- Email: [email protected]
Predictive score for the development of hepatocellular carcinoma in patients with liver cirrhosis
HEPATOLOGY Vol. 34, No. 4, Pt. 2, 2001
AASLD ABSTRACTS
179A
21
22
HIGH HETEROGENEITY OF GENE EXPRESSION PROFILES IN HCV AND HBV RELATED HEPATOCELLULAR CARCINOMA. Oona Delpuech, Faculty Necker INSERM U 370, Paris France; Jean Feuillard, Unit of Biological H4matology HOpital Avicenne, Bobigny France; Dina Kremsdorf, Christian Brechot, Faculty Necker INSERM U370, Paris France
SYSTEMIC ADMINISTRATION OF LIPOSOME-ENCAPSULATED OK-432 (BIOLOGICAL RESPONSE MODIFIER) PROLONGS THE SURVIVAL OF RATS W I T H HEPATOCELLULAR CARCINOMA THROUGH THE INDUCTION OF IFN-y-PRODUCING HEPATIC LYMPHOCYTES. Takafumi Ichida, Kazuhiro Uehara, Satoshi Sugahara, Toru Ishikawa, Satoshi Yamagiwa, Yuhei Yoshida, Minoru Nomoto, Hisami Watanabe, Toru Abo, Niigata University School of Medicine, Niigata Japan; Masashi Katoh, Hiroshi Satoh, Niigata University Medical Hospital, Niigata Japan; Hitoshi Asakura, Niigata University School of Medicine, Niigata Japan
Despite some recent studies based on eDNA microarray technology, it is still unclear whether or not gene expression profile in HCV or HBV-related HCCs shows some specificity and whether or not development of HCC on a background of cirrhosis influences such a gene profile. To address this issue we have analysed: 1) HCV and HBV related HCCs and: 2) HCCs developed on cirrhotic and non cirrhotic tissue. Methods: W e compared gene expression profiles using Cancer 1.2 microarray (CLONTECH, 1,176 genes) of paired tumour and non-turnout tissues from 15 patients, with HCV(n=8), HBV(n=6) or HCV+HBV (n= 1) related HCCs, HCCs developed on cirrhotic (n= 11) or non-cirrhotic ( n = 4 ) tissues. Experiments have been performed in duplicate and analysed by hierarchical clustering. Comparison of gene expression profiles was performed to identify genes related to the viral etiology or to the development of HCC on a background of cirrhosis. A statistical analysis was performed using Chi2 test. Results: 1/Among the 775 genes differencially expressed between tumour and non-tumour tissues, only 3 genes were found up regulated and 10 down regulated in 40% and more of the experiments. 2/ The hierarchical clustering analysis showed the heterogeneity of gene expression profiles in HCCs. 3/However, the individual analysis showed that the expression of a few- genes was found related to HCV or HBV- related HCCs (such as BIGH3, SOCS2, cyclin-dependant kinase inhibitor 1, insulin induced protein l ) and to the development of HCC on cirrhosis or not (such as: alpha 1 catemn, adehyde oxidase) . Conclusions: Our resuhs l / s h o w the heterogeneity in gene expression profile in Hepatocellular carcinoma, suggesting multiple regulating pathways involved in HCC development, 2/suggest abnormal expression of only a limited number of genes in HCC, in contrast to other human cancer 3 / d o not support the hypothesis of a specific pattern of gene expression in HCV related liver carcinogenesis.
