辅助性子宫切除在妊娠滋养细胞肿瘤治疗中的作用

目的:评价辅助性子宫切除在妊娠滋养细胞肿瘤(GTN)治疗中的作用。方法:回顾性分析1990～2010年在我院接受辅助性子宫切除的GTN患者71例的临床资料,评价子宫切除的指征、住院时间、化疗疗程以及血β-HCG的变化,并比较接受了全子宫切除术+化疗(手术+化疗组)的低危GTN患者与仅进行了化疗的低危GTN患者的临床资料和结局。结果:671例GTN患者中有71例接受了辅助性子宫切除。主要手术指征为患者要求(54.9%)、耐药病灶(39.4%)和出血(19.7%)。与仅行化疗的低危GTN患者相比,接受辅助性子宫切除在住院时间、化疗疗程、血β-HCG降至正常的时间方面明显缩短。结论:虽然化疗是治疗GTN的主要手段,在某些特殊情况下,辅助性子宫切除仍然在GTN的治疗中起重要作用。     

FAV方案和FAEV方案作为初始方案治疗高危型妊娠滋养细胞肿瘤的疗效比较

目的:探讨FAV方案(氟脲苷+放线菌素D+长春新碱)和FAEV方案(氟脲苷+放线菌素D+依托泊苷+长春新碱)作为初始化疗方案治疗高危型妊娠滋养细胞肿瘤(GTN)的效果。方法:回顾性分析2005年1月至2013年12月在北京协和医院接受初次化疗且首选方案为FAV方案(FAV组,32例)或FAEV方案(FAEV组,42例)的高危型GTN患者的治疗情况,并比较两组治疗方案的疗效。结果:FAV组和FAEV组达到血清学完全缓解(SCR)的患者分别为21例和30例(65.6%vs71.4%,P=0.59),其中FAV组有2例复发,FAEV组无复发(9.5%vs 0,P=0.17)。FAV组和FAEV组对化疗耐药(NR)的患者分别为9例和6例(28.1%vs 14.3%,P=0.14),因难以耐受化疗副反应而更改方案的患者分别为2例和6例(6.3%vs 14.3%,P=0.27)。两组中的耐药患者和出现难以耐受化疗副反应的患者经过更换化疗方案,并且部分患者接受辅助手术治疗后均获得SCR。FAV组中耐药患者,FAEV、EMA-CO和EMA-EP 3种方案的SCR率分别为42.9%(3/7)、71.4%(5/7)和50.0%(1/2)。FAEV组中耐药患者,EMA-CO和EMA-EP方案的SCR率分别为66.7%(4/6)和100.0%(2/2)。结论:FAV和FAEV方案作为高危型GTN患者的一线治疗方案是有效的,两种方案的疗效和患者耐受性无明显差异。对于FAV或FAEV方案耐药的高危型GTN患者,EMA-CO均可作为理想的二线化疗方案。     

动脉栓塞治疗妊娠滋养细胞肿瘤6例临床分析

目的:探讨妊娠滋养细胞肿瘤(GTN)患者行动脉栓塞术治疗的方法及疗效。方法:回顾分析2008年1月至2014年12月在复旦大学附属妇产科医院因大出血或子宫动静脉瘘而行子宫或股动脉栓塞术治疗的6例妊娠滋养细胞肿瘤患者的临床资料。结果:6例GTN患者中有转移病灶伴大出血者3例,超声检查发现子宫动静脉瘘3例;入院即刻行栓塞术者4例,化疗后行栓塞术者2例;6例患者均栓塞成功。栓塞后接受手术治疗3例。所有患者均采用单药或多药联合化疗平均9.12个疗程(6~11疗程),5例(83.3%)患者完全缓解。结论:有出现转移病灶、子宫动静脉瘘时采用动脉栓塞术及时安全有效地控制病情,为全身化疗提供治疗机会。栓塞术后全身化疗仍是GTN的主要治疗方法。     

复发性卵巢未成熟畸胎瘤34例分析

目的 总结复发性卵巢未成熟畸胎瘤的特点，以及复发后手术和化疗的经验。方法 对1949～2001年北京协和医院34例复发性卵巢未成熟畸胎瘤进行回顾性分析。结果 去除5例复发后仅行活检者，其余29例随访中位数时间84个月(1l-286个月)，28例存活(96．7％)。复发后手术病理级别为0级共23例次，17例次未再复发，未复发的患者中9例次术后未化疗，其余8例次接受1～8个疗程化疗。复发后手术病理级别为1～3级共17例次。9例次未再复发，均接受方案为VAC(长春新碱、环磷酰胺、阿霉素)≥4个疗程或顺铂(DDP)为主的联合化疗≥3个疗程。结论 复发性卵巢未成熟畸胎瘤的治疗效果是很好的。理想的再次肿瘤细胞减灭术非常重要，复发后病理级别未转为0级者应辅以规范有效的化疗(DDP为主的联合化疗至少3个疗程)。0级畸胎瘤可以不化疗，但仍有增长可能，应密切随诊，延长随诊时间，积极手术。     

