Management of Isolated Liver Metastases

Rebecca Miksad

Douglas W. Hanto

Robert D. Timmerman

Steven Come

INTRODUCTION

About half of patients with metastatic breast cancer develop liver metastases, a finding that generally portends a poor prognosis (1): median overall survival (OS) of 4 to 22 months (1, 2). Since liver metastases commonly occur in the setting of concurrent extrahepatic metastases, liver involvement is generally considered to be a manifestation of disseminated disease, and patients are usually treated with systemic therapy (2) (see Chapters 33 and 34). However, in modern studies a minority of patients manifests metastatic breast cancer limited to the liver (1, 2). Localized, liver-directed treatments have achieved some success in other cancers, especially when disease is limited to the liver, and these approaches have been applied to breast cancer patients. Published studies have evaluated the safety and benefit of a range of liver-directed treatments: hepatic resection (Table 81-1), radiofrequency ablation (RFA), transarterial chemoembolization (TACE) with and without drugeluting beads, radioembolization, intraarterial chemotherapy, stereotactic body radiation therapy (SBRT), brachytherapy, and interstitial laser therapy (ILT) (Table 81-2). However, the comparative efficacies of these approaches remain unknown because there are limited prospective studies and no randomized controlled trials (Table 81-3). Furthermore, identifying appropriate patients remains a challenge, and the carefully selected patients in published series may represent good prognosis subgroups independent of the therapeutic approach.

Nevertheless, studies suggest that treatment of metastatic breast cancer limited to the liver may benefit some patients. In addition, as improvements in systemic therapies offer better control of metastatic disease and longer survival, more patients may need localized management of liver metastases. This chapter reviews localized, liver-directed treatment of hepatic metastases in breast cancer, details specific clinical considerations for each treatment option, and describes the available data for common and emerging approaches.

GENERAL SELECTION CRITERIA FOR LIVER-DIRECTED TREATMENT OF METASTATIC BREAST CANCER IN THE LIVER

Careful patient selection is essential to optimize the riskbenefit ratio of liver-directed approaches for all cancers with liver involvement (Table 81-4). The patients most likely to benefit from liver-directed approaches have a good overall prognosis (see Chapters 30 and 31) such that progression of other disease sites and/or comorbidities do not negate any disease control achieved in the liver. General criteria are controlled primary disease, limited metastatic disease in the liver (both number and size of lesions), longer disease-free intervals, a younger age, and a higher performance status (1, 2 and 3). On the basis of the more extensive experience of liver metastases in colorectal cancer, the presence of extrahepatic metastatic or residual primary breast cancer is commonly (4), although not always (5, 6), considered a contraindication to liver-directed therapy.

Evaluation before proceeding with liver-directed therapy for metastases from breast cancer should define the extent of disease as well as the potential responsiveness to systemic therapy. These factors may aid risk assessment and decision making regarding the role of liver-directed and/or systemic therapy. Commonly used imaging studies are (i) computed tomography (CT) of the chest to rule out pulmonary and mediastinal disease; (ii) triphasic CT scan of the abdomen and pelvis to evaluate the number and location of liver metastases in order to facilitate procedure planning and to rule out other intraabdominal disease; and (iii) a bone scan to rule out bone metastases. A PET scan may be useful to identify extrahepatic disease (see Chapters 33 and 34), and magnetic resonance imaging (MRI) may provide more detailed evaluation of the liver.
The need for thorough staging was highlighted in a series of 90 breast cancer patients evaluated for resection of liver metastases: 60% were deemed ineligible preoperatively because of extrahepatic metastases, 22% had unresectable extrahepatic disease at exploratory laparotomy, and only 10% ultimately underwent resection (7).

