Surgical Strategies for Colorectal Liver Metastases: Central Approaches

Colorectal cancer is the third most common cancer and the third leading cause of mortality worldwide [1]. Approximately half of patients suffering from colorectal cancer will develop liver metastases during the course of the disease [1,2]. Palliative chemotherapy was the only option for patients with liver metastases from colorectal cancer (MHCC) for many years [2,3].

Unfortunately, systemic treatment alone is associated with devastating outcomes, with 5-year survival rates < 5% [3-5]. Over time, advances in chemotherapy and surgical technique meant that liver resection became a valid option for patients with MHCC, leading to long-term survival with 5-year rates of 30% to 40%. [6-8].

The treatment of MHCCs needs to be considered as a multidisciplinary effort, and the development of an effective chemotherapy regimen is an essential factor that allows the utilization of major liver transplants and resections [9].

Despite the eminent role of chemotherapy in the treatment of MHCC, the objective of this review is to focus on the evolution of central surgical strategies in the battle against MHCC, a long and eventful journey, from palliation alone, to liver transplant in very selected patients.

The first reports of liver resections for metastatic disease date back to the end of the 19th century, as described by Keen et al., in 1899 [10]. The first detailed reports on liver surgery in MHCCs are often attributed to Richard Cattel in 1940 [11,12]. In a comprehensive review, Fineberg et al. also recognized pioneering results in the first part of the 20th century by Wendel (1911), Honjo and Wangesteen (1949), and Lortat-Jacob (1952) [13].

Despite the widely negative attitude toward resections in systemic disease at the time, those early studies showed surprisingly promising results. In this context, Foster et al. published a review of 132 liver resections in 1970, including more than 80 patients with MHCC [14].

A perioperative mortality of 6%, along with 2- and 5-year survival rates of 47% and 21%, respectively, were exceptional results for that time, providing hope that resection for MHCC could be accepted. Encouraged by these results, several groups continued to reproduce these favorable results in the following 2 decades, revealing 5-year survival rates after resection of MHCCs of up to 40% [4,13-19].

Those studies increasingly challenged the prevalent paradigm that palliation was the only option for patients presenting with MHCC. But the crucial question of whether surgery added a survival benefit to the natural history of the disease remained unanswered. A critical evaluation by Wagner et al. raised the concern: “The natural history of untreated cancer is the standard by which the effectiveness of any treatment must be measured” [4]. This group not only postulated that patients with a single untreated MHCC could have a 3-year survival of 20%, but also highlighted the essential role of accurate staging and meticulous selection of patients undergoing liver resection [4 ].

Several comparative studies challenged the claims of Wagner et al., and found an extremely poor 5-year survival for untreated MHCC, between 0% and 4% [20-26]. It became apparent that the interpretation and validity of all of these studies were strongly limited by their retrospective design, warning of the need for randomized trials [27].

Encouraging reports of long-term survival after liver resections for MHCC had already raised the expectations of surgeons and patients. Therefore, a randomization of patients, withholding treatment to 1 group of patients, was considered unethical and was never performed [28]. From this, liver surgery was – at least in some centers – a valid option in the treatment spectrum of MHCC.

The evolution of liver resections for MHCCs depended largely on technical achievements in liver surgery, particularly hemorrhage control and knowledge of anatomy.

While early mortality rates for hepatectomies approached a dramatic 20% mark, technical progress improved the figures below 5%, an acceptable target in centers dedicated to hepatobiliary surgery [29]. That tremendous progress encouraged surgeons at the time, and motivated them to expand the scope of resections.

Although major liver resections for MHCC were described in the early 20th century, the prevailing opinion was that patients with solitary liver metastases would ideally be the ones to benefit from surgery [17].

In 1970, Wilson et al., presented favorable results for patients resected with a single MHCC, and poor oncological results if multiple metastases had been resected [17]. Although the role of surgery for MHCC became widely accepted, a relevant question arose: Which patient should be resected?

