晚期胰臟癌新治療之發展

Abstract

胰臟癌是一個高度致命的疾病，具有相當的醫療迫切需求，以腺癌為主，五年存活率僅5%，初診斷時僅20%病患可以接受治癒性手術。過去gemcitabine及5-FU或是5-FU類似物是最常被使用於晚期胰臟癌的藥物，但腫瘤反應率不高。目前的標準第一線併用化療處方為FOLFIRINOX以及gemcitabine併用nab-paclitaxel，但兩個處方皆具有相當程度的副作用。標靶藥物在胰臟癌的發展則是相當有限，唯一在晚期胰臟癌被核可的藥物為erlotinib，但存活改善極微。至於放射線治療雖然可以在局部腫瘤獲得不錯的控制效果，但在晚期胰臟癌的治療定位上一直難有定論。 TP53基因在胰臟癌突變的比例約七成，p53異常與細胞週期、細胞凋亡與細胞老化等功能失去調控有關。而MDM2是p53主要的負調控因子，一方面是扮演p53的E3泛素接合酶，另一方面可以抑制p53的轉錄活化功能，MDM2尚可調控癌細胞中與p53無關的促細胞存活機轉。Gemcitabine為一個核苷類似物，透過阻斷DNA複製以及促使DNA斷裂造成細胞死亡。過去其他學者研究發現gemcitabine的抗藥性與p53相關訊息傳導途徑失去調控可能有關，而且MDM2-p53這個迴路在胰臟癌可能是失去正常調控的，因此，MDM2也可能在gemcitabine的抗藥性機轉中扮演重要角色。藉著臨床研究以及細胞株研究，探索MDM2-p53調控迴路在胰臟癌的機轉，可以對於標靶治療在胰臟癌的發展提供重要線索。 本論文包含兩個主題，茲分述如下: 根據MDM2及p53在gemcitabine的抗藥性扮演重要角色的假設，首先設計了第一個研究，為一個回溯性臨床病理檢體研究。病患必須具有完整病歷資料及影像資料，且具有病理學診斷為胰臟腺癌，可取得病理組織檢體及過去曾針對胰臟癌接受以gemcitabine為基礎的化療也是必要條件。方法是利用福馬林固定石蠟包埋的腫瘤組織切片，使用免疫組織染色後做分析，必須具有至少10%癌細胞免疫染色陽性才視為陽性個案。研究結果發現MDM2陽性個案佔21.9%，p53陽性個案佔51.8%，MDM2陽性個案預後顯著較差(中位存活期為3.7個月比5.8個月，P= 0.048)，但p53免疫染色結果與預後無關。MDM2陽性表現與第一線以gemcitabine為基礎的化療後惡化或是死亡顯著相關，但p53免疫染色結果無統計顯著相關。除此之外，MDM2陽性個案在初診斷時有較高的C反應蛋白數值，但p53免疫染色結果無統計顯著相關。 根據上述臨床病理研究結果設計了胰腺癌體外細胞模式(第二個研究)，也就是利用胰臟癌細胞株探索MDM2與預後不佳的可能機轉。在實驗用的細胞株中，未處理gemcitabine前的ribonucleotide reductase regulatory subunit M2、deoxycytidine kinase、RelA/p65、p53、p21、 MDM2及MDM4/MDMX基礎表現無明顯與gemcitabine的抑制效果相關。其中Panc-1細胞株具有MDM2表現，同時又具有TP53基因的突變，顯示對於gemcitabine具有中度的抗藥性(IC50: 1.17±0.30 μM, 72-hr MTT assay)，經過暫時性MDM2之DNA質體轉染以及短時間(30分鐘)處理gemcitabine後，在clonogenic assay中可觀察到過度表現MDM2的Panc-1細胞群落較控制組多，顯示MDM2的過度表現可能使Panc-1胰腺癌細胞株對gemcitabine產生抗藥性。 為了探索放射線治療在晚期胰臟癌的角色，設計了第三個研究，以回溯性的臨床病患資料分析放射線治療在晚期胰臟癌的使用情形，結果發現使用放射線治療的病患中位存活顯著較長(14.6個月比8.1個月，P < .001)，在第一線治療選擇使用放射線治療與期別較早及CA 19-9較低有關。顯然的，回溯性研究受限於病人選擇與腫瘤負荷的大小等因素的限制，無法做出有力的結論。為了實際進一步探索放射線治療的臨床效益，設計了第四個研究，為第一期臨床試驗，採用3+3的設計，在轉移性胰臟癌病患首先使用S-1 (50–70 mg/m2/day)併用短療程及低劑量的放射線治療(2.5–3.6 Gy/day for 10–12 fractions)，完成後經過短時間的休息，再給予gemcitabine (1000 mg/m2 on days 1 and 15)併用S-1 (60–100 mg/day on days 1–7 and 15–21)，但是增加給藥期間的間隔以降低副作用，每四周為一個療程，主要試驗終點為S-1併用放射線治療的最大耐受劑量以及安全性，但由於收案進度緩慢以及提早結束試驗，並未找到最大耐受劑量。總共收案10位病患，結果S-1併用放射線治療的血液副作用輕微，而最常見的非血液副作用為輕度的腸胃不適，而局部腫瘤控制率為100%。症狀緩解也是相當顯著，疼痛控制率為56%，腸道未阻塞率為90%，新發生膽道阻塞率為17%。最終整體中位存活期為9.8 (95%CI, 7.3-12.4)個月。結果僅顯示S-1併用放射線治療對晚期胰臟癌是可耐受的治療組合，毒性尚屬輕微。 總結來說，胰臟癌腫瘤組織表現MDM2的病患預後較差且與疾病惡化相關，且和接受gemcitabine為主化療的胰腺癌病人存活呈現負相關，胰臟癌細胞株過度表現MDM2會造成gemcitabine的抗藥性。未來，持續探索MDM2的抑制或調控以進一步提昇gemcitabine的效果，將可能為胰腺癌的治療增加一線曙光。

