A biologic risk model for stage I lung cancer: Immunohistochemical analysis of 408 patients with the use of ten molecular markers☆☆☆★

Abstract 

Objective: The standard treatment of patients with stage I non–small cell lung cancer is resection of the primary tumor; however, the recurrence rate is 28% to 45%. This study evaluates a panel of molecular markers in a large population of patients with stage I non–small cell lung cancer to determine the prognostic value of each marker and to create a biologic risk model. Methods: Pathologic specimens were collected from 408 consecutive patients after complete resection for stage I non–small cell lung cancer at a single institution, with follow-up of at least 5 years. A panel of 10 molecular markers was chosen for immunohistochemical analysis of the primary tumor on the basis of differing oncogenic mechanisms. Local tumor expansion requires growth regulating proteins (epidermal growth factor receptor, the protooncogene erb-b2); apoptosis proteins (p53, bcl-2); and cell cycle regulating proteins (retinoblastoma recessive oncogene, KI-67). Local tumor invasion requires angiogenesis (factor viii). The development of distant metastases involves the expression of adhesion proteins (CD-44, sialyl-Tn, blood group A). Cox proportional hazards regression analysis was used to construct an independent risk model for cancer recurrence and death. Results: Multivariable analysis demonstrated significantly elevated risk for the following molecular markers: p53 (hazard ratio, 1.68; P = .004); factor viii (hazard ratio, 1.47 P = .033); erb-b2 (hazard ratio, 1.43; P = .044); CD-44 (hazard ratio, 1.40; P = .050); and retinoblastoma recessive oncogene (hazard ratio, 0.747; P = .084). Conclusions: Five molecular markers were associated with the risk of recurrence and death, representing independent metastatic pathways: apoptosis (p53), angiogenesis (factor viii), growth regulation (erb-b2), adhesion (CD-44), and cell cycle regulation (retinoblastoma recessive oncogene). This study demonstrates the validity of this molecular biologic risk model in patients with stage I non– small cell lung cancer. (J Thorac Cardiovasc Surg 1999;117:736-43)