- Email: [email protected]
Liver transplantation for hepatocellular carcinoma
Liver Transplantation for Hepatocellular Carcinoma P. Neuhaus, S. Jonas, W.O. Bechstein, C. Wex, N. Kling, U. Settmacher, and H. Al-Abadi,
I
N THE THERAPY of primary hepatocellular carcinoma confined to the liver, treatment options aim at ablation of the tumor by surgical resection or destruction in situ. The various possible procedures are applied according to tumor extent and functional hepatic reserve; that is, tumor stage or number/diameter of nodules and stage of liver cirrhosis, respectively. However, TNM stages and UICC classification of hepatocellular carcinoma must be considered with caution for therapeutic decisionmaking, because they comprise biologically different tumors in common categories; for example, carcinomas with and without vascular invasion in T stages 1 to 3 or lymph node–positive and –negative tumors in UICC stage III. In hepatocellular carcinoma in cirrhosis, the evaluation of hepatic function generally follows the Child–Pugh classification, whereas the significance of the Okuda stages, which consider functional as well as oncologic (although not very specific) criteria has decreased over the past decade. A more profound understanding of the anatomic and physiologic bases of liver surgery, especially of segmental anatomy and hepatic regeneration, as well as a marked progress of anesthesiology and surgical technique, resulted in a substantial increase in total number and safety of hepatic resections. Various resective procedures with different degrees of segment- or lobectomies have represented the standard in the treatment of malignant neoplasia of the liver. Postoperative mortality as well as fatalities due to recurrence of the tumor and complications of cirrhosis have impaired outcome, especially in the treatment of hepatocellular carcinoma in cirrhosis. Other therapeutic practices, such as total hepatectomy and liver transplantation or percutaneous ethanol injection, have resulted in favorable survival figures in some series, and could at least emphasize the need for a revision of standardized therapeutic strategies.
there are now a variety of theoretical arguments in favor of total hepatectomy and liver transplantation as an oncologic therapeutic concept: 1. A total hepatectomy should always be considered as formally curative, unless the cancer has extended beyond the hepatic confines. In particular, atypical or wedge resections performed due to limited functional hepatic reserve frequently do not succeed in achieving sufficiently cleared resection margins. 2. Today, postoperative mortality after liver transplantation is low, or its potential can even be ignored, whereas it ranges from 3% to 15% after liver resections performed in patients suffering from cirrhosis. 3. Transplantation of the liver offers the advantage of simultaneous therapy for treatment of the primary disease. About 80% of all hepatocellular carcinomas develop within cirrhotic liver tissue, which is considered as a risk factor for malignant transformation. 4. An increased radicality and a simultaneous treatment of the primary disease will not only result in prevention of recurrent disease, but also in prevention of de novo hepatocellular carcinomas. 5. Decreased mortality due to liver cirrhosis, but also—as will be shown later—to oncologic reasons, is likely to be reflected by favorable survival rates over the long term. 6. A total hepatectomy offers the possibility of complete histopathologic staging for detection of multicentricity or small satellite nodules. In contrast, a staging based on radiomorphologic criteria (ie, computed tomography, lipiodole ultrasonography, angiography, or magnetic resonance tomography) will probably miss some smaller satellites in the remnant liver and generate false negative results.
TOTAL HEPATECTOMY AND LIVER TRANSPLANTATION
Conversely, some arguments may advocate against total hepatectomy and liver transplantation in the treatment of hepatocellular carcinoma in cirrhosis. Graft shortage and high costs represent more general problems, demonstrating the narrow limits that have to be considered even if indication for transplantation is justified. The progressive growth of the tumor during the waiting period is a more specific problem, particularly because neoadjuvant thera-
Liver transplantation for hepatocellular carcinoma and other malignancies carries the risk of early recurrence with usually poor overall 5-year survival rates of 30% to 40%, based on UNOS and ELTR data. In the early era of liver transplantation, the surgical strategy for the treatment of hepatocellular carcinoma in cirrhosis followed a misconception in selecting patients suffering from advanced and therefore irresectable cancers as transplant candidates. Although the original expectation of increased survival in patients suffering from advanced liver tumors did not hold true,
From the Clinic of General, Visceral, and Transplantation Surgery, Charite´, Medical Faculty of the Humboldt-University Berlin, Campus Virchow-Clinic, Berlin, Germany. Address reprint requests to Dr Peter Neuhaus, Clinic of General, Visceral and Transplantation Surgery, Charite´, Medical Faculty of the Humboldt-University of Berlin, Campus VirchowClinic, Augustenburger Platz 1, D-13353 Berlin, Germany.
