Surgical and Interventional Management of Liver Metastasis

Abstract

Colorectal cancer is one of the most common cancers diagnosed worldwide. While the incidence of colorectal cancer has been declining since the adoption of screening colonoscopy, the findings of liver metastasis are still found in up to 25% of patients at diagnosis. The management of liver metastasis has evolved over the past two to three decades, and survival rates have improved secondary to improved systemic therapy, surgical options, and local therapies. In this article, we aim to review the available surgical and ablative options for management of colorectal liver metastasis, as well as appropriate imaging and patient selection.

Colorectal cancer (CRC) is one of the most common cancers diagnosed worldwide. In the United States, it is the third most common cancer in both males and females. The overall incidence of colon cancer is estimated to be 104,279 per year and rectal cancer is 45,230 per year. Together, colon and rectal cancer account for 52,980 deaths per year. ¹ Overall incidence of CRC has declined that is attributed to increased use of colonoscopy and death rate from CRC has dropped by 53% in males and 59% in females. ¹

The most common site of metastasis in CRC is the liver. Approximately 20 to 25% of patients are diagnosed with metastatic disease to the liver at the time of initial diagnosis. ² Additionally, the incidence of colorectal liver metastasis (CRLM) is 12.8% in 5 years from diagnosis. ³ Management of CRLM has evolved significantly in the past two to three decades and subsequently overall survival (OS) has improved. Previously, stage IV CRC was considered incurable and treatment was limited to palliative systemic therapy. ⁴ With advances in systemic chemotherapy and biologic agents, introduction of surgical resection and widening of indications for resection, as well as, increasing variety of local therapies, metastatic CRC (mCRC) is becoming a chronic disease in many patients. The purpose of this article is to evaluate and discuss the surgical and interventional management of liver metastasis in CRC.

Imaging

Timely detection of metastasis during initial workup of a newly diagnosed CRC as well as during surveillance is crucial for adequate management. The latest National Comprehensive Cancer Network guidelines (NCCN) (Version 1.2022) considers computed tomographic (CT) imaging modality as first line. Imaging of the chest, abdomen, and pelvis should be obtained to evaluate for locoregional and distant disease. This should be performed with intravenous (IV) iodinated contrast and oral contrast. Chest imaging does not require IV contrast; however, it is often done when in conjunction with abdominal and pelvic imaging. Magnetic resonance imaging (MRI) with gadolinium-based contrast agent can be performed instead if patient has a contraindication to IV iodinated contrast. Positron emission tomography/computed tomography is not indicated in the workup of metastasis of CRC and it should not be done in place of a contrast-enhanced CT or MRI. Per NCCN guidelines, MRI with hepatobiliary contrast agent is preferred over CT for better characterization of liver lesions if liver directed therapy or surgery is indicated. ^{5 6} Multiple recent studies have demonstrated higher per-lesion sensitivity of hepatobiliary contrast-enhanced MRI versus contrast-enhanced CT of 94.9 and 74.2%, respectively. This was also demonstrated with lesions smaller than 1 cm with a sensitivity of 85.7 and 50%, respectively. ⁷ It is our practice to obtain a dual-phase liver protocoled CT scan after initial CT scan identifying the metastatic hepatic lesions. This allows for high-quality imaging that can be used in operative planning to assess for oncologic appropriate resections, while also defining the metastatic lesion's relationship to the portal vein, hepatic artery, hepatic vein, and biliary tree. We use MRI selectively in patients with fatty liver disease, underlying fibrosis, or those who have been treated with neoadjuvant chemoradiation with a clinical response. These lesions are better characterized with MRI especially if less than 1 cm.

Staging

The American Joint Committee on Cancer (AJCC) 8th edition cancer staging system classification of distant metastasis is as follows:

• M0: No distant metastasis based on imaging. No evidence of tumor in distant sites or organs.

• M1: Metastasis to one or more distant site or organs or peritoneal metastasis is identified.

  • M1a: Metastasis to one site or organ is identified without peritoneal metastasis.

  • M1b: Metastasis to two or more sites or organs identified without peritoneal metastasis.

  • M1c: Metastasis to peritoneal surface is identified alone, or with other site or organ metastases.

