Long-term oral branched chain amino acids in patients undergoing chemoembolization for hepatocellular carcinoma: a randomized trial

Authors


Dr R. T.-P. Poon, Department of Surgery, University of Hong Kong Medical Centre, Queen Mary Hospital, 102 Pokfulam Road, Hong Kong, China.
E-mail: poontp@hkucc.hku.hk

Summary

Background : Patients undergoing transarterial chemoembolization for hepatocellular carcinoma have advanced tumour or severe cirrhosis and frequently have associated protein-calorie malnutrition. The role of nutritional supplements for such patients is unclear.

Aim : To investigate, in a randomized controlled trial, any benefit of the long-term administration of branched chain amino acids in patients undergoing chemoembolization for hepatocellular carcinoma.

Methods : Forty-one patients received oral branched chain amino acids for up to four courses of chemoembolization and 43 patients did not receive any nutritional supplement. Morbidity, liver function, nutritional status, quality of life and long-term survival were compared between the two groups.

Results : The administration of branched chain amino acids resulted in a lower morbidity rate compared with the control group (17.1% vs. 37.2%, P = 0.039). In particular, the group given branched chain amino acids showed a significantly lower rate of ascites (7.3% vs. 23.2%, P = 0.043) and peripheral oedema (9.8% vs. 27.9%, P = 0.034). Significantly higher serum albumin, lower bilirubin and a better quality of life were observed after chemoembolization in the group given branched chain amino acids. However, there was no significant difference in survival between the two groups.

Conclusions : Nutritional supplementation with oral branched chain amino acids is beneficial in increasing the serum albumin level, reducing the morbidity and improving the quality of life in patients undergoing chemoembolization for inoperable hepatocellular carcinoma.

Introduction

Hepatocellular carcinoma (HCC) is one of the most common malignancies, ranking fifth in the world.1 The management of HCC is complicated by the presence of underlying cirrhosis in most patients. Hepatic resection provides a chance of cure, but is only applicable in a small proportion of patients.2, 3 Although liver transplantation is an alternative curative treatment for patients with small HCCs associated with severe cirrhosis, its application is restricted by the severe shortage of liver donors.4 For most patients with advanced HCC confined to the liver, or poor liver function due to underlying cirrhosis, locoregional therapies, such as transarterial chemoembolization (TACE), transarterial radiotherapy, ethanol injection and radiofrequency ablation, offer a chance of effective palliation or even long-term survival.5 TACE is widely used for the treatment of unresectable HCC. Although early trials failed to demonstrate its benefit in prolonging survival,6, 7 more recent randomized studies have shown that it is effective in improving the survival of patients with inoperable HCC.8, 9

Most patients with HCC have underlying cirrhosis, which is frequently associated with a state of protein-calorie malnutrition.10, 11 In addition, patients with HCC may suffer from a tumour-induced increase in protein catabolism,12 which may further aggravate their nutritional status. The administration of branched chain amino acids (BCAAs) has been shown to correct malnutrition associated with cirrhosis in both animal and human studies.13, 14 Previous randomized controlled studies have demonstrated that parenteral or oral nutritional support with BCAA-enriched preparations in patients undergoing hepatic resection for HCC significantly improves their post-operative nutritional status and reduces the morbidity rate and hospital stay.15, 16 Furthermore, it has been demonstrated that long-term oral administration of BCAAs after resection of HCC significantly increases the serum albumin level and reduces the incidence of ascites or peripheral oedema.17

Patients with inoperable HCC undergoing locoregional therapy have either more advanced tumours or more severe cirrhosis than those undergoing hepatic resection, and thus are more likely to suffer from malnutrition. Nutritional support may play an important role in the management of these patients. However, thus far, there has been no study on nutritional support in patients undergoing non-surgical management of HCC. We conducted a prospective randomized trial to evaluate any benefit of long-term oral supplementation with BCAAs in patients with unresectable HCC undergoing TACE.

