現今的免疫治療對於MSI-H (microsatellite instability-high,微衛星不穩定)之轉移性大腸直腸癌有突破性的療效,但由於MSI-H之轉移性大腸直腸癌僅占1.8%至4%,因此化學治療合併標靶治療仍舊是佔96%以上的MSS (microsatellite stable,微衛星穩定)轉移性大腸直腸癌之標準治療。目前至少已經有兩個第三期臨床試驗,顯示免疫治療合併化學治療與標靶治療並無法顯著增加存活期。因此,嶄新的合併治療處方以及探尋免疫治療的新型生物標記,是現今MSS轉移性大腸直腸癌一項亟待突破的重要課題。現今已知其他種類癌症常用之免疫治療生物標記:PD-L1 (programed cell death ligand-1,細胞程式性凋亡配體1),對於轉移性大腸直腸癌並不能做為免疫治療生物標記,而MHC class I (major histocompatibility complex class I,第一型主要組織相容型複合體)之表現則是抗原表現細胞(antigen presenting cells)與T細胞毒殺效應的重要環節,有少數研究指出MHC class I 對於轉移性大腸直腸癌可能是一預後因子。探究MHC class I與免疫微環境之交互作用,可能為未來之免疫治療嶄新生物標記。 以MHC class I為中心,我們依序設計兩大部分的研究。第一部份,目前已知化學治療與標靶藥物會影響腫瘤的免疫微環境(tumor immune microenvironment)以及免疫表面因子。初步資料顯示irinotecan能藉由增加癌細胞的內質網壓力(endoplasmic reticulum stress)引發後續的免疫源發細胞凋亡(immunogenic cell death, ICD)。而oxaliplatin可激發癌細胞上的MHC class I之表現。此類反應都能激發T細胞的免疫毒殺反應。因此探究化學治療「前」、「後」腫瘤細胞上免疫表面因子之動態變化,對於未來設計免疫合併治療有不可或缺的重要性。第二部分,目前初步證據已知,IFN-γ (interferon-γ,干擾素γ)/ Janus kinase (JAK)/ STAT1 (signal transducer and activator of transcription 1,轉錄訊息傳遞與活化子1)路徑活化程度乃是免疫治療之重要的預測性生物標記之一。IFN-γ/JAK/STAT1不僅是抗原表現路徑中的樞紐,更藉此影響下游的MHC class I表現。因此更深入探究大腸癌細胞STAT1的表現,對於了解免疫治療的後續機轉與調控策略之研發至為重要。 依據以上兩大命題,我們設計與執行了兩個主要方向之研究。第一主題的研究,我們首先使用流式細胞儀(flow cytometry)測試了三株大腸癌細胞株:SW480、HT29、與COLO-320。這三株大腸癌細胞株同樣都是MHC class I與NK細胞配體(natural-killer cell ligands)表現量之基礎值都很低。在接受IFN-γ刺激後,這三株細胞株的MHC class I表現量也都顯著增加,尤其是HLA-A。相反地,其NK細胞配體都對IFN-γ刺激毫無反應。後續實驗中我們使用三大類治療大腸直腸癌的化學治療藥物,包括oxaliplatin,5-FU,以及irinotecan的活性代謝物SN-38,來應用在SW480細胞株。我們發現這三個化學治療藥物都能增加大腸癌細胞上MHC class I(尤其是HLA-A)的表現量,這其中又以SN-38增加幅度最為顯著。此外,oxaliplatin與5-FU對於MHC class I提升量與藥物濃度呈正相關。相對地,SN-38在極低的劑量就有效增加MHC class I表現,其增加幅度甚至接近IFN-γ的效果,而SN-38增加MHC class I在中劑量時達到最高效果,高劑量後效果則略微降低。接著我們使用西方點墨法(Western blot)來分析抗原表現路徑中的各個訊號,進而發現化學治療藥物刺激MHC class I之機轉,主要藉由刺激TAP1與TAP2 (transporters associated with antigen processing 1 and 2,表現轉移子1與2)。ICP-47(infected cell protein 47,感染細胞蛋白-47)是單純皰疹病毒的產物,可以直接抑制人類的TAP1與TAP2。在使用Xfect將ICP-47轉染入SW480細胞後,也確實會讓原本可被化學治療藥物激發的MHC class I表現量下降。在免疫功能分析的實驗中,我們也證實藉由增加癌細胞上MHC class I之表現,SN-38能夠顯著增加單核球衍生之樹突細胞(monocyte-derived dendritic cells, MoDCs)對癌細胞進行吞噬作用。最後,我們也從臨床檢體檢測,罹患轉移性大腸直腸癌的患者在接受第一線化學治療合併標靶治療「之前」與「之後」,進由分析成對的腫瘤切片,其腫瘤細胞上的MHC class I表現量確實有大幅增加,而增加的MHC class I主要是HLA-A與HLA-B。本主題研究結果證實了化學治療藥物可以提升大腸直腸癌細胞的免疫反應。 第二主題的研究,我們首先利用流式細胞儀加以測試更多不同大腸癌細胞株:SW620與DLD-1。發現與SW480等細胞株不同的是,雖然所有的大腸癌細胞株上其MHC class I與PD-L1之基礎值皆呈現低表現量,然而在IFN-γ刺激後,SW620與DLD-1細胞株對IFN-γ刺激完全沒有反應。後續使用西方點墨法分析發現IFN-γ的下游訊息:STAT1與pSTAT1 (phosphorylated STAT1,磷酸化轉錄訊息傳遞與活化子1)之基礎表現量,在SW480等細胞株都有正常表現,但在SW620與DLD-1細胞株中STAT1與pSTAT1之基礎表現量則都下降,進而使下游的訊息傳遞路徑,包括MHC class I等之表現量也同步下降。更進一步,我們發現蛋白酶體(proteasome)抑制劑,尤其是已可在臨床上使用的bortezomib,能有效恢復pSTAT1的表現量,同時也增加了下游的訊號:包括IRF-1(interferon regulatory factor-1)與MHC class I表現量。我們後續使用PerkinElmer Opal多重染色平台對所收集的六十個轉移性大腸直腸癌患者之腫瘤檢體進行檢測。我們發現腫瘤細胞中高表現的STAT1,同時也與該腫瘤高活性免疫微環境呈正相關。在高表現STAT1腫瘤檢體中,其腫瘤細胞與腫瘤周邊細胞都表現較高的MHC class I與PD-L1,並且腫瘤周邊淋巴球(tumor infiltrating lymphocytes, TILs) 數量也顯著地提升,同時包括CD4與CD8淋巴球。最後我們將這些臨床檢體使用NanoString RNA定量平台加以分析,我們應證了這兩個平台的結果是相同的。在PerkinElmer Opal多重染色平台檢測顯示高表現STAT1的腫瘤檢體,其使用NanoString平台也顯示該腫瘤檢體有較高的STAT1 mRNA表現量。更進一步,我們也驗證了高表現STAT1的腫瘤檢體,其MHC class I,包含了HLA-A,HLA-E,與HLA-G的mRNA表現量都顯著地比低表現STAT1的腫瘤檢體來得高。最後,我們從NanoString平台結果發現高表現STAT1的腫瘤檢體中有顯著較高的IFN-γ mRNA表現量。 總和以上結論,我們的研究提供了嶄新的轉移性大腸直腸癌免疫治療策略以及免疫治療生物標記。我們的結果證明大腸癌細胞之抗原表現路徑,尤其是MHC class I之表現與細胞免疫微環境呈現高度正相關,並且也顯著較高機率引發抗原表現細胞對癌細胞進行吞噬作用。我們進一步證明化學治療藥物能有效提升癌細胞上MHC class I之表現,而最高的刺激效果來自SN-38,而非oxaliplatin或5-FU。在分析更多臨床檢體後,我們也發現MHC class I之表現受控於抗原表現路徑上游的STAT1。腫瘤細胞中高表現的STAT1,同時與高活性免疫微環境尤其是MHC class I之表現呈正相關。針對高表現STAT1腫瘤,合併免疫治療與化學治療,特別是irinotecan,能夠增強免疫治療的效果。而在低表現STAT1腫瘤,合併蛋白酶體抑制劑是一個可行的調控策略,尤其是bortezomib,可能進一步活化腫瘤的免疫微環境。由於我們研究結果證明MHC class I與STAT1可做為轉移性大腸直腸癌免疫治療之嶄新生物標記,後續我們仍需前瞻性之臨床試驗來驗證其可行性。
