Fax: (011) 81-6-6879-3259.
Interferon-α and 5-fluorouracil combination therapy after palliative hepatic resection in patients with advanced hepatocellular carcinoma, portal venous tumor thrombus in the major trunk, and multiple nodules
Version of Record online: 16 OCT 2007
Copyright © 2007 American Cancer Society
Volume 110, Issue 11, pages 2493–2501, 1 December 2007
How to Cite
Nagano, H., Miyamoto, A., Wada, H., Ota, H., Marubashi, S., Takeda, Y., Dono, K., Umeshita, K., Sakon, M. and Monden, M. (2007), Interferon-α and 5-fluorouracil combination therapy after palliative hepatic resection in patients with advanced hepatocellular carcinoma, portal venous tumor thrombus in the major trunk, and multiple nodules. Cancer, 110: 2493–2501. doi: 10.1002/cncr.23033
- Issue online: 19 NOV 2007
- Version of Record online: 16 OCT 2007
- Manuscript Accepted: 9 JUL 2007
- Manuscript Revised: 3 JUL 2007
- Manuscript Received: 26 FEB 2007
- Cancer Research from the Ministry of Culture and Science
- Ministry of Health and Welfare in Japan
- hepatocellular carcinoma;
- hepatic resection;
- interferon type 2 receptor;
- portal vein thrombosis;
- arterial infusion chemotherapy
The authors reported previously the beneficial effects of interferon (IFN)-α/5-fluorouracil (5-FU) combination therapy for patients with advanced hepatocellular carcinoma (HCC) who have tumor thrombi in the major portal branches. In this report, the authors describe the results from IFN/5-FU chemotherapy for patients who underwent palliative hepatic resection for advanced HCC with tumor thrombus in the main trunk of the portal vein and multiple nodules in the whole liver. In addition, they evaluated the correlation between the response to such therapy and expression of IFN-α type 2 receptor (IFNAR2).
From October 1999 to December 2004, 30 patients with advanced HCC, tumor thrombi in the main trunk of the portal vein, and multiple nodules in the whole liver (Vp4 and grade 3 intrahepatic metastases) were recruited for this study. They underwent palliative hepatic resection followed by at least 2 courses of IFN/5-FU. IFNAR2 expression levels were determined by immunohistochemistry.
No major treatment-related complications were noted. An objective response was noted in 10 patients (33.3%) and included a complete response in 6 patients (20%), a partial response in 4 patients (13.3%), no response in 1 patient (3.3%), and progressive disease in 19 patients (63.4%). IFNAR2 expression was detected in 20 of 30 patients (66.7%). There was a significant difference in overall survival between patients with positive and negative IFNAR2 expression cases (P < .0025), and a significant correlation was observed between IFNAR2 expression and response to IFN/5-FU combination therapy (P = .0199).
Adjunct IFN/5-FU therapy is a promising modality for patients with advanced HCC, tumor thrombi in the major trunk, and multiple nodules after palliative hepatic resection. The results from this study indicated that the response to such therapy seemed to be correlated with IFNAR2 expression. Cancer 2007. © 2007 American Cancer Society.
The prognosis for patients with advanced hepatocellular carcinoma (HCC) remains poor, particularly in patients who have tumor thrombi in the major trunk of the portal vein (Vp4).1–3 The mortality rate is very high in patients with unresectable tumors, and their quality of life is poor because of intractable ascites or esophageal bleeding. Even in patients who have resectable HCC, the prognosis is extremely poor despite aggressive surgery.4, 5 In such a situation, conventional therapies generally have no clinical effect on HCC associated with portal vein tumor thrombi (PVTT) because of poor efficacy and possible complications.6, 7 Arterial infusion chemotherapy also has been attempted, but its effectiveness still is unsatisfactory.9, 10 Therefore, a new strategy is required for patients who have intractable HCC and tumor thrombi in the major branch of the portal vein.