Patients and methods

Selection of patients and randomization

Between July 1998 and December 2000, 88 patients with newly diagnosed unresectable HCC, undergoing TACE at the Department of Surgery, University of Hong Kong Medical Centre, Queen Mary Hospital, Hong Kong, were recruited into this single-centre, randomized controlled trial. The diagnosis of HCC was based on histology, cytology or elevated serum α-foetoprotein levels (> 400 ng/mL), with typical imaging findings on computed tomography scan and arteriography. The patients had unresectable disease due to bilobar tumours or inadequate liver function reserve. The selection criteria for TACE were as follows: (i) absence of extra-hepatic metastasis; (ii) absence of vascular contraindications (hepatic artery thrombosis, main portal vein thrombosis or arteriovenous shunting); (iii) absence of hepatic encephalopathy, ascites refractory to diuretics or variceal bleeding within 3 months; (iv) a serum bilirubin level of less than 50 µmol/L and a serum albumin level above 25 g/L; and (v) a Karnofsky performance score of 50 or above.18 Patients with a history of other treatments for the tumour or acute tumour rupture were excluded from the study.

Patients were randomized into two groups: those who received nutritional supplementation with BCAAs and a control group. The control group took their usual diet and the BCAA group took 50 g Aminoleban EN (Otsuka Pharmaceutical Company, Tokyo, Japan) twice a day in the morning and evening in addition to the usual diet. No dietary restrictions were imposed unless the patient developed hepatic encephalopathy, in which case the protein intake was restricted. Aminoleban EN was started 1 week before first TACE and continued for up to 1 year. This allowed a maximum of four sessions of TACE during which Aminoleban EN was consumed in the BCAA group. No placebo was given in the control group because the special taste of Aminoleban EN made it impossible to find a placebo with a similar taste that could truly ensure blinding of the patients. The randomization was performed without stratification by drawing consecutively numbered sealed envelopes assigning the patients to each arm. The protocol was approved by the Ethics Committee of our institution and informed consent was obtained from all participating patients.

Preparation of oral BCAAs

Aminoleban EN contains a high content of BCAAs and a small amount of other amino acids with a Fischer ratio of approximately 38. Each 100 g of Aminoleban EN contains 27.0 g protein (13.0 g as amino acids, 13.0 g as peptides, 1.0 g as casein), 62.1 g dextran and 7.0 g rice oil, which produces 420 kcal of energy. Specifically, each 100 g of Aminoleban EN contains 3.2 g valine, 4.0 g leucine and 3.8 g isoleucine. It is also complemented with various minerals and vitamins. The composition of Aminoleban EN is given in Table 1.

Table 1.  Composition of Aminoleban EN (per 50 g)
  • *

     Minerals include trace amounts of magnesium sulphate, calcium glycerophosphate, potassium iodide, potassium chloride, sodium dihydrogen phosphate dihydrate, sodium ferrous citrate, cupric sulphate, zinc sulphate and manganese sulphate.

  • † Vitamins include retinol palmitate, ergocalciferol, bisbentiamine, riboflavin, pyridoxine HCl, cyanocobalamin, folic acid, sodium l-ascorbate, tocopherol acetate, phytonadione, calcium pentothenate, nicotinamide and biotin.

Amino acids
 l-Isoleucine1.923 g
 l-Leucine2.037 g
 Lysine.HCl0.243 g
 l-Threonine0.133 g
 l-Valine1.602 g
 l-Arginine.HCl0.302 g
 l-Histidine.HCl0.188 g
 l-Tryptophan73.5 mg
Gelatine hydrolysate6.5 g
Rice oil3.5 g
Dextrin31.03 g
Minerals*
Vitamins

Procedure of TACE

TACE was given using a standardized protocol, details of which have been described in previous reports from our institution.8, 19 Briefly, patients were fasted overnight, and prophylactic antibiotics (intravenous amoxicillin–clavulanic acid, 1.2 g) were given before the procedure. TACE was performed by catheterization via the femoral artery under local anaesthesia, with superselective catheterization of the hepatic artery feeding the tumour, unless there were bilobar tumours, in which case chemoembolization was performed in the appropriate hepatic artery. Depending on the tumour size, various amounts of an emulsion of cisplatin (1 mg/mL) and Lipiodol (Lipiodol Ultrafluide, Laboratoire Guerbet, Aulnay-sous-Bois, France), in a volume ratio of 1 : 1, were injected under fluoroscopic monitoring. The maximum volume of the emulsion injected was 60 mL (containing 30 mg of cisplatin). This was followed by embolization with gelatine-sponge particles mixed with gentamicin. Patients resumed their diet after the procedure and were discharged the next day unless complications occurred. Oral antibiotics (amoxicillin–clavulanic acid, 375 mg three times daily) and an H2 blocker (famotidine, 20 mg twice daily) were prescribed for 5 days after discharge.

