To evaluate the effectiveness of immunotherapy using PSK, an immunomodulator, a controlled study was implemented from February 1987 to November 1989 at 16 institutions in Japan. Postoperative adjuvant therapy completely accomplished was continued for at least 3 months until to the tumor progression. The aim of the present retrospective study was to evaluate the indications for PSK therapy by using 653 patients drawn from this controlled study, who were completely accomplished therapies. Two hundred twenty-two patients underwent surgery alone, 184 received gastrectomy plus complete PSK therapy, 146 received gastrectomy plus complete chemotherapy, and 101 received gastrectomy plus complete chemotherapy plus PSK. There was a significant difference in survival between patients receiving gastrectomy alone and PSK, but not between those receiving chemotherapy and chemotherapy + PSK. There was also a significant difference in survival between patients with and without PSK. Among CEA-negative patients, there was a significant difference in survival between patients receiving gastrectomy alone and those receiving PSK, while among CEA-positive patients, there was a significant difference in survival between both gastrectomy alone and PSK, and chemotherapy and chemotherapy + PSK. Using the Cox's proportional -hazard model , we identified that PSK therapy prolonged survival times in patients with abnormal levels of IAP, ASP, SA, and CEA, while in those with normal level of ACT with statistical significance, and that the stata of CEA and SA levels might be best predictive values of response to PSK treatment in gastric cancer.

soluble glycoproteins (ASP), and sialic acid, before immunotherapy using protein-bound polysaccharide K (PSK) can predict the clinical outcome in patients with gastric cancer. Seven hundred and twelve patients were enrolled in the study from February 1987 to November 1989 at 16 institutions throughout Japan. All patients were followed over a period of 5 years. Among the 712 patients, 694 (97.5%) eligible patients with biopsy-proven gastric adenocarcinoma underwent gastrectomy and were randomly assigned to receive supportive therapy as follows: gastrectomy alone, gastrectomy plus PSK, gastrectomy plus chemotherapy (CT), and gastrectomy+CT+PSK. The 5-year follow up rate was 94.5%. There was a significant difference in survival between patients receiving gastrectomy with or without PSK who showed abnormal serum levels of CEA (log-rank test, P=0.0175), and both CEA and APR (log-rank test, P=0.0431), and between those receiving PSK and not receiving PSK in those with parameter point 3 (log-rank test, P=0.0404). These results indicate that PSK therapy has a beneficial effect in gastric cancer patients with high levels of CEA and APR. Ann Cancer Res Ther 7(1): 43~49, 1998/Received 13 Apr 1998 Accepted 1 May 1998

In order to clarify the immunogenetical background, host factors in oncology, human leukocyte antigen (HLA) and tumor necrosis factor (TNF) beta alleles as prognostic, preventive and therapeutic indicators were investigated in 712 patients with a histologic diagnosis of adenocarcinoma of the stomach treated with gastrectomy. HLA and TNF beta alleles were tested serologically and by DNA-PCR typing. The absence of HLA Cw1 antigen may represent resistant and prognostic factors. HLA-B51, B61 and TNF beta 10.5 kb homozygote alleles are therapeutic, survival and prognostic factors. Considering the relation with lymph node metastasis, HLA-DR4 antigen and HLA-DRB 1*0405 allele were found to be risk factors for lymph node metastasis in poorly differentiated adenocarcinoma. TNF beta 10.5 kb homozygote allele also represented a risk factor for lymph node metastasis. TNF beta 5.5 kb homozygote allele was considered a resistant factor for lymph node metastasis in poorly differentiated adenocarcinoma. HLA and TNF beta alleles can play an important role as prognostic, preventive and therapeutic indicators in gastric cancer. Therefore, TNMH (TNM with host factor) should be proposed as a new approach.