Prognostic value of carcinoembryonic antigen and CYFRA21-1 in patients with pathological stage I non-small cell lung cancer.

BACKGROUND: The aim of this retrospective study was to assess the prognostic value of serum tumor markers (carcinoembryonic antigen (CEA) and CYFRA21-1) in patients with pathologic (p-) stage I non-small cell lung cancer (NSCLC) undergoing complete resection.

METHODS: Two hundred and seventy-five patients (163 males, 112 females, mean age 67.1 years) with p-stage I NSCLC who underwent complete resection at our institution between April 1999 and October 2004 were examined. Patients who had received preoperative chemotherapy or radiotherapy were excluded, as were patients who had multiple malignancies including multiple lung cancer. The serum levels of tumor markers were measured using commercially available immunoassays within 1 month before surgical resection. Serum levels of CEA and CYFRA21-1 higher than 5.0 and 2.8 ng/ml, respectively, were considered as positive according to the manufacture's instructions.

RESULTS: The histological classification was adenocarcinoma in 193 patients, squamous cell carcinoma in 71, large cell carcinoma in 5, and other histological type in 6. One hundred and fifty-seven patients had T1 disease and 118 patients had T2 disease. The positive ratio of CEA and CYFRA21-1 was 25.7% and 13.7%, respectively, and in relation to histological type was 27.8% and 7.8% in adenocarcinoma, and 20.6% and 28.4% in squamous cell carcinoma. The overall 5-year survival rate was 79.3%. With a median follow-up of 35.5 month for surviving patients, those with initial CYFRA21-1 serum levels higher than 2.8 ng/ml had a significantly worse prognosis (p=0.0041). Patients with an elevated preoperative CEA level exceeding 5.0 ng/ml had a shorter disease-free survival period (p=0.0003). In patients with adenocarcinoma, a CEA level above 5.0 ng/ml was associated with shorter survival and early recurrence, whereas CYFRA21-1 showed no such association. In patients with squamous cell carcinoma, elevated preoperative CEA was not related to survival and recurrence. In these patients, preoperative CYFRA21-1 level exceeding 2.8 ng/ml was associated with a poorer outcome, whereas preoperative CYFRA21-1 level was not associated with cancer recurrence.

CONCLUSION: The patients with p-stage I adenocarcinoma whose preoperative CEA level was high might be considered as good candidates for adjuvant chemotherapy. The prognostic value of CYFRA21-1 could not be confirmed for stage I NSCLC, and preoperative CYFRA21-1 level was not useful in selecting the candidates for adjuvant chemotherapy.