Dr T. Huo, Division of Gastroenterology, Department of Medicine, Taipei Veterans General Hospital, Taipei, Taiwan. E-mail: email@example.com
Background : Transarterial chemoembolization (TACE) and percutaneous acetic acid injection (PAI) are effective loco-regional therapies for hepatocellular carcinoma (HCC).
Aim : To compare the therapeutic efficacy of TACE vs. PAI for unresectable HCC.
Methods : A total of 310 patients with unresectable HCCs (size ≤6 cm) undergoing TACE (n = 195) or PAI (n = 115) were studied prospectively. Overall and progression-free survivals were measured endpoints.
Results : The overall survival was not significantly different between the two groups (P = 0.508). Among 129 patients with large (3.1–6 cm) HCCs, the overall survival was significantly better for the TACE group (P = 0.018). Cox multivariate analysis showed that Child-Pugh B [relative risk (RR): 4.2, 95% confidence interval (CI): 2.3–7.7, P < 0.001] and PAI therapy (RR: 1.4, 95%: 1.0–1.9, P = 0.057) were poor prognostic predictors; the progression-free survival was also significantly better in the TACE group (P = 0.038). Among 181 patients with small (≤3 cm) HCCs, there was no significant difference of overall survival (P = 0.265) or progression-free survival (P = 0.146) between the two groups; Child-Pugh B was the only prognostic factor predicting a decreased survival (RR: 2.8, 95% CI: 1.7–4.8, P < 0.001).
Conclusions : Patients with large HCC undergoing TACE tend to have a more favourable long-term outcome. For small HCC, either TACE or PAI therapy could be recommended as the primary treatment modality.
Hepatocellular carcinoma (HCC) is one of the common malignant tumours in the world with a global increasing annual incidence as recently reported.1, 2 Surgical resection is the standard treatment modality to achieve a long-term survival.3–5 However, resection is possible in only a small proportion of patients due to compromised hepatic reserve and multifocality of tumours. Liver transplantation is another treatment option especially for patients with decompensated cirrhosis,6 but potential recipients far outnumber donors. Alternative non-surgical methods include transarterial chemoembolization (TACE),7–11 and percutaneous ultrasound (US)-guided ablation therapy including injection of ethanol12–18 or acetic acid,19–22 and radio-frequency thermal ablation.23
For patients with unresectable diseases, the goal of palliative treatment is to control symptoms and prolong survival.10, 11, 24 TACE using iodized oil and chemotherapeutic agents combines the effect of targeted chemotherapy with that of ischaemic necrosis induced by arterial embolization. It could be repeatedly administered and can prolong survival in patients with unresectable HCC.25, 26 However, some reports suggest that by using 15–50% acetic acid, percutaneous acetic acid injection (PAI) has stronger necrotizing power compared with that of pure ethanol,20, 21 and may be equally effective or even superior to ethanol injection in treating small HCC.19, 22 Both TACE and percutaneous injection therapy have a well-documented loco-regional anti-tumoral effect and are two most commonly employed methods for HCC. However, for patients with unresectable lesions, the choice of the primary treatment modality between transarterial vs. percutaneous approach remains unclear in terms of long-term survival. In this study, we investigated the therapeutic efficacy of TACE and PAI in a large cohort of patients with unresectable HCC. Risk factors linked to survival and the optimal management for such patients were established.
Patients and methods
From April 1998 to October 2002, patients who were diagnosed as HCC and treated in our centre were studied and their medical profiles were prospectively collected. To compare the therapeutic efficacy between TACE and PAI, the following criteria were used to select eligible patients: (1) not indicated for resection; (2) tumour nodule measured ≤6 cm in greatest dimension; (3) the number of tumour nodules was three or less; (4) Child-Pugh class A or B; (5) has not been treated with other loco-regional therapy or active treatment (treatment naive); (6) no main portal vein involvement or extrahepatic metastasis. The HCC nodule(s) was considered unresectable if there were multifocal lesions which made extended resection necessary to eradicate all tumours, or the hepatic reserve was insufficient with an indocyanine green 15-min retention rate >30%. The diagnosis of HCC was histologically verified by needle biopsy, or based on the findings of typical radiological features in at least two image examinations including US, contrast-enhanced dynamic computed tomography (CT), magnetic resonance imaging (MRI) and hepatic angiography, or by a single positive imaging technique associated with serum α-fetoprotein (AFP) level >400 ng/mL.27 The severity of the underlying cirrhosis was classified according to the Child-Pugh system. The underlying hepatitis was considered hepatitis B virus infection if patients were seropositive for hepatitis B surface antigen (HBsAg, RIA kit; Abbott Laboratories, North Chicago, IL, USA) on at least two occasions.
