Surgical therapy for primary liver tumours – resection and transplantation
The two most common malignant liver tumours are hepatocellular carcinoma (HCC) and intrahepatic cholangiocellular carcinoma (intrahepatic cholangiocellular adenocarcinoma [iCCA]). Primary liver tumours can in principle arise from all histogenetic cellular elements present in the liver. HCC is a primary liver-specific tumour with hepatocytic differentiation; the cell of origin is the hepatocyte at various stages of differentiation, whereas iCCA is defined as an intrahepatically located malignancy with biliary differentiation [1]. Intrahepatic cholangiocarcinomas can arise in any part of the intrahepatic bile duct, from segmental bile ducts to the smallest ramifications of the bile ducts and tubular structures [2].
HCCs are relatively rare but, due to their historically relatively poor prognosis, are among the ten most common causes of cancer-related death [3]. In Germany, around 8,300 new cases occur per year, with an almost equal number of deaths. The mean age at onset is 70 years in men and 74 years in women [1].
According to large statistics, iCCA account for approximately 5–10 % of all primary malignant liver tumours, with the prevalence showing a wide geographical variation. High-incidence areas lie in Asia, where the endemic occurrence of aetiologically relevant worms (Opisthorchis viverrini and Clonorchis sinensis) is documented. In the United States, which is classified as a low-incidence area, 1–2 new cases/100,000 inhabitants are assumed (these figures could be the same as ours). The average age of patients with iCCA is 55 years. Men are generally affected more frequently than women. The incidence of iCCA is steadily increasing [1].
Surgical therapy for iCCA
iCCA frequently remain asymptomatic for a long time and therefore usually reach a considerable size and extent by the time of diagnosis [4, 5, 6]. A distinction is made between the two rare periductal-infiltrating and intraductal-growing forms and the more common mass-forming type [7]. The two first-mentioned variants usually become apparent through a congestion of the downstream bile ducts (jaundice), whereas the mass-forming type clinically usually presents as a large solitary or multifocal, often confluent mass that frequently has contact with large vessels or infiltrates them. Cholestasis with jaundice is rare and usually results from compression of the hepatic bifurcation, less commonly from direct tumour infiltration [4, 5, 8, 9].
At present, radical surgical removal of the tumour tissue represents the only curative treatment for iCCA. Since iCCA predominantly arise in a non-cirrhotic liver, the required extensive resections are often feasible. Across all tumour stages, 5-year survival rates between 21 and 45 % are achieved after R0 resection [5, 6, 9, 10, 11, 12]. Distant metastasis, multifocality, lymph node metastases and vascular invasion are important prognostically relevant factors after R0 resection of an iCCA [6, 9, 12, 13, 14].
Analogous to other gastrointestinal tumour diseases, surgical therapy for iCCA is increasingly embedded in multimodal concepts. In the BILCAP study ("Capecitabine compared with observation in resected biliary tract cancer"), a median survival of 53 months was found after resection + adjuvant therapy compared with 36 months after surgical therapy alone [15]. Adjuvant chemotherapy with capecitabine is currently regarded as the standard. Data are also available on the efficacy of a neoadjuvant therapy. In a French multicentre analysis, comparable results were found after secondary resection of initially unresectable or borderline-resectable iCCA to those after resection of initially resectable iCCA, but markedly better results than after systemic chemotherapy alone [11]. Further studies also suggest that even an R1 resection in combination with subsequent chemotherapy can achieve superior survival data compared with chemotherapy alone. The indication for a palliative resection (debulking, R2 resection) is given only in individual cases [5, 6, 10, 16].
In the case of isolated intrahepatic tumour recurrences after potentially curative resection (usually within the first 2 years in about half of the patients), local ablative or surgical treatment has increasingly been used in recent years as an alternative to systemic therapy. This has made it possible to achieve results that are comparable with primary resection [10, 17].
Liver transplantation (LTX) currently has only a minor role in iCCA. In patients with a so-called "very early stage iCCA" (solitary iCCA < 2 cm), 5-year survival rates of 65 % can be achieved [18]. In the current guidelines of the German Medical Association on organ transplantation, LTX in patients with a CCA is recommended only within the framework of clinical studies [19].
Surgical therapy for HCC
In over 85 % of cases, HCC arise in a cirrhotic liver. The presence and degree of cirrhosis as well as its underlying disease are decisive for diagnosis, therapy and prognosis [8, 16, 20].
The German HCC guideline, currently undergoing revision, as well as the European guideline recommend LT as the therapy of choice for HCC in cirrhosis [16, 20]. It is contraindicated in extrahepatic tumour manifestations as well as in infiltration of large hepatic vessels detectable on imaging [19].
