- Email: [email protected]
Management of primary liver cell carcinoma
Management of Primary Liver Cell Carcinoma M.Balasegaram,
MB, FRCS, FRCS (Eng), FRACS, FACS, Kuala Lumpur, Malaysia
Primary liver cell carcinoma is very common in Malaysia, constituting about 4 per cent of all malignant lesions reported at the Institute of Medical Research, Kuala Lumpur, Malaysia [I]. Although the population of Malaysia is multiracial with 50 per cent Malays, 38 per cent Chinese, 10 per cent Indians, and the remaining 2 per cent indigenous, the Chinese have a very high incidence (68 per cent) of primary liver cell carcinoma. The causative factor is unknown, although 85.4 per cent of these patients also have cirrhosis of the liver. The daily intake of aflatoxins has been shown to be related to cancer of the liver in Thailand [2] and Kenya [3]. Available evidence does not eliminate the possibility that mycotoxins are also involved in the causes of primary liver cell carcinoma in Malaysia. Research on this aspect is currently being conducted in the Second Surgical Division of the General Hospital, Kuala Lumpur. Moreover, Australian hepatitis-B antigen has been found in 31.7 per cent of our patients of all races with liver cell carcinoma, although the incidence in routine blood donors is about 8 per cent. This is a significant finding, but there is insufficient evidence to assess its role in this disease. Further research on the cause of primary liver cell carcinoma is being carried out in various parts of Africa and Asia including Malaysia. In our study, 40 per cent of the patients with cirrhosis of the liver were alcoholics and 20 per cent were severely malnourished. From the Department of Surgery. General Hospital, Kuala Lumpur, Malaysia. Reprint requests should be addressed to Professor M. Balasegaram, No. 5, Jalan Liew Weng Chee, off Jalan Yap Kwan Seng. Kuala Lumpur. 04-06, Malaysia.
Volume 130, July 1975
Clinical Data Between January 1964 and July 1974, 352 patients with primary liver cell carcinoma were treated. These patients were referred from various parts of Malaysia. The ratio of males to females was 4:l. Their ages ranged from one to eighty-six years, the vast majority of the patients being between forty-five and sixty years. Fortytwo patients were admitted with an acute abdomen after rupture of the hepatoma. Fifteen of these patients died before any effective treatment could be carried out. Thirty-eight patients were admitted with bleeding esophageal or fundic varices associated with cirrhosis of the liver, and fifty-two patients were admitted with end stage liver failure. Diagnosis. Apart from the history and physical findings, the diagnostic aids used included hepatic scanning with indium 113 M using blood pool studies, celiac and splenic macroaggregate studies with iodine 131-tagged albumin, peritoneoscopy, angiography, and specific enzyme investigations. Among the specific serologic tests, measurement of serum alpha-fetoproteins in conjunction with proline hydroxylase has allowed preoperative detection in 94.6 per cent of our patients with primary liver cell carcinoma. Contrary to other reported series [4] we have had no false-positive results in the 126 patients who underwent this test. We have also found that after complete resection of a hepatic tumor, the serum alpha-fetoprotein level becomes negative; however, it remains positive if the tumor is incompletely removed. Furthermore, follow-up studies have shown that a return to positive of the alpha-fetoprotein level signifies recurrence of the tumor. Based on this experience, we believe that these two serologic tests will eventually not only prove to be a good screening test, but also prove to be effective in evaluating the extent of the tumor. All these investigations have allowed preoperative diagnosis of the nature, location, and extent of hepatic tu-
33
Balasegaram
TABLE
I
Types of Hepatic
Resection Performed
Operation
Left lobectomy Left hemihepatectomy Extended left hemihepatectomy Right lobectomy Right hemihepatectomy Extended right hemihepatectomy Right hemihepatectomy and left lobectomy Total
*. . 4 2 2
.. ..