With increasingly precise cross-sectional imaging, the focus shifted from technical limitations to meticulous patient selection. The resectability criteria initially used consisted mainly of the dissemination pattern and size of the metastases.

In that era of “disease-focused resectability criteria,” tumor size, more than 4 lesions, and multilobar and extrahepatic disease were considered contraindications to surgery [30,31]. Fulfillment of this criterion was attributed to increasing resection rates with free margin (R0) and, consequently, allowing a 5-year survival rate of more than 20% [31].

At the same time as these surgical advances, another crucial element in the fight against MHCC, systemic chemotherapy, was constantly evolving. In 1996, the group around Henri Bismuth pushed the limits of resectability by “downstaging” previously unresectable disease using chemotherapy regimens based on 5-fluoruracil and oxyplatin [30].

This multimodal treatment of a previously unresectable disease resulted in a favorable survival of 40%. These promising results of combined treatment modalities against MHCC, surgery and systemic treatment, meant that resectability criteria had to be reconsidered [18,32].

Traditionally, a margin of 1 cm has been considered crucial for an R0 resection [31]. Over time, it became clear that “subcentimeter” margins show similarly favorable outcomes, and should not preclude patients from undergoing liver resection [33,34]. That new interpretation of R0 resection pushed the extent of resections even further, and resectability changed from a “disease-focused” perspective to a focus on the future liver remnant (FHR).

Regardless of the number and size of metastases, a functionally sufficient RHF, with preserved inflow and outflow as well as biliary drainage, became the only limitation for liver resections [35,36].

In liver volumetry, a RHF of approximately 30% of the initial liver volume has become widely accepted as the cut-off point for safe liver resections in a healthy liver [37]. In livers with extensive exposure to chemotherapy and/or underlying liver disease (steatosis, fibrosis, cirrhosis), an RHF of 40% or 50% should be considered [37].

Unfortunately, a considerable portion of patients suffer from extensive bilobar disease, which exceeds the limits of a sufficient RHF. Innovative surgeons tried to overcome that obstacle by taking advantage of the liver’s ability to regenerate.

As is known since the myth of Prometheus, the ancient Greeks were already familiar with the phenomenon of liver regeneration [38]. However, it took until 1920 for the first experimental proof that liver regeneration occurs in livers where contralateral portal vein ligation (PVL) was available [39].

Rous and Larimore, of the Rockefeller Institute , New York, using an animal model (rabbit), were able to clearly demonstrate the compensatory growth of the right liver after ligating the left portal vein, and the corresponding shrinkage of the left liver, within a few weeks [ 39]. As liver surgery was struggling at the time with technical problems, rather than limitations resulting from insufficient RHF, that concept was not followed for many decades.

In the late 1980s, Masatoshi Makuuchi of the National Cancer Center , Tokyo, made a similar observation in patients with perihilar cholangiocarcinoma and portal vein invasion [40]. He observed ipsilateral atrophy of the affected hemiliver, and corresponding contralateral hypertrophy, as a reaction of the liver to preserve liver function.

That group suggested the intentional use of such portal flow shunting preoperatively, by embolization of the portal vein to induce hepatic hypertrophy, which is widely known as portal vein embolization (PVE) [40].

Currently, percutaneous PVE can be considered a safe intervention with a low complication rate, and is generally performed by interventional radiologists. The volumetric increase in RHF usually ranges between 30% and 40% [41,42]. The group around Henri Bismuth presented their 10-year experience with EVP in the context of MHCC, in 2000 [43].

Using a mixture of enbucrylate and lipiodol, PVE was successfully performed in 30 patients. Despite a success rate of 100%, and a median hypertrophy interval of 7 weeks, only 63% (n = 19) of patients ultimately underwent resection. However, patients who were resected achieved a 5-year survival of 40%, comparable to that of initially resectable patients.

