Locoregional surgical and interventional therapies for advanced colorectal cancer liver metastases: expert consensus statements

HPB

DOI:10.1111/j.1477-2574.2012.00597.x

ORIGINAL ARTICLE

Locoregional surgical and interventional therapies for advanced colorectal cancer liver metastases: expert consensus statements Eddie K. Abdalla1, Todd W. Bauer2, Yun S. Chun3, Michael D'Angelica4, David A. Kooby5 & William R. Jarnagin4 1 Department of Surgery, Lebanese American University, Beirut, Lebanon, 2Department of Surgery, University of Virginia Health System, Charlottesville, VA, USA, 3Department of Surgical Oncology, Fox Chase Cancer Center, Philadelphia, PA, USA, 4Department of Surgery, Memorial Sloan–Kettering Cancer Center, New York, NY, USA and 5Department of Surgery, Emory University School of Medicine, Atlanta, GA, USA

Abstract Selection of the optimal surgical and interventional therapies for advanced colorectal cancer liver metastases (CRLM) requires multidisciplinary discussion of treatment strategies early in the trajectory of the individual patient's care. This paper reports on expert consensus on locoregional and interventional therapies for the treatment of advanced CRLM. Resection remains the reference treatment for patients with bilateral CRLM and synchronous presentation of primary and metastatic cancer. Patients with oligonodular bilateral CRLM may be candidates for one-stage multiple segmentectomies; two-stage resection with or without portal vein embolization may allow complete resection in patients with more advanced disease. After downsizing with preoperative systemic and/or regional therapy, curative-intent hepatectomy requires resection of all initial and currently known sites of disease; debulking procedures are not recommended. Many patients with synchronous primary disease and CRLM can safely undergo simultaneous resection of all disease. Staged resections should be considered for patients in whom the volume of the future liver remnant is anticipated to be marginal or inadequate, who have significant medical comorbid condition(s), or in whom extensive resections are required for the primary cancer and/or CRLM. Priority for liver-first or primary-first resection should depend on primary tumour-related symptoms or concern for the progression of marginally resectable CRLM during treatment of the primary disease. Chemotherapy delivered by hepatic arterial infusion represents a valid option in patients with liver-only disease, although it is best delivered in experienced centres. Ablation strategies are not recommended as first-line treatments for resectable CRLM alone or in combination with resection because of high local failure rates and limitations related to tumour size, multiplicity and intrahepatic location. Received 16 August 2012; accepted 20 August 2012

Correspondence Eddie K. Abdalla, Lebanese American University, University Medical Centre, Rizk Hospital, Zahar Street, Beirut 11-3288, Lebanon. Tel: + 961 1 200 800. Fax: + 961 1 200 816. E-mail: [email protected]

Approaches to bilateral colorectal cancer liver metastases Historically, criteria for the resectability of colorectal cancer liver metastases (CRLM) included factors such as a tumour size of

Proceedings of the Consensus Conference on Multidisciplinary Treatment of Colorectal Cancer Liver Metastases sponsored by the Americas HepatoPancreato-Biliary Association and co-sponsored by the Society of Surgical Oncology, the Society for Surgery of the Alimentary Tract and the University of Texas M. D. Anderson Cancer Center held in San Francisco, CA, USA; January 18, 2012.

HPB 2013, 15, 119–130

<5 cm, the presence of fewer than four lesions, and unilateral distribution.1–3 However, contemporary studies show that the bilateral distribution of CRLM is not an independent predictor of poor oncologic outcome, but, rather, presents a technical challenge to the achievement of a margin-negative resection and the simultaneous preservation of sufficient functional liver parenchyma.4 Contemporary consensus supports the suggestion that ability to remove all metastatic deposits leaving an adequate liver remnant is key to the definition of resectability.5 Modern chemotherapy and biologic agents have further expanded the proportion of patients with bilateral CRLM eligible for resection by downsizing tumours; response to chemotherapy is used by many investigators as a surrogate marker of tumour biologic behaviour and

© 2013 International Hepato-Pancreato-Biliary Association

120

may have the potential to contribute to selection for hepatectomy in some patients.6–8 For patients with multiple bilateral CRLM that cannot be safely or completely resected in a single procedure, two-stage hepatectomy is a technical approach that has been shown to accomplish the complete resection of disease that is otherwise considered unresectable to facilitate longterm survival comparable with that in patients who undergo initial resection.6 These favourable results are likely, in large part, to reflect selection because all patients who completed two-stage hepatectomy demonstrated a radiographic and significant histologic response to chemotherapy. It should be emphasized that an intention-totreat analysis which included patients who did not complete all stages of therapy showed impressive 5-year overall survival superior to that of matched patients who did respond to chemotherapy but did not undergo surgery, further emphasizing the potential benefit of liver resection in this setting.6 This approach typically entails initial treatment with systemic chemotherapy with or without a biologic agent, with reassessment after four to six chemotherapy cycles. The first-stage resection is typically performed when cross-sectional imaging demonstrates response to treatment or stable disease. Most commonly, first-stage hepatectomy includes the complete resection of all metastases from the future liver remnant (FLR) in the form of minor resections that avoid hepatic pedicle dissection or mobilization of the liver to be resected in the second stage.9 Portal vein embolization (PVE) may be required if the FLR volume is insufficient. After 4–6 weeks (without interval chemotherapy), or later (with chemotherapy); re-imaging is performed to assess liver regeneration and second-stage major hepatectomy may be performed, possibly followed by additional chemotherapy.6 Neither cytotoxic agents nor bevacizumab impair liver regeneration after PVE.10,11 In a study by Mentha et al. evaluating patients undergoing two-stage hepatectomy, mostly without interval chemotherapy, histologic examination demonstrated the increased proliferation of tumour cells at the periphery of lesions after second-stage resection.12 Some investigators therefore recommend interval chemotherapy, whereas others do not in order to mitigate chemotherapy-associated hepatotoxicity.6,9 If chemotherapy is selected, PVE may also be performed during this time without fear of reduced efficacy. The in situ split procedure is a recently reported, novel method for inducing rapid remnant liver hypertrophy by ligating and dividing the portal vein, and splitting the liver parenchyma along the intended transection line; a second procedure, to be conducted during the same hospitalization, is required to remove the specimen once hypertrophy of the liver remnant has occurred. Recently, Schnitzbauer et al. described this procedure in 25 patients with various malignancies, including 14 with CRLM.13 Although this technique may have a role in selected patients, enthusiasm must be tempered by the high perioperative mortality rate of 12% (mainly related to liver failure and infection) and absence of large studies or longterm results.

HPB 2013, 15, 119–130

HPB

One-stage hepatectomy with multiple parenchyma-sparing resections is widely practised and, when technically feasible, can be considered a standard approach.14 One-stage hepatectomy requires high-quality preoperative imaging and mastery of intraoperative ultrasound and segmental hepatic resections.15 Advantages of one-stage resection include low published morbidity and mortality rates and the avoidance of a second laparotomy.14 The selection of a one- or two-stage approach in a patient with bilateral CRLM is challenging and should be individualized based on disease burden, potential for positive margins, the anatomic distribution of disease and the nature of any extrahepatic resections that will be performed. A greater disease burden targeted using the multiple-wedge approach carries the risk for narrow resection margins and higher recurrence rates. In a report by Kokudo et al., 57% of patients with four or more metastases preferentially treated with one-stage multiple sub-segmentectomies developed recurrence in the remnant liver.16 However, many patients were amenable to repeat resection and, in patients with five or more metastases resected in one to three hepatectomies, 5-year overall survival after the first hepatectomy was 45%. The role of ablation as an adjunct to resection in patients with bilateral CRLM remains controversial.17 Studies combining ablation with resection usually reflect outcomes in patients with more advanced intrahepatic disease burdens and profiles that many would consider unresectable, and show 3-year overall survival rates of 47-51%.17–19 A multi-institution study of 125 patients who underwent resection with ablation demonstrated median disease-free and overall survival of 14.7 months and 34.8 months, respectively, which were shorter than those of patients undergoing resection alone (median overall survival: 50.5 months).19 However, the two groups were not comparable because patients in the former were more frequently associated with adverse prognostic factors, including synchronous multinodular bilateral disease. Intraoperative ablation combined with resection may expand the number of patients eligible for liver-directed therapy because normal liver parenchyma can be spared, allowing for contralateral major resections. Selection for the use of intraoperative ablation is critical and should be limited to small (ⱕ3 cm, preferably ⱕ1 cm) tumours away from major vessels.20 The short-term benefits of ablation should thus be balanced with the high longterm risk for recurrence. When the disease is refractory to contemporary chemotherapeutic agents, intra-arterial therapies may downsize metastases sufficiently to allow resection in some patients.21 However, data are limited on the safety of hepatectomy after yttrium-90 radioembolization, which may induce portal hypertension, parenchymal fibrosis and liver atrophy. Chemotherapy delivered by hepatic arterial infusion (HAI) has also been used successfully to convert unresectable to resectable disease.22,23 Very advanced disease is the norm in these patients, who often require ablation combined with resection, and in whom disease recurrence rates are as high as 90%.22 In patients in whom bilateral CRLM is sufficiently downsized to allow operative therapy after systemic or intra-arterial therapies,

