Is any single in situ assay of tumour response adequate?

The different assays available for measuring the response of undisturbed tumours in situ after therapy are reviewed. These are: animal survival time, regression rate of tumours, regrowth delay, local tumour control and loss of incorporated radioactivity. The relative advantages and disadvantages of each assay are reviewed in terms of cost-effectiveness and the relevance of the data they yield. For comparisons of different treatment modalities any single assay seems adequate provided a dose-response relationship can be demonstrated. The assay of choice will depend upon: the dose-range to be investigated, the amount of prior information that is required and the skills and apparatus that are available. No single assay is clearly best, but survival time and regression rate studies probably yield the least valuable information. If the main question is the absolute number of cells surviving a particular treatment, or the mechanisms leading to a given response, no single assay will yield as much information as a combination of several in situ techniques, together with excision assays. For clinically oriented questions, however, a single assay may be adequate. The choice of an appropriate tumour model is the most important factor in determining the relevance of the data obtained from mice for man.