透過基因擴增導致的 MYCN 過度活化是高風險神經母細胞瘤的核心特徵，儘管接受了密集治療，該疾病的臨床預後依然不佳。在MYCN擴增的腫瘤中，極光激酶 A（Aurora kinase A， AURKA）可透過蛋白質交互作用穩定 MYCN，進而促進惡性腫瘤進展。在本研究中，資料庫分析顯示，同時高表達MYCN與AURKA的神經母細胞瘤患者，其存活預後顯著差於僅單獨高表達任一基因的腫瘤患者。鑑於兒童惡性腫瘤通常是由胚胎谱系特性而非累積突變所驅動，本研究將兒童癌症世代分析擴展至尋找其他發育相關的兒童腫瘤時，鑑定出了一組伴隨MYCN與AURKA同時高表達的軟組織肉瘤；其中，在尤文氏肉瘤（Ewing’s sarcoma）的共同高表達與整體存活率呈現強烈的負相關。在神經母細胞瘤與尤文氏肉瘤細胞株中，MYCN的表達水平與對AURKA 異位調節抑制劑（包括alisertib和6K465）的敏感性呈正相關。為評估其臨床轉譯潛力，本研究在 SK-N-BE(2) CDX（細胞株來源的異種移植）模型中，比較了alisertib與 DBPR728（6K465的前體藥物）的抗腫瘤療效。結果顯示，每週給藥一次300 mg/kg 的 DBPR728，比每日給藥一次50 mg/kg 的alisertib 能產生更持久的腫瘤抑制效果。為了尋找此類MYCN/AURKA雙重抑制劑敏感性相關的預測性基因特徵，本研究利用有限的細胞毒性與轉錄組圖譜，透過彈性網路（Elastic Net）k-fold重複交叉驗證開發了預測模型。將這些模型應用於癌症細胞株百科全書（CCLE）與 St. Jude 資料庫時發現，藥物敏感性與參與神經相關途徑的基因具有相關性。這些結果為兒童癌症患者的分層提供了潛在依據。總體而言，這些發現將MYCN與AURKA的共同活化定義為一種治療脆弱性，為進一步研發針對MYCN與AURKA高表達之兒童癌症的靶向藥物提供了有力支持。;Excessive activation of MYCN via gene amplification is a defining feature of high-risk neuroblastoma， which continues to have poor clinical outcomes despite intensive therapy. In MYCN-amplified tumors， Aurora kinase A (AURKA) contributes to promote malignancies by stabilizing MYCN via protein-protein interactions. In this study， database analyses revealed that neuroblastoma co-overexpressing MYCN and AURKA exhibit significantly worst survival outcomes compared with tumors overexpressing either one gene. Given that pediatric malignancies are frequently driven by embryonic lineage identities rather than accumulated mutations， expanding pediatric cancer cohort analysis to identify other developmentally related pediatric tumor identified a subset of soft-tissue sarcoma with concurrent MYCN and AURKA co-overexpression， among which Ewing’s sarcoma showed a strong negative correlation between co-overexpression and overall survival. Across neuroblastoma and Ewing’s sarcoma cell lines， MYCN expression levels positively correlated with sensitivity to AURKA allosteric inhibitors， including alisertib and 6K465. To evaluate translational potential， the antitumor efficacy of alisertib and DBPR728 (the prodrug of 6K465) in the SK-N-BE(2) CDX model were compared. DBPR728 administered at 300 mg/kg once weekly produced more durable tumor suppression than alisertib at 50 mg/kg once daily. To identify predictive gene signatures associated with sensitivity to AURKA allosteric inhibitors， predictive models using Elastic Net k-fold repetitive cross-validation were developed on a limited set of cytotoxicity and transcriptomic profiles. Applying these models to Cancer Cell Line Encyclopedia (CCLE) and St. Jude datasets revealed that sensitivity correlates with genes involved in the neuronal-related pathways. These results suggest a potential basis for pediatric cancer patient stratification. Collectively， these findings define coactivation of MYCN and AURKA as a therapeutic vulnerability， warranting further development of AURKA-targeting agents for pediatric cancers.