© 2013 International Hepato-Pancreato-Biliary Association

HPB

resection should encompass all initial sites of disease if possible. ‘Adjuvant hepatectomy’ or incomplete removal of just some CRLM should be avoided because the currently available data do not demonstrate any oncologic benefit of such an approach. Consensus statements 1 Resection is the reference standard treatment for bilateral CRLM. 2 One-stage surgery with multiple resections of multiple bilobar lesions is an effective strategy with which to treat appropriately selected patients with bilateral metastases. 3 Two-stage hepatectomy with perioperative chemotherapy ⫾ PVE (or portal vein ligation) may allow the complete resection of disease in patients who are otherwise considered unresectable. 4 Ablation may be a useful adjunct for managing extensive bilateral CRLM not amenable to complete resection, but proper selection is critical and higher local recurrence rates must be recognized. 5 After downsizing with preoperative systemic or intra-arterial therapy, the goal of a curative-intent hepatectomy remains to remove all initial and currently known sites of disease if possible. Incomplete removal of CRLM is not recommended.

Approaches to the synchronous presentation of colorectal cancer and CRLM Of the 50% of patients with colorectal cancer (CRC) who develop liver metastases,24 a large proportion present with synchronous disease.4,25–29 Many studies have found that synchronous CRLM is associated with worse survival than metachronous disease,1,25,30–33 although others have shown no difference.27,34,35 Patients with synchronous CRLM often have advanced disease and decisions on their management can be complex, mandating a multidisciplinary approach that includes the expertise of colorectal and hepatic surgeons and medical oncologists. Because the timing of chemotherapy, hepatic resection and colorectal resection must be carefully planned and sequenced, it is imperative that all relevant specialties participate in clinical decision making from the beginning. Role and timing of colon and liver resection One of the main considerations in the management of patients with synchronous CRLM concerns the roles and timings of resections of the primary tumour and liver metastases. Resection of the primary tumour does not improve survival in patients in whom liver metastases are not resected.36–39 Therefore, among patients with unresectable CRLM, primary tumour resection is reserved for patients with symptoms that cannot be controlled with less invasive techniques such as endoscopic stenting. The reported 5-year survival rates of 40–58% following CRLM resection have made resection the standard of care in patients who can be rendered disease-free.27,40–42 For patients with initially unresectable CRLM, systemic chemotherapy is the primary therapy, but the

HPB 2013, 15, 119–130

121

possibility of reducing the tumour burden sufficiently to allow resection should always be considered. Adam et al.43 demonstrated that treatment with FOLFOX [folinic acid (leucovorin), fluorouracil (5-FU), oxaliplatin] or FOLFIRI [folinic acid (leucovorin), 5-FU, irinotecan] converted 12.5% of patients with unresectable CRLM to resectable status. In selected patients, higher rates of conversion to resectable intrahepatic disease have been reported after HAI treatment combined with systemic chemotherapy than after systemic chemotherapy alone.22,23 For patients with a resectable primary tumour and resectable synchronous CRLM, there are three options in regard to sequencing the operative components: (i) staged resection with colon resection first; (ii) staged resection with liver resection first, and (iii) simultaneous resection of the primary tumour and CRLM. Regardless of the sequencing of the procedures, in all patients it is critically important that an adequate oncologic resection at both sites (liver and primary disease) be performed. Historically, simultaneous resection was associated with higher rates of morbidity and mortality and therefore was not advocated.26,31 As a result of continued improvements in anaesthetic and surgical management, mortality in even major hepatic resections has decreased to very low rates, prompting many centres to revisit the safety of simultaneous resection. Thelen and colleagues44 reported higher operative mortality after simultaneous (10%) versus staged (1.1%) resections; all of the lethal events in the former group occurred after major hepatic resections. By contrast, several recent series have reported equal rates of mortality in simultaneous and staged resections. Yan et al.45 compared outcomes in 73 patients undergoing simultaneous resection with those in 30 patients undergoing staged resection and found mortality rates of 0% in both groups, although about three quarters of patients in both groups were submitted to major hepatic resection. Martin and colleagues46 reported results in 230 patients in a contemporary series who underwent resection for synchronous CRLM (70 simultaneous, 160 staged resections), in whom the decision on type of resection was based upon the complexity of the two resections and the overall comorbidity of the patient. The two groups were equivalent in age and comorbid conditions, lesion numbers and sizes, and extent of liver resection. The mortality rate was 2% in both groups and complication rates (56% versus 55%) and the severity of complications were equivalent in both groups. The simultaneous resection group had a significantly shorter length of stay (LoS) in hospital compared with the combined lengths of both hospitalizations in the staged resection group (LoS: 10 days versus 18 days; P = 0.001). Furthermore, Turrini et al. reported that patients submitted to simultaneous resection were able to start chemotherapy sooner after surgery and were more likely to complete the treatment course.47 In a recent meta-analysis of 14 published reports on 2204 patients, Chen et al.48 showed that simultaneous resection was associated with a shorter hospital LoS [weighted mean difference -4.77 days, 95% confidence interval (CI) -7.26 to -2.28; P < 0.01] and lower morbidity (odds ratio 0.71, 95% CI 0.57–0.88; P = 0.002) compared

© 2013 International Hepato-Pancreato-Biliary Association

HPB

122

with staged resections, with no difference in longterm survival. A recent non-randomized study comparing simultaneous versus staged resection in synchronous CRLM demonstrated equivalent (or lower) morbidity, equivalent mortality rates, and shorter hospital LoS for simultaneous resection.49 Unfortunately, the studies addressing this particular issue are limited by a high degree of selection bias arising from the fact that patients in whom complication rates were expected to be higher (as a result of greater comorbidity, marginal FLR volume, etc.) are generally selected for staged resections. The most that can be concluded from this body of work is that, with careful patient selection and selective use of PVE, simultaneous colon and liver resections appear to result in a shorter LoS and equivalent morbidity and mortality rates compared with staged resections. It appears to be safe to combine a straightforward colon resection (segmental or hemi-colectomy) with a hepatectomy in which up to four segments are removed, or minor hepatectomy with any colorectal resection; in combinations of extended intestinal resections (i.e. rectal resections, total abdominal colectomy) with major hepatectomy, morbidity risks are significantly greater.50 From a technical perspective, in patients undergoing simultaneous resection, liver resection is generally performed first under low central venous pressure conditions, followed by the colon resection. In patients who are deemed unsuitable for synchronous resection, the ‘colorectal-first’ approach has historically been favoured because of the perceived risks for bleeding, obstruction, perforation or progression of the primary tumour.44,45,47,48,51 In patients with haemorrhage or obstruction, resection of the primary disease should, in nearly all cases, be performed without considering a simultaneous resection. Likewise, patients who present with a perforation require urgent surgical intervention, should almost never undergo simultaneous treatment of the liver disease, and are at significant risk for peritoneal carcinomatosis that will mitigate any anticipated benefit of CRLM resection. An alternative to the synchronous approach and classic ‘colorectal-first’ approach is the ‘liver-first’ approach (or ‘reverse’ approach), initially described by Mentha et al.52 This approach is useful in patients with a rectal primary tumour, in whom possible treatments can be sequenced as follows: systemic chemotherapy first, followed by liver resection, followed by chemoradiation to the rectal primary tumour, followed by resection of the rectal primary tumour. The radiation component may be individualized as the resulting reduction in risk for local recurrence may be considerably smaller than the risk for distant recurrence.53 The benefit of this approach is that treatment of the metastatic disease is not delayed by chemoradiation.52 The ‘liver-first’ approach may also be useful in advanced CRLM that may progress to become unresectable while off chemotherapy, although this possibility should not drive clinical decision making in most patients. Brouquet et al. analysed outcomes in 156 patients with synchronous CRLM who underwent resection via ‘combined’, ‘colorectal-first’ or ‘liver-first’ approaches49 and found that the last of these was associated with morbidity, mortality and overall survival rates