Pancreatic cancer is highly lethal with great medical unmet needs. The majority of patients have adenocarcinoma. The 5-year overall survival (OS) rate is 5%. At initial diagnosis, only 20% of patients have resectable diseases. Gemcitabine and 5-FU/5-FU analogs are the mostly utilized agents until recent years with low response rate (RR). The standard first-line combination chemotherapy regimens, such as FOLFIRINOX or gemcitabine plus nab-paclitaxel, are associated with significant toxicities. The development of targeted therpy is quite limited. The only approved drug is erlotinib with minimal OS benefit. Good local control can be achieved with radiotherapy (RT). However, the role of RT in advanced pancreatic cancer is under debate. TP53 is mutated in 70% of pancreatic cancer. The dysfunction of p53 is associated with dysregulated cell cycle, apoptosis, and senescence. MDM2 is the major negative regulator of p53, functions both as an E3 ubiquitin ligase and an inhibitor of p53 transactivation. In addition, MDM2 also regulates p53-independent pro-survival signaling in cancer cells. Gemcitabine, a nucleoside analog, works by blocking DNA replication and inducing DNA breaks, which results in cell death. Other investigators found that the resistance of gemcitabine was probably associated with defective p53-associated pathways. The functional intactness of the MDM2-p53 loop may be defective in pancreatic cancer. Therefore, MDM2 may be also implicated in the resistance of gemcitabine. To explore the implication of MDM2-p53 loop in clinical setting and in vitro study may provide us important clues for development of targeted therapy in pancreatic cancer. This thesis included two themes. These studies were described as follows: Based on the hypothesis that both of MDM2 and p53 are implicated with the resistance of gemcitabine, the retrospective study (Study 1) was proposed. Patients with complete medical records, imaging data and histopathologic diagnosis of pancreatic adenocarcinoma were selected. Archival tissues and prior exposure to gemcitabine-based chemotherapy for pancreatic cancer were required. Formalin-fixed, paraffin-embedded tumor tissue sections were stained with immunohistochemistry. When at least 10% of the tumor cells had positive staining, MDM2 and p53 expression were considered positive. As a result, expression of MDM2 and p53 was found in 21.9% and 51.8% patients, respectively. MDM2 was significantly associated with poor prognosis (median OS 3.7 vs 5.8 months, P = .048), while p53 had no prognostic significance. MDM2 but not p53 was significantly associated with progressive disease or death after first-line gemcitabine-based chemotherapy. In addition, MDM2 but not p53 expression was significantly associated with high C-reactive protein (CRP) level at initial diagnosis. Following the clinical finding of MDM2, in vitro cell line study of pancreatic adenocarcinoma (Study 2) was performed to explore the mechanism of MDM2-associated poor prognosis. Among the cell lines, the baseline expression of ribonucleotide reductase regulatory subunit M2, deoxycytidine kinase, RelA/p65, p53, p21, MDM2, and MDM4/MDMX was not associated with gemcitabine sensitivity. Panc-1, a cell line with MDM2 expression and TP53 mutation, was moderately resistant to gemcitabine (IC50: 1.17±0.30 μM, 72-hr MTT assay). After transient tranfection of MDM2 DNA plasmid and short-term (30 mins) exposure to gemcitabine in the clonogenic assay, gemcitabine resistance was demonstrated with increased colonies in Panc-1 cells with MDM2 overexpression. To explore the role of radiotherapy in advanced pancreatic cancer, the retrospective study (Study 3) about the utilization of radiotherapy was proposed. The median OS was significantly better in patients with RT than those without it (14.6 vs. 8.1 months, P < .001). The choice of RT in pancreatic cancer for first-line therapy was significantly associated with initial stage and CA 19-9 level. Obviously, the retrospective study, which was limited to potential bias from patient selection and tumor burden, remained inconclusive. To explore the clinical feasiblity of RT combined with S-1, the phase I study (Study 4) was proposed to use S-1-based, short-course and low-dose RT followed by gemcitabine plus S-1 (GS) in metastatic pancreatic cancer. The 3+3 trial design was adopted. First, patients received concurrent chemoradiotherapy (CCRT) with RT (2.5–3.6 Gy/day for 10–12 fractions) and S-1 (50–70 mg/m2/day). After a short rest, gemcitabine (1000 mg/m2 on days 1 and 15) and S-1 (60–100 mg/day on days 1–7 and 15–21), were administered in a 4-week cycle. The primary endpoint was the maximum tolerated dose (MTD) and safety profiles of CCRT. As a result, the MTD of CCRT was not identified due to the slow recruitment and premature termination. Ten patients were enrolled. In the CCRT part, hematological toxicity was mild. The most common non-hematological toxicities were mild gastrointestinal upsets. The local disease-control rate (DCR) was 100% after CCRT. Palliation was significant with the pain control rate of 56%, free-from-bowel obstruction rate of 90%, and new biliary obstruction rate of 17%. The median OS was 9.8 (95%CI, 7.3-12.4) months. These results demonstrated that S-1-based CCRT was tolerable in metastatic pancreatic cancer with manageable toxicities. In summary, MDM2 but not p53 expression in pancreatic tumor tissues was associated with poor prognosis and disease progression. MDM2 overexpression was associated with in vitro gemcitabine resistance. In the future exploration, enhancement of gemcitabine efficacy through inhibition of MDM2 may shed light on the therapy in advanced pancreatic cancer