TABLE 81-1 Data for Hepatic Resection of Metastatic Breast Cancer to the Liver (selected)

Authors	Year	Type of Study	No. of Breast Cancer Patients (n)	Postoperative Mortality (%)	Median Survival (mos)	5-Year Survival (%)
Maksan et al. (7)	2000	CS	90	0	—	51
Selzner et al. (10)	2000	CS	17	6	25	22
Yoshimoto et al. (15)	2000	CS	25	—	34	27
Elias et al. (9)	2003	CS	54	0	34	50 (3 yrs)
Vlastos et al. (4)	2004	CR	31	0	63	61
D’Annibale et al. (3)	2005	CS	18	0	32	30
Adam et al. (6)	2006	CR	85	0	32	37
Martinez et al. (19)	2006	CR	20	—	32	33
Caralt et al. (60)	2008	CR	12	0	35.9	33
Lubrano et al. (16)	2008	CR	16	0	42	33
Thelen et al. (18)	2008	CR	39	—	—	42
Hoffmann K (14)	2010	CS	41	0	58	48
Abbott et al. (13)	2012	CS	86	0	57	43.6
van Walsum et al. (12)	2012	CR	32	0	55	37
Dash (—) represents unknown/not reported; CS, case series; CR, chart review; mos, months; n, number; yrs, years.

Palliative liver-directed treatment may also be beneficial if liver metastases impair quality of life. However, recent improvements in outcomes for endocrine-responsive, HER2/neu-amplified cancers and palliative single-agent chemotherapy have improved the effectiveness and tolerability of systemic therapy (8). Therefore, the risks and benefits of liver-directed therapies should be evaluated in context of potential systemic treatment options (2) (see Chapters 72, 73, 74, 75, 76, 77, 78 and 79).

SURGICAL APPROACH

Hepatic Resection

Hepatic resection (metastectomy) is the most commonly available liver-directed option for breast cancer patients. The morbidity and mortality of hepatic resection has declined significantly over the past two decades because of improvements in (i) understanding of intrahepatic segmental anatomy; (ii) imaging techniques (three-dimensional CT and intraoperative ultrasound) to characterize the tumor; (iii) anesthetic management; (iv) surgical techniques (preoperative portal vein embolization, segmental and anatomic resections, vascular inflow occlusion, maintenance of low central venous pressure, devices for safer division of the liver parenchyma and for maintenance of hemostasis); (v) laparoscopic hepatic resection approaches; (vi) understanding of negative risk factors (steatosis, remnant liver volume, and preoperative chemotherapy); and (vii) postoperative care. As a result, surgical resection is a technically safe option for most patients with metastatic breast cancer limited to the liver.

Patients deemed to be surgical candidates after preoperative screening undergo additional evaluation in the operating room. Prior to hepatic resection, patients are often explored to rule out extrahepatic, intraabdominal disease. Intraoperative ultrasound may identify additional liver lesions not imaged preoperatively, characterize the exact location of the lesion(s), and define the proximity of lesions to venous structures. The value of this additional exploration was demonstrated by a series of 108 breast cancer patients considered for hepatic resection after extensive preoperative evaluation with imaging (6). Over a 20-year period, 23% were found to have unresectable extrahepatic or hepatic disease during abdominal exploration, and an additional 13% had unexpected, but resectable, intraabdominal disease (6). Of the 85 patients who ultimately underwent hepatic resection, an R0 (microscopically negative margin) resection was attained in only 65%, while an R1 microscopically positive margin) resection was achieved in 18%, and an R2 resection (macroscopically positive margin) was carried out in 17% (6).

Selection Criteria for Hepatic Resection

In addition to the general selection criteria outlined above, hepatic resection candidates must have lesions that can be completely resected while leaving an adequately sized liver remnant along with its hilum (i.e., vascular and ductal continuity to the body). Because the function and architecture of the liver are integrated, adequate liver function can be maintained if there is a critical volume of intact liver and a contiguous bile duct system (20% of a normal liver, 40% of the liver if steatosis is present). If a small liver remnant is anticipated, a patient may benefit from preoperative portal vein embolization (right or left) of the lobe to be resected. This causes hypertrophy of the opposite lobe (the lobe that will become the liver remnant), thereby decreasing the risk of postoperative hepatic insufficiency.