© 1999 by Elsevier Science Inc. 655 Avenue of the Americas, New York, NY 10010
0041-1345/99/$–see front matter PII S0041-1345(99)01712-6 469
Transplantation Proceedings, 31, 469–471 (1999)
470
pies have not yet been shown to be effective. However, an uncertain prognosis over the long term due to a diverse morbidity and a hardly definable rate of recurrence is associated with the posttransplant immunosuppressive treatment. The increased risk of intrahepatic recurrence and limited survival under immunosuppression has been shown in a study by Yokoyama et al who found a much shorter tumor volume doubling time (TVDT) of recurrent hepatocellular carcinomas after transplantation than after resection (33 6 7 vs 274 6 79 days).1 This finding was supported in a detailed study by Panis et al who injected colorectal tumor cells into the portal vein of rats.2 Eight weeks after tumor cell injection only rats without liver tumors were selected for randomization into three groups. One served as a control group, another received cyclosporine, and in one a standardized liver resection was performed. The noteworthy results showed a significantly higher rate of liver tumors in the cyclosporine-treated group (80%) when compared with the control group (29%). Moreover, not only immunosuppression but also liver resection was associated with a significantly higher numbers of rats suffering from liver tumors (62%) in comparison to controls—an effect possibly due to the influence of cytokines and growth factors.
PATIENT SELECTION
Bismuth et al showed that total hepatectomy and liver transplantation restricted to small hepatocellular carcinomas with diameters of less than 3 cm, comprising only one or two nodules (ie, a group that, until that time had been thought to be the most suitable for resection), showed favorable outcome.3 The results of Bismuth’s group have been confirmed in another study of six French centers showing a 60% rate of 5-year disease-free survival in patients undergoing liver transplantation.4 This figure is significantly higher than the 14% 5-year disease-free survival rate in patients undergoing liver resection. Independent prognostic factors were again tumor size and the number of tumor nodules. In addition, vascular invasion could be identified as a prognostic factor in the study by Selby et al, where hepatocellular carcinomas of UICC stages I to III carried an acceptable 5-year survival of about 60%, whereas UICC stage IVa, reflecting vascular invasion and disseminated tumor nodules, showed significantly poorer prognosis (5-year survival of 11%).5 The results of the aforementioned studies do not originate from prospective, randomized trials. Therefore, the groups undergoing resection or transplantation are certainly not comparable, and it must be emphasized that liver transplantation should only be considered as a treatment option for a selected subset of patients.6 Based on previous studies, some transplant centers, including ours, selected patients for liver transplantation with tumors of less than 3 or 5 cm in cirrhotic livers and no signs of vascular invasion. The 5-year survival rates were
NEUHAUS, JONAS, BECHSTEIN ET AL
virtually the same as in patients undergoing liver transplantation for benign diseases. OUTCOME
Our experience after 1015 liver transplantations over a 10-year period consists of 73 patients suffering from a hepatocellular carcinoma in cirrhosis. During the same period, 63 patient underwent hepatic resection for hepatocellular carcinoma in cirrhosis and another 52 for a hepatocellular carcinoma without associated liver cirrhosis. The 5-year survival rates were 74% in patients undergoing liver transplantation and 31% in noncirrhotic patients, in whom a liver resection had been performed. In cirrhotic patients undergoing hepatic resections, the 4-year survival rate was approximately 25%. However, no 5-year survivors have been observed so far. Similar to reports of other groups, these are not the results of a prospective, randomized study. Therefore, the resection and transplantation groups are probably not comparable. An analysis of the different shares of UICC stages within the treatment groups discloses a high proportion of UICC I/II as well as IVa stages in the transplant population. Conversely, UICC stage III represents the largest percentage in both resection groups, whereas small tumors (UICC I/II) can almost exclusively be observed in patients suffering from liver cirrhosis. Considering