TACE was repeated every 3 months and was withheld or discontinued when vascular contraindications, liver failure, severe complications or progressive disease developed. Liver failure was defined as a rise in bilirubin to > 50 µmol/L, the development of ascites refractory to diuretics or the development of hepatic encephalopathy.19 Patients who had TACE discontinued because of progressive tumours or liver failure were treated conservatively without any further anti-tumour treatment.

Follow-up and assessment of outcome

In addition to the imaging studies and routine blood tests (complete blood count, coagulation profile, liver biochemistry and renal function test) required before TACE, all patients had a baseline assessment that included nutritional status (body weight, body mass index, triceps skin-fold thickness, mid-arm circumference and grip strength), indocyanine green clearance test20 and quality of life assessment using the Functional Assessment of Cancer Therapy — General (FACT-G) questionnaire,21 which has been shown to be a useful tool for the study of the quality of life in HCC patients in a previous study by the authors.22 Any complications after each TACE treatment were recorded. All patients were followed up every fortnight with clinical assessment, complete blood count, liver and renal biochemistry. Physical examination was performed during each follow-up with a particular emphasis on oedema, ascites and signs of hepatic encephalopathy. Ascites was confirmed with ultrasound or computed tomography scan findings. The clinicians assessing the patients in the out-patient clinic were blind to the randomization group of the patients. Anthropometric measurements, indocyanine green clearance test and quality of life were evaluated every 3 months before the next TACE session. Tumour response was monitored by serial α-foetoprotein levels, angiography and computed tomography scan every 3 months. The compliance to the intake of Aminoleban EN was monitored during each follow-up visit by a research nurse and was satisfactory in all patients assigned to the BCAA group. The dietary intake of both groups of patients was monitored with the aid of a diet diary. The study protocol was completed after 1 year, but the patients were allowed to continue further TACE if appropriate.

The primary outcome measure was the post-TACE morbidity. Morbidity was defined as adverse effects that developed within 4 weeks after each session of TACE, including deterioration of liver function that resulted in the development or worsening of ascites, peripheral oedema or hepatic encephalopathy. Transient fever, vomiting and abdominal pain immediately after TACE were not included. The morbidity rate of each group was defined as the overall frequency of morbidity after any TACE session during the four TACE sessions (or fewer when TACE was terminated for the reasons indicated above) in the study protocol. Secondary outcome measures included treatment mortality (death within 4 weeks after a session of TACE), re-admission to hospital for adverse events, liver function, nutritional status, quality of life and long-term survival.

Statistical analysis

The cumulative rate of morbidity of TACE, as defined above, during a maximum of four courses of TACE, was estimated to be about 50% based on the results of previous studies from our institution and others.6, 8, 23, 24 In accordance with our previous study of nutritional support in patients undergoing hepatic resection for HCC,15 we assumed that a reduction in the morbidity rate by half was necessary to indicate the efficacy of nutritional support with BCAAs. Forty-four patients were required in each group to detect a difference at a level of statistical significance of 0.05 and a power of 0.80.

Continuous variables were expressed as the median (range) and compared using the Mann–Whitney U-test. Categorical variables were compared using the chi-squared test or Fisher's exact test where appropriate. Survival was computed using the Kaplan–Meier method and compared using the log-rank test. A P value of 0.05 or less by the two-tailed test was considered to indicate statistical significance. All statistical analyses were performed using statistical software (SPSS 9.05 for Windows, SPSS, Inc, Chicago, IL, USA).

Results

Eighty-eight patients were randomized either to the BCAA group (n = 44) or the control group (n = 44). Three patients assigned to the BCAA group and one patient assigned to the control group were secondarily excluded from the analysis. In the BCAA group, one patient who changed his mind and opted for liver transplantation instead of TACE after randomization was excluded. Another patient who did not receive TACE due to arteriovenous shunting found in angiography was also excluded. A third patient withdrew from the study before first TACE. In the control group, one patient was excluded because he was found to have main portal vein thrombosis at the time of angiography and did not receive chemoembolization. Thus, the data in this study consisted of observations made in 41 patients in the BCAA group and in 43 patients in the control group.