Introduction: Biological response modifier, OK-432, is known to increase the host anti-minor response. We previously reported that systemic administration of liposome-encapsulated OK-432 (OK-Lipo) specifically induced hepatic lymphocytes in mice (Yamagiwa S, et al.J. Gastroenterol. Hepatol. 2000; 15: 542-49). Here, we aimed to investigate the anti-tumor effect of OK-Lipo on hepatocellular carcinomas (HCC) in experimental rats. During experiment, what kind of hepatic lymphocytes will be involved for the long survival effect? This stud), is anticipated some possibility of cancer prevention from necro-inflammatory condition and immunotherapy. Materials and Methods: Diethylnitrosamine was administered for 12 weeks to all rats (n= 36). Rats were divided into 3 groups of twelve rats each. One group was injected with OK-Lipo from week 5 (OK-Sw group) and another from week 9 (OK-9w group). A control group was injected with saline from week 5 (Non-OK group). At week 13, five rats from each group were used for histological analysis and immunofluorescence assays (surface phenotypic analysis and intracellular cytokine analysis of the mononuclear cells in the liver, spleen and peripheral blood). The remaining rats were observed for the remainder of their survival period. Resuhs: The mean survival times of Non-OK, OK-5w, and OK-9w groups differed significantly (98.0 + 5.3 days, 116.0 _+ 5.8 days, and i06.0 + 5.4 days, respectively, p<0.01). Macroscopic tumor growth in the liver was markedly different between groups. Histological examination revealed many apoptotic tumor cells, infiltration of lymphocytes and macrophages in the OK-5w group. The two-color immunofluorescence assay showed that the proportion of natural killer (NK) cells (p<0.05) and IFN-y-producing cells (p<0.05) in the liver was significantly increased in the OK-Sw group. Discussion: These findings show that systemic administration of OK-Lipo contributed to prolong the survival of rats with HCC. OK-Lipo induced NK ceils and IFN-y-producing cells specifically- in the liver, and these cells seemed to reduce hepatocarcinogenesis and tumor growth. Our study showed that OK-Lipo induced NK cells and IFN-'y-producing cells specifically in the liver. These ceils probably inhibit hepatocarcinogenesis and tumor development by enhancing celIular immunity, prolonging the survival of rats with HCC. In clinical use, inhibition of HCC in patients with chronic hepatitis or cirrhosis is expected from the long-term administration of OK-Lipo. Though further investigations are needed to determine the way of administration and DDS of OK-Lipo~ it may be a useful treatment for both the immunoprevention and immunotherapy of HCC.
23
24
PREDICTIVE SCORE FOR THE DEVELOPMENT OF HEPATOCELLULAR CARCINOMA IN PATIENTS W I T H LIVER CIRRHOSIS. Manuel Rodriguez, Liver Unit, Hospital Central de Asturias, Oviedo Spain; Rosario F Velazquez, Hospital Central de Asturias, Oviedo Spain; Carmen A Navascues, Hospital de Cabuenes, Gijon Spain; Antonio Linares, Ramon Perez, Nieves G Sotorrio, Isabel Martinez, Monica G Espiga, Luis Rodrigo, Hospital Central de Astnrias, Oviedo Spain
OVER EXPRESSION OF MACROPHAGE MIGRATION INHIBITORY FACTOR ON THE HEPATOCYTES IN THE PRECANCEROUS AND CANCEROUS STATE: RELEVANCE TO 1MMUNOPATHOGENESIS OF HEPATOCELLULAR CARCINOMA. Akbar Fazle, Norio Horiike, Hiroshi Matsubara, Michiataka Kojiro, Masanori Abe, Morikazu Onji, Ehime Univ Sch of Medicine, Ehime Japan
BACKGROUND: The risk of developing hepatocellular carcinoma (HCC) in patients with liver cirrhosis varies in different series. Programs for early diagnosis of HCC have been widely accepted, although their benefit still remains unclear. A greater knowledge of the risk factors for development of HCC in patients with liver cirrhosis could improve the efficacy of such programs. AIMS: To identify groups of patients with liver cirrhosis with different risk for developing HCC during a 4 year period. METHODS: Between 1.9.92 and 1.9.99, 463 patients with liver cirrhosis (Child's grade A or B, and 40-65 years old), were included in a surveillance program for HCC based on abdominal uhrasonography and AFP determination every 3 or 6 months. The etiology of the cirrhosis was: Alcoholic (n=273), HCV infection ( n ~ 137), HBV infection ( n = 33) and others (n= 20). Patients were followed up until the appearance of HCC, death, inclusion on a waiting list for liver transplant or lost in the course of follow-up. The predictive value of different risk-prognostic variables was evaluated by means of the Kaplan-Meier Survival and Cox Regression Analysis. RESULTS: Thirty eight patients developed HCC during a follow-up period of 4 years, tn the multivariate analysis, 4 variables showed independent predictive value for the development of HCC: age ~> 55 years, presence of HCV antibodies, prothrombin activity -< 75%, and platelets < 7 5 × 103/ram3. With the coefficients obtained in the Cox regression, a predictive score, ranged from 0 to 4.71, was designed. The score identified 2 groups of patients: 1)Low-risk group, when it was ~<2.3 points (n = 270, 4 HCC, cumulative incidence of HCC at 4 years of 2.3%); 2)High-risk group, when the score was > 2.3 (n = 193, 34 HCC, cumulative incidence of HCC at 4 years of 30.1%)(p=0.0001). CONCLUSIONS: A simple score made up of four clinical and biological variables, allowed us to distinguish two groups of cirrhotic patients with high and low risk for the development of hepatocellular carcinoma. This score can be useful to stablish a subset of cirrhotic patients in whom a surveillance program for early detection of HCC can be unjustified.
Background/Aim: Macrophage migration inhibitory factor (MIF), a T-ceil derived cytokine, is related with the induction of inflammation and immunity as well as cell growth and ceil differentiation. A role of MIF during carcinogenesis has been postulated because MIF is over expressed at the malignant tissues from patients with adenocarcinoma of the lung, uveal melanoma, prostatic tumor and breast carcinoma. However, none of the studies have checked the expression of MIF in precancerous and as well as as in cancerous tissue. Hepatocellular carcinoma (HCC) represents the malignant transformation of hepatocytes and usually occur as a consequence of chronic inflammation of the liver. Most of the patients with HCC suffer from liver cirrhosis (LC) before developinmg HCC and LC represent a definite precancerous state of HCC. Subjects and Methods: MIF in the sera was estimated from a total of 55 patients with hepatocellular carcinoma (HCC), 26 patients with liver cirrhosis (LC), 41 patients with chronic hepatitris (CH), 29 patients with gastrointestinal cancer (esophageal cancer; 14, colon cancer: 15) and 45 normal controls. Peripheral blood mononuclear cells (PBMC) from patients with HCC and normal controls were cultured for 5 days to see their capacity to produce MIF. The levels of MIF was estimated in the sera and culture supernatant by a highly sensitive ELISA method in which the microtiter plates were coated with mouse monoclonal antibody to human MIF and was expressed as ng/ml. Tissue expression of MIF was studied by enzyme-immunohistochemistry using biotin-conjugated anti-human MIF antibody. As a part of antigen retrieval technique, tissues were autoclaved to get proper signaling for MIF. Results: The mean levels of MIF were significantly higher in the sera from patients with HCC (25.6± 15.3 ng/ml), LC (18.9 +_ 10.7 ng/m) and CH (13.8_+8.5 ng/ml) compared to gastrointestinal cancer (6.8+_7.5 ng/ml) aud normal controls (5.6-+ 1.2 ng/ml) (p<0.01). Serial study revealed that the levels of MIF is the sera was increased along with the progression of luver diseases. Immunohistochemical study showed that few infihrating mononuclear cells in the liver tissues from patients with CH expressed MIF. On the other hand, MIF staining was almost undetectable on the hepatocytes from patients with CH. In contrast, most of the hepatocytes from patients with LC expressed very high levels of MIF. Malignant hepatocytes in the cancer nodules and hepatocytes in the noncancerous cirrhotic nodules form patients with HCC also showed very strong expression of MIF. Concliusions: This study has shown that MIF is over expressed on the malignant hepatocytes from patients with HCC. The over expression of MIF on the hepatocytes from patients with LC, a precancerous state of HCC, indicates that MIF over expression may be related with the hepatocarcinogenesis. The levels of of MIF in the sera from patients with CH and LC may be used as a marker of disease progression.