the different causes of death, such as postoperative mortality, tumor recurrence, etc., liver transplantation compared favorably to the resection groups, irrespective of the association with liver cirrhosis. After liver transplantation, no operative fatalities occurred. Tumor recurrence and other causes of death were at least cut in half over the long term. The pattern of other causes of death was also different when comparing transplantation to resection. Recurrent hepatitis B and de novo malignancies were most eminent in eight patients dying after liver transplantation from causes other than tumor recurrence. Liver failure was the leading cause of death after hepatic resection, not only during the immediate postoperative course, but also later. The retrospective analysis of prognostic parameters within the selected transplant population did not reveal significant differences in regard to whether the tumors were solitary or multiple (5-year survival 80% vs 72%). A tumor size less than or exceeding 3 cm also did not impact survival in the patients of our study (5-year survival 71% vs 80%). Eighteen patients with hepatocellular carcinomas in cirrhosis, detected incidentally after transplantation, appeared to fare worse than patients with preoperatively known carcinomas. The respective 5-year survival figures were 59% compared with 84%. However, this difference did not reach statistical significance. The finding of poorer outcome after transplantation of incidental tumors is in contrast to previous reports from other groups. It should be noted that half of the incidentally detected hepatocellular carcinomas were indeed UICC stage I/II tumors, whereas the other half could be categorized as UICC stage IVa. Upon comparison of all patients undergoing liver trans-
LT FOR HEPATOCELLULAR CARCINOMA
plantation for hepatocellular carcinomas in cirrhosis of UICC stages I to III with those suffering from UICC stage IVa carcinomas, superior survival (P 5 .10) could be observed in the group with less advanced tumors. The respective 5-year survival figures were 83% vs 48%. Longterm survival of patients suffering from hepatocellular carcinomas in cirrhosis (UICC stages I to III) was comparable with the 5-year survival rate of patients who had undergone liver transplantation for benign indications (5year survival 86%). The favorable overall long-term prognosis is comparable to the results reported by other groups applying selection criteria that do not adhere stringently to the TNM classification, but to size and number of tumor nodules. Moreover, known vascular invasion and extrahepatic manifestations are considered contraindications to liver transplantation. Mazzaferro et al (Milan) and Llovet et al (Barcelona) reported outcomes of 48 and 58 patients, respectively. Their long-term survival rates were 75% and 74%.7,8 OUTLOOK
Additional prognostic parameters as well as multimodal treatment options are required in the future to improve results for the more advanced (eg, UICC stage IVa) tumors. A microscopic tumor cell dissemination has been investigated for many gastrointestinal malignancies, although not for hepatocellular carcinoma by bone marrow aspiration and immunocytochemical staining. Our preliminary findings, which had been obtained from a group of patients undergoing liver resection for hepatocellular carcinoma, did not give any conclusive results.9 However, patients suffering from associated liver cirrhosis appeared to have a lower frequency of bone marrow metastases than patients without associated liver cirrhosis. This finding merits further investigation. Other tools of interest for characterization of tumors prior to transplantation might be DNA histocytometry or single-cell photometry, whereas flow cytometry appears to be inaccurate in detecting DNA aneuploidy. In a series of 54 patients in whom hepatic resections were performed for hepatocellular carcinomas a significant correlation of postoperative survival and DNA ploidy was observed.10 All patients with an euploid tumor cell pattern survived 2 years or more, whereas the aneuploid group showed very poor short-term prognosis of less than 20%. Currently, a prospective investigation of core biopsies from hepatocellular carcinoma patients scheduled for transplantation in our clinic is underway. So far, multimodal treatment options in conjunction with liver transplantation have been confined to pilot trials involving transarterial chemoembolization and postoperative chemotherapy.11–13 Olthoff et al suggested doxoru-
