Apparent diffusion coefficient values and response to brachytherapy in bulky cervical cancer: A retrospective study (169)

Abstract

<b>Objectives:</b> Local control can be difficult to achieve for patients with bulky cervical cancers. Cellularity inferred from apparent diffusion coefficient (ADC) values measured by diffusion MRI (dMRI) has been shown to predict tumor outcomes in other cancers. This study aimed to evaluate whether distributions of ADC from post-external beam radiation therapy (EBRT)/ prebrachytherapy (BT) dMRI can be used to predict a complete metabolic response (CR) to definitive chemoradiation in patients with bulky cervical cancer. <b>Methods:</b> A retrospective, single-institution study was conducted among patients with cervical cancer requiring interstitial BT for bulky disease (>30cc). Gross residual cervical cancer volume on post-EBRT/pre-BT single b-value clinical dMRI was contoured in 3D Slicer and reviewed by a radiation oncologist for accuracy; 90th percentile ADC values were generated from these contours. Patients were stratified based on response to BT (defined as CR on PET-CT at three months post-BT). Two-tailed t-tests were used to compare ADC values between the CR and non-CR groups. <b>Results:</b> Among a total of 47 patients, 29 (62%) were CR. The 90th percentile ADC value was significantly (p = 0.014) higher in CR (2.46 ± 0.44 µm2/ms, <i>n</i> = 29) than non-CR (2.14 ± 0.36 µm2/ms, <i>n</i> = 18). <b>Conclusions:</b> Clinical dMRI may serve to risk-stratify patients undergoing ISBT for bulky residual cervical cancer. ADC values in tumors are associated with tissue cellularity, and a higher 90th percentile ADC value may represent a higher degree of tumor necrosis post-EBRT. If our results can be replicated in larger prospective studies, percentile ADC values from post-EBRT dMRI may be used to predict a complete response to ISBT. As standard practice includes post-EBRT/ pre-BT imaging with tumor contouring, this could allow radiation oncologists to identify patients in whom escalation or additional therapy may be desired, based on the individual likelihood of complete response.