Hope it helps.

来源: lao_liu 于 2012-05-23 12:00:18 [] [博客] [旧帖] [给我悄悄话] 阅读数 : (165726 bytes)
回答: 心路 五 由 四十号高速 于 2012-05-23 09:13:07
Malignant gestational trophoblastic disease: Staging and treatment
All topics are updated as new evidence becomes available and our peer review process is complete.
Literature review current through: Apr 2012. | This topic last updated: Feb 15, 2012.

INTRODUCTION — Gestational trophoblastic disease (GTD) defines a group of conditions that arise from an aberrant fertilization event. The resulting proliferative process has the potential to develop into an invasive malignant neoplasm. The spectrum of GTD includes:

 

 

The last four categories comprise the malignant forms of GTD. In general, malignant GTN (also called trophoblastic neoplasia) is one of the most chemotherapy-responsive and highly curable cancers, even in the setting of widespread metastatic disease.

The staging and treatment of the three most common malignant forms of GTD, including persistent/invasive GTN, choriocarcinoma, and PSTT, will be reviewed here. Other aspects of GTD, including molar pregnancy, are discussed separately. (See "Gestational trophoblastic disease: Pathology" and "Gestational trophoblastic disease: Epidemiology, clinical manifestations and diagnosis" and "Gestational trophoblastic disease: Management of hydatidiform mole".)

SPECTRUM OF DISEASE — Malignant GTD is diagnosed when there is clinical, radiologic, pathologic, and/or hormonal evidence of persistent gestational trophoblastic tissue. Most commonly, the diagnosis is made following a molar pregnancy, but a malignant form of GTD can occur after any type of gestation. Rarely, the antecedent pregnancy cannot be determined.

Any of the malignant forms of GTD can metastasize. When metastases occur, the most common site is the lungs, which are involved in over 80 percent of patients. Vaginal metastases are present in up to 30 percent. Hepatic and cerebral metastases are less common, occurring in approximately 10 percent of patients. They are most commonly associated with choriocarcinoma following a nonmolar pregnancy. The majority of patients with hepatic and cerebral metastases have concurrent pulmonary and/or vaginal metastasis.

The frequency of metastatic disease at presentation varies according to the type of malignant GTD. What follows is a brief description of the three types of malignant GTD, focusing on clinical behavior and metastatic potential. The clinical features and diagnosis of these disorders is discussed in detail elsewhere. (See "Gestational trophoblastic disease: Epidemiology, clinical manifestations and diagnosis", section on 'Malignant GTD'.)

 

  • Persistent/invasive gestational trophoblastic neoplasia (GTN) — Approximately 20 percent of patients with a complete molar pregnancy develop GTN [1]. Approximately 15 percent of patients with GTN have disease that is limited to the uterus, while 4 percent have evidence of metastatic spread at the time of diagnosis. By comparison, only 2 to 4 percent of partial moles develop GTN, which is generally nonmetastatic [2].

    Predictors of GTN in women with a complete mole include a pre-evacuation serum level of the beta subunit of human chorionic gonadotropin (beta-hCG) >100,000 mIU/mL, clinical and/or pathologic evidence of marked trophoblastic overgrowth (including excessive uterine size), and maternal age over 35 years [3].
  • Choriocarcinoma — Although most often a sequela of a molar pregnancy, choriocarcinoma may develop after any gestational event. Approximately 50 percent of all cases of gestational choriocarcinoma follow a hydatidiform mole, 25 percent follow a spontaneous abortion or ectopic pregnancy, and 25 percent follow a term pregnancy. GTD that develops after a nonmolar pregnancy is almost always histologically a choriocarcinoma (rarely PSTT, see below).

    Choriocarcinoma is a highly malignant tumor. Most lesions begin in the uterus, although many cases of extrauterine disease in the absence of a primary uterine tumor have been described. At the time of diagnosis, the majority of patients have metastatic disease, most commonly to the lung. Because of this propensity for early dissemination, most women with choriocarcinoma require combination chemotherapy to achieve disease remission.
  • Placental site trophoblastic tumors — PSTTs are rare malignant neoplasms that originate from intermediate cytotrophoblast cells. PSTTs most commonly arise months to years after a term gestation, but can occur after a spontaneous abortion or a molar pregnancy. Patients present with abnormal uterine bleeding, a mass in the endometrial cavity, or amenorrhea. These tumors secrete small amounts of beta-hCG; as a result, a large tumor burden may be present before serum beta-hCG increases [4].

    Anatomic disease extent, as indicated by the FIGO tumor stage, is the strongest predictor of outcome for PSTT (see 'Staging and pretreatment evaluation' below) [5-7]. Other factors associated with a poorer outcome include older age, long interval between antecedent pregnancy and diagnosis of PSTT [8], large tumor burden, high pretreatment levels of beta-hCG, and the histologic features of deep myometrial invasion, high mitotic index, coagulative tumor necrosis, and tumor cells with clear cytoplasm.

    In contrast to choriocarcinoma, the tumor is confined to the uterus in most cases. However, as many as 30 percent of patients with PSTT present with overt metastatic disease [8,9]. The mortality rate approaches 50 percent when metastases are present [10], but prolonged remission is possible with combination chemotherapy, particularly if it is started early (ie, within one week of surgery) [5,11,12].The stage-specific prognosis of PSTT was addressed in a population based series of 62 women diagnosed with PSTT in the United Kingdom between 1976 and 2006 [8]. The probability of 10-year survival for those with stage I, II, or III/IV disease was 90, 52, and 49 percent, respectively.

 

DIAGNOSIS

Following a molar pregnancy — Following evacuation of a complete or partial molar pregnancy, the diagnosis of malignant GTD is usually based upon a rising or stable serum level of beta-hCG. In most such cases, the diagnosis is made clinically rather than histologically by examination of tissue.

The specific diagnostic criteria vary around the world. Most centers in the United States utilize the following [13]:

 

  • Plateauing of beta-hCG levels over at least three weeks
  • A 10 percent or greater rise in beta-hCG for three or more values over at least two weeks
  • Persistence of beta-hCG six months after molar evacuation, or
  • The histologic identification of choriocarcinoma

 

The treatment for either persistent/invasive GTN or choriocarcinoma is generally the same in that both are usually treated with chemotherapy. However, select patients with persistent/invasive GTN may also be candidates for observation. The rationale for observation is discussed below. (See 'Observation' below.)

Following a nonmolar pregnancy — Serum hCG monitoring is not routinely performed after nonmolar pregnancies, unless the woman has had a previous molar pregnancy. Women who develop GTD after a nonmolar pregnancy usually undergo evaluation with serum hCG and ultrasound only after they become symptomatic, which may be months to years after the pregnancy.

Because it may not be clear whether the GTD is a new event (complete or partial mole) or represents the malignant sequelae (choriocarcinoma or PSTT) of an antecedent pregnancy, histologic confirmation of ultrasound findings that are suggestive of malignant GTD is required unless there is evidence of metastatic spread. The presence of overt metastatic disease excludes the possibility of a complete or partial hydatidiform mole, and combination chemotherapy is indicated.

A more detailed discussion of the diagnosis of choriocarcinoma or PSTT after a nonmolar pregnancy is presented elsewhere. (See "Gestational trophoblastic disease: Epidemiology, clinical manifestations and diagnosis", section on 'Malignant disease'.)

Once the diagnosis has been confirmed, the stage of disease must be determined prior to making a decision regarding appropriate therapy.

STAGING AND PRETREATMENT EVALUATION — For patients with cancers other than GTN, outcome stratification and selection of appropriate therapy is typically based upon the anatomic extent of disease, which is typically ascertained by surgical staging and histologic examination of the resected specimen. The extent of disease spread in turn determines the disease "stage", which is then used to select treatment.

Malignant GTD differs from this usual paradigm in a number of ways:

 

  • Prognosis is dependent upon other factors that are not reflected in the anatomic disease extent. These include age, the interval between the gestational event and the persistent disease, and serum beta-hCG concentration.
  • In most cases, the diagnosis is based upon clinical and biochemical parameters; histologic confirmation and surgical staging is frequently not obtained.
  • Particularly in patients with choriocarcinoma, metastatic spread to distant organs, in particular the lungs, can occur early, even in the absence of disease in the uterus or pelvis.

