鄭安理臺灣大學：臨床醫學研究所林家齊Lin, Chia-ChiChia-ChiLin2007-11-272018-07-062007-11-272018-07-062006http://ntur.lib.ntu.edu.tw//handle/246246/55537膀胱癌最常見的細胞型態為泌尿上皮細胞癌（urothelial carcinoma）（>90%）。膀胱癌根據原發腫瘤侵犯的深度可分為：T1為腫瘤侵犯至lamina propria、T2為腫瘤侵犯至膀胱肌肉層、T3為腫瘤侵犯至膀胱周圍脂肪、T4為腫瘤侵犯至膀胱周圍器官。其中T1稱為表淺性膀胱癌（superficial bladder cancer），其標準治療為經尿道腫瘤切除術，復發後若仍為表淺性，可考慮膀胱灌注藥物（如BCG、doxorubicin、mitomycin-C）以減少復發的機會；T2稱為肌肉層侵犯膀胱癌（muscle-invasive bladder cancer），其標準治療，在美國多為根除性膀胱切除手術（radical cystectomy），在歐洲多為全膀胱放射治療（whole bladder irradiation），至於手術前後化學治療的角色仍待進一步的評估；M1稱為遠端轉移膀胱癌（metastatic bladder cancer），其標準治療為全身性化學治療，在美國多使用MVAC處方，在歐洲多使用CMV處方。 台灣泌尿上皮細胞癌的特點在於上泌尿道（即腎盂及輸尿管）相對於膀胱的比例較高，手術切除腫瘤常包括腎臟切除，導致腎功能不佳。 在本論文中，我們試圖將5-fluorouracil（5-FU），一個在泌尿上皮細胞癌過去並不常使用的藥物，與cisplatin為基礎的化學治療做結合。將分成三個部份來討論：第一，我們發展新的以cisplatin及5-FU為基礎的第一線處方；第二，我們進一部探討這些新處方，包括分析可以預測反應率或存活期的臨床特徵、測試這些新處方在特殊晚期病患（例如；病患接受腎臟移植又合併有轉移性移形上皮細胞癌）的可行性、測試這些新處方在特殊非晚期病患（例如；肌肉層侵犯膀胱一型上皮細胞癌的病患想保留膀胱）的可行性；第三，我們發展第二線處方來治療對cisplatin及5-FU為基礎的治療失敗的轉移性移形上皮細胞癌。Urothelial carcinoma (UC) is the second most common genitourinary malignancy. In 2002, the crude incidence in Taiwan is 7 cases/100,000/year and the mortality 2 cases/100 000/year. Most cases present with superficial disease. Nevertheless, 20% have muscle-invasive disease at presentation, while 5% present with metastatic disease. In addition, 50% of patients undergoing radical surgery for muscle-invasive UC will relapse. Relapses cannot be cured by local treatment in the majority of cases. Therefore, surgically inoperable disease comprises three distinct groups: inoperable disease due to local extension or grossly involved pelvic or paraortic lymph node (stages T4b, N2-3), metastatic disease at diagnosis, or recurrent disease after radical surgery. Systemic chemotherapy represents the most useful option for patients with surgically incurable disease. The landmark of systemic chemotherapy in advanced UC was the development of the combination of methotrexate, vinblastine, doxorubicin and cisplatin (MVAC) at the Memorial Sloan-Kettering Cancer Center (MSKCC) in 1983. Its activity against UC has been considerable with response rates (RR) of >50%, 3-year survival of 20-25% and median survival of >1 year [Sternberg et al., 1988]. Furthermore, MVAC was shown to be superior to single-agent cisplatin in a randomized trial establishing the role of combination chemotherapy in advanced UC [Loehrer, Sr. et al., 1992]. Several MVAC-like regimes have also been studied. The combination of cisplatin, methotrexate and vinblastine (CMV) was contemporary with MVAC and was developed at Stanford University. In a randomized study of 214 patients it was shown to be superior to the combination of methotrexate and vinblastine [Mead et al., 1998], underlying the importance of cisplatin in prolonging survival and establishing the routine use of cisplatin-based combination chemotherapy. The omission of vinblastine from MVAC led to the MD Anderson Cancer Center (MDACC) CISCA regimen (cisplatin, cyclophosphamide and doxorubicin). The CISCA regimen was found to be inferior to MVAC in a randomized trial of 110 patients [Logothetis et al., 1990]. Therefore, most centers worldwide have considered MVAC regimen as the standard treatment for advanced UC cancer for almost two decades. Despite superior outcomes severe toxicity significantly limits MVAC usefulness for advanced UC. Ten to 20% grade 3 and 4 mucositis, 20 to 30% grade 3 and 4 leukopenia, and a 3 to 4% toxic death rate are consistently seen in randomized trials. Perhaps most importantly only a small subset of patients treated with MVAC has long-term disease-free survival (3.7% at 6 years) [Saxman et al., 1997]. Two strategies have therefore been used to improve over MVAC: the use of granulocyte-colony stimulating factor (G-CSF) and the study of new combinations. The use of G-CSF can reduce myelotoxicity but also allowed for the administration of MVAC in a more dose-dense fashion [Gabrilove et al., 1988; Loehrer, Sr. et al., 1994; Seidman et al., 1993; Logothetis et al., 1995]. The results of phase II studies indicated that the efficacy of MVAC might be improved by dose