Table 2 shows the baseline characteristics of the two groups of patients. Most patients (88.1%) had positive serology for hepatitis B surface antigen. The two groups were well balanced with regard to age, sex, presence of co-morbid illnesses, liver function, renal function, body weight, tumour characteristics and Okuda staging.25

Table 2.  Baseline characteristics of the study patients
 BCAA group
(n = 41)
Control group
(n = 43)
  • BCAA, branched chain amino acid.

  • P > 0.05 for all variables when the two groups were compared.

  • *

     Continuous data are expressed as the median and range; otherwise values indicate the number of patients.

  •  Includes diabetes mellitus and chronic respiratory, cardiovascular or renal diseases.

Age (years)*59 (24–84)59 (27–80)
Sex (male/female)39/239/4
Positive serum hepatitis B surface antigen3737
Presence of co-morbid illness1916
Presence of ascites12
Presence of peripheral oedema22
Haemoglobin (g/dL)*13.2 (8.5–16.7)13.4 (8.0–18.0)
Platelet count (× 109/L)*108 (36–610)115 (47–506)
Serum bilirubin (µmol/L)*16 (4–49)15 (4–43)
Serum albumin (g/L)*35 (25–48)36 (24–46)
Serum creatinine (µmol/L)*90 (64–273)90 (39–167)
Body weight (kg)*60 (43–89)59 (38–78)
Weight loss > 10%78
Body mass index (kg/m2)*22.3 (16.6–29.6)22.6 (16.7–30.3)
Diameter of largest tumour (cm)*7.0 (1.5–16.0)6.5 (2–17.5)
Presence of multiple tumours1215
Presence of bilobar tumours2224
Presence of tumour thrombus in portal vein branch118
Okuda stage (I/II)30/1131/12

Both groups received a median number of two courses (range, 1–4) of TACE during the study period (P = 0.487). The 41 patients in the BCAA group received a total of 104 courses of chemoembolization, whereas the 43 patients in the control group received a total of 96 courses of chemoembolization. The dose of cisplatin–Lipiodol emulsion injected in one course was similar in the BCAA group (median, 20 mL; range, 2–60 mL) and the control group (median, 18 mL; range, 4–60 mL) (P = 0.462). Eighteen patients in the BCAA group and 14 patients in the control group completed four courses of TACE. For those patients who could not complete four courses of TACE during the study period, the reasons for termination of TACE included tumour progression (15 in the BCAA group, 15 in the control group), severe deterioration of liver function (four in the BCAA group, 10 in the control group), development of arteriovenous shunting (two in the BCAA group, one in the control group) and severe adverse effects (two in the BCAA group, three in the control group).

Post-chemoembolization morbidity and mortality

Table 3 shows the post-TACE complications in the two groups. The overall morbidity rate in the BCAA group was significantly lower than that in the control group (17.1% vs. 37.2%, P = 0.039). Compared with the control group, the BCAA group had a significantly lower frequency of ascites (7.3% vs. 23.2%, P = 0.043) and peripheral oedema (9.8% vs. 27.9%, P = 0.034), which were the most common causes of morbidity after TACE. More severe complications, such as gastrointestinal bleeding from peptic ulcer, variceal bleeding, hepatic encephalopathy, liver abscess, tumour rupture and renal failure, were relatively rare, with no significant differences between the two groups (Table 3). Hence, the reduction in the morbidity rate in the BCAA group was related mainly to the reduced frequency of ascites and peripheral oedema.

Table 3.  Complications of transarterial chemoembolization
 BCAA group
(n = 41)
Control group
(n = 43)
  • BCAA, branched chain amino acid.

  • Some patients had more than one complication.

  • *

     New development or deterioration of pre-existing ascites or peripheral oedema that required diuretic therapy.

  •  P < 0.05 between BCAA group and control group; no significant differences between the two groups in other complications.

Ascites*3 (7.3%)10 (23.2%)
Peripheral oedema*4 (9.8%)12 (27.9%)
Peptic ulcer bleeding1 (2.4%)3 (7.0%)
Variceal bleeding1 (2.4%)0 (0%)
Hepatic encephalopathy0 (0%)1 (2.3%)
Liver abscess0 (0%)1 (2.3%)
Tumour rupture1 (2.4%)1 (2.3%)
Renal failure0 (0%)1 (2.3%)
Any complication(s)7 (17.1%)16 (37.2%)

There were three deaths (7.0%) within 1 month after TACE in the control group, but none in the BCAA group (P = 0.241). In the control group, two patients died of liver failure and one patient died of tumour rupture after TACE.