Although there are limited data, the combination of hepatic resection and RFA has been explored for metastatic disease in both lobes or when one or more lesions are technically unresectable (4). In addition, while patients with extrahepatic metastases are traditionally excluded from resection, some series include patients with controlled extrahepatic disease (6).

TABLE 81-2 Data for Nonsurgical Locoregional Treatments for Metastatic Breast Cancer to the Liver (selected)

Authors		Year	Type of Study	No. of Breast Cancer Patients with Liver Metastases (n)	Median Survival (mos)	Survival (Time point)
Radiofrequency Ablation (RFA)
@	Livraghi et al. (34)	2001	NRT	24	—	96% (4 to 44 mos)
@	Lawes et al. (35)	2006	NRT	19	—	41% (2.5 yrs)
@	Gunabasham et al. (36)	2007	NRT	14	—	64% (1 yrs)
@	Sofocleous et al. (5)	2007	CR	12	60	30% (5 yrs)
@	Meloni et al.^a (33)	2009	CR	52	29.9	27% (5 yrs)
@	Jakobs et al. (37)	2009	CR	43	58.6	—
@	Carrafiello et al. (38)	2011	CR	13	10.9 (mean)	—
Transarterial Chemoembolization (TACE)
@	Li et al. (27)	2005	CR	TACE = 28	TACE = 28	TACE = 13.01% (3 yrs)
@	@	@	@	SC = 20	SC = 18	SC = 11.29% (3 yrs)
@	Buijs et al. (20)	2007	CR	14	25	35% (3 yrs)
@	Cho et al. (61)	2010	CR	10	12	—
@	Vogl et al. (24, 25)	2010 (2011)	NRT	208 (TACE + LITT: 161)	18.5 (TACE + LITT: 32.5, mean)	33% (3 yrs) (TACE + LITT: 36.6% [3 yrs])
@	Duan et al. (26)	2011	CR	TACE + SC = 44 SC = 43	—	TACE + SC = 47.6% (3 yrs) SC = 7.4% (3 yrs)
@	Martin et al. (22)^b 2012	CS	40	47	—
Hepatic Arterial Infusion (HAI)
@	Fraschini et al. (28) 1987	NRT	31	11
@	Ikeda et al. (62) 1999	NRT	28	25.3	—
@	Camacho et al. (29) 2007	NRT	10	—
Radioembolization
@	Bangash et al. (63) 2007	NRT	27	<25% tumor burden: 9.4 mos >25% tumor burden: 2.0 mos	—
@	Hoffmann et al. (59)	2010	CR	16	—	—
Stereotactic Body Radiation Therapy (SBRT)
@	Wulf et al. (52)	2001	NRT	6	—
@	Herfarth et al. (47)	2004	NRT	14	—	—
@	Katz et al. (54)	2007	CR	16	—	about 50% (2 yrs)
@	Milano et al. (51)	2012	NRT	39 breast cancer patients with 13 liver metastases treated with SBRT	—	47% (6 yrs) for all breast patients
Brachytherapy
@	Wieners et al. (58)	2011	NRT	41	—	60% (18 mos)
@	Collettini et al. (57)	2012	CR	37	18	—
Interstitial Laser Therapy (ILT)
@	Mack et al. (55)	2004	NRT	232	4.3 yrs	41% (5 yrs)
^a Includes 9 patients from Livraghi study (34).
^bTACE with drug-eluting beads loaded with doxorubicin.
Dash (—) represents unknown/not reported; NRT, nonrandomized trial; CS, case series; CR, chart review; mos, months; yrs, years; TACE, transarterial chemoembolization; SC, systemic chemotherapy; LITT, laser-induced chemotherapy.