Several recent studies have indicated the beneficial effects of interferon (IFN)-α-based combination chemotherapies for HCC10–15 despite the lack of satisfactory results from IFN-α monotherapy.16 We also reported on the clinical efficiency of IFN-α and 5-fluorouracil (5-FU) combination therapy for advanced HCC with PVTT and intrahepatic metastasis.17–20 IFN-α suppresses the proliferation of all type I IFN receptor type 2 (IFNAR2)-positive cancer cell lines in vitro, an effect that is mediated through its high affinity to IFNAR2. Studies from our laboratories demonstrated that IFNAR2 expression in HCC tissues may be a useful predictor of response to IFN/5-FU combination therapy.19
The current study was an extension to our previous work,18–20 in which we examined the clinical effects of the combination therapy of subcutaneous IFN-α and arterial infusion of 5-FU after palliative hepatic resection in 30 patients who had HCC associated with Vp4 and multiple intrahepatic metastases (IM3). We also investigated the correlation between response to this therapy and expression of IFNAR2.
MATERIALS AND METHODS
Patients and Selection Criteria
The current investigation was a single-arm, open-label study that was based on our previous reports.18–20 Between October 1999 and December 2004, 30 patients with advanced HCC were enrolled. All patients had radiologically confirmed tumor thrombi in the main trunk of the portal vein (Vp4) and IM3. The diagnosis was based on liver function tests, serum α-fetoprotein (AFP), serum protein induced by vitamin K absence or antagonist-II (PIVKA-II), and imaging techniques, which included computed tomography (CT) scans, magnetic resonance imaging (MRI), hepatic angiography, and arterial portography. Consequently, these 30 patients underwent palliative reduction surgery with tumor thrombectomy in the main trunk of the portal vein to reduce tumor volume and to reopen the portal blood flow. IFN-α and 5-FU combination therapy for remnant multiple hepatomas was carried out after surgery. We used the following eligibility criteria for the selection of patients. 1) absence of extrahepatic metastases, 2) granulocyte count > 2500/μL or < 12,000/μL, 3) erythrocyte count > 8.0 g/dL, 4) platelet count > 8×104/μL, 5) glutamic oxaloacetic and pyruvic transaminase levels < 100 IU/L, 6) total bilirubin < 1.4 g/dL, 7) blood urea nitrogen < 30 mg/dL, 8) serum creatinine < 1.5 mg/dL, 9) successful implantation of intra-arterial catheter and drug delivery system, and 10) an Eastern Cooperative Oncology Group (ECOG) performance status from 0 to 2.21 These eligibility criteria were based on our previous studies.18, 19 All patients signed informed consent documents that were approved by the Institutional Review Board attesting that they were aware of the investigational nature of the study and were willing to try the combination therapy.
Treatment Protocol of IFN/5-FU Combination Therapy
In each of 30 patients, an intra-arterial catheter was inserted through the gastroduodenal artery during surgery or through the subclavian or femoral artery with a subcutaneously implanted drug-delivery system.22 Each patient received subcutaneous IFN-α (OIF; Otsuka Pharmaceutical Company, Tokushima, Japan) and an intra-arterial infusion of 5-FU (Kyowa Hakko Company, Tokyo, Japan). One cycle of treatment consisted of 4 weeks. IFN-α (5 × 106 U [5 MU]) was administered subcutaneously on Days 1, 3, and 5 of each week, resulting in a total dose of 60 MU per cycle. Continuous infusion chemotherapy (5-FU, 300 mg/m2 per day) through the proper hepatic artery was administered in the first and second weeks through a catheter connected to a subcutaneously implanted drug-delivery system. A 2- or 3-week rest period (cessation of drug therapy) separated the treatment cycles. All anticancer therapies were discontinued when adverse effects reached level 2 of the ECOG classification21 (with the exception of platelet and leukocyte counts < 40,000/mm3 and < 2000/mm3, respectively, because these parameters often were low before treatment because of associated cirrhosis).18
Evaluation of Response to IFN/5-FU Combination Therapy
A pretreatment evaluation was conducted at the commencement of the IFN-α/5-FU protocol, and a posttreatment evaluation was conducted after the completion of 2 cycles of treatment, almost 3 months later. The evaluation included CT or MRI studies and an assessment of changes in serum tumor markers, such as AFP and PIVKA-II. All patients had their results compared at these 2 time points for the evaluation of antitumor effects. The objective response was classified according to ECOG criteria.21 A complete response (CR) was defined as normalization of tumor marker levels and disappearance of all tumors and portal vein thrombosis on CT and/or MRI studies. A partial response (PR) was defined as a decrease in tumor marker levels and a decrease between 50% and 99% in 2-dimensional tumor measurement. No change (NC) represented a decrease < 50% or anincrease < 25% in tumormeasurements, and progressivedisease (PD) represented an increase > 25%. In addition, we evaluated progression-free and overall survival rates. Follow-up was from 15 to 75 months.