471
bicine pre-, intra-, and postoperatively with a significant increase of survival when compared with historic controls.12 However, tumor stages were not well defined and 3-year survival rates did not compare favorably with those published in many series without adjuvant therapy. Prospective, randomized trials have not been published so far, although they are in progress at different centers. The Paris group conducted a retrospective analysis of their experience with transarterial chemoembolization (TACE) and failed to show a survival benefit in patients with pretransplant TACE.13 Interestingly, they could demonstrate significant survival when dividing patients treated with TACE into a group that responded with tumor necrosis to another group not developing necrosis. Therefore, it should be determined whether chemoembolization might serve as a selection criterion prior to liver transplantation. CONCLUSION
Total hepatectomy and liver transplantation allows patients suffering from small hepatocellular carcinoma in cirrhosis a more favorable prognosis than hepatic resection. This result reflects a lower rate of tumor recurrence as well as decreased mortality due to the underlying cirrhosis. A selection of patients based on tumor size and tumor stage is oncologically justified. However, vascular invasion is unpredictable and other selection criteria (eg, DNA ploidy) are required. Adjuvant treatment options may prove beneficial, but have not been investigated extensively or in prospective, randomized trials. The few studies published have not yet generated conclusive results. REFERENCES 1. Yokoyama I, Carr B, Saitsu H, et al: Cancer 68:2095, 1991 2. Panis Y, Ribeiro J, Chretien Y, et al: Br J Surg 79:221, 1992 3. Bismuth H, Chiche L, Adam R, et al: Ann Surg 218:145, 1993 4. Michel J, Suc B, Montpeyroux F, et al: J Hepatol 26:1274, 1997 5. Selby R, Kadry Z, Carr B, et al: World J Surg 19:53, 1995 6. Jonas S, Bechstein WO, Kling N, et al: Dtsch Med Wschr 122:617, 1997 7. Mazzaferro V, Regalia E, Doci R, et al: N Engl J Med 334:693, 1996 8. Llovet JM, Bruix J, Fuster J, et al: Hepatology 27:1572, 1998 9. Jonas S, Weinrich M, Steinbrich R, et al: Z Gastroenterol 36:78, 1998 10. Al-Abadi H, Jonas S, Bechstein WO, et al: Langenbecks Arch Chir (in press) 11. Stone MJ, Klintmalm BG, Polter D, et al: Gastroenterology 104:196, 1993 12. Olthoff KM, Rosove MH, Shackleton CR, et al: Ann Surg 221:734, 1995 13. Majno PE, Adam R, Bismuth H, et al: Ann Surg 226:4, 1997
I
N THE THERAPY of primary hepatocellular carcinoma confined to the liver, treatment options aim at ablation of the tumor by surgical resection or destruction in situ. The various possible procedures are applied according to tumor extent and functional hepatic reserve; that is, tumor stage or number/diameter of nodules and stage of liver cirrhosis, respectively. However, TNM stages and UICC classification of hepatocellular carcinoma must be considered with caution for therapeutic decisionmaking, because they comprise biologically different tumors in common categories; for example, carcinomas with and without vascular invasion in T stages 1 to 3 or lymph node–positive and –negative tumors in UICC stage III. In hepatocellular carcinoma in cirrhosis, the evaluation of hepatic function generally follows the Child–Pugh classification, whereas the significance of the Okuda stages, which consider functional as well as oncologic (although not very specific) criteria has decreased over the past decade. A more profound understanding of the anatomic and physiologic bases of liver surgery, especially of segmental anatomy and hepatic regeneration, as well as a marked progress of anesthesiology and surgical technique, resulted in a substantial increase in total number and safety of hepatic resections. Various resective procedures with different degrees of segment- or lobectomies have represented the standard in the treatment of malignant neoplasia of the liver. Postoperative mortality as well as fatalities due to recurrence of the tumor and complications of cirrhosis have impaired outcome, especially in the treatment of hepatocellular carcinoma in cirrhosis. Other therapeutic practices, such as total hepatectomy and liver transplantation or percutaneous ethanol injection, have resulted in favorable survival figures in some series, and could at least emphasize the need for a revision of standardized therapeutic strategies.
