管腔型乳腺癌是最常見(jiàn)的乳腺癌亞型，大約占全部病例的三分之二，晚期一線標(biāo)準(zhǔn)治療方案為內(nèi)分泌治療聯(lián)合CDK4/6抑制劑，中位無(wú)進(jìn)展生存大約2～2.5年。不過(guò)，目前尚無(wú)針對(duì)CDK4/6抑制劑耐藥的明確標(biāo)準(zhǔn)治療方案。后續(xù)治療方案包括內(nèi)分泌治療±針對(duì)特定基因突變（包括PI3K-AKT-mTOR通路靶點(diǎn)、BRCA突變或ESR1突變）靶向治療或再用CDK4/6抑制劑，中位無(wú)進(jìn)展生存都不超過(guò)6個(gè)月。隨著疾病進(jìn)展，對(duì)內(nèi)分泌治療敏感性降低，治療策略只能逐漸轉(zhuǎn)向以化療為主的治療方式。抗體綴合藥物徹底改變管腔型乳腺癌的治療格局，DESTINY-Breast04研究證實(shí)德曲妥珠單抗將HR陽(yáng)性HER2低表達(dá)乳腺癌化療失敗患者中位無(wú)進(jìn)展生存延長(zhǎng)至10.1個(gè)月，DESTINY-Breast06研究證實(shí)德曲妥珠單抗可改善HER2低表達(dá)和超低表達(dá)乳腺癌經(jīng)過(guò)充分內(nèi)分泌治療但是尚未化療患者中位無(wú)進(jìn)展生存。雖然德曲妥珠單抗和戈沙妥珠單抗等抗體綴合藥物有效，但是仍然離不開(kāi)細(xì)胞毒性化療藥物。根據(jù)管腔型乳腺癌分子特征進(jìn)行精準(zhǔn)治療，尤其找出生物學(xué)預(yù)測(cè)標(biāo)志物仍然至關(guān)重要，值得進(jìn)一步研究。

諸多研究努力闡明管腔型乳腺癌的分子圖譜。50基因表達(dá)亞型及其風(fēng)險(xiǎn)模型是乳腺癌分子分型的里程碑，但是根據(jù)mRNA分型不能充分反映組織異質(zhì)性和細(xì)胞多樣性。國(guó)際乳腺癌分子分型聯(lián)盟METABRIC數(shù)據(jù)庫(kù)僅包括1977至2005年診斷患者，可能無(wú)法充分反映當(dāng)代臨床治療標(biāo)準(zhǔn)。同樣，由于缺乏多組學(xué)數(shù)據(jù)，美國(guó)紐約紀(jì)念醫(yī)院斯隆凱特林癌癥中心MSKCC數(shù)據(jù)庫(kù)對(duì)全面探索基因組事件的價(jià)值有限。雖然臨床蛋白質(zhì)組腫瘤分析聯(lián)盟CPTAC已彌補(bǔ)多組學(xué)數(shù)據(jù)不足，但是樣本量有限，尤其管腔型乳腺癌復(fù)發(fā)患者，限制其全面探索腫瘤異質(zhì)性的能力。這些分子模型無(wú)法準(zhǔn)確地分析腫瘤或發(fā)現(xiàn)個(gè)體生物學(xué)特征及其治療靶點(diǎn)。

為了應(yīng)對(duì)這些挑戰(zhàn)，復(fù)旦大學(xué)附屬腫瘤醫(yī)院此前對(duì)管腔型乳腺癌患者樣本進(jìn)行大規(guī)模并行測(cè)序，并結(jié)合代謝組學(xué)和蛋白質(zhì)組學(xué)分析，利用大規(guī)模多組學(xué)數(shù)據(jù)相似性網(wǎng)絡(luò)融合（SNF）確定管腔型乳腺癌SNF復(fù)旦分型：經(jīng)典管腔型SNF1、免疫原型SNF2、增殖型SNF3、RTK驅(qū)動(dòng)型SNF4。這些SNF分型通過(guò)癌癥基因組圖譜TCGA、METABRIC和CPTAC等獨(dú)立數(shù)據(jù)集得到進(jìn)一步驗(yàn)證，表明SNF分型具有可靠性和普適性。不過(guò)，多組學(xué)測(cè)序的成本和復(fù)雜性可能限制其臨床應(yīng)用。為了克服該難題，復(fù)旦大學(xué)附屬腫瘤醫(yī)院開(kāi)發(fā)數(shù)字病理分類(lèi)方法，利用病理全切片圖像人工智能卷積神經(jīng)網(wǎng)絡(luò)對(duì)分子分型進(jìn)行分類(lèi)，并捕獲和整合重要的形態(tài)學(xué)特征，例如腫瘤浸潤(rùn)淋巴細(xì)胞、癌相關(guān)成纖維細(xì)胞等，交叉驗(yàn)證曲線下面積達(dá)0.87（SNF1）、0.81（SNF2）和0.78（SNF3和SNF4），表明根據(jù)數(shù)字病理人工智能深度學(xué)習(xí)進(jìn)行分型區(qū)分具有可行性。這些數(shù)據(jù)回顧分析工作為管腔型乳腺癌分型和個(gè)體化治療奠定基礎(chǔ)，但是根據(jù)SNF分型系統(tǒng)進(jìn)行人工智能輔助精準(zhǔn)治療對(duì)管腔型乳腺癌的療效仍需進(jìn)一步臨床前瞻驗(yàn)證。