 

Pretreatment evaluation — Patients who have or who are suspected of having persistent GTN, choriocarcinoma, or PSTT must undergo a thorough evaluation prior to institution of therapy. In preparation for treatment, blood tests are obtained to assess renal and hepatic function, peripheral blood counts, and baseline serum hCG levels.

Radiographic evaluation includes pelvic ultrasound, both to look for evidence of retained trophoblastic tissue, and to evaluate the pelvis for local spread. Chest imaging is also required, as the lungs are the most common site of metastatic disease. Chest CT is more sensitive than chest x-ray; pulmonary metastases can be detected by chest CT in up to 40 percent of patients with a negative chest x-ray [14]. Chest CT is not mandatory, however, particularly if detection of occult pulmonary metastases will not alter the treatment plan.

In the absence of pulmonary and vaginal involvement, brain and liver metastases are rare. Asymptomatic patients with normal chest and pelvic imaging do not require further imaging of brain or liver [15]. CT scan or magnetic resonance imaging (MRI) of the brain is recommended in women with persistent disease who have vaginal or lung metastases and in all patients with choriocarcinoma. Cerebral involvement can also be assessed by measuring beta-hCG levels in the cerebrospinal fluid (CSF). However, the sensitivity is not sufficient to replace brain imaging [16].

As long as the clinical picture is compatible with the diagnosis of malignant GTD, metastatic lesions need not be biopsied for confirmation because of the high vascularity of most tumors and the risk for hemorrhage.

Staging and prognostic stratification — The International Federation of Gynecology and Obstetrics (FIGO) initially developed a four-tiered anatomic staging system for GTN based upon disease distribution:

 

  • Stage I — All patients with persistently elevated beta-hCG levels and tumor confined to the uterus.
  • Stage II — The presence of tumor outside of the uterus, but limited to the vagina and/or pelvis.
  • Stage III — Pulmonary metastases with or without uterine, vaginal, or pelvic involvement.
  • Stage IV — All other metastatic sites (eg, brain, liver, kidneys, gastrointestinal tract).

 

All stages were further stratified as A, B or C depending on the presence or absence of one or both of two risk factors for recurrence (beta-hCG ≥100,000 mIU/mL, or the detection of disease more than six months followed termination of an antecedent pregnancy).

Clinical or prognostic staging evolved after it was recognized that a number of clinical and prognostic factors other than anatomic site of disease involvement influence cure rates [17]. In 2002, FIGO approved a revision of the staging system for GTN that was subsequently adopted by the American Joint Committee on Cancer (AJCC) (table 1) [13,18-20]. The basic FIGO stages I to IV were retained to describe the anatomic distribution of disease. The prior A, B, and C risk modifier subgroups were eliminated, and replaced by a modification of the World Health Organization (WHO) scoring system called the Prognostic Scoring Index [21-23].

In addition to its prognostic utility, the revised staging system is also capable of predicting which patients are likely to respond poorly to single-agent chemotherapy. A prognostic score of 7 or higher is considered a high-risk score; all of these patients are more likely to be resistant to single agent therapy and require combination chemotherapy. Patients with scores under 7 are considered low-risk and can usually be managed using single-agent chemotherapy.

The benefit of using the prognostic score to distinguish between clinically high-risk and low-risk disease applies primarily to FIGO stage II and III disease [21,24]. The prognostic score adds little to the selection of treatment for women with FIGO stage I or stage IV disease:

 

  • In general, women with FIGO stage I disease generally have low-risk scores, and >90 percent achieve remission with single-agent chemotherapy [24].
  • The majority of patients with FIGO stage IV disease have high-risk scores and are resistant to single-agent chemotherapy [21,25]. Most have choriocarcinoma following a nonmolar pregnancy; they often have a delayed diagnosis and large tumor burden.

 

The prognostic utility of the scoring index for patients with PSTT is less certain; some authors recommend treatment adapted to disease stage alone [8].

TREATMENT OVERVIEW — The approach to treatment takes into account the stage and risk score as defined above. While many patients diagnosed with malignant GTD will require single-agent or combination chemotherapy, some may be candidates for observation. These options are discussed below. An overview of treatment by stage is also provided below. (See 'Overview of treatment by stage' below.)

Observation — Most patients with persistent/invasive gestational trophoblastic neoplasia (GTN) are diagnosed solely on the basis of an elevated beta-HCG level. Traditionally, these patients are treated with chemotherapy, but one large observational study suggests that a subset of these patients may be safely observed.

For women with persistent/invasive GTN solely on the basis of a persistently elevated hCG, we suggest observation, provided they are willing and able to follow a careful schedule of surveillance with blood tests every two weeks. Patients in whom serial hCG evaluations are not possible, who do not demonstrate a normalization of their hCG within 12 months (provided they remain clinically stable without symptoms of GTN), or whose levels continue to rise should undergo definitive treatment. (See 'Chemotherapy regimens' below.)

This approach was illustrated in a retrospective analysis involving almost 14,000 women treated for a hydatidiform mole between 1993 and 2008 [26]. Following primary treatment, 92 percent had spontaneous normalization of their hCG levels within six months of evacuation and 7 percent required chemotherapy. Fewer than 1 percent (n=76) had persistently high hCG levels. Of these patients, 66 (87 percent) continued under surveillance and the remaining 13 percent underwent chemotherapy with the following outcomes reported:

 

  • hCG normalized in 98 percent of patients undergoing continued surveillance. None of these patients developed GTN requiring chemotherapy.
  • Treatment with chemotherapy (methotrexate and folinic acid) induced an 80 percent remission rate. Two patients never reached a normal hCG despite chemotherapy, but remained asymptomatic with 3 and 15 years of follow-up.
  • There were no deaths in either group with a median follow-up of 2 years (range, 0.6 to 15 years).

 

Chemotherapy regimens — Chemotherapy is a major component of curative therapy for the majority of women diagnosed with malignant GTD. As a group, malignant gestational trophoblastic tumors are exquisitely sensitive to chemotherapy, and they represent one of the only cancers for which single agent therapy is still in use. The choice of single agent versus combination chemotherapy depends on risk category, disease stage, and previous drug treatment.

What follows is a brief description of the various chemotherapy regimens that are used for treatment of malignant GTD, including efficacy and side effects. Treatment recommendations according to disease stage and prognostic risk score are presented below (see 'Overview of treatment by stage' below).

Single-agent chemotherapy — Single agent chemotherapy is widely accepted for low-risk malignant GTD. In the United States, the most commonly used drugs for monotherapy are methotrexate and dactinomycin. Etoposide and 5-fluorouracil (5-FU) are other active single agents.

Methotrexate with and without leucovorin — The dramatic cure of women with metastatic choriocarcinoma using single agent methotrexate (MTX) was first reported in 1956; at present, it remains a mainstay of single-agent treatment for low-risk disease [27].

MTX is commonly given with leucovorin (folinic acid) for "rescue" [28-30]. Leucovorin calcium, an agent that has no inherent antitumor activity, is able to protect against MTX toxicity in normal cells by overcoming the defect in folate metabolism induced by MTX. (See "Therapeutic use of high-dose methotrexate".)

With the doses of MTX commonly used for malignant GTD (50 to 60 mg/m2 per dose), leucovorin rescue is less critical since the use of leucovorin is more important when MTX doses are over 150 mg/m2 [31]. To be maximally effective, leucovorin rescue should be started within 24 hours of MTX dosing; effective rescue is difficult after about 40 hours. One approach is to withhold leucovorin unless measured serum levels of MTX are beyond a threshold value at a certain time after drug administration (eg, 10 micromol/L or higher at 24 hours after infusion of MTX [32,33]).

Several dosing regimens for single agent MTX with or without leucovorin are in clinical use, and there is little evidence to support the superiority of any one over another [34,35]:

 

  • Infusional MTX is administered as a single IV bolus dose of MTX 100 mg/m2 over 30 minutes followed by a 12-hour infusion of MTX 200 mg/m2, with or without leucovorin rescue [32,33,36,37]. This regimen is associated with minimal toxicity (and no alopecia), but it is somewhat inconvenient. At least some data suggests that use of this regimen is associated with a greater likelihood of needing salvage chemotherapy than the eight-day regimen [37].
  • An eight-day regimen of MTX (1 mg/kg IM every other day for four doses) and leucovorin calcium (0.1 mg/kg, given once, 24 hours after each dose) is easy to administer and well tolerated [37]. It is associated with remission rates in excess of 90 percent for patients with stage I GTD and approximately 70 percent of those with low-risk stage II or III disease [35,38-41] (see 'Overview of treatment by stage' below).

