以Aurora 激酶為標靶之肝癌分子治療探索

Abstract

研究背景 肝細胞癌是全世界造成癌症死亡最主要的癌症之一。對於無法進行外科手 術、無法進行局部治療或是已轉移的肝癌患者，通常會選擇化學療法作為治療；然而，肝癌細胞卻幾乎對所有的化療藥物具有高度的抗藥性。最近肝癌的分子標靶治療已有若干進展。Sorafenib因為在兩個重要的第三期安慰劑控制的臨床試驗中證實其確實可以延長晚期肝癌患者的存活時間，因此成為晚期肝癌治療第一個取得官方適應症的突破性藥物。不過，sorafenib 在末期肝癌的抗腫瘤活性嚴格說來仍然非常有限，客觀腫瘤反應率僅僅只有2至3％。因此，發展嶄新的肝癌標靶治療藥物仍是迫切需要的。 在本論文中我們嚐試探討Aurora kinases，一種和細胞有絲分裂相關蛋白質激酶(mitotic kinase)，作為肝癌分子治療標靶的可能性。非整倍體染色體(aneuploidy)與染色體重排(chromosome rearrangements)是許多人類癌症的重要基因特徵。在人類癌症的發生過程中(hepatocarcinogenesis)，基因組失衡(genomic imbalances)是經常被發現的異常。有絲分裂激酶(mitotic kinases)因為是有絲分裂關卡checkpoint、有絲分裂中紡錘體功能(spindle function)、染色體與細胞質分離 (chromosome segregation and cytokinesis)的重要調控者，所以其異常表現可能是許多細胞癌化的重要原因。而Aurora 激酶是有絲分裂激酶家族最重要的一群成員，因為可能有致癌作用，加上Aurora 激酶發現具有成為癌症治療分子標靶的潛力，而成為最近癌症研究領域的一大焦點。 本研究的第一部分是 Aurora激酶小分子抑制劑在肝癌的前臨床研究。先前臺大醫院病理部鄭永銘副教授與許輝吉教授的研究，發現Aurora A激酶在肝癌組織有過度表現的現象，而且Aurora A激酶過度表現的肝癌的惡性度更高、腫瘤分期更晚期，表示Aurora A激酶在肝癌的發生與疾病進展扮演重要的角色。因此，我們認為Aurora激酶在肝癌的治療上可能有重要的臨床意義。首先，過度表現的Aurora A激酶和其他Aurora 激酶家族的成員，可能是肝癌患者重要的預後因子，也是潛在的分子治療的標靶。其次，在我們開始著手這一系列研究之初，我們就已經掌握若干國際性大藥廠已在發展針對Aurora激酶的小分子抑制劑藥物，有了治療學的武器，我們的研究成果就有較高的可能性轉變成臨床可行的療法來治療肝癌。近年來在若干國際性大藥廠的實驗室發展了許多Aurora 激酶的小分子抑制劑，這些小分子抑制劑在臨床前研究顯示出它們有優越的抗腫瘤的效果。雖然Aurora 激酶抑制劑的臨床研究已進入第二期臨床試驗，但是在肝癌治療方面的潛在應用仍缺乏相關研究。因此，我們決定對Aurora 激酶抑制劑在肝癌分子標靶治療的可能角色作進一步的探討。 第二部分是探索Aurora激酶過度表現在肝癌的臨床病理意義。目前在哺乳類動物中有三個較常見的Aurora kinases，即Aurora A、Aurora B、及Aurora C。Aurora A與Aurora B激酶在受質催化區域(catalytic domain)有高度的同源序列(sequence homology)，且在許多人類的癌症上都有過度表現的情形。先前的研究顯示Aurora A與Aurora B激酶的表現量不管是過高或過低都會造成染色體數目不穩定，並進而造成基因組不穩定的現象。即使Aurora A與Aurora B序列相似，兩者在染色體基因座，細胞內的位置，功能與訊號受質皆有差異。 Aurora A kinases之基因座落在染色體20q13.2，而Aurora B kinases基因坐落在染色體17p13.1。 Aurora A kinase 蛋白位於中心體的紡錘體極，且在中心體成熟與紡錘體分裂扮演重要的角色。許多Aurora kinases的抗癌治療研究興趣剛開始都集中在Aurora A的部份，但近來許多研究也證明其他Aurora kinases抑制劑也可抑制腫瘤的活性，如Aurora B可利用基因的方法達到相似的作用。 因此，區別Aurora A與Aurora B在癌症中的角色與臨床上的差異性是目前急需研究的目標。本部分研究的目的是釐清在肝癌中Aurora B與Aurora A的表現在臨床病理的差異性，與其兩者與兩個基因穩定性及肝癌發生有關的分子因子-p53及β-catenin的突變之相關性。我們在本研究的的一部分發現Aurora A和Aurora B 激酶的抑制劑(pan-Aurora kinase inhibitor)，VE - 465，在肝癌的細胞株與動物實驗中展現了具體的抗肝癌效果。然而，即使Aurora A與Aurora B序列相似，兩者在染色體基因座，細胞內的位置，功能與訊號受質皆有差異。