Objectives: Copious literature shows that in lung cancer many serum markers, especially the cytokeratin degradation products, correlate with the extent of disease. In 1995, we suggested the possibility of predicting the resectability of non–small cell lung cancer by measuring the plasma level of the tissue polypeptide antigen, a marker of the cytokeratin family. This study was designed (1) to confirm the earlier data in a new prospective evaluation, (2) to comparatively assess another classic biomarker (ie, the carcinoembryonic antigen), and (3) to incorporate their results into the preoperative evaluation of non–small cell lung cancer. Methods: We analyzed the database of a single institution over a 5-year period (1994-1998) in a community-based hospital and second referral level institution for a province of 500,000 people. The database included 124 consecutive patients (105 men) with pathologically documented lung cancer (50% with adenocarcinoma) accurately staged, clinically judged operable or potentially operable, and eventually operated on. Anthropometric, clinical, and laboratory data (including the carcinoembryonic antigen and tissue polypeptide antigen serum levels) and the results of a complex staging workup were prospectively recorded. Receiver-operating characteristic curves and diagnostic formulas were used for data analysis. Results: Computed tomography of the thorax, upper part of the abdomen, and brain was the most accurate preoperative method to assess tumor resectability (receiver-operating characteristic area: 0.76, 95% confidence intervals: 0.67-0.86, P =.000; accuracy rate: 77%, confidence intervals: 69%-84%). Tissue polypeptide antigen was also predictive for tumor resectability (receiver-operating characteristic area: 0.62, 95% confidence intervals: 0.51-0.73, P =.035; accuracy rate at a threshold level of 110 U/L: 65%, 95% confidence intervals: 56%-73%). Carcinoembryonic antigen was diagnostic only at the extreme values of its distribution (accuracy rate at a level up to 10 ng/mL: 69%, 95% confidence intervals: 60%-77%). The probability of finding resectable disease at the time of the operation increased from 78% (baseline computed tomography–based probability) to 83% when the concentration of tissue polypeptide antigen was lower than 90 U/L and to 85% when the concentration of carcinoembryonic antigen was below 10 ng/mL. The probability of discovering an advanced disease increased from 68% (baseline computed tomography–based probability) to 89% when tissue polypeptide antigen levels were abnormal and to 100% when carcinoembryonic antigen concentrations were higher than 10 ng/mL. Conversely, the predictability of computed tomography was diminished by contrasting biomarker results, requiring further clinical investigations. Conclusions: Computed tomography remains the gold standard for the preoperative evaluation of non–small cell lung cancer, although it may significantly underestimate the real tumor extension. The addition of the easy and inexpensive tissue polypeptide antigen test (with or without carcinoembryonic antigen) is capable of correcting this underestimation and helps to decide whether to completely rely on computed tomography or order additional clinical investigations.J Thorac Cardiovasc Surg 2001;122:891-9