经宫腹腔镜诊断的妊娠滋养细胞肿瘤预后分析

目的:初步探讨宫腹腔镜用于妊娠滋养细胞肿瘤(GTN)诊断的安全性和对肿瘤预后的影响.方法:回顾性分析经宫腹腔镜诊断为GTN的11例病例资料,其中绒毛膜癌9例,胎盘部位滋养细胞肿瘤2例.结果:11例中行宫腔镜下清宫或病灶切除2例,腹腔镜下病灶切除术9例.手术平均时间48.2±32.4分钟(15～120分钟),中位出血量30ml(5～800ml),无术中和术后并发症.术后1例采用腹腔镜手术治疗,10例接受以氟尿嘧啶为主的联合化疗3～6疗程,平均4.4±1.0疗程,血清β-HCG下降至正常水平平均需化疗2.0±0.8疗程(1～3疗程),其中3例联合经腹手术治疗.所有病例均获得完全缓解,平均随诊31.3±18.7月(3～56月),均无瘤存活.结论:对于临床特征不典型的GTN病例,官腔镜和(或)腹腔镜检查是可供选择且安全的诊断方法,对预后无不利影响.     

多囊性肾脏发育不良胎儿的遗传学检测及妊娠结局分析

目的:评估预防性子宫切除对40岁以上葡萄胎患者恶变为妊娠滋养细胞肿瘤(GTN)的治疗效果。方法:回顾性分析南京医科大学附属妇产医院2012年1月至2020年5月期间收治40岁以上葡萄胎患者共78例,其中预防性子宫切除患者40例、期待治疗患者38例,对相关临床资料进行分析。结果:预防性子宫切除组葡萄胎恶变率(37.5%)与期待治疗组(50.0%)比较,差异无统计学意义((印)P(正)>0.05);恶变组的平均停经时间(10.92±3.65周)较治愈组(8.36±4.35周)长,差异有统计学意义((印)P(正)<0.05);期待治疗后GTN化疗前血β-hCG数值、化疗疗程及耐药率较预防性子宫切除后GTN高,差异有统计学意义((印)P(正)<0.05)。结论:对于40岁以上的葡萄胎患者,预防性子宫切除可以提高GTN的早期诊断率,降低化疗量,且无增加转移病灶的风险,可以作为有效的治疗策略。     

56例复发性卵巢癌无瘤生存期影响因素及治疗后预后分析

目的:探讨复发性卵巢癌患者无瘤生存期(DFI)相关影响因素并分析两种治疗方案患者的生存预后。方法:回顾性分析56例复发性卵巢癌患者临床资料。按复发后治疗方法不同分为二次肿瘤细胞减灭术联合术后化疗22例(手术组),单纯化疗34例(化疗组)。结果:1病理类型、组织学分级、临床分期、初次术后化疗疗程数及初次术后残余病灶大小与患者DFI有关(P0.05),年龄与DFI无关(P0.05);多因素分析提示临床分期、初次术后残余病灶大小是DFI独立影响因素,临床分期越早、初次术后残余病灶越小,DFI越长。2手术组较化疗组复发后中位生存时间明显延长,分别为30月与16月(χ~2=10.849,P=0.010)。复发后化疗组1、2、3、4年生存率分别为65%,32%,8%,0,手术组分别为95%,75%,29%,0;手术组复发后生存率较化疗组高,差异有统计学意义(P0.05)。结论:复发性卵巢癌患者DFI与病理类型、组织学分级、临床分期、初次术后化疗疗程数及初次术后残余病灶大小相关,临床分期、初次术后残余病灶大小是DFI的独立影响因素。二次肿瘤细胞减灭术联合化疗可提高患者复发后的近期生存率。     

