Since the early 1970s, several studies have shown that high carcinoembryonic antigen (CEA) levels were a potential marker of poor prognosis in nonsmall cell lung cancer (NSCLC). Abnormally elevated CEA levels were reported in 30% to 70% of patients with NSCLC and were most frequently observed in patients with adenocarcinoma and advanced stage carcinoma. Despite its potential value, CEA was often falsely elevated in smokers and in patients with restrictive or obstructive pulmonary disease. Consequently, the stage of disease and performance status of the patients remained better predictors of survival than CEA, and in 1997, The American Thoracic Society and the European Respiratory Society did not recommend the routine measurement of CEA for patients with NSCLC, because it did not influence the management of these patients. However, the past decade has been associated with significant changes in the mode of presentation and treatment of NSCLCs. Adenocarcinoma has become more frequent than squamous cell carcinoma in North America and Asia, low-dose chest computed tomography has been popularized, the incidence of bronchioloalveolar carcinoma has been rising (particularly in nonsmokers), and chemotherapy has been proven beneficial in patients with early stage NSCLC. These changes have brought a whole new set of questions in our current management of patients with stage IA and IB NSCLC. Do all patients with very early stage NSCLC require a lobectomy or would segmentectomy be an appropriate operation in some patients? Do all bronchioloalveolar carcinomas have to be treated similarly to other NSCLCs, or is there a spectrum of disease between bronchioloalveolar carcinoma and adenocarcinoma? Do all patients with early stage NSCLC require chemotherapy in addition to surgery, and should they receive it before or after surgery? Refinements to our current evaluation of patients with early stage NSCLC is needed to answer some of these questions. The measurement of tumor markers could potentially be part of this evaluation. Carcinoembryonic antigen is an inexpensive and readily available serum marker that has been shown to be one of the few independent preoperative predictors of survival in patients with small adenocarcinoma. High preoperative CEA level has also been shown to be associated with a higher risk of nodal metastasis in patients with NSCLC, and particularly in patients with clinical stage IA adenocarcinoma. The combination of a normal preoperative CEA level with a high tumor disappearance rate, which is determined by the ratio of the tumor area between the mediastinal and the lung windows on chest computed tomography, was recently shown to preoperatively predict the selection of a group of patients with peripheral adenocarcinoma that had no nodal metastasis [1Takamochi K, Nagai K, Yoshida J, et al. Pathologic N0 status in pulmonary adenocarcinoma is predictable by combining serum carcinoembryonic antigen level and computed tomographic findings? J Thorac Cardiovasc Surg 2001;122:325–30.Google Scholar]. This study from Okada and colleagues looking at the role of histology and smoking status on the prognostic value of CEA in patients with clinical stage I NSCLC reiterates the importance of CEA as an independent preoperative prognostic variable along with age, gender, and size of the tumor [2Okada M, Nishio W, Sakamoto T, et al. Effect of histologic type and smoking status on interpretation of serum carcinoembryonic antigen value in non–small cell lung carcinoma. Ann Thorac Surg 2004;78:1004–10.Google Scholar]. The authors eventually demonstrate that elevated preoperative CEA level had a predictive impact on prognosis only in nonsmokers with adenocarcinoma; this group also included previous smokers that had stopped smoking before their diagnosis of NSCLC. This finding may explain some of the discrepancies that were observed in the past and the low accuracy of elevated CEA level to predict outcome when it is indifferently applied to all patients with NSCLC. The importance of postoperative CEA level on outcome in patients with adenocarcinoma or squamous cell carcinoma and pathologically proven stage I disease is also briefly reported in the study. Unfortunately, the authors did not directly compare the outcome of patients that normalized their CEA level postoperatively to those that did not. However, the persistence of elevated CEA level despite apparently complete surgical resection is likely an important finding that is related to the persistence of undetected micrometastasis in mediastinal lymph nodes or elsewhere in the body. Hence, this group of patients is certainly at high risk of recurrence and could potentially benefit most from aggressive adjuvant chemotherapy. In conclusion, CEA is a tumor marker that may help to stratify patients with early stage NSCLC. The prognostic potential appears to be valid, mainly in nonsmokers or previous smokers, and it seems to be particularly useful in patients with adenocarcinoma. As the treatment of patients with early stage NSCLC is being refined in prospective studies, preoperative and postoperative CEA level appear to be an important factor to consider.