The median hospital stay for each course of TACE was 2 days (range, 1–6 days) in both groups (P = 0.862). In addition to the scheduled hospital admissions for TACE, five patients (12.2%) in the BCAA group and 15 patients (34.9%) in the control group had at least one re-admission during the study period for adverse events after TACE. The re-admission rate of the BCAA group was significantly lower than that of the control group (P = 0.032). There were a total of six and 16 re-admissions for adverse events in the BCAA group and the control group, respectively. The causes of these re-admissions included ascites (two in the BCAA group, seven in the control group), persistent fever (one in the BCAA group, two in the control group), peptic ulcer bleeding (one in the BCAA group, three in the control group), variceal bleeding (one in the BCAA group), hepatic encephalopathy (one in the control group), liver abscess (one in the control group), tumour rupture (one in each group) and renal failure (one in the control group). The median hospital stay for each re-admission was 3 days (range, 1–7 days) in the BCAA group and 4 days (range, 2–9 days) in the control group (P = 0.233).

Liver function, nutritional status and quality of life

Table 4 shows the dietary intake, liver function parameters, anthropometric measurements and quality of life scores of the patients in the two groups before and at 3, 6, 9 and 12 months after first TACE. Analysis of the diet diaries of the two groups revealed that the dietary intake, excluding Aminoleban EN in the BCAA group, was comparable in terms of daily calories and nitrogen intake at all time points. The baseline liver function parameters (serum bilirubin, albumin, aspartate aminotransferase, prothrombin time and indocyanine green retention at 15 min) were comparable between the two groups. The BCAA group had significantly higher serum albumin levels at 3 months (median 36 vs. 30 g/L, P = 0.008), 6 months (median 35 vs. 30 g/L, P = 0.024) and 9 months (median 34 vs. 29 g/L, P = 0.042) after first TACE compared with the control group. In addition, the BCAA group had significantly lower serum bilirubin levels at 3 months (median 18 vs. 29 µmol/L, P = 0.013) and 6 months (median 20 vs. 30 µmol/L, P = 0.026) after first TACE. There were no significant differences in aspartate aminotransferase, prothrombin time and indocyanine green retention at 15 min between the two groups at any time points.

Table 4.  Dietary intake, liver function, nutritional status and quality of life at different time points
 Pre-treatment
(BCAA, n = 41;
control, n = 43)*
3 months
(BCAA, n = 37;
control, n = 36)*
6 months
(BCAA, n = 32;
control, n = 28)*
9 months
(BCAA, n = 25;
control, n = 23)*
12 months
(BCAA, n = 21;
control, n = 20)*
  • BCAA, branched chain amino acid; FACT-G, Functional Assessment of Cancer Therapy — General.

  • *

     Number of surviving patients at each time point available for study.

  •  P < 0.05 between BCAA group and control group; no significant differences in other comparisons.

