JVIR twitter

Monday, February 4, 2019

Feasibility of Boosted Radioembolization for Hepatocellular Carcinoma Larger than 5 cm 

Clinical question
Are Y90 doses larger than 150 Gy beneficial in the treatment of HCCs measuring more than 5 cm?

Take-away point
Boosted Y90 doses seem to show promising results, with 80% complete response of the target tumor, however 15% of patients suffered symptomatic biliary strictures requiring treatment.

Kim HC, Kim YJ, Lee JH, Suh KS, Chung JW. Feasibility of Boosted Radioembolization for Hepatocellular Carcinoma Larger than 5 cm. J Vasc Interv Radiol. 2019 Jan;30(1):1-8. doi: 10.1016/j.jvir.2018.07.002. Epub 2018 Oct 4. PubMed PMID: 30293734.

Click here for abstract

Study design
Retrospective review 20 patients

Funding source
Self-funded, unfunded

Seoul National University Hospital

This retrospective study evaluated the outcomes of radioembolization with glass microspheres using doses larger than 150 Gy for the treatment of HCCs larger than 5 cm. Other inclusion criteria included nodular tumor, BCLC stage A/B and tumors treated with a single Y90 session. Patients were excluded if the tumor was infiltrative, smaller than 5 cm, BCLC stage C/D, 2 treatment sessions for bilobar HCC, previous treatment for HCC and extra-hepatic metastasis.

Twenty, non-consecutive patients were included, 13 patients had a single tumor, 7 multiple tumors. Average tumor diameter was 7.6 cm (range 5.1-13 cm). Radioembolization was performed in a) segmenta/sub-segmental branches, if the tumor was peripheral, b) in lobar arteries if the tumor was central and small branches arising from the lobar branch fed the tumor and c) dominant tumor feeding artery or caudate artery if the tumor had a dominant artery or the tumor was central and supplied by the caudate artery. The authors sought to treat the target liver volume with doses that ranged from 180-240 Gy, with an intended tumor dose of 300 Gy.

The mean percentage of treated liver volume divided by whole liver volume was 61% (33-79%), mean total radiation activity was 4.96GBq (1.63-9.15GBq), mean target perfused tissue dose 263.5 Gy (156.2-550.6 Gy), median number of vials 4 (2-7 vials). 19.3% of vials were injected into a lobar artery, 45.8 % into a segmental artery and 35% into a sub-segmental artery. Five extra-hepatic feeding arteries were treated (inferior phrenic and right internal mammary artery. Seven patients suffered abdominal/chest pain requiring medication, 3 patients developed biliary strictures that required biliary drainage and antibiotic therapy, 1 patient developed asymptomatic biliary dilation. All 4 patients that presented with biliary stricture underwent treatment of the caudate artery.

Median follow up was 11.6 months (6.3 – 22 months). Complete response (CR) was seen in 60% of patients at 3 months, based on best tumor response, CR was seen in the entire liver and for the primary index tumor in 70% and 80% of patients, respectively. Local progression-free survival and progression free survival rates were 94% and 74% at 1 year respectively. 25% of patients underwent surgical resection, 1 patient had partial response at 1 month on CT and underwent resection 1.6 months after radioembolization, and on pathologic evaluation there was 60% necrosis. 4 patients had CR on a 3 month CT and underwent resection after 3 months showing complete necrosis on pathologic examination.

The authors highlight that the median total activity used in this study, 4.88 GBq, is higher than what has been reported for radiation segmentectomy (0.95-1.59 GBq) and for the previously reported boosted dose (2.9 GBq), with no instances of radiation induced liver disease (RILD). By applying surgical criteria for HCC surgical resection, if 40% of the total liver is not treated with Y90, the risk of RILD can be minimized even if a boosted dose is used for treatment. In this study the mean percentage of untreated liver was 38.9%. Complications in this study included seven patients that suffered chest/abdominal discomfort classified as Grade 2, 3 patients with a biliary stricture requiring antibiotics and drainage, and 1 patient with asymptomatic biliary stricture. The authors reinforce that care should be taken when treating a caudate artery branch, and that further studies are needed to identify the adequate dose to maximize tumor death and minimize biliary damage.

This study is limited by the retrospective nature of the study, the small sample size, the non-consecutive treatment of patients, and the lack of long term follow up.

The authors conclude that boosted dose radioembolization of large HCC shows favorable results, but with a relatively high biliary complication rate.

Figure- A) Hypervascular central HCC measuring 5.2 cm B) MIP image from C arm CT showing the hypervascular tumor being supplied by 3 major branches (marked by the arrows). C) 1 month follow up CT scan shows no viable tumor, D) MRI at 6 months shows no residual tumor enhancement and biliary dilation (arrowhead).


Boosted doses of Y90 microspheres appear to show promising results in the treatment of large HCCs. CR for the entire liver and for the primary tumor index was seen in 70% and 80% respectively, with local progression free survival and progression free survival rates at 1 year of 94% and 74% respectively. The authors highlight that from a technical standpoint, boosted doses of Y90 can be safely injected by sub-selecting the segmental and sub-segmental branches feeding the tumor, or by additional treatment of the dominant tumor supplying artery or caudate artery. The doses injected were clearly above previously published doses, with no instances of RILD, however if the caudate artery is treated with high doses there could be a high risk of biliary strictures requiring treatment. Larger studies are required to be able to confirm the results as well as maximization of caudate artery doses to maximize tumor death and minimize biliary damage.

Post Author
Carlos J. Guevara, MD
Assistant Professor
Department of Radiology, Interventional Radiology Division
University of Texas Health Sciences, Houston

No comments:

Post a Comment