Reagents and Immunohistochemistry
Rabbit polyclonal antihuman IFNAR2 antibody (OCT4813; Otsuka Pharmaceutical Company) and its blocking peptide were prepared according to the report by Novick et al.23 The expression of IFNAR2 was examined in all 30 resected tumor samples by immunohistochemistry, which was carried out according to the method described previously by investigators in our laboratories.19, 24–26 All slides were interpreted by 1 of 2 investigators (H.W. or H.O.) in a blinded manner without knowledge of the clinical or pathologic parameters.
The Breslow-Gehan-Wilcoxon univariate test was used to examine the possible correlations between the effect of therapy (CR/PR vs NC/PD), Child-Pugh score, serum AFP, serum PIVKA-II, Okuda score, Cancer of the Liver Italian Program (CLIP) score,3 and the expression of IFNAR2. Survival curves were constructed using the Kaplan-Meier method. Differences in distribution between groups were compared using the chi-square test, and differences in mean values were calculated with the Student t test. All data were expressed as the mean ± standard error of the mean (SEM). A P value < .05 denoted a statistically significant difference.
The preoperative clinical characteristics of the participating patients are summarized in Table 1. The median age of patients was 56 years (range, 29–71 years). AFP and/or PIVKA-II expression levels were abnormal in all 30 patients. Preoperative liver function tests (mean ± SEM values) were as follows; serum albumin, 3.6 ± 0.4 g/dL; serum total bilirubin, 1.0 ± 0.1 mg/dL; prothrombin time, 71.7% ± 12.3%; hepaplastin test, 75.4% ± 16.2%; and indocyanine green retention rate at 15 minutes, 18.5% ± 11.0%.
|Patient||Age||Sex||T||M||N||Vp||IM||Stage||Alb, g/dL||T.Bil, mg/dL||PT/HPT,%||ICGR-15,%||AFP, ng/mL||PIVKA-II mAU/mL||Virus|
Clinical Response to Combination Therapy
Table 2 summarizes the operative procedure, postoperative pathologic diagnosis, and clinical response to IFN/5-FU combination therapy. All patients completed at least ≥2 cycles of IFN/5-FU combination therapy after palliative surgery. For patients who had a clinical response, we continued this combination therapy; whereas, for patients who had no response, we stopped treatment after the completion of the second cycle because of extensive progression of HCC.
|Patient||Operation||Histology||Response to IFN/5-FU||IFNAR2|
|1||Extended left lobectomy||Ed III (poor)||Chr glisonitis||PR||1|
|2||Right lobectomy||Ed IV (undiff)||Normal||PD||1|
|3||Extended posterior segmentectomy||Ed III (poor)||B′−||PD||1|
|4||Extended posterior segmentectomy||Ed II (mod)||B′−||PR||1|
|5||Extended left lobectomy||Ed III (poor)||CIH||CR||1|
|6||Extended posterior segmentectomy||Ed III (poor)||B′+||CR||1|
|7||Left lobectomy||Ed III (poor)||B′+||PD||0|
|8||Right lobectomy||Ed III (poor)||B−||PD||1|
|9||Extended right lobectomy||Ed III (poor)||CIH||CR||1|
|10||Anterior segmentectomy||Ed III (poor)||B+||PD||1|
|11||Extended left lobectomy||Ed III (poor)||B′+||PD||1|
|12||Left lobectomy||Ed III (poor)||B−||PR||2|
|13||Left lobectomy||Ed IV (undiff)||B′−||PD||1|
|14||Right lobectomy||Ed III/IV (poor)||B+||PD||0|
|15||Extended left lobectomy||Ed III/IV (poor)||CIH||PD||0|
|16||Right lobectomy||Ed III/IV (poor)||CIH||CR||2|
|17||Right lobectomy||Ed III (poor)||CIH||CR||1|
|18||Right lobectomy||Ed III (poor)||CIH||PD||0|
