A re-evaluation of carcinoembryonic antigen (CEA) as a serum marker for breast cancer: A prospective longitudinal study

Purpose: Carcinoembryonic antigen (CEA) is still a widely used test for mon itoring breast cancer, although recent reports discourage its routine use b ecause of low sensitivity. This is a prospective study evaluating the effic acy of CEA and CA 15.3 in monitoring breast cancer. Experimental Design: Serum CEA and CA 15.3 were measured in 2191 patients w ith either benign (n = 738) or malignant (n = 1453) breast diseases. Five h undred and forty-nine patients were monitored during postsurgical follow-up for either a minimum of 5 years or until time of recurrence. Fifty-three p atients with metastases were also monitored during chemotherapy. Results: Elevated CEA and CA 15.3 levels were found in 16.7% and 33.0% of p atients, respectively. CEA sensitivity rose to 41.3% and CA 15.3 sensitivit y rose to 80.8% in metastatic patients. The adjunct of CEA increased the CA 15.3 sensitivity by 6% in the overall population and by only 2.1% for pati ents with metastases. During postsurgical follow-up, CEA was elevated in 38 .0% and CA 15.3 in 70.2% of patients with recurrence. The combination of CE A and CA 15.3 increased the overall sensitivity by only 1.4%. Longitudinal monitoring of 53 metastatic patients undergoing chemotherapy demonstrated t hat, when positive, both CEA and CA 15.3 paralleled response to treatment, although CA 15.3 was a significantly more powerful marker for determining r esponse to treatment. The cost effectiveness ratio of CEA was clearly less favorable than that of CA 15.3. Conclusions: CEA monitoring should be considered an expensive and inefficie nt method of follow-up evaluation for breast cancer patients, and it provid es no additional value when used in combination with CA 15.3.