Abstract #0334

Assessing Chemotherapy Response in Metastatic Ovarian Cancer: The Value of Histogram Analysis of Apparent Diffusion Coefficients

Stavroula Kyriazi¹, David J. Collins¹, Robert L. Davidson¹, Veronica A. Morgan¹, Sharon L. Giles¹, Catherine J. Simpkin¹, Stan B. Kaye², Nandita M. deSouza¹

¹Cancer Imaging Centre, Institute of Cancer Research & The Royal Marsden NHS Foundation Trust, Sutton, Surrey, United Kingdom; ²Gynaecological Oncology, Institute of Cancer Research & The Royal Marsden NHS Foundation Trust, Sutton, Surrey, United Kingdom

This study investigates ADC histogram-derived parameters as predictive biomarkers of chemotherapy response in forty-five patients with advanced ovarian cancer. Mean ADC, 10^th, 25^th, 50^th, 75^th and 90^th centile points (C10, C25, C50, C75 and C90), and histogram skew were calculated in 126 primary ovarian and metastatic peritoneal, omental and visceral lesions, before and after the 1^st, 3^rd and 6^th cycle of treatment. A linear mixed model was used to analyze differences between responders and non-responders across timepoints . Morphological (RECIST) and biochemical (serum CA125) response were associated with an early and sustained increase of ADC values and decrease of skew (right histogram shift and tendency to normalization). The parameter with the highest discriminant accuracy was percentage C25 change (AUC=.791 and .818 post-1^st and -3^rd cycle respectively).