Application of tumor markers CEA, TPA, and SCC-Ag in patients with low-risk FIGO stage IB and IIA squamous cell carcinoma of the uterine cervix

Objective. The aim of this study was to determine the potential clinical ut ility of tumor markers carcinoembryonic antigen (CEA), tissue polypeptide a ntigen (TPA), and squamous cell carcinoma antigen (SCC-Ag) in patients with FIGO stage IB and IIA squamous cell carcinoma of the uterine cervix with l ow-risk clinicopathologic factors (negative lymph node metastasis, no lymph ovascular space involvement, no bulky tumor size, no parametrial invasion, no deep stromal invasion, and well-differentiated cellular histology). Methods. A retrospective study was performed on 558 patients with FIGO stag e IB-IIA and pathology-proven invasive squamous cell carcinoma of the uteri ne cervix, treated at the Veterans General Hospital, Taipei, between Decemb er 1986 and November 1990. Serum specimens were drawn before operation. A t otal of 140 assessable patients were enrolled into the study (including 109 stage IB patients and 31 stage IIA patients; all patients had no clinicopa thologic risk factors and had at least one tumor marker datum). Survival cu rves were constructed according to the Kaplan-Meier method and survival cur ves were compared using the log-rank test. Results. In univariate analysis of survival, CEA, TPA, and SCC-Ag all have roles in the prediction of prognosis. In Cox proportional hazards model usi ng CEA, TPA, and SCC-Ag as covariates, TPA demonstrated the most significan t risk factor (P = 0.031). Conclusions. We concluded that preoperative evaluation of serum TPA might b e of great value in the prediction of survival of patients without any clin icopathologic risk factors and this special group of patients should be pai d much attention in the follow-up period. From this study, preoperative ele vation of TPA defines a group of otherwise low-risk invasive cervical cance r patients who are at high risk for recurrence. Adjuvant therapy might be n ecessary for this special subset of patients. A prospective study with a la rger sample should be conducted to prove this particular finding. (C) 2000 Academic Press.