Systemic Therapy for Metastatic Colorectal Cancer

The benefits of neoadjuvant therapy in the setting of resectable metastatic disease include decrease in tumor size, eradication or control of micrometastatic disease, assessment of tumor response to specific chemotherapy treatment, and better chemotherapy tolerance. Nonetheless, patients are at risk for disease progression, which may be rendered unresectable further down, along with the multitude of side effects of chemotherapy, including chemotherapy-associated liver injury. ⁸

Currently, first-line chemotherapy for CRC with CRLM includes fluorouracil, leucovorin and oxaliplatin (FOLFOX); fluorouracil, leucovorin, irinotecan (FOLFURI), capecitabine plus oxaliplatin (XELOX); and fluorouracil, leucovorin, oxaliplatin and irinotecan (FOLFOXFIRI). Biologic agents are also being used in combination with above as targeted therapy for certain mutations.

The European Organization for Research and Treatment of Cancer (EORTC) 40983 trial was a multicenter randomized control trial that set to determine if perioperative chemotherapy in resectable CRLM increased survival rate. Patients were randomized to receive six cycles of FOLFOX prior to surgery and an additional six cycles after surgery versus surgery upfront. The study demonstrated a significant increase in progression-free survival (PFS) in the perioperative group compared with surgery alone for eligible patients. It did not, however, show a significant difference in OS. ⁹ Optimal treatment for CRLM is still debated and further development in systemic therapy is still needed.

Surgical Management

Treatment for mCRC has evolved significantly in the past four decades. Just over 20 years ago, life expectancy for mCRC was 9 months, with a 5-year survival rate of 3%. ¹⁰ Liver resection has been performed with curative intent in the past. However, historically, only 20% of patients with CRLM were considered resectable. ² More recently, advances in systemic therapy, biologic agents, local therapies, and expansion of the definition of resectability, have increased the number of patients eligible for resection. OS rate in resection of solitary colorectal metastasis has been shown to be up to 71%. ¹¹ Recent meta-analysis also demonstrated 5-year median survival rate to be 38%. ¹² However, despite these advances, careful patient selection requires an individualized approach in a multidisciplinary setting. This should include colorectal surgery, hepatobiliary surgery, medical oncology, radiation oncology, pathology, and radiology.

Resectability of Colorectal Liver Metastasis

Determining resectability of CRLM is a complex process that involves determination of oncologic and technical resectability and patient fitness. It is crucial for surgeons to carefully evaluate preoperative imaging. Determining the location of the tumor, size, and proximity to important structures will dictate if surgery is feasible. The main components of technical resectability are ¹³ :

• 1) Complete resection with tumor-free margins (R0 resection).

  There has been extensive debate regarding optimal tumor resection margins. Previously, 1 cm margins were used to determine resectability; however, numerous studies have challenged this concept. A multi-institutional study in 2005 demonstrated that the width of a negative resection margin does not impact the OS. ² In 2015, the Expert Group on OncoSurgery management of Liver Metastasis (EGOSLIM) stated that the goals of therapy are safe resection margins, and a minimal surgical clearance margin of 1 mm has been suggested as sufficient.

• 2) Preservation of at least two disease-free liver segments with intact portal and arterial inflow, as well as venous and biliary outflow, with an optimal future liver remnant (FLR).

With this said, a small lesion in the periphery of the liver can be resected with a nonanatomical resection with minimal complications in most healthy patients. However, as lesions get larger, there are numerous lesions presents, bilobar disease, or lesions in close proximity to main vascular and biliary structures; anatomic resection may be indicated, and a larger amount of liver will need to be resected. A safe FLR is at least 25% of the preoperative liver volume in patients with normal liver parenchyma. Patients with compromised liver function require a higher FLR. If prior chemotherapy was administered, an FLR of at least 30% is required. Patients with liver disease like steatosis and cirrhosis need at least 40%. FLR is evaluated using functional testing like hepatobiliary scintigraphy to determine the uptake and excretion and can be combined with single-photon emission computed tomography-computed tomography to differentiate functional from nonfunctional liver tissue. ¹⁴

It is important to understand that even in the setting of a resectable liver metastasis consistent with the above criteria, if a patient has uncontrolled extrahepatic disease, for example unresectable primary, peritoneal disease, widespread pulmonary disease, and extensive nodal disease, this will render the metastasis oncologically unresectable.

There are situations in which resection is feasible; however, the patient's FLR is deemed insufficient. Portal vein occlusion is an option to increase resectability in those with borderline FLR. This can be done by means of portal venous embolization (PVE). The reasoning behind PVE is to occlude the portal vein of the disease side, to cause hypertrophy of the healthy or future remaining liver parenchyma. Four weeks after procedure, imaging is obtained to determine degree of hypertrophy and kinetic growth. There is, however, a risk of oncologic progression during this time.

