Objective: Two decades of research have resulted in modest gains in progression-free and overall survival for ovarian cancer patients. Peer reviewed studies randomizing these patients to various standard chemotherapies consistently demonstrate remarkably similar survival, highlighting the therapeutic equivalence and ambiguity that exists with empiric treatment. Chemoresponse assays were developed to determine an individual's in vitro tumor response to various chemotherapy agents, providing sensitivity and resistance information to guide patient treatment. Utilizing a two arm clinical trial design approach, the objective of this study was to conduct a comparative analysis of the impact that a chemoresponse assay (ChemoFx®, Precision Therapeutics, Inc.) has on overall survival (OS) as compared to patients treated by non-assay informed empiric choice from previously accrued drug trials from 4 large cooperative group studies.

Methods: Patient characteristics from 4 cooperative group studies (duBios JNCI 2003; Pfisterer JNCI 2006; duBois JCO 2006; Bookman 2009 JCO) totaling >7,000 patients were used as the empiric control arm and compared to a 192 patient cohort from a chemoresponse assay study (Herzog 2010 AJOG). These arms were compared to confirm that the assay-informed arm was representative of large trial patient populations in advanced ovarian cancer. In addition, annual rates for OS were compared among all 5 patient cohorts. An additional analysis was conducted comparing patient outcomes from the assay-informed arm with one of the cohorts from the empirically-treated arm (Bookman 2009 JCO). Patients from the assay-informed arm were categorized by in vitro response as sensitive (S), intermediate (I), or resistant (R) to the drug administered, and Kaplan Meier survival curves were created for each response category. These survival curves were compared to the Kaplan Meier curves from the selected empirically-treated cohort.

Results: Patient characteristics were very similar between the empirically-treated and the assay-informed arms, though 11-16% more patients were optimally debulked in each of the 4 empirically-treated cohorts. Additionally, with the exception of one of the cohorts (Bookman 2009 JCO), patient stage was lower for the empirically-treated arm as compared to the assay-informed arm, with approximately 10% of patients earlier staged. The assay-informed arm experienced a 10% improvement in median OS (48 vs. 44 months). At the time of completion of the assay-informed study (6 years), the difference in OS rates between the 2 arms was 10%, with 39% OS for the assay-informed vs. 29% in the empirically-treated arm. The largest cohort from the empirically treated arm (Bookman 2009 JCO) included 4,312 patients and a stage distribution similar to the assay-informed arm. It was chosen for further analysis to see survival differences when patients in the assay-informed arm were separated by response category. Median OS by response category for the assay-informed was 72.5, 48.6, and 28.2 months, for S (n=20), I (n=133), and R (n=39), respectively (P=0.03; hazard ratio, 0.70; 95% CI, 0.50-0.97), as compared to 44 months for the patient subset from the empirically-treated arm.

Conclusion: Realizing study design limitations and despite a generally worse prognosis, patients from the assay-informed arm experienced increased median OS when compared to the empirically-treated arm. This difference was further augmented when the assay informed arm was compared directly to one of the cohorts from the empirically treated population. These results support the use of chemoresponse assays as predictors of clinical response and OS in epithelial ovarian cancer as well as for choosing between equivalent treatments in clinical practice.

Citation Format: Thomas J. Herzog, James Fiorica, James W. Orr, Jr., Robert Holloway, Mike Gabrin, Edward C. Grendys, Jr.. A comparison of patient outcomes in advanced ovarian cancer: Assay-informed vs. empirically treated. [abstract]. In: Proceedings of the AACR Special Conference on Advances in Ovarian Cancer Research: From Concept to Clinic; Sep 18-21, 2013; Miami, FL. Philadelphia (PA): AACR; Clin Cancer Res 2013;19(19 Suppl):Abstract nr A85.