(P037) Impact of Serial DWI and ADC Measurements in Assessment of Brain Metastases Treated With Neurosurgical Resection and Intraoperative Cesium- 131 Brachytherapy: Results of a Prospective Trial

Menachem Z. Yondorf, BS, A. Gabriella Wernicke, MD, MSc, Bhupesh Parashar, MD, Theodore H. Schwartz, MD, John A. Boockvar, MD, Phillip Stieg, MD, Susan Pannullo, MD, Dattatreyudu Nori, MD, K.S. Clifford Chao, MD, Ilhami Kovanlikaya, MD; Weill Cornell Medical College

Background and Objectives: Intraoperative Cs-131 brachytherapy has been demonstrated to be safe, well tolerated, and convenient for patients, rendering high local control and minimal toxicity for patients with brain metastases (Wernicke et al.). Early identification of residual or recurrent metastasis is crucial to apply potentially more effective therapy. Interpretation of conventional magnetic resonance imaging (MRI) is a challenge in differentiating residual tumor and possible inflammatory reactions. In this study, we reviewed serial MRI images along with diffusion-weighted images (DWIs) for distinct local changes and assessed their clinical relevance and measured apparent diffusion coefficient (ADC) values.

Materials and Methods: As part of the institutional review board (IRB)-approved prospective study, 26 patients with brain metastases treated with Cs-131 brachytherapy were evaluated with MRI exams, which included: T2-weighted, flair, diffusion-weighted, gradient recalled echo (GRE), or susceptibility weighted imaging (SWI); and T1-weighted sequences with and without contrast, before and after the surgery (the first within 48 hours and subsequent ones every 4–8 weeks). GRE/SWI images were used to rule out hemorrhage or other sources of susceptibility artifacts that cause high signal intensities on DWI. ADC values were measured and normalized by dividing contralateral normal-appearing brain parenchyma.

Results: On follow-up MRI, delayed enhancement was seen in previously ADC low areas immediately postoperatively and was interpreted as ischemic/inflammatory changes due to surgical trauma. Depending on its size and localization, some of these changes ended up as encephalomalacia in the brain parenchyma. Contrast enhancement (CE) in lower ADC areas other than the previously low ADC should be considered suspicious for a local recurrent tumor (there have been no cases of local recurrent tumor).
There were nine cases that showed increased enhancement 4–7 weeks after implantation, with higher ADC values other than surgical changes in the parenchyma around the cavity. The enhancement pattern was mostly rim-shaped, but some nodular pattern was noticed. These cases were interpreted as being suspected for recurrent tumor, and follow-up MRI was recommended. Mean ADC ratio was 1.71 (range: 1.46–1.78). In all cases, CE revealed decreased enhancement in the follow-up exams, with higher ADC values. We propose that this enhancement pattern with high ADC values is due to the inflammatory effects of Cs-131 seeds in the cavity. All of the higher ADC values seen due to surgical trauma decreased on the follow-up exams. New restricted diffusion was found in two patients: one with an abscess in the cavity and one with an additional subdural metastasis in the surgical incision. T2 hyperintensities, which reflected vasogenic edema around the metastatic tumor, decreased in 24/26 cases, mostly in 4 weeks after surgery. In two cases, T2 hyperintensities increased in the follow-up studies due to additional metastasis close to the region of original metastasis.

Conclusion: Serial DWI and measurement of ADC values on brain metastasis treated with surgical removal and brachytherapy might be very useful in differentiating ischemic necrotic inflammatory processes from residual or local recurrent neoplasms. A new enhancement observed around the surgical cavity after brachytherapy should be interpreted in the context of the serial DWIs, including preoperative evaluation.