本研究除了探討在肝癌組織Aurora B表現的意義和作用外，也將探討以抑制Aurora B 激酶為治療肝癌的策略是否有效可行。而抑制Aurora B 激酶的方法為使用一個對Aurora B 激酶抑制具有專一性的小分子藥物—AZD1152 (AstraZeneca Pharmaceuticals,Inc.；Macclesfield, UK)。本部分研究也設法分析Aurora 激酶的過度表現和肝癌細胞血管入侵(vascular invasion)和轉移之間的相關性，同時也研究以抑制Aurora 激酶為治療策略，在預防肝癌細胞的血管入侵和轉移的可能效果。 第三部分是Aurora激酶小分子抑制劑在肝癌的藥物合併治療。目前已有一些前臨床研究顯示以Aurora 激酶抑製劑為本的組合性藥物處方是有可能造成治療上的突破。 Aurora 激酶抑制劑和化療藥物及其他分子標靶藥物可能會強化彼此的抗腫瘤效應，但其分子機轉大部分仍有待探討。在第一部分的研究中，我們發現VE-465這種pan-Aurora kinase抑制劑具有在肝癌細胞抗腫瘤的效果。我們也發現VE-465會造成p-ERK的upregulation，推測這是一種肝癌細胞抵抗細胞凋亡的機轉。而sorafenib則是第一個被證實對提高晚期肝癌患者存活率有益的藥物。由於sorafenib是一種有效的Raf-MEK-ERK路徑抑制物，在肝癌治療中更是目前臨床上最重要的治療藥物，因此我們嘗試研究合併VE-465和sorafenib的效應，探討這兩種藥物在肝癌細胞中是否具有細胞毒殺協同作用。我們的假設是阻斷若干肝癌細胞抵抗VE-465導致細胞凋亡的機轉路徑，可能可以提高VE-465治療肝癌的療效。 研究方法 在第一部分 Aurora激酶小分子抑制劑在肝癌的前臨床研究中，我們嘗試尋找並研究具有臨床應用可能性的Aurora kinase抑制劑。我們與發展該類小分子抑制劑最快的兩家國際大藥廠(Merck & Co., Inc., Whitehouse Station, NJ, USA AstraZeneca Pharmaceuticals，Macclesfield, UK)聯繫並提出合作研究發展計畫，並順利取得重要且具有臨床發展性的研究藥物—Aurora kinase小分子抑制劑。在此部分研究裡，我們用肝癌細胞的活體內與活體外模型檢測一個嶄新pan-Aurora kinase小分子抑制劑，VE-465，是否有能力抑制Aurora kinase的表現並檢驗其可能的抗肝癌作用已及機轉。 在第二部分探索Aurora激酶過度表現在肝癌的臨床病理意義的研究中，我們選取160個手術切除的肝癌組織檢體，這些病例是於1987年1月至1997年12月所蒐集；病患皆在台大醫院接受完整的病理評估和定期追蹤。我們以反轉錄聚合酶鏈鎖反應 (reverse transcription-polymerase chain reaction，RT-PCR)來測量肝癌細胞以及非腫瘤肝臟組織樣本的Aurora A和Aurora B mRNA的表現。另外，本研究160個病人肝癌檢體中，分別有134個和150個肝癌檢體用來分析p53和β-catenin的基因突變。此分析的方法是藉由直接定序p53 exon 2至exon 11和β-catenin的exon 3。所有的160名病患都接受超過五年以上的追蹤或是追蹤至死亡；最後一次的追蹤日期是2008年11月，當時仍有37名病患存活。其中有149名病患 (93%) 被診斷為有早期腫瘤復發的情況。 2、Fisher's exact test和log-rank test用來作單變數分析；早期腫瘤復發(ETR)、腫瘤大小、腫瘤分期和惡性腫瘤程度等變數則利用複羅吉斯迴歸模型 (multiple logistic regression model) 做多變數分析。存活時間分析則利用多重Cox 比例風險模型 (multiple Cox’s proportional hazards models)進行。血管侵犯與遠端轉移部分的研究採用子題A (Aurora B 激酶過度表現在肝癌的意義)中相同的160個手術切除的肝癌組織檢體分析。有無血管侵犯的現象根據病理組織學檢查的結果判定。是否有肝外轉移則根據臨床追蹤判斷，轉移的發現與最終診斷則依賴影像工具、病理檢查或是升高的AFP。所有的160名病患持續追蹤至少5年以上或是死亡，其中有59名患者有肝外腫瘤轉移的情況。 在第三部分Aurora激酶小分子抑制劑在肝癌的藥物合併治療研究中，我們首先用MTT法評估在不同藥物處理下細胞的存活率，將肝癌細胞種在96-well培養皿中，隔夜培養後加入不同濃度的VE-465和sorafenib，再培養72小時後，加入MTT，再以OD540測量藥物對細胞的毒殺作用。細胞存活率的表示方法是將所測得的吸光值除以背景吸光值，再比較各個不同濃度之間的細胞存活數。每個獨立的細胞存活率測量都會重覆做3至6次。至於VE-465 和sorafenib的潛在的協同生長抑制作用則以Chou-Talalay median effect analysis來推估。