Chapter 84A Radioembolization for liver tumors

Overview

Interventional oncology is a field that is establishing its role in the management of liver tumors. The techniques used in this field are targeted to the tumors to allow delivery of toxic chemotherapeutic, radiotherapeutic, and thermal doses with minimal toxicity to normal structures; this minimizes the systemic toxicity associated with these therapeutic options and allows delivery of doses to the tumor that would not be possible using systemic delivery methods. In this chapter, the general concepts associated with radioembolization and its specific use in different primary and secondary malignancies of the liver will be discussed (Table 84A.1)

Table 84A.1 Radionuclide Carrier Combinations for Radioembolization

Vascular Anatomy of the Liver and Its Tumors (See Chapter 1B)

Portal Vein

The normal hepatic parenchyma is predominantly supplied by the portal vein. This unique vascularization lets the hepatocytes carry out their metabolic functions on the numerous substances absorbed from the gastrointestinal (GI) tract. It also makes the liver a common site of metastasis for GI cancers.

Hepatic Artery

The proper hepatic artery that supplies the liver is a continuation of the common hepatic artery, a branch of the celiac trunk. The proper hepatic artery branches into the right and left hepatic arteries, which supply the corresponding lobes of the liver. The cystic artery supplying the gallbladder usually arises from the right hepatic artery, and the right and left hepatic arteries give rise to the segmental branches; these eventually branch into the small vessels found in the portal triads.

Unlike the normal hepatic parenchyma, liver tumors are supplied primarily by hepatic arterial blood and are hypervascular structures compared with the surrounding uninvolved parenchyma (see Chapter 86). Hepatic tumors may parasitize arterial blood flow from the arteries that supply segments adjacent to the segment the tumor is in and those of surrounding organs, such as the stomach. This relative arterial hypervascularity of the tumors with respect to the normal parenchyma has been the basic principle behind transarterial therapies for targeting liver tumors.

History of Radioembolization

External Radiation (See Chapter 84B)

The use of external radiation for liver tumors has traditionally held limited value because of the sensitivity of the normal hepatic parenchyma to the radiation dose (Ingold et al, 1965; Geschwind et al, 2004). A dose greater than 35 Gy has led to the development of a radiation-induced liver disease (RILD); a clinical syndrome of ascites, anicteric hepatomegaly, and elevation of liver enzymes. Conformal-beam therapy, which uses a three-dimensional approach rather than broad axial plane techniques, has been shown to minimize the toxicity to normal hepatic parenchyma (Dawson et al, 1999). However, even when using conformal-beam therapy, the radiation delivered to normal hepatic parenchyma limits the maximum dose that can be delivered to the tumor without compromising safety.

External-beam radiation had limitations in treating liver tumors in specific locations; this includes tumors near the dome of the liver, which carry the risk of exposing the lungs to radiation and increasing the risk of radiation pneumonitis, and tumors in the caudate lobe in close proximity to the porta hepatis, where risk of damage to the major biliary and vascular structures is a consideration. Dose fractionation, dose delivery to the tumor in fractions, is beneficial in targeting radiosensitive and radioresistant malignant cells at different sessions, but multiple treatments are required. The normal tissue complications probability (NTCP) is an important parameter when calculating radiation dose delivery (Dawson et al, 2002), but its use has been challenged (Langer et al, 1998). Respiratory movement must be considered to minimize the incidence of normal tissue complications, although recent advances in the technology of external-beam radiation have been shown to increase safety and efficacy of this technique, such as stereotactic body radiotherapy, proton radiotherapy, and carbon ion radiotherapy; these are establishing the role of external-beam radiotherapy in the management of liver tumors (Dawson & Guha, 2008; Fukumitsu et al, 2009; Nomiya et al, 2008).

Radioembolization

Given the distinct vasculature of hepatic tumors described above, radioembolization provides a mode of delivering localized radiotherapy to the tumor. A radiation dose of up to 150 Gy can be administered, minimizing the complications of external radiation. Radioembolization has been shown to have promising outcomes in primary and secondary hepatic malignancies by numerous investigators.

Radioembolization was first studied by Nolan and Grady (1969) using yttrium-90 oxide (⁹⁰Y₂O₃) contained in a metal particle 50 to 100 µm in size. Their study consisted of a small number of patients but showed a favorable response, observed by the reduction in size of palpable masses. The next ⁹⁰Y study was published in 1982 by Mantravadi and colleagues and studied the effect and distribution of ⁹⁰Y after whole-liver delivery via the hepatic artery; the authors concluded that patients with hypervascular tumors are much more likely to benefit from this treatment. A dose-escalation study on animals was performed, which formed the basis of the Phase I dose-escalation evaluations in humans (Wollner et al, 1987). Shepherd and colleagues (1992) conducted a Phase I dose-escalation study using ⁹⁰Y microspheres in 10 patients. Extrahepatic shunting was assessed using technetium-99m–labeled macroaggregated albumin (^99mTc-MAA). In contrast to earlier studies, none of the patients in this study experienced hematologic toxicity, a fact that emphasizes the value of assessing angiographic findings and extrahepatic shunting before treatment (Shepherd et al, 1992). The technical aspects of dosimetry, radioassays, safety, and efficacy were discussed by many studies that followed (Andrews et al, 1994; Yan et al, 1993). Lau and colleagues (1994) showed that tumor response was proportional to the dose delivered, with improved survival in patients receiving more than 120 Gy. Recent data on the safety and efficacy of this treatment are discussed below.