    A second course of treatment (either with methotrexate or an alternative active single agent), which is required in 10 to 30 percent of patients, can be administered if the serum level of beta-hCG does not fall by one log within 18 days, or if the value plateaus for more than two weeks before returning to normal. (See 'Monitoring during and after chemotherapy' below.)

    Acute toxicity with this regimen was addressed in a series of 250 patients [40]. There was no alopecia. Nausea was mild in all cases, and complicated fewer than 15 percent of all treatment cycles, vomiting occurred in less than 5 percent. One-fourth of patients developed grade 1 or 2 mucositis, and an additional one-fourth had grade 1 or 2 blepharitis or conjunctivitis. Hematologic toxicity was mild and infrequent: grade 3 or 4 neutropenia or thrombocytopenia complicated fewer than 1 percent of treatment cycles.
  • A regimen of weekly MTX (30 to 50 mg/m2 IM weekly) was developed by the Gynecologic Oncology Group (GOG) [42]. Treatment is repeated until the serum beta-hCG concentration falls to normal (less than 5 mIU/mL) for three consecutive weeks.

    This regimen may be similarly effective but less toxic than the eight-day regimen in patients with nonmetastatic malignant GTD [42-44], although few comparative studies have been conducted (see below). In one series, the complete biochemical response rate was 81 percent after a median of seven weeks (range 3 to 19) [42]. Thirteen of 63 patients developed grade 1 or 2 neutropenia, while three had platelet nadirs below 140,000/microL. There were no other significant toxic effects.

    A randomized GOG study comparing weekly MTX with pulsed biweekly dactinomycin showed that the complete response rate with weekly MTX was particularly low (9 percent; 1 of 11 patients) for women with high-risk disease (WHO score 5-6) or choriocarcinoma [45]. It should be noted, however, that the MTX dose utilized in this study was 30 mg/m2 per week, ie, the lowest weekly regimen used in clinical practice. (See 'Staging and prognostic stratification' above.)

 

Although significant toxicity with MTX is infrequent, hematologic indices should be carefully monitored during therapy. Normal renal and liver function should be demonstrated prior to each treatment, as methotrexate is entirely excreted by the kidney, and it may be hepatotoxic. Furthermore, methotrexate has also been associated with idiopathic interstitial pneumonitis. (See "Chemotherapy hepatotoxicity and dose modification in patients with liver disease", section on 'Antimetabolites' and "Methotrexate-induced lung injury" and "Crystal-induced acute kidney injury (acute renal failure)".)

Dactinomycin — Older retrospective series suggested that single agent dactinomycin (actinomycin D) was as effective but more toxic than MTX monotherapy when used for initial treatment of patients with low-risk GTD. Over time, five-day regimens (12 mcg/kg or 0.5 mg total dose daily for five days, repeated every two weeks [46,47]) have evolved to "pulsed" regimens using single higher doses (typically 1.25 mg/m2) administered every two weeks. These regimens are less toxic and easier to administer; however, whether they are as effective as the five-day regimen remains a point of controversy:

 

  • In one report, the five-day regimen was associated with a 94 percent rate of complete and sustained remission in women with nonmetastatic disease; the corresponding value for those with metastatic disease was 67 percent [48].
  • A similar complete remission rate (94 to 100 percent) for nonmetastatic disease was achieved with dactinomycin 1.25 mg/m2 (or 40 microgram/kg) intravenously every other week in at least two retrospective series [49,50]. However, others report a 20 to 40 percent failure rate of initial therapy with biweekly dactinomycin [45,51,52]. In such cases, successful salvage may be accomplished with the alternative five-day dactinomycin regimen or with single agent MTX [46,47].

    This biweekly regimen was directly compared to single agent MTX (30 mg/m2 weekly) in a trial conducted by the GOG in 240 women with “low-risk” GTN, which included some women with choriocarcinoma [45]. The initial complete response rate was significantly higher with dactinomycin (70 versus 53 percent), an advantage that persisted when the analysis was restricted to women with a WHO score of 0 to 4 and non-choriocarcinoma histology (73 versus 58 percent). The data suggest that pulsed dactinomycin may be more effective than the 30 mg/m2 dose of weekly MTX. However, most women who failed initial treatment received the alternative agent, and in the final analysis, there was only one disease recurrence in each arm. Dactinomycin was associated with significantly more frequent and severe (grade 2) nausea, grade 1 to 2 vomiting, low-grade gastrointestinal tract toxicity, rash and alopecia, and neutropenia.

 

There are no randomized trials comparing dactinomycin to the more commonly used multi-day MTX regimens. At least in the United States, the main use of dactinomycin is for the treatment of MTX-resistant nonmetastatic malignant GTD [39,42,47]. In this setting, the five consecutive day treatment protocol is preferred given the low remission rates with the single higher dose regimen.

Dactinomycin is considered a highly emetogenic agent, and prophylaxis using modern antiemetic regimens is indicated prior to each dose. (See "Prevention and treatment of chemotherapy-induced nausea and vomiting", section on 'Cisplatin'.)

Etoposide — Single agent etoposide (100 mg/m2 daily for five days every 10 days until remission is achieved) is another effective monotherapy regimen, which is used mainly for salvage of MTX failures [53]. Single agent etoposide is also an acceptable option for primary therapy for low-risk disease [35,54]. Because of the risk of secondary malignancy (leukemia), etoposide is less often used as an upfront regimen than either MTX or dactinomycin. However, the risk is low, probably less than 1 percent if the cumulative etoposide dose is <2000 mg/m2. (See "Treatment-related toxicity in men with testicular germ cell tumors", section on 'Leukemia'.)

5-Fluorouracil — 5-fluorouracil (5-FU) has been the preferred single agent for low-risk malignant GTD in China, inducing a complete remission in 93 and 86 percent of patients with stage I or II disease, respectively [55]. Although not standardized, a typical regimen consists of 5-FU 30 mg/kg per day for 10 days, repeated every 28 days [56,57].

Choice of agent — A randomized trial comparing dactinomycin, 1.25 mg/m2 biweekly, to methotrexate, 30 mg/m2 weekly, has demonstrated a superior response rate for dactinomycin over methotrexate, 69 percent versus 53 percent (p=0.015) [58]. However, the clinical impact of this trial is limited due to concern about the low dose of MTX.

Comparative efficacy of chemotherapy regimens was also evaluated in a retrospective review of 247 patients undergoing single agent therapy for low-risk malignant GTD using one of the following four regimens [35]:

 

 

The primary remission rate was significantly higher with etoposide or dactinomycin (90 and 84 percent, respectively) than with either MTX regimen (60 percent with the eight-day regimen and 74 percent with the five-day regimen). However, overall disease-free survival (87 percent) was similar for all regimens, reflecting the success of salvage chemotherapy [35]. Treatment-related toxicity with etoposide and dactinomycin was actually less than that with the MTX regimens.

Summary — Malignant GTD is a chemotherapy-sensitive disease. Single agent chemotherapy alone can achieve remission in over 90 percent of patients with stage I disease, and over 80 percent of women with low-risk stage II and III disease (see 'Overview of treatment by stage' below). The definition of remission is three consecutive normal hCG levels over a 14 to 21 day period. (See 'Monitoring during and after chemotherapy' below.)

The best regimen has not been determined, although some data suggest there is a higher likelihood of requiring further chemotherapy in patients who are treated with the infusional rather than eight day regimen of MTX. Largely based upon the GOG trial discussed above [42], 2004 guidelines from the American College of Obstetricians and Gynecologists (ACOG) recommend weekly MTX as the treatment of choice until beta-hCG values have reached normal levels, followed by an additional one to two weeks of therapy [15]. However, broad consensus on this issue is lacking [57,59]. Others prefer weekly MTX therapy or pulsed dactinomycin.

Multiagent chemotherapy — Combination chemotherapy is used for disease that is refractory to single-agent therapy, for newly diagnosed high-risk malignant GTD (defined as stage IV disease or stage II/III disease with a high prognostic risk score, (table 1)), and for PSTT (see 'Overview of treatment by stage' below).