Two-stage hepatectomy (HDE), with and without portal vein occlusion (EVP or LVP), is a strategy introduced to address high tumor burden of MHCCs and insufficient RHF. This strategy was first described by Adam et al., in 2000, and consists of a two-stage surgery. In the first stage of the operation, as many metastases as possible are removed. After an interval of 2 to 14 months, allowing liver regeneration, the second stage of the operation is performed [44].

Most cases received chemotherapy during that period, and the authors reported a median overall survival of 31 months after HDE [44]. In contrast to such “staged tumor removal” without portal vein occlusion, Jaeck et al. first implemented EVP as a regenerative boost after metastasectomies, as part of a staged resection of MHCCs, in 2004. [Four. Five].

Belghiti et al. suggested a concept of HDE where contralateral metastasectomies (subsequently called “cleanup”) are combined with an ipsilateral LVP in stage I [46]. This group even combined LVP and RHF clearance with resection of the primary tumor of colorectal or neuroendocrine origin.

That study consisted of 20 patients (12 with MHCC and 8 with neuroendocrine liver metastases), of whom 15 patients (75%) underwent complete hepatectomy in the second stage. Since no major complications occurred, that new strategy appeared to be safe and feasible [46].

Comparing the degree of hypertrophy of EVP and LVP before 2-stage surgery, the increase in liver volume was 35% after EVP, and 38% after LVP [41]. The use of HDE quickly inspired hepatobiliary surgeons around the world, and those results were subsequently reproduced in several studies [43,45,47-58]. Overall, approximately 70% of patients completed resection, resulting in a 5-year survival of 42% [59].

The development of HDE was a great success, further pushing the limits of resectability. In contrast, a 70% resection rate still leaves 30% of patients unresectable. The main reasons for failure to complete an HDE are growth of hepatic tumor mass during the interval, and insufficient increase in RHF volume [59].

In most reports, the interval between portal vein occlusion and resection is between 6 and 8 weeks. Several authors raised concerns that inducing liver hypertrophy via a regenerative stimulus may also increase tumor growth during the interval [60-62].

It is true that most of those studies have small sample sizes and are retrospective in nature, but still the concept of chemotherapy after portal vein occlusion was established, to prevent possible tumor growth during the intervals [43,63 ].

Still, some groups expressed concern that “interval” chemotherapy may affect liver regeneration [64-66]. However, that could not be proven, because resection rates for HDE remained in a similar range, both for those who received interval chemotherapy and those who did not [43,59].

In 2007, a group in Germany developed a novel strategy in the management of MHCC, combining right portal vein ligation with parenchymal resection, during stage 1 of a staged hepatectomy [67]. That additional parenchymal resection resulted in accelerated liver hypertrophy and allowed completion of the hepatectomy within a week of the initial procedure [68,69].

Within the hepatobiliopancreatic (HBP) community, such a novel strategy raised hopes that many patients might be amenable to resection, compared with conventional HDE [68]. Lang et al., presented 3 cases using this new strategy in 2011 [67].

Shortly thereafter, the results of the first multicenter study investigating the outcomes of the procedure, which was initially called “ in situ division ,” were published by Schnitzbauer et al. [68].

That pioneering study demonstrated 74% RHF hypertrophy after a median of 9 days, and a 100% staged resectability rate, in otherwise unresectable neoplasms [68]. Subsequently, de Santibáñez and Clavien suggested the acronym ALPPS (for Associating Liver Partition and Portal Vein Ligation for Staged Hepatectomy ), for this novel procedure, which was finally adopted in the HBP literature [69].

Despite the excitement resulting from unprecedented liver regeneration after ALPPS, some were skeptical about the procedure due to initial reports of high mortality [68-70].

Therefore, efforts were made to make the procedure less invasive and overcome safety concerns. Parenchymal transection or approaches replacing the section with a tourniquet showed promising results, with significantly lower complication rates, and comparable liver hypertrophy [71-75]. Alternatively, laparoscopic variants, and the use of radiofrequency to mimic parenchymal transection, the so-called R-ALPPS, have been reported with encouraging results [76,77].

