A southwest oncology group study on the use of a human tumor cloning assay for predicting response in patients with ovarian cancer

Abstract

A total of 211 patients with epithelial ovarian cancer (168 with tumors refractory to prior chemotherapy and 43 with no prior chemotherapy) from 33 different Southwest Oncology Group institutions had their tumors sampled and specimens shipped to two central laboratories for drug-sensitivity testing in a human tumor cloning assay. The 168 patients with a prior history of chemotherapy failure (median of four prior chemotherapeutic agents) were treated with the most effective agent(s) found in the cloning assay (23 patients), and those patients whose tumors did not form colonies in vitro or did not manifest any sensitivity to agent(s) were treated with a clinician's choice of agent(s) (101 patients). The remaining 44 of the 168 patients were not treated with chemotherapy because of deteriorating performance status or early death. The complete and partial response rate in patients treated according to assay results was 28% versus 11% for the patients treated according to clinician's choice (P = 0.03). There was no statistically significant difference in survival between the two options (6.25 versus 7 months, respectively). The 43 patients with no history of prior chemotherapy were all treated with standard combination chemotherapy, and their clinical response was compared with their in vitro sensitivity to the same agents. Overall there was a 100% true-positive rate and 100% true-negative rate for the seven evaluable patients. From these data the authors conclude that use of the human tumor cloning assay may increase the response rate but not the survival for selected patients with advanced chemotherapy-refractory ovarian cancer. The study is weakened, however, by the many steps of patient selection necessitated by inadequate tumor colony formation in vitro and the inability to treat all patients (because of early death or a rapid decline in performance status). The assay does appear to be worthy of additional study for predicting response to combination chemotherapy in patients without a prior history of chemotherapy. Finally the use of central chemosensitivity testing laboratories is feasible for testing in vitro predictive assays in a cooperative group setting.