WORKSHOP– PRIMÄR OCH SEKUNDÄR LEVERCANCER

Mikael Öman, Kir klin Norrlands Universitetssjukhus, Umeå

Två intensiva dagar inleddes med Tore Schersténs sammanfattning av leverkirurgins utveckling de senaste 100 åren. Yuman Fong och Bernard Nordlinger redovisade sina resultat inom traditionell leverkirurgi, lokal tumör destruktion samt vid kombinationsbehandling med kemoterapeutika. Resultaten inom Sverige redovisades och dessa befanns snarlika de internationella. Det görs relativt få leverresektioner i Sverige och en slutsats vid mötet var att fler patienter med kolorektala lever-metastaser borde dra nytta av leverkirurgi. Med de kriterier för resektabilitet som redovisades torde 20 % av patienterna med kolorektala levermetastaser bli aktuella för resektion. Om hälften av denna grupp blev opererade, skulle ungefär 300 patienter bli åtgärdade med leverresektion för kolorektala metastaser varje år. För närvarande opereras ungefär 100 patienter. Uppenbarligen kommer alltså inte alla patienter till "lever-kirurgen" för bedömning. En arbetsgrupp bildades med syftet att intensifiera samarbetet med kolorektala kirurger och med onkologer, som rimligtvis bör vara de som identifierar dessa patienter. Man enades om att gå med i en EORTC studie av resektabla levermetastaser samt planera en studie avseende kombinationsbehandling av primärt icke-resektabla levermetastaser.

Tillfälle till rekreation (reinkarnation?) gavs när hela sällskapet gav sig ut på forsränning på Vindelälven under mycket högt vattenstånd. Några deltagare var i stark närkontakt med det våta elementet, men kunde räddas till fortsatt medverkan i mötet. Lyckligtvis var deltagarantalet intakt även efter de värsta forsarna, varför det föreligger goda möjligheter att de föreslagna protokollen aktiveras. Nedan följer ett starkt förkortat sammandrag av de två dagarnas seminarium. Mötet avhandlades på engelska, varför nedanstående referat är detta språk.

Historical background

Tore Scherstén set off the meeting with a historical background. The evolvement of liver surgery during the last 100 years has been tremendous. The mortality and morbidity earlier was not acceptable and cure rates was unknown. Advances in anaesthesiology, physiology, microbiology and radiology have made it possible to perform liver surgery with good results.

In 1898, Cantlie made the first description of the middle hepatic vein. His name is still honoured in "Cantlies line". McIndoe and Councellour did in 1927 describe the right and left liver lobes, and in 1939, Ton That Tung explored the venous vascularisation and venous outflow. However, in 1951, Carl-Herman Hjortsjö did the most important contribution with his documentation of the intrahepatic biliary duct system and the vascular tree. Three years later, Couinaud described the segments of the liver, the surgical anatomy, which forms the basis of liver surgery.

The first planned hepatic resection where the patient survived, was performed by Langenbuch in Germany 1888. As early as 1899, Keen reported from US a series of 76 liver resections (37 liver tumours). In 1908 Pringle described a method to control bleeding during surgery with compression of the portal inflow, but all eight patients subjected to the procedure died. Wendle was in 1911 the first to do a successful right lobe hepatectomy using the achieved knowledge of vessels and with hilar ligation. On the 16^th October, 1951 Lortat-Jacob, using the principles of Hjortsjö and Couinaud, did the first anatomical resection. Lin advocated in 1958 the non-anatomical finger fracturing technique for resection. This technique was fast but bloody and the estimated blood loss for a lobectomy was 2000 ml.

In 1962 Pack documented liver-regeneration after liver resection, estimating three to six months to full regeneration. However, 10 years later, Blumgart made a resection after trauma and documented regeneration to full-size liver already after ten to eleven days. Pack and Miller did the first segmental resection in 1961, and this technique was refined by McBride in 1972. Bismuth introduced in 1982 the intraoperative ultrasonography for segmental resection. Pichlmayr described the ex situ tumour resection in 1990. In 1993, Wayand introduced the first laparoscopic lobe resections.

There are many resection-techniques with no one superior to another. Five year survival rates in the range of 30-50 % is mostly dependent on the selection of patients. Tore Scherstén concluded that only with adjuvant treatment the outcome of liver surgery for primary or metastatic livercancer will be possible to improve.

