Monday, July 21, 2008

Treatment of Cholangiocarcinoma

Preoperative preparation
Jaundiced patients require vitamin K.
Resective procedures for cholangiocarcinomas are major procedures and, therefore, patients should have good functional status especially if they are elderly.

Preoperative biliary decompression
In general, it is preferable to avoid stents, if possible. Many surgeons find the presence of any biliary stent a hindrance to determining the proximal tumor extent intraoperatively. On the other hand, cholestasis, liver dysfunction, and biliary cirrhosis develop rapidly with unrelieved obstruction. The extent of liver dysfunction is one of the main factors that increase postoperative morbidity and mortality following surgical resection.
  • The chance of postoperative liver failure may be lessened by preoperative decompression, especially decompression of the side to be retained.
  • Decompression has the dual purpose of allowing that side to recover function and actually to hypertrophy.
  • On the other hand, stents may introduce bacteria and cause cholangitis.
  • Selective percutaneous decompression is an accepted strategy in Japan and multiple stents are often inserted.

A reasonable strategy is to proceed to surgery in younger patients (age less than 70 years) without serious comorbidities who have been jaundiced for less than 4 weeks, whose serum bilirubin is less than 10 mg/dL and whose future remnant liver will be > 40% of total liver mass.
  • Also, such patients should not have had biliary instrumentation, which always contaminates the obstructed biliary tract.
  • For the remainder, we routinely decompress the side of the liver to be retained and wait until the serum bilirubin falls to 3 mg/dL.

Preoperative portal vein embolization

  • Because the achievement of histologically negative resection margins is so critical to outcome, preoperative portal vein embolization (PVE) has been used in an attempt to increase the limits of safe resection.
  • The intent of PVE is to induce lobar hypertrophy in patients who have a predicted postoperative liver remnant volume of <25>
  • By allowing a larger resection volume to be carried out safely, PVE may permit a margin-negative resection in patients who otherwise would be considered unresectable because of concerns about insufficient postoperative residual liver volume.

Hilar cholangiocarcinoma

Criteria for resectability — The traditional guidelines for resectability of cholangiocarcinoma in the United States include :

  • Absence of retropancreatic and paraceliac nodal metastases or distant liver metastases
  • Absence of invasion of the portal vein or main hepatic artery (although some centers support en bloc resection with vascular reconstruction)
  • Absence of extrahepatic adjacent organ invasion
  • Absence of disseminated disease
The goals of surgical resection are to remove the tumour with negative resection margins and to perform a portal and coeliac node dissection.
  • In order to achieve this, hemihepatectomy with resection of the caudate lobe as well as the suprapancreatic extrahepatic bile duct and the portal and coeliac lymph nodes is required.
  • Resection of the caudate lobe is needed because, when the tumour is Bismuth 2 or greater, the orifices of the caudate ducts may be involved with tumour.
  • Hemihepatectomy is required when the tumour involves the sectional ducts (Bismuth 3) on one side of the liver or the artery or portal vein to one side of the liver.
  • Theoretically, resection of a Bismuth 2 tumour could be accomplished by resection of bile ducts with the caudate lobe but, in practice, negative margins are much more likely to be accomplished by a hemihepatectomy with resection of the caudate lobe.
  • Therefore, surgical practice has evolved in the past 20 years to the point at which there is a consensus that virtually all hilar cholangiocarcinomas should be resected by hemihepatectomy and caudate resection.
  • Low-lying Bismuth 1 tumours, in which a negative upper margin can be obtained on the common hepatic duct, i.e. below the level of the orifices of the caudate ducts, may be resected without removing liver tissue, but these instances are uncommon.
Unfortunately, cholangiocarcinomas have a well-known propensity to spread microscopically for long distances along the bile duct.
  • A positive resection margin occurs in 20–40% of cases even when liver resection has been done.
  • The resection margin may be extended by dividing through the termination of the sectional ducts, but there is a limit to how far this can be carried.
  • Mobilizing the portal vein in the umbilical fissure aids this manoeuvre on the left as the bile ducts to the left lateral section pass behind this vein.
  • The operation can also be extended by taking the left medial section with the right liver (right trisectionectomy) or the right anterior section with the left liver (left trisectionectomy).
  • The operation may also be extended by vascular resections.
    • Involvement of the portal vein may require replacement of the anterior wall of this structure or circumferential resection with anastomosis of the main portal vein to the left portal vein or, less commonly, the right portal vein with or without an interposition graft.
  • It has been proposed to remove the anterior wall of the portal vein as a routine in order to obtain a higher rate of negative margins.
  • This recommendation lacks supporting randomized data and is not routinely followed.
  • Arterial reconstruction of the proper hepatic artery is less commonly performed.
  • Pancreatoduodenectomy is performed in up to 15% of cases in some series in order to obtain negative lower margins or to remove potentially malignant nodes.
  • Reconstruction of the biliary tract is by a Roux-en-Y hepaticojejunostomy.
  • The Roux limb is made at least 60 cm in length to avoid reflux of intestinal contents into the biliary tree.
The important trends are as follows.
  • Portal vein embolization is becoming more common as a preoperative preparation.
  • Some 15–50% of patients are found to be unresectable at surgery but, in recent years, the figure has been decreasing towards the lower limit of this range.
  • More procedures include a major hepatic resection because the chance of achieving an R0 resection is substantially higher when a major resection is performed.
  • Mortality rates are falling.