7
1
9
2
2 26
1 4 (15.3 per cent)
mars [5], thus eliminating the need for exploratory laparotomy. These investigations have also established a more rational approach to the management of primary liver cell carcinoma. Treatment and Results No Treatment. One hundred thirty-nine patients had no form of treatment. Thirty-four patients left the hospital against medical advice before any treatment could be begun. Thirty-seven of the thirty-eight patients admitted with coexisting cirrhosis and bleeding varices died of hemorrhage. The remaining sixty-seven patients were admitted with end stage disease, fifteen of whom had ruptured hepatoma. Despite resuscitation with blood, plasma, or other intravenous fluids, all died within two weeks of admission. Curative Resection. Although complete resection of the involved lobe is the ideal form of treatment, because of the extensive involvement of both lobes of the liver and the multicentric characteristics of the tumor, the indications for resection in our patients with primary liver cell carcinoma are very limited. This finding is in keeping with those of other series[6,7]. Severe impairment of liver function with marked jaundice and significant ascites in addition to associated cirrhosis pre-
TABLE
II
Results of Resection Result
Postoperative death Lost to follow-up study Died within 1 year Died within 2 to 3years Died within 3 to 4 years Alive 2 to 3 years Alive 5 years Total
34
Number of Patients 4 1 6 4 2 4 5 26
eluded any form of curative surgery [8]. Fewer than 7 per cent of the 352 hepatomas were resectable. Lin [9], however, reported a rate of exploration of 75.2 per cent and of resectability, 34.7 per cent. The presence of infiltration or thrombosis of the portal vein and inferior vena cava as shown by angiographic studies was a further contraindication to operation. Tables I and II show the extent of resection, survival, and mortality rate of the patients who underwent resection. Palliative Treatment. Since the remaining 93 per cent of the 352 hepatomas were nonresectable, various forms of palliative treatment were attempted because the average period of survival of untreated patients was between two and four months from the time the patient was first seen. After treatment, these patients were assessed as follows: (1) by objective measurement of changes that occurred on physical examination of tumor size, liver scanning, needle biopsy of the liver, angiography, and serial determinations of liver function; (2) by subjective evidence of improvement, including an increase in weight with a return of appetite, relief of abdominal pain, and a sense of well-being; (3) by autopsy study, which indicated that most of the tumor that responded to treatment had decreased in size or showed necrotic changes, although microscopic evidence of viable tumor tissue was still present. Regional chemotherapy: Twenty-five patients had regional chemotherapeutic infusions of methotrexate or 5-fluorouracil via the hepatic or the right epiploic artery using the standard dosage [IO]. Serious complications developed, including leakage, displacement of the catheter, and hemorrhage, in addition to the severe side-effects of the drugs, especially bone marrow depression. Most of these patients died within one to three months. This form of treatment has since been abandoned. Similar poor results have been reported in Asian patients [7,11,12], although Watkins [13] and Sullivan [14] and their co-workers have reported reasonably good results in their American series. Complete hepatic dearterialization: Complete hepatic dearterialization, including ligation of all the collateral and accessory arterial supplies from the diaphragm, falciform ligament, abdominal wall, and adjacent organs, was performed in sixty patients in the three years from 1970 to 1972. This procedure is based on the principle that because a liver tumor derives its blood supply almost exclusively from the hepatic artery and its collaterals, necrosis of tumor tissue occurs after dearterialization [15]. On the basis of previous experience [16],
The American Journal of Surgery
Liver Cell Carcinoma
our indications for such a procedure were: (1) absence of infiltration or thrombosis of the portal vein; (2) absence of severe jaundice or marked ascites; (3) presence of normal hepatic tissue between areas of multicentric tumor. Although complete hepatic dearterialization has a definite place in the palliative treatment of primary liver cell carcinoma when these indications are strictly observed [16] (Table III), the disadvantages of this procedure are that preoperative hepatic angiography is of vital importance and liga.tion of all collaterals is necessary, which was often difficult, incomplete, and time-consuming. Most of the adjacent organs were extremely vascular and often not easily identified. It also was subsequently found that collateral circulation and an increase in portahepatic anastomosis eventually developed. Our experience also showed that although necrosis and regression of the tumor occurred, viable malignant cells were always present, which eventually caused recurrence. Furthermore, this procedure was feasible in only 12.6 per cent of the patients in this study. Multiple dose therapy: Because resection and complete hepatic dearterialization were contraindicated, sixty patients with advanced hepatomas received multiple dose therapy [17] between the years 1972 and 1973, Our regimen of multiple dose therapy for advanced carcinoma of the liver normally consisted of a combination of three of the four drugs in average dosages as follows: (1) intravenous methotrexate, 0.5 mg per kg of body weight, twice weekly in ten doses; (2) cyclophosphamide, 5 mg per kg of body weight, twice weekly in ten doses; (3) intravenous vincristine, 0.035 mg per kg of body weight, twice weekly in ten doses; (4) 5fluorouraci1, 5 mg per kg of body weight, daily in ten doses. Of the fifty-six patients, twenty-eight had methvincristine, and cyclophosphamide, otrexate, whereas the others had methotrexate, cyclophosphamide, and 5-fluorouracil. Only seven of the sixty patients survived eight to twelve months. Seven others lived four to eight months, whereas fourteen patients died within four to six months. Twenty patients survived less than four months. Three patients died during treatment and the rest died within one month. (Table IV.) However, subjective improvement was noted in twenty patients, including an increase in weight, return of appetite, improved sense of well-being, and more importantly, return to work. Because of the severe sideeffects, such as hemorrhage, alopecia, skin changes and soreness of the mouth, drug cost, and the rela-
Volume 130, July 1975
TABLE
III
Results of Complete Dearterialization
Period of Survival
Number of Patients ___. 4 8
10 days to 4 weeks 6 weeks 4 to 6 months 7 to 12 months
14 15
12 to 18 months
11
18 to 24 months
8
Hepatic
Remarks ~~___ _ One died of hemorrhage Three died of thrombosis of the portal vein Died of liver failure Died of tumor recurrence and metastasis Died of tumor recurrence and metastasis Died of tumor recurrence and metastasis
60
Total
tively poor results in our series, this form of treatment has been discontinued. Current therapy: Based on the experience in this series, our current therapy for primary liver cell carcinoma is as follows: (1) When the tumor is confined to one lobe and there is no evidence of liver failure, jaundice, or marked cirrhosis, resection of the involved lobe or extended hemihepatectomy is performed. (2) In the presence of multicentric tumor, only ligation of the hepatic artery is carried out at the level of the gastroduodenal junction, if no portal vein thrombosis, cirrhosis, severe jaundice, or liver failure is present. A week after ligation of the hepatic artery, chemotherapy is begun. The chemotherapy we currently use consists of a combination of 200 mg of 5-fluorouracil and 40 mg of methotrexate intravenously in a saline drip biweekly. After healing of the operative wound, the patient is subjected to supervoltage radiotherapy, and a total dosage of about 2,500 to 3,000 r is administered in a period of four weeks. It must be stressed that supportive therapy, especially administration of essential amino acids and large doses of vitamins, are of vital importance during treatment. Simultaneously, immunotherapy, using bacillus Calmette-G&in (6,000,OOO organisms) is adminis-
TABLE _____
IV
Results
of Multiple _
Period of Survival Died during therapy 20 to 30 days 1 to 4 months 4 to 6 months 6 to 8 months 8 to 10 months Total
(10 days)
Dose Therapy Number of Patients 4 8 20 14 7
7 60
Remarks No response No response No response No response Partial response Partial response
35
Balasegaram
TABLE V
Results of Current
Survival
__ ___ Died during treatment Died 1 to 4 months Alive 4 to 6 months Alive 6 to 8 months Total
Therapy
Number of Patients ~~ 2 7 18 15 42 .-___.
Remarks .~ ~~~____~
. No response Good response Good response
tered subcutaneously in divided doses for five weeks. Its value has, however, not yet been established. Whenever hepatic arterial ligation is not feasible, for reasons mentioned earlier, a combination of chemotherapy (5fluorouracil and methotrexate), radiotherapy, and immunotherapy is the treatment of choice. Thus far, forty-two patients have had this form of treatment. Results are listed in Table V. Although the prescribed dosage of radiotherapy was 5,500 r because the hepatic artery had not been ligated, in all cases the radiotherapist assessed the dosage for each patient to obtain a balance between necrosis of the tumor and total liver failure. Remarkable objective and subjective response occurred in fifteen patients, with almost complete regression of the tumor. However, this form of treatment cannot yet be assessed adequately since it was started in early 1974. Comments
Based on results of this study, it is obvious that the current management of primary liver cell carcinoma is far from adequate. A review of our current form of treatment after a period of two years is planned. It is interesting to note that of the 352 patients, 139 or 42.8 per cent had no treatment. Reported series in most countries in Africa and Asia have shown that the rate of resectability for primary liver cell carcinoma is relatively low and the long-term survival is disappointingly poor. Western investigators [12,18,19], however, have reported fairly good long-term survival with subjective improvement after complete hepatic dearterialization. Hepatic arterial ligation followed by regional infusion of chemotherapeutic drugs through the distal end of the hepatic artery has been reported, especially in the United States and Europe [12,13]. Our results, however, have been disappointing. Several factors, including diet, advanced stage of the disease, resistance of the host, and parasitic infection, may contribute to this disparity. Moreover, it is well-known that Asian patients have a poor prognosis whatever form of treatment for pri-