|19||Extended right lobectomy||Ed III (poor)||CIH||PD||1|
|20||Right lobectomy||Ed III (poor)||B′−||PD||0|
|21||Right lobectomy||Ed III (poor)||CIH||PD||0|
|22||Left lobectomy||Ed III (poor)||CIH||CR||1|
|23||Right lobectomy||Ed III (poor)||CIH||PD||0|
|24||Right lobectomy||Ed III (poor)||B′−||PR||1|
|25||Extended left lobectomy||Ed III (poor)||B+||NC||1|
|26||Right lobectomy and pancreatoduodenectomy||Ed IV (undiff)||Normal||PD||0|
|27||Right lobectomy||Ed III (poor)||CIH||PD||0|
|28||Right lobectomy||Ed III/IV (poor)||B′−||PD||1|
|29||Right lobectomy||Ed III (poor)||CAH+||PD||1|
|30||Right lobectomy||Ed III (poor)||CAH+||PD||0|
At the start of IFN/5-FU treatment, all 30 patients had multiple intrahepatic lesions in the residual liver after palliative resection. The average and median tumor size of the largest nodule were 16 mm and 15 mm, respectively (n = 30 patients; range, 10–32 mm), as detected on CT or MRI studies. With regard to the clinical effect, 10 patients (33.3%) had an objective response, 6 patients (20%) had a CR, 4 patients (13.3%) had aPR, 1 patient (3.3%) had NC, and 19 patients (63.3%) had PD. With respect to the time to disease progression, the median progression-free survival was 2 months, and the 1-, 2-, and 3-year progression-free survival rates were 20%, 16%, and 0%, respectively. Furthermore, the median overall survival was 9.5 months, and the 1-, 2-, and 3-year survival rates were 40%, 28.5% and 21.4%, respectively. The median progression-free survival for patients who had a response (CR/PR; n = 10 patients) was 17.5 months, and the median progression-free survival of patients who had NC/PD (n = 20 patients) was 2 months. The 1-, 2-, and 3-year progression-free survival rates of patients who had CR/PR were 60%, 48%, and 0%, respectively; and all 3 rates were 0% for the patients who had NC/PD. The tumor burden generally was very small because of the excision of the main tumor by reduction surgery. However, there was no correlation between antitumor effect and tumor size in the remnant liver.
The median survival was 29 months for patients who achieved a CR/PR (n = 10 patients) and 6 months for patients who had NC/PD (n = 20 patients). The median follow-up of the patients who survived was 32 months. The 1-, 2-, and 3-year survival rates for patients who achieved a CR/PR were 100%, 77.8%, and 58.3%, respectively; and the rates for patients who had NC/PD were 10%, 0%, and 0%, respectively. The time to progression and overall survival curves are shown in Figures 1 and 2, respectively. There were significant differences in the time to progression and overall survival between responders (CR/PR) and nonresponders (NC/PD; P < .0001).
None of our patients developed side effects related to catheter insertion or subcutaneous implantation of the drug-delivery system. Grade 1 leukopenia, thrombocytopenia, or myelosuppression was observed in 8 patients, but none of those episodes forced the termination of therapy or required treatment with granulocyte-colony stimulating factor. Other adverse effects, including stomatitis or diarrhea, were mostly grade 1 and clinically manageable in general. Fever was observed commonly but was controlled easily by nonsteroidal anti-inflammatory drugs before IFN injection. No depression because of IFN administration was observed in any of the 30 patients.