Liver resections – the French experience

Bernard Nordlinger showed in his results of all types of resection of colorectal liver metastases an overall mortality rate of 1-2 % for large resection and zero % for minor resections and a 5-year survival of 25%. His material on re-resections with 143 repeated resections performed within 4 to 88 months, mostly major resections (56%), had a median survival of 19 months and a 5-year survival rate of 34 %. That implies that patients with recurrent disease have the same chance for survival as patients with first time metastases. In dealing with selection of the candidates for liver resections, 1568 patients had several prognostic factors evaluated in a scoring system. The factors associated with poor survival are cancers having serosa involvement, a delay in resection, the number of metastases more than 5 cm, a clear resection margin and elevated CEA levels.

Liver resections – the US experience

Yuman Fong continued with the theme on how to analyse the patient. He uses the Clinical Risk Score (CRS), a decision score that is considering a tumour diameter more than 5 cm, tumour presence less than 12 months, extrahepatic growth and a CEA more than 200 ng/ml to a poor prognosis. Another way to stage patients is by using PET scanning. The theory is that a "hot" tumour on PET has worse prognosis. The 18F-FDG isotope analogue of glucose is directed toward cancer cells, with higher glucolytic activity and faster transport in the cell than normal cells. The isotope is transported into the cell but fosforylated inside the cell and unable to get out. The drawbacks of the method are too low sensitivity for extrahepatic disease and inability to detect lesions less than one cm. Furthermore, if chemotherapy has been given previously, the sensitivity of PET gets lower, then only ten percent of liver tumours less than one cm is actually identified up.

Concerning repeated liver resections, Yuman Fong reported good survival, however, the anatomy is tricky and resections are usually small. Only five percent become candidates for this. As an alternative, he advocated other ablative techniques, such as RF or laser, to be tested in protocols.

Primary hepatocellular cancer

In a study by Fong that recruited patients with primary hepatocellular cancer from 1991 to 1998, 412 patients were included of which 234 were explored. Of these, 154 had a resection, 5 got treated with infusion pump and 143 were subjected to RF ablation. Patients resected had a 30% 5-year survival (median 3 years). Patient treated with ablation had a 20 % 5-year survival (median 2 years). Explored patients having no other treatment had zero percent 5-year survival and a median survival of 1 year. Significant factors of survival was found AFP level (correlates with vascular invasion), Child class, Okuda stage and size more than five cm diameter. Age, cirrhosis type, gender and number of tumours had no significant impact on survival.

Systemic chemotherapy

Bengt Glimelius continued by giving an overview of current therapy of colorectal liver metastases, which he divided into three groups. Firstly, the few patients (less than ten percent) with a limited disease, tumours less than six cm in diameter and involving only a few segments suitable for surgery, do all have favourable prognosis. Secondly, the small group of patients with relatively limited disease that can turn resectable after systemic chemotherapy. Thirdly, all others (the majority) in whom cytostatic drugs prolong median survival but not cure rates.

Chemotherapy may make unresectable liver metastases resectable and possibly reduce the risk of recurrence. The rationale for this, is that about 50 % of recurrences occur in the liver, possibly due to dormant cancer cells in the remaining part of the other lobe, and that 5-FU in experimental studies has reduced the recurrence rates. Chemotherapy in advanced colorectal cancer has increased median survival from 5 to 14 months but the 5-year survival is still only two percent.

Regarding systemic chemotherapy, Bengt Glimelius and Bengt Gustavsson analysed the current status and concluded that for primary hepatocellular cancer there is no indication for chemotherapy. For cholangiocarcinoma, there is sometimes indication for chemotherapy with an estimated clinical benefit of about 30%. The drugs analysed are fluorouracil plus leucovorin (FU/LV), gemcitabine or oxaliplatin. For colorectal cancer , combination therapy with irinotecan plus FU/LV or oxaliplatin plus FU/LV has been proven effective and should be evaluated for treatment of liver metastases.

5-FU is one of oldest chemotherapeutic agents. It is a prodrug, which needs to be activated by enzymes. It works as an inhibitor of thymidilate-syntetase (TS). After a bolus dose of 5-FU, a great variation in plasma concentration occurs. By blocking the degradation, a high plasma concentration is achieved for 24 hrs. 5-FU is used in combination with Leucovorin, which increases the intracellular concentration of tetrahydrofolate. TS can be used as a marker for 5-FU response, low TS correlates to a response to 5-FU. New inhibitors of TS are raltitrexed (a folate-based inhibitor) and capecitabine (an oral fluroropyrimidine), both with the drawback of being more toxic than 5-FU.

