Proliferation indices as molecular pharmacodynamic endpoints in evaluation of anticancer drug effect in human solid tumors.

PURPOSE: The present study compared proliferative indices, i.e. incorporation of DNA precursor (i.e. thymidine or TdR, and bromodeoxyuridine or BrdU) and expression of proliferating cell nuclear antigen (PCNA), as molecular pharmacodynamic endpoints in evaluation of anticancer drug effect in human solid tumors.
METHODS: Tumor specimens obtained from patients were grown as histocultures. After treatment with doxorubicin, mitomycin C, and/or paclitaxel, cells labeled by [3H]TdR were identified using autoradiography, and cells labeled by BrdU and PCNA were identified using immunohistochemical techniques. Drug effect was measured as reduction of DNA precursor-labeled cells or PCNA-expressing cells.
RESULTS: The results indicate that (a) the two DNA precursors, TdR and BrdU, labeled the same cells and resulted in identical pharmacodynamics, (b) the pharmacodynamics established using inhibition of DNA precursor incorporation were qualitatively and quantitatively different from the pharmacodynamics established using inhibition of PCNA expression, (c) the inhibition of PCNA expression was erratic in some tumors, and (d) the differences in pharmacodynamics established using the two end points are drug-specific, with greater differences for paclitaxel than for mitomycin C.
CONCLUSIONS: The erratic results measured by the PCNA labeling method suggest that this method may be less reliable than the conventional DNA precursor labeling method. The finding of identical pharmacodynamics of doxorubicin and paclitaxel established using BrdU and [3H]TdR indicates that the two precursors are interchangeable. Because the methodology for detecting BrdU incorporation requires less time and does not require the use of radioactivity, we conclude that inhibition of BrdU incorporation represents a useful endpoint for evaluating the antiproliferative activity of anticancer drugs in human solid tumors.