Surgery for Nonresectable Metastatic Colorectal Cancer

Fig. 15.1
Results of chemotherapy: CT scan demonstrating. (a) Tumor burden prechemotherapy treatment; (b) tumor burden postchemotherapy treatment

The efficacy of FOLFOX and FOLFIRI has been confirmed in large single-center series. These regimens are considered effective in facilitating hepatic resection in selected, initially nonresectable patients. Increasingly, however, the trend is to use a combination of three chemotherapy agents (all cytotoxic agents or two cytotoxic agents and one biologic agent). For example, in the phase III CRYSTAL trial, which included 1,217 patients, combined use of cetuximab with FOLFIRI (5-fluorouracil, irinotecan, leucovorin) increased response rates (59 % vs. 43 %, P = 0.004) and PFS (HR 0.68, CI 0.50–0.94, P = 0.02) in patients with K-ras wild-type (wt) tumors and increased R0 resection rates of patients with initially unresectable metastatic CRC (4.8 % with FOLFIRI + cetuximab vs. 1.7 % with FOLFIRI alone [includes both K-ras wt and mutant tumor status]) [16]. The OPUS trial has obtained similar results (FOLFOX ± cetuximab vs. standard chemotherapy alone). The response rate in patients with K-ras wild-type tumors was 61 % with the addition of cetuximab vs. 37 % with standard chemotherapy [17]. Another randomized phase II multicenter study (the CELIM study) of cetuximab plus FOLFOX6 or cetuximab plus FOLFIRI in the neoadjuvant setting of nonresectable metastatic CRC confined to liver found response rates of 68 % in the FOLFOX6 arms and 57 % in the FOLFIRI arms. In a combined analysis of both arms, response rate was 70 % in patients with wild-type K-ras tumors. R0 resections were performed in 34 % of patients [18].

In addition, surgeons have begun to examine combinations of three times cytotoxic chemotherapy plus antibody treatment with bevacizumab or cetuximab. Randomized trials have shown that combination of a biologic agent with an oxaliplatin- or irinotecan- based regimen can increase efficacy and also the rate of secondary resection of metastatic lesions [19]. The combination of cetuximab with a chronomodulated FOLFOXIRI regimen resulted in an 85 % response rate and a 75 % resection rate. However, dose reduction was necessary because of unacceptable rates of diarrhea, and a less conservative definition of nonresectability was used [20]. Further studies are needed to show an advantage over FOLFOXIRI or chemotherapy plus cetuximab.

As for initially nonresectable CRLM to conventional systemic therapy, a number of studies have shown very significantly results. A retrospective study evaluated 151 initially nonresectable CRLM patients to first-line conventional chemotherapy, who then underwent combining therapy with cetuximab [21]. After a median of six cycles of combining therapy with cetuximab, 25 (16 %) of those patients underwent surgery. After a median of 16 months follow-up, 23 of the 25 patients (92 %) were alive and 10 (40 %) were disease-free. Median OS durations from initiation of cetuximab therapy were 20 months and PFS durations were 13 months. Similarly, in a single-arm study, initially unresectable CRLM patients to tritherapy with fluorouracil/leucovorin, irinotecan, and oxaliplatin, 82 % of patients could have R0 resection. Complete clinical remission rate postoperative was 79 %, and 2-year survival rate was 83 % following triple cytotoxic chemotherapy [22]. Intra-arterial Chemotherapy

The interest in using intra-arterial chemotherapy in neoadjuvant setting has also progressively increased as it has been demonstrated to have a high response rate in both the first- and second-line settings. In a Clavien et al. study, 6 (26 %) of 23 previously treated patients were induced resectability using HAI-FUDR with or without leucovorin. The actuarial 3-year survival rates were 84 % for neoadjuvant therapy responders compared to 40 % for nonresponders [23]. In a Memorial Sloan-Kettering hospital study [24], 44 patients with extensive liver metastases using HAI-FUDR and dexamethasone plus oxaliplatin-based systemic chemotherapy as part of two phase I trials in 44 extensive liver metastases patients. The study population in this trial had a high number of patients with more than four metastases, metastases greater than 5 cm, more than 25 % liver involvement with tumor, a CEA level greater than 10 ng/dl, and previous chemotherapy exposure. Despite these negative parameters, the objective response rate was 82 %; therefore, 9 (20 %) of the 44 patients underwent complete gross resection of tumor and a median survival for all patients of 26 months. The current, original data from several clinical trials using the oxaliplatin or irinotecan via HAI have been promising.