AASLD ABSTRACTS
179A
21
22
HIGH HETEROGENEITY OF GENE EXPRESSION PROFILES IN HCV AND HBV RELATED HEPATOCELLULAR CARCINOMA. Oona Delpuech, Faculty Necker INSERM U 370, Paris France; Jean Feuillard, Unit of Biological H4matology HOpital Avicenne, Bobigny France; Dina Kremsdorf, Christian Brechot, Faculty Necker INSERM U370, Paris France
SYSTEMIC ADMINISTRATION OF LIPOSOME-ENCAPSULATED OK-432 (BIOLOGICAL RESPONSE MODIFIER) PROLONGS THE SURVIVAL OF RATS W I T H HEPATOCELLULAR CARCINOMA THROUGH THE INDUCTION OF IFN-y-PRODUCING HEPATIC LYMPHOCYTES. Takafumi Ichida, Kazuhiro Uehara, Satoshi Sugahara, Toru Ishikawa, Satoshi Yamagiwa, Yuhei Yoshida, Minoru Nomoto, Hisami Watanabe, Toru Abo, Niigata University School of Medicine, Niigata Japan; Masashi Katoh, Hiroshi Satoh, Niigata University Medical Hospital, Niigata Japan; Hitoshi Asakura, Niigata University School of Medicine, Niigata Japan
Despite some recent studies based on eDNA microarray technology, it is still unclear whether or not gene expression profile in HCV or HBV-related HCCs shows some specificity and whether or not development of HCC on a background of cirrhosis influences such a gene profile. To address this issue we have analysed: 1) HCV and HBV related HCCs and: 2) HCCs developed on cirrhotic and non cirrhotic tissue. Methods: W e compared gene expression profiles using Cancer 1.2 microarray (CLONTECH, 1,176 genes) of paired tumour and non-turnout tissues from 15 patients, with HCV(n=8), HBV(n=6) or HCV+HBV (n= 1) related HCCs, HCCs developed on cirrhotic (n= 11) or non-cirrhotic ( n = 4 ) tissues. Experiments have been performed in duplicate and analysed by hierarchical clustering. Comparison of gene expression profiles was performed to identify genes related to the viral etiology or to the development of HCC on a background of cirrhosis. A statistical analysis was performed using Chi2 test. Results: 1/Among the 775 genes differencially expressed between tumour and non-tumour tissues, only 3 genes were found up regulated and 10 down regulated in 40% and more of the experiments. 2/ The hierarchical clustering analysis showed the heterogeneity of gene expression profiles in HCCs. 3/However, the individual analysis showed that the expression of a few- genes was found related to HCV or HBV- related HCCs (such as BIGH3, SOCS2, cyclin-dependant kinase inhibitor 1, insulin induced protein l ) and to the development of HCC on cirrhosis or not (such as: alpha 1 catemn, adehyde oxidase) . Conclusions: Our resuhs l / s h o w the heterogeneity in gene expression profile in Hepatocellular carcinoma, suggesting multiple regulating pathways involved in HCC development, 2/suggest abnormal expression of only a limited number of genes in HCC, in contrast to other human cancer 3 / d o not support the hypothesis of a specific pattern of gene expression in HCV related liver carcinogenesis.