Dietary calorie intake (kcal/kg body weight per day)
 BCAA group28.8 (24.2–34.5)28.0 (24.6–33.5)28.2 (25.0–32.5)28.0 (25.0–33.2)27.2 (24.3–32.4)
 Control group29.0 (24.6–33.6)28.6 (25.3–34.2)28.0 (25.0–33.2)27.8 (23.5–32.4)27.5 (24.0–33.2)
Dietary nitrogen intake (g/kg body weight per day)
 BCAA group0.16 (0.13–0.19)0.16 (0.14–0.19)0.15 (0.12–0.21)0.14 (0.11–0.18)0.14 (0.13–0.18)
 Control group0.16 (0.12–0.20)0.15 (0.12–0.19)0.16 (0.12–0.20)0.15 (0.12–0.19)0.14 (0.12–0.18)
Albumin (g/L)
 BCAA group35 (25–48)36 (28–44)35 (26–40)34 (25–42)33 (26–40)
 Control group36 (24–46)30 (20–42)30 (20–40)29 (22–38)29 (23–39)
Bilirubin (µmol/L)
 BCAA group16 (4–49)18 (8–52)20 (9–40)24 (8–62)22 (10–70)
 Control group15 (4–43)29 (5–84)30 (12–94)31 (10–83)29 (12–78)
Aspartate aminotransferase (iu/L)
 BCAA group47 (14–184)64 (31–336)53 (20–215)47 (26–141)52 (35–316)
 Control group48 (8–256)66 (40–435)60 (28–357)54 (29–245)60 (28–304)
Prothrombin time (s)
 BCAA group13.6 (10.0–19.1)14.8 (11.1–20.2)15.2 (10.9–22.1)15.4 (10.8–21.4)15.6 (11.0–22.0)
 Control group13.4 (10.8–17.4)15.0 (11.4–20.6)15.2 (12.2–22.4)15.6 (11.6–22.0)16.0 (12.2–24.3)
Indocyanine green retention at 15 min (%)
 BCAA group20.8 (3.4–75.6)21.8 (5.8–73.9)24.7 (7.3–89.3)25.2 (3.4–70.5)25.8 (5.8–72.4)
 Control group19.8 (3.2–55.1)22.4 (20–72.9)25.8 (8.5–78.9)27.8 (6.1–82.4)29.2 (6.8–76.5)
Body weight (kg)
 BCAA group60 (43–89)60 (43–87)61 (43–87)62 (49–72)60 (46–78)
 Control group59 (42–78)59 (45–73)59 (48–73)61 (48–76)61 (50–80)
Mid-arm circumference (cm)
 BCAA group24.5 (20.0–29.0)24.0 (18.0–29.0)24.0 (20.0–29.5)25.0 (21.0–28.0)24.5 (20.0–28.5)
 Control group24.5 (19.0–31.0)22.5 (19–28.0)23.0 (21.0–29.0)23.0 (21.5–28.0)23.0 (21.0–28.0)
Triceps skin-fold (mm)
 BCAA group8.0 (4.0–17.0)8.0 (3.0–16.0)7.5 (4.0–16.0)8.0 (4.0–15.0)8.5 (4.0–15.5)
 Control group7.5 (3.0–17.0)7.0 (4.0–14.0)7.0 (4.0–16.0)7.0 (3.5–15.0)7.0 (4.0–15.0)
Hand grip strength (kg)
 BCAA group20 (9–34)22 (13–33)22 (15–32)22 (10–34)20 (11–32)
 Control group20 (7–33)17 (10–31)17 (13–31)18 (12–32)17 (12–32)
FACT-G score
 BCAA group87 (65–98)93 (68–99)91 (71–99)90 (73–97)89 (68–96)
 Control group86 (74–99)85 (65–96)84 (66–94)84 (68–96)84 (70–96)

The two groups had a similar nutritional status before first TACE in terms of body weight, mid-arm circumference, triceps skin-fold thickness and hand grip strength (Table 4). The hand grip strength of the BCAA group was significantly better than that of the control group at 3 months (median 22 vs. 17 kg, P = 0.032) and 6 months (median 22 vs. 17 kg, P = 0.046) after first TACE. There were no significant differences between the two groups in the other parameters. The mid-arm circumference was higher in the BCAA group than in the control group at all four time points after first TACE, but the differences did not reach statistical significance.

The quality of life score by the FACT-G questionnaire was similar between the two groups before first TACE. However, a significantly better quality of life, as reflected by the higher FACT-G scores, was observed in the BCAA group at all four time points after first TACE (Table 4).

Survival

Figure 1 shows the cumulative survival curves of the two groups of patients by the time of data analysis. The median follow-up from the time of first TACE was 29 months (range, 18–44 months) in the BCAA group and 30 months (range 18–43 months) in the control group. There was no significant difference in survival (median 12.3 vs. 10.4 months, P = 0.294) between the two groups. By the time of data analysis, 29 patients (70.7%) in the BCAA group and 36 patients (83.7%) in the control group had died (P = 0.155). In the BCAA group, 22 patients died of progression of cancer, six of liver failure and one of an unrelated medical condition. In the control group, 22 patients died of tumour progression, 12 of liver failure and two of unrelated medical conditions.

Figure 1.

Cumulative survival curves in the branched chain amino acid (BCAA) group (full line) and the control group (broken line) (log-rank test, P = 0.294).