The most commonly used multiagent chemotherapy regimens and the application of these regimens to various clinical settings are discussed below.

EMA/CO — Despite the absence of randomized trials to prove its superiority, the combination of etoposide, methotrexate, and dactinomycin followed by cyclophosphamide and vincristine (EMA/CO) has become the preferred regimen for initial treatment of high-risk GTD in most countries [60-64]. The components of this regimen are [60]:

 

  • Etoposide — 100 mg/m2 IV over 30 minutes on days 1 and 2
  • Methotrexate — 100 mg/m2 IV push followed by 200 mg/m2 IV over 12 hours on day 1
  • Dactinomycin — 0.5 mg IV bolus on days 1 and 2
  • Leucovorin calcium — 15 mg orally every 12 hours for four doses, starting 24 hours after start of MTX
  • Cyclophosphamide — 600 mg/m2 IV on day 8
  • Vincristine (Oncovin®) — 1.0 mg/m2 IV on day 8

 

The regimen is repeated every two weeks until remission (ie, normalization of beta-hCG and disappearance of all radiographically evident disease), and then generally continued for an additional three cycles (six weeks).

The efficacy of EMA/CO was evaluated in a study of 272 women with high-risk malignant GTD, 121 of whom had received prior chemotherapy (86 with chemotherapy-resistant disease) [62]. The overall complete remission rate was 78 percent, but the five-year overall survival rate was 86 percent, reflecting successful salvage of some nonresponders with cisplatin-based combination therapy with or without bleomycin (see 'Second-line chemotherapy' below).

EMA/CO is a relatively well-tolerated regimen. In a series of 45 patients treated with EMA/CO for high-risk GTD, alopecia was universal, but there was no life-threatening toxicity [63]. Grade 3 to 4 hematologic toxicity was seen in less than 2 percent of chemotherapy cycles, and only one patient required a one-week treatment delay because of prolonged neutropenia. Gastrointestinal symptoms were mild; only one patient developed grade 3 toxicity requiring hospitalization. More than one-half retained their fertility.

However, in addition to these short-term chemotherapy-related risks, this intensive regimen is associated with a small risk of secondary malignancy [62,65]. In the above series of 272 women with high-risk disease, five (1.8 percent) developed a secondary malignancy [62]. There were two acute myeloid leukemias, two cervical cancers, and one gastric adenocarcinoma.

Modifications to the EMA/CO regimen for patients with cerebral metastases are discussed in detail below (see 'Central nervous system metastases' below).

EMA — The EMA regimen uses the same doses as for the EMA/CO regimen, but without the cyclophosphamide and vincristine. Although they have never been compared directly in a randomized trial, retrospective series suggest that treatment with etoposide, methotrexate, and dactinomycin (EMA) may be as effective as EMA/CO; whether it is less toxic is unclear [66,67].

In two series totaling 77 women with high-risk disease, 74 and 75 percent achieved remission with EMA [66,67]. The toxicity profile differed between the studies, however. In one series, alopecia was nearly universal, grade 4 leukopenia or thrombocytopenia occurred in only 5 and 6 percent of patients, respectively, and nausea and/or vomiting were frequent but mild [67].

In contrast, the second report noted a more severe toxicity profile [66]. Grade 2 or worse nausea, emesis, or stomatitis developed in 29, 30 and 37 percent, respectively. Fifty-one percent experienced grade 2 or 3 anemia, 8 percent grade 2 or higher thrombocytopenia and 64 percent had grade 3 or 4 neutropenia; in six cases this was complicated by sepsis.

MAC — Triple therapy with MAC (methotrexate, dactinomycin plus either chlorambucil [original regimen] or cyclophosphamide [modified regimen]) was once the preferred first-line treatment for patients with high-risk malignant GTD [68,69], but this regimen fell out of favor for first-line therapy because of low durable remission rates (in some reports as low as 50 percent) [70,71].

The comparative efficacy of EMA/CO versus MAC (cyclophosphamide) was evaluated in a retrospective review from Korea that compared outcomes among 40 women receiving first-line MAC for high-risk disease and 96 who received EMA/CO (using a weekly rather than every two week regimen) [72]. The remission rate with EMA/CO was 91 percent, and the median number of courses needed to achieve remission was 8.5 (8.5 weeks). In contrast, the remission rate with MAC was 68 percent, and the median number of courses required to achieve remission was 10.7 (32 weeks). Furthermore, there were no treatment-related deaths during EMA/CO compared to seven with MAC.

Summary — Despite the absence of randomized trials to prove its superiority, EMA/CO has emerged as the regimen of choice for initial treatment of high-risk GTD. The regimen is repeated every two weeks until remission (ie, normalization of beta-hCG and disappearance of all radiographically evident disease), and then generally continued for an additional three cycles (six weeks).

Second-line chemotherapy — Additional or salvage chemotherapy is required in approximately 20 to 25 percent of women who are resistant to initial chemotherapy, are intolerant of the upfront regimen, or develop recurrent disease after chemotherapy. Almost all patients with relapsed stage I or low-risk disease, and approximately 60 to 70 percent of those with relapsed high-risk disease, can be successfully salvaged with additional chemotherapy alone [71,73-76]. Patients who fail to enter initial remission have a worse outcome (five-year survival 43 percent in one report [77]) than do those who relapse after an initial response.

Several regimens have activity in this setting.

MAC and EMA — Patients with low-risk disease who are resistant to single-agent therapy are often treated with MAC (methotrexate, dactinomycin plus cyclophosphamide) in order to avoid etoposide and the risk of secondary leukemia (see 'MAC' above and 'Salvage after single-agent therapy' below) [69].

The modified MAC III regimen consists of the following [69]:

 

  • Methotrexate 1 mg/kg per day IM on days 1, 3, 5, and 7
  • Leucovorin calcium 0.1 mg/kg per day IM, IV, or orally on days 2, 4, 6, and 8
  • Actinomycin 12 micrograms/kg per day IV on days 1 through 5
  • Cyclophosphamide 3 mg/kg per day IV on days 1 through 5

 

Cycles are repeated every 21 days.

At our institution, we favor this approach, while others use the EMA regimen in this setting (see 'EMA' above).

In contrast, women who relapse after EMA/CO are usually treated with a platinum-based regimen [75]. Taxanes are often incorporated into third-line treatment.

EMA/EP — The most commonly used regimen for patients who are refractory to EMA, EMA/CO or MAC is EMA/EP. The EMA regimen is different for EMA/EP as compared to EMA/CO in that the etoposide and dactinomycin are administered for one day rather than two [73,74,78]:

 

 

The regimen is repeated every two weeks.

The efficacy of EMA/EP for salvage therapy was shown in a series of 42 women with persistent/invasive gestational trophoblastic neoplasia (GTN) or metastatic placental site trophoblastic tumor (PSTT, n = 8) that was either refractory to or recurred after initial therapy with EMA/CO (see above) [73]. For the women with GTN, a complete remission was obtained in 21 of 22 who were resistant to previous EMA/CO and who had a nearly normal serum beta-hCG, all 12 who were resistant or relapsed after EMA/CO and had elevated serum beta-hCG concentrations, and all three who relapsed after an initial response to EMA/CO. Four of the eight patients with PSTT were alive and in remission after EMA/EP. The cumulative overall survival for all patients was 81 percent.

Treatment-related toxicity with this intensive weekly treatment regimen can be prominent. In this report, as an example, rates of grade 3 or 4 anemia, neutropenia and thrombocytopenia were 21, 68, and 40 percent, respectively [73]. Cumulative myelosuppression prompted treatment delays in 88 percent and dose reductions in 38 percent. However, hematopoietic growth factor support was only applied to 13 patients (31 percent). Ten patients had multiple grade 3 or 4 toxicities.

Low blood cell counts at the beginning of a scheduled subsequent course of chemotherapy do not mandate dose reductions or treatment delays unless the previous cycle was complicated by febrile neutropenia or some other significant clinical event. The use of colony stimulating factors to maintain dose intensity and avoid unnecessary treatment delays is discussed below (see 'Importance of maintaining dose intensity' below).