These technical refinements, combined with meticulous patient selection, could consolidate the role of ALPPS in the armamentarium against MHCC [78-80]. The only randomized trial available comparing ALPPS with HDE, in the MHCC setting, is the LIGRO trial [78,81].

The authors report an impressive resection rate of 92% for patients undergoing ALPPS, compared with 80% for patients undergoing conventional HDE, including 12 patients who crossed over to the ALPPS group due to unsatisfactory hypertrophy.

Perioperative mortality rates remained comparable in both groups. The authors reported a significant survival benefit for patients undergoing ALPPS, with a median survival of 46 months, compared with 26 months for patients randomized to HDE [81]. A recent study from the international ALPPS registry, which included 510 patients, found that ALPPS for previously unresectable MHCC was performed with a mortality rate below 5%, and achieved a median survival of 39 months [82].

Based on the available evidence, the ALPPS procedure appears to increase the resection rate for initially unresectable MHCC, by more than 90%, and has the potential to decrease the interval between stages to approximately 2 weeks [78,83-89 ]. Eventually, an acceptable mortality rate can be achieved for selected patients in specialized tertiary centers.

Prompted by the initially high mortality rates of the ALPPS procedure, alternative methods were developed. The idea that additional trauma increases hepatic hypertrophy was adopted by another strategy, so-called portal embolization associated with arterial ligation [90].

In a proof-of-principle study, a pioneering group found efficient regeneration of the RHF, and achieved a 100% resection rate in patients with bilobar MHCC [90]. However, the concept of completely devascularizing the hepatic lobe, with the potential for necrosis, prevented that method from gaining common acceptance.

Guiu et al. described a liver venous deprivation technique in 2016 [91]. This procedure consists of simultaneous embolization of the portal vein and the corresponding hepatic veins. Preliminary results, in a small, heterogeneous cohort, are encouraging, showing hypertrophy rates comparable to ALPPS [92].

The minimally invasive nature of that procedure was the main advantage over ALPPS, since it strictly uses an endovascular approach in the first stage. To the authors’ knowledge, 2 multicenter prospective studies, 1 French (HYPER-LIV01 trial) and 1 Dutch (DRAGON trial), are currently evaluating the value of the hepatic venous deprivation technique in regenerative liver surgery.

A different concept to address MHCCs is not to preoperatively increase RHF, but rather to preserve as much of the liver parenchyma as possible. Increasing anatomical knowledge and the use of intraoperative ultrasound allowed Minagawa et al. to present promising results for limited multiple liver resections [94].

They found that patients with ≥4 MHCC had a 10-year survival of 29% when parenchyma-sparing hepatectomy was performed. Likewise, Kokudo et al. demonstrated comparable oncological outcomes between limited non-anatomical liver resections and major anatomical resections for MHCC [95]. In the early 2000s, Torzilli et al., developed another concept of parenchymal preservation, using intraoperative ultrasound [96].

The technique follows the minimum requirements of resection margins (1 mm) in MHCC, maximizing – therefore – the conservation of the liver parenchyma. In fact, replacing a staged major hepatectomy with multiple minor hepatectomies is a reasonable strategy and requires in-depth knowledge of intrahepatic anatomy.

From a technical point of view, the detachment of the tumor from the intrahepatic vascular structures [96,97], as well as the recognition of the communicating veins that determine hepatic outflow, have been considered as crucial elements for a successful procedure [98 ].

The development of multiple non-anatomical liver resections, rather than a major hepatectomy, also prompted the use of high-resolution 3D imaging software in the strategic planning of highly complex procedures. Some centers even make 3D printing models for better planning of the surgical procedure. In the future, real-time stereotaxic liver surgery could become as important as it is today in neurosurgery [99-101].

Those complex surgical strategies, in combination with perioperative chemotherapy, demonstrated beneficial oncological outcomes for resected MHCC. However, in theory, an exchange of the entire liver would give the highest probability of an R0 resection.

