Rates of cell proliferation in adenomatous, suppressed, and normal parathyroid tissue: implications for pathogenesis.

In previous studies, the birth rate of new cells in parathyroid adenomas measured at the time of surgical excision was shown to be much too low to account for growth of the tumors from a single cell in the time available, but comparison with normal rates was not possible. We measured the prevalence of cells expressing the Ki-67 antigen, a cell cycle marker, in 55 parathyroid adenomas using the MIB-1 antibody and microwave antigen retrieval; in 22 cases, separate measurements were made in nonadenomatous tissue from the same glands. In 10 cases complete maps of the gland profile were reconstructed to study the distribution of labeled cells. The proportion of Ki-67-positive cells, estimated by systematic random sampling, was used to calculate cell birth rate assuming a duration of Ki-67 expression of 24 h; the results were compared to rates previously determined in normal parathyroid glands by the same method. The geometric mean cell birth rate was 9.97%/yr, about double the normal rate of 5.4%/yr, but less than a third of the cases had values above the normal range. The corresponding value in nonadenomatous tissue was 2.58%/yr, about half the normal rate. In 10 cases studied in more detail, the cell birth rate was 12.3%/yr in the peripheral regions and 6.2%/yr in the central regions, a value not significantly different from normal. The results in adenomas are in reasonable agreement with previous estimates of cell birth rate of 13.7%/yr using [3H]thymidine labeling and 6.4%/yr using prevalence of the mitotic karyotype. The proportion of Ki-67-positive cells using unbiased sampling was about 50 times smaller than that in previous studies using selective sampling. Cell birth rates at the time of excision were about 20-25 times lower than initial rates estimated from modeling tumor growth by the Gompertz function. We conclude that 1) cell birth rate in parathyroid adenomas has fallen substantially during the growth of the tumors and is only modestly greater than normal; 2) the fall in cell birth rate had been greater in the central and presumably older regions of the adenoma than in the peripheral and presumably younger regions; 3) nonadenomatous tissue was suppressed with respect to its proliferative as well as its secretory function, presumably as a result of hypercalcemia; and 4) the progressive fall in cell birth rate, despite the accumulation of mutations that are supposed to increase cell birth rate, is most readily explained by the set-point hypothesis.