The incidence of breast cancer: analysis of the age dependence

The incidence of female breast cancer was studied as a function of age in 71 different populations throughout the world. Incidence data was normalized for differences in frequency of occurrence. Median normalized incidence increased exponentially from age 20 to 45. After age 45 the rate of incidence decreased, and we suggest that menopause may be responsible for the rate reduction. Populations were not randomly distributed about the total population median normalized incidence at the various age intervals. Populations were classified according to their age-incidence patterns in 3 groups: Group 1 (with median % incidence after age 69=50): Group 2 (with median % incidence after age 69=47), and Group 3 (with median % incidence after age 69=31). At most age intervals, the normalized incidence in Group 1 populations was less than the total median, while the normalized incidence in Group 3 populations was greater than the total median. Although we cannot exclude the possibility that the variation in age-incidence patterns among the population groups is an artifact of tumor registration errors, evidence suggests that such an artifact is unlikely. Breast cancer occurs more frequently in Group I populations than it does in Group 3 populations. We suggest that Group 3 populations may share an element of protection from, or that Group I populations may share an element of risk for, breast cancer.

The oral glucose tolerance test: a comparison of the time points on the basis of limit values, normal dispersion, and reproducibility

Time points in the glucose tolerance test (GTT) are compared on the basis of limit values, dispersion within a reference population, and reproducibility. We suggest using the distance between a limit value and the median reference value as a measure of the magnitude of abnormality. The distance between 140 mg/100 ml and the median fasting plasma glucose value is chosen as a standard distance and limits for other points in the GTT are calculated to equal this standard distance of abnormality. We suggest that the probability of correctly interpreting an individual result is directly related to the reproducibility of the test and inversely related to the percentage of the total range of values which is dispersed among the normal population. The ratio of reproducibility to percentage normal dispersion is proposed as an index of the probability of correctly interpreting an individual result. According to this index, and probability of correct interpretation varies in order: fasting plasma glucose concentration greater than 3-h greater than 2-h greater than 0.5-h greater than 1-h plasma glucose concentration.

The incidence of male genital tumors: a cellular model for the age dependence

Most human tumors, including most male genital tumors, exhibit an exponential increase in incidence with advancing age of the host. This exponential age-incidence pattern can be explained by the accumulation of mutations in the stem cells of the tissues of tumor origin. The age-incidence pattern for testicular tumors, however, is unique with a large linear increase in incidence from age 14 to 30 and a linear decline in incidence from age 30 to 60. After age 60, the incidence of testicular tumors remains low and constant. The probability of testicular tumorigenesis is determined by the susceptibility of male germ cells to neoplastic mutation and/or the neoplastic mutagenicity of the male germ cell environment. Since there is no evidence for an environmental mutagen which is specific for male germ cells, and since male germ cells are unusually susceptible to mutation, we interpret the variation in testicular tumor incidence with age as a reflection of the susceptibility of male germ cells to neoplastic mutation. Cell are most susceptible to mutation during genome replication and we propose a model for testicular tumorigenesis which is consistent with the available data on male germ cell proliferation and with the data on testicular tumor incidence.

Glycohemoglobin and glucose tolerance tests compared as indicators of borderline diabetes

We concurrently measured glycohemoglobin and performed 3-h oral (100 g) glucose tolerance tests on 69 ambulatory patients suspected of having abnormal carbohydrate metabolism. The patients were divided into two groups: (a) The 37 patients for whom the results were normal had plasma glucose concentrations of 0.70--1.15 milligram during fasting and 0.70--1.23 g/L 2 h after glucose ingestion. (b) Borderline diabetics exceeded one or both of these limits. The range of glycohemoglobin in the normal group was 3.0--4.7% of total hemoglobin. Of the 21 borderline diabetics, 11 had increased glycohemoglobin (4.8--8.0%). The difference in tolerance test results between borderline diabetics with and without increased glycohemoglobin was insufficient to predict the status of glycohemoglobin. We suggest a tentative definition for latent diabetes: increased glycohemoglobin in the presence of normal or borderline-abnormal glucose concentration in plasma collected during fasting.

Glycohemoglobin and glucose tolerance tests compared as indicators of borderline diabetes

We concurrently measured glycohemoglobin and performed 3-h oral (100 g) glucose tolerance tests on 69 ambulatory patients suspected of having abnormal carbohydrate metabolism. The patients were divided into two groups: (a) The 37 patients for whom the results were normal had plasma glucose concentrations of 0.70--1.15 milligram during fasting and 0.70--1.23 g/L 2 h after glucose ingestion. (b) Borderline diabetics exceeded one or both of these limits. The range of glycohemoglobin in the normal group was 3.0--4.7% of total hemoglobin. Of the 21 borderline diabetics, 11 had increased glycohemoglobin (4.8--8.0%). The difference in tolerance test results between borderline diabetics with and without increased glycohemoglobin was insufficient to predict the status of glycohemoglobin. We suggest a tentative definition for latent diabetes: increased glycohemoglobin in the presence of normal or borderline-abnormal glucose concentration in plasma collected during fasting.