there are now a variety of theoretical arguments in favor of total hepatectomy and liver transplantation as an oncologic therapeutic concept: 1. A total hepatectomy should always be considered as formally curative, unless the cancer has extended beyond the hepatic confines. In particular, atypical or wedge resections performed due to limited functional hepatic reserve frequently do not succeed in achieving sufficiently cleared resection margins. 2. Today, postoperative mortality after liver transplantation is low, or its potential can even be ignored, whereas it ranges from 3% to 15% after liver resections performed in patients suffering from cirrhosis. 3. Transplantation of the liver offers the advantage of simultaneous therapy for treatment of the primary disease. About 80% of all hepatocellular carcinomas develop within cirrhotic liver tissue, which is considered as a risk factor for malignant transformation. 4. An increased radicality and a simultaneous treatment of the primary disease will not only result in prevention of recurrent disease, but also in prevention of de novo hepatocellular carcinomas. 5. Decreased mortality due to liver cirrhosis, but also—as will be shown later—to oncologic reasons, is likely to be reflected by favorable survival rates over the long term. 6. A total hepatectomy offers the possibility of complete histopathologic staging for detection of multicentricity or small satellite nodules. In contrast, a staging based on radiomorphologic criteria (ie, computed tomography, lipiodole ultrasonography, angiography, or magnetic resonance tomography) will probably miss some smaller satellites in the remnant liver and generate false negative results.
TOTAL HEPATECTOMY AND LIVER TRANSPLANTATION
Conversely, some arguments may advocate against total hepatectomy and liver transplantation in the treatment of hepatocellular carcinoma in cirrhosis. Graft shortage and high costs represent more general problems, demonstrating the narrow limits that have to be considered even if indication for transplantation is justified. The progressive growth of the tumor during the waiting period is a more specific problem, particularly because neoadjuvant thera-
Liver transplantation for hepatocellular carcinoma and other malignancies carries the risk of early recurrence with usually poor overall 5-year survival rates of 30% to 40%, based on UNOS and ELTR data. In the early era of liver transplantation, the surgical strategy for the treatment of hepatocellular carcinoma in cirrhosis followed a misconception in selecting patients suffering from advanced and therefore irresectable cancers as transplant candidates. Although the original expectation of increased survival in patients suffering from advanced liver tumors did not hold true,
From the Clinic of General, Visceral, and Transplantation Surgery, Charite´, Medical Faculty of the Humboldt-University Berlin, Campus Virchow-Clinic, Berlin, Germany. Address reprint requests to Dr Peter Neuhaus, Clinic of General, Visceral and Transplantation Surgery, Charite´, Medical Faculty of the Humboldt-University of Berlin, Campus VirchowClinic, Augustenburger Platz 1, D-13353 Berlin, Germany.
© 1999 by Elsevier Science Inc. 655 Avenue of the Americas, New York, NY 10010
0041-1345/99/$–see front matter PII S0041-1345(99)01712-6 469
Transplantation Proceedings, 31, 469–471 (1999)
470
pies have not yet been shown to be effective. However, an uncertain prognosis over the long term due to a diverse morbidity and a hardly definable rate of recurrence is associated with the posttransplant immunosuppressive treatment. The increased risk of intrahepatic recurrence and limited survival under immunosuppression has been shown in a study by Yokoyama et al who found a much shorter tumor volume doubling time (TVDT) of recurrent hepatocellular carcinomas after transplantation than after resection (33 6 7 vs 274 6 79 days).1 This finding was supported in a detailed study by Panis et al who injected colorectal tumor cells into the portal vein of rats.2 Eight weeks after tumor cell injection only rats without liver tumors were selected for randomization into three groups. One served as a control group, another received cyclosporine, and in one a standardized liver resection was performed. The noteworthy results showed a significantly higher rate of liver tumors in the cyclosporine-treated group (80%) when compared with the control group (29%). Moreover, not only immunosuppression but also liver resection was associated with a significantly higher numbers of rats suffering from liver tumors (62%) in comparison to controls—an effect possibly due to the influence of cytokines and growth factors.