HPB 2013, 15, 119–130

similar to those of the other strategies. A total of 5% of patients in the ‘liver-first’ group developed primary tumour-related complications; all of these patients had demonstrated nearly obstructing primary tumours prior to chemotherapy. This study showed that outcomes in patients with colon and rectal primary tumours were similar regardless of treatment sequence. The ‘simultaneous’, ‘colorectal-first’ and ‘liver-first’ approaches all have roles in the treatment of patients with synchronous CRLM. The particular approach used must be tailored to the patient through multidisciplinary discussion that prioritizes therapy for the most pressing disease component, minimizes the risk for complications and expedites the use of systemic therapy. Consensus statements 1 Patients with synchronous CRLM should undergo formal multidisciplinary evaluation by colorectal and hepatic surgeons and medical oncologists prior to the initiation of any treatment. 2 In many patients with synchronous CRLM, simultaneous resection by experienced surgeons is feasible, safe and effective; however, staged resection should be considered for patients with a marginal or inadequate FLR volume, those with significant morbidity-related risks, or when extensive operations are anticipated at both sites. 3 Priority in staged resections may be given to colorectal-first or liver-first strategies based on concern for complications related to the primary tumour or the progression of marginally resectable CRLM during treatment of the primary tumour. Role and timing of chemotherapy in synchronous CRLM A more detailed discussion of the timing of chemotherapy and liver resection (preoperative/neoadjuvant, perioperative, postoperative/adjuvant) is included in the accompanying third Expert Consensus Statement presented by Schwarz et al.54 in this issue of HPB. However, several issues relevant to the treatment of patients with synchronous CRLM warrant specific discussion. The concept of selection for resection based on response to chemotherapy has gained popularity in this group of patients with CRLM. In the great majority of patients, tumours either decrease in size or remain stable following treatment with systemic chemotherapy, a finding reported in virtually every prospective study completed using modern chemotherapy.55–58 Response to chemotherapy is associated with a lower incidence of non-therapeutic laparotomy in the prospective European Organization for Research and Treatment of Cancer (EORTC) study (95% resected in the chemotherapy arm versus 89% resected in the nochemotherapy arm),55 and in multivariate analyses of retrospective data.59 Conversely, in patients with extensive disease (e.g. four or more CRLM), progression on chemotherapy is associated with shorter longterm survival compared with patients with stable or responding disease (5-year overall survival: 37% in responders, 30% in patients with stable disease, 8% in patients

© 2013 International Hepato-Pancreato-Biliary Association

HPB

with progression).43 Progression, however, is rare (7% in the EORTC trial) and progression that precludes resection is even less common (4%)55 and thus few patients will be denied resection based on response to chemotherapy. If chemotherapy is to be given preoperatively, a short course (four to six cycles) is recommended in view of considerations of hepatotoxicity.60 Chemotherapy can be safely delivered to patients with intact primary tumours.49,61 A series of 233 patients [of whom 78 (33%) had rectal primary lesions] with stage IV CRC and intact primary tumours treated by systemic chemotherapy revealed that 11% developed primary tumour-related complications and 3.5% required a bypass or stoma as a result (other primary disease-related complications were managed with resection, stent or radiation).62 When symptoms are minimal or absent, and when the colonoscope can traverse the lesion, systemic chemotherapy for synchronous CRLM and untreated primary tumours can be safely delivered.49,61 Given that the majority of patients respond to chemotherapy (in both the primary tumour and CRLM), consideration must be given to marking lesions that might be difficult to locate after a major chemotherapy response. India ink tattoo of small primary tumours is recommended to facilitate localization at the time of resection as some lesions will nearly disappear following shortcourse treatment. Similarly, small non-subcapsular liver metastases that would otherwise be amenable to limited liver resection should be considered for marking (e.g. with a metallic coil by the interventional radiologist) in case they disappear or become difficult to localize after systemic therapy.63 For primary tumours that present with minimal bleeding or demonstrate significant mucosal erosion, treatment with bevacizumab is not generally recommended for the first cycle(s) of therapy in view of the risks for bleeding and perforation.64 Development of colonic obstruction in the setting of chemotherapy-associated neutropenia is associated with significant risk for morbidity and mortality, and thus resection of the primary tumour prior to chemotherapy should be strongly considered in patients with impending obstruction.49 In conclusion, data from several studies indicate that the administration of up to six cycles of neoadjuvant FOLFOX is a reasonable strategy for use in patients with resectable CRLM. Other data support the safety of chemotherapy in the presence of an intact primary tumour. A key to this approach involves limiting pre-resection chemotherapy to 2–3 months to prevent the occurrence of chemotherapy-induced liver injury. Patients at intermediate or high risk for recurrence (based on the clinical risk score)65 may benefit from preoperative chemotherapy. Consensus statement 4 In patients with synchronous CRLM and asymptomatic primary tumours, preoperative chemotherapy (of 2–3 months’ duration) is relatively safe and should be considered in patients at high risk for recurrent disease.

HPB 2013, 15, 119–130

123

Intra-arterial therapies Rationale for hepatic arterial therapy The liver is the sole site of metastatic disease in many patients with CRC. Normal liver parenchyma is perfused predominantly by portal venous blood, whereas metastatic tumours of >1 mm in size are supplied primarily by the hepatic arterial system.66,67 The administration of therapy via the hepatic arterial system enhances drug delivery to tumour tissue and substantially reduces the occurrence of systemic side-effects.66,67 Hepatic arterial infusion therapy Hepatic arterial infusion chemotherapy is administered either through surgically placed ports or through an implantable continuous infusion pump, both of which provide more durable delivery than percutaneous catheters.68 Arterial infusion catheters are best inserted into the gastroduodenal artery.69 Hepatic arterial infusion pumps are best placed at centres with considerable experience in this technique by surgeons who have placed at least 25 pumps.70 In this circumstance, complications can be minimized, and pump function preservation can be as high as 91% at 1 year and 84% at 2 years.70 Floxuridine is an antimetabolite chemotherapy agent most commonly used in HAI therapy and is the most thoroughly studied drug for this purpose. It has a half-life of <10 min and 95% of drug is extracted on its first pass through the liver.71 Radioactive HAI floxuridine distributes a 15-fold higher concentration of chemotherapy to the tumour than to normal liver parenchyma and a 400-fold higher concentration of chemotherapy to the tumour than is observed with systemic administration.72 5-Fluorouracil (5-FU) has been the preferred agent in Europe (usually delivered via surgically placed ports)73,74 and oxaliplatin has been used more recently in HAI therapy with favourable results.22,75,76 Systemic toxicity is low with HAI-administered floxuridine. Aberrant gut perfusion can result in ulceration and diarrhoea.77 Biliary toxicity is the most common treatment-related complication of floxuridine-based HAI chemotherapy and occurs because the biliary system is also perfused by the hepatic artery.78 Elevated liver enzymes warrant dose modification.79 Simultaneous administration of dexamethasone reduces the biliary toxicity of floxuridine-based HAI.80 The addition of concomitant systemic chemotherapy can increase the toxicity of HAI pump therapy.81 Efficacy in unresectable disease Hepatic arterial infusion chemotherapy has been extensively studied as first-line treatment for unresectable CRLM in at least 10 randomized studies.82 Although response rates to HAI are higher than those in non-HAI-treated control subjects, improvements in survival have been inconsistent, with the best being a 4.4-month advantage in patients treated with floxuridinebased HAI compared with those receiving systemic 5-FU and

© 2013 International Hepato-Pancreato-Biliary Association

HPB

124

leucovorin in a well-performed trial conducted by the Cancer and Leukaemia Group B (CALGB) reported in 2006.83 In this trial, quality of life in HAI recipients was superior to that in those treated with systemic therapy only. In the HAI-treated group, median time to hepatic progression was longer than in the non-HAI group (9.8 months versus 7.3 months), whereas time to extrahepatic progression was shorter (7.7 months versus 14.8 months). It is important to note that floxuridine-based HAI was administered without systemic therapy in this trial, which is not the current standard approach. An EORTC trial comparing outcomes in 290 patients on either systemic therapy or HAI-administered 5-FU and leucovorin showed no difference in median survival between the groups; 37% of patients in the HAI group received no regional therapy, underscoring the complexity of this modality.74 A meta-analysis of 10 selected randomized trials demonstrated a greater tumour response to HAI therapy than systemic therapy alone (43% versus 18%; P < 0.0001), but no significant overall survival difference (15.9 months versus 12.4 months; P = 0.24).82 There may be an advantage to combining HAI with systemic chemotherapy in patients in whom first-line systemic chemotherapy alone has failed84 and in a preoperative approach in patients with initially unresectable disease.23 In the latter scenario, response rates as high as 92% have been reported.23 A recent French Phase II trial of 26 oxaliplatin-naïve patients with CRLM treated with infusional port oxaliplatin and systemic 5-FU/leucovorin showed median overall and disease-free survival of 27 months each.75

Adjuvant approach Postoperative floxuridine-based HAI (with systemic 5-FU/ leucovorin) following a negative margin hepatic resection has been compared with i.v. 5-FU/leucovorin alone in a randomized controlled trial of 156 patients.85 At 2 years, the HAI group had improved overall survival (86% versus 72%; P = 0.03), but the 10-year follow-up failed to show a sustained, statistically significant survival advantage, despite significantly longer progressionfree survival (31.3 months versus 17.2 months; P = 0.02) and greater overall survival (41% versus 27%; P = 0.10) in the HAItreated group.86 No other randomized study to date has shown HAI chemotherapy to result in a significant survival advantage in the adjuvant setting. Although limited by the heterogeneity of the studies it examined, a 2009 Cochrane review of seven randomized trials of adjuvant HAI of floxuridine or 5-FU therapy for CRLM failed to show a survival benefit for this therapy over adjuvant systemic therapy alone.87 A recent Phase II trial conducted through the National Surgical Adjuvant Breast and Bowel Project (NSABP) investigated the combination of floxuridine-based HAI and systemic oxaliplatin and capecitabine in the adjuvant setting and reported a 2-year survival rate of 85%.88 Studies assessing the role of HAI chemotherapy in association with modern systemic chemotherapy regimens and different infusional agents are ongoing.

HPB 2013, 15, 119–130

Consensus statements 1 Hepatic arterial infusion therapy is a promising option for the provision of palliative or adjuvant therapy in patients with CRLM, but is best administered in experienced centres. 2 In patients with unresectable CRLM, floxuridine-based HAI therapy significantly improves tumour response and hepatic progression-free survival, but the survival advantage of floxuridine-based HAI over contemporary systemic chemotherapy remains questionable. 3 Randomized controlled trials are warranted to examine the potential advantages of and indications for HAI therapy with updated systemic chemotherapy regimens in patients with resectable and unresectable disease. Radioembolization Radiotherapy can induce cytotoxicity in malignant liver tumours; however, the value of standard external beam radiation applied to hepatic malignancies is limited by toxicity to surrounding hepatic parenchyma. Radioembolization is a form of brachytherapy in which a high-energy beta-emitting radiation source (yttrium-90) is coupled with an appropriately sized (~30 mm diameter) embolic particle. This technique allows the delivery of a therapeutic dose to the tumour vasculature and attempts to minimize damage to uninvolved tissue.89 SIR-Spheres® (Sirtex Medical Pty Ltd, Sydney, NSW, Australia) are resin-based microspheres approved by the US Food and Drug Administration (FDA) for use in the treatment of CRLM with adjuvant intra-arterial floxuridine. The basis of FDA approval was a Phase III trial of floxuridine-based HAI administered with and without resin microspheres in 72 patients with unresectable CRLM.90 The combination therapy was associated with an improved time to progression (15.9 months versus 9.7 months; P < 0.01) over floxuridine-based HAI alone. A subsequent multicentre Phase II trial of yttrium-90 resin microsphere radioembolization was performed in 50 patients refractory to chemotherapy in whom previous oxaliplatin- and irinotecan-based systemic regimens had failed.91 Response or stable disease was observed in 48% of patients treated and median overall survival was 12.6 months in these heavily pretreated patients. In this trial yttrium-90 treatment was well tolerated, although longterm hepatic toxicity was not evaluated. An ongoing Phase III trial, SIRFLOX [NCT00724503 (http://www.clinicaltrials.gov)], is evaluating yttrium-90 intra-arterial treatment as first-line therapy for unresectable CRLM treated with FOLFOX6 and SIR-Spheres® versus FOLFOX6 alone, and is expected to complete in 2012. TheraSpheres® (MDS Nordion, Ottawa, ON, Canada) are glassbased microspheres that have a humanitarian device exemption for the treatment of primary and metastatic liver tumours. There are limited data on the role of this agent in CRLM. A Phase II trial of TheraSpheres® as salvage therapy in patients in whom systemic chemotherapy had failed was performed in 27 patients, with a positron emission tomography response measured at 88% in the first treated hepatic lobe.92 A randomized Phase III multicentre

© 2013 International Hepato-Pancreato-Biliary Association

HPB

clinical trial, known as the ‘Efficacy Evaluation of TheraSphere following Failed First-Line Chemotherapy in Metastatic Colorectal Cancer’ (EPOCH) trial, will open soon at up to 30 sites worldwide, with a target enrolment of approximately 350 patients (http://clinicaltrials.gov/ct2/show/NCT01483027). Consensus statement 4 Yttrium-90 transarterial radioembolization is effective and has shown promising early results in the palliative management of unresectable CRLM, and can be considered for patients with liver-only disease in whom at least one standard line of systemic chemotherapy fails. First-line therapy with yttrium-90 should only be administered in clinical trials; longterm toxicity potential requires further clinical evaluation. Chemoembolization The approaches to and reports on traditional transcatheter arterial chemoembolization (TACE) in CRLM are ill-defined and variable. As such, this discussion will focus on drug-eluting beads delivering irinotecan (DEBIRI; Biocompatibles UK Ltd, Farnham, UK), data for which are emerging from ongoing clinical trials.93,94 These are polyvinyl alcohol hydrogel beads loaded with irinotecan.95 The beads are delivered via a selective transcatheter arterial administration and lodge in the interface between the tumour and normal parenchyma. The drug is slowly released into the tumour environment, providing relatively selective drug exposure. Irinotecan was selected because of its chemical properties, charge and ability to be loaded on the beads. An initial experience in 20 patients demonstrated a reduction of >50% in serum carcinoembryonic antigen levels in 12 patients and reduced lesion enhancement on contrasted axial imaging in 80% of patients.96 Right upper quadrant pain has been the main effect of toxicity.97 Subsequently, a multicentre registry of 55 heavily pretreated patients reported rates of hepatic progression-free survival and overall survival at 1 year of 75% for each.98 A Phase II randomized trial of 60 patients comparing treatment with FOLFOX ⫾ bevacizumab plus DEBIRI, with treatment with FOLFOX ⫾ bevacizumab (http://clinicaltrials.gov/ct2/show/NCT00932438) is currently accruing. Although it is too early to speculate on the benefits of this therapy, it appears to be well tolerated and offers the potential for a standardized mode of chemoembolization for CRLM. Consensus statement 5 Traditional TACE lacks data and standardization as a modality for treatment of CRLM. DEBIRI is a newer form of TACE, which shows promising early results, but requires further study before its use can be recommended.

Ablation strategies including radiofrequency ablation, microwave ablation and external beam radiotherapy Hepatic resection is associated with 5-year survival rates of up to 58% and cure rates of approximately 20%.27,40,99 Results of this

HPB 2013, 15, 119–130

125

level are not reported with any other therapy. Local failure at the resection margin is uncommon, with most large series reporting rates of significantly <5%.42 Although rates of morbidity in partial hepatectomy for CRLM have remained stable at 30–40%, 90-day mortality has diminished substantially to 1% in high-volume centres.100,101 Furthermore, survival appears to be improving.100 Collectively, these data have established hepatic resection as the reference standard locoregional treatment for CRLM. The thermal ablation of tumours utilizes image guidance to deliver extreme temperatures to a tumour and its surrounding tissue. The advantages of thermal ablation include its adaptability to minimally invasive approaches, the ability to spare liver parenchyma and a low morbidity rate. Thermal ablation can be performed percutaneously, laparoscopically or at laparotomy. At 60 °C there is immediate cell death; typically, ablation zones are created with temperatures in excess of this threshold. Radiofrequency ablation (RFA) is the most commonly used form of thermal ablation in the treatment of liver tumours. In RFA, needles placed in and around tumours deliver alternating electrical current in the radiofrequency range that generates heat. At high temperatures, tissue desiccates, limiting current flow and impairing further ablation.102–104 In properly selected patients, RFA-associated morbidity is generally <10%, regardless of whether RFA is performed surgically or percutaneously.105 Mortality in RFA is <1%.105 These advantages are balanced by a number of limitations to the use of RFA, which is generally ineffective in tumours of >3 cm in size because of high local recurrence rates.17,102–104 The heat generated by RFA can injure adjacent structures (e.g. the diaphragm, gallbladder, gastrointestinal tract, major bile ducts) and RFA should not be used to treat tumours adjacent to these locations. Lastly, the efficacy of RFA is limited by the presence of large blood vessels as a result of their high flow and the related ‘heat sink’ effect, which protects adjacent cells from thermal ablation.102–104 It is therefore important to note that unresectable tumours are not generally amenable to thermal ablation because the factors that make a tumour unresectable (number, size, proximity to major blood vessels or bile ducts, etc.) also make it unsuitable for ablation. Local recurrence after RFA remains a major problem. Factors associated with higher local recurrence rates include large tumour size, proximity to major blood vessels, diagnosis (CRLM being worse) and percutaneous approach.20,102–104 In a large metaanalysis, tumour size and the percutaneous approach were independently associated with higher recurrence rates. Recurrence rates in tumours of >3 cm in size ranged from 25% to 50% regardless of approach. For tumours of <3 cm in size, local recurrence rates in surgical and percutaneous ablation were 4% and 16%, respectively.20 In single-arm percutaneous RFA series, local recurrence rates have ranged from 18% to 50% and are related to tumour size.20,102–105 Recurrence rates in open RFA have typically been <15% with some series reporting rates of <10%.20,102–105 The best local control rates reported for RFA (and other ablation techniques) appear to be achieved at open operation, probably