Discussion

Several retrospective studies have demonstrated that TACE is an effective treatment for inoperable HCC confined to the liver, with a tumour response rate of 30–60%.23, 24, 26 Two early randomized trials failed to demonstrate a significant benefit of TACE with Lipiodol in prolonging survival compared with conservative management in patients with advanced HCC.6, 7 These two studies showed a tumour response rate of 53% and 24%, respectively, but the potential survival benefit of the anti-tumour effect of TACE in these trials appeared to be counteracted by its deleterious effect on liver function, with a liver failure rate of more than 50% in both studies. Two more recent randomized trials, however, demonstrated a significant survival benefit of TACE compared with conservative treatment,8, 9 and a recent meta-analysis also concluded that this is an effective treatment in prolonging the survival of patients with advanced HCC.27 Although TACE is largely considered as a palliative treatment for HCC, long-term survival in excess of 5 years is possible with this treatment in a small proportion of patients with advanced tumours.28 However, TACE is associated with a high rate of complications, including liver failure, which not only leads to morbidity but may also limit its survival benefit.6, 19, 24 Hence, the search for effective measures to reduce the morbidity of this treatment is of considerable importance. This study evaluated the effect of nutritional supplementation using BCAAs on the morbidity in patients with unresectable HCC undergoing TACE.

To our knowledge, this is the first study to evaluate the role of nutritional support in patients undergoing non-surgical therapy for HCC. A few previous studies have demonstrated the benefit of nutritional supplementation using BCAAs in reducing morbidity after hepatectomy for HCC.15–17 Nutritional support is conceivably important in patients undergoing non-surgical therapy because these patients usually have more advanced tumour and cirrhosis, and thus a more severe nutritional problem. The current study showed that nutritional supplementation using BCAAs is effective in reducing the morbidity after TACE. The reduction in morbidity resulted mainly from the reduced frequency of ascites and peripheral oedema, which, in turn, might be attributable to the better maintained serum albumin levels in the BCAA group compared with the control group. However, as the development of ascites and peripheral oedema in cirrhotic patients is influenced by sodium intake, and dietary restriction was not imposed in this study, it is not clear whether the reduced frequency of ascites and peripheral oedema in the BCAA group was the result of decreased sodium intake or a real effect of the BCAA supplement. A previous study of long-term supplementation using the same oral BCAA preparation, in patients who had undergone hepatic resection for HCC, also showed an increased serum albumin level and reduced frequency of ascites and peripheral oedema in the BCAA group compared with the control group.17 In the current study, the decrease in morbidity also led to a significantly lower re-admission rate in the BCAA group. Although we did not include a cost-effectiveness analysis in the design of this study, the decrease in hospital stay as a result of re-admission for adverse events in the BCAA group is a favourable factor in terms of the cost–benefit ratio of this nutritional therapy.

In addition to the higher serum albumin levels, the BCAA group also had lower serum bilirubin levels 3 and 6 months after first TACE. However, there were no significant differences in two other liver function parameters, namely the prothrombin time and indocyanine green retention rate, between the two groups. With regard to the nutritional parameters, the hand grip strength was significantly higher in the BCAA group than in the control group at 3 and 6 months after first TACE. The mid-arm circumference was similar in the two groups before treatment, but was higher in the BCAA group than in the control group at all time points after first TACE, although the difference did not reach statistical significance. There was no obvious difference in the body weight, which was not a reliable indicator of nutritional status in the study patients because of the presence of fluid retention in a substantial proportion of patients.

This study also compared the quality of life between the two groups. Quality of life is an important outcome measure of cancer therapy.21 To our knowledge, however, none of the previous trials on TACE or other therapies for HCC have assessed the quality of life of the study patients as an outcome. Our study showed significantly improved quality of life at all time points after first TACE in the BCAA group compared with the control group. Because the study was not placebo-controlled or blind to the patients, it could be argued that there may be subjective bias in the quality of life evaluation. This is a limitation of the study. When we designed the study, we gave consideration to a blind study using a placebo, but found that this was not practicable because we could not find a placebo that tasted similar to Aminoleban EN to effectively blind the patients. Nevertheless, the significant improvement in morbidity and objective parameters, such as serum albumin levels and bilirubin levels, in the BCAA group compared with the control group suggests that the better quality of life in the BCAA group could at least be partly ascribed to the use of BCAAs.