Treatment and follow-up
The TACE was carried out according to Seldinger's technique of arterial embolization.28–30 The equipment for selective common or proper hepatic artery angiography was LCN (GE Medical System, Waukesha, WI, USA) or Diagnost 3 (Philips, Amsterdam, the Netherlands). Hepatic arteriography and superior mesenteric arterial portovenography were performed to define the size and location of tumour nodules. During the sequential scanning of the liver, radiocontrast medium (Telebrix, Laboratoire Guerbet, Aulnay-Sous-Bois, France; Ultravist, Schering, Berlin, Germany) 100–150 mL was injected using a power injector (CT9000 ADV, Liebel-Flarsheim, UK) to evaluate the vascularity of the tumour (tumour stain). The arteries supplying the tumour were catheterized superselectively followed by infusion of a mixture of 20–30 mg adriamycin (Carlo Erba, Milan, Italy) and 5–10 mL Lipiodol (Laboratoire Guerbet, Aulnay-Sous-Bois, France). The aim was to deliver a sufficient amount of emulsion to the tumoral areas without retrograde flow in each treatment cycle. Under fluoroscopic control, the feeding arteries were subsequently embolized with 2–3 mm strips of Gelfoam (Upjohn, Kalamazoo, MI, USA) until complete flow stagnation was achieved.
The equipment used for PAI was a commercially available ultrasound scanner with a puncture probe with guide device (Aloka, Tokyo, Japan). Sterile 50% acetic acid (Merck, Darmstadt, Germany) was injected with a 22-gauge spinal needle.22 Treatment was administered twice a week in an inpatient setting. Typically, one or two injections at a dose of 2–5 mL acetic acid were given during each treatment session. Acetic acid was slowly injected until the echogenic area appearing immediately after injection covered the entire tumour. After the injection was complete, the needle was left in place for 1–2 min to prevent reflux of acetic acid into peritoneal cavity. Two to four sessions were given for the lesions <3 cm, and three to five sessions were given for the lesions between 3 and 6 cm in each treatment cycle. For patients with multiple tumours, PAI was performed first on the main tumour (one with the largest diameter), followed by treating the smaller ones until all tumours were successfully treated.
Post-treatment follow-up included US scan and measurement of serum AFP levels every 2 months, and contrast-enhanced dynamic CT every 3–4 months. MRI and/or hepatic angiography were carried out as supplemental examinations. Residual viable tumour or new tumour occurring elsewhere in the liver was detected by the dynamic CT showing contrast enhancement during arterial phase or hepatic angiography showing tumour staining. Additional treatment cycle was administered once tumour recurrence or viable tumour was found in either treatment group. When patients developed a diffuse, infiltrative HCC, they were considered non-responsive and not treated.
The primary end points were progression-free survival and overall survival. Progression-free survival was calculated from the date of entry into the study to the date of disease progression, death, or the last follow-up visit. Overall survival was calculated from the date of entry into the study to the date of death or the last follow-up visit. Disease progression was defined as an increase of more than 25% in the tumour size of one or more lesions present at the start of therapy or the appearance of new intrahepatic or extrahepatic lesions.
Statistical methods and risk factor analysis
All statistical analyses were carried out using SPSS for Windows Release 11.0.1. Chi-square test or Fisher's exact test (two-tailed) was used for categorical data, and Mann–Whitney ranked sum test was used for continuous data. The survival was estimated by the Kaplan–Meier method and compared by the log-rank test. Patient- and tumour-related variables including age, sex, underlying hepatitis, Child-Pugh class, number and size of tumour, serum levels of albumin, bilirubin and AFP, prothrombin time ratio and treatment modality, were included for analysis. Factors that were significant or marginally significant (P < 0.1) in univariate survival analysis were entered into the Cox proportional hazard model to determine the adjusted risk ratios.