Morbidity rates are still high, reflecting the magnitude of the procedure.
R0 resection is necessary for cure; almost no patients who have had R1 resection survive 5 years without recurrence.
Even with R0 resection, the overall 5-year survival is only 20–35%.
Factors affecting survival include
  • R0 resection,
  • en bloc hepatectomy,
  • absence of tumour in lymph nodes,
  • tumour grade and tumour type with IG type (papillary) having better survival than other types.
The need to perform a vascular resection is associated with a poorer outcome in some series. The value of pancreatoduodenectomy is unclear at this time. Most patients requiring this addition to the procedure have had a relatively short survival.

Lower duct tumours
The rationale and extent of the procedure is the same as that used for pancreatic carcinoma.
Distal lesions are usually treated with pancreaticoduodenectomy (Whipple procedure).
  • Often, the tissue of origin of the tumour, i.e. whether pancreatic, ampullary or bile duct in origin, is uncertain until after the specimen is examined pathologically and, even then, doubt can remain.
  • The main complication of this procedure is a fistula from the pancreatic–jejunal anastomosis, which occurs in 5–10% of patients.
  • Biliary fistulas occur in about 2% of patients.
  • Patients rarely die from these complications today because of improvements in diagnostic and interventional radiology, intensive care and treatment of infection.
Resection of part of the stomach is no longer required for lower duct cholangiocarcinoma, although there is little or no difference in short-term outcome or quality of life between the pyloruspreserving and standard types of pancreaticoduodenectomy.
  • Many patients require pancreatic enzyme replacement after this procedure, but few become diabetic.
Five-year overall survival in recent case series varies from 16% to 37%, and results have not improved in the last decade. Overall, the results of pancreatoduodenectomy for lower bile duct cancer are much the same as those for adenocarcinoma of the pancreas and unlike those for ampullary and duodenal cancers, which have a much better prognosis.

Intrahepatic cholangiocarcinomas
The principles of treatment are as for other malignant intrahepatic hepatic lesions.
  • The tumour must be resected with a margin of normal tissue to obtain microscopically free resection margins (a 1-cm tumour-free resection margin is the goal), yet leave enough normally functioning liver tissue behind for the patient to have adequate liver function in the postoperative period.
  • The size of the resection may vary from a single segment or less to resection of three of the four hepatic sections.
  • The role of resection of extrahepatic lymph nodes is unclear, but it is being done with increasing frequency. Lymph node positivity is common in tumours over 4.5 cm in size, but rare in patients with smaller tumours, and lymphadenectomy might reasonably be limited to large tumours.
Four recent series of patients with MF intrahepatic cholangiocarcinoma treated by surgical resection in 60, 35, 52 and 104 patients reported 5-year survival rates of 29%, 33%, 34%
and 10% respectively.
Multivariate analyses of prognostic risk factors identified -
  • symptomatic patient,
  • positive surgical margin,
  • multiple tumours,
  • vascular invasion,
  • lymph node metastases
  • high microvessel counts indicative of angiogenesis as predictors of poor outcome.
  • Mucin (MUC)4 expression in carcinoma tissues is also associated with poor outcome in MF intrahepatic cholangiocarcinoma.
Fewer data are available for the less common IG and PI types of intrahepatic cholangiocarcinoma.
  • The IG type has the best prognosis.
Liver transplantation
Liver transplantation has been performed for intrahepatic and upper duct cholangiocarcinoma, but the results have been disappointing until recently.
  • In one series published in 1993,only 3 of 14 patients (21%) lived more than 28 months after the procedure.
  • Even in a more recent series, in which transplantation was performed in patients with cholangiocarcinoma in the setting of PSC, the 5-year survival was only 35%.
  • While these are not poor results for a visceral cancer, they must be evaluated with the knowledge that many patients with endstage chronic liver disease are dying while on a waiting list for liver transplantation, and the comparative survival rate in this group of patients would be expected to be about 90%.
  • However, two American centres have reported better results for hilar cholangiocarcinoma.
The Mayo Clinic group reported on 56 patients who entered a trial of liver transplantation for unresectable cholangiocarcinoma or cholangiocarcinoma in the setting of PSC.
  • The patients were staged by EUS and also by staging laparotomy.
  • All patients received neoadjuvant chemoradiation.
  • Approximately half the patients were transplanted, with a 5-year actuarial survival of 82% in transplanted patients.
  • The other group using a similar approach achieved a 5-year survival rate of 50%.
Adjuvant chemotherapy or radiotherapy
One randomized controlled trial of adjuvant therapy using mitomycin C and 5-fluorouracil vs. surgery alone enrolled 139 patients with cholangiocarcinoma at various levels. There was no survival benefit, although there was benefit in a group of patients with gallbladder cancer treated with the same regimen. External beam radiotherapy was associated with improved postoperative survival in patients with hilar cholangiocarcinoma in one non-randomized trial. However, patients in the control group were treated at an earlier time period or tended to be in poor general condition. In an earlier study in which groups of patients with hilar cholangiocarcinoma were more comparable, external beam radiation did not result in
improved survival. Intraluminal brachytherapy does not extend survival and is associated with an increased incidence of complications.