15.3.2 Adjunctive Techniques Employed to Reduce the Liver Volume to Be Resected Tumor Ablation Techniques

The use of ablative techniques enables the possibility to avoid resection of healthy parenchyma around tumors, thus permitting treatment of a greater number of lesions. Efficacy of these local ablative techniques is considered superior to that of chemotherapy alone.

Radiofrequency ablation (RFA) – This technique uses heat (200 to 2 MHz) to destroy solid organ tumors. It is considered as a curative treatment for hepatocellular carcinoma smaller than 3 cm. It has also been proven to be an effective technique to treat colorectal liver metastases (Fig. 15.2). Risk factors for failure of RFA are tumor volume (>3 cm), centrally located metastases and proximity to large vessels [25, 26], age above 55 years, and percutaneous approach. Morbidity is low (range 0–33 %) but, when occurring, can be very deleterious. RFA procedure when performed in combination with surgery increases the resectability and curability for patients in whom hepatic resection alone is not curative. It has reported that adding RFA to liver resection could be well tolerated with a perioperative morbidity and mortality comparable to those seen after resection alone [27]. For metastases considered as unresectable, RFA combined with hepatic resection can achieve a median survival as high 37 months [28].


Fig. 15.2
Schematic demonstration of the tumor radiofrequency ablation (RFA) used as adjunctive procedure to liver surgery

Cryotherapy – This method causes destruction of tumoral cells by direct cellular freezing and indirectly through vascular thrombosis and tissue anoxia (Fig. 15.3). It has proven to be effective in liver metastases of colorectal cancer in terms of survival, and results of such treatment combined with hepatic resection for patients not eligible for hepatic resection alone have shown a 5-year survival rate of 24 %, better than those obtained by palliative chemotherapy [8, 29]. Local recurrence at the site of cryotherapy occurs in 5–44 % of patients, and it has been found that the rate increases when treating multiple lesions (>8), large lesions (>3 cm), or tumors located to major blood vessels (blood warmth may impair the freezing process). Recurrence rate is however higher than after liver resection and RFA. Edge cryotherapy is utilized in some centers to achieve a 1 cm margin after hepatectomy. However, with the emergence of radiofrequency ablation, the use of cryotherapy in liver metastases has become limited.


Fig. 15.3
Intraoperative cryotherapy ablation as adjunctive procedure to liver resection

15.3.3 Techniques Employed to Improve Remnant Liver Volume Preoperative Portal Vein Embolization

The first cause of mortality after liver resection remains liver failure. This complication appears when the liver remnant volume is too small to cope with the postoperative metabolic demands. Criteria defining this state are encephalopathy, hyperbilirubinemia, and coagulopathy, which are frequently associated with renal insufficiency. It is accepted that the risk of liver failure is considerable when remnant liver is less than 30 % (normal parenchyma) and less than 40 % (pathological parenchyma). In this context, embolizing one side of the portal venous system that induces the contralateral liver lobe hypertrophy (the future remnant liver) can reduce the incidence of postoperative liver failure.

This phenomenon (unilateral hypertrophy) was initially observed in intrahepatic cholangiocarcinomas, where compression of a portal branch caused atrophy of segments downstream to this branch and compensatory hypertrophy of the remnant liver. Makuuchi was the first to utilize this observation by occluding the right branch of the portal vein before a right hepatectomy. The future resected liver atrophied while the future remnant liver grew in size. Many products have been used to occlude the portal vein including gelatin sponge, coils, cyanoacrylate, and alcohol. None has shown an advantage in inducing atrophy or hypertrophy of the liver. Recently, transient portal vein embolization with reabsorbable occlusive products has been successfully developed and employed [30].

There are several ways of occluding the portal vein. Portal vein embolization can be done percutaneously under radioscopic control (Fig. 15.4) by punctioning the contralateral branch of the portal vein to be embolized (e.g., left branch of PV). The product is injected or deployed in the contralateral side (e.g., right branch of PV) in the same direction of the bloodstream. The ipsilateral approach is another alternative method performed by punctioning the right portal vein and injecting the product in counter stream direction.


