Portal vein embolization for arterio-portal shunt closure in hepatocellular carcinoma to facilitate selective internal radiation therapy

Abstract

BACKGROUND

Selective internal radiation therapy (SIRT) has been widely used in the treatment of hepatocellular carcinoma (HCC). However, marked intrahepatic shunt (IHS) secondary to arterio-portal shunt (APS) often precludes safe SIRT administration.

AIM

To evaluate the efficacy and safety of portal vein embolization (PVE) in reducing IHS and enabling SIRT therapy in HCC patients with APS.

METHODS

This retrospective study enrolled seven HCC patients with APS confirmed by hepatic angiography during mapping. PVE was performed to close the APS outlet and single photon emission computed tomography/computed tomography was performed to determine the distribution of (99m) technetium-labeled macroaggregated albumin. These patients meeting the criteria for SIRT subsequently underwent Yttrium-90 (Y-90) resin microspheres administration. Changes in liver function and complications were monitored during the 6-month follow-up period.

RESULTS

Complete APS occlusion was achieved in six patients (n = 6, 6/7). The hepatopulmonary shunt rate was below 20% (4.65%-16.95%) in five of six patients with APS occluded, and Y-90-SIRT was subsequently administered in these patients. No severe procedural complications occurred.

CONCLUSION

PVE is a safe and effective strategy to reduce IHS in HCC patients with APS, thereby improving eligibility for SIRT.

Key Words: Hepatocellular carcinoma; Portal vein embolization; Arterio-portal shunt; Hepatopulmonary shunt; Macroaggregated albumin; Selective internal radiation therapy

Core Tip: Marked arterio-portal shunt (APS) can be a contraindication for selective internal radiation therapy (SIRT) due to the risk of radiation-induced liver toxicity or pneumonitis. In this study, portal vein embolization (PVE) was performed to close the APS outlet in seven hepatocellular carcinoma (HCC) patients, with complete occlusion achieved in six (n = 6, 6/7). Five patients subsequently completed SIRT. PVE is a safe and effective strategy to reduce intrahepatic shunting in advanced HCC patients with APS, thereby improving eligibility for SIRT.

INTRODUCTION

Hepatocellular carcinoma (HCC) ranks as the fifth most common malignant tumor worldwide and the fourth leading cause of cancer-related mortality[1]. The global incidence of HCC continues to rise, with most cases associated with chronic liver disease, viral hepatitis, liver cirrhosis, and metabolic dysfunction. HCC is insidious in onset; most patients present at an advanced stage with lost opportunities for curative therapies, including surgical resection, liver transplantation, and tumor ablation[2-4]. Palliative locoregional and systemic therapies are therefore the mainstay treatment, including transarterial chemoembolization (TACE), selective internal radiation therapy (SIRT), hepatic artery infusion chemotherapy, targeted therapy and immunotherapy[5].

Yttrium-90 (Y-90) SIRT (radioembolization) delivers internal radiation via resin or glass microspheres into tumor-feeding arteries, achieving targeted tumor killing with minimal normal liver injury. It is increasingly applied for intermediate-advanced HCC due to favorable efficacy and safety profiles[6-11]. However, tumor vascular invasion[12], prior biopsy[13,14], or TACE[15] may induce hepatic arterio-portal shunt (APS), leading to marked intrahepatic shunt (IHS) and hepatopulmonary shunt (HPS)[16]. Significant IHS/HPS can cause radiation-induced liver disease, pneumonitis, or insufficient tumor dose deposition, thus precluding or restricting SIRT.

Traditional APS management includes transcatheter arterial embolization (TAE)[17-21] and systemic therapies[22,23]. Portal vein embolization (PVE) is conventionally used to increase future liver remnant volume before extended hepatectomy. We first reported the use of PVE to resolve severe APS and facilitate safe SIRT in a case report published in December 2023[24]. In the present study, we extended the sample size to further validate the efficacy, feasibility, and safety of PVE for APS closure to enable SIRT in advanced HCC.

MATERIALS AND METHODS

Patients and selection criteria

This retrospective study was approved by the hospital ethics committee. All patients provided written informed consent for participation. A review was conducted of 151 HCC patients who underwent (99m) technetium-labeled macroaggregated albumin [(99m)Tc-MAA] scintigraphy before planned SIRT.

The inclusion criteria were: (1) Aged 18-85 years; (2) Eastern Cooperative Oncology Group (ECOG) performance score of 0-1; (3) Child-Pugh class of A or B; (4) HCC diagnosed based on histological examination, contrast-enhanced computed tomography (CT), or magnetic resonance imaging findings; and (5) Diagnosis of APS by hepatic angiography and treated by PVE to close the APS outlet before (99m)Tc-MAA scintigraphy from March 2023 to October 2024. Exclusion criteria were: (1) ECOG score > 1; (2) Child-Pugh grade C; (3) Presence of other types of cancer; and (4) Lack of complete medical information.

