SS 36 - GI 3: Liver - Checking in on Inhibitors, inSPECTing Liver Function, and Making Progress Oligo-ly
Presenter(s)
S. Apisarnthanarax1, S. R. Bowen1, C. Grassberger1, Y. He1, J. Fu1, Y. Kim1, K. Thonglert2, N. S. M. Wong3, S. Cui1, M. C. Matesan4, H. J. Vesselle4, and M. Nyflot1; 1Department of Radiation Oncology, University of Washington/Fred Hutchinson Cancer Center, Seattle, WA, 2Division of Radiation Oncology, Department of Radiology, King Chulalongkorn Memorial Hospital (KCMH), Bangkok, Thailand, 3Department of Clinical Oncology, Tuen Mun Hospital, Tuen Mun, Hong Kong, 4Department of Radiology, Division of Nuclear Medicine, University of Washington, Seattle, WA
Purpose/Objective(s): A critical challenge in (RT) for liver cancer is accurately assessing both global and regional liver function to improve risk stratification, personalize RT planning, and characterize normal liver injury. We conducted a prospective study to evaluate baseline and longitudinal liver function in RT-treated liver cancer patients using 99mTc-sulfur colloid (SC) SPECT/CT imaging.
Materials/Methods: Patients with primary or metastatic liver cancer receiving 5-fraction photon SBRT or 15-fraction proton therapy were enrolled. SC SPECT scans were performed at 3 timepoints: pre-treatment (pre-Tx), mid-treatment (at fraction 3 of 5 or 8 of 15, mid-Tx), and 1 month post-treatment (post-Tx). Global liver function and spatial heterogeneity were quantified at each timepoint using total liver function (TLF = functional liver volume ratio at 60% SC uptake threshold × mean liver to spleen ratio) and coefficient of variation (CV), respectively. Spearman correlation, Wilcoxon signed rank, and Mann Whitney U tests evaluated TLF and Child-Pugh (CP) associations, longitudinal changes and group differences. TLF and CP associations with liver failure-specific survival (LFSS) were assessed using Cox regression (cause specific hazard approach) and log rank tests.
Results: Among 38 enrolled patients, 36 with =2 SC SPECT scans were analyzed. Most had hepatocellular carcinoma (89%) and cirrhosis (92%) with CP class A (67%) or B/C (33%). Median pre-Tx TLF inversely correlated with CP scores (A5 = 0.64, A6 = 0.30, B7-C10 = 0.18;
r = -0.61, p = 0.002) and were lower in CP-B/C vs. CP-A (p=0.02) and CP-A6 vs. CP-A5 (p=0.003). Over time, TLF progressively declined while CV increased compared to baseline: at mid-Tx, TLF decreased by 29% and CV increased by 5%; at post-Tx, TLF decreased by 56% and CV increased by 42% (Table). The most significant changes occurred in CP-A patients post-Tx. Absolute mid- and post-Tx TLF values, but not changes in TLF, correlated with CP score changes at post-Tx (p=0.003/0.007) and LFSS (TLF binarized at median, HR=undefined, p=0.02). Patients with mid-Tx TLF > 0.29 had 2-year LFSS of 100% compared to 64% for those with TLF < 0.29. Baseline CP scores and mid- and post-Tx CP changes did not correlate with LFSS.
Conclusion: SC SPECT imaging detects early global and regional changes in liver function during and after RT that are not appreciable with CP scoring. Absolute liver function metrics during and after RT, analogous to the surgical liver remnant concept, correlate with liver failure-related survival. Further investigation of these imaging biomarkers is underway to guide personalized adaptive RT in liver cancer.
Table 1: Longitudinal TLF and CV. All values median ± STD. Significant change from preTx: *p<0.05, **p<0.01, ***p<0.001 | Pre-Tx | | Mid-Tx | | Post-Tx | |
| TLF | CV | TLF | CV | TLF | CV |
All | 0.41 ± 0.34 | 0.38 ± 0.10 | 0.29 ± 0.30* | 0.40 ± 0.12 | 0.18 ± 0.31*** | 0.54 ± 0.13* |
CP-A | 0.44 ± 0.38 | 0.37 ± 0.09 | 0.39 ± 0.29* | 0.40 ± 0.08 | 0.29 ± 0.29** | 0.54 ± 0.11*** |
CP-B/C | 0.18 ± 0.18 | 0.38 ± 0.13 | 0.15 ± 0.20 | 0.46 ± 0.17 | 0.09 ± 0.10* | 0.52 ± 0.14** |