PVB and PEB — For women resistant to both EMA/CO and EMA/EP, the combination of cisplatin (20 mg/m2 IV daily for five days, every three weeks), vinblastine (0.15 mg/kg IV on days 1 and 2, every three weeks), and bleomycin (30 units IV on day 2, then weekly) (PVB) is effective [79-81]. Although durable remissions have been achieved in up to 60 percent of refractory patients, surgery may be needed to remove sites of resistant disease (see below).

Another commonly used salvage regimen with a similar degree of efficacy is BEP (bleomycin plus etoposide and cisplatin, (table 2)) [74,82].

In general, treatment-related toxicity is less severe with these regimens than with the intensive weekly EMA/EP regimen. Nevertheless, essentially all patients experience myelosuppression (particularly neutropenia), fatigue, hair loss, and at least mild-to-moderate nausea.

As with the EMA-EP regimen, low blood cell counts at the beginning of a scheduled subsequent course of chemotherapy do not mandate dose reductions or treatment delays unless the previous cycle was complicated by febrile neutropenia or some other significant clinical event. The use of colony stimulating factors to maintain dose intensity and avoid unnecessary treatment delays is discussed below (see 'Importance of maintaining dose intensity' below).

Bleomycin-related pulmonary toxicity is discussed elsewhere. (See "Bleomycin-induced lung injury".)

Other regimens — A variety of other regimens can produce durable remissions in the setting of refractory disease, including:

 

 

Importance of maintaining dose intensity — Treatment delay or dose reduction, which is usually prompted by chemotherapy-induced neutropenia, may contribute to tumor resistance and failure of a regimen. Since the goal of therapy is cure of disease, unnecessary treatment delays and dose reductions must be avoided if at all possible. Recombinant hematopoietic growth factors (eg, granulocyte colony-stimulating factor, G-CSF) are an important tool to maintain adequate dose intensity and to prevent unnecessary dose reductions [93].

Guidelines from the American Society of Clinical Oncology suggest primary prophylaxis using CSFs (ie, prophylactic administration during the first cycle) to prevent the development of febrile neutropenia in patients who have a high risk of febrile neutropenia based upon age, medical history, disease characteristics, and the anticipated myelotoxicity of the regimen [93]. Secondary prophylaxis (administration during a subsequent cycle) is recommended for patients who experience a neutropenic complication (including delayed recovery) in which a reduced dose or treatment delay may compromise disease-free or overall survival, or treatment outcome. The goal of hematopoietic growth factor support is to avoid dose reduction and unnecessary treatment delay. (See "Use of granulocyte colony stimulating factors in patients with chemotherapy-induced neutropenia" and "Use of granulocyte colony stimulating factors in patients with chemotherapy-induced neutropenia", section on 'Summary and recommendations'.)

In general, we routinely use G-CSF for secondary and not primary prophylaxis. When it is needed, we suggest the use of G-CSF on days 3 through 6 and 9 through 14 when using EMA/CO or EMA/EP, and on days 6 through 14 when using VIP/ICE, PVB or BEP [94]. Alternatively, a pegylated formulation of G-CSF (pegfilgrastim) is available that has a prolonged half-life. This permits the administration of a single dose rather than daily administration. (See "Use of granulocyte colony stimulating factors in patients with chemotherapy-induced neutropenia".)

OVERVIEW OF TREATMENT BY STAGE — Recommendations for treatment of malignant GTD vary according to stage and risk status (table 1).

Stage I disease — The overall cure rate for patients with nonmetastatic disease is over 90 percent [36,42,95-97]. The selection of primary therapy for stage I GTN is based upon the patient's desire to preserve fertility. If a woman does not wish to preserve fertility, hysterectomy may be performed, with one course of adjuvant single-agent chemotherapy added for treatment of any occult metastases. Either MTX or dactinomycin can be given, although we prefer MTX [15]. The addition of chemotherapy after hysterectomy does not increase the risk of perioperative bleeding or infection [11].

Contemporary outcomes can be illustrated by a study of 482 patients treated at the New England Trophoblastic Disease Center (NETDC) for stage I GTN between June 1965 and December 2001; 441 (91.5 percent) achieved complete remission with single-agent chemotherapy [36]. All 41 remaining patients who were resistant to single-agent treatment attained remission with combination chemotherapy.

The excellent outcomes among women treated for stage I GTD between 1996 and 1998, and reported to FIGO, are outlined in the table (table 3) [98].

Salvage after single-agent therapy — Between 9 and 33 percent of patients treated with single-agent chemotherapy for low-risk malignant GTD will require salvage chemotherapy because of either resistance to the first-line drug or intolerable adverse effects. Generally accepted criteria for the definition of resistance to first-line single-agent chemotherapy include any of the following:

 

  • Failure of the serum beta-hCG concentration fall by one log within 18 days
  • A plateau or increase in hCG for more than two weeks before normal values are obtained on three consecutive weekly specimens [42,47]
  • The detection of new metastases [42,47]

 

For women who are resistant to either single-agent MTX or dactinomycin, single-agent treatment with the other drug is preferred prior to initiating combination chemotherapy. The efficacy of sequential single-agent therapy was demonstrated in a review of 92 patients with low-risk metastatic malignant GTD [96]. Primary remission was achieved with initial single-agent therapy (primarily MTX) in 67 percent; a second sequential single-agent (usually dactinomycin) was given because of drug resistance in 22 patients (22 percent) or drug toxicity in 10 patients (11 percent). All 92 patients were cured, with only one requiring multiagent chemotherapy.

Patients who are resistant to single-agent therapy are often treated with MAC, in order to avoid the long-term risks associated with etoposide (see 'MAC' above). Another option is the EMA regimen (see 'EMA' above). Over 90 percent of these patients attain durable remissions.

Stage I PSTT — The role of postoperative chemotherapy is more controversial in patients with stage I PSTT. In general, response to chemotherapy is more variable than with other malignant forms of GTD [8,10,99-102].

A lack of benefit for chemotherapy in addition to surgery in patients with stage I disease was suggested in a retrospective review of 62 patients with PSTT who were treated in the United Kingdom over a 30-year period [8]. Of the 34 who had stage I disease, treatment consisted of surgery alone in 17 and surgery plus chemotherapy in 16 (one patient had chemotherapy alone after refusing surgery). The chemotherapy regimens were EMA/CO, EMA/EP, or methotrexate plus actinomycin and etoposide. The 10-year overall survival was not significantly higher in the group receiving chemotherapy (93 versus 91 percent). The authors concluded that surgery alone was an acceptable strategy for stage I disease, unless there are risk factors for disease recurrence (eg, >48 months since antecedent pregnancy, vascular or deep myometrial invasion, serosal involvement, high mitotic index, or persistently raised postoperative hCG concentration). (See 'Hysterectomy' below.)

Low-risk stage II and III disease — Patients with low-risk stage II and III GTN (WHO prognostic score less than or equal to 6, (table 1)) are managed with primary single-agent chemotherapy using either MTX or dactinomycin (see 'Single-agent chemotherapy' above). Remission rates of 80 percent or higher can be achieved without incurring the toxicity of multiagent therapy [46,49]. Long-term survival rates are as high as 84 percent, reflecting the success of salvage therapy [22,68,95,97,98,103,104].

Salvage therapy — The criteria for the definition of resistance to first-line single-agent chemotherapy are the same as described above for stage I disease (see 'Salvage after single-agent therapy' above).

Patients who are resistant to single-agent therapy generally receive combination therapy. The MAC regimen, which avoids the use of etoposide (which is associated with a small risk of leukemia, 1 percent), is commonly utilized in this setting. EMA is another treatment option (see 'MAC' above and 'EMA' above).

Surgical reduction of tumor burden by either hysterectomy or local resection (in women desiring to retain fertility) could be considered for women with low-risk metastatic disease who do not respond to single-agent chemotherapy. This approach is usually considered for women who do not desire future fertility and who have disease seen on imaging of the uterus, particularly if there is evidence of invasion into the myometrium. While salvage combination chemotherapy is still required, this approach might lower the number of chemotherapy courses required for cure (see 'Role of surgery' below).

High-risk stage II and III disease — Women with FIGO stage II and III disease and a WHO score of 7 or higher (table 1) are at high risk for chemotherapy resistance and disease recurrence [11,22,23,105,106]. These patients should be managed with primary combination chemotherapy, typically EMA/CO (see 'EMA/CO' above). In a series of 272 women with high-risk malignant GTD who were treated with EMA/CO, 78 percent entered complete biochemical remission, and only 17 percent developed resistance to EMA/CO.