intensification. In order to change the toxicity profile and possibly increase the efficacy of MVAC, other combinations based on platinum have been studied. The development of novel platinum-based combinations has focused on taxanes and gemcitabine, mainly because of their single-agent activity and their synergic effects with platinum. The gemcitabine-cisplatin combination (GC) has been evaluated in three studies that involved a total of 121 patients and used different administration schedules. All studies showed high overall RR and complete response (CR) rate, while median overall survival (OS) was also encouraging at 12-14 months [von der Maase et al., 1999; Kaufman et al., 2000a; Moore et al., 1999]. With longer follow-up, the median OS was 13.2 months and estimated 4-year survival was 13% [Stadler et al., 2002]. The combination of docetaxel and cisplatin every 3 weeks has also been evaluated in three studies. In a total of 123 patients, the overall RR ranged from 52% to 62% and the median OS was reported to be 8-14 months [Dimopoulos et al., 1999; Sengelov et al., 1998; Garcia del Muro et al., 2002]. Regimens of combined paclitaxel and carboplatin, usually given every 3 weeks, were evaluated in three studies, involving a total of 77 patients. The overall RR ranged from 21% to 65% with the median OS around 9 months [Small et al., 2000; Redman et al., 1998; Zielinski et al., 1998]. The promising efficacy and the favorable toxicity profile shown in the above-mentioned studies led to four more recent randomized studies comparing MVAC with other combinations of platinum. An EORTC study showed that intensified MVAC (HD MVAC) with G-CSF support was less toxic than classic MVAC. The RR, CR rate and progression-free survival (PFS) were superior for HD MVAC but the difference in OS was non-significant in the initial publication [Sternberg et al., 2001b]. With longer follow-up, the initial results have been confirmed with PFS and morality hazard ratios of 0.73 and 0.76, respectively and 2-year and 5-year survival rates of 36% compared with 26% and 21% compared with 13% for HD MVAC and classic MVAC, respectively [Sternberg et al., 2006]. In another randomized study the combination of GC was found to be equally effective with regard to response time to progression and survival but less toxic than classic MVAC [von der Maase et al., 2000]. This trial was powered to detect a 33% (4 months) difference in survival between the arms; it was not designed to prove that the treatments were equivalent, which would have required a much larger trial. Long-term results of this study have also been published recently, confirming similar 5-year survival and PFS rates for GC (13% and 9.8%) and MVAC (15.3% and 11.3%) [von der Maase et al., 2005b]. In the third randomized study the combination docetaxel–cisplatin was compared with G-CSF supported MVAC. Docetaxel–cisplatin was found to be inferior to MVAC, although the difference in survival did not reach statistical significance due to an imbalance in performance status between the two arms [Bamias et al., 2004]. The phase III study designed to compare MVAC with paclitaxel-carboplatin never completed accrual and, therefore, no conclusions can be drawn [Dreicer et al., 2004]. The results of the former two studies indicate that the standard in chemotherapy for advanced UC may be changing. As shown above the last decade has been characterized by the use of chemotherapy doublets, some of which were compared with standard MVAC. Nevertheless, the development of several active agents, as well as the wider application of G-CSF, led to the study of three or more drug combinations in advanced UC. The first approach is to add a new agent, either a taxane or gemcitabine, to the ifosfamide-cisplatin regimen. The first triplet was developed at the MSKCC, who tested the combination of ifosfamide, paclitaxel, and cisplatin (ITP). They reported a 68% RR and an impressive median survival of 18 months [Bajorin et al., 2000]. Based on these results and the Norton-Simon hypothesis of chemotherapy sensitivity (Non-cross resistant sequenced chemotherapy has been shown to improve outcomes in breast cancer.), the same group sequentially administered five agents to facilitate drug delivery [Dodd et al., 2000]: six cycles of doxorubicin and gemcitabine (AG) were followed by four cycles of ITP. Fifty-six patients have been treated at the MSKCC with an overall RR of 73%, which is higher than expected with either AG or ITP [Maluf et al., 2000]. The second triplet was developed at the MDACC, who tested the combination of ifosfamide, gemcitabine, and cisplatin. They reported a 41% RR [Pagliaro et al., 2002]. The second approach is to add a platinum, to the taxane-gemcitabine regimen. Bellmunt et al. [Bellmunt et al., 2000] and the Spanish Oncology Genitourinary Group have conducted a phase I/II trial of the combination of gemcitabine, cisplatin, and paclitaxel (GCP). The overall RR and CR rates at the phase II part of the study were 76% and 26%, respectively, and the median survival was 15.6 months. In a randomized phase II study, Lorusso et al. [Lorusso et al., 2005] reported an RR of 43% with a CR rate of 12% among 42 patients treated with GCP. Response rates were similar to those of GC. Nevertheless, this study was underpowered to detect significant differences in survival. Pectasides et al. [Pectasides et al., 2002] studied a weekly regimen of gemcitabine, cisplatin, and docetaxel in 35 patients. A RR of 65% and a median survival of 15.5 months were reported in a population of predominantly good prognosis (77% PS 0 or 1). Hussain et al. [Hussain et al., 2001] conducted a phase II trial evaluating the efficacy of the combination gemctabine-carboplatin-paclitaxel in patients with advanced UC. Most of the 49 patients who were enrolled would have been eligible for cisplatin-based chemotherapy. The overall RR was 68%, with a CR rate of 32% and a median survival of 14.7 months. In a similarly designed study, Hainsworth et al. [Hainsworth et al., 2005] failed to duplicate these results. Lower RR (43%), CR rate (12%) and median survival (11 months) in 60 patients with similar prognostic features were reported. The third approach is to add a non-platinum, to the taxane-gemcitabine regimen. The preliminary results of a phase I/II study by Lara et al. [Lara, Jr. et al., 2004] reported the use of a non-cisplatin triplet built up with paclitaxel, methotrexate and gemcitabine. In 21 evaluable patients the overall RR was 57% (6 CRs and 3 PRs), median overall and progression-free survival was 18 and 6.3 months, respectively and the most common toxicity was neutropenia. Data from triplet studies so far demonstrate the feasibility of such combinations, since treatment was generally well tolerated. The precise role of these combinations will be clarified by the EORTC randomized trial (30987), which compared GCP with GC and has recently completed accrual [de Wit R., 2003]. In this thesis, we incorporated 5-fluorouracil (5-FU), a chemotherapeutic agent uncommonly used in advanced UC, to cisplatin-based chemotherapy. Three aspects will be addressed: In the first part, we developed two new first-line regimens based on cisplatin and 5-FU. In the second part, we defined the effect of pretreatment patient characteristics on clinical outcome, tested the feasibility of these new regimens in the setting of renal allograft recipients with metastatic UC, and incorporated these new regimens to a bladder-conserving strategy in patients with muscle-invasive bladder UC. In the third part, we developed second-line regimens for UC after failure to these new cisplatin and 5-FU-based regimens.封 面 口試委員、指導教授與所長簽名表 國家圖書館碩博士授權書 致 謝 目 錄 中文摘要（Abstract in Chinese） 1 緒 論（Introduction） 2 研究方法與材料（Materials and Methods） 23 結 果（Results） 29 討 論（Discussion） 35 展 望（Perspectives） 51 論文英文簡述（Abstract in English） 60 參考文獻（References） 76 圖 表（Tables and Figures） 94 附 錄（Appendix） 1251657140 bytesapplication/pdfen-US膀胱癌移形上皮細胞癌抗癌合併化學治療處方太平洋紫杉醇bladder neoplasmscarcinoma, transitional cellantineoplastic combined chemotherapy protocolsfluorouracilpaclitaxel[SDGs]SDG3texthttp://ntur.lib.ntu.edu.tw/bitstream/246246/55537/1/ntu-95-D89421002-1.pdf