In recent decades, the indication for liver transplantation (LT) has expanded to a subgroup of primary malignant diseases of the liver, under strict selection criteria [102]. Still, transplant services are reluctant to consider LT for patients suffering from metastatic disease such as MHCC.

It is obvious that the problem of lack of organs and historically poor outcomes remain major obstacles to considering MHCCs as a possible indication for LT [103]. In a rather experimental setting, some pioneers explored the role of TH for MHCCs 3 decades ago [104]. Data from the European Liver Transplant Registry , including n = 50 patients who underwent LT for MHCC before 1995, suggest a 5-year survival of 18% [103].

The fortunate donor situation in Norway and improved LT outcomes led to a revival of transplantation for MHCCs [105]. In the SECA-I trial, including 21 patients undergoing LT for MHCC, between 2006 and 2011, Hagness et al. reported a 5-year survival of 60%.

However, recurrence rates remained as high as 91% (n = 19/21). A tumor diameter > 5.5 cm, increased carcinoembryonic antigen (CEA) > 80 mg/L, tumor progression during chemotherapy, and an interval of less than 2 years from resection of the primary tumor were identified as risk factors for a poor outcome. poor oncology. These criteria are frequently mentioned as Oslo criteria.

In 2017, a subsequent multi-institutional retrospective cohort study reported equivalent oncological outcomes for n = 12 patients transplanted between 1995 and 2015 [106]. Of note, preoperative chemotherapy was administered in the majority of patients ( n = 11/12), and almost all patients ( n = 10/11) showed a response to systemic treatment before transplantation.

The subsequent SECA-II trial from the Norwegian group had more stringent inclusion criteria, and recently reported a 5-year survival rate of 83% [107]. In addition to the aforementioned Oslo criteria, only patients with an adequate radiological response (at least 10%) to chemotherapy were included in that study. Therefore, the authors concluded that LT provides the longest reported overall survival in highly selected patients with unresectable liver metastases [107].

The eminent role of selection criteria was highlighted by the Oslo group in another article [108]. A comprehensive evaluation, including the Oslo criteria, clinical Fong Score , and tumor metabolic activity on positron emission tomography computed tomography, identified patients with the most favorable outcomes [108].

Meanwhile, the shortage of donor organs remains the essential limitation for wider application of LT for MHCCs. Therefore, alternative strategies are being explored to increase the pool of accessible liver grafts.

In 2015, Line et al. suggested the RAPID technique [109]. The concept of the RAPID procedure is to transplant a small auxiliary left lateral liver graft (segments 2+3) and ligate the right portal vein, followed by hepatectomy of the native liver in a second stage, after sufficient graft regeneration.

The biggest advantage of this technique is that the remaining extended right graft can be transplanted to another patient. The original RAPID technique suggests left lateral resection (segments 2+3) of the native liver, before orthotopic transplantation of a segment 2/3 graft.

Recent modifications include a left recipient hepatectomy, allowing inclusion of the orifice of the middle hepatic vein in the vein graft anastomosis, to achieve optimal outflow [110]. The RAPID technique has been explored in deceased (DD-RAPID) and living (LD-RAPID) donors, with both showing promising preliminary results [111].

It is noteworthy that, since MHCC are not yet considered a standard indication for LT, it is of utmost importance to guarantee the safety of the living donor. In this context, a left lateral donation can be considered a less invasive alternative than the donation of a right lobe, because it is more commonly performed in transplants between adults with a living donor. Currently, ongoing trials aim to determine the safety and oncological efficacy of this new strategy.

Ravaioli et al., from Bologna, took the concept of RAPID, and performed a heterotopic transplant in the splenic fossa after a splenectomy, without manipulation of the native liver [112,113]. After sufficient graft regeneration, total hepatectomy of the native liver was performed. This procedure was later called “heterotopic transplantation of segments 2 and 3 using the splenic vein and artery after splenectomy, with delayed total hepatectomy (RAVAS).

