From November 1988 to May 1996 a prospective randomized stud!: was undertak
en to assess the efficacy of superselective intra-arterial chemotherapy for
surgically proved unresectable gastric carcinoma. Each patient had undergo
ne endoscopy as well as abdominal and pelvic CT scanning for staging. Patie
nts with evidence of liver metastasis, peritoneal carcinomatosis, enlarged
retroperitoneal lymph nodes, or locally advanced disease beyond curative re
section a ere excluded from the study. A total of 386 patients with potenti
ally curable disease were randomized to one of three treatment groups: (I)
control; (2) systemic intravenous chemotherapy; or (3) superselective intra
-arterial chemotherapy. On completion of preoperative chemotherapy, all pat
ients underwent operative exploration with curative intent. A total of 74 c
onsecutive patients were found to be unresectable, as evidenced by the pres
ence of liver metastasis, peritoneal carcinomatosis, enlarged retroperitone
al lymph nodes, or locally extensive disease not detected by preoperative C
T scanning. The median survival time in the control group and after intrave
nous chemotherapy was only 91 and 96 days, respectively, as compared to 401
days in the patients receiving intra-arterial chemotherapy. The results co
nfirmed that superselective intra-arterial chemotherapy conferred a highly
significant survival advantage compared to control or systemic intravenous
chemotherapy adjusted for all patient characteristics (P < 0.0001).