Fig. 15.4
(a) Radiological demonstration of percutaneous portal vein embolization; (b) intraoperative demonstration of the right lobe atrophy with a contralateral lobe hypertrophy

Surgical approaches have also been described. An ileocolic vein approach is sometimes performed through a mini-laparotomy. The catheter is pushed to the desired portal branch through the superior mesenteric vein accessed by an ileocolic vein.

Finally, portal vein occlusion can be performed as a part of a two-step resection (see Two-Stage Hepatectomy). There are reports of laparoscopic portal vein ligation in the same operative time as primary resection [31]. The results are comparable to those of radiological or laparotomy PVE/PVL. Mortality of PVE or PVL is exceptional. Morbidity is mainly due to the so-called post-embolization syndrome which is characterized by nausea and vomiting.

Standard chemotherapy does not seem to impact compensatory hyperplasia after PVE [32]. Similarly bevacizumab does not impact regeneration after PVE procedure [33]. Regeneration failure after PVE can be considered as a predictive factor for liver failure after hepatectomy as PVE is a stress test for liver regeneration. Indocyanine green clearance and three-dimensional CT scan have shown to be effective monitoring investigations for compensatory liver hypertrophy. Timing of hepatectomy after portal vein embolization is highly variable. Most frequently, the interval between the two procedures is 3–6 weeks. During this interval, chemotherapy must be resumed except for bevacizumab. Indeed there are studies underlining the risk of tumor growth in the future remnant liver though literature is controversial. Elias was the first to report a growth of metastases in the non-embolized lobe after PVE in the waiting time before surgery. Others have reported growth of metastases in the embolized lobe. However, these studies only compare pre- and post-PVE volumes and therefore do not show an increase in growth speed after PVE. To prevent the growth of metastases in the future remnant liver, association of PVE to chemoembolization, as practiced for hepatocellular carcinoma by Japanese teams, may be a treatment option. Indeed, the arterial buffer response activated after PVE could be responsible for the growth of the metastases in both lobes via an increased oxygen and nutrient support. Nevertheless, morbidity of this procedure especially biliary necrosis could be considerable, and a very careful evaluation of this strategy is always necessary. The most important figure is how this potential growth could impact on curability of the patients. In a recent meta-analysis, it is shown that, after PVE, 17 % of the embolized patients do not undergo liver resection, two thirds of which due to disease progression [34]. Two-Stage Hepatectomy


This strategy is reserved for extremely difficult cases of multinodular, bilobar disease, not manageable with standard liver resections, often requiring more than 70 % of the functional parenchyma to be sacrificed. The main principle of this strategy is sequential resection by two-staged hepatectomies. The objective of first hepatectomy is to render the eventual remnant liver parenchyma tumor-free, as a result of which, the second liver resection becomes feasible and potentially curative. The success of this procedure depends upon liver regeneration between the two interventions, which in turn allows the second resection to be performed with acceptable risks. Generally, patients can be classified into three possible groups: patients with multinodular, unilobar metastatic disease who require resection of up to 70 % of the functional parenchyma; patients with multinodular, bilobar metastatic disease, whose resection leaves no more than three nodules in the remnant liver; and lastly patients with multiple bilobar lesions, in which a planed resection would live more than three nodules or any nodule larger than 3 cm in the remnant liver. For the first group, preoperative portal vein embolization followed by liver resection provides the best surgical treatment, whereas liver resection combined with intraoperative local ablation therapy is the choice for the second group of patients. On the other hand, patients belonging to the later group would be best treated with two-stage hepatectomy; hence, it is this group of patients which would benefit the most from the recently developed strategy.

Preoperative and Intraoperative Assessment

Patients’ evaluation should include liver US, contrasted CT, or MR imaging of the abdomen and pelvis (preferred investigative modalities). It is essential for planning the resection extensiveness by determining the intrahepatic extent of the disease and its relation with important vascular and biliary structures. Local recurrence at the primary site should be excluded by performing a colonoscopy.

The functional capacity of the liver should be measured by the indocyanine green (ICG) test, to determine the necessary remaining liver volume after hepatic resection. Usually, to perform a safe hepatic resection, in the patients with absence of prolonged chemotherapy and normal ICG clearance, the remnant liver should have >30 % of functional parenchyma. For another, the functional parenchyma volume should be >40 % in patients with prolonged chemotherapy or abnormal ICG clearance.

During the intervention, the abdominal cavity is carefully explored, and if any extrahepatic tumor deposit or suspicious lymph node(s) is suspected, a frozen-section histological examination should be performed. The second step involves a manual palpation and intraoperative US examination of the liver to determine the degree of the metastatic disease affecting it. Intraoperative US is a mandatory part of the operation as it can give additional information in up to 89 % and may contribute to change therapeutic plans in up to 42 % of the cases [35]. Its guidance is invaluable, particularly when precise mapping of the anatomical relationship between metastases and main hepatic vessels is required. Its sensitivity approaches 90 %, and if there are no contraindications to surgery, the hepatic resection should begin.

Technical Aspects

Types of Resection

Depending on the pattern of the metastatic disease, different approaches can be used to decide which lobe is operated on first and often this decision is made intraoperatively. From a general point of view, the type of resection performed during the first stage can be one of the two described below.

Left metastatic resection (anatomical or non-anatomical) and right PVE – This approach consists of metastatic clearance on the left liver combined with right portal vein embolization. In addition, in patients with a primary colorectal tumor in place, a colectomy is performed during the same intervention (first stage). The initial steps of this procedure are identical to those of standard hepatectomies. However, considering that the ultimate intention is to proceed with a second liver resection, minimum dissection at the site to be resected during the second stage should be performed in order to minimize the fibrotic adhesions. Therefore, the division of triangular ligament and excessive dissection of hilar structures of the contralateral lobe is avoided. Following the mobilization of the concerned site of the liver and the control of the portal structures, the resection of the metastases commences, aiming not only to achieve the oncological target (complete clearance of tumor lesions from the left lobe with microscopically free margins) but also at the maximal conservation of the tumor-free liver parenchyma. The resection of the liver is done in the usual fashion by using either ultrasonic dissector or Kelly clamp with intermittent inflow occlusion combined with the use of low central venous pressure anesthesia. Major supplying vessels are ligated along the parenchyma transection, taking care in minimizing in maximum the blood loss which has been proven to be an independent risk factor in the postoperative outcome [4]. Following the resection, the procedure continues with the exposure of the right branch of the portal vein. The exposure is obtained by a lateral approaching of the free edge of the hepatoduodenal ligament. Knowing that position of the portal vein is posterior helps to direct the dissection toward it, minimizing excessive tissue disruption. Obtaining the control of the right branch of PV just distal to the bifurcation is followed by its ligation/division and by absolute alcohol injection (10–15 cc) into the distal end. This step has a double purpose: first, it triggers the growth of the remnant parenchyma of the left lobe, and second it prevents a cavernous transformation of the right portal system [36]. Important point of this maneuver is to ligate the vein before the injection in order to prevent proximal spilling of the alcohol which can result in thrombosis of the common portal trunk. Our practice is to check the result of embolization intraoperatively by US/Doppler examination. Demonstration of the absence of flow in the portal system as well as visualization of a newly formed thrombus would confirm the result of the procedure. The procedure is terminated after the completion of the embolization. A drain is left in the resected site of the liver, and the abdomen is closed with nonabsorbable sutures (interrupted). Subsequently, after an interval of 3–4 weeks (interval to allow liver regeneration), chemotherapy is commenced, following which the second resection (stage) is performed, aiming at a complete metastatic clearance by resecting the right lobe or individual segments harboring the remaining metastatic lesions. Our approach in deciding the appropriate time for the second liver resection is based on the amount of liver regeneration and the control of remnant tumor by chemotherapy. Although this technique is advantageous as it limits the dissection only at the hepatic pedicle, its applicability can at times be limited by the tumor volume, often making the use of local ablation therapies (RFA, cryotherapy) necessary.

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Jul 30, 2017 | Posted by in ABDOMINAL MEDICINE | Comments Off on Surgery for Nonresectable Metastatic Colorectal Cancer

Full access? Get Clinical Tree

Get Clinical Tree app for offline access