2025年12月4日，美國(guó)《細(xì)胞》旗下《癌細(xì)胞》在線發(fā)表復(fù)旦大學(xué)附屬腫瘤醫(yī)院范蕾①??、張文娟①、龔悅①、金希、趙珅、于寶華、吉芃、劉西禹、陳力、賀敏、劉引、隋辛怡、馬林曉曦、朱秀之、楊帆、葛麗萍、吳松陽(yáng)、吳炅、余科達(dá)、柳光宇、胡欣、楊文濤、王中華??、江一舟??、邵志敏??、北京腫瘤醫(yī)院李惠平①和王楠、重慶大學(xué)附屬腫瘤醫(yī)院曾曉華①和張寧寧、中國(guó)醫(yī)科大學(xué)附屬第一醫(yī)院滕月娥①和石晶、遼寧省腫瘤醫(yī)院孫濤、南昌市人民醫(yī)院陳文艷和汪云、中山大學(xué)腫瘤防治中心王樹(shù)森、西安交通大學(xué)第一附屬醫(yī)院楊謹(jǐn)、上海市第一婦嬰保健院莊志剛、南通大學(xué)附屬醫(yī)院倪蘇婕和何志賢、蘇北人民醫(yī)院符德元、福建省腫瘤醫(yī)院宋傳貴、吉林大學(xué)第一醫(yī)院呂錚、恒瑞醫(yī)藥梁倩男和沈煜、復(fù)旦大學(xué)附屬中山醫(yī)院梁斐等學(xué)者的LINUX（BCTOP-L-M05）研究報(bào)告，首次對(duì)管腔型晚期乳腺癌CDK4/6抑制劑耐藥患者人工智能輔助SNF復(fù)旦分型精準(zhǔn)治療進(jìn)行前瞻驗(yàn)證。

• LINUX (BCTOP-L-M05): SNF Platform Study of HR+/HER2-advanced Breast Cancer (NCT05594095)

• Official Title: Precision Platform Study of HR+/HER2-advanced Breast Cancer Based on SNF Typing (A Prospective, Open-label, Multi-center, Phase II Platform Study)

該多中心探索性貝葉斯優(yōu)化隨機(jī)對(duì)照二期平臺(tái)研究于2022年12月30日至2024年5月23日從全國(guó)6個(gè)中心入組管腔型晚期乳腺癌CDK4/6抑制劑耐藥患者105例，通過(guò)人工智能輔助數(shù)字病理分類(lèi)進(jìn)行SNF復(fù)旦分型，再按2比1隨機(jī)分入精準(zhǔn)治療組或單藥化療組。

精準(zhǔn)治療組70例：

• SNF1：依維莫司＋內(nèi)分泌治療

• SNF2：卡瑞利珠單抗＋法米替尼＋單藥化療

• SNF3：氟唑帕利＋單藥化療

• SNF4：阿帕替尼＋單藥化療

單藥化療組35例：

• 卡培他濱、白蛋白紫杉醇、長(zhǎng)春瑞濱、艾立布林四選一

結(jié)果，精準(zhǔn)治療組與單藥化療組相比，客觀緩解率顯著較高：

• SNF1：10%比0%

• SNF2：65%比30%

• SNF3：40%比30%

• SNF4：70%比20%

兩組患者3～4級(jí)治療相關(guān)不良事件發(fā)生率相似，均為37%。

因此，該小樣本二期臨床研究結(jié)果表明，對(duì)于管腔型晚期乳腺癌CDK4/6抑制劑耐藥患者，人工智能輔助SNF復(fù)旦分型精準(zhǔn)治療與單藥化療相比，臨床獲益顯著，尤其對(duì)于SNF2和SNF4亞型而言，值得開(kāi)展大樣本三期臨床研究進(jìn)一步驗(yàn)證。

Cancer Cell. 2025 Dec 4. IF: 44.5

Precision treatment with artificial intelligence assisted subtyping enhances therapeutic efficacy in HR+/HER2- breast cancer: The LINUXtrial.