紫杉醇联合铂类治疗持续耐药及复发性妊娠滋养细胞肿瘤的疗效分析

目的:评价紫杉醇联合铂类药物化疗方案治疗持续耐药及复发性妊娠滋养细胞肿瘤(GTN)患者的疗效及安全性。方法:回顾性分析2006年1月至2013年1月,在北京协和医院接受紫杉醇联合铂类化疗方案治疗的25例持续耐药及复发性GTN患者的治疗情况及最终治疗结局。结果:25例持续耐药及复发性GTN患者共接受了115疗程的紫杉醇联合铂类的化疗方案,平均每例4.6±2.2(2~10)疗程,具体包括紫杉醇+顺铂(TP)方案52疗程、紫杉醇+卡铂(TC)方案56疗程及紫杉醇+依托泊苷/紫杉醇+顺铂(TE/TP)方案7疗程。在停止化疗时,血清学完全缓解14例,部分缓解4例,治疗无效7例,完全缓解率为56.0%(14/25),总缓解率为72.0%(18/25),血清学完全缓解后的复发率为35.7%(5/14),平均复发时间为95.4±18.4天(约3.2个月)。紫杉醇联合铂类方案的毒副反应主要为骨髓抑制、消化道反应、肝肾损伤及过敏等,发生Ⅲ~Ⅳ度骨髓抑制患者比例为48.0%,未发生致死性副反应。结论:紫杉醇联合铂类对持续耐药及复发性GTN患者是可供选择的化疗方案。     

肺叶切除术治疗妊娠滋养细胞肿瘤肺转移的疗效分析

目的 评价肺叶切除术治疗妊娠滋养细胞肿瘤(GTN)肺转移的疗效.方法 对1995年1月-2005年12月间北京协和医院收治的62例凶GTN肺转移行肺叶切除术患者的临床病理资料进行回顾性分析.根据术前临床治疗情况将上述患者分为复发性GTN(A组,10例)、耐药性GTN(B组,28例)和化疗过程中血清人绒毛膜促件腺激素β亚单位(β-hCG)水平呈对数下降、化疗效果满意但肺部病灶持续存在者(C组,25例),其中1例患者分别因耐药与复发两次接受肺叶切除术治疗,放同时进入A组和B组.结果 62例患者总的完伞缓解率为89%(55/62),其中A、B、C组完全缓解率分别为90%(9/10)、79%(22/28)和100%(25/25),B组明显低于C组(P=0.024),其他组间比较,差异则均无统计学意义(P>0.05).3组患者的复发率分别为2/8、15%(3/20)和0.3组高危[即国际妇产科联盟(FIGO)GTN评分≥7分]患者比例分别为90%(9/10)、82%(23/28)和44%(11/25),C组明显低于A、B组(P<0.05);B组患者术前化疗疗程数(7个疗程)明显多于A、C组(分别为3和5个疗程;P<0.05);A、B组患者术前血清β-hCG水平未达正常所占百分比[分别为50%(5/10)61%(17/28)]明显高于C组[为12%(3/25);P<0.05];而3组术后病理阳性率分别为60%(6/10)、36%(10/28)和12%(3/25),C组明显低于A、B组(P<0.05).结论 肺叶切除术对于GTN肺转移是一种有效的治疗方法 .对于肺部病灶相对局限的耐药和复发患者,建议在化疗后适时行肺叶切除术;而对于化疗过程中血清β-hCG水平呈对数下降、化疗效果满意而肺部病灶持续存在的初治患者,町严密随诊,暂不必选择手术治疗.     

依托泊苷联合顺铂治疗高危、耐药及复发性妊娠滋养细胞肿瘤的疗效分析

目的探讨依托泊苷联合顺铂（EP）方案治疗高危、耐药及复发性妊娠滋养细胞肿瘤（GTN）的疗效及安全性。方法回顾性分析接受EP方案化疗的39例GTN患者的临床资料,其中初治高危患者25例,耐药患者9例,复发患者5例。39例患者中10例患者辅以手术治疗,所有患者均随访,观察其疗效及毒副反应,并追踪继发性肿瘤的发生情况和其中30例保留生育功能患者的生育情况。结果39例GTN患者共接受了221个疗程的EP方案化疗,平均疗程数5．7个,总的完全缓解率为74％（29／39）。其中,25例初冶高危患者总疗程数139个,平均疗程数5．6个,19例完全缓解,6例耐药,完全缓解率为76％（19／25）;9例耐药患者化疗总疗程数55个,平均疗程数为6．1个,6例获完全缓解,3例再次耐药,完全缓解率为6／9;5例复发患者化疗总疗程数27个,平均疗程数为5．4个,4例完全缓解,1例耐药死亡,完全缓解率为4／5。该方案主要的毒副反应是骨髓抑制、消化道反应及脱发,骨髓抑制较轻,均为Ⅰ-Ⅲ度,未发生致死性毒副反应。30例保留生育功能患者共获妊娠8例次,其中人工流产2例次,足月妊娠分娩6例次,共获新生儿6个,均无先天性畸形发生,随访期内（随访时间最长者5年）小儿牛长发育正常。所有存活患者尢继发性肿瘤发生。结论EP方案作为一种安全、有效的化疗方案,口丁用于高危、耐药及复发性GTN患者的治疗。     

化疗联合手术治疗高危型妊娠滋养细胞肿瘤25例分析

目的:分析高危型妊娠滋养细胞肿瘤治疗中手术的重要性。方法:收集我院自2000年1月～2010年8月间收治的高危型妊娠滋养细胞肿瘤患者的临床资料,分析其临床特点及手术联合化疗的临床转归。结果:25例中13例行次广泛子宫切除加双附件切除,2例行全子宫切除,其余10例分别行子宫病灶切除或子宫外转移灶切除术,除2例自动出院失访外,23例均获完全缓解,临床缓解率92%,平均化疗疗程5.8个。结论:在高危型妊娠滋养细胞肿瘤中应用强有力化疗的同时正确选择手术可有效地控制病情,减少化疗疗程,降低化疗副反应,提高治愈率。     

Japanese trial for classification of gestational trophoblastic disease

OBJECTIVE: To evaluate the clinical usefulness of the Japanese Diagnostic Score to differentiate choriocarcinoma clinically without histologic findings from persistent gestational trophoblastic disease (GTD). STUDY DESIGN: We reviewed the clinical records and histologic reports on all 809 patients with persistent GTD treated with surgery and chemotherapy in Japan. There were 347 cases of choriocarcinoma and 462 cases of invasive mole with histologic confirmation. We retrospectively applied the Japanese Diagnostic Score to all patients for detection of choriocarcinoma in persistent trophoblastic disease. RESULTS: The sensitivity of the score for choriocarcinoma was 92.2%. The specificity was 93.5%. This retrospective study showed that the accuracy of this scoring system to differentiate true malignant choriocarcinoma clinically from both low risk and high risk gestational trophoblastic neoplasia without histologic findings was 92.9%. CONCLUSION: Our trial to differentiate choriocarcinoma clinically from persistent GTD without histologic findings using a unique scoring system was successful. Proper management in the early stages strongly influences the outcome of these diseases. This scoring system should be very useful in comparing the incidence and survival rate of choriocarcinoma between nations.     

Role of surgery in the management of high-risk gestational trophoblastic neoplasia

OBJECTIVE: To evaluate the role of surgery in the management of high-risk gestational trophoblastic neoplasia. STUDY DESIGN: Twenty-four (48%) of 50 patients treated with etoposide, methotrexate, actinomycin D, cyclophosphamide and vincristine (EMA-CO) regimen as primary or secondary chemotherapy for high-risk gestational trophoblastic neoplasia between 1986 and 2005 underwent 28 adjuvant surgical procedures. The procedures included hysterectomy (17), lung resection (5), salpingectomy (1), uterine wedge resection (1), small bowel resection (1), suturing of the liver or uterus for bleeding (2) and uterine artery embolization (1). RESULTS: Twenty-one (87.5%) of 24 patients who had surgical procedures as part of their treatment for high-risk disease survived. Fifteen (88%) of 17 patients undergoing hysterectomy were cured. Four (80%) of 5 patients who had resistant foci of choriocarcinoma in the lung were cured by pulmonary resection. The patients who had suturing of the uterus, uterine artery embolization, small bowel resection and salpingectomy for bleeding as well as the patient who had uterine wedge resection of resistant choriocarcinoma survived. CONCLUSION: Adjuvant surgical procedures, especially hysterectomy and pulmonary resection for chemotherapy-resistant disease as well as procedures to control hemorrhage, are important components in the management of high-risk gestational trophoblastic neoplasia. Twenty-four (48%) of 50 patients with high-risk gestational trophoblastic neoplasia in this series underwent surgical procedures, and 21 (87.5%) were cured.     