The development of immune checkpoint inhibitors has undergone considerable progress for patients with microsatellite instability-high (MSI-H) metastatic colorectal cancer (mCRC), which only comprises 1.8%–4.0% of all mCRC patients. Chemotherapy and targeted therapy remain the primary treatment for more than 96% of patients with mCRC with microsatellite stable (MSS) tumors. Various combination strategies have been established to treat this common form of mCRC; however, two phase III randomized trials provided disappointing results. Novel strategies, especially combination strategies, for immunotherapy in mCRC with MSS tumors, are crucial for current researches. Chemotherapy and targeted therapy might change the tumor immune microenvironment and immune surface markers. In the first part, we have known that the irinotecan increases the endoplasmic reticulum (ER) stress of tumor cells, and subsequently induces immunogenic cell death (ICD). Oxaliplatin increases the expression of major histocompatibility complex (MHC) class I. Both actions enhance the efficacy of cytotoxic T cells. Research on immune reactions before and after systemic agents for mCRC is warranted for new immunotherapy targets. In the second part, there have some biomarkers been identified as predictors for immunotherapy, with the most critical biomarker being the interferon-γ (IFN-γ)/ Janus kinase (JAK)/ signal transducer and activator of transcription (STAT)1 pathway. The IFN-γ/JAK/STAT1 pathway plays a crucial role in the antigen processing pathway and the subsequent dynamic change of downstream signals, including MHC class I. However, the role of the IFN-γ/JAK/STAT1 pathway in mCRC has received little attention. Identification of the detailed interactions between STAT1 might provide new insights for immunotherapy in mCRC. According to prior hypotheses, we designed and executed two main projects to prove our concepts. In the first project, we tested three colon cancer cell lines, SW480, COLO 320 and HT29, through flow cytometry. These cell lines all initially exhibited low expression of MHC class I and NK cell ligands. IFN-γ specifically stimulated the expression of MHC class I, particularly HLA-A. In opposite, NK cell ligands were totally irresponsive to IFN-γ stimulation. Further, we established that chemotherapy agents, particularly 7-ethyl-10-hydroxycamptothecin (SN-38) which was the active metabolite of irinotecan, stimulated the expression of stimulatory MHC class I alleles through stimulation the pathway of transporters associated with antigen processing 1 and 2 (TAP1 and TAP2) in SW480 cell line. The infected cell protein 47 (ICP47), which was a herpes simplex virus 1 (HSV-1) product, could specifically inhibit human TAP1 and TAP2. After transfection of ICP47 with Xfect, the stimulation effects of SN-38 on MHC class I and HLA-A were diminished by ICP47 due to blockade the upregulation of TAP1 and TAP2. In the functional assay, SN-38 significantly