Procedures

Firstly, hepatic angiography was performed to determine the presence of APS and identify the responsible portal vein branch (Figure 1A and B). The contralateral or ipsilateral portal vein branches were punctured with an EV needle (Hakko, Nagano, Japan) under ultrasound guidance. A 0.035-inch guide wire was inserted into the portal vein, and then a 5F catheter (Hanaco Medical, Tianjin, China) was placed in the main portal vein for angiography to show all branches and confirm the responsible portal vein branch that showed fistula with the hepatic artery. Subsequently, the 5F catheter was inserted into the ipsilateral first-order branch of the portal vein, and a 2.8-F microcatheter (Boston Scientific Corporation, Marlborough, MA, United States) was superselected into the distal branch of the responsible portal vein. Next, an N-butyl cyanoacrylate (NBCA) (Compont, Beijing, China) and iodized oil (Lipiodol; Andre Guerbet, Aulnay-Sous-Bois, France) mixture (1:1-1:3) was injected and the microcatheter was gradually pulled out until the responsible portal vein was completely embolized (Figure 1C). In order to prevent regurgitation of the mixed fluid, the proximal portal vein was embolized with interlock microcoils (Tornado, Cook, Bloomington, IN, United States; Azur, Terumo, Somerset, NJ, United States) in some patients. Hepatic arteriography was performed again to confirm closure of the APS (Figure 1D). Super-selective hepatic arteriography was implemented by cannulation of the tumor feeding arteries with a new 2.8-F microcatheter. Cone beam CT with contrast administration through the microcatheter was conducted to determine tumor feeding arteries. (99m)Tc-MAA (Xinke, Shanghai, China) was injected through the tumor-feeding arteries and single photon emission CT (SPECT)/CT (Symbia T16; Siemens Healthcare, Germany) was performed to determine the distribution of (99m) Tc-MAA (Figure 2). Patients meeting the criteria for SIRT subsequently underwent SIR-Spheres^® Y-90 resin microspheres (Sirtex Medical Pty Ltd, St Leonards, NSW, Australia) administration the following week using the Medical Internal Radiation Dose partition model, in accordance with established procedure guidelines[25].

Figure 1 Arterio-portal shunt detection on hepatic arteriography and closure by portal vein embolization. A: Hepatic arteriography detected Arterio-portal shunt (APS); B: Hepatic right arteriography with microcatheter; C: Embolization of the responsible portal vein with N-butyl cyanoacrylate and iodized oil mixture; D: Repeated hepatic right artery angiography with a microcatheter showed that the APS had been completely blocked.

Figure 2 Single photon emission computed tomography/computed tomography scintigraphy following injection of (99m) technetium-labeled macroaggregated albumin ((99m)Tc-MAA) into the tumor-feeding artery. A: Planar scintigraphy to calculate the hepatopulmonary shunt. B: Transverse, sagittal and coronal single photon emission computed tomography/computed tomography imaging of the distribution of (99m)Tc-MAA in tumor lesions, normal liver tissue and other systemic regions.

RESULTS

Between March 2023 and October 2024, mapping ((99m)Tc-MAA scintigraphy before planned SIRT) was performed in 151 HCC patients in Shanghai Eastern Hepatobiliary Surgery Hospital, of which nine had APS (9/151, 5.96%) on hepatic angiography.

Of these nine patients with APS, three had their fistulas successfully blocked using only the PVE method. Six patients initially received TAE to occlude arterio-portal fistulas. Subsequent hepatic arteriography revealed successful occlusion in two patients and failure in four patients. The four failed patients then underwent remedial PVE. In summary, seven patients including 3 patients who received PVE only and four patients who received remedial PVE were included (Figure 3).

Figure 3 Flowchart of closure of the hepatic arterio-portal shunt in hepatocellular carcinoma patients during the mapping procedure. TAE: Transcatheter arterial embolization; PVE: Portal vein embolization.

The demographics and clinical characteristics of the seven study patients (7 males and 0 females; median age, 59 years; range, 44-61 years) are summarized in Table 1. The Barcelona Clinic Liver Cancer stage was B in one patient (1/7) and C in six patients (6/7). The success rate of PVE for occluding APS was 85.71% (6/7). No abnormal distribution of (99m)Tc-MAA was found in the normal livers of the six patients on SPECT/CT. The HPS rate was below 20% (4.65%-16.95%) in five of six patients with APS occluded, and Y-90-SIRT was subsequently administered in these cases. The remaining one patient (No. 4) had a high HPS rate of 53.87%, thus failed mapping and did not receive SIRT treatment. In all seven patients who underwent PVE for embolization of APS, none of the severe complications associated with PVE, such as intra-abdominal hemorrhage, upper gastrointestinal bleeding, hepatic failure, or ascites, were observed postoperatively. All five patients who completed SIRT treatment did not develop any Y-90-related severe complications such as radiation-induced liver injury or radiation pneumonitis in the 6-month follow-up period.