The authors reported the case of a 40-year-old patient with unresectable synchronous MHCC, in whom HDE was attempted. However, the patient developed severe bile leak after the first stage, and the second stage of the operation could not be performed. As a salvage procedure, a rejected pediatric graft, with insufficient liver volume for LT, was heterotopically implanted into the splenic fossa.

Unlike RAPID, a tourniquet was placed over the main portal vein to direct flow toward the heterotopic graft. Within a 2-week interval, the graft-to-body weight ratio increased from 0.6 to 1, making native hepatectomy possible. The patient had an uneventful postoperative course, and there were no signs of tumor recurrence for 8 months postoperatively [113].

Avoiding manipulation of the tumor-bearing liver appears to be the advantage conferred by the RAVAS and RAPID procedures. However, as seen in the past, heterotopic liver transplantation is much more prone to vascular complications, particularly with regard to outflow leading to Budd-Chiari syndrome [114]. Although 1 patient was successfully treated with the RAVAS method, that procedure is currently viewed as experimental.

Despite these encouraging preliminary reports on LT for MHCC, there are currently no data on long-term outcomes. Therefore, transplantation for MHCC cannot be considered a standard treatment, and remains restricted to strict criteria within clinical trials. However, ongoing clinical trials may contribute to improving knowledge of the role of TH in the treatment of MHCC.

The use of these complex techniques in staged liver surgery and transplantation, along with more and more targeted chemotherapy, have dramatically improved survival for a previously palliative condition. Despite this gain in overall survival, recurrence is a frequently observed phenomenon. In transplant patients, recurrence rates are as high as 90%. Similarly, complex staged resections, such as ALPPS, show recurrence rates of 71% [82].

Consistent evidence from both staged resection and LT demonstrated that tumor biology, as well as response to chemotherapy, are factors inherent to the tumor that drive aggressiveness. Among them are: N/KRAS mutational status, serum ACE levels, location of the primary tumor (right colon tumors have a more aggressive biology), and tumor dynamics in the chemotherapy-free interval [82,107,108]. These criteria should currently be considered in each type of staged liver resection and transplantation.

Knowledge of tumor biology and response to chemotherapy are also important factors in evaluating the treatment of recurrences. Similar to the initial treatment of MHCCs, the treatment of recurrences has changed dramatically from palliation to attempted cure.

At present, most surgeons would agree that patients with favorable tumor biology and good overall health should be eligible for the same aggressive treatment of recurrences as initially employed for MHCC. This includes the use of repeat surgery for liver recurrences, but also for lung metastases in selected cases.

Patients with a stable tumor situation, where resection is not possible from a technical point of view, can be evaluated for LT. Furthermore, local ablative therapies, including radiofrequency ablation, microwave ablation, and irreversible electroporation, have evolved and can be applied intraoperatively and percutaneously. Last but not least, the key to success is systemic control of the disease, using a wide selection of highly efficient chemotherapeutic agents.

Taking these developments together, MHCC and its recurrences can be treated with a multimodal armamentarium and can be viewed as a chronic disease rather than an instantly fatal disease, in patients with favorable tumor biology and good systemic control of the disease.

Globally, surgical evolution has undergone a long journey in the last century, from palliation to the consideration of living donor transplantation in highly selected patients. With significant advances in surgical techniques as well as systemic treatment, metastatic colorectal cancer has become a potentially curable disease. However, tumor biology and patient selection remain crucial in optimizing treatment for that patient cohort [115].

Future directions include further expansion of liver surgery, along with more effective and targeted therapies, making individualized treatment possible. Technically, liver surgery has reached a stage where RHF size is the only limitation to further drive resectability, when LT is not applicable. Future attempts, such as pharmacological induction of liver growth with non-carcinogenic compounds, ex situ expansion of liver tissue, and repopulation of acellular livers, are examples that may overcome too small a RHF.