A total of 363 patients were identified according to the selection criteria. Among them, 53 patients who were treated with combination therapy using sequential TACE and PAI in a previous study were excluded from analysis,30 and the remaining 310 patients were allocated to either treatment after full explanation of the potential treatment-related complications according to the standards of the Declaration of Helsinki. There were 195 patients who chose to undergo TACE and the remaining 115 patients who did not receive TACE underwent PAI therapy. Their survival data were censored at the time of the last visit, and all statistical results were entered into the multivariate model for adjustment. The comparison of the baseline demographics between the two groups of patients is given in Table 1. Patients in TACE group more often had Child-Pugh class A reserve (75% vs. 57%, P = 0.002), a lower serum total bilirubin level (1.3 mg/dL vs. 1.6 mg/dL, P = 0.001) and multinodular lesions (47% vs. 34%, P = 0.024) than the patients in the PAI group. Comparisons among other parameters showed insignificant differences.
Table 1. Baseline demographics of study patients in two treatment groups
TACE, transarterial chemoembolization; PAI, percutaneous acetic acid injection. Numerical data are expressed as mean ± s.d.
No. of patients
67 ± 10
67 ± 10
Child-Pugh class (%)
Size of main tumour (%)
≤3 cm/>3 cm
Number of tumour (%)
Serum albumin (g/dL)
3.7 ± 0.5
3.5 ± 0.5
Serum bilirubin (mg/dL)
1.3 ± 0.5
1.6 ± 0.6
Prothrombin time ratio
1.0 ± 0.2
1.1 ± 0.3
Serum AFP (ng/mL)
1335 ± 7808
762 ± 3260
55 (2–98 033)
46 (2–30 422)
Number of treatment cycles
2.9 ± 0.9
3.1 ± 1.4
Hospital stay (days)
9.6 ± 3.1
9.9 ± 3.5
No serious complications that necessitated intensive care were encountered immediately after PAI or TACE therapy. Most patients experienced nausea and mild to moderate abdominal pain during or after either treatment, which could be controlled with additional analgesics. Elevation of liver enzymes was transient without overt jaundice. None of the patients in either group developed significant bleeding that was directly related to the procedure and required blood transfusion. Post-treatment fever (>38 °C) was usually mild and subsided within 3–4 days after appropriate management.
The overall 1-, 3- and 5-year survival rates were 90, 58 and 27% for TACE group and 91, 54 and 20% for PAI group, respectively (P = 0.508, Figure 1). Among all patients, Child-Pugh class B [relative risk (RR): 3.8, 95% confidence interval (CI): 1.9–7.4, P < 0.001] was the only independent prognostic factors predicting a poor survival in the adjusted Cox multivariate analysis. Treatment modality was not significantly associated with the outcome.
Overall survival and progression-free survival stratified by tumour size.
The distribution of the proportion of patients undergoing TACE and PAI according to size and number of tumour and Child-Pugh class is given in Table 2. Among the 129 patients with large (>3 cm) HCCs, patients in the TACE group more frequently had Child-Pugh class A reserve (78% vs. 51%, P = 0.014). Stratified analysis in this group showed that the survival was significantly better in the TACE group. The 1-, 3- and 5-year survival rates were 88, 56 and 34% for TACE group and 81, 33 and 0% for the PAI group, respectively (P = 0.018, Figure 2a). Patients in the TACE group tended to have a better progression-free survival than in the PAI group (P = 0.038, Figure 2b). Adjusted Cox multivariate analysis in this group disclosed that Child-Pugh class B (RR: 4.2, 95% CI: 2.3–7.7, P < 0.001) and PAI treatment (RR: 1.4, 95%: 1.0–1.9, P = 0.057) were independent prognostic factors predicting a decreased survival.
Table 2. Distribution of the proportion of patients undergoing TACE and PAI according to size and number of tumour and Child-Pugh class
For the 181 patients with small (≤3 cm) HCCs, there was no significant difference of overall survival between the two treatment groups. The 1-, 3- and 5-year survival rates were 92, 59 and 20% for TACE group and 97, 68 and 37% for PAI group, respectively (P = 0.265, Figure 3a). The progression-free survival between the two groups of patients was comparable (P = 0.146, Figure 3b). Adjusted Cox multivariate analysis in this group disclosed that Child-Pugh class B (RR: 2.8, 95% CI: 1.7–4.8, P < 0.001) was the only independent prognostic factor predicting a decreased survival.