藥物處理後的組合指數(combination indices, CI)則以CompuSyn 軟體 (CompuSyn, Inc., Paramus, NJ, USA)計算。組合指數CI值小於1代表VE-465加sorafenib有藥物協同作用(synergistic effect)，組合指數CI值等於1代表加性效應(additive effect)，組合指數CI值大於1則代表拮抗作用(antagonistic effect)。 結果 第一部分 Aurora激酶小分子抑制劑在肝癌的前臨床研究 為了確認VE-465對肝癌細胞存活的影響，我們投予不同濃度的VE-465處理Huh-7和HepG2肝癌細胞株，我們觀察到這兩株細胞株細胞的存活會受到VE-465抑制，而且這些肝癌細胞生長受抑制的程度是和藥物濃度呈現正相關的關係 (dose-dependent)。此外，肝癌細胞生長受抑制的程度也會受到藥物處理時間長短的影響。VE-465會引起Huh-7和HepG2組蛋白H3的去磷酸化，其去磷酸化的程度會受到VE-465濃度的影響。 VE-465可能會干擾肝癌細胞的有絲分裂。我們因此分析暴露在VE-465下的肝癌細胞在有絲分裂時期其紡錘體和染色體在形態上的變化。我們發現以1 μM VE-465處理的Huh-7會有大量不正常的細胞分裂前中期細胞，這些細胞都會有受VE-465影響而具有較不成熟中心體和紡錘體雙極性異常。在1 μM VE-465處理下的HepG2細胞株也觀察到類似的異常有絲分裂。VE-465對肝癌細胞在細胞週期上的影響會使Huh-7細胞在第24小時大量累積含有4N DNA的細胞，隨後的48小時和72小時則出現含有8N DNA的細胞。HepG2在第24小時，VE-465促使細胞產生含有4N 以上DNA的細胞，在第48和72小時則受到proficient postmitotic checkpoint的調控導致有8N DNA的細胞較少。在藥物引起細胞凋亡部分，VE-465引起Huh-7和HepG2細胞的凋亡程度會隨VE-465濃度增加而上升。細胞暴露在10 μM VE-465經過72小時後，造成46%的Huh-7細胞凋亡(處在sub-G1時期)、HepG2則有24% 的細胞凋亡。關於VE-465在活體內試驗對肝癌腫瘤生長抑制的影響，裸鼠接受一天兩次VE-465的腹腔注射，其劑量為15、25和35 mg/kg/天，裸鼠的平均腫瘤的大小和控制組做比較分別減少了59%、59%和77% (p < 0.005)。 第二部分 探索Aurora激酶過度表現在肝癌的臨床病理意義 在基因放大的指數期檢測160個經手術切除的單一性肝癌細胞檢體，其中有98個檢體 (61%) 有Aurora B mRNA過度表現的現象。肝癌患者若有Aurora B的過度表現情況會有較差的五年存活率 (P < 0.0001)及較常發生術後肝癌早期復發 (P < 0.0001)。在多變數分析結果中，ETR (術後肝癌早期復發)事件事影響病患存活時間最重要的危險因子。分析術後肝癌早期復發事件，則發現Aurora B的過度表現、腫瘤大於5cm、較晚期之腫瘤分期和年紀較輕的患者(55歲以下) 皆為造成病人術後肝癌早期復發統計上顯著的獨立危險因子。這部分的分析主要的發現是：Aurora B過度表現會造成較高比例的高度惡性腫瘤以及較多的肝癌早期復發事件。Aurora B的過度表現透過對ETR事件和腫瘤惡性度的影響，再導致接受根除性肝癌切除手術病患的存活狀況變差。 肝癌組織若同時過度表現Aurora A和Aurora B則會造成較多比例的患者有血清AFP升高 (≥200 ng/mL，71%)、較大型的腫瘤 (>5 cm，72%)、較高的腫瘤惡性度(第二至第四級)的腫瘤 (94%)、較晚期(第IIIA至IV期)的腫瘤 (82%)。此外，肝癌組織若同時過度表現Aurora A和Aurora B也會造成較多p53 基因變異 (64%)、較少β-catenin 基因變異(8%) 和四組病人中最差的五年存活率 (19%)。另外，我們發現無論肝癌組織是否有Aurora A過度表現、是否有p53或β-catenin基因變異存在，肝癌中的Aurora B過度表現均會導致較短的病人存活時間、較低的病患五年存活率。 我們進一步探討抑制Aurora B激酶對肝癌細胞生長活性的影響。我們發現AZD1152-HQPA具有濃度依賴性抑制肝癌細胞活性的能力。AZD1152 – HQPA的72小時50％細胞抑制存活率濃度（IC50）在Huh- 7和Hep3B細胞株分別為 16.72 ± 2.44 nM和4.79 ± 1.03 nM。AZD1152-HQPA抑制組蛋白H3在Ser10位置的磷酸化是具有濃度依賴性的。然而AZD1152-HQPA卻完全不會抑制Aurora A激酶在T288位置的磷酸化。AZD1152 – HQPA會誘導肝癌細胞的細胞週期紊亂、誘導肝癌細胞凋亡。 Aurora激酶的表現和肝癌血管侵襲有正相關性，單變數分析發現肝癌血管侵犯除與血清中AFP升高、腫瘤大於5cm、較高的腫瘤組織惡性度有關外，也與Aurora A的過度表現、及Aurora B的過度表現成正相關。Aurora激酶的表現和肝癌遠處轉移也具相關性；單變數分析檢驗發現肝癌遠處的腫瘤轉移與血清中AFP升高、腫瘤大於5cm、較高的腫瘤組織惡性度、較晚的腫瘤分期、以及Aurora A的過度表現、Aurora B的過度表現有關。以Cox比例風險模型 (Cox’s proportional hazards model) 來分析影響病患，發現有肝癌血管侵犯的獨立危險因子，包括腫瘤大於5cm (OR：3.893，P = 0.0005)、Aurora B的過度表現(OR：2.659，P = 0.0183)。而影響病患肝癌根除性手術後遠處轉移的獨立危險因子則有較晚期之腫瘤分期 (stage III-IV，OR：6.945，P < 0.0001)、及Aurora A的過度表現(OR：4.195，P = 0.0027)。 第三部分Aurora激酶小分子抑制劑在肝癌的藥物合併治療 將Huh-7細胞株以0.3 μM VE-465處理(Huh-7 對VE-465的72小時MTT IC50約為1-3μM)，可以發現在24小時以後，ERK的磷酸化表現即加強，此一ERK路徑的活化現象可持續到至少48小時。同樣地，將HepG2細胞株以1 μM VE-465處理(HepG2 對VE-465的72小時MTT IC50約為3-5μM)，可以發現其ERK的磷酸化表現亦在24小時以後加強，此ERK路徑的活化現象可持續到72小時。 關於VE-465 和sorafenib潛在的協同生長抑制作用的推估，我們發現在Huh-7(Huh-7 是對VE-465相對最敏感的細胞株，對VE-465的72小時MTT IC50約為1-3μM) 和PLC/RPF/5(PLC/RPF/5是對VE-465相對最不敏感的細胞株，對VE-465的72小時MTT IC50為＞20μM) ，VE-465加sorafenib的組合指數CI值均小於1。亦即此合併藥物處方VE-465和sorafenib在對VE-465有效與沒有效的肝癌細胞株均呈現出明顯的抗癌細胞生長的協同效應。 結論 第一部分 Aurora激酶小分子抑制劑在肝癌的前臨床研究 台大病理學研究所鄭等先前的研究已指出在所有測試的8種肝癌細胞株和224例人類肝癌細胞組織中有137例 (61%) 會有Aurora A激酶的過度表現。Aurora A激酶的過度表現和更後期的肝癌(第三與第四期)、惡性程度較高的腫瘤與有著相統計上有意義的相關性。這些研究結果均顯示著Aurora 激酶有機會成為治療肝癌的分子標的。 在這項研究中，我們發現，VE-465，一種新型的Aurora kinase抑制劑，可能是一種很有發展潛力的肝療治療藥物。首先，在測試肝癌細胞存活率時我們發現，VE-465對於抑制腫瘤肝癌細胞存在著濃度和時間依賴性。VE-465於肝癌細胞株Huh-7及HepG2的IC50值均遠低於小鼠模型中血漿內可達到的VE-465濃度。此外，我們研究了VE-465對於Aurora訊息傳遞路徑的影響。我們發現，Huh - 7和HepG2肝癌細胞株中，組蛋白histone H3於Ser10的磷酸化在VE-465的處理下明顯下降。在動物實驗中，給予VE-465的也導致Aurora訊息傳遞路徑的抑制、腫瘤生長速度的下降、及細胞凋亡。根據上述的結果，我們推測Aurora 激酶可能有機會成為治療肝癌的標靶分子。 有一些研究顯示，Aurora A和Aurora B激酶是不同的標靶治療分子目標。然而也有研究顯示，Aurora激酶 A和B的共同抑制劑(pan-Aurora kinase inhibitor)與使用基因方式僅阻斷Aurora B激酶的細胞株表現型幾乎一致。在目前研究中，我們發現，VE - 465誘導的細胞週期變化類似於使用RNA干擾造成的Aurora B激酶功能喪失。然而，VE-465也能抑制中心體成熟和紡綞體兩極化，此現象又符合Auroa A激酶的功能不足。是否VE-465所額外造成的Aurora A激酶抑制能夠造成其療效的進步則仍待澄清。 綜合上述討論，我們證明了VE - 465治療在肝癌細胞造成顯著的Aurora 訊息傳遞路徑抑制，從而導致有缺陷的有絲分裂、細胞週期阻滯、細胞DNA複製但有絲分裂失敗的endoreduplication現象、細胞凋亡，並最終抑制癌細胞的生長。我們的研究結果表明，VE - 465和其他可能的小分子Aurora磷酸酶抑制劑可能能用於治療肝癌。進一步的臨床研究來驗證Aurora激酶抑制劑於肝癌患者的治療是必要的。 第二部分 探索Aurora激酶過度表現在肝癌的臨床病理意義 為了進一步說明Aurora B激酶在肝癌細胞癌化過程扮演的角色，並探索其與Aurora A激酶之間的相互作用，我們分析了160個經手術切除的單一性肝細胞癌Aurora A和Aurora B的mRNA，在這160例中其過度表現的現象別分佔了63%和61% 。重要的是，Aurora B激酶基因表達與臨床病理參數的單因素分析中，有多項密切相關，包括高AFP水平（P < 0.0001），腫瘤大小大於五公分（P = 0.021），較高的惡性組織病理分級（P = 0.0007），以及更高的腫瘤分期（P < 0.0001）。多變項分析中，我們發現，Aurora B激酶過度表達與較晚期的肝癌腫瘤分期（第IIIA，IIIB期和IV）有相關，而第III期和第IV期的腫瘤會表現出不同程度的血管侵犯和微觀肝內腫瘤蔓延（勝算比為7.439， P = 0.0003）。 這些結果顯示，過度表現的Aurora B激酶與肝癌的腫瘤侵犯性和肝內轉移性，有高度的統計相關性。 在本部分研究中，我們發現，Aurora B激酶的過度表現在肝癌代表了病人有較高機率發生腫瘤早期復發，這個腫瘤早期復發的機率相較於沒有Aurora B激酶過度表現的病人大於2倍（勝算比OR為8.71; 95％ CI為4.02-18.91; P = 0.0001）。 Aurora B激酶的過度表現與較晚期腫瘤分期（第IIIA，IIIB期和第IV期）的肝癌有統計上的正相關，而肝癌患者具有Aurora B激酶的過度表現，五年存活率比沒有Aurora B激酶過度表現的患者明顯較差（勝算比OR為4.19; 95％ CI為2.11-8.36; P = 0.0001）。藉著多變項分析可證實，Aurora B激酶是一個腫瘤早期復發的獨立危險因子（勝算比OR為4.679， P = 0.0011）。 本部分實驗中，我們也發現AZD1152 – HQPA在肝癌細胞有良好的抗癌效果。 AZD1152 - HQPA治療導致顯著的Aurora B激酶活性被抑制，從而導致細胞週期紊亂，細胞凋亡和抑制肝癌細胞生長。我們的研究結果證明，對Aurora B激酶具選擇性的小分子抑制劑是有潛力的治療肝癌分子標靶藥物。然而，是否選擇性單獨針對Aurora B激酶抑制，或是同時抑制Aurora A 與Aurora B激酶，這兩種治療策略在肝癌孰優孰劣，非常需要我們設計更深入的轉譯與臨床研究探索。 在針對肝癌血管侵犯的多變項分析中，我們發現，肝癌血管侵犯與腫瘤大小大於五公分呈現正相關（勝算比為3.893， P = 0.0005）。另外，肝癌血管侵犯與Aurora B 激酶的過度表現（勝算比為2.659， P = 0.0183）亦呈現正相關，但和較好的組織學惡性度(grade I) 亦呈現負相關（勝算比為0.175， P = 0.0087）。這些結果顯示，Aurora B激酶是一個肝癌血管侵犯的獨立危險因子。在針對肝癌遠處轉移的多變項分析中則發現，肝癌遠處轉移與較晚期的腫瘤分期（勝算比為6.945，P < 0.0001）、Aurora A 激酶的過度表現（勝算比為4.195， P = 0.0027）呈現正相關。這部分分析則顯示，Aurora A激酶是一個肝癌肝癌遠處轉移的獨立危險因子。 第三部分Aurora激酶小分子抑制劑在肝癌的藥物合併治療 在本部分的研究中我們發現，VE-465會造成p-ERK的upregulation，我們推測這是一種肝癌細胞抵抗細胞凋亡的機轉。而sorafenib是一種有效的Raf-MEK-ERK路徑抑制物，在肝癌治療中也是臨床重要的治療藥物，因此我們接著嘗試合併VE-465和sorafenib處理肝癌細胞，結果發現這兩種藥物在Huh - 7和PLC/PRF/5細胞中具有細胞毒殺協同作用。我們的假說是sorafenib阻斷Raf-MEK-ERK路徑的活化，等於阻斷若干肝癌細胞抵抗VE-465導致細胞凋亡的機轉路徑，因此得以提高VE-465治療肝癌的療效。

論文英文簡述 (Abstract in English) Molecular therapeutics is the most promising field of contemporary cancer treatment modalities. Several molecule-targeted drugs have been used clinically in the treatment of lymphoma, and cancer of the lung, breast, colon, head and neck . Nevertheless, the molecular therapy of hepatocellular carcinoma (HCC), one of the most common fatal cancers in Taiwan, has not yet achieved any significant progress. Previous study indicated that molecular targets of HCC may be specific and different from other epithelial malignancies. Thus, searching for more effectivetherapeutic modalities such as modulating cellular signaling is mandatory for the exploration of molecular therapy in HCC. One set of mitotic kinases called Aurora kinases, which are thought to be key regulators of chromosome duplication, segregation, and cytokinesis. Are known to play critical roles in the development of a variety of human cancer, including HCC (clarify).There are three members of human Aurora kinase family, including Aurora A, B, and C. Many studies have shown amplification and overexpression of Aurora kinases in various human cancers in vitro and in vivo . The roles of overexpression of these kinases in oncogenesis have also been demonstrated.

Frequent overexpression of Aurora A has been demonstrated in HCC, which plays important role in the tumor progression of HCC. Recent studies have shown that the inhibition of Aurora kinases can block cell proliferation and induce cell death via apoptosis in multiple types of human tumors. Profound repression of tumor growth was also achieved by inhibition of Aurora kinases in vivo. More importantly, clinically applicable inhibitors of Aurora kinases, such as VX-680, are quickly surfacing on the pipelines of key manufacturers. Thus, exploring the potential roles of these agents in the molecular therapy of HCC is both, is both academically and clinically imperative. Aurora A and Aurora B kinases share a high degree of sequence homology in their catalytic domains, and the overexpression has been identified in many human cancers. Despite their sequence similarity, Aurora A and Aurora B differ in chromosomal gene loci, subcellular localization, cellular functions, and signaling substrates. The enthusiasm of exploring Aurora kinases as anticancer therapeutic targets initially centered on Aurora A. However, recent studies have demonstrated that several Aurora kinase inhibitors exhibit anticancer activity resembling that of Aurora B disruption induced by genetic methods. Therefore, determination of the distinctive roles of Aurora A and Aurora B in carcinogenesis and their individual clinical significance is mandatory.

