Gastric cancer remains a major health care problem in the United States. While the number of newly diagnosed patients has remained essentially constant over the past 10-15 years, the incidence of proximal gastric cancers (which are more virulent) is rising rapidly. On a global basis, cancer of the stomach is one of the most prevalent malignancies known to man. While surgery remains the mainstay of potentially curative therapy, preoperative staging techniques are unsatisfactory, complete resection is possible in only a minority of patients, and survival rates even for these are poor. A substantial fraction of patients present with advanced inoperable cancer. There is a paucity of data regarding the biology of this important disease. We plan (1) To test the hypothesis that the outcome of high risk patients with local regional gastric cancer can be improved by intensive pre and postoperative adjuvant chemotherapy, given systemically and intraperitoneally (IP). High risk patients will be identified preoperatively using endoscopic ultrasonography and flow cytometry in addition to conventional staging techniques. The preoperative chemotherapy regimen used will be the FAMTX (fluorouracil, methotrexate, adriamycin, leucovorin) regimen, which has been shown in a random assignment MSKCC study to be well tolerated and capable of inducing complete remissions. The IP regimen of cisplatin and FU has already been studied at MSKCC in the postoperative adjuvant setting. (2) The biology of gastric cancer will be intensely studied in the same patient population entering the clinical trials, Using information obtained from earlier studies showing nonrandom chromosomal abnormalities as a guide, we will focus molecular genetic studies on abnormal genes on t;he appropriate chromosomes, and correlate genetic perturbations with clinical outcome. Biological data will be entered into an existing prospective computer data base containing clinical information on surgical and chemotherapy protocols involving the same patients. Sophisticated biostatistical techniques will be used to analyze the potential correlations between clinical and laboratory data.