Recurrence of cholangiocarcinoma is often local, suggesting that adjuvant chemotherapy and/or radiation would be beneficial but, currently, there is no evidence to support its use outside clinical trials.

The purpose of palliation is to relieve jaundice and pruritus and to extend life.
  • Jaundice and pruritus are treated by stenting, usually by endoscopic or percutaneous means or, less commonly, by surgery.
  • Several randomized trials of surgical bypass vs. endoscopic intubation for lower duct tumours have been published.
  • The three earlier trials favoured endoscopic stenting.
  • Two of these trials were quite small and, in the third, the surgical procedures were performed by registrars; surgical complication rates were unusually high by current standards.
  • The reintervention rate was high in the stent group.
  • There was no difference in survival.
  • In a more recent trial of patients found to be unresectable by staging laparoscopy, surgical bypass was favoured over stenting with the surgical group achieving a longer survival.
The current surgical consensus is that a surgical bypass should be performed by an experienced HPB surgeon in younger patients without distant metastases or obvious peritoneal disease.
  • When unresectability is determined at laparotomy, a double bypass is performed to decompress the biliary tree and bypass the duodenum, which may become obstructed.
  • Surgical decompression of hilar tumours may also be accomplished by bypass to the segment 3 duct, the so-called Bismuth–Corlette procedure.
  • This provides decompression of the left liver.
  • It is usually only done when a patient is found to be unresectable at the time of exploration.
Decompression with stents may be achieved by endoscopic or percutaneous means, with the former favoured because it is less invasive.
  • Metal stents are generally used because of improved stent patency rates.
  • In the hilum, plastic stents also have the theoretical disadvantage that their solid walls may block smaller side-branches.
  • For lesions below the bifurcation, a single stent will decompress the entire liver.
  • Bilateral stents or forked stents have been used to decompress tumours obstructing the confluence.
  • However, it seems that a unilateral stent is often just as effective provided that it is placed selectively, based upon a preprocedure MRCP, on the side of the liver that will result in decompression of the most functional liver.
  • It is essential that the side of the liver not to be stented is not cannulated as that often results in cholangitis on that side, and this is associated with much poorer results.
  • Should inadvertent cannulation of the side of the liver occur on the side that was not intended to be decompressed, then it is best to employ bilateral stents in order to avoid cholangitis.
Intraluminal brachytherapy has not improved results and, as reported in the adjuvant setting, is associated with more complications.

Photodynamic therapy — Photodynamic therapy (PDT) involves the injection of an intravenous porphyrin photosensitizer followed by the endoscopic application of light (of a specific wavelength) to the tumor bed. The interaction between light and the photoagent causes tumor cell death, presumably by the generation of oxygen free radicals.

  • Initial uncontrolled series suggested that in addition to facilitating biliary decompression in patients with locally advanced disease, that survival might be improved in patients who underwent PDT.
  • It is thought that the survival benefit is related to prolonged relief of obstruction rather than to any reduction in tumor mass.
  • PDT is now being studied preoperatively as a means of improving the likelihood of achieving a margin-negative resection. Unfortunately, treatment is not widely available.

No comments: