Negative serum carcinoembryonic antigen has insufficient accuracy for excluding recurrence from patients with Dukes C colorectal cancer: analysis with likelihood ratio and posttest probability in a follow-up study

Abstract

Purpose: This study was designed to determine the efficacy of carcinoembryonic antigen (CEA) monitoring for screening patients with colorectal cancer by using posttest probability of recurrence.

Methods: For this study, 348 (preoperative serum CEA level elevated: CEA+, n = 119; or normal: CEA-, n = 229) patients who had undergone potentially curative surgery for colorectal cancer were enrolled. After five-year follow-up with measurements of serum CEA levels and imaging workup, posttest probabilities of recurrence were calculated.

Results: Recurrence was observed in 39 percent of CEA+ patients and 30 percent in CEA- patients, and CEA levels were elevated in 33.3 percent of CEA+ patients and 17.5 percent of CEA- patients. With obtained sensitivity (68.4 percent, CEA+; 41 percent, CEA-), specificity (83 percent, CEA+; 91 percent, CEA-) and likelihood ratio (test positive: 4.0, CEA+; 4.4, CEA-; and test negative: 0.38, CEA+; 0.66, CEA-), posttest probability given the presence of CEA elevation in the CEA+ and CEA- was 72.2 and 65.5 percent, respectively, and that given the absence of CEA elevation was 20 and 22.2 percent, respectively.

Conclusions: Whereas postoperative CEA elevation indicates recurrence with high probability, a normal postoperative CEA is not useful for excluding the probability of recurrence.