PATIENT SELECTION
Bismuth et al showed that total hepatectomy and liver transplantation restricted to small hepatocellular carcinomas with diameters of less than 3 cm, comprising only one or two nodules (ie, a group that, until that time had been thought to be the most suitable for resection), showed favorable outcome.3 The results of Bismuth’s group have been confirmed in another study of six French centers showing a 60% rate of 5-year disease-free survival in patients undergoing liver transplantation.4 This figure is significantly higher than the 14% 5-year disease-free survival rate in patients undergoing liver resection. Independent prognostic factors were again tumor size and the number of tumor nodules. In addition, vascular invasion could be identified as a prognostic factor in the study by Selby et al, where hepatocellular carcinomas of UICC stages I to III carried an acceptable 5-year survival of about 60%, whereas UICC stage IVa, reflecting vascular invasion and disseminated tumor nodules, showed significantly poorer prognosis (5-year survival of 11%).5 The results of the aforementioned studies do not originate from prospective, randomized trials. Therefore, the groups undergoing resection or transplantation are certainly not comparable, and it must be emphasized that liver transplantation should only be considered as a treatment option for a selected subset of patients.6 Based on previous studies, some transplant centers, including ours, selected patients for liver transplantation with tumors of less than 3 or 5 cm in cirrhotic livers and no signs of vascular invasion. The 5-year survival rates were
NEUHAUS, JONAS, BECHSTEIN ET AL
virtually the same as in patients undergoing liver transplantation for benign diseases. OUTCOME
Our experience after 1015 liver transplantations over a 10-year period consists of 73 patients suffering from a hepatocellular carcinoma in cirrhosis. During the same period, 63 patient underwent hepatic resection for hepatocellular carcinoma in cirrhosis and another 52 for a hepatocellular carcinoma without associated liver cirrhosis. The 5-year survival rates were 74% in patients undergoing liver transplantation and 31% in noncirrhotic patients, in whom a liver resection had been performed. In cirrhotic patients undergoing hepatic resections, the 4-year survival rate was approximately 25%. However, no 5-year survivors have been observed so far. Similar to reports of other groups, these are not the results of a prospective, randomized study. Therefore, the resection and transplantation groups are probably not comparable. An analysis of the different shares of UICC stages within the treatment groups discloses a high proportion of UICC I/II as well as IVa stages in the transplant population. Conversely, UICC stage III represents the largest percentage in both resection groups, whereas small tumors (UICC I/II) can almost exclusively be observed in patients suffering from liver cirrhosis. Considering the different causes of death, such as postoperative mortality, tumor recurrence, etc., liver transplantation compared favorably to the resection groups, irrespective of the association with liver cirrhosis. After liver transplantation, no operative fatalities occurred. Tumor recurrence and other causes of death were at least cut in half over the long term. The pattern of other causes of death was also different when comparing transplantation to resection. Recurrent hepatitis B and de novo malignancies were most eminent in eight patients dying after liver transplantation from causes other than tumor recurrence. Liver failure was the leading cause of death after hepatic resection, not only during the immediate postoperative course, but also later. The retrospective analysis of prognostic parameters within the selected transplant population did not reveal significant differences in regard to whether the tumors were solitary or multiple (5-year survival 80% vs 72%). A tumor size less than or exceeding 3 cm also did not impact survival in the patients of our study (5-year survival 71% vs 80%). Eighteen patients with hepatocellular carcinomas in cirrhosis, detected incidentally after transplantation, appeared to fare worse than patients with preoperatively known carcinomas. The respective 5-year survival figures were 59% compared with 84%. However, this difference did not reach statistical significance. The finding of poorer outcome after transplantation of incidental tumors is in contrast to previous reports from other groups. It should be noted that half of the incidentally detected hepatocellular carcinomas were indeed UICC stage I/II tumors, whereas the other half could be categorized as UICC stage IVa. Upon comparison of all patients undergoing liver trans-
LT FOR HEPATOCELLULAR CARCINOMA