Surgical Approach

Traditionally, open liver resection (OLR) was the standard of care. With great advances in minimally invasive surgery (MIS), it is becoming the preferred technique when possible. Studies have consistently shown that oncologic results in MIS approach are congruent to those with OLR. Additionally, MIS has known benefits in respect to recovery time, pain control, and hospital stay. In a multicenter randomized trial, Fichtinger et al described laparoscopic hemihepatectomy to be superior to open hemihepatectomy in terms of time to functional recovery and length of hospital stay. They did not find a significant difference in resection margins, complications, or OS at 3-year follow-up. ¹⁵ MIS intervention will highly depend on the location of the disease, the availability of appropriate laparoscopic or robotic tools needed, as well as surgeon expertise and comfort. When able, an MIS approach should be offered.

Resectable Liver Metastasis

Synchronous CRLM is defined as liver metastasis detected at or before diagnoses of the primary tumor. ¹⁶ The current NCCN guidelines (Version 1.2022) recommend either upfront surgery with adjuvant therapy, or neoadjuvant therapy, followed by surgery and adjuvant therapy for resectable metastasis in colon cancer. ⁵ Thus, the decision should be made after a thorough individualized evaluation and discussion by a multidisciplinary team.

Regardless of systemic therapy, currently, surgical intervention is the gold standard for cure of mCRC with CRLM. The question then becomes what is the optimal timing for surgical intervention. Whether simultaneous resection or colon or liver-first approach should be performed is highly dependent on the specific patient presentation and comorbidities. It must take into account the burden of liver disease, as well as the planned colon operation, and most importantly the fitness of the patient for surgery.

Abelson et al analyzed a total of 1,430 patient who underwent surgery for synchronous CRLM. Seventy-six percent of patients underwent simultaneous resection, and the other 24% underwent either colon or liver first. There was no significant difference in major events or anastomotic leaks in patients who underwent simultaneous surgery as compared with a staged approach. Furthermore, simultaneous resection approach showed reduced health care utilization. ¹⁷ Similarly, Martin et al had shown that perioperative mortality was similar and overall complications were less with simultaneous approach. This was mainly achieved by avoidance of a second laparotomy. ^{18 19} Based on these findings, a simultaneous resection is a reasonable and safe option for patients with an asymptomatic primary and resectable CRLM.

A special situation is patients with a mid or low rectal cancer presenting with synchronous liver metastases. Traditionally, these patients may have been considered for liver first approach, as long course chemoradiation with 5-fluorouracil placed these patients at risk of progression of the CRLM during the treatment of the primary tumor. ³

Giuliante et al evaluated a total of 7,360 patients who underwent primary first (4,415), liver first (552), and simultaneous resections (2,393) in the LiverMetSurvey registry between 2000 and 2017. Rectal cancer patients were more likely to undergo a liver-first approach compared with the other two groups. The liver-first approach showed a clear survival advantage over the primary first and simultaneous approaches specifically in patients with multiple bilobar metastasis. Three-year survival rate was above 65% for liver-first approach in these patients, compared with 50 to 60% in the other two groups. Patients with unilobar disease had similar results with all three approaches. ²⁰

With the recent gain in popularity of total neoadjuvant therapy (TNT) for locally advanced rectal cancer, the approach to these patients is evolving. In the Rectal cancer And Preoperative Induction therapy followed by Dedicated Operation (RAPIDO) trial, patients were randomized to treatment with short course radiation followed by nine cycles of FOLFOX then total mesorectal excision (TME) (experimental group) or long course radiation followed by TME then adjuvant chemotherapy (standard of care group). At 3 years after randomization, the cumulative probability of disease-related treatment failure was 23.7% in the experimental group versus 30.4% in the standard of care group ( p  = 0.019). ²¹ Administration of chemotherapy upfront will change the approach to rectal cancer patients with liver metastases. While it is uncommon to combine a low rectal cancer surgery with a major hepatectomy, with downstaging of liver disease after TNT, smaller resections and ablative therapies are more feasible.

Ablative Therapies for Metastatic Colorectal Cancer to the Liver

Surgical resection is the treatment of choice for most patients with CRLM. Unfortunately, only 20% of patients with isolated CRLM undergo surgery. ²² Factors such as disease burden, tumor location, function liver reserve, and patient functional status preclude some patients from undergoing hepatic resection. For these patients, nonresectional therapies exist and may be appropriate to treat CRLM with curative intent. Current techniques most commonly utilized for ablation of liver lesions include radio frequency ablation (RFA) and microwave ablation (MWA). Although surgical resection is considered the favored approach to CRLM, based on current literature, there is poor evidence that nonresectional therapy is inferior to surgical resection. ²³

Radio Frequency Ablation

RFA is the most commonly used ablative technique today. This involves placing an electrode directly into the neoplasm via ultrasound or CT guidance and delivering radio frequency currents in a concentrated manner. This results in heat production, with tissue injury and necrosis. This can be performed in an open surgical manner, laparoscopically, or percutaneously, with no approach being proven to be superior to another. ²⁴ Ideally, tumors appropriate for RFA should be less than or equal to 3cm in size; this takes into account the ability of the applicator to achieve an ablative margin of at least 5 to 10 mm without overlapping ablation. In one study, tumors more than 3 cm in size were an independent predictor of shorter long tumor progression free survival (LTPFS). Local tumor progression rates after RFA for lesions less than or equal to 3 cm versus more than 3 to 5 cm were 15 and 61%, respectively. ²⁵ In addition to size criteria, lesions must not be located adjacent to large vascular structures or bile ducts, due to the potential for inadvertent injury and heat sink loss due to rapid blood flow. In one study, multivariate logistic regression analysis showed that presence or absence of a large peritumoral vessel (≥3mm) is an independent and the dominant predictor of treatment outcome. In the nonperivascular group, five of 74 (7%) had either incompletely treated tumor (manifested within 6 months) or local recurrence beyond 6 months. In the perivascular group, 15 of 31 (48%) had incompletely treated or locally recurrent tumor. ²⁶ A systematic review of the literature on RFA for CRC liver metastases reported a wide range of 5-year survival (14–55%) and local recurrence rates (3.6–60%). ²⁷

Microwave Ablation

MWA uses the same technology as RFA, but with smaller energy frequencies resulting in faster ablation times and higher therapeutic temperatures with a larger ablation zone, as well as less susceptibility to heat-sink effect, compared with RFA. For these reasons, MWA has gained in popularity recently. In one single-center, retrospective study, 57 patients were treated with percutaneous MWA over a 10-year period. At a median follow-up of 42 months, LTP was only 4%. LTPFS at 1, 3, and 5 years was 93, 58, and 39% and median LTPFS was 48 months. OS at 1, 3, and 5 years was 96, 66, 47% and median OS was 52 months. ²⁸ In this study, tumor size was not associated with shortened OS, although all tumors were less than 5 cm in size. In one small randomized trial comparing MWA to resection, 30 patients with potentially resectable hepatic metastases from CRC were randomly assigned to laparotomy with ultrasound-guided MWA or surgical resection. The authors found no difference in OS between the two groups (mean survival 27 months following MWA vs. 25 months following surgical resection; p  = 0.83).

Unresectable Liver Metastasis

Management of patients with CRLM deemed unresectable can be further characterized into potentially resectable and unequivocally unresectable. Patients with potentially resectable disease typically present with limited liver metastasis with involvement of critical structures or insufficient FLR. ^{5 6} Preoperative systemic and local therapies are used to downsize tumor burden to convert to resectable disease. Patients with high liver disease burden are unlikely to achieve an R0 resection despite systemic therapy with favorable response. ²⁹

The NCCN guidelines recommend upfront chemotherapy plus/minus antiepithelial growth factor receptor (anti-EGFR) or antivascular endothelial growth factor as first line in unresectable CRLM. ^{5 6} The EORTC 40983 phase III trial showed a 40% response rate and an absolute increase in PFS of 7.3% at 3 years with doublet therapy FOLFOX without increase in operative mortality. ²² In addition, incorporation of target therapy like bevacizumab, panitumumab, and cetuximab based on the KRAS (Kirsten rat sarcoma) and BRAF (v-raf murine sarcoma viral oncogene homolog B1) molecular status of the tumor has been shown to improve response rate. ³⁰ The OLIVIA (Bevacizumab plus mFOLFOX-6 or FOLFOXIRI in patients with initially unresectable liver metastases from colorectal cancer) trial demonstrated an increase in resection rate, R0 rates and PFS with addition of bevacizumab to triplet chemotherapy. ³¹ Addition of EGFR blockade has also shown promising results. Patients with RAS wild-type tumors had improved conversion to resectability and R0 rates compared with systemic therapy alone, when adding anti-EGFR. ³²

While undergoing chemotherapy, re-evaluation should occur after 2 months and every 2 months thereafter. Patients who are converted to resectable metastasis should undergo synchronized or staged resection with additional adjuvant therapy. ^{5 6} In the setting of bevacizumab use in conversion therapy, a minimum of 6 weeks between the last dose and surgery is necessary, as well as 6 to 8 weeks postoperatively prior to reinitiation if indicated. ^{5 6}