Lei Fan, Wen-Juan Zhang, Hui-Ping Li, Xiao-Hua Zeng, Yue-E Teng, Yue Gong, Xi Jin, Shen Zhao, Tao Sun, Wen-Yan Chen, Shu-Sen Wang, Jin Yang, Zhi-Gang Zhuang, Su-Jie Ni, Zhi-Xian He, De-Yuan Fu, Chuan-Gui Song, Zheng Lv, Qian-Nan Liang, Bao-Hua Yu, Jing Shi, Nan Wang, Xin-Rui Liang, Ning-Ning Zhang, Yun Wang, Peng Ji, Xi-Yu Liu, Li Chen, Min He, Yin Liu, Xin-Yi Sui, Lin-Xiaoxi Ma, Xiu-Zhi Zhu, Fan Yang, Li-Ping Ge, Song-Yang Wu, Jiong Wu, Ke-Da Yu, Guang-Yu Liu, Xin Hu, Yu Shen, Zheng Pang, Jian-Fei Wang, Fei Liang, Wen-Tao Yang, Zhong-Hua Wang, Yi-Zhou Jiang, Zhi-Ming Shao; BCTOP investigators.

Fudan University Shanghai Cancer Center, Shanghai, China; Shanghai Medical College, Fudan University, Shanghai, China; Peking University Cancer Hospital and Institute, Beijing, China; Chongqing University Cancer Hospital, Chongqing, China; The First Hospital of China Medical University, Shenyang, China; Cancer Hospital of Dalian University of Technology, Liaoning Cancer Hospital & Institute, Shenyang, China; Nanchang People's Hospital, Nanchang, China; Sun Yat-sen University Cancer Center, Guangzhou, China; The First Affiliated Hospital of Xi'an Jiaotong University, Xi'an, China; Shanghai First Maternity and Infant Hospital, School of Medicine, Tongji University, Shanghai, China; The Affiliated Hospital of Nantong University, Nantong, China; Northern Jiangsu People's Hospital Affiliated to Yangzhou University, Yangzhou, China; Fujian Cancer Hospital (Fujian Branch of Fudan University Shanghai Cancer Center), Fuzhou, China; The First Hospital of Jilin University, Changchun, China; Jiangsu Hengrui Pharmaceuticals Co., Ltd., Shanghai, China; Zhongshan Hospital, Fudan University, Shanghai, China.

HIGHLIGHTS

• LINUX platform trial explores AI-assisted subtyping-based precision strategies

• Subtyping-based precision therapies demonstrate efficacy in HR+/HER2- breast cancer

• Subtyping-based precision treatments achieve primary endpoint in SNF2/4 subtypes

• Precision treatments show a manageable safety profile

We report the results of LINUX (NCT05594095), a multicenter, randomized, controlled phase II platform trial aiming to identify effective precision treatments for hormone receptor-positive/human epidermal growth factor receptor 2-negative metastatic breast cancer after resistance to cyclin-dependent kinase 4/6 inhibitor. A total of 105 patients were categorized into four similarity network fusion (SNF) subtypes by artificial intelligence-assisted classification and randomly assigned to receive subtyping-based precision therapy (N = 70) or treatment of physician's choice (N = 35). Results demonstrate superior primary endpoint of objective response rates in the subtyping-based groups compared to controls: 10% versus 0% for SNF1, 65% versus 30% for SNF2, 40% versus 30% for SNF3, and 70% versus 20% for SNF4. Grade 3-4 treatment-related adverse events occurred in 37% of both groups. These findings highlight the clinical benefits of subtyping-based precision therapies, particularly for SNF2 and SNF4 subtypes, warranting further validation in phase III trials.

KEYWORDS: HR+/HER2-, breast cancer, precision treatment, AI-assisted, subtyping-based therapy

DOI: 10.1016/j.ccell.2025.11.003

來(lái)源：SIBCS

免責(zé)聲明：本公眾號(hào)旨在為醫(yī)療衛(wèi)生專(zhuān)業(yè)人士提供腫瘤領(lǐng)域資訊參考，促進(jìn)專(zhuān)業(yè)交流。內(nèi)容僅供信息交流，不可替代專(zhuān)業(yè)醫(yī)療指導(dǎo)，亦非診療建議。 據(jù)此操作，風(fēng)險(xiǎn)自擔(dān)。轉(zhuǎn)載內(nèi)容均已標(biāo)注來(lái)源及作者。如有版權(quán)問(wèn)題，請(qǐng)留言聯(lián)系，我們將及時(shí)處理。