Gestational trophoblastic neoplasia and evidence-based medicine

OBJECTIVE: To determine whether protocols for the management of gestational trophoblastic neoplasia conform to the principles of evidence-based medicine. STUDY DESIGN: Protocols for the management of low- and high-risk gestational trophoblastic neoplasia were examined to determine to what extent they conformed to the principles of evidence-based medicine. RESULTS: Nearly all current chemotherapy regimens for gestational trophoblastic neoplasia are based on the experience of management of various risk groups, variously defined. Some prospective, randomized studies were flawed by faulty selection criteria. Local population variations may influence the results of management. CONCLUSION: The management of trophoblastic neoplasia is based on physician experience. Nearly all prospective, randomized studies have been flawed. There is a need for carefully planned prospective studies with stringent inclusion criteria to determine the most effective and cost-effective and least toxic therapy, particularly for low-risk neoplasia.     

Clinicopathologic characteristics and subsequent pregnancy outcome in 139 complete hydatidiform moles

OBJECTIVE: The most common form of gestational trophoblastic disease is the complete hydatidiform mole (CHM). The study reports our experience of clinicopathologic characteristics and subsequent pregnancy outcome of patients with CHM. STUDY DESIGN: One hundred fifty-one subsequent cases with initial diagnosis of CHM were re-evaluated histopathologically. Clinical characteristics, the need for chemotherapy and subsequent pregnancy outcome were evaluated. RESULTS: Twelve out of 151 cases were re-evaluated as hydropic abortion, as partial hydatidiform moles or were insufficient for morphologic examination and therefore excluded from further analysis. The leading clinical symptoms of the remaining 139 cases were irregular vaginal bleeding (67%) and uterine enlargement (41%). Twenty-six patients (19%) required chemotherapy because of gestational trophoblastic neoplasia (GTN; low-risk: 23 out of 26). All patients were cured successfully. The subsequent pregnancy rate was 15% (21/139). Five patients suffered from abortions, 12 women delivered a healthy offspring. Four women presented with recurrent CHM with a spontaneous normalization of HCG levels after D&C. CONCLUSIONS: The clinical and morphologic diagnosis of CHM is a challenge, and diagnosis as well as treatment should be multidisciplinary and centralised. One fifth of CHM are at risk of a GTN, but the cure rate is 100% with adequate management. Pregnancy outcome following CHM is complicated by an increased risk of abortion.     

Duration of human chorionic gonadotropin surveillance for partial hydatidiform moles

OBJECTIVE: Partial hydatidiform moles infrequently progress to gestational trophoblastic neoplasia. The purpose of this study was to determine the optimal duration of human chorionic gonadotropin surveillance. STUDY DESIGN: We retrospectively reviewed the clinical follow-up of all women who were diagnosed with partial hydatidiform mole at our institution from 1983 to 2003. RESULTS: One hundred sixty-three patients were identified with a median age of 23 years (range, 14-42 years). Seventy-four patients (45%) attained undetectable levels of human chorionic gonadotropin; none of the patients had gestational trophoblastic neoplasia. Forty patients completed the 6 months of recommended follow-up; 6 patients conceived during surveillance, and 28 patients did not return for any further office visits 1 to 5 months after achieving remission. Eighty-three patients (51%) were lost to follow-up before normalization of human chorionic gonadotropin. Six women (4%) had stage I gestational trophoblastic neoplasia during surveillance. CONCLUSION: Our results support the suggestion that a single undetectable human chorionic gonadotropin level after evacuation is sufficient follow-up to ensure remission in patients with partial hydatidiform moles.     

Changes of clinical features in hydatidiform mole: analysis of 113 cases

OBJECTIVE: To investigate the changes of the clinical features of hydatidiform mole. STUDY DESIGN: A total of 113 cases of hydatidiform mole treated in Peking Union Medical College Hospital during 1989-2006 were reviewed retrospectively, and a comparison was made to historic data from 1948-1975 using the chi2 test. RESULTS: The median age was 28 years (range, 20-55). The median gestational age was 90.2 days. Vaginal bleeding remains the most common presenting symptom, occurring in 94 of 113 cases (83.2%). Of 113 cases, 52 (46%) presented with excessive uterine size. Preeclampsia, hyperemesis, hemoptysis and theca lutein cysts occurred in 4 of 113 (3.5%), 12 of 113 (10.6%), 4 of 113 (3.5%) and 19 of 113 cases (16.8%), respectively. The incidence of postmolar trophoblastic neoplasia was 21% (24 of 113). Compared to historic data, the incidence of vaginal bleeding and preeclampsia were statistically lower (p < 0.005). The incidence of postmolar gestational trophoblastic neoplasia was increased moderately without statistical significance compared to historic data. CONCLUSION: Because of the wide use of ultrasonography and serum human chorionic gonadotropin test, current patients with hydatidiform mole have been diagnosed earlier in gestation and the clinical features have changed. Patterns of medical practice should be changed as well.     

