We investigated the relationship between ErbB-2 and HLA in order to clarify the clinical and genetic factors related to Japanese patients with lung cancer. Thirty-nine of the 73 lung cancer patients (53.4%) had elevated levels of ErbB-2. Only seven of 23 (30. 4%) patients with small cell carcinoma had elevated ErbB-2 levels. The prevalence of ErbB-2 positivity was highest (23 of 32; 71.8%) in patients with adenocarcinoma, while that in patients with squamous cell carcinoma was 50% (9 of 18). The frequencies of HLA A33, B44, B62, and B75 were lower in the lung cancer patients than in the control group. HLA-DR9 was higher in frequency in lung cancer patients than in the healthy controls (P<0.05), but HLA-DR6 was lower in frequency in lung cancer patients than in controls (P<0.01). DRB1*0901 was significantly higher in frequency in lung cancer patients than in controls (P<0.05). On the other hand, DRB1*0802, DRB1*1302 and the DRB1*14 group (*1401, *1403, *1405, *1406, and *1407) were completely absent in lung cancer patients. The frequencies of HLA B35, B52, B62, DRB1*0404, and DRB1*0406 were higher in the ErbB-2-positive lung cancer patients than in the ErbB-2-negative lung cancer patients. However, these types of HLA were not included in significant frequencies in our group of lung cancers. Our results suggest that some HLA-antigens/alleles participate in the pathogenesis of lung cancer in Japanese patients. In addition, the relationship between HLA-associated genetic factors and ErbB-2 seems to be weak. These findings suggest that ErbB-2 is correlated with prognostic factors for lung cancer independently of HLA-associated genetic factors.