Correlation Between IFNAR2 Immunostaining Pattern and Prognosis
For each tissue section, the intensity of IFNAR2 immunostaining was scored on a scale from 0 to 2, in which 0 represented no or faint immunostaining, 1 represented moderate staining, and 2 indicated strong staining, based on our previous reports.19, 24, 25 Table 2 shows the IFNAR2 expression level in each of the 30 patients. IFNAR2 expression was noted in 10 of 30 patients (33.3%). The median progression-free survival rate was 8 months for IFNAR2-positive patients and 2 months for IFNAR2-negative patients. The time to progression survival rates at 1, 2, and 3 years for IFNAR2-positive patients (n = 20 patients) were 30%, 24%, and 0%, respectively, and were significantly higher than the respective rates for IFNAR2-negative patients (n = 10 patients; 0% for all 3 rates; P = .0038) (Fig. 3A). The median overall survival rate was 16 months for IFNAR2-positive patients and 5.5 months for IFNAR2-negative patients. The overall survival rates at 1 year, 2 years, and 3 years among IFNAR2-positive patients (n = 20 patients; 60%, 42.8%, and 32.1%, respectively) were significantly higher than the respective rates among IFNAR2-negative patients (n = 10 patients; 0% for all 3 rates; P < .0025) (Fig. 3B).
Clinical and Pathologic Correlations
Finally, we compared the responders (CR/PR; n = 10 patients) with the nonresponders (NC/PD; n = 20 patients) in terms of serum AFP (within normal range; < 5 ng/mL), serum PIVKA-II (normal range; < 45 mAU/mL), Child-Pugh scores, Okuda scores, CLIP scores, and IFNAR2 expression in univariate analysis (Table 3). Serum AFP, PIVKA-II, Child-Pugh scores, Okuda scores, and CLIP scores did not correlate with the response to combination therapy. Conversely, IFNAR2 expression correlated significantly with the response to IFN/5-FU combination therapy (P = .0199). Thus, the expression level of IFNAR2 was the sole factor that influenced the response to the combination therapy.
|Characteristic||No. of patients||P|
|CR/PR (n = 10)||NC/PD (n = 20)|
With regard to the patient selection criteria followed in the current study27 and in our previous investigations,18–20 we considered the presence of 3 types of advanced HCC with PVTT in the main trunk for the analysis of tumor progression (Fig. 3). The 3 types were defined as follows: type I, PVTT with multiple nodules in the bilateral lobes; type II, PVTT with a huge mass in 1 lobe and no intrahepatic metastatic nodules in the other lobe; and type III, PVTT with a huge mass in 1 lobe and multiple intrahepatic metastatic nodules in the other lobe. Patients with type I PVTT received IFN/5-FU combination treatment: An antitumor effect was noted in 43.7% of patients, and a significant survival benefit was noted in 55 patients from our previous study.19 Patients with type II PVTT underwent hepatic surgery to remove the huge mass followed by IFN/5-FU combination treatment as a postoperative adjunct. In this series, 100% survival rate at 1 year was achieved in 15 patients.20 The patients in the current study had type III PVTT, which is considered the most advanced stage of HCC. In such patients, the main trunk of the portal vein already is packed with PVTT, and they have rapid worsening to liver failure because of the decrease in portal blood flow. These patients are prone to rupture of esophageal varices because of increased portal venous pressure. In general, most patients with advanced HCC can be treated only with best supportive care. However, for selected patients with type III PVTT who have liver function good enough to endure hepatic lobectomy, a multimodal treatment that includes surgery may be possible. Consequently, the 30 patients in the current study underwent palliative reduction surgery, which consisted of bisegmentectomy or trisegmentectomy with extirpation of PVTT to reduce tumor volume and to reopen the portal blood flow. IFN/5-FU combination therapy for remnant multiple hepatomas in the residual liver was carried out after surgery. With regard to these 30 patients, none developed any major complications, and they started the IFN/5-FU combination therapy from 3 to 5 weeks after surgery. We also demonstrated the beneficial effects of IFN/5-FU combination therapy in our patients. The efficacy of such treatment was 33.3% in our patients with highly advanced HCC. Thus, the combination treatment with IFN-α and 5-FU after hepatic palliative surgery had a marked antitumor effect with an encouraging response rate. Furthermore, the clinical response translated into survival benefits, as shown in Figures 1 and 2.