© 2013 International Hepato-Pancreato-Biliary Association

HPB

126

because this allows for the mobilization of the liver and the use of direct hepatic intraoperative ultrasound to guide probe placement. Intraoperative ablation is also limited by tumour size, as indicated by recurrence rates of ⱕ20% for tumours of 1–3 cm in size and ⱕ50% for tumours of >3 cm in size.106 The utility and limitations of RFA combined with resection are discussed in the section entitled ‘Approaches to bilateral colorectal cancer liver metastases’. Comparisons between RFA and resection have generally been limited to retrospective series that have attempted to match patients. These retrospective comparative studies have shown substantially higher local recurrence rates for RFA (16–60% versus 0–24%) and better longterm survival in resected patients.102–104 Unfortunately, these comparisons are hampered by tremendous selection bias and confounding factors, and conclusions about survival rates are not possible. One recently published study proposed percutaneous RFA as a ‘test of time’ to manage selected patients with resectable CRLM. In this study, 134 tumours were treated in 88 patients with limited tumour burdens and >40% of RFA-treated tumours recurred locally. Resection was not performed in two thirds of patients because they either remained free of disease (many with repeat RFA) or had unresectable distant progression.107 In this study, the avoidance of a surgical procedure was considered the goal of RFA, but whether this should be the desired outcome is disputable. In this series, RFA failed in >40% of cases and it was unclear whether patients could be effectively salvaged. Therefore, this approach has not been uniformly accepted. Another recent concept refers to the use of laparoscopic RFA in patients with resectable CRLM. A recent example of such an approach was illustrated in a study of 64 patients with resectable metastases who underwent laparoscopic RFA. Local failure occurred in 7% of patients and median recurrence-free survival was 15 months. Median and 5-year actuarial survival were 4.3 years and 49%, respectively, which may be considered comparable with outcomes in resection series.108 Laparoscopic RFA in wellselected patients may be a promising strategy, but has not been sufficiently studied. Microwave ablation (MWA) is a newer technology that utilizes high-frequency electromagnetic radiation to create thermal damage and coagulation necrosis. This form of ablation carries less risk for charring and a heat sink effect and creates larger ablation zones more rapidly. As MWA is a more powerful energy mediator, there are concerns over injury to adjacent structures. The use of MWA to treat CRLM has not been well studied. Most series reflect outcomes for a large variety of liver tumours. In sum, these studies show complication rates ranging from 6% to 30%, with the higher rates occurring in patients undergoing laparotomy and other procedures.102,109,110 Local recurrence rates vary hugely from 3% to 50%, but the two largest series reported low rates of 3% and 6%, respectively.109,110 Published MWA series report survival rates, but the heterogeneity of diseases treated in each series precludes the clarification of any conclusions regarding survival data.102,109,110

HPB 2013, 15, 119–130

External beam radiation therapy (EBRT) has recently been proposed as a potential locoregional therapy in CRLM and other liver tumours. Historically, EBRT was not used in the liver because the therapeutic window between its tumoricidal benefits and liverspecific toxicity was too small. Recent advances in imaging, stereotactic body radiation (SBRT) and corrections for respiratory motion have increased the potential role of EBRT in the treatment of liver tumours. Furthermore, although it has not been well studied, EBRT may be effective and safe when used near major blood vessels; however, the potential for major bile duct toxicity remains a concern. Overall, the use of SBRT for liver tumours has been minimally studied and only a few studies have focused on its use in CRLM. A few well-designed dose-escalation studies have demonstrated that SBRT can be delivered to liver tumours without dose-limiting toxicity in doses of ⱕ60 Gy. Unfortunately, local progression rates at 2 years at the initial lower doses have been high, ranging from 23% to 43%;111–114 higher doses may afford better local control rates. In a recent dose-escalation study, nine patients with malignant liver tumours demonstrated a 90% response rate and a 2-year local control rate of 100% at a dose of 60 Gy in five fractions.114 Consensus statements 1 Ablation strategies are inadequately studied and plagued by high local failure rates, and are limited by tumour size, tumour multiplicity and location, and thus are not recommended as first-line treatments for resectable CRLM. 2 Ablation strategies play a role in highly selected patients with small, appropriately located tumours when resection is not feasible or safe, but should be considered as second-line locoregional therapy to hepatic resection. 3 Prospective trials comparing ablative techniques or comparing resection with ablation in well-defined patients are needed to define the role of ablation in the treatment of CRLM in the future. Conflicts of interest None declared.

References 1. Scheele J, Stang R, Altendorf-Hofmann A, Paul M. (1995) Resection of colorectal liver metastases. World J Surg 19:59–71. 2. Gayowski TJ, Iwatsuki S, Madariaga JR, Selby R, Todo S, Irish W et al. (1994) Experience in hepatic resection for metastatic colorectal cancer: analysis of clinical and pathologic risk factors. Surgery 116:703–710; discussion 710–711. 3. Ekberg H, Tranberg KG, Andersson R, Lundstedt C, Hagerstrand I, Ranstam J et al. (1986) Determinants of survival in liver resection for colorectal secondaries. Br J Surg 73:727–731. 4. Minagawa M, Makuuchi M, Torzilli G, Takayama T, Kawasaki S, Kosuge T et al. (2000) Extension of the frontiers of surgical indications in the treatment of liver metastases from colorectal cancer: longterm results. Ann Surg 231:487–499.

© 2013 International Hepato-Pancreato-Biliary Association

HPB

5. Charnsangavej C, Clary B, Fong Y, Grothey A, Pawlik TM, Choti

127

20. Mulier S, Ni Y, Jamart J, Ruers T, Marchal G, Michel L. (2005) Local

MA. (2006) Selection of patients for resection of hepatic colorectal

recurrence after hepatic radiofrequency coagulation: multivariate meta-

metastases: expert consensus statement. Ann Surg Oncol 13:1261–

analysis and review of contributing factors. Ann Surg 242:158–171.

1268.

21. Sharma RA, Van Hazel GA, Morgan B, Berry DP, Blanshard K, Price D

6. Brouquet A, Abdalla EK, Kopetz S, Garrett CR, Overman MJ, Eng C

et al. (2007) Radioembolization of liver metastases from colorectal

et al. (2011) High survival rate after two-stage resection of advanced

cancer using yttrium-90 microspheres with concomitant systemic

colorectal liver metastases: response-based selection and complete

oxaliplatin, fluorouracil, and leucovorin chemotherapy. J Clin Oncol

resection define outcome. J Clin Oncol 29:1083–1090.

25:1099–1106.

7. Blazer DG 3rd, Kishi Y, Maru DM, Kopetz S, Chun YS, Overman MJ

22. Goere D, Deshaies I, de Baere T, Boige V, Malka D, Dumont F et al.

et al. (2008) Pathologic response to preoperative chemotherapy: a new

(2010) Prolonged survival of initially unresectable hepatic colorectal

outcome endpoint after resection of hepatic colorectal metastases.

cancer patients treated with hepatic arterial infusion of oxaliplatin

J Clin Oncol 26:5344–5351.

followed by radical surgery of metastases. Ann Surg 251:686–691.

8. Adam R, Wicherts DA, de Haas RJ, Ciacio O, Levi F, Paule B et al.

23. Kemeny NE, Melendez FD, Capanu M, Paty PB, Fong Y, Schwartz LH

(2009) Patients with initially unresectable colorectal liver metastases: is

et al. (2009) Conversion to resectability using hepatic artery infusion

there a possibility of cure? J Clin Oncol 27:1829–1835.

plus systemic chemotherapy for the treatment of unresectable

9. Adam R, Miller R, Pitombo M, Wicherts DA, de Haas RJ, Bitsakou G et al. (2007) Two-stage hepatectomy approach for initially unresectable colorectal hepatic metastases. Surg Oncol Clin N Am 16:525–536. 10. Zorzi D, Chun YS, Madoff DC, Abdalla EK, Vauthey JN. (2008) Chemo-

liver metastases from colorectal carcinoma. J Clin Oncol 27:3465– 3471. 24. Steele G Jr, Ravikumar TS. (1989) Resection of hepatic metastases from colorectal cancer. Biologic perspective. Ann Surg 210:127–138.

therapy with bevacizumab does not affect liver regeneration after portal

25. Fong Y, Fortner J, Sun RL, Brennan MF, Blumgart LH. (1999) Clinical

vein embolization in the treatment of colorectal liver metastases. Ann

score for predicting recurrence after hepatic resection for metastatic

Surg Oncol 15:2765–2772.

colorectal cancer: analysis of 1001 consecutive cases. Ann Surg

11. Goere D, Farges O, Leporrier J, Sauvanet A, Vilgrain V, Belghiti J. (2006) Chemotherapy does not impair hypertrophy of the left liver after right portal vein obstruction. J Gastrointest Surg 10:365–370. 12. Mentha G, Terraz S, Morel P, Andres A, Giostra E, Roth A et al. (2009) Dangerous halo after neoadjuvant chemotherapy and two-step hepatectomy for colorectal liver metastases. Br J Surg 96:95–103. 13. Schnitzbauer AA, Lang SA, Goessmann H, Nadalin S, Baumgart J, Farkas SA et al. (2012) Right portal vein ligation combined with in situ

230:309–318; discussion 318–321. 26. Bolton JS, Fuhrman GM. (2000) Survival after resection of multiple bilobar hepatic metastases from colorectal carcinoma. Ann Surg 231:743–751. 27. Choti MA, Sitzmann JV, Tiburi MF, Sumetchotimetha W, Rangsin R, Schulick RD et al. (2002) Trends in longterm survival following liver resection for hepatic colorectal metastases. Ann Surg 235:759– 766.

splitting induces rapid left lateral liver lobe hypertrophy enabling

28. de Santibanes E, Lassalle FB, McCormack L, Pekolj J, Quintana GO,

two-staged extended right hepatic resection in small-for-size settings.

