REVIEW

Short-term in vitro drug sensitivity tests for cancer chemotherapy. A summary of correlations of test result with both patient response and survival

Andrew G Bosanquet, Bath Cancer Research Unit, Wolfson Centre, Royal United Hospital, Bath, UK

Forum Trends in Experimental and Clinical Medicine 1994; Vol 4 (2): pp 179-195.

In vitro drug sensitivity tests have been investigated for over 40 years as a method for identifying the sensitivity of individual patients to anticancer chemotherapy. Two ways to assess the accuracy of these tests are comparison of test result with clinical response (clinical correlations) and comparison with patient survival. This review summarises 4,250 clinical correlations including 510 from the differential staining cytotoxicity assay (overall predictive accuracy (PA) = 83%), 489 from metabolic function assays (PA = 81%) and 779 from the 3H-thymidine assays (PA = 74%). Where assessed, patient survival was significantly longer in patients treated with assay-sensitive compared with assay-resistant drugs. Initial publications of results often show a higher PA compared with those reporting larger numbers of clinical correlations. This has been termed the initial enthusiasm factor, and is similar to the higher publication rate observed for "positive" or "interesting" clinical trial results. Controlled clinical trials comparing standard therapy with chemotherapy chosen by in vitro assay result, including both an assessment of quality of life and an economic evaluation, would encourage the use of these assays. This mass of data is sufficient for initiating the widespread use of these assays; other laboratory tests have been incorporated into routine clinical practice with less confirmatory evidence than that presented here.