mary liver cell carcinoma is carried out [2,18,20]. The technical problems of hepatic transplantation largely have been overcome so that this could be the ultimate solution. However, problems with immunologic factors, tissue rejection, and organ preservation, including government legislation, must be solved before hepatic transplantation can become a routine procedure. It is likely that in countries such as Malaysia, where hepatic carcinoma is common, hepatic transplantation will not be feasible because of the excessive number of patients who would require this treatment, in addition to the problems of the costs involved and the training of personnel. It must be remembered that in these developing countries other important health problems take priority. Needless to say, our attention should be directed toward detecting the definite causative factors of hepatic carcinoma. More importantly, development of a good screening test to detect early carcinoma of the liver could be an alternate answer. Summary
A review of 352 patients with primary liver cell carcinoma treated by the author is presented. The poor rate of resectability (7 per cent) has necessitated various forms of treatment over the years. These are described in detail. Based on this experience, the current form of treatment for nonresectable carcinoma is summarized. Although it is too early to assess this form of treatment, initial results appear to be promising. A second report in the near future is planned. Acknowledgment: I wish to thank the DirectorGeneral, Ministry of Health, Federation of Malaysia, for permission to publish this paper and Mrs Jessie Cardoza for typing this manuscript. Addendum
Since this report was prepared, a total of ninetyeight patients have been treated by the current therapy. More recently, bacillus Calmette-Guerin (30,000,OOO organisms) is injected directly into the tumor tissue at operation in addition to subcutaneous therapy. Silver clips are then inserted at the sites of the multicentric tumor to assess regression of the tumor and allow administration of selective radiotherapy to the malignant cells. The Silver clips also permit percutaneous transhepatic injection of bacillus Calmette-Guerin postoperatively daily for four days, the dosage being 20,000,OOO organisms.
The
American Journal of Surgery
Liver Cell Carcinoma
References II. 1. Balasegaram M: Hepatic resection. Br J Surg 55: 168, 1968. 2. Plengvanit V, Limwonges K, Viranuvatti V, Hiianant S, Chearanai 0: Treatment of primary carcinoma of the liver by hepatic artery ligation. Preliminary report of 40 cases. Liver Research, p 490. Antwerp, Belgium, Tijdschrift voor Gastroenterologie, 1967. 3. Bagshawe A: Symposium in the International Liver Conference in South Africa, January 23-26, 1973. Personal communication. 4. Purves LR, Bersohn El, Geddes EW: Serum alpha-feto proteins and primary cancer of the liver in man. Cancer 25: 1261. 1970. 5. Balasegaram M: Hepatic surgery: present and future. Ann R Co/l Surg Engl47: 139, 1970. 6. Hsia SS, Tseng HH. Wu HK, Huang HC: Appraisal on liver resection in the treatment of primary hepatic carcinoma. Chin Med J 8: 248, 1965. 7. Tung TT: Personal communication, Hanoi, 197 I. 8. Balasegaram M: Hepatic surgery: a review of a personal series of 95 major resections. Aust NZ J Surg 42: 1, 1972. 9. Lin TY: Primary cancer of the liver. Stand J Gasfroenterol (Sup@) 6: 223, 1970. 10. Lange M, Falkson G, Geddes EW: Hepatic artery catheteri-
Volume 130, July 1975
12. 13.
14.
15. 16. 17.
18.
19. 20.
sation and infusion chemotherapy for primary liver cell carcinoma. In press. Chan KT: Management of primary liver carcinoma. Ann R CollSurgEngl41: 253. 1967. Foster JH: Survival after liver resection for cancer. Cancer 26: 493, 1970. Watkins E Jr, Khazel AH, Nahra KS: Surgical basis for arterial infusion chemotherapy of disseminated carcinoma of the liver. Surg Gynecol Obstet 138: 580. 1970. Sullivan RD, Watkins E Jr, Norcross JW:‘Chemotherapy of liver cancer by prolonged hepatic artery infusion. N Engl J Med270: 321, 1964. Breedis C, Young G: The blood supply of neoplasms in the liver. Am J Pathol30: 969, 1954. Balasegaram M: Complete hepatic dearterialization for primary carcinoma of the liver. Am J Surg 124: 340, 1972. Tucker WG, Talley RW, Brown Lee RW, Burrows JH, Stott PB, Moorhead EL, Sandiego EL: Preliminary trial with combination of cyclophosphamide, vincristine. 5-fluorouracil. Cancer Chemother Reo 52: 593, 1968. Bengmark ST, Almersjo b, Engevik L, Hafstrom 0: Surgical management of tumours of the liver. Chirurg 39: 320, 1968. Fortner J: Personal communication, 1974. Mori M. Masuda M, Miyanaga T: Hepatic artery ligation and tumor necrosis in the liver. Surgery 59: 359, 1966.