It should be noted, however, that the remaining 20 of 30 patients (66.7%) in our study did not respond to the combination therapy. Among the 20 nonresponders, there was only 1 patient with NC despite the mostly chemoresistant disease. We believe that this finding may be attributed as follows: The HCC in this series was far advanced, and HCC progression was extremely rapid and aggressive despite palliative reduction surgery. Under such conditions, almost all nonresponders died within 12 months; 12 of 20 patients (60.0%) died within 6 months. For nonresponders to this treatment, however, the survival was too short to allow the receipt of another treatment modality. Therefore, accurate prediction of chemosensitivity is desirable not only to prevent the loss of a limited chance for another possible treatment but also to avoid potentially serious side effects. However, currently, there are no suitable markers with which to distinguish between patients who are likely and patients who are unlikely to respond to this combination chemotherapy.
Several mechanisms for the anticancer effects of IFN-α, with or without 5-FU, have been proposed.28–37 We demonstrated previously that IFN-α and 5-FU synergistically inhibited tumor cell proliferation with cell cycle arrest38 and induced apoptosis by regulating apoptosis-related molecules.39 We also reported that tumor necrosis factor-related apoptosis inducing ligand, its receptor pathway,40 and Fas and the Fas-ligand pathway41 partially contributed to the antitumor effects of IFN-α and 5-FU combination therapy. Moreover, IFN-α suppressed proliferation in all type I IFNAR2-positive HCC cell lines in vitro through mechanisms related to apoptosis or cell cycle inhibition.42 The importance of IFNAR2 expression for the anticancer effect of IFN/5-FU was highlighted in a similar situation in our previous report.38, 39, 43 These findings suggest that the antineoplastic effects of IFN-α are likely to be mediated through its high-affinity membrane type I receptor, IFNAR2.44 In this regard, we postulated that IFNAR2 expression in HCC tissues may be a useful predictor with which to distinguish between potential responders and nonresponders to IFN/5-FU combination therapy. On the basis of these results, we investigated the correlation between IFNAR2 expression and the effect of IFN/5-FU combination therapy using immunohistochemical analysis, and the results showed a good correlation.
Several markers for the prediction of tumor recurrence and prognosis have been identified for patients with HCC. Levy and Sherman45 reported that the CLIP classification for HCC is easier to implement and more accurate than the Okuda classification. In addition, Koike et al.46 suggested that the serum PIVKA-II level is the most useful clinical parameter for predicting the development of portal vein invasion. To investigate the applicability of these clinical parameters, AFP, PIVKA-II, Okuda scores, and CLIP scores were used in the current study to assess the clinical effects of IFN/5-FU combination therapy. The results indicated that expression of IFNAR2 was the only significant predictor of clinical outcome of IFN/5-FU combination therapy; and our survival analysis indicated a significant role of IFNAR2 expression on prognosis. These results suggest that the expression of IFNAR2 may be a potentially useful predictor of response to IFN/5-FU combination therapy. In our recent report using microarray analysis, several genes involved in IFN signaling transduction were identified as useful for molecular prediction of response to IFN/5-FU combination therapy.47
In conclusion, the current study has demonstrated the efficacy of IFN/5-FU combination therapy after surgery for patients with advanced HCC who have tumor thrombi in major branches of the portal vein. The results also indicated that the clinical response to such therapy is correlated significantly with the expression of IFNAR2 in patients with HCC.
We thank Dr. Yasukazu Ohmoto from the First Institute of New Drug Research and Otsuka Pharmaceutical Company, Ltd. for providing antihuman interferon-α receptor 2 antibody (OCT4813)
- 1Liver Cancer Study Group of Japan. The General Rules for the Clinical and Pathological Study of Primary Liver Cancer, 4th ed. Tokyo, Japan: Kanehara-Shuppan, 2000.
- 7The safety and efficacy of transcatheter arterial chemoembolization in the treatment of patients with hepatocellular carcinoma and main portal vein obstruction: a prospective controlled study. Cancer. 1997; 79: 2087–2094., , , et al.
- 13Combined therapy consisting of intraarterial cisplatin infusion and systemic interferon-(for hepatocellular carcinoma patients with major portal vein thrombosis or distant metastasis. Cancer. 2000; 88: 1986–1991., , , et al.
- 46Des-gamma-carboxy prothrombin as a useful predisposing factors for the development of portal venous invasion in patients with hepatocellular carcinoma: a prospective analysis of 227 patients. Cancer. 2001; 91: 561–569., , , et al.