441 Chemotherapy response score: correlation with preoperative serological and radiological assessment of response and clinical implications in ovarian cancer patients

Abstract

<h3>Introduction/Background</h3> The 3-Tier Chemotherapy Response Score (CRS) was developed to quantify the response after neoadjuvant chemotherapy (NACT) in high-grade serous ovarian cancer patients undergoing interval debulking surgery. CRS3 (optimal response) identifies patients with a longer progression-free (PFS) and overall survival (OS) compared to patients with a CRS1/2 (no or minimal response/partial response). We critically evaluated the clinical value of CRS and compared its predictive power to standard serological (CA125) and radiological response in patients with advanced epithelial ovarian cancer. <h3>Methodology</h3> A retrospective analysis of 277 patients, who received primary chemotherapy for advanced epithelial ovarian cancer was performed. CRS, serological, and radiological findings, and pathological complete remission (pCR) were correlated to PFS and OS. <h3>Results</h3> Only 62.1% (172/277) patients treated with NACT could be assessed by CRS, as the CRS score can only be determined in patients that undergo interval debulking surgery, have a representative biopsy of the omentum, and have tumours with a high-grade serous histology. In patients with CRS3 (n=50) a longer median PFS and OS was observed compared to patients with CRS1/2 (n=122) (31.2 vs. 18.9, P&lt;0.001; 55.0 vs. 36.1 months, P=0.050). Patients with serological and radiological complete response showed longer PFS (23.0 vs. 14.4, P=0.011; 21.4 vs. 9.6 months, P&lt;0.001) and OS (49.5 vs 29.0, P=0.003; 45.0 vs. 12.9 months, P&lt;0.001). Patients with a pCR had the best median PFS (52.8 months), even compared to non-pCR CRS3 (27.8 months). In the total study cohort, serological and radiological complete response was better at predicting PFS (hazard ratio 2.23 and 2.77). Radiological complete response was better at predicting OS (hazard ratio 2.34). <h3>Conclusion</h3> In this study, evaluation of response to chemotherapy by CRS was not superior to conventional methods (CA125 or radiology). Independent of the used evaluation method, response to NACT was predictive of PFS and OS. Conventional methods should even be considered more clinically relevant, as these can be applied to all ovarian cancer patients receiving upfront chemotherapy, while only 62% of patients in our cohort could be assessed by CRS. Conventional response assessment, based on radiology and/or CA125, is used to evaluate whether a patient should be offered IDS and can, similar to CRS, be used to predict PFS and OS. As CRS has no influence on the treatment of patients undergoing NACT for ovarian cancer, the added value of response assessment using CRS is negligible. <h3>Disclosures</h3> JPR has no conflict of interest to declare. TB has been an advisor for Tesaro and received research grant from Amgen, non-financial support from Amgen, MSD, Roche, and Tesaro, outside the submitted work. MLM, AT, KUW, SH have no conflict of interest to declare. PH reports grants and personal fees from Astra Zeneca and Roche, personal fees from Sotio, grants and personal fees from Tesaro and GSK, personal fees from Stryker, Zai Lab, and MSD, grants from Public funding (DKH, DFG, EU), personal fees from Clovis, and Immunogen, grants from Boehringer Ingelheim, Medac, and Genmab, outside the submitted work. AdB reports personal fees from Roche, Astra Zeneca, Tesaro, Clovis, Pfizer, Genmab, Pharmar, and Biocad, outside the submitted work.