37
MB, FRCS, FRCS (Eng), FRACS, FACS, Kuala Lumpur, Malaysia
Primary liver cell carcinoma is very common in Malaysia, constituting about 4 per cent of all malignant lesions reported at the Institute of Medical Research, Kuala Lumpur, Malaysia [I]. Although the population of Malaysia is multiracial with 50 per cent Malays, 38 per cent Chinese, 10 per cent Indians, and the remaining 2 per cent indigenous, the Chinese have a very high incidence (68 per cent) of primary liver cell carcinoma. The causative factor is unknown, although 85.4 per cent of these patients also have cirrhosis of the liver. The daily intake of aflatoxins has been shown to be related to cancer of the liver in Thailand [2] and Kenya [3]. Available evidence does not eliminate the possibility that mycotoxins are also involved in the causes of primary liver cell carcinoma in Malaysia. Research on this aspect is currently being conducted in the Second Surgical Division of the General Hospital, Kuala Lumpur. Moreover, Australian hepatitis-B antigen has been found in 31.7 per cent of our patients of all races with liver cell carcinoma, although the incidence in routine blood donors is about 8 per cent. This is a significant finding, but there is insufficient evidence to assess its role in this disease. Further research on the cause of primary liver cell carcinoma is being carried out in various parts of Africa and Asia including Malaysia. In our study, 40 per cent of the patients with cirrhosis of the liver were alcoholics and 20 per cent were severely malnourished. From the Department of Surgery. General Hospital, Kuala Lumpur, Malaysia. Reprint requests should be addressed to Professor M. Balasegaram, No. 5, Jalan Liew Weng Chee, off Jalan Yap Kwan Seng. Kuala Lumpur. 04-06, Malaysia.
Volume 130, July 1975
Clinical Data Between January 1964 and July 1974, 352 patients with primary liver cell carcinoma were treated. These patients were referred from various parts of Malaysia. The ratio of males to females was 4:l. Their ages ranged from one to eighty-six years, the vast majority of the patients being between forty-five and sixty years. Fortytwo patients were admitted with an acute abdomen after rupture of the hepatoma. Fifteen of these patients died before any effective treatment could be carried out. Thirty-eight patients were admitted with bleeding esophageal or fundic varices associated with cirrhosis of the liver, and fifty-two patients were admitted with end stage liver failure. Diagnosis. Apart from the history and physical findings, the diagnostic aids used included hepatic scanning with indium 113 M using blood pool studies, celiac and splenic macroaggregate studies with iodine 131-tagged albumin, peritoneoscopy, angiography, and specific enzyme investigations. Among the specific serologic tests, measurement of serum alpha-fetoproteins in conjunction with proline hydroxylase has allowed preoperative detection in 94.6 per cent of our patients with primary liver cell carcinoma. Contrary to other reported series [4] we have had no false-positive results in the 126 patients who underwent this test. We have also found that after complete resection of a hepatic tumor, the serum alpha-fetoprotein level becomes negative; however, it remains positive if the tumor is incompletely removed. Furthermore, follow-up studies have shown that a return to positive of the alpha-fetoprotein level signifies recurrence of the tumor. Based on this experience, we believe that these two serologic tests will eventually not only prove to be a good screening test, but also prove to be effective in evaluating the extent of the tumor. All these investigations have allowed preoperative diagnosis of the nature, location, and extent of hepatic tu-
33
Balasegaram
TABLE
I
Types of Hepatic
Resection Performed
Operation
Left lobectomy Left hemihepatectomy Extended left hemihepatectomy Right lobectomy Right hemihepatectomy Extended right hemihepatectomy Right hemihepatectomy and left lobectomy Total
*. . 4 2 2
.. ..