Part I Preclinical studies of Aurora kinase inhibitor in the treatment of advanced hepatocellular carcinoma In mammals, there are three highly related Aurora kinases: Aurora A, B, and C. Overexpressions of Aurora A and Aurora B have been identified in many human cancers. Because of their roles in mitotic control, genomic instability and tumorigenesis, Aurora kinases have attracted much attention as potential cancer therapeutic targets. The inhibition of Aurora A by antisense oligonucleotides was first shown to effectively repress the tumor growth of lymphoma and pancreatic cancer cells. Later on, several Aurora kinase inhibitors were discovered, including VX-680 (MK-0457), ZM447439 , Hesperadin, PHA-680632, AZD1152, and MLN8054. VX-680 was the first Aurora kinase inhibitor showing broad antitumor activity in vitro and in vivo, and the foremost Aurora kinase inhibitor to be studied in clinical trials. Although clinical studies of Aurora kinase inhibitors have already reached phase II trials, their potential application in the treatment of HCC remains to be explored. We previously demonstrated frequent overexpression of Aurora A kinase, and its correlation with higher tumor grade and more advanced tumor stage in HCC. The unmet medical need for effective HCC treatment prompted us to investigate the therapeutic potential of Aurora kinase inhibitors. In this study, we tested VE-465, an analog of VX-680 with comparable potency in Aurora kinase inhibition, in both in vitro and in vivo HCC models. The results showed that VE-465 effectively inhibited Aurora kinase activity and induced mitotic disturbance, endoreduplication, apoptosis, and growth suppression of HCC. Our observations suggest that VE-465 is a promising therapeutic agent in HCC and deserves further investigation. The overexpression of Aurora A was associated with high-grade and high-stage tumors, and p53 mutation. These aggressive tumor phenotypes are characteristics of HCC with chromosome instability and imply that overexpression of Aurora kinases contributes to progression in human HCC. Furthermore, we found that overexpression of Aurora A contributed to worse patient survival. Consistently, Sistayanarain et al. reported that Aurora B transcripts were detectable in 12 (71%) out of 17 HCC cases. Okada et al.reported that a novel anticancer substance, MK-615, inhibited growth of HCC cells by suppressing Aurora A kinase activity. These findings indicate that Aurora kinases are potential therapeutic targets in HCC. In this study, we have demonstrated that VE-465, a novel Aurora kinase inhibitor, is a promising therapeutic agent in HCC. Firstly, we found that VE-465 suppressed tumor cell viability in the tested liver cancer cell lines in a concentration- and time-dependent manner. The IC50 values of VE-465 for Huh-7 and HepG2 were far below the plasma concentrations of VE-465 achievable in mouse models. Secondly, we showed that VE-465repressed the Aurora signaling evidenced by the steady downregulation of histone H3 (Ser10) phosphorylation in Huh-7 and HepG2 cells. Thirdly , VE-465 administration led to the inhibition of Aurora signaling, tumor growth suppression, and apoptosis in animal studies Taken together,these results suggest that Aurora kinases represent potential novel therapeutic targets in HCC. To better understand the anticancer effects of VE-465, we examined the morphologic changes in mitosis, cell cycle progression, and cell death in VE-465 treated liver cancer cells. VE-465 treatment led to the formation of monopolar spindles and the failure of centrosome maturation in Huh-7 and HepG2, which are remiscent of the functional deficiency of Aurora A. We also observed the detachment of chromosomes from the spindle structures in VE-465 treated HepG2 cells, implying dysregulation of Aurora B function. These mitotic disturbances induced by the inhibition of Aurora kinases may underlie the cytotoxic effects of VE-465 in HCC cells. Following aberrant mitosis induced by Aurora kinases inhibition, cancer cells may be arrested in a pseudo G1 state controlled by a p53-dependent postmitotic checkpoint. Consistent with this suggestion, we demonstrated that endoreduplication was more readily elicited in the p53-mutated Huh-7 cells (codon 220 mutation) compared with the wild-type p53 HepG2 cells. The p53 function is also regarded as a major determinant of apoptosis induced by Aurora kinase inhibitors, but the regulatory mechanisms remain elusive. We showed that sub-G1 peaks induced by VE-465 in Huh-7 were greater than those in HepG2 cells, which may be at least in part attributed to the loss of p53 checkpoints in Huh-7 cells. In summary, we demonstrated that VE-465 treatment resulted in profound inhibition of Aurora signaling in liver cancer cells, which led to defective mitosis, cell cycle arrest, endoreduplication, apoptosis, and eventually suppression of cancer cell growth. Our results suggest that VE-465 and probably other small-molecule inhibitors of Aurora kinases are promising in the treatment of HCC. Further clinical studies are needed to validate the therapeutic potential of Aurora kinase inhibitors in HCC patients.