Introduction: Biological response modifier, OK-432, is known to increase the host anti-minor response. We previously reported that systemic administration of liposome-encapsulated OK-432 (OK-Lipo) specifically induced hepatic lymphocytes in mice (Yamagiwa S, et al.J. Gastroenterol. Hepatol. 2000; 15: 542-49). Here, we aimed to investigate the anti-tumor effect of OK-Lipo on hepatocellular carcinomas (HCC) in experimental rats. During experiment, what kind of hepatic lymphocytes will be involved for the long survival effect? This stud), is anticipated some possibility of cancer prevention from necro-inflammatory condition and immunotherapy. Materials and Methods: Diethylnitrosamine was administered for 12 weeks to all rats (n= 36). Rats were divided into 3 groups of twelve rats each. One group was injected with OK-Lipo from week 5 (OK-Sw group) and another from week 9 (OK-9w group). A control group was injected with saline from week 5 (Non-OK group). At week 13, five rats from each group were used for histological analysis and immunofluorescence assays (surface phenotypic analysis and intracellular cytokine analysis of the mononuclear cells in the liver, spleen and peripheral blood). The remaining rats were observed for the remainder of their survival period. Resuhs: The mean survival times of Non-OK, OK-5w, and OK-9w groups differed significantly (98.0 + 5.3 days, 116.0 _+ 5.8 days, and i06.0 + 5.4 days, respectively, p<0.01). Macroscopic tumor growth in the liver was markedly different between groups. Histological examination revealed many apoptotic tumor cells, infiltration of lymphocytes and macrophages in the OK-5w group. The two-color immunofluorescence assay showed that the proportion of natural killer (NK) cells (p<0.05) and IFN-y-producing cells (p<0.05) in the liver was significantly increased in the OK-Sw group. Discussion: These findings show that systemic administration of OK-Lipo contributed to prolong the survival of rats with HCC. OK-Lipo induced NK ceils and IFN-y-producing cells specifically- in the liver, and these cells seemed to reduce hepatocarcinogenesis and tumor growth. Our study showed that OK-Lipo induced NK cells and IFN-'y-producing cells specifically in the liver. These ceils probably inhibit hepatocarcinogenesis and tumor development by enhancing celIular immunity, prolonging the survival of rats with HCC. In clinical use, inhibition of HCC in patients with chronic hepatitis or cirrhosis is expected from the long-term administration of OK-Lipo. Though further investigations are needed to determine the way of administration and DDS of OK-Lipo~ it may be a useful treatment for both the immunoprevention and immunotherapy of HCC.
23
24
PREDICTIVE SCORE FOR THE DEVELOPMENT OF HEPATOCELLULAR CARCINOMA IN PATIENTS W I T H LIVER CIRRHOSIS. Manuel Rodriguez, Liver Unit, Hospital Central de Asturias, Oviedo Spain; Rosario F Velazquez, Hospital Central de Asturias, Oviedo Spain; Carmen A Navascues, Hospital de Cabuenes, Gijon Spain; Antonio Linares, Ramon Perez, Nieves G Sotorrio, Isabel Martinez, Monica G Espiga, Luis Rodrigo, Hospital Central de Astnrias, Oviedo Spain
OVER EXPRESSION OF MACROPHAGE MIGRATION INHIBITORY FACTOR ON THE HEPATOCYTES IN THE PRECANCEROUS AND CANCEROUS STATE: RELEVANCE TO 1MMUNOPATHOGENESIS OF HEPATOCELLULAR CARCINOMA. Akbar Fazle, Norio Horiike, Hiroshi Matsubara, Michiataka Kojiro, Masanori Abe, Morikazu Onji, Ehime Univ Sch of Medicine, Ehime Japan
BACKGROUND: The risk of developing hepatocellular carcinoma (HCC) in patients with liver cirrhosis varies in different series. Programs for early diagnosis of HCC have been widely accepted, although their benefit still remains unclear. A greater knowledge of the risk factors for development of HCC in patients with liver cirrhosis could improve the efficacy of such programs. AIMS: To identify groups of patients with liver cirrhosis with different risk for developing HCC during a 4 year period. METHODS: Between 1.9.92 and 1.9.99, 463 patients with liver cirrhosis (Child's grade A or B, and 40-65 years old), were included in a surveillance program for HCC based on abdominal uhrasonography and AFP determination every 3 or 6 months. The etiology of the cirrhosis was: Alcoholic (n=273), HCV infection ( n ~ 137), HBV infection ( n = 33) and others (n= 20). Patients were followed up until the appearance of HCC, death, inclusion on a waiting list for liver transplant or lost in the course of follow-up. The predictive value of different risk-prognostic variables was evaluated by means of the Kaplan-Meier Survival and Cox Regression Analysis. RESULTS: Thirty eight patients developed HCC during a follow-up period of 4 years, tn the multivariate analysis, 4 variables showed independent predictive value for the development of HCC: age ~> 55 years, presence of HCV antibodies, prothrombin activity -< 75%, and platelets < 7 5 × 103/ram3. With the coefficients obtained in the Cox regression, a predictive score, ranged from 0 to 4.71, was designed. The score identified 2 groups of patients: 1)Low-risk group, when it was ~<2.3 points (n = 270, 4 HCC, cumulative incidence of HCC at 4 years of 2.3%); 2)High-risk group, when the score was > 2.3 (n = 193, 34 HCC, cumulative incidence of HCC at 4 years of 30.1%)(p=0.0001). CONCLUSIONS: A simple score made up of four clinical and biological variables, allowed us to distinguish two groups of cirrhotic patients with high and low risk for the development of hepatocellular carcinoma. This score can be useful to stablish a subset of cirrhotic patients in whom a surveillance program for early detection of HCC can be unjustified.