Vaccaro C et al. (2002) Simultaneous colorectal and hepatic resections

Ann Surg 255:405–414.

for colorectal cancer: postoperative and longterm outcomes. J Am Coll

14. Torzilli G, Procopio F, Botea F, Marconi M, Del Fabbro D, Donadon M

Surg 195:196–202.

et al. (2009) One-stage ultrasonographically guided hepatectomy for

29. Minagawa M, Yamamoto J, Miwa S, Sakamoto Y, Kokudo N, Kosuge T

multiple bilobar colorectal metastases: a feasible and effective alterna-

et al. (2006) Selection criteria for simultaneous resection in patients with

tive to the two-stage approach. Surgery 146:60–71.

synchronous liver metastasis. Arch Surg 141:1006–1012; discussion

15. Gold JS, Are C, Kornprat P, Jarnagin WR, Gonen M, Fong Y et al. (2008)

1013.

Increased use of parenchymal-sparing surgery for bilateral liver metas-

30. Schlag P, Hohenberger P, Herfarth C. (1990) Resection of liver metas-

tases from colorectal cancer is associated with improved mortality

tases in colorectal cancer – competitive analysis of treatment results in

without change in oncologic outcome: trends in treatment over time in

synchronous versus metachronous metastases. Eur J Surg Oncol

440 patients. Ann Surg 247:109–117.

16:360–365.

16. Kokudo N, Imamura H, Sugawara Y, Sakamoto Y, Yamamoto J, Seki M

31. Nordlinger B, Guiguet M, Vaillant JC, Balladur P, Boudjema K,

et al. (2004) Surgery for multiple hepatic colorectal metastases. J Hepa-

Bachellier P et al. (1996) Surgical resection of colorectal carcinoma

tobiliary Pancreat Surg 11:84–91.

metastases to the liver. A prognostic scoring system to improve case

17. Abdalla EK, Vauthey JN, Ellis LM, Ellis V, Pollock R, Broglio KR et al. (2004) Recurrence and outcomes following hepatic resection, radiofrequency ablation, and combined resection/ablation for colorectal liver metastases. Ann Surg 239:818–825; discussion 825–827. 18. Tanaka K, Shimada H, Nagano Y, Endo I, Sekido H, Togo S. (2006) Outcome after hepatic resection versus combined resection and microwave ablation for multiple bilobar colorectal metastases to the liver. Surgery 139:263–273.

selection, based on 1568 patients. Association Française de Chirurgie. Cancer 77:1254–1262. 32. Tsai MS, Su YH, Ho MC, Liang JT, Chen TP, Lai HS et al. (2007) Clinicopathological features and prognosis in resectable synchronous and metachronous colorectal liver metastasis. Ann Surg Oncol 14: 786–794. 33. Beppu T, Sakamoto Y, Hasegawa K, Honda G, Tanaka K, Kotera Y et al. (2012) A nomogram predicting disease-free survival in patients with

19. de Jong MC, van Vledder MG, Ribero D, Hubert C, Gigot JF, Choti MA

colorectal liver metastases treated with hepatic resection: multicentre

et al. (2011) Therapeutic efficacy of combined intraoperative ablation

data collection as a Project Study for Hepatic Surgery of the Japanese

and resection for colorectal liver metastases: an international, multi-

Society of Hepato-Biliary-Pancreatic Surgery. J Hepatobiliary Pancreat

institutional analysis. J Gastrointest Surg 15:336–344.

Sci 19:72–84.

HPB 2013, 15, 119–130

© 2013 International Hepato-Pancreato-Biliary Association

HPB

128

34. Adam R, Bismuth H, Castaing D, Waechter F, Navarro F, Abascal A

50. Reddy SK, Pawlik TM, Zorzi D, Gleisner AL, Ribero D, Assumpcao L

et al. (1997) Repeat hepatectomy for colorectal liver metastases. Ann

et al. (2007) Simultaneous resections of colorectal cancer and synchro-

Surg 225:51–60; discussion 62.

nous liver metastases: a multi-institutional analysis. Ann Surg Oncol

35. Bockhorn M, Frilling A, Fruhauf NR, Neuhaus J, Molmenti E, Trarbach T

14:3481–3491.

et al. (2008) Survival of patients with synchronous and metachronous

51. Fahy BN, D'Angelica M, DeMatteo RP, Blumgart LH, Weiser MR, Ostro-

colorectal liver metastases – is there a difference? J Gastrointest Surg

vnaya I et al. (2009) Synchronous hepatic metastases from colon

12:1399–1405.

cancer: changing treatment strategies and results of surgical interven-

36. Huh JW, Cho CK, Kim HR, Kim YJ. (2010) Impact of resection for primary colorectal cancer on outcomes in patients with synchronous colorectal liver metastases. J Gastrointest Surg 14:1258–1264. 37. Benoist S, Pautrat K, Mitry E, Rougier P, Penna C, Nordlinger B. (2005) Treatment strategy for patients with colorectal cancer and synchronous irresectable liver metastases. Br J Surg 92:1155–1160. 38. Tebbutt NC, Norman AR, Cunningham D, Hill ME, Tait D, Oates J et al. (2003) Intestinal complications after chemotherapy for patients with unresected primary colorectal cancer and synchronous metastases. Gut 52:568–573. 39. Michel P, Roque I, Di Fiore F, Langlois S, Scotte M, Teniere P et al. (2004) Colorectal cancer with non-resectable synchronous metastases: should the primary tumour be resected? Gastroenterol Clin Biol 28:434–437.

tion. Ann Surg Oncol 16:361–370. 52. Mentha G, Roth AD, Terraz S, Giostra E, Gervaz P, Andres A et al. (2008) ‘Liver first’ approach in the treatment of colorectal cancer with synchronous liver metastases. Dig Surg 25:430–435. 53. Butte JM, Gonen M, Ding P, Goodman KA, Allen PJ, Nash G et al. (2012) Patterns of failure in patients with early-onset (synchronous) resectable liver metastases from rectal cancer. Cancer. doi: 10.1002/ cncr.27567. Epub 2012 April 19. 54. Schwarz RE, Berlin JD, Lenz HJ, Nordlinger B, Rubbia-Brandt L, Choti MA. (2013) Systemic cytotoxic and biological therapies of colorectal liver metastases: expert consensus statement. HPB 15:106–115. 55. Nordlinger B, Sorbye H, Glimelius B, Poston GJ, Schlag PM, Rougier P et al. (2008) Perioperative chemotherapy with FOLFOX4 and surgery

40. Abdalla EK, Adam R, Bilchik AJ, Jaeck D, Vauthey JN, Mahvi D. (2006)

versus surgery alone for resectable liver metastases from colorectal

Improving resectability of hepatic colorectal metastases: expert con-

cancer (EORTC Intergroup trial 40983): a randomized controlled trial.

sensus statement. Ann Surg Oncol 13:1271–1280.

Lancet 371:1007–1016.

41. Fernandez FG, Drebin JA, Linehan DC, Dehdashti F, Siegel BA, Stras-

56. Chua TC, Saxena A, Liauw W, Kokandi A, Morris DL. (2010) Systematic

berg SM. (2004) Five-year survival after resection of hepatic metastases

review of randomized and non-randomized trials of the clinical response

from colorectal cancer in patients screened by positron emission

and outcomes of neoadjuvant systemic chemotherapy for resectable

tomography with F-18 fluorodeoxyglucose (FDG-PET). Ann Surg 240:438–447; discussion 447–450.

colorectal liver metastases. Ann Surg Oncol 17:492–501. 57. Adam R, Pascal G, Castaing D, Azoulay D, Delvart V, Paule B et al.

