Click here to


Are you sure ?

Yes, do it No, cancel

Mid-Treatment Gd-EOB-DTPA-Enhanced MRI and Interleukin 6 Cytokine as Biomarkers of Radiation-Induced Liver Toxicity in Metastatic Liver Patients

A Ajdari1*, Y Xie1 , C Richter2 , T Hong1 , T Bortfeld1 , (1) Massachusetts General Hospital, Boston, MA, (2) OncoRay, Dresden, Germany


(Thursday, 7/18/2019) 7:30 AM - 9:30 AM

Room: Stars at Night Ballroom 2-3

Purpose: In the era of precision medicine and personalized radiation treatment (RT), there is an ever-growing need to find predictive biomarkers of treatment response in patients. Here we investigate the potentials of using mid-treatment MR images and inflammatory cytokines as biomarkers of liver toxicity.

Methods: Eleven intrahepatic metastatic patients who had proton stereotactic body radiotherapy (SBRT) to the liver lesion were retrospectively analyzed. Two Gd-EOB-DTPA (a hepatobiliary-directed contrast agent)-enhanced MR scans as well as three inflammatory cytokines (interleukin 6 [IL-6], IL-8, and tumor necrosis factor α [TNF- α]) were acquired during the RT course. Deformable image registrations were done among mid-treatment (fx4 and 5) MR images and the planning CT. MR signal changes and delivered dose were then calculated for each voxel. Mid-treatment changes in the expression of the cytokines were calculated with respect to the pre-treatment baseline. Liver toxicity was assessed at 3 months post-RT, using Child-Pugh (CP) and ALBI score. Patients were subsequently classified into high-risk (HR) and low-risk (LR) groups. Statistical analysis was performed to compare the changes in the MR signals and cytokine expressions between these groups.

Results: On average, high-risk patients had lower high-dose/low-dose mid-treatment signal changes (i.e., decreased/increased signal in high-dose/low-dose). In CP classification, there was a significant difference in MR signal change between two group means (0.61 and 1.04 for HR and LR groups; p-value=0.005). The ALBI classification showed more pronounced difference (0.61 vs. 1.11, p-value = 0.002). High-risk patients also showed larger IL-6 changes during their treatment (86% vs. 0.33%, p-value=0.01).

Conclusion: Using mid-treatment MR scans and interleukin 6 as biomarkers, it is possible to predict the risk of acute liver toxicity, already during the RT course. This biomarker information can be potentially used for adaptive planning and RT plan personalization.


MRI, NTCP, Radiobiology


TH- response assessment : MR imaging-based

Contact Email