Carcinoembryonic Antigen Monitoring to Detect Recurrence of Colorectal Cancer: How Should We Interpret the Test Results?

To the Editor: It is routine clinical practice, supported by national guidelines in both North America and Europe, to measure blood carcinoembryonic antigen (CEA)1 to detect recurrence of colorectal cancer during follow-up after primary treatment. Blood CEA is usually measured every 3–6 months, and patients with a CEA concentration above an absolute threshold (5 μg/L according to American Society of Clinical Oncology guidelines) are investigated further by radiological imaging. However, the evidence underpinning both guidelines and routine practice is weak. We recently reported the interim results of the Follow-up After Colorectal Surgery (FACS) trial, a clinical trial comparing different types of posttreatment follow-up in 1200 patients with colorectal cancer (1). This trial confirmed that measuring CEA is an effective way of detecting recurrence at an early stage, thus increasing the number of recurrences that can be treated with curative intent. However, the threshold we applied to define an abnormal CEA concentration (7 μg/L above the patient's postoperative concentration at trial entry) was more conservative than current guidelines, so we decided to reevaluate our data to assess retrospectively whether we could have done better by applying a different strategy to interpret the CEA result and trigger further investigation. We assessed 3 strategies for interpreting the CEA result based on (a) the result of each test individually; (b) the difference between the individual test result and the patient's own postoperative baseline; and (c) the trend in results over time, calculated by linear regression of the log-transformed CEA values. For each strategy, we assessed the outcome of applying different cutoff thresholds by ROC analysis. In the CEA arms of the FACS trial, blood CEA was measured every 3 months for 2 years after primary treatment was complete, followed by every 6 months for 3 years. All CEA measurements were made by use of a Siemens Centaur XP analyzer using a chemiluminescence immunoassay, at the John Radcliffe Hospital in Oxford, U.K. (which participates in the national external quality assurance scheme). The analysis we report here is based on 6259 individual CEA measurements in 559 patients (90 of whom experienced recurrence). Patients with <2 CEA measurements, repeat tests, and measurements taken after confirmation of cancer recurrence were excluded. The ROC curves for each strategy are shown in Fig. 1. This figure illustrates why CEA should not be used as the only method of monitoring, since no strategy achieved 100% diagnostic sensitivity even when applying cutoff thresholds with unacceptably low levels of diagnostic specificity. The figure also shows that calculating the difference from the patient's baseline concentration is no better than considering the result of each test individually [area under the curve (AUC) 0.73 and 0.74]. However, taking account of the trend in CEA concentrations over time was substantially better than either of the other 2 strategies (AUC 0.90, 95% CI 0.85–0.95). The CEA slope that provided maximal diagnostic sensitivity and specificity was a monthly increase of 1.02 μg/L. A review of the literature shows that this observation is not new. One of the first papers drawing attention to the value of CEA in detecting covert recurrence of colorectal cancer, published in The Lancet 40 years ago, stressed the importance of the velocity of change (2). A number of subsequent papers have repeated the message that the rate of change in CEA concentrations over time (i.e., the slope of the curve) has diagnostic and prognostic value over and above the absolute concentration at any 1 time (3). However, this has never been taken up as routine practice, perhaps because CEA is still perceived as a diagnostic test and its potential as a monitoring test (i.e., a diagnostic test with a time dimension) (4) has not been widely recognized. Estimating trend in a series of test results is made much easier by the increasing use of electronic records in clinical practice. For example, in the UK all medical records in primary care facilities (where CEA monitoring was done in the FACS trial) are computerized, and reporting results in terms of trend is no harder than reporting the individual test result. However, estimating the trend in CEA for an individual does require several measurements to be performed. An important clinical implication of these findings is the need to consider increasing the frequency of CEA testing, particularly at the beginning of follow-up (5). The results we report here are based on a retrospective analysis (i.e., we analyzed all the measurements that we had available for each individual) and therefore require prospective validation in an independent dataset before this strategy can be implemented in clinical practice. We would welcome collaboration to achieve this end. carcinoembryonic antigen Follow-up After Colorectal Surgery area under the curve.