Background: Most patients with primary epithelial ovarian cancer (PEOC) treated with carboplatin/paclitaxel will relapse between one to two years. Our purpose was to define the optimal calculation method for the adenosine triphosphate-tumor chemosensitivity assay (ATP-TCA) applied to PEOC treated with platinum-containing chemotherapy and to analyze its predictive relevance.
Materials and methods: ATP-TCA results from 80 PEOC specimens were analyzed applying three different methods: 50% inhibition concentration, sensitivity index (IndexSUM), and area under curve by testing multiple cut-off levels. Correlation between in vitro results and clinical outcome was performed for 61 (76%) patients by univariative and multivariative analysis. Tumor recurrence 6 months after chemotherapy was classified as platinum-resistance.
Results: The IndexSUM set at > 250 had the highest test sensitivity, specificity, positive and negative predictive value of 90%, 43%, 62% and 81%, respectively. Patients whose tumors were shown to be resistant by ATP-TCA had a higher risk for recurrence (RR) compared to those who tested as sensitive (p < 0.003, RR = 3.3, 95% CI = 1.2-9.4). This result was confirmed after adjustment for FIGO stage by logistic regression (p < 0.004, Odds ratio = 8.3, 95% CI = 1.9-35.5). In multivariate analysis ATP-TCA and the FIGO-stage were independent predictive factors of early recurrence.
Conclusion: ATP-TCA results in combination with the use of IndexSUM > 250 are best able to predict platinum resistance.