7
1
9
2
2 26
1 4 (15.3 per cent)
mars [5], thus eliminating the need for exploratory laparotomy. These investigations have also established a more rational approach to the management of primary liver cell carcinoma. Treatment and Results No Treatment. One hundred thirty-nine patients had no form of treatment. Thirty-four patients left the hospital against medical advice before any treatment could be begun. Thirty-seven of the thirty-eight patients admitted with coexisting cirrhosis and bleeding varices died of hemorrhage. The remaining sixty-seven patients were admitted with end stage disease, fifteen of whom had ruptured hepatoma. Despite resuscitation with blood, plasma, or other intravenous fluids, all died within two weeks of admission. Curative Resection. Although complete resection of the involved lobe is the ideal form of treatment, because of the extensive involvement of both lobes of the liver and the multicentric characteristics of the tumor, the indications for resection in our patients with primary liver cell carcinoma are very limited. This finding is in keeping with those of other series[6,7]. Severe impairment of liver function with marked jaundice and significant ascites in addition to associated cirrhosis pre-
TABLE
II
Results of Resection Result
Postoperative death Lost to follow-up study Died within 1 year Died within 2 to 3years Died within 3 to 4 years Alive 2 to 3 years Alive 5 years Total
34
Number of Patients 4 1 6 4 2 4 5 26
eluded any form of curative surgery [8]. Fewer than 7 per cent of the 352 hepatomas were resectable. Lin [9], however, reported a rate of exploration of 75.2 per cent and of resectability, 34.7 per cent. The presence of infiltration or thrombosis of the portal vein and inferior vena cava as shown by angiographic studies was a further contraindication to operation. Tables I and II show the extent of resection, survival, and mortality rate of the patients who underwent resection. Palliative Treatment. Since the remaining 93 per cent of the 352 hepatomas were nonresectable, various forms of palliative treatment were attempted because the average period of survival of untreated patients was between two and four months from the time the patient was first seen. After treatment, these patients were assessed as follows: (1) by objective measurement of changes that occurred on physical examination of tumor size, liver scanning, needle biopsy of the liver, angiography, and serial determinations of liver function; (2) by subjective evidence of improvement, including an increase in weight with a return of appetite, relief of abdominal pain, and a sense of well-being; (3) by autopsy study, which indicated that most of the tumor that responded to treatment had decreased in size or showed necrotic changes, although microscopic evidence of viable tumor tissue was still present. Regional chemotherapy: Twenty-five patients had regional chemotherapeutic infusions of methotrexate or 5-fluorouracil via the hepatic or the right epiploic artery using the standard dosage [IO]. Serious complications developed, including leakage, displacement of the catheter, and hemorrhage, in addition to the severe side-effects of the drugs, especially bone marrow depression. Most of these patients died within one to three months. This form of treatment has since been abandoned. Similar poor results have been reported in Asian patients [7,11,12], although Watkins [13] and Sullivan [14] and their co-workers have reported reasonably good results in their American series. Complete hepatic dearterialization: Complete hepatic dearterialization, including ligation of all the collateral and accessory arterial supplies from the diaphragm, falciform ligament, abdominal wall, and adjacent organs, was performed in sixty patients in the three years from 1970 to 1972. This procedure is based on the principle that because a liver tumor derives its blood supply almost exclusively from the hepatic artery and its collaterals, necrosis of tumor tissue occurs after dearterialization [15]. On the basis of previous experience [16],
The American Journal of Surgery
Liver Cell Carcinoma
our indications for such a procedure were: (1) absence of infiltration or thrombosis of the portal vein; (2) absence of severe jaundice or marked ascites; (3) presence of normal hepatic tissue between areas of multicentric tumor. Although complete hepatic dearterialization has a definite place in the palliative treatment of primary liver cell carcinoma when these indications are strictly observed [16] (Table III), the disadvantages of this procedure are that preoperative hepatic angiography is of vital importance and liga.tion of all collaterals is necessary, which was often difficult, incomplete, and time-consuming. Most of the adjacent organs were extremely vascular and often not easily identified. It also was subsequently found that collateral circulation and an increase in portahepatic anastomosis eventually developed. Our experience also showed that although necrosis and regression of the tumor occurred, viable malignant cells were always present, which eventually caused recurrence. Furthermore, this procedure was feasible in only 12.6 per cent of the patients in this study. Multiple dose therapy: Because resection and complete hepatic dearterialization were contraindicated, sixty patients with advanced hepatomas received multiple dose therapy [17] between the years 1972 and 1973, Our regimen of multiple dose therapy for advanced carcinoma of the liver normally consisted of a combination of three of the four drugs in average dosages as follows: (1) intravenous methotrexate, 0.5 mg per kg of body weight, twice weekly in ten doses; (2) cyclophosphamide, 5 mg per kg of body weight, twice weekly in ten doses; (3) intravenous vincristine, 0.035 mg per kg of body weight, twice weekly in ten doses; (4) 5fluorouraci1, 5 mg per kg of body weight, daily in ten doses. Of the fifty-six patients, twenty-eight had methvincristine, and cyclophosphamide, otrexate, whereas the others had methotrexate, cyclophosphamide, and 5-fluorouracil. Only seven of the sixty patients survived eight to twelve months. Seven others lived four to eight months, whereas fourteen patients died within four to six months. Twenty patients survived less than four months. Three patients died during treatment and the rest died within one month. (Table IV.) However, subjective improvement was noted in twenty patients, including an increase in weight, return of appetite, improved sense of well-being, and more importantly, return to work. Because of the severe sideeffects, such as hemorrhage, alopecia, skin changes and soreness of the mouth, drug cost, and the rela-
Volume 130, July 1975
TABLE
III
Results of Complete Dearterialization
Period of Survival
Number of Patients ___. 4 8
10 days to 4 weeks 6 weeks 4 to 6 months 7 to 12 months
14 15
12 to 18 months
11
18 to 24 months
8
Hepatic
Remarks ~~___ _ One died of hemorrhage Three died of thrombosis of the portal vein Died of liver failure Died of tumor recurrence and metastasis Died of tumor recurrence and metastasis Died of tumor recurrence and metastasis
60
Total
tively poor results in our series, this form of treatment has been discontinued. Current therapy: Based on the experience in this series, our current therapy for primary liver cell carcinoma is as follows: (1) When the tumor is confined to one lobe and there is no evidence of liver failure, jaundice, or marked cirrhosis, resection of the involved lobe or extended hemihepatectomy is performed. (2) In the presence of multicentric tumor, only ligation of the hepatic artery is carried out at the level of the gastroduodenal junction, if no portal vein thrombosis, cirrhosis, severe jaundice, or liver failure is present. A week after ligation of the hepatic artery, chemotherapy is begun. The chemotherapy we currently use consists of a combination of 200 mg of 5-fluorouracil and 40 mg of methotrexate intravenously in a saline drip biweekly. After healing of the operative wound, the patient is subjected to supervoltage radiotherapy, and a total dosage of about 2,500 to 3,000 r is administered in a period of four weeks. It must be stressed that supportive therapy, especially administration of essential amino acids and large doses of vitamins, are of vital importance during treatment. Simultaneously, immunotherapy, using bacillus Calmette-G&in (6,000,OOO organisms) is adminis-
TABLE _____
IV
Results
of Multiple _
Period of Survival Died during therapy 20 to 30 days 1 to 4 months 4 to 6 months 6 to 8 months 8 to 10 months Total
(10 days)
Dose Therapy Number of Patients 4 8 20 14 7
7 60
Remarks No response No response No response No response Partial response Partial response
35
Balasegaram
TABLE V
Results of Current
Survival
__ ___ Died during treatment Died 1 to 4 months Alive 4 to 6 months Alive 6 to 8 months Total
Therapy
Number of Patients ~~ 2 7 18 15 42 .-___.
Remarks .~ ~~~____~
. No response Good response Good response
tered subcutaneously in divided doses for five weeks. Its value has, however, not yet been established. Whenever hepatic arterial ligation is not feasible, for reasons mentioned earlier, a combination of chemotherapy (5fluorouracil and methotrexate), radiotherapy, and immunotherapy is the treatment of choice. Thus far, forty-two patients have had this form of treatment. Results are listed in Table V. Although the prescribed dosage of radiotherapy was 5,500 r because the hepatic artery had not been ligated, in all cases the radiotherapist assessed the dosage for each patient to obtain a balance between necrosis of the tumor and total liver failure. Remarkable objective and subjective response occurred in fifteen patients, with almost complete regression of the tumor. However, this form of treatment cannot yet be assessed adequately since it was started in early 1974. Comments
Based on results of this study, it is obvious that the current management of primary liver cell carcinoma is far from adequate. A review of our current form of treatment after a period of two years is planned. It is interesting to note that of the 352 patients, 139 or 42.8 per cent had no treatment. Reported series in most countries in Africa and Asia have shown that the rate of resectability for primary liver cell carcinoma is relatively low and the long-term survival is disappointingly poor. Western investigators [12,18,19], however, have reported fairly good long-term survival with subjective improvement after complete hepatic dearterialization. Hepatic arterial ligation followed by regional infusion of chemotherapeutic drugs through the distal end of the hepatic artery has been reported, especially in the United States and Europe [12,13]. Our results, however, have been disappointing. Several factors, including diet, advanced stage of the disease, resistance of the host, and parasitic infection, may contribute to this disparity. Moreover, it is well-known that Asian patients have a poor prognosis whatever form of treatment for pri-