Table 1 Characteristics of 7 included patients using portal vein embolization to block arterio-portal shunt.

Patient No.	Sex	Age (year)	Child-Pugh	BCLC	TAE	PVE	APS closure	HPS (%)	SIRT	Complication
1	Male	59	A6	C	No	Yes	Yes	4.65	Yes	No
2	Male	48	A6	C	Yes	Yes	No	/	No	No
3	Male	61	A5	B	Yes	Yes	Yes	16.42	Yes	No
4	Male	44	A6	C	Yes	Yes	Yes	53.87	No	No
5	Male	60	A5	C	No	Yes	Yes	16.95	Yes	No
6	Male	45	A6	C	Yes	Yes	Yes	5.21	Yes	No
7	Male	59	A6	C	No	Yes	Yes	8.50	Yes	No

BCLC: Barcelona Clinic Liver Cancer; TAE: Transcatheter arterial embolization; PVE: Portal vein embolization; APS: Arterio-portal shunt; HPS: Hepatopulmonary shunt; SIRT: Selective internal radiation therapy.

DISCUSSION

HCC combined with hepatic APS is commonly encountered in liver tumors treated with interventional therapy. APS exacerbates portal hypertension and increases the incidence of severe complications, such as esophagus varicose rupture, refractory ascites, and hepatic encephalopathy[26]. In addition, HCC patients with severe APS are usually unsuitable for SIRT treatment as radioactive microspheres can flow through the fistula to normal liver tissue resulting in poorer deposition of radioactive microspheres in the tumor tissue and aggravating the normal liver tissue radiation-induced injury. To facilitate subsequent TACE or SIRT, various methods have been attempted to occlude significant APS, including systemic therapy[22,23], TAE[17-21], and portal vein balloon occlusion[19]. However, the optimal method has not yet been elucidated, and other new techniques may be necessary to successfully alleviate the IHS and HPS caused by APS.

PVE is a widely used technique for liver regeneration, which can completely embolize the outlet of the APS, indicating its potential to facilitate SIRT by reducing IHS and HPS. In the present study, NBCA-embolization of the corresponding portal vein branches to block APS achieved success in six of seven (85.71%) patients, and only one patient with multiple intrahepatic arterio-portal fistulas failed. It is worth noting that four patients with failed prior TAE received remedial PVE which was successful in three. Systemic anti-angiogenic agents have been used to manage APS or arteriovenous shunt (AVS). Theysohn et al[27] reported that sorafenib reduced HPS in four of seven patients; however, three patients became ineligible for SIRT due to tumor progression and deteriorated liver function. This strategy is unsuitable for patients with rapidly growing tumors and short life expectancy, who may lose the opportunity for SIRT. Balancing the timing of anti-angiogenic therapy and SIRT remains challenging. TAE is currently the most commonly used interventional treatment method for APS. In the study by Izaki et al[28], TAE was performed in 15 patients with portal vein tumor thrombus (PVTT) and APS to the main portal trunk or first-order branch, and five patients with hepatic vein tumor thrombus and AVS, with radiation therapy added in suitable cases. Shunts disappeared in seven of 13 patients available for follow-up. Murata et al[19] reported that only two of seven patients treated with TAE alone achieved complete APS occlusion, and portal vein balloon occlusion combined with TACE achieved durable complete APS closure in nine of 14 patients. This result is consistent with our findings and confirms the potential of occluding APS via the portal vein approach. PVE is faster and more efficient in blocking APS, shortens waiting time, and does not compromise SIRT efficacy. Furthermore, no severe PVE-related complications, such as liver failure, gastrointestinal hemorrhage, or abdominal bleeding, were observed in this study, indicating its good safety performance in managing APS.

This study has several limitations that should be acknowledged. First, this was a retrospective observational study with a small sample size, which limits the statistical power and generalizability of the findings. Largescale prospective randomized controlled trials are needed to confirm the safety and efficacy of PVE for APS occlusion. Second, we did not quantitatively measure pulmonary shunt fraction before and after embolization; therefore, we cannot confirm whether PVE effectively reduces intrapulmonary shunting. Third, the long term effects of PVE on tumor metastasis risk remain unclear. Portal vein occlusion may theoretically affect hematogenous tumor spread, and its impact on the progression of PVTT and patient survival requires further investigation. Finally, the study lacks long term oncologic outcomes and direct comparative data with other conventional APS occlusion strategies, limiting definitive conclusions regarding clinical superiority.

CONCLUSION

Our clinical practice suggests the feasibility and safety of the closure of APS with PVE during the mapping procedure to enable subsequent SIRT in HCC patients.

ACKNOWLEDGEMENTS

The authors thank all patients and clinical staff who participated in and supported this study.