Percutaneous injection therapy using acetic acid is an effective treatment modality for HCC.19–22 However, previous literature reporting the therapeutic efficacy of PAI is rather limited, and few studies have specifically compared the therapeutic efficacy between TACE and PAI for unresectable HCC.31 The current study included a large series of patients who underwent PAI or TACE for early to moderately advanced stage HCC to allow for a detailed comparison. Our results suggest that the overall survival was not significantly different between patients receiving either treatment. However, patients with large HCCs treated with TACE may have a better clinical outcome as shown in the survival analysis stratified by tumour size.
Percutaneous injection using ethanol or acetic acid is effective for HCC. Acetic acid has a strong ability to penetrate cells and can dissolve lipids and extract collagen from intra-tumoral septa and capsules that frequently contain viable cancer cells.20, 21 Therefore, acetic acid is superior to ethanol in treating HCC as the texture of tumour and its resistance to perfusion with ethanol may make percutaneous ethanol injection less effective.19 In addition, our recent observation showed that PAI has the advantage of reduced number of treatment sessions and a shorter hospital stay compared with that of percutaneous ethanol injection.22 A major limitation of percutaneous injection therapy is that it is preferred only for small HCC. There has been only one study to date that has assessed the effect of PAI for large (>3 cm) HCC;21 however, the case number in that report was considered too small to draw a firm conclusion. In this study, we have measured the treatment effect by comparing the progression-free survival and overall survival between patients treated with TACE and PAI. We found that the anti-tumoral effect of PAI tended to be less satisfactory among patients with large HCCs, as the TACE group had a better progression-free survival and, as a result, a favourable long-term outcome in this group. TACE might induce ischaemic tumour necrosis and is considered a major palliative treatment option for patients with unresectable HCCs.25, 26 It has the advantage of targeting single- or multi-nodular hypervascular HCCs in one treatment procedure. Hence, it is noteworthy that patients who underwent palliative percutaneous or transarterial anti-cancer therapy frequently may have viable (not completely ablated) tumours or might develop new lesions during the follow-up. However, it should not be considered a specific drawback for TACE or PAI therapy. Patients undergoing curative partial hepatectomy also often had a tumour recurrence after resection,3, 4 possibly related to the multiple metachronous or unnoticed synchronous tumours that could not even be detected by highly sensitive imaging modalities.32, 33
Tumour size is an important prognostic predictor in HCC patients undergoing surgical resection,3, 4 liver transplantation,6 TACE7–9 and percutaneous injection therapy.14–17 When the survival analysis was stratified by tumour size, we found that treatment modality was not a prognostic predictor in the small HCC group, whereas PAI therapy was a potential risk factor predicting a poor survival among patients with large HCCs. These findings are in accordance with previous studies suggesting that the treatment effect of percutaneous injection therapy was closely linked with tumour size and large HCCs should not be treated solely with PAI therapy.15, 16, 30, 34
A major limitation of this study is that the treatment allocation was not randomly assigned. To overcome this drawback, the risk factor analysis has been performed and adjusted with the Cox multivariate model to determine the independent effect of the prognostic predictors. Among patients with large HCCs, those who underwent PAI therapy had a decreased survival at a marginal significance after adjustment (P = 0.057). As patients in this group more often had Child-Pugh B cirrhosis which could confound the results, further randomized trials are needed to justify our findings.
Both TACE and PAI are relatively ineffective for advanced stage (size >6 cm) HCC, and these patients were excluded from analysis in this study. Histological studies of resected HCC specimens after TACE showed that complete necrosis occurred in only 22–50% of HCC lesions and that TACE is particularly effective in treating tumours <5 cm in diameter.35–37
The majority of HCC patients have coexisting chronic hepatitis or liver cirrhosis upon diagnosis.38 Decompensated liver cirrhosis (Child-Pugh class B) was consistently identified as a strong negative prognostic predictor in this study, and this result is in line with most previous literature. The liver functional reserve may gradually deteriorate after anti-cancer treatment and could act independently to affect the long-term outcome, given the tumours might have been adequately controlled.
In conclusion, our results indicate that the overall and progression-free survival rates are fairly comparable between patients undergoing TACE and PAI. However, patients with large HCCs treated with TACE tend to have a more favourable long-term outcome. For patients with small HCCs, TACE and PAI similar therapeutic efficacy and either method could be recommended as the primary treatment modality. Patients with decompensated cirrhosis have a decreased long-term survival. Careful pretreatment evaluation and patient selection are necessary to achieve a better outcome.
This study was supported by a grant from the Taipei Veterans General Hospital, Taipei, Taiwan.