promoted the phagocytosis of colon cancer cells by monocyte-derived dendritic cells (MoDCs). This result indicated that the upregulation of MHC class I on SW480 cells induced by SN-38 made the SW480 cells more vulnerable to immune surveillance. We then confirmed that the expression of MHC class I, significantly increased after first-line chemotherapy and targeted therapy in the samples of a real-world patient with de novo mCRC. In the second project, we established that although the immune surface markers were significantly increased after IFN-γ stimulation on SW480 as mentioned above, these immune surface markers were notably unresponsive on the SW620 and DLD-1 cell lines. We discovered that the downstream signal of IFN-γ, the STAT1 and phosphorylated STAT1 (pSTAT1), were downregulated in the SW620 cell line at baseline. We verified that the STAT1/pSTAT1 could be restored through the application of proteasome inhibitors, including clinically applicable bortezomib. The expression of downstream signals of STAT1, including interferon regulatory factor-1 (IRF-1) and MHC class I, were also up-regulated by application of proteasome inhibitors. The similar results were reproduced in DLD-1 colon cancer cell line. We then analyzed sixty real word’s samples from patients with mCRC with the PerkinElmer Opal multiplex system. The positive tumor STAT1 expression was significantly associated with higher PD-L1 expression and higher MHC class I expression, on tumor cells and also non-tumor cells. The tumor infiltrating lymphocytes (TILs) also increased in the positive STAT1 group with significantly more cells expressed CD4 or CD8. We further subjected these patients’ tumors samples to NanoString analysis. Consistent with the results from the PerkinElmer Opal multiplex system, the expressions of all specific MHC class I alleles were higher in the positive-STAT1 group, and the differences were especially significant for HLA-A, HLA-E, and HLA-G. The NanoString analysis also verified the reason that high STAT1 and MHC class I expression was strongly correlated with IFN-γ expression. The IFN-γ expression was significantly higher in the positive-STAT1 group than in the negative-STAT1 group. Our research establishes novel pathways and novel biomarkers of immunotherapy for mCRC. We have proved that the MHC class I pathway is strongly correlated to highly active cell immunity and phagocytosis by antigen presenting cells. We then demonstrate that the expression of MHC class I could be elevated by application of chemotherapy agents SN-38, the active metabolite of irinotecan, rather than oxaliplatin or 5-FU. Further, we establish novel biomarker of immune microenvironment for mCRC, the STAT1. In a real-world cohort of mCRC samples, we identify two subgroups of mCRC according to STAT1 expression levels on tumor cells, high STAT1 group and low STAT1 group. For high STAT1 expression tumors, combination of chemotherapy agents, especially irinotecan, with immunotherapy could potentially increase the response. For low STAT1 expression tumors, addition of bortezomib, a proteasome inhibitor, into new immunotherapy combinations might be helpful. Our results provide the rationale for further translational researches and clinical trials of immunotherapy combinations for mCRC treatment.