⁹⁰YTTRIUM Microspheres

Overview

⁹⁰Yttrium is the radioactive element most commonly used for radioembolization. It is a pure β-emitter with a half-life of 64.2 hours, and it decays into the stable element zirconium-90. Tissue penetration of the emissions ranges from 2.5 to 11 mm.

Pretreatment Evaluation

Clinical Evaluation

Clinical evaluation is required to stratify patients according to the Eastern Cooperative Oncology Group (ECOG) performance status, and those with a score greater than two are not considered ideal candidates for this treatment.

Laboratory Workup

The laboratory workup is important to assess the pretreatment functional status of the liver. This should include assessment of liver enzymes (alanine aminotransferase [ALT], aspartate aminotransferase [AST], alkaline phosphatase), bilirubin, albumin, and tumor markers such as α-fetoprotein (AFP).

Pretreatment Angiography and Coil Embolization

Radioembolization requires pretreatment diagnostic mesenteric angiography. The aortogram, superior mesenteric angiogram, and celiac trunk angiogram allow the interventional radiologist an opportunity to study the vascular anatomy of the liver and its surrounding structures in detail. The patency of the portal vein and the presence of arterioportal shunting are assessed. The inadvertent spread of the microspheres is prevented by a meticulous study of the vascular anatomy of the liver and collateral nontarget flow (Covey et al, 2002). Coil embolization of nontarget vessels may be necessary to decrease the unintended deposition of microspheres. Some examples of vessels that may need to be embolized are inferior esophageal, left inferior phrenic, accessory left gastric, supraduodenal, and retroduodenal arteries. The minimal incidence of complications following coil embolization and the grave clinical consequences associated with the inadvertent deposition of microspheres in the stomach, duodenum, or pancreas favor prophylactic coil embolization before radioembolization in select cases.

^99mTechnetium Macroaggregated Albumin Scan

The ^99mTc-MAA scan is used to assess splanchnic and pulmonary shunting. This pretreatment nuclear scan is important to prevent certain complications associated with the treatment. The proximity of the duodenum and stomach to the liver may make it difficult to assess shunting to these portions of the GI tract by using nuclear medicine scans alone. Thus, it is important to correlate the findings of angiography to the findings of the ^99mTc-MAA scan. The lung shunt fraction (LSF) is used to calculate the dose delivered to the lungs, and adjustment for this parameter minimizes the risk of radiation pneumonitis that may be seen after radioembolization.

All hepatic vessels are assessed during the angiogram, and the arteries feeding the tumor are studied in detail. As the tumor may parasitize blood flow from surrounding vessels, it is necessary to study its vascular supply in detail. Failure to recognize a vessel supplying the tumor may lead to the incomplete targeting of the lesion and treatment failure.

Available Devices

Tumor Involvement of Liver (%)	Dose (Whole Liver)
≤25	2 GBq
25-50	2.5 GBq
>50	3 GBq

* Sirtex Medical, Woburn, MA

The model of dosimetry for SIR-Spheres is based on whole-liver infusion. The calculated gigabecquerels of the whole liver are multiplied by the percentage of the target site as a proportion of the whole liver. The most widely used body surface area (BSA) method is as follows:

A is the activity in gigabecquerels, BSA is the body surface area in meters squared, and % tumor burden is the percentage of the liver involved by tumor.

New Concepts

Radiation Segmentectomy

Recently, the concept of radiation segmentectomy has been discussed. This is schematically shown in Figure 84A.1B. The current dosimetry models assume lobar infusion; however, if the tumor is localized to two or fewer liver segments, the interventional radiologist can infuse the dose at the segmental level. This maximizes the dose delivery to the tumor and minimizes delivery of radioactive microspheres to the normal parenchyma (Riaz et al, 2010), an approach that has proven to be a safe and effective method of treatment delivery.

Extended Shelf Life Microspheres

Lewandowski and colleagues (2009b) recently published a report on the concept of extended shelf life microspheres, which can be used for large or multifocal tumors. As stated above, the difference in vials of different activities is the number of microspheres. If the vials of high activities (more microspheres) are allowed to stay on the shelf longer, the activity per microsphere will decrease. This allows the delivery of a higher number of microspheres to larger volumes of the liver with minimal radiotoxicity to the normal parenchyma (Lewandowski et al, 2009b).

Radiation Lobectomy

An animal study concluded that portal fibrosis was seen after the use of radioactive microspheres (Wollner et al, 1988). Gaba and colleagues (2009) recently published a comprehensive analysis on the concept of radiation lobectomy, whereby infusion at the lobar level was used to treat the tumors. Because of fibrosis of the normal parenchyma, decreases in volumes of the treated lobe were seen. Furthermore, compensatory increases in the volumes of the untreated lobes were also observed.

New Concepts

Blood Flow Patterns for Dose Calculation

Kennedy and colleagues (2010) have recently published their analysis on computer modeling of ⁹⁰Y resin microsphere transport in the hepatic arterial tree. They conclude that computer simulations of both blood flow patterns and microsphere dynamics have the potential to provide insight on methods to optimize microsphere implantation into hepatic tumors while sparing normal tissue.