Salvage therapy — Additional chemotherapy will be required in approximately 20 to 25 percent of high-risk patients. Patients who are resistant to EMA/CO are most often treated with EMA/EP (see 'Second-line chemotherapy' above). In one study, among women who developed resistance to EMA/CO chemotherapy, successful salvage with a cisplatin-based combination regimen with or without surgery was possible in 70 percent [62] (see 'Role of surgery' below). As a result of successful salvage therapy, overall survival rates are 90 percent or greater in contemporary series [60].

Stage II/III PSTT — Patients with stage II or III PSTT are usually managed by combined surgery and combination chemotherapy [8].

Stage IV disease — Patients with stage IV disease are a high-risk group with a propensity for rapidly progressive disease and chemoresistance. The use of primary combination chemotherapy in conjunction with selective use of radiation and surgical treatment has resulted in significantly improved remission rates in patients with stage IV disease. Before 1975, only 30 percent of patients with stage IV disease achieved remission. After 1975, when the concept of early intensive multiagent treatment was introduced, a dramatic improvement was observed in the survival of patients with stage IV disease.

At present, about 80 percent of patients with stage IV disease achieve an initial disease remission, and 45 to 60 percent maintain a durable remission (table 3) [75,98,107-109]. Therefore, these patients should all be treated with curative intent.

Combination therapy with EMA/CO is recommended as initial therapy for patients with stage IV disease (see 'EMA/CO' above). Similar to stage II and III high risk patients, resistant disease is treated primarily with EMA/EP (see 'Second-line chemotherapy' above).

Specific considerations in the treatment of selected metastatic sites (brain, liver, lung and vaginal metastases) are addressed in detail below (see 'Treatment of selected metastatic sites' below).

ROLE OF SURGERY — The role of surgery in patients with all stages of malignant GTD is to excise sites of bulky and/or resistant tumor, and less frequently, to treat complications of the disease such as hemorrhage or bowel obstruction [110].

Uterine evacuation — The utility of repeat suction curettage in patients with persistent postmolar GTN is controversial and remains an area of debate. A second curettage may have a debulking effect and possibly lower the number of chemotherapy courses required for cure [111].

There is increasing interest in the potential for surgery alone as curative treatment for stage I low-risk disease [105]. There is some evidence that patients with low-risk stage I disease who undergo a second curettage may not require any chemotherapy [111,112]. However, interpretation of the published literature is difficult because of the varying definitions of persistent disease [111,112]. In our view, repeat curettage is associated with risks of uterine perforation, hemorrhage, infection, intrauterine adhesions, and anesthetic complications, and should only be performed if there is evidence of retained tissue in the uterus.

In the case of PSTT, ultrasound imaging may permit differentiation between two types of disease, hypervascular and hypovascular [10,113]. This is clinically useful because massive bleeding following dilation and curettage has been reported with the hypervascular type. Thus, dilation and curettage should be avoided when there is prominent vascularity within the tumor.

Hysterectomy — For women with choriocarcinoma who do not desire future fertility, hysterectomy is usually performed before chemotherapy. Hysterectomy prevents the persistence of drug-resistant local disease, and can shorten the duration and amount of chemotherapy required to produce remission [95,114].

In contrast to choriocarcinoma, hysterectomy is the primary therapy for stage I or II PSTT because the tumor is usually limited to the uterus, and the response to chemotherapy is more variable than with other malignant forms of GTD [8,10,99-102]. For women at high risk of recurrent disease, adjuvant chemotherapy (generally combination chemotherapy rather than single agent therapy) is warranted. In addition to the presence of overt metastatic disease, other features that suggest high risk include duration greater than two years from an antecedent pregnancy and a high mitotic index (over 5 mitoses per 10 high power fields) [5].

Chemotherapy-resistant disease — Hysterectomy should also be considered in the subset of women who have chemotherapy-resistant disease, particularly if the histology is PSTT [8,62,74,110,115-120]. In such cases, it can contribute to a potential cure: long-term survival has been reported in 70 to 97 percent of cases in small series.

In the case of PSTT, additional postoperative chemotherapy should be considered if there is residual tumor in the surgical specimen [8]. In this series, 61 percent of patients with PSTT who were treated with primary chemotherapy showed no or an incomplete response to chemotherapy [8]. Long-term disease control was achieved in 12 of 13 patients with only an incomplete response to initial chemotherapy, eight after surgery alone and four after surgery followed by more chemotherapy.

Hysterectomy may also be necessary to control uterine bleeding or ongoing sepsis due to infection of necrotic tumor.

Local excision — For women who desire to retain fertility, more conservative surgical therapy may be considered, even in patients with a localized PSTT. Successful treatment of localized disease by hysterotomy, local tumor excision, and uterine reconstruction has been described [121]. Disease localization via ultrasound, MRI, and/or arteriography is often helpful for planning of local resection.

Presurgical evaluation — There is no proven benefit to performing a hysterectomy if the uterus has no radiographically visualized lesions. Thus, radiographic imaging, with magnetic resonance imaging (MRI) and/or ultrasonography, should be performed in the prehysterectomy evaluation of women being considered for this procedure [122,123].

However, serum beta-hCG concentration usually has to exceed 700 mIU/mL to detect myometrial invasion by MRI; thus, false negative results may be obtained in this setting [122]. There is also a documented lag time between the fall in serum beta-hCG (tumor regression) and complete radiologic absence of disease. As a result, false positive results can be obtained if imaging is performed too early.

Surgical excision of metastases — Surgery also has a role in the management of metastatic disease sites (see 'Treatment of selected metastatic sites' below).

TREATMENT OF SELECTED METASTATIC SITES

Central nervous system metastases — Cerebral metastases are uncommon overall, but they occur in 9 to 21 percent of patients with metastatic GTN [124,125] Brain metastases are associated with a poorer prognosis than pulmonary or vaginal metastases. In addition, patients who develop brain metastases while on treatment or relapse in the brain after an initial complete remission have a worse prognosis than do those who present with brain metastases before treatment [124,126,127] Nevertheless, some of these patients can be cured.

There are two main therapeutic options for these patients. In the past, institution of whole brain radiotherapy (RT, 20 to 30 Gy in 2 Gy daily fractions) was recommended concurrent with the initiation of chemotherapy [108,109,124,128,129]. In addition to shrinking the brain metastases, concomitant cranial irradiation increases the methotrexate concentration within the CNS [130], reduces the risk of cerebral hemorrhage prior to eradication of tumor, and may improve survival [126,128]. However, the use of concurrent methotrexate and cranial irradiation also increases the likelihood of treatment-related toxicity, especially leukoencephalopathy. (See "Complications of cranial irradiation".)

An alternative is high-dose EMA/CO with or without intrathecal MTX [16,75,108,126,131,132]. In one study, 30 of 35 patients experienced sustained remission with intensive combination chemotherapy alone that included high-dose intravenous and intrathecal MTX [131].

A typical regimen is as follows [132]:

 

  • Etoposide — 100 mg/m2 IV over 60 minutes on days 1 and 2
  • Methotrexate — 1000 mg/m2 IV over 24 hours on day 1
  • Dactinomycin — 0.5 mg IV bolus on days 1 and 2
  • Leucovorin calcium — 30 mg IM or orally every 12 hours for three days, starting 32 hours after
  • Cyclophosphamide — 600 mg/m2 IV on day 8
  • Vincristine — 1.0 mg/m2 IV on day 8
  • Intrathecal MTX — 12.5 mg total dose on day 8 followed by oral leucovorin calcium 15 mg at 24 and 36 hours. The regimen is repeated every two weeks.

 

The higher dose of parenteral MTX allows for adequate levels of MTX within the cerebrospinal fluid (CSF) [75,108,131,132]. The benefit of this approach was shown in a series in which 13 of 15 patients treated with high-dose EMA/CO plus intrathecal MTX achieved a sustained remission without the use of whole brain irradiation [132].

The need for intrathecal therapy in these patients is controversial [16,75,107]. Although treatment protocols in some institutions omit cranial RT and instead administer intrathecal MTX [16,126,133], at our institution, high-dose parenteral MTX is preferred over IT MTX and cranial RT.