With still compensated liver function, resection represents an alternative to transplantation. However, owing to the limited functional reserve, more extensive resections are only rarely possible. Resections in liver cirrhosis are furthermore associated with an increased perioperative risk, which is particularly heightened by portal hypertension. Studies suggest that morbidity can be reduced by a laparoscopic approach [21]. Analyses report 5-year survival rates after resection of small and solitary HCCs of between 30 and 55%, and in highly selected subgroups even up to over 75% [22, 23, 24]. Depending on the cause of the liver cirrhosis, the recurrence risk within 5 years after resection is 60–80%, as resection cures the HCC lesion but not the underlying disease (hence also the recommendation for LT). An effective adjuvant therapy after R0 resection for HCC in cirrhosis has so far been largely lacking. For neoadjuvant therapy there is as yet no solid evidence [25]. Recurrences after HCC resection in cirrhosis are frequently confined to the liver, so that in these cases a repeat resection or an LT can be performed on an individual basis. 5-year survival rates of 60% for repeat resection and of up to 80% after salvage LT are reported [20, 26, 27].
For HCC in a non-cirrhotic liver, resection represents the therapy of choice. In stage M1, there is usually no longer an indication for resection; isolated pulmonary or adrenal metastases may, where appropriate, be an exception. After R0 resection of an HCC in a non-cirrhotic liver, 5-year survival rates between 26 and 60% are described [8].
Despite potentially curative resection, tumour recurrence also occurs in HCC in more than half of patients within the first 2 years. Mostly, multifocal intrahepatic or also combined intra- and extrahepatic recurrences are found. Isolated intrahepatic tumour recurrences that can be resected again are rare. In cases of unresectability of locally advanced tumours or tumour recurrences, an LT can also be considered for HCC in a non-cirrhotic liver. A European multicentre analysis with over 100 patients showed, in these cases after LT, a 5-year survival of 49% and a disease-free survival of 43% [28].
Technical aspects in the surgery of iCCA and HCC
In resections of primary liver tumours, all technical possibilities of liver surgery are used, taking into account the residual functional capacity of the liver [5, 6, 8, 9, 10, 12]. In a healthy liver, up to 80% of the liver volume can be resected.
The standard procedure for HCC and iCCA in a non-cirrhotic liver is anatomical liver resection. In addition to the classic (extended) hemihepatectomy, segmentectomies and less widespread techniques such as sectorectomies or mesohepatectomies can also be considered for smaller tumours [5, 6, 7, 8, 9, 10, 11, 12]. The aim is R0 resection while maintaining a safety margin, for the size of which, however, no solid evidence exists.
Particularly in non-cirrhotic parenchyma, the liver can be conditioned for resection, whereby hypertrophy rates of the FLR (functional liver remnant volume) of up to 40% can be achieved. Common techniques are portal vein embolisation and, rarely, operative portal vein ligation [29]. The technique of the “in-situ split” or “associating liver partition and portal vein ligation for staged hepatectomy” (ALPPS), which enables volume increases of the FLR of 60 to over 100% within a week, is performed only very cautiously because of the sometimes considerable perioperative complication rate [30].
In addition to extensive liver resections, vascular resections and, particularly in iCCA, bile duct resections are frequently indicated [5, 6, 8, 12, 31, 32]. Provided that an overall curative concept appears achievable, these technical extensions of the operation nowadays no longer represent a contraindication to resection, but do entail a not inconsiderable perioperative mortality of up to 10%.
In HCC in cirrhosis, predominantly small resections and segmentectomies, more rarely also hemihepatectomies, are performed [22, 23, 24].
In primary liver carcinomas, lymphadenectomy is primarily of diagnostic-prognostic significance. In resectable HCC, the incidence of lymph node metastasis is overall low at 5–10%, with HCC in cirrhosis metastasising lymphatically less frequently [8, 13, 14]. In iCCA, the incidence of lymph node metastases, at approximately 20–40%, is considerably higher than in HCC.
For the resection of hepatobiliary tumours, lymphadenectomy is predominantly required, although the underlying data situation for this is weak. Since the lymphatic drainage of the liver is very complex and variable, a standardisation of lymphadenectomy for liver tumours has so far also been lacking. Various studies have shown that the detection of lymph node metastases has a negative influence on the prognosis of tumours of the liver (including liver metastases), the pancreas, and the biliary tract [13, 33]. In addition to the total number of involved lymph nodes, the ratio of involved to examined lymph nodes (lymph node ratio) also appears to possess prognostic significance, at least in intrahepatic (ICC) and perihilar cholangiocarcinoma (PHCC, Klatskin).
Nevertheless, it is consensus that the liver hilum should always be included in a lymphadenectomy. However, there is uncertainty about how "aggressively" the dissection of lymph and connective tissue in the liver hilum should be performed.
In liver cirrhosis, hilar lymphadenectomy is associated with increased morbidity: venous bleeding due to portal congestion, frequent respiratory or cardiovascular complications, wound and intra-abdominal infections [13]. Occasionally, pronounced lymph leaks or massive ascites development occur postoperatively. Due to the various postoperative risks, a very precise weighing of the benefits and risks of lymphadenectomy is necessary in liver cirrhosis.