Stage I nonsmall cell lung cancer. A multivariate analysis of treatment methods and patterns of recurrence.

BACKGROUND: Nonsmall cell lung cancer (NSCLC) has become the leading cause of cancer-related deaths in women and men in the United States, with more than 157,000 estimated deaths in 1995. Surgical resection remains the mainstay of therapy in Stage I and II disease. However, local and distant recurrence account for the disappointing survival rates after resection. Appropriate selection of surgical procedures and effective use of adjuvant therapies will depend upon the elucidation of prognostic factors that predict for recurrence. METHODS: A detailed analysis was undertaken to evaluate surgical therapy and to define risk factors associated with recurrence and cancer death in 289 consecutive patients with NSCLC who were diagnosed, resected and followed at the Duke University Medical Center from January 1, 1980, until December 31, 1988. These patients had no evidence of metastases on head and chest/abdominal computed tomograms and radionuclide bone scans before resection. Resected specimens from these patients pathologic verification of Stage I disease. Follow-up was complete in all cases through 8/1/94 (median, 61 months). Variables analyzed included age, sex, smoking history, presenting signs and symptoms, operative procedure, histopathology, hospital course including complications, and the time and location of any recurrence or cancer death. RESULTS: The 30-day mortality rate was 5 of 289 (1.7%), with minor and major morbidity rates of 17% and 9%, respectively. Statistical comparison of lobectomy (193) wedge resection (75) and pneumonectomy (21) revealed significantly (P < 0.04) smaller tumors (T1), more comorbidity, and fewer complications for wedge resection patients. A trend (P < 0.09) toward an increased rate of local/regional recurrence and no difference in survival was also observed for wedge resection. One hundred five patients died of cancer (13-month median time to recurrence) for an actual 5-year survival of 63%. Significant univariate predictors of early recurrence and decreased survival (P < 0.01) were: male sex, the presence of symptoms, hemoptysis, chest pain, type of cough, tumor size in cm and by T-classification, visceral pleural invasion, high mitotic index, and vascular invasion. Significant (P < 0.05) multivariate independent variables for early recurrence and cancer death were the presence of symptoms, vascular invasion, pleural invasion, high mitotic index, and tumor size greater than 3 cm. CONCLUSION: Current surgical therapy for stage I NSCLC has an acceptable morbidity and mortality rate. The current data also stratify patients with Stage I NSCLC into high and low risk populations that can be used in future randomized trials of adjuvant therapy.

Duke Scholars

Author David Harold Harpole Jr. Surgery, Cardiovascular and Thoracic Surgery

Author James Emmett Herndon II Biostatistics & Bioinformatics, Division of Biostatistics