Protein-calorie malnutrition in liver cirrhosis is related to reduced dietary intake, impaired digestion, malabsorption and impaired protein synthesis in the liver.10, 11 In patients with advanced HCC undergoing TACE, the nutritional status may be further compromised by the malignancy. Cirrhotic patients are characterized by an imbalance of amino acid metabolism, with decreased BCAAs as a result of enhanced uptake in the muscle, and increased plasma aromatic amino acids because of reduced metabolism in the liver.29 This imbalance in BCAAs and aromatic amino acids has been implicated in the pathogenesis of hepatic encephalopathy, and the use of BCAAs has been found to have some beneficial effect in cirrhotic patients with chronic hepatic encephalopathy.30 In addition, BCAA therapy has been shown to have an anti-catabolic effect in cirrhotic patients, resulting in decreased muscle protein catabolism and increased hepatic protein synthesis.14, 29 BCAAs can be utilized directly by muscle, brain, heart and liver as an energy substrate, and may therefore have a beneficial effect in meeting the increased energy requirement in cirrhotic patients with HCC undergoing TACE.31 The use of BCAAs as an energy substrate has also been shown to have a protein-sparing effect in cirrhotic patients.32 The anti-catabolic effect and enhanced hepatic protein synthesis associated with BCAA administration in cirrhotic patients probably account for the better nutritional status and increased serum albumin level in the BCAA group in the current study. Two previous studies have shown that long-term oral BCAA administration improves protein metabolism and the nutritional status of cirrhotic patients without HCC.33, 34

TACE induces damage to the non-tumorous liver, especially with repeated courses, resulting in a significant deterioration of liver function.6, 7, 19, 23, 24 In several retrospective and prospective studies, worsening of liver function, with ascites, jaundice or hepatic encephalopathy, has been observed in 50–100% of patients after TACE.6, 7, 23, 24 Liver atrophy has been demonstrated in computed tomography scan after repeated courses of TACE.35 The embolizing particles and chemotherapeutic agent may cause significant ischaemic and cytotoxic damage to the non-tumorous liver. Recovery of liver function depends on liver regeneration between each course of TACE. However, protein-calorie malnutrition can impair liver regeneration. The administration of BCAAs has been shown to enhance liver regeneration after partial hepatectomy.36 A previous study has demonstrated that oral administration of BCAAs, using the same preparation as in our study, contributes to a more rapid improvement of liver function, in terms of serum albumin and bilirubin levels, after hepatic resection for HCC.16 Our study revealed similar findings in the group with BCAA supplementation. However, as the current study did not include a group with an isonitrogenous standard protein solution supplement, it is not possible to demonstrate that the benefits observed in the BCAA group, compared with the control group, were specifically due to BCAA, rather than a general effect of supplementation with any type of protein solution in a group of patients with protein malnutrition. A recent randomized double-blind trial, however, has demonstrated that oral BCAA supplementation for 1 year improves the nutritional parameters, reduces the serum bilirubin level and reduces the prevalence and severity of ascites in patients with advanced cirrhosis, compared with a control group given an isonitrogenous conventional protein supplement.37

A study on BCAA therapy in cirrhotic patients has suggested that long-term oral administration of BCAAs has a beneficial effect on patient survival by improving protein nutritional status and delaying mortality from liver failure.33 In the current study, there were fewer deaths in the long term from liver failure in the BCAA group compared with the control group (six and 12 patients, respectively). However, this did not result in a significant difference in the overall cumulative survival between the two groups, because the majority of deaths in both groups were related to tumour progression. Two experimental studies have found that high BCAA levels in culture medium can suppress the in vitro growth of HCC, and suggest that amino acid imbalance therapy, with an enrichment of BCAAs, may be useful in the treatment of HCC.38, 39 However, in this study, the numbers of deaths from tumour progression were similar between the two groups, and BCAA supplementation did not have any apparent anti-tumour effect.

In conclusion, this study shows that long-term oral administration of BCAAs in patients with unresectable HCC undergoing TACE is beneficial in increasing the serum albumin level, reducing hyperbilirubinaemia, reducing morbidity, reducing the number of hospital re-admissions, improving nutritional parameters, as demonstrated by an increased hand grip strength, and improving the quality of life. This is the first study to demonstrate a positive effect of nutritional support in HCC patients undergoing non-surgical therapy. More studies should be conducted to further elucidate the role of nutritional support in various types of non-surgical therapy for HCC.

Ancillary