42. Pawlik TM, Scoggins CR, Zorzi D, Abdalla EK, Andres A, Eng C et al.

(2004) Tumour progression while on chemotherapy: a contraindication

(2005) Effect of surgical margin status on survival and site of recurrence

to liver resection for multiple colorectal metastases? Ann Surg

after hepatic resection for colorectal metastases. Ann Surg 241:715– 722; discussion 722–724.

240:1052–1061; discussion 1061–1064. 58. Gallagher DJ, Zheng J, Capanu M, Haviland D, Paty P, Dematteo RP

43. Adam R, Delvart V, Pascal G, Valeanu A, Castaing D, Azoulay D et al.

et al. (2009) Response to neoadjuvant chemotherapy does not predict

(2004) Rescue surgery for unresectable colorectal liver metastases

overall survival for patients with synchronous colorectal hepatic metas-

downstaged by chemotherapy: a model to predict longterm survival. Ann Surg 240:644–657; discussion 657–658. 44. Thelen A, Jonas S, Benckert C, Spinelli A, Lopez-Hanninen E, Rudolph

tases. Ann Surg Oncol 16:1844–1851. 59. Pawlik TM, Assumpcao L, Vossen JA, Buijs M, Gleisner AL, Schulick RD et al. (2009) Trends in non-therapeutic laparotomy rates in patients

B et al. (2007) Simultaneous versus staged liver resection of synchro-

undergoing surgical therapy for hepatic colorectal metastases. Ann

nous liver metastases from colorectal cancer. Int J Colorectal Dis

Surg Oncol 16:371–378.

22:1269–1276. 45. Yan TD, Chu F, Black D, King DW, Morris DL. (2007) Synchronous resection of colorectal primary cancer and liver metastases. World J Surg 31:1496–1501.

60. Karoui M, Penna C, Amin-Hashem M, Mitry E, Benoist S, Franc B et al. (2006) Influence of preoperative chemotherapy on the risk of major hepatectomy for colorectal liver metastases. Ann Surg 243:1–7. 61. Mentha G, Majno PE, Andres A, Rubbia-Brandt L, Morel P, Roth AD.

46. Martin RC 2nd, Augenstein V, Reuter NP, Scoggins CR, McMasters KM.

(2006) Neoadjuvant chemotherapy and resection of advanced synchro-

(2009) Simultaneous versus staged resection for synchronous colorec-

nous liver metastases before treatment of the colorectal primary. Br J

tal cancer liver metastases. J Am Coll Surg 208:842–850; discussion 850–852.

Surg 93:872–878. 62. Poultsides GA, Servais EL, Saltz LB, Patil S, Kemeny NE, Guillem JG

47. Turrini O, Viret F, Guiramand J, Lelong B, Bege T, Delpero JR. (2007)

et al. (2009) Outcome of primary tumour in patients with synchronous

Strategies for the treatment of synchronous liver metastasis. Eur J Surg

stage IV colorectal cancer receiving combination chemotherapy without

Oncol 33:735–740.

surgery as initial treatment. J Clin Oncol 27:3379–3384.

48. Chen J, Li Q, Wang C, Zhu H, Shi Y, Zhao G. (2011) Simultaneous vs.

63. Zalinski S, Abdalla EK, Mahvash A, Vauthey JN. (2009) A marking

staged resection for synchronous colorectal liver metastases: a meta-

technique for intraoperative localization of small liver metastases before

analysis. Int J Colorectal Dis 26:191–199.

systemic chemotherapy. Ann Surg Oncol 16:1208–1211.

49. Brouquet A, Mortenson MM, Vauthey JN, Rodriguez-Bigas MA,

64. Galfrascoli E, Piva S, Cinquini M, Rossi A, La Verde N, Bramati A

Overman MJ, Chang GJ et al. (2010) Surgical strategies for synchro-

et al. (2011) Risk/benefit profile of bevacizumab in metastatic colon

nous colorectal liver metastases in 156 consecutive patients: classic,

cancer: a systematic review and meta-analysis. Dig Liver Dis 43:286–

combined or reverse strategy? J Am Coll Surg 210:934–941.

294.

HPB 2013, 15, 119–130

© 2013 International Hepato-Pancreato-Biliary Association

HPB

129

65. Merkel S, Bialecki D, Meyer T, Muller V, Papadopoulos T, Hohenberger

82. Mocellin S, Pilati P, Lise M, Nitti D. (2007) Meta-analysis of hepatic

W. (2009) Comparison of clinical risk scores predicting prognosis after

arterial infusion for unresectable liver metastases from colorectal

resection of colorectal liver metastases. J Surg Oncol 100:349–357.

cancer: the end of an era? J Clin Oncol 25:5649–5654.

66. Lucke B, Breedis C, Woo ZP, Berwick L, Nowell P. (1951) Differential

83. Kemeny NE, Niedzwiecki D, Hollis DR, Lenz HJ, Warren RS, Naughton

growth of blood-borne metastatic tumours in liver and lung (experi-

MJ et al. (2006) Hepatic arterial infusion versus systemic therapy for

ments with rabbit V-2 carcinoma). Am J Pathol 27:729–730.

hepatic metastases from colorectal cancer: a randomized trial of effi-

67. Ackerman NB. (1974) The blood supply of experimental liver metastases. IV. Changes in vascularity with increasing tumour growth. Surgery 75:589–596. 68. Ensminger WD. (2002) Intrahepatic arterial infusion of chemotherapy: pharmacologic principles. Semin Oncol 29:119–125. 69. Allen PJ, Stojadinovic A, Ben-Porat L, Gonen M, Kooby D, Blumgart L et al. (2002) The management of variant arterial anatomy during hepatic arterial infusion pump placement. Ann Surg Oncol 9:875–880. 70. Allen PJ, Nissan A, Picon AI, Kemeny N, Dudrick P, Ben-Porat L et al. (2005) Technical complications and durability of hepatic artery infusion pumps for unresectable colorectal liver metastases: an institutional experience of 544 consecutive cases. J Am Coll Surg 201:57–65. 71. Ensminger WD, Gyves JW. (1983) Clinical pharmacology of hepatic arterial chemotherapy. Semin Oncol 10:176–182. 72. Sigurdson ER, Ridge JA, Kemeny N, Daly JM. (1987) Tumour and liver drug uptake following hepatic artery and portal vein infusion. J Clin Oncol 5:1836–1840.

cacy, quality of life, and molecular markers (CALGB 9481). J Clin Oncol 24:1395–1403. 84. Kemeny N, Gonen M, Sullivan D, Schwartz L, Benedetti F, Saltz L et al. (2001) Phase I study of hepatic arterial infusion of floxuridine and dexamethasone with systemic irinotecan for unresectable hepatic metastases from colorectal cancer. J Clin Oncol 19:2687–2695. 85. Kemeny N, Huang Y, Cohen AM, Shi W, Conti JA, Brennan MF et al. (1999) Hepatic arterial infusion of chemotherapy after resection of hepatic metastases from colorectal cancer. N Engl J Med 341: 2039–2048. 86. Kemeny NE, Gonen M. (2005) Hepatic arterial infusion after liver resection. N Engl J Med 352:734–735. 87. Nelson R, Freels S. (2006) Hepatic artery adjuvant chemotherapy for patients having resection or ablation of colorectal cancer metastatic to the liver. Cochrane Database Syst Rev (4)CD003770. 88. Alberts SR, Roh MS, Mahoney MR, O'Connell MJ, Nagorney DM, Wagman L et al. (2010) Alternating systemic and hepatic artery infusion

73. Stagg RJ, Venook AP, Chase JL, Lewis BJ, Warren RS, Roh M et al.

therapy for resected liver metastases from colorectal cancer: a North

(1991) Alternating hepatic intra-arterial floxuridine and fluorouracil: a

Central Cancer Treatment Group (NCCTG)/ National Surgical Adjuvant

less toxic regimen for treatment of liver metastases from colorectal

Breast and Bowel Project (NSABP) phase II intergroup trial, N9945/CI-

cancer. J Natl Cancer Inst 83:423–428.

66. J Clin Oncol 28:853–858.

74. Kerr DJ, McArdle CS, Ledermann J, Taylor I, Sherlock DJ, Schlag PM

89. Kennedy AS, Nutting C, Coldwell D, Gaiser J, Drachenberg C. (2004)

et al. (2003) Intrahepatic arterial versus intravenous fluorouracil and

Pathologic response and microdosimetry of (90)Y microspheres in man:

folinic acid for colorectal cancer liver metastases: a multicentre rand-

review of four explanted whole livers. Int J Radiat Oncol Biol Phys

omized trial. Lancet 361:368–373.

60:1552–1563.

