Breast cancer at menopausal ages: An explanation of observed incidence changes

N examining data on breast cancer from I the Danish Cancer Registry for 1942 to 1944, Clemmesen2 noted a remarkable and stastically significant fall in the incidence curve between the age groups 45 to 49 and 50 to 54. From age 55 onward, the upward trend in incidence with age continued. A similar phenomenon has since been observed in later material from the same source reported by Clemmesen arid Nielsen,3 and in data from England and the United States.lV Clemmesen2 suggested that the "break" in the incidence curve may be related to menopausal changes, and most later authors who have observed the phenomenon have implicated changes with age in hormonal or other physiological factors affecting susceptibility to breast cancer.l! 6 , However, the hypothesis does not seem a strong one. It seems unlikely that a menopausal change would be responsible for so temporary a reduction in susceptibility. Furthermore, the phenomenon has a somewhat evanescent quality that is difficult to reconcile under the hypothesis with the constancy of menopause. For example, in a composite graph presented by Dorn and Cutler,s the break in the incidence curve was evident in data derived from the "ten metropolitan areas" survey of 1937 to 1939; Denmark, 1943 Denmark, to 1947;; New York (exclusive of the City of New York), 1945 to 1947; and Connecticut, 1947 to 1951, but was not seen in data from the "ten metroc politan areas" survey of 1947 to 1948, or in that from Iowa for 1950. Earlier data from New York (exclusive of the City of New York), published by Clemmesenz did not show the break. Several studies published prior to 1940, in which age distributions of large series of cases were presented,sn 10.179 18 failed to show