plantation for hepatocellular carcinomas in cirrhosis of UICC stages I to III with those suffering from UICC stage IVa carcinomas, superior survival (P 5 .10) could be observed in the group with less advanced tumors. The respective 5-year survival figures were 83% vs 48%. Longterm survival of patients suffering from hepatocellular carcinomas in cirrhosis (UICC stages I to III) was comparable with the 5-year survival rate of patients who had undergone liver transplantation for benign indications (5year survival 86%). The favorable overall long-term prognosis is comparable to the results reported by other groups applying selection criteria that do not adhere stringently to the TNM classification, but to size and number of tumor nodules. Moreover, known vascular invasion and extrahepatic manifestations are considered contraindications to liver transplantation. Mazzaferro et al (Milan) and Llovet et al (Barcelona) reported outcomes of 48 and 58 patients, respectively. Their long-term survival rates were 75% and 74%.7,8 OUTLOOK
Additional prognostic parameters as well as multimodal treatment options are required in the future to improve results for the more advanced (eg, UICC stage IVa) tumors. A microscopic tumor cell dissemination has been investigated for many gastrointestinal malignancies, although not for hepatocellular carcinoma by bone marrow aspiration and immunocytochemical staining. Our preliminary findings, which had been obtained from a group of patients undergoing liver resection for hepatocellular carcinoma, did not give any conclusive results.9 However, patients suffering from associated liver cirrhosis appeared to have a lower frequency of bone marrow metastases than patients without associated liver cirrhosis. This finding merits further investigation. Other tools of interest for characterization of tumors prior to transplantation might be DNA histocytometry or single-cell photometry, whereas flow cytometry appears to be inaccurate in detecting DNA aneuploidy. In a series of 54 patients in whom hepatic resections were performed for hepatocellular carcinomas a significant correlation of postoperative survival and DNA ploidy was observed.10 All patients with an euploid tumor cell pattern survived 2 years or more, whereas the aneuploid group showed very poor short-term prognosis of less than 20%. Currently, a prospective investigation of core biopsies from hepatocellular carcinoma patients scheduled for transplantation in our clinic is underway. So far, multimodal treatment options in conjunction with liver transplantation have been confined to pilot trials involving transarterial chemoembolization and postoperative chemotherapy.11–13 Olthoff et al suggested doxoru-
471
bicine pre-, intra-, and postoperatively with a significant increase of survival when compared with historic controls.12 However, tumor stages were not well defined and 3-year survival rates did not compare favorably with those published in many series without adjuvant therapy. Prospective, randomized trials have not been published so far, although they are in progress at different centers. The Paris group conducted a retrospective analysis of their experience with transarterial chemoembolization (TACE) and failed to show a survival benefit in patients with pretransplant TACE.13 Interestingly, they could demonstrate significant survival when dividing patients treated with TACE into a group that responded with tumor necrosis to another group not developing necrosis. Therefore, it should be determined whether chemoembolization might serve as a selection criterion prior to liver transplantation. CONCLUSION
Total hepatectomy and liver transplantation allows patients suffering from small hepatocellular carcinoma in cirrhosis a more favorable prognosis than hepatic resection. This result reflects a lower rate of tumor recurrence as well as decreased mortality due to the underlying cirrhosis. A selection of patients based on tumor size and tumor stage is oncologically justified. However, vascular invasion is unpredictable and other selection criteria (eg, DNA ploidy) are required. Adjuvant treatment options may prove beneficial, but have not been investigated extensively or in prospective, randomized trials. The few studies published have not yet generated conclusive results. REFERENCES 1. Yokoyama I, Carr B, Saitsu H, et al: Cancer 68:2095, 1991 2. Panis Y, Ribeiro J, Chretien Y, et al: Br J Surg 79:221, 1992 3. Bismuth H, Chiche L, Adam R, et al: Ann Surg 218:145, 1993 4. Michel J, Suc B, Montpeyroux F, et al: J Hepatol 26:1274, 1997 5. Selby R, Kadry Z, Carr B, et al: World J Surg 19:53, 1995 6. Jonas S, Bechstein WO, Kling N, et al: Dtsch Med Wschr 122:617, 1997 7. Mazzaferro V, Regalia E, Doci R, et al: N Engl J Med 334:693, 1996 8. Llovet JM, Bruix J, Fuster J, et al: Hepatology 27:1572, 1998 9. Jonas S, Weinrich M, Steinbrich R, et al: Z Gastroenterol 36:78, 1998 10. Al-Abadi H, Jonas S, Bechstein WO, et al: Langenbecks Arch Chir (in press) 11. Stone MJ, Klintmalm BG, Polter D, et al: Gastroenterology 104:196, 1993 12. Olthoff KM, Rosove MH, Shackleton CR, et al: Ann Surg 221:734, 1995 13. Majno PE, Adam R, Bismuth H, et al: Ann Surg 226:4, 1997