Craniotomy — Craniotomy is reserved for life-threatening complications (eg, hemorrhage and the need for acute decompression). The incidence of intracranial hemorrhage from metastatic malignant GTD is appreciably higher than with other primary and secondary intracranial neoplasms.

Craniotomy and resection of drug-resistant lesions is only rarely required and justified only for patients who do not have metastatic disease elsewhere. In such case, surgery may provide an opportunity for complete remission, even in patients with chemotherapy-resistant brain metastases [15,124].

Prognosis — More than 50 percent of women with isolated brain metastases may be cured using either treatment approach [16]. As a result, concerns for long-term toxicity should be a prime consideration in choosing therapy. The prognosis is poorer if there are coexisting hepatic metastases (see below).

Hepatic metastases — Hepatic metastases present a particularly difficult and challenging problem. Because of the hypervascular nature of these tumors (particularly choriocarcinomas), biopsy may prompt life-threatening hemorrhage.

These women also typically have a poor prognosis, even with modern intensive chemotherapy. In one series, the five-year survival rates for women with hepatic metastases alone or in combination with brain metastases were 27 and 10 percent, respectively [134].

Combination chemotherapy can induce a partial response in most women [135,136]. Hepatic resection and/or selective embolization or occlusion of the hepatic arteries may be required in selected cases to control bleeding or excise resistant tumor [137-139].

Pulmonary metastases — An isolated pulmonary nodule that is resistant to chemotherapy may be treated by thoracotomy with wedge resection [97,140-142]. Criteria that predict a favorable outcome from surgical resection include [117,140,142,143]:

 

  • Absence of other systemic metastases
  • Unilateral solitary lung nodule
  • No uterine involvement
  • Urine beta-hCG concentration less than 1000 mIU/mL [140], serum beta-hCG concentration less than 1500 mIU/mL [143]

 

Thus, prior to proceeding with surgical management, other sites of persistent disease must be ruled out. Often, nonviable fibrotic nodules, which may persist indefinitely after tumor regression [144], can be distinguished from viable tumor by PET scan before surgery is contemplated [145].

Chemotherapy is often administered perioperatively, either prior to or after the operation to eradicate any occult metastases and to reduce tumor dissemination.

Vaginal metastases — Vaginal metastases are usually located on the anterior wall and are friable, vascular, and prone to hemorrhage. They have been reported in 4 to 9 percent of malignant GTD [146,147]. One-third of patients experience significant bleeding [146,147].

Bleeding from vaginal metastases can be controlled by packing, followed by a wide local excision if necessary. An alternative approach is arteriographic embolization of the vaginal branch of the hypogastric artery [148].

Surgical management of metastases in other intraabdominal organs — If chemotherapy has been ineffective in controlling metastatic disease in other organs in the abdomen or pelvis, surgical resection may be considered in selected cases [17,149,150]. In a series of 33 patients undergoing various surgical salvage procedures for chemorefractory GTN, successful surgical salvage was most likely in those who had one preoperative disease site, underwent salvage surgery within one year of initial diagnosis, non-choriocarcinoma histology, and a total WHO prognostic scoring index of <8 [149]. (See 'Staging and prognostic stratification' above.) In another report of 61 patients with chemoresistant GTN undergoing surgical salvage, the clinical factors that predicted failure of salvage surgery were age over 35, hCG value >10 IU/mL, an antecedent nonmolar pregnancy, and metastases outside of the lungs [150].

MONITORING DURING AND AFTER CHEMOTHERAPY — All women with GTD should be monitored with serial measurements of serum beta-hCG, beginning at weekly intervals during therapy. The approximate biologic half-life of hCG is 1.5 to 3 days, and serum levels should fall exponentially (by at least one log within 18 days). A slower rate of decline suggests the possibility of chemoresistance, although there is no consensus or clear guideline as to the optimal cutoff for determining chemoresistance or the management of patients with a slower than expected tumor marker decline [37,41,151,152]. The French Trophoblastic Disease Reference Center in Lyon defines chemotherapy resistance only as an increase or a plateau in two consecutive hCG values over a two-week interval [153]. As described above, other generally accepted criteria include failure of the serum hCG value to fall by one log within 18 days, and detection of new metastases [42,47].

A disease remission requires three consecutive normal hCG values (less than 5 mIU/mL) over a period of 14 to 21 days. When weekly or biweekly chemotherapy regimens are used, the regimen is continued until the third negative result is obtained. When infusional MTX is given as a one-time dose, hCG levels are then followed weekly, with additional therapy given only if an inadequate response is observed (see 'Single-agent chemotherapy' above).

After remission is achieved, serum beta-hCG should be measured monthly until monitoring has shown one year of normal hCG levels [15]. Some centers continue biannual titers indefinitely for high-risk individuals, although 85 to 95 percent of recurrences, which are detected as new elevations in serum beta-hCG, occur within the first 18 months [100,117]. Subgroups that might fit into this high-risk category include women with extreme treatment resistance who required multiple regimens of combination therapy, those with advanced stage choriocarcinoma, particularly with chemoresistance, and patients who have late recurrences.

The hCG assay used for monitoring should detect hyperglycosylated hCG, nicked hCG, and nicked hCG missing the C-terminal extension on beta-hCG, as these are the major (and sometimes the only) sources of hCG immunoreactivity [154,155]. Failure to measure these molecules may result in a false negative test. (See "Gestational trophoblastic disease: Management of hydatidiform mole", section on 'Human chorionic gonadotropin'.)

Recurrent disease — After attaining a normal hCG level, the risk of tumor relapse in women with persistent GTN is 3 to 9 percent [86]. The median time from remission to relapse is about six months [21,156]. Many patients with recurrent disease can be cured, and such cases should always be approached with curative intent. The optimal therapy for recurrent disease varies with the initial regimen and responsiveness (see 'Overview of treatment by stage' above).

Contraception — It is essential that women use contraception both during and for the entire duration of beta-hCG follow-up [15]. Estrogen-progestin contraceptives are preferred because of their low failure rate and relatively low incidence of irregular bleeding, which may confuse the clinical picture. They are both safe and effective in women with GTD [157]. Intrauterine contraceptive devices are not recommended because of the risk of uterine perforation. (See "Overview of the use of estrogen-progestin contraceptives".)

FUTURE PREGNANCY — Pregnancy should be avoided for at least one year following treatment for gestational trophoblastic neoplasia (GTN). As noted above, recurrences are most common within the first year, and the diagnosis of relapse is likely to be delayed in the presence of pregnancy.

Pregnancies that occur within six months of completing chemotherapy may be at increased risk of miscarriage [158], and possibly stillbirths, and repeat moles [159].

Beyond six months, chemotherapy does not generally impact future fertility, the rate of spontaneous abortion or congenital malformations with future pregnancies, or child development following a subsequent term pregnancy [158,160-164]. In a report of 445 long-term survivors after chemotherapy for GTN, 97 percent of women who desired fertility after receiving MTX conceived, and 86 percent had at least one live birth [164]. Women who got three or more drugs were less likely to have a live birth than those who received one or two drugs.

Another retrospective cohort study noted no difference in adverse maternal outcomes (spontaneous abortion, repeated mole, spontaneous preterm labor, preeclampsia, placental disease) or in perinatal outcomes (stillbirth, fetal malformation, preterm birth) among women who conceived after receiving combination versus single agent chemotherapy [158]. (See "Overview of fertility and pregnancy in cancer survivors".)

Persistent disease is more common in patients who have had repeated molar pregnancies (incidence 11 percent after one and 29 percent after two complete or partial moles) [165]. A repeat molar pregnancy is not, however, necessarily an indication for chemotherapy. In one report, chemotherapy was only necessary in 6 percent of second molar pregnancies [166].

SUMMARY AND RECOMMENDATIONS — Malignant gestational trophoblastic disease (GTD) is diagnosed when there is clinical, radiologic, pathologic, and/or hormonal evidence of persistent gestational trophoblastic tissue after a gestational event. The spectrum of malignant GTD includes (see 'Spectrum of disease' above):

 

  • Persistent/invasive gestational trophoblastic neoplasia (GTN)
  • Choriocarcinoma
  • Placental site trophoblastic tumor (PSTT)

 

Diagnosis and pretreatment evaluation

The diagnosis of persistent GTN is typically made in women who have persistently elevated serum levels of beta-human chorionic gonadotropin (beta-hCG) following a molar pregnancy. Choriocarcinoma, a highly malignant tumor with the propensity for early dissemination, may develop after any gestational event. The rare PSTT most commonly arises months to years after a term gestation, but can occur after a spontaneous abortion or a molar pregnancy (see 'Diagnosis' above).

