Cocktail gene therapy for hepatocellular carcinoma and liver cirrhosis

論述重點： 免疫療法及抗血管新生基因治療在許多肝癌研究中都已經被證實對於小型肝癌有效，然而這些研究中所使用的動物模型常常不能真正反應人類肝癌的情境。這些動物肝癌常是植入肝癌細胞而長成之腫瘤，動物本身肝臟很健康。而人類肝癌一般是多發，體積較大，而且常常伴隨有病毒性肝炎或是肝硬化。所以為了開發肝癌療法之用，一個同時帶有肝癌以及病毒性肝炎或是肝硬化的動物模型便顯得非常重要。本研究首先利用土撥鼠來驗證合併免疫療法及抗血管新生基因治療的雞尾酒療法是否有更好的療效，慢性感染土撥鼠肝炎病毒之土撥鼠會自然產生肝癌，而這樣的疾病表現非常接近人類肝癌。在土撥鼠模式上驗證雞尾酒療法之後，我們進一步試圖發展出一個同時具有肝硬化的肝癌模型。我們利用不同的誘發時間長短，而造成肝癌或是同時合併肝癌及肝硬化等兩個不同動物模型，接著在同時帶有肝癌及肝硬化的大鼠身上，我們使用表現干擾素α基因的重組腺病毒來進行同不對肝癌及肝硬化的基因治療。 方法 本研究使用能表現抗血管新生因子（pigment epithelium-derived factor and endostatin）及細胞激素(granulocyte macrophage colony-stimulating factor and interleukin)的重組腺病毒來治療土撥鼠肝癌。治療後以腹部超音波或電腦斷層掃瞄來進一步追蹤腫瘤大小。為了進一步發展出帶有肝硬化的肝癌動物模型，我們改良了傳統上使用diethylnitrosamine (DEN)的誘發方法，根據動物體重變化，等比例調整藥物在飲水中的濃度量，自六週齡之後連續餵食。這結果使得大白鼠在六周誘發後，便產生出多發肝癌，此時如果停止誘發，再六週後將回復正常肝臟機能。若動物餵藥超過九周之後，此時除發展出肝癌外，同時會出現肝硬化的表現。這兩種模式造成的肝癌數量及大小相當，不過肝臟之硬化程度差別很大。而且在肝硬化組中的動物，其肝臟纖維化為不可逆變化，非常接近人類疾病模式。而我們接下來則使用這樣的動物模型來進行以腺病毒攜帶干擾素α的基因治療，試圖來同步治療肝癌及肝硬化。 結果 我們的實驗結果證實，抗血管新生基因治療或是免疫治療均能造成一定程度的抗肝癌療效。而更進一步合併兩者的雞尾酒療法，在不增加總腺病毒用量的條件下，比起任何單一成分基因療法，更能得到加成的抗癌效益。特別是在大型腫瘤上，抗癌效益的增加更是明顯。我們在接受免疫治療的土撥鼠腫瘤內可看見大量CD3+ T淋巴球浸潤，而在抗血管新生基因治療組則觀察到腫瘤內血管明顯減少，而接受雞尾酒療法的土撥鼠腫瘤怎可看見兩個情形兼而有之以外，更有比單一療法更高的自然殺手細胞活性表現，以即便異議制分子的表現減少。至於肝硬化治療部分，我們的研究結果證實，由於腺病毒容易感染肝細胞的特質，干擾素α基因治療使得干擾素α大量局部表現於肝臟，而有統計意義地減少了肝腫瘤體積，同時減輕了大鼠肝硬化程度。在作用機制的探討方面，我們發現干擾素α基因治療能造成免疫調節抗細胞增殖，以及誘發腫瘤凋亡活動，而達到抑制腫瘤生長的目的。同時干擾素α治療也能經由抑制transforming growth factor-β (TGF-β) 以及 tissue inhibitor of metalloproteinase-1 (TIMP-1)表現，而造成肝纖維化的減輕。 討論 免疫療法與抗血管新生治療的合併使用，展現出比任何一個單一療法更強力的抗癌效益。這證實了合併這兩個不同機制的基因治療能達到加成的作用。而經過在土撥鼠這個如此貼近人類的動物模型上驗證之後，我們可以期待將來可能在臨床上成為一個輔助抗癌療法，經由大幅縮小腫瘤，來增加手術切除率，改善病人治療預後，或是延長生命。除了土撥鼠模型之外，我們發展出來的合併肝癌肝硬化之大鼠模型，也是一個研究肝癌治療策略時，很理想的動物模式。而我們以干擾素α基因治療來作為驗證，同時也得到以腺病毒載體在肝內表現干擾素α一法能同步治療肝癌及肝硬化的滿意結果。Background & Aims: Cytokine and antiangiogenic gene therapies have proven effective in many implanted hepatocellular carcinoma(HCC) models with small tumor burdens. However, these models may not realistically reflect human HCCs, which are typically large, multifocal and often noted on a background of hepatitis or liver cirrhosis. So, to develop anti-HCC therapeutic strategies, an animal model bearing multifocal liver tumors accompanied with viral hepatitis or cirrhosis is a preferred experimental setting. This study utilized woodchucks to investigate the efficacy of a cocktail-type gene therapy strategy. The animals bore spontaneously-occurring HCCs induced by woodchuck hepatitis virus infection, and therefore closely resembled human HCCs. In the second part, we tried to develop a tumor-bearing rat model with liver cirrhosis. By adjusting the duration of administration of diethylnitrosamine (DEN), the animals could be induced to develop HCC alone, or HCC and liver cirrhosis simultaneously. Methods: Adenoviruses encoding antiangiogenic factors [pigment epithelium-derived factor and endostatin] or cytokines [granulocyte macrophage colony-stimulating factor and interleukin ] were delivered, either separately or in combination, into woodchuck livers bearing HCCs via the intrahepatic artery. Tumor sizes were monitored by sequential abdominal ultrasound examination and CT. To develop HCC-bearing cirrhotic animal model, we modified the feeding protocol by adjusting the DEN intake amount weekly based on BW of the animals. The results show that animals fed with DEN for 6 consecutive weeks develop multifocal HCC with normal liver parenchyma; whereas animals fed with DEN for 9 consecutive weeks develop multifocal HCC complicated with moderate cirrhosis. The tumor burdens between these two groups are comparable, determined by the ratio of liver weight/BW, the estimated volume of all visible tumor nodules, and the density of glutathione S-transferase placental form-positive foci. On the other hand, liver cirrhosis occurs slightly and transiently in the former group, but severely and irreversibly in the latter group, shown by immunohistochemical staining of α-smooth muscle actin, the bile flow rate, and the liver hydroxyproline content, then adenoviruses encoding interferon-α(IFN-α) were delivered into rats in the latter group for treatment of liver cirrhosis and HCC simultaneously. Results: Our results demonstrate that monotherapy with antiangiogenic genes or cytokine genes had significant therapeutic effects. Furthermore, cocktail-type therapies containing both types of genes displayed synergistic antitumor effects under the same viral doses, particularly for large tumor loads. Effector cell analyses based on flow cytometry demonstrate that the cytokine-based gene therapies resulted in increased CD3+ T-cell infiltration of the HCCs. Examination of microvessel density also showed that tumor vasculature was significantly reduced by the antiangiogenesis-based gene therapies. No evidence of adenovirus-induced hepatic toxicity was found in these animals. As for the research about treatment of liver cirrhosis, Our results demonstrated that targeting of IFN-α expression to the liver significantly reduced liver tumor volume and ameliorated liver cirrhosis. Mechanistic studies revealed that IFN-α gene therapy induced immunomodulatory, antiproliferative, and proapoptotic activities that were effective in the control of tumor growth, and reduced the expressions of transforming growth factor-β (TGF-β) and tissue inhibitor of metalloproteinase-1 (TIMP-1), leading to amelioration of liver cirrhosis. Conclusions: The combination of immunotherapy and antiangiogenic gene therapy exhibited the most significant antitumor effect compared to either therapy alone. This indicates a strong synergism induced by the combination of two different therapeutic strategies. Using the clinically relevant model, future clinical application as a neoadjuvant therapy would be encouraged. Except the woodchuck model, the HCC-bearing rat model with cirrhosis is an ideal model to study therapeutic strategies that can treat both diseases simultaneously, and our results suggest that IFN-α gene therapy is a promising strategy to treat HCC patients who have concomitant liver cirrhosis