Second cancer following cancer of the digestive system in Connecticut, 1935-82

Abstract

The risk of developing a second primary cancer was evaluated in approximately 64,000 persons diagnosed with cancer of the digestive system in Connecticut during 1935-82. Significant excesses of all second cancers combined were observed following cancer of the esophagus (58 observed vs. 33 expected), small intestine (41 vs. 24), and colon (2,268 vs. 1,714). A slight excess of multiple primaries was observed following cancer of the liver and biliary tract (47 vs. 40). The observed number of second cancers was nearly equal to the expected number for persons initially diagnosed with cancers of the stomach (251 vs. 258), rectum (952 vs. 941), and pancreas (40 vs. 40). Persons with initial cancers of the small intestine, colon, and rectum also had excess second cancers arising primarily in the colon, which suggested the influence of common etiologic factors or possibly misclassified metastases in some. Shared dietary, socioeconomic, or hormonal factors may explain the excess of uterine and ovarian cancers among patients with colon cancer and the excess of breast cancer among patients with colon and rectal cancers. Oral and respiratory cancers occurred more frequently than expected in persons with an initial esophageal cancer, which is likely due to common risk factors of cigarette smoking or alcohol intake, or both. The elevations in cancer of the prostate among males with cancers of the esophagus, small intestine, colon, rectum, liver/biliary, and pancreas are probably artifacts associated with increased medical surveillance of cancer patients. The prostate cancer excesses were limited to the first year after diagnosis of the initial cancer or decreased over time for all but cancer of the colon and small intestines. Increased medical surveillance may also contribute to the excess renal and bladder cancers seen within 5 years of diagnosis of stomach cancer. Excesses were also seen for second pancreatic cancer among small intestine and liver/biliary cancer patients and second kidney and brain cancers among those with colon cancer. The deficits of stomach and rectal cancer among persons initially diagnosed with the same tumors, respectively, were anticipated because surgical removal of the organ is the primary form of treatment. Patients with rectal cancer also had deficits of stomach and pancreatic cancers. Future research should clarify the role of diet, alcohol, metabolic and endocrine factors, and host susceptibility on the risk of second neoplasms following cancer of the digestive system.