Can more effective tumour destruction contribute to more cures, was the question Bengt Gustavssson posed. Two studies will try to answer that question. A Nordic protocol is proposed for non-resectable liver metastases of colorectal cancer. The patients will get combination-chemotherapy and the responders are randomised to surgical and/or local treatment (to accomplish radical tumour-destruction) or to non-surgical treatment. There is also an ongoing EORTC study of patients with resectable liver metastases from colorectal origin with pre and postoperative chemotherapy with oxaliplatin plus 5FU/LV versus surgery alone with a few Swedish centres recruiting patients.

Local tumour destruction

Kalle Tranberg introduced his presentation with the statement that too few patients with colorectal metastases are presented to liver surgeons. With our present criteria 20 % of patients with liver metastases from colorectal cancer should be candidates for liver resection. If only half of these are resectable, there should be performed 300 liver resections per year. However only 100 patients per year are resected in Sweden. Obviously, there is a large number of patients who never will be presented to a liver-surgeon.

How do we treat patients we cannot offer a traditional liver-resection? Minimally invasive treatment with local tumour destruction hypothetically has advantages in offering less immunosuppression, lower release of growth factors and minor shedding of tumour cells. There are interesting effects of local tumour destruction, with an induction of the immunologic response, described already in 1981 when cryotherapy in rats with subcutaneous tumours could reduce the risk of distant metastases.

The indications for local tumour destruction are tumours some distance away from the portal triad (because complications arise from the bile ducts), less than 4 tumours with a diameter of less than five cm, and no extrahepatic growth. There are several methods for local tumour destruction. Hyperthermic treatment is applied with a number of interventions such as laser (with the benefit of real time monitoring by MRI), high intensive ultrasound, radio frequency and microwave coagulation. Cryotherapy is less used now because it requires operation to locate the tumours accurately and moreover it has a rather high complication rate. We can also treat the tumour with chemical means such as injection of alcohol and acetic acid, and with irradiation, both internal and external.

The discussion following this presentation, emphasised that these therapies are experimental and should only be used in clinical protocols.

Regional chemotherapy

Lo Hafström continued with an updating analysis of regional chemotherapy. The liver is the most frequent location for spread disease with 20 % synchronous and 30 % metachronous metastases.

The rationale for regional therapy is the high clearance rate of 5-FU in the liver, 5-FU, the drug mostly used in regional chemotherapy of colorectal liver metastases, is extracted to 90% in the liver, leading to high local drug concentration.. There is a relation between response-rate and concentration of the drug in the target-organ. The systemic complication rate is very low.

Meta-analyses of six studies with regional infusion therapy, including 654 patients demonstrates that Hepatic Artery Infusion(HAI) gives a 41% and intravenous (IV) a 14% tumour response rate, and the median survival of 16 months in the HAI-group is better than in the IV-group (12 months). In a study by Hafström et al. Intra Portal Infusion (IPI) vs. Best Supportive Care (BSC) revealed a benefit for the treated group with a median survival of 17 months versus 8 months. We have compared HAI to IPI but closed the study prematurely because of a significantly better survival in the HAI group (15 months) versus IPI group (10 months). In this study, patients with extrahepatic cancer were included, explaining the somewhat shorter survival. A trial comparing the best regional therapy to the best systmei c therapy should be considered, Lo Hafström concluded.

Liver transplantation

Michael Olausson reviewed the field of liver transplantation. PBC, PSC, alcohol cirrhosis, hepatitis B and C and amyloidosis are the most common indications for liver transplantation. Controversial diagnoses for transplantation are hepatitis, primary liver cancer and metastases. In Sweden there are 12 donors per million and this is sufficient for the current transplantation-need. Other techniques are split liver (possibility to share organs), living donors (segment 2-3, adult to child), right lobe donation (adult to adult), portocaval hemi transplantation’s (if no splanchnic nor sufficient collateral veins are present) and auxiliary liver transplantation’s (using segment 2-3 temporarily to overcome the risk of hyperacute rejection of renal allograft).

The overall 5-year survival after liver transplantation is today 80%. Transplantation is the best treatment for alcohol cirrhosis with a 80% 5-year survival. Even though cholangiocellular carcinoma and primary hepatocellular carcinoma is a doubtful indication, the 5-year survival is 30%.

In the case of PHCC smaller than five cm in diameter, the long term results are in favour of transplantation. The results are even better with light cirrhosis, but if liver cancer exceeds five cm in diameter, resection is the choice.

Liver transplantation for liver metastases are not recommended , but an exception for neuroendocrine tumours was mentioned.