mary liver cell carcinoma is carried out [2,18,20]. The technical problems of hepatic transplantation largely have been overcome so that this could be the ultimate solution. However, problems with immunologic factors, tissue rejection, and organ preservation, including government legislation, must be solved before hepatic transplantation can become a routine procedure. It is likely that in countries such as Malaysia, where hepatic carcinoma is common, hepatic transplantation will not be feasible because of the excessive number of patients who would require this treatment, in addition to the problems of the costs involved and the training of personnel. It must be remembered that in these developing countries other important health problems take priority. Needless to say, our attention should be directed toward detecting the definite causative factors of hepatic carcinoma. More importantly, development of a good screening test to detect early carcinoma of the liver could be an alternate answer. Summary
A review of 352 patients with primary liver cell carcinoma treated by the author is presented. The poor rate of resectability (7 per cent) has necessitated various forms of treatment over the years. These are described in detail. Based on this experience, the current form of treatment for nonresectable carcinoma is summarized. Although it is too early to assess this form of treatment, initial results appear to be promising. A second report in the near future is planned. Acknowledgment: I wish to thank the DirectorGeneral, Ministry of Health, Federation of Malaysia, for permission to publish this paper and Mrs Jessie Cardoza for typing this manuscript. Addendum
Since this report was prepared, a total of ninetyeight patients have been treated by the current therapy. More recently, bacillus Calmette-Guerin (30,000,OOO organisms) is injected directly into the tumor tissue at operation in addition to subcutaneous therapy. Silver clips are then inserted at the sites of the multicentric tumor to assess regression of the tumor and allow administration of selective radiotherapy to the malignant cells. The Silver clips also permit percutaneous transhepatic injection of bacillus Calmette-Guerin postoperatively daily for four days, the dosage being 20,000,OOO organisms.
The
American Journal of Surgery
Liver Cell Carcinoma
References II. 1. Balasegaram M: Hepatic resection. Br J Surg 55: 168, 1968. 2. Plengvanit V, Limwonges K, Viranuvatti V, Hiianant S, Chearanai 0: Treatment of primary carcinoma of the liver by hepatic artery ligation. Preliminary report of 40 cases. Liver Research, p 490. Antwerp, Belgium, Tijdschrift voor Gastroenterologie, 1967. 3. Bagshawe A: Symposium in the International Liver Conference in South Africa, January 23-26, 1973. Personal communication. 4. Purves LR, Bersohn El, Geddes EW: Serum alpha-feto proteins and primary cancer of the liver in man. Cancer 25: 1261. 1970. 5. Balasegaram M: Hepatic surgery: present and future. Ann R Co/l Surg Engl47: 139, 1970. 6. Hsia SS, Tseng HH. Wu HK, Huang HC: Appraisal on liver resection in the treatment of primary hepatic carcinoma. Chin Med J 8: 248, 1965. 7. Tung TT: Personal communication, Hanoi, 197 I. 8. Balasegaram M: Hepatic surgery: a review of a personal series of 95 major resections. Aust NZ J Surg 42: 1, 1972. 9. Lin TY: Primary cancer of the liver. Stand J Gasfroenterol (Sup@) 6: 223, 1970. 10. Lange M, Falkson G, Geddes EW: Hepatic artery catheteri-
Volume 130, July 1975
12. 13.
14.
15. 16. 17.
18.
19. 20.
sation and infusion chemotherapy for primary liver cell carcinoma. In press. Chan KT: Management of primary liver carcinoma. Ann R CollSurgEngl41: 253. 1967. Foster JH: Survival after liver resection for cancer. Cancer 26: 493, 1970. Watkins E Jr, Khazel AH, Nahra KS: Surgical basis for arterial infusion chemotherapy of disseminated carcinoma of the liver. Surg Gynecol Obstet 138: 580. 1970. Sullivan RD, Watkins E Jr, Norcross JW:‘Chemotherapy of liver cancer by prolonged hepatic artery infusion. N Engl J Med270: 321, 1964. Breedis C, Young G: The blood supply of neoplasms in the liver. Am J Pathol30: 969, 1954. Balasegaram M: Complete hepatic dearterialization for primary carcinoma of the liver. Am J Surg 124: 340, 1972. Tucker WG, Talley RW, Brown Lee RW, Burrows JH, Stott PB, Moorhead EL, Sandiego EL: Preliminary trial with combination of cyclophosphamide, vincristine. 5-fluorouracil. Cancer Chemother Reo 52: 593, 1968. Bengmark ST, Almersjo b, Engevik L, Hafstrom 0: Surgical management of tumours of the liver. Chirurg 39: 320, 1968. Fortner J: Personal communication, 1974. Mori M. Masuda M, Miyanaga T: Hepatic artery ligation and tumor necrosis in the liver. Surgery 59: 359, 1966.
37