Part II Exploring the clinicopathologic significances of Aurora kinases overexpression in hepatocellular carcinoma We previously showed that Aurora A was overexpressed in 137 (61%) of 224 human HCCs and that the overexpression of Aurora A was associated with aggressive tumor characteristics and poor prognosis of patients. Furthermore, we demonstrated that VE-465, a novel pan-Aurora kinase inhibitor, had anticancer effects in preclinical experimental models of human HCC. These findings indicated that Aurora kinases may be important biomarkers and potential therapeutic targets in HCC. There are three highly related Aurora kinases in mammals, Aurora A, B, and C. Aurora A and Aurora B share a high degree of sequence homology in their catalytic domains, and their overexpressions have been identified in many human cancers. Despite their sequence similarity, Aurora A and Aurora B differ in chromosomal gene loci, subcellular localization, cellular functions, and signaling substrates. The Aurora A kinase gene is localized to chromosome 20q13.2, and Aurora B kinase to chromosome 17p13.1. (Repeatitive)Aurora A kinase protein is localized in the centrosome and spindle poles and plays important roles in centrosome maturation and spindle assembly. Aurora B kinase, which is a chromosome passenger protein localized in the centromeres during early mitosis and then at the spindle midzone after anaphase, is essential for chromosome biorientation, function of the spindle assembly checkpoint, and cytokinesis. The enthusiasm of exploring Aurora kinases as anticancer therapeutic targets initially centered on Aurora A, but recent studies have demonstrated that several Aurora kinase inhibitors exhibit anticancer activity resembling that of Aurora B disruption induced by genetic methods. Therefore, determination of the distinctive roles in carcinogenesis and individual clinical significance of Aurora A and Aurora B is mandatory. The aims of this study were to elucidate the clinicopathologic significance of Aurora B expression and Aurora A expression in HCC and to correlate their expression with p53 and β-catenin mutations, the two most frequently mutated genes in HCC. Despite the sequence homology and common association with mitotic regulatory events, Aurora A and Aurora B differ in the subcellular localization and signaling substrates , and hence possess essentially distinctthe functions. We have reported that Aurora A is highly expressed in HCC and that the overexpression is closely associated with aggressive tumor phenotypes and worse patient prognosis. But the clinicopathologic significance of Aurora B in HCC progression remains to be clarified. In this study, we demonstrated that overexpressions of Aurora A and Aurora B were detected in 63% and 61% of 160 surgically resected, primary unifocal HCCs, respectively. By multivariate analyses, we showed that Aurora B overexpression was associated with high-stage (stages IIIA, IIIB, and IV) HCCs, which exhibit vascular invasion and various extents of intrahepatic spread (OR, 7.439; P = 0.0003). These findings suggest that overexpression of Aurora B is associated with tumor invasion and intrahepatic metastasis of HCC, as having been shown in Aurora A.Our findings suggest that Aurora B overexpression serves as a useful marker predicting ETR and hence poor prognosis.

In the present study, we showed AZD1152-HQPA, an Aurora B selective inhibitor, exerted anticancer effects in HCC cells. AZD1152-HQPA treatment resulted in profound inhibition of Aurora B signaling, which in turn led to cell cycle disturbance, apoptosis, and growth suppression in HCC cells. Our results suggest that Aurora B selective inhibitors are potential drugs for HCC treatment, confirming the observation that AZD1152 is a novel promising therapeutic approach for HCC. We previously demonstrated that Aurora B overexpression correlated well with higher histology grade and more advanced stage of hepatocellular carcinoma (HCC). In this part of study, we sought to analyze the association between Aurora B overexpression and vascular invasion/metastasis of HCC, as well as the potential of AZD1152, a novel and selective Aurora B kinase inhibitor in preventing vascular invasion and metastasis of HCC. Univariate analysis showed that vascular invasion and metastasis were both associated with younger age (