Background/Aim: Macrophage migration inhibitory factor (MIF), a T-ceil derived cytokine, is related with the induction of inflammation and immunity as well as cell growth and ceil differentiation. A role of MIF during carcinogenesis has been postulated because MIF is over expressed at the malignant tissues from patients with adenocarcinoma of the lung, uveal melanoma, prostatic tumor and breast carcinoma. However, none of the studies have checked the expression of MIF in precancerous and as well as as in cancerous tissue. Hepatocellular carcinoma (HCC) represents the malignant transformation of hepatocytes and usually occur as a consequence of chronic inflammation of the liver. Most of the patients with HCC suffer from liver cirrhosis (LC) before developinmg HCC and LC represent a definite precancerous state of HCC. Subjects and Methods: MIF in the sera was estimated from a total of 55 patients with hepatocellular carcinoma (HCC), 26 patients with liver cirrhosis (LC), 41 patients with chronic hepatitris (CH), 29 patients with gastrointestinal cancer (esophageal cancer; 14, colon cancer: 15) and 45 normal controls. Peripheral blood mononuclear cells (PBMC) from patients with HCC and normal controls were cultured for 5 days to see their capacity to produce MIF. The levels of MIF was estimated in the sera and culture supernatant by a highly sensitive ELISA method in which the microtiter plates were coated with mouse monoclonal antibody to human MIF and was expressed as ng/ml. Tissue expression of MIF was studied by enzyme-immunohistochemistry using biotin-conjugated anti-human MIF antibody. As a part of antigen retrieval technique, tissues were autoclaved to get proper signaling for MIF. Results: The mean levels of MIF were significantly higher in the sera from patients with HCC (25.6± 15.3 ng/ml), LC (18.9 +_ 10.7 ng/m) and CH (13.8_+8.5 ng/ml) compared to gastrointestinal cancer (6.8+_7.5 ng/ml) aud normal controls (5.6-+ 1.2 ng/ml) (p<0.01). Serial study revealed that the levels of MIF is the sera was increased along with the progression of luver diseases. Immunohistochemical study showed that few infihrating mononuclear cells in the liver tissues from patients with CH expressed MIF. On the other hand, MIF staining was almost undetectable on the hepatocytes from patients with CH. In contrast, most of the hepatocytes from patients with LC expressed very high levels of MIF. Malignant hepatocytes in the cancer nodules and hepatocytes in the noncancerous cirrhotic nodules form patients with HCC also showed very strong expression of MIF. Concliusions: This study has shown that MIF is over expressed on the malignant hepatocytes from patients with HCC. The over expression of MIF on the hepatocytes from patients with LC, a precancerous state of HCC, indicates that MIF over expression may be related with the hepatocarcinogenesis. The levels of of MIF in the sera from patients with CH and LC may be used as a marker of disease progression.