Second cancer following cancer of the respiratory system in Denmark, 1943-80

Abstract

A 10% increased risk of developing a second cancer was observed among approximately 36,000 persons reported to the Danish Cancer Registry with a cancer of the respiratory system during 1943-80. This estimate is markedly influenced by a striking tendency by physicians not to report or the Cancer Registry not to accept a report of a second lung cancer following a primary lung cancer (14 observed vs. 99 expected). A significant 30% excess of all second cancer was seen after laryngeal cancer (368 vs. 282), whereas the 22% excess following cancer of the nasal cavities and paranasal sinuses did not quite reach the level of statistical significance (95% CI = 0.9-1.6). For cancers of the lung and larynx, second cancers arose mainly in the buccal cavity, bladder, kidney (after lung cancer only) and lung (after laryngeal cancer only). These second cancers may be due to common carcinogenic factors, most likely tobacco. Elevated risks of second cancers of the breast, cervix uteri, and other female genital organs were found consistently. Radiotherapy may have contributed to the increased risk of breast cancer, but the excess risk of cancer of the female genital organs other than the cervix was unexpected. Although not significant, the risk of esophageal cancer following cancer of the larynx was below expectation (1 vs. 4.1), which was surprising because alcohol consumption and smoking are thought to be common risk factors for these 2 sites. Significant excesses of pancreatic cancer were observed following cancers of the lung, larynx, and nasal cavities, which might be due to more careful medical surveillance of these patients or to common risk factors such as cigarette smoking. Finally, the risk of a patient developing liver cancer after lung cancer was significantly elevated (22 vs. 11.6). This increase is unlikely to be due to misdiagnosed metastases from the lung, inasmuch as the risk was generally elevated throughout the observation period.