Background: Changes in the subsite distribution of malignant and benign colorectal tumors over the last few years have been reported in several series. They may be related to changes in diagnostic accuracy or to real changes in incidence.
Methods: Trends in incidence and subsite distribution of colorectal cancers and polyps between 1978 and 1988 have been analyzed using data from the Cancer Registry of the Canton Vaud, Switzerland (530,000 inhabitants in 1981), which has adopted standardized methods for identification and registration not only of malignant, but also of benign colorectal lesions.
Results: Age-standardized incidence rates for malignant tumors of the ascending colon in men increased from 4.6/100,000 in 1978-80 to 6.4/100,000 in 1987-88, and in women from 4.9 to 6.5. Incidence was approximately stable for transverse, descending, and sigmoid colon, whereas a decline was observed for "other and unspecified" colon cancers. Rates for rectal cancer declined by over 10% in both sexes, although the trends were not linear across calendar periods in each sex. Overall colorectal cancer incidence was relatively stable in both sexes in the Vaud population. In terms of proportional distribution, the percentage of cases in the ascending colon increased from 27% in 1978-83 to 33% 1984-88 (P < 0.05). Reliable incidence data for polyps were available for 1979 and from 1982 to 1988. Rates for ascending colon polyps increased from 1.0/100,000 in 1979-83 to 5.0/100,000 in 1987-88 in men, and from 0.4 to 2.7 in women. Incidence rates were higher, although to a lesser extent, for transverse colon and for descending colon polyps in men only. As in the case of malignant tumors, incidence rates declined between the early and the late 1980s for rectal polyps as well as for "other and unspecified" colon polyps. Overall, colorectal polyp incidence was stable around 45/100,000 for men but increased from 19 to 27/100,000 for women. With reference to proportional distributions, significant increases were observed for ascending (from 8.4% to 16.8%) and, to a lower degree, transverse colon (from 8.4% to 11.0%). No appreciable change was observed for any other subsite, except a decline for "other and unspecified" colonic polyps.
Conclusions: The observation of similar changes in distribution for benign and malignant tumors may suggest the importance of improved diagnostic accuracy (particularly total colonoscopy) for lesions arising in the proximal colon.