA) is a tumor marker widely used for nonsmall cell lung cancer (NSCLC). The aim of this study was to evaluate changes in serum CEA levels as a surrogate marker for tumor response to chemotherapy in NSCLC. METHODS: From 1995 through 2005, we retrospectively analyzed 24 NSCLC patients who had high serum CEA levels (>5 ng/ml) and who received chemotherapy followed by surgery. We compared serum CEA levels with tumor response, as defined by Response Evaluation Criteria in Solid Tumors (RECIST) or World Health Organization (WHO) criteria, as well as with histological response. RESULTS: Serum CEA levels after chemotherapy significantly decreased in patients who achieved partial response, defined by RECIST or WHO criteria (p = 0.004 and p = 0.008, respectively), when compared with the CEA levels before chemotherapy. In contrast, there was no significant difference in CEA levels in patients with either stable disease or no response to chemotherapy. They decreased significantly, however, in patients in whom less than one-third of tumor cells was viable by pathological examination, but not in patients in whom more than a third was viable (p = 0.008). Using the receiver-operating characteristic (ROC) curve analysis, we found that a 60% reduction of CEA levels was an appropriate cutoff value in predicting a good response to chemotherapy. When the value was set at that level, the sensitivity of CEA for RECIST was 82%, and the specificity was 69%. CONCLUSION: Serum CEA concentration was a useful surrogate marker for the evaluation of tumor response to chemotherapy and seemed to be comparable with RECIST in NSCLC patients who had elevated CEA levels prior to treatment.
PURPOSE:Carcinoembryonic antigen (CEA) is a tumor marker widely used for nonsmall cell lung cancer (NSCLC). The aim of this study was to evaluate changes in serum CEA levels as a surrogate marker for tumor response to chemotherapy in NSCLC. METHODS: From 1995 through 2005, we retrospectively analyzed 24 NSCLCpatients who had high serum CEA levels (>5 ng/ml) and who received chemotherapy followed by surgery. We compared serum CEA levels with tumor response, as defined by Response Evaluation Criteria in Solid Tumors (RECIST) or World Health Organization (WHO) criteria, as well as with histological response. RESULTS: Serum CEA levels after chemotherapy significantly decreased in patients who achieved partial response, defined by RECIST or WHO criteria (p = 0.004 and p = 0.008, respectively), when compared with the CEA levels before chemotherapy. In contrast, there was no significant difference in CEA levels in patients with either stable disease or no response to chemotherapy. They decreased significantly, however, in patients in whom less than one-third of tumor cells was viable by pathological examination, but not in patients in whom more than a third was viable (p = 0.008). Using the receiver-operating characteristic (ROC) curve analysis, we found that a 60% reduction of CEA levels was an appropriate cutoff value in predicting a good response to chemotherapy. When the value was set at that level, the sensitivity of CEA for RECIST was 82%, and the specificity was 69%. CONCLUSION: Serum CEA concentration was a useful surrogate marker for the evaluation of tumor response to chemotherapy and seemed to be comparable with RECIST in NSCLCpatients who had elevated CEA levels prior to treatment.