Diagnosis and treatment of gestational trophoblastic disease: ACOG Practice Bulletin No. 53

Gestational trophoblastic disease comprises a spectrum of interrelated conditions originating from the placenta. Other terms often used to refer to these conditions include gestational trophoblastic neoplasia and gestational trophoblastic tumor. Histologically distinct disease entities encompassed by this general terminology include complete and partial hydatidiform moles, invasive moles, gestational choriocarcinomas, and placental site trophoblastic tumors. Before the advent of sensitive assays for human chorionic gonadotropin (hCG) and efficacious chemotherapy, the morbidity and mortality from gestational trophoblastic disease were substantial. At present, with sensitive quantitative assays for beta-hCG and current approaches to chemotherapy, most women with malignant gestational trophoblastic disease can be cured and their reproductive function preserved. The purpose of this document is to address current evidence regarding the diagnosis, staging, and management of gestational trophoblastic disease.     

Gestational trophoblastic disease: the significance of vaginal metastases

PURPOSE: To evaluate patients with vaginal lesions in gestational trophoblastic disease and determine prognostic and therapeutic implications applicable to management. METHODS: Twelve patients among 75 cases of gestational trophoblastic neoplasia were analyzed retrospectively between 1990 and 1997. Vaginal metastases were documented by physical examination and biopsy. Two patients received MAC III regimen (5 and 7 courses), 4 patients received EMA-CO regimen for 2 to 11 courses, while 6 were administered methotrexate alone. Remission was defined as 3 weekly beta hCG levels below assay sensitivity (<5 mIU/ml). RESULTS: The mean age of the patients was 25.4 years. While 10 of the patients presented initially with hemorrhage and bloody leukorrhea, the remaining 2 were diagnosed during a routine study of hydatidiform mole. The sites of involvement were almost always the anterior distal vaginal wall. Five cases had additional lung and 1 case had lung, liver, spleen and brain metastases. Three of the patients who received methotrexate as monotherapy did not respond to therapy and were switched to EMA-CO. Overall survival was 91.6%. One patient died in the first month of the initial therapy. CONCLUSION: The presence of large vaginal metastases should be classified as a high-risk factor and these patients must be treated by multiple agent chemotherapy.     

Secondary chemotherapy for high-risk gestational trophoblastic neoplasia

OBJECTIVE: To determine the efficacy of secondary chemotherapy after failure of initial treatment for high-risk gestational trophoblastic neoplasia. METHODS: Twenty-six patients with high-risk gestational trophoblastic neoplasia based on WHO criteria who failed primary treatment or relapsed from remission and received secondary chemotherapy were identified from the records of the Brewer Trophoblastic Disease Center. Initial chemotherapy consisted of etoposide, high-dose methotrexate with folinic acid, actinomycin D, cyclophosphamide and vincristine (EMA-CO) in 10 patients and methotrexate/actinomycin D-based chemotherapy without etoposide in 16 patients. Secondary chemotherapy consisted mainly of platinum-etoposide combinations with methotrexate and actinomycin D (EMA-EP), bleomycin (BEP), or ifosfamide (VIP, ICE). Adjuvant surgery and radiotherapy were used in selected patients. Clinical response and survival as well as factors affecting survival were analyzed retrospectively. RESULTS: The overall survival has 61.5% (16/26). Of the 10 patients who failed primary treatment with EMA-CO, 9 (90%) had complete clinical responses to secondary chemotherapy with EMA-EP (3) or BEP (6), and 6 (60%) were placed into lasting remission. Of the 16 patients who failed primary treatment with methotrexate/actinomycin D-based chemotherapy without etoposide, 10 (63%) had complete clinical responses to BEP (8), VIP (1) and ICE (1), and 10 (63%) achieved long-term remission. Adjuvant surgical procedures were performed on 15 patients as a component of their therapy; eight (73%) of 11 patients who underwent hysterectomy, five (62%) of eight patients who had pulmonary resections, and one patient who had wedge resection of resistant choriocarcinoma from the uterus survived. Survival was significantly influenced by both hCG level at the start of secondary therapy and sites of metastases. CONCLUSION: Patients with persistent or recurrent high-risk gestational trophoblastic neoplasia who develop resistance to methotrexate-containing treatment protocols should be treated with drug combinations employing a platinum agent and etoposide with or without bleomycin or ifosfamide.