75. Ducreux M, Ychou M, Laplanche A, Gamelin E, Lasser P, Husseini F

90. Gray B, Van Hazel G, Hope M, Burton M, Moroz P, Anderson J et al.

et al. (2005) Hepatic arterial oxaliplatin infusion plus intravenous chemo-

(2001) Randomized trial of SIR-Spheres plus chemotherapy vs. chemo-

therapy in colorectal cancer with inoperable hepatic metastases: a trial

therapy alone for treating patients with liver metastases from primary

of the gastrointestinal group of the Federation Nationale des Centres de Lutte contre le Cancer. J Clin Oncol 23:4881–4887.

large bowel cancer. Ann Oncol 12:1711–1720. 91. Cosimelli M, Golfieri R, Cagol PP, Carpanese L, Sciuto R, Maini CL et al.

76. Boige V, Malka D, Elias D, Castaing M, De Baere T, Goere D et al. (2008)

(2010) Multicentre phase II clinical trial of yttrium-90 resin microspheres

Hepatic arterial infusion of oxaliplatin and intravenous LV5FU2 in

alone in unresectable, chemotherapy refractory colorectal liver metas-

unresectable liver metastases from colorectal cancer after systemic chemotherapy failure. Ann Surg Oncol 15:219–226.

tases. Br J Cancer 103:324–331. 92. Lewandowski RJ, Thurston KG, Goin JE, Wong CY, Gates VL, Van

77. Gluck WL, Akwari OE, Kelvin FM, Goodwin BJ. (1985) A reversible

Buskirk M et al. (2005) 90Y microsphere (TheraSphere) treatment

enteropathy complicating continuous hepatic artery infusion chemo-

for unresectable colorectal cancer metastases of the liver: response

therapy with 5-fluoro-2-deoxyuridine. Cancer 56:2424–2427.

to treatment at targeted doses of 135–150 Gy as measured by

78. Northover JM, Terblanche J. (1979) A new look at the arterial supply of the bile duct in man and its surgical implications. Br J Surg 66:379–384.

[18F]fluorodeoxyglucose positron emission tomography and computed tomographic imaging. J Vasc Interv Radiol 16:1641–1651.

79. Kemeny MM, Battifora H, Blayney DW, Cecchi G, Goldberg DA, Leong

93. Fiorentini G, Aliberti C, Tilli M, Mulazzani L, Graziano F, Giordani P et al.

LA et al. (1985) Sclerosing cholangitis after continuous hepatic artery

(2012) Intra-arterial infusion of irinotecan-loaded drug-eluting beads

infusion of FUDR. Ann Surg 202:176–181.

(DEBIRI) versus intravenous therapy (FOLFIRI) for hepatic metastases

80. Kemeny N, Seiter K, Niedzwiecki D, Chapman D, Sigurdson E, Cohen A et al. (1992) A randomized trial of intrahepatic infusion of fluorodeoxyuridine with dexamethasone versus fluorodeoxyuridine alone in the treatment of metastatic colorectal cancer. Cancer 69:327–334.

from colorectal cancer: final results of a phase III study. Anticancer Res 32:1387–1395. 94. Martin RC 2nd, Scoggins CR, Tomalty D, Schreeder M, Metzger T, Tatum C et al. (2012) Irinotecan drug-eluting beads in the treatment of

81. Kemeny N, Capanu M, D'Angelica M, Jarnagin W, Haviland D, Demat-

chemo-naïve unresectable colorectal liver metastasis with concomitant

teo R et al. (2009) Phase I trial of adjuvant hepatic arterial infusion (HAI)

systemic fluorouracil and oxaliplatin: results of pharmacokinetics and

with floxuridine (FUDR) and dexamethasone plus systemic oxaliplatin, 5-fluorouracil and leucovorin in patients with resected liver metastases from colorectal cancer. Ann Oncol 20:1236–1241.

HPB 2013, 15, 119–130

phase I trial. J Gastrointest Surg 16:1531–1538. 95. Fiorentini G, Poddie DB, Cantore M, Giovanis P, Guadagni S, De Giorgi U et al. (2001) Locoregional therapy for liver metastases from colorectal

© 2013 International Hepato-Pancreato-Biliary Association

HPB

130

cancer: the possibilities of intra-arterial chemotherapy, and new hepatic-directed modalities. Hepatogastroenterology 48:305–312.

radiofrequency ablation as treatment of colorectal liver metastases. Eur J Cancer 45:1748–1756.

96. Fiorentini G, Aliberti C, Turrisi G, Del Conte A, Rossi S, Benea G et al.

105. Mulier S, Mulier P, Ni Y, Miao Y, Dupas B, Marchal G et al. (2002)

(2007) Intra-arterial hepatic chemoembolization of liver metastases from

Complications of radiofrequency coagulation of liver tumours. Br J Surg

colorectal cancer adopting irinotecan-eluting beads: results of a phase II clinical study. In Vivo 21:1085–1091. 97. Aliberti C, Tilli M, Benea G, Fiorentini G. (2006) Transarterial chemoembolization (TACE) of liver metastases from colorectal cancer using irinotecan-eluting beads: preliminary results. Anticancer Res 26 (5B): 3793–3795. 98. Martin RC, Joshi J, Robbins K, Tomalty D, Bosnjakovik P, Derner M et al. (2011) Hepatic intra-arterial injection of drug-eluting bead, irinote-

89:1206–1222. 106. Kingham TP, Tanoue M, Eaton A, Rocha FG, Do R, Allen P et al. (2011) Patterns of recurrence after ablation of colorectal cancer liver metastases. Ann Surg Oncol 19:834–841. 107. Livraghi T, Solbiati L, Meloni F, Ierace T, Goldberg SN, Gazelle GS. (2003) Percutaneous radiofrequency ablation of liver metastases in potential candidates for resection: the ‘test-of-time approach’. Cancer 97:3027–3035.

can (DEBIRI) in unresectable colorectal liver metastases refractory to

108. Hammill CW, Billingsley KG, Cassera MA, Wolf RF, Ujiki MB, Hansen

systemic chemotherapy: results of multi-institutional study. Ann Surg

PD. (2011) Outcome after laparoscopic radiofrequency ablation of

Oncol 18:192–198.

technically resectable colorectal liver metastases. Ann Surg Oncol 18:

99. Tomlinson JS, Jarnagin WR, DeMatteo RP, Fong Y, Kornprat P, Gonen M et al. (2007) Actual 10-year survival after resection of colorectal liver metastases defines cure. J Clin Oncol 25:4575–4580. 100. House MG, Ito H, Gonen M, Fong Y, Allen PJ, DeMatteo RP et al. (2010)

1947–1954. 109. Iannitti DA, Martin RC, Simon CJ, Hope WW, Newcomb WL, McMasters KM et al. (2007) Hepatic tumour ablation with clustered microwave antennae: the US Phase II trial. HPB 9:120–124.

Survival after hepatic resection for metastatic colorectal cancer: trends

110. Martin RC, Scoggins CR, McMasters KM. (2010) Safety and efficacy of

in outcomes for 1600 patients during two decades at a single institution.

microwave ablation of hepatic tumours: a prospective review of a 5-year

J Am Coll Surg 210:744–752; discussion 752–755. 101. Lupinacci R, Penna C, Nordlinger B. (2007) Hepatectomy for resectable colorectal cancer metastases – indicators of prognosis, definition of resectability, techniques and outcomes. Surg Oncol Clin N Am 16: 493–506. 102. Rocha FG, D'Angelica M. (2010) Treatment of liver colorectal metastases: role of laparoscopy, radiofrequency ablation, and microwave coagulation. J Surg Oncol 102:968–974. 103. Hompes D, Prevoo W, Ruers T. (2011) Radiofrequency ablation as a treatment tool for liver metastases of colorectal origin. Cancer Imaging 11:23–30. 104. Stang A, Fischbach R, Teichmann W, Bokemeyer C, Braumann D. (2009) A systematic review on the clinical benefit and role of

HPB 2013, 15, 119–130

experience. Ann Surg Oncol 17:171–178. 111. Dawood O, Mahadevan A, Goodman KA. (2009) Stereotactic body radiation therapy for liver metastases. Eur J Cancer 45:2947–2959. 112. van der Pool AE, Mendez Romero A, Wunderink W, Heijmen BJ, Levendag PC, Verhoef C et al. (2010) Stereotactic body radiation therapy for colorectal liver metastases. Br J Surg 97:377–382. 113. Chang DT, Swaminath A, Kozak M, Weintraub J, Koong AC, Kim J et al. (2011) Stereotactic body radiotherapy for colorectal liver metastases: a pooled analysis. Cancer 117:4060–4069. 114. Rule W, Timmerman R, Tong L, Abdulrahman R, Meyer J, Boike T et al. (2011) Phase I dose-escalation study of stereotactic body radiotherapy in patients with hepatic metastases. Ann Surg Oncol 18:1081– 1087.

© 2013 International Hepato-Pancreato-Biliary Association