In vitro assays of chemotherapeutic sensitivity.

: The concept of designing an in vitro assay to predict in vivo antineoplastic drug activity that would provide the medical oncologist with the necessary data to define beneficial drug regimens is appropriate; however, the optimal assay has been elusive over the last 3 decades. It is hoped that information gained from attempts to design such an assay has provided further refinements that will bring the goal in reach. The initial studies of drug-induced cell cytotoxicity employing changes in cell metabolism or the cell's ability to exclude supravital dye or reduction in the incorporation of radiolabeled precursors into DNA, RNA, or proteins provided the starting point for developing such an assay. Although initial enthusiasm existed with each of these assays, it soon became apparent that their predictive value was not sufficiently specific to warrant broad application. Modification of the dye exclusion assay or the combination of the clonogenic assay with radio precursor incorporation may provide better predictability. Confirmation of these assays awaits completion of randomized clinical trials. More recently, led by the appreciation of a subset of self-renewing cells--that is, "stem cells" present in the bone marrow--Salmon and colleagues reported on the pertinent applications of the clonogenic assay in predicting in vivo patient responses to chemotherapy. Since this report, considerable advances in development, improvement, and application of the clonogenic assay have occurred. This assay has been applied to preclinical screening of new antineoplastic agents, cytogenetic analysis of human tumor specimens, and the identification of growth factors and hormones for different tumor types. Despite these major advances in applying and solving technical problems associated with the assay, major problems continue to exist, the foremost being the overall poor growth of most tumor specimens in the assay such that in vitro chemosensitivity data can be obtained only in 30% to 40% of specimens. Indeed, because only this fraction grows, it is important to demonstrate that "growth itself" in the assay is not an independent prognostic factor. Further, pharmacologic considerations will have to be completed for each drug such that in vitro studies of drug exposure and drug/drug interaction will mimic the clinical situation. Constant critical analysis of this and other assays will no doubt lead to improvements, particularly their use as tools for biologic studies. Currently, insufficient data on prospective trials evaluating in vitro assays in predicting clinical responses are available.(ABSTRACT TRUNCATED AT 400 WORDS)