Background: Prognostic implications of overexpression of the HER-2 gene in nonsmall cell lung carcinoma (NSCLC) are a matter of controversy. Many conflicting results have been reported from different laboratories.
Methods: A metaanalysis of published studies was performed for this quantitative review of the effects of HER-2 overexpression on survival among patients with NSCLC. Of 44 articles initially selected, 20 articles fulfilled eligibility criteria. DerSimonian-Laird random effects analysis was used to estimate the effects of HER-2 overexpression on survival differences (the survival rate among patients without HER-2 overexpression minus the survival rate among patients with HER-2 overexpression) at endpoints of 1 years, 3 years, and 5 years after resection of NSCLC.
Results: In total, 2579 patients were included in the final analysis. Overall, HER-2 positivity differed according to histologic type and included 38% of patients with adenocarcinoma, 16% of patients with squamous cell carcinoma, and 18% of patients with large cell carcinoma (P < 0.0001). The combined survival differences in patients with NSCLC at 1 year, 3 years, and 5 years, respectively, were 2.7% (95% confidence interval [95% CI], 1.3-6.7%; P = 0.1787), 15.2% (95% CI, 5.8-24.5%; P = 0.0015), and 16.4% (95% CI, 7.9-14.8%; P = 0.0001), suggesting significant poorer survival at 3 years and 5 years among patients with HER-2 overexpression. In patients with adenocarcinoma, the combined survival difference at 5 years was 26.0% (95% CI, 16.0-36.1%; P < 0.0001), suggesting a particularly strong survival impact for HER-2 overexpression.
Conclusions: A significant, unfavorable prognostic effect of HER-2 overexpression in NSCLC was evident from the metaanalysis. However, because several studies that found no significant difference were excluded by the current eligibility criteria, caution is needed in interpreting the results.
(c) 2005 American Cancer Society.