In addition to serum levels of beta-hCG, pretreatment evaluation includes pelvic ultrasound and chest imaging. CT or MRI of the brain is recommended in women who have symptoms suggestive of brain metastases, those with vaginal or lung metastases, and in all patients with choriocarcinoma. Prognosis and treatment selection depends on the anatomic extent of disease as well as on clinical factors that are included in the prognostic scoring index (table 1) (see 'Staging and pretreatment evaluation' above).

Treatment — The following represents our general approach to therapy.

Observation

For women with persistent/invasive gestational trophoblastic neoplasia on the basis of a persistently elevated hCG, we suggest observation after surgery rather than chemotherapy, provided they are willing and able to follow a careful schedule of surveillance with blood tests every two weeks (Grade 2C). Patients in whom serial hCG evaluations are not possible, who do not demonstrate a normalization of their hCG within 12 months, or whose levels continue to rise should undergo definitive treatment as discussed above. (See 'Observation' above.)

Role of uterine surgery

For women with choriocarcinoma who do not desire future fertility, we suggest hysterectomy in addition to chemotherapy (Grade 2C). Hysterectomy prevents the persistence of drug-resistant local disease and can shorten the duration and amount of chemotherapy required to produce remission. (See 'Hysterectomy' above.)

For this same reason, we also discuss the option of hysterectomy with women who have persistent/invasive GTN that is resistant to single agent chemotherapy.

For women with choriocarcinoma who wish to preserve future fertility, we administer chemotherapy alone. We suggest local resection after chemotherapy only for those women who have evidence of persistent uterine disease after single agent chemotherapy, especially if there is evidence of myometrial invasion on pelvic ultrasound (Grade 2C). Although these women also receive salvage combination chemotherapy, this approach can shorten the duration and amount of chemotherapy required to produce remission. (See 'Local excision' above.)

For women with stage I and II PSTT, we recommend hysterectomy as the treatment of choice (Grade 1B). PSTT is usually limited to the uterus, and the response to chemotherapy is more variable than with choriocarcinoma. (See 'Hysterectomy' above.)

Hysterectomy may also be indicated in the subset of women who have chemotherapy-resistant disease, particularly if the histology is PSTT. (See 'Chemotherapy-resistant disease' above.)

Chemotherapy — As a group, malignant gestational trophoblastic tumors are exquisitely sensitive to chemotherapy, and they represent one of the only cancers for which single-agent therapy is still in use. The exception to this general rule is PSTT, which is less chemotherapy responsive than other types of GTN.

Assessment of response — All women with GTD should be monitored with weekly serial measurements of serum beta-hCG during therapy. (See 'Monitoring during and after chemotherapy' above.)

 

  • Remission is defined as three consecutive normal hCG values (less than 5 milli-international units/mL) over 14 to 21 days.
  • When weekly or biweekly single agent chemotherapy regimens are used, the regimen is continued until the third negative result is obtained. When infusional MTX is given as a one-time dose, hCG levels are followed weekly, with additional therapy given only if an inadequate response is observed.
  • When combination chemotherapy with EMA/CO is used, the regimen is repeated every two weeks until normalization of beta-hCG, and then generally continued for an additional three cycles (six weeks).

    After remission is achieved, serum beta-hCG should be measured monthly until monitoring has shown one year of normal hCG levels.

 

Management by stage

Stage I

For women with stage I malignant GTD except those with PSTT, we suggest one course of single agent chemotherapy (Grade 2B). (See 'Stage I disease' above.) We prefer methotrexate rather than dactinomycin. No clearly superior methotrexate regimen has emerged, although some data suggest there is a higher likelihood of requiring further chemotherapy in women who are treated with the infusional rather than the eight-day regimen. (See 'Single-agent chemotherapy' above.)

If markers remain elevated, or if they do not fall by one log over 18 days, or if they plateau for more than two weeks before returning to normal, we suggest a second course of single-agent therapy rather than combination chemotherapy (Grade 2C). We administer a course of dactinomycin if methotrexate was initially used, and methotrexate if dactinomycin was initially given. (See 'Salvage after single-agent therapy' above.)

For patients who are resistant to single-agent therapy, combination chemotherapy is indicated. We suggest MAC rather than EMA/CO to limit the potential for etoposide-related leukemogenesis (Grade 2C). (See 'Second-line chemotherapy' above.)

For women with stage I PSTT, hysterectomy is the treatment of choice (see above). We recommend adjuvant chemotherapy after hysterectomy for women who are at high risk for disease recurrence (>48 months since antecedent pregnancy, vascular or deep myometrial invasion, serosal involvement, high mitotic index, or persistently raised postoperative hCG concentration) (Grade 1B). In such cases, we recommend initial combination rather than single-agent chemotherapy (Grade 1B). (See 'Stage I PSTT' above.)

We suggest EMA/CO rather than MAC in this setting (Grade 2C). (See 'EMA/CO' above.)

For women who are resistant to EMA/CO, we recommend a platinum-based regimen such as EMA/EP (Grade 1B). (See 'EMA/EP' above.)

Stage II and III, WHO score ≤6

For patients with low-risk FIGO stage II and III GTN (WHO prognostic score less than or equal to 6, (table 1)), we recommend one course of single-agent chemotherapy (Grade 1B). (See 'Low-risk stage II and III disease' above.) Both methotrexate and dactinomycin can be considered for upfront treatment; at our institution, we prefer infusional methotrexate in conjunction with leucovorin rather than dactinomycin. (See 'Single-agent chemotherapy' above.)

For women who are resistant to first-line single-agent therapy, we suggest combination chemotherapy rather than a second trial of an alternative single agent (Grade 2C). We suggest MAC rather than EMA/CO to limit the potential for etoposide-related leukemogenesis (Grade 2C). (See 'Second-line chemotherapy' above.)

For women who are resistant to EMA/CO, we recommend a platinum-based regimen such as EMA/EP (Grade 1B). (See 'EMA/EP' above.)

Stage II and III, WHO score >7

For women with high-risk FIGO stage II and III disease (WHO score of 7 or higher) we recommend initial combination rather than single agent chemotherapy (Grade 1B). We suggest EMA/CO rather than MAC in this setting (Grade 2C). (See 'EMA/CO' above.)

For women who are resistant to EMA/CO, we recommend a platinum-based regimen (Grade 1B). At our institution, we use EMA/EP. (See 'EMA/EP' above.)

Stage IV

For women with overt distant metastases, we recommend initial combination chemotherapy rather than single agent chemotherapy (Grade 1B). We suggest EMA/CO rather than MAC in this setting (Grade 2C). (See 'EMA/CO' above.)

As with stage II and III GTD, resistant disease is treated with a platinum-based regimen. (See 'EMA/EP' above.)

 

  • Central nervous system disease

    For women who present with cerebral metastases, we suggest high-dose EMA/CO (with 1000 mg/m2 of MTX IV) rather than cranial RT or intrathecal MTX (Grade 2C). (See 'Central nervous system metastases' above.)

    We reserve craniotomy for life-threatening complications and for drug-resistant lesions in patients who have no extracranial disease. (See 'Craniotomy' above.)

 

 

  • Isolated liver metastases

    Biopsy of hepatic metastases should be avoided if at all possible. We reserve hepatic resection for control of bleeding or for excision of resistant tumor. (See 'Hepatic metastases' above.)

 

 

  • Isolated lung metastases

    We restrict thoracotomy and wedge resection to women with chemotherapy-resistant isolated pulmonary metastases who fulfill all of the following criteria (see 'Pulmonary metastases' above):

 

 

  • Absence of other systemic metastases
  • Disease limited to one lung
  • No uterine involvement

 

所有跟帖： 

• I'm glad I can attach a such long article. -lao_liu- ♂ 给 lao_liu 发送悄悄话 lao_liu 的博客首页 (306 bytes) () 05/23/2012 postreply 12:08:15

请您先登陆，再发跟帖！