Initial management of high-risk gestational trophoblastic neoplasia

INTRODUCTION — Gestational trophoblastic disease (GTD) defines a group of conditions that arise from an aberrant fertilization event. When GTD recurs following surgical management, or when there is evidence of metastatic disease, it is called gestational trophoblastic neoplasia (GTN) and comprises four subtypes of disease:

●Invasive mole

●Choriocarcinoma

●Placental site trophoblastic tumor

●Epithelioid trophoblastic tumor

The initial treatment of high-risk GTN is discussed here. The pathology, epidemiology, clinical manifestations, and staging of GTD are discussed separately. In addition, the management of low-risk and recurrent GTN are also discussed separately.

●(See "Hydatidiform mole: Epidemiology, clinical features, and diagnosis".)

●(See "Gestational trophoblastic disease: Pathology".)

●(See "Hydatidiform mole: Treatment and follow-up".)

●(See "Initial management of low-risk gestational trophoblastic neoplasia".)

DEFINITION OF HIGH-RISK DISEASE — High-risk invasive mole and choriocarcinoma are characterized by the International Federation of Gynecology and Obstetrics (FIGO) stage and the World Health Organization (WHO) Prognostic Scoring System (table 1) (see "Gestational trophoblastic neoplasia: Epidemiology, clinical features, diagnosis, staging, and risk stratification", section on 'Staging and risk assessment'):

●Stage IV disease

●Stages II and III with risk score >6

While FIGO staging is also used for patients with placental site trophoblastic tumor (PSTT) or epithelioid trophoblastic tumor (ETT), the WHO risk scoring system is not used for such patients. Management of PSTT and ETT is also different than invasive mole and choriocarcinoma and is described below. (See 'Metastatic PSTT or ETT' below.)

APPROACH TO TREATMENT — GTN is uniquely sensitive to chemotherapy, which is the major treatment modality for patients with high-risk disease. The exception to this is women with placental site trophoblastic tumor (PSTT) or epithelioid trophoblastic tumor (ETT), which are relatively resistant to chemotherapy as compared with choriocarcinoma and invasive mole. As such, the primary treatment may be a combination of surgery and chemotherapy. Regardless, in the treatment of high-risk GTN, other modalities such as surgery and radiation therapy (RT) may be indicated in addition to chemotherapy.

For patients in whom combination chemotherapy is indicated, treatment may result in ovarian insufficiency. Therefore, we advocate for the use of oral contraceptives or gonadotropin-releasing hormone agonists to suppress the pituitary glands' production of luteinizing hormone, which aims to protect the ovaries from the toxicity of chemotherapy and also suppress the pituitary production of human chorionic gonadotropin (hCG), which in our clinical experience, could falsely suggest the presence of active disease if not suppressed [1].

Chemotherapy — Patients with high-risk GTN are likely to develop drug resistance if single-agent therapy is administered. Therefore, these patients are commonly treated with multi-agent regimens. Evidence of the greater likelihood of resistance to single-agent chemotherapy comes from early work at the National Cancer Institute where only 36 percent of patients with high-risk GTN achieved remission with single-agent treatment [2].

Our preferred regimen for these patients is etoposide, methotrexate (MTX), plus actinomycin D (ActD) alternating with cyclophosphamide and vincristine (EMA-CO) because it results in complete response rates between 71 and 78 percent and long-term survival rates of 85 to 94 percent [3-12]. However, a 2012 Cochrane review has found that regimens that incorporate etoposide and cisplatin are effective options, though the lack of randomized trials prevented an analysis to define the optimal regimen [13].

EMA-CO — MTX, plus ActD, alternating with cyclophosphamide and vincristine (EMA-CO) has emerged as the regimen of choice for initial treatment of high-risk GTN [14]. This is predominantly based on retrospective data that consistently show it is active in high-risk GTN and is associated with a low toxicity profile (table 2) [13,15].

The components of this regimen are [11]:

●Etoposide – 100 mg/m² IV over 30 minutes on days 1 and 2

●MTX – 100 mg/m² IV bolus followed by 200 mg/m² IV over 12 hours on day 1

●ActD – 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 – 1.0 mg/m² IV on day 8

Although EMA-CO is the most widely used regimen, there have been no randomized trials to demonstrate that it should be the preferred regimen. However, compared with other combination regimens, it appears to be as effective (if not more so) and better tolerated.

This was shown in one report that included over 200 women with high-risk GTN treated with a variety of multi-agent regimens [15]:

●MTX plus ActD and folinic acid (MA, administered between 1971 and 1995)

●MTX, ActD, cyclophosphamide, doxorubicin, melphalan, hydroxyurea, and vincristine (CHAMOCA, administered between 1982 and 1995)

●MTX, ActD, and chlorambucil (MAC, administered between 1971 and 1982)

●EMA-CO (administered between 1985 and 1995)

Of these four regimens, EMA-CO resulted in [15]:

●The highest remission rate (91 percent) compared with MA, CHAMOCA, and MAC (63, 76, and 68 percent, respectively) and required the fewest number of courses to attain remission

●The lowest mortality rate (9 percent versus 37, 22, and 33 percent, respectively)

In a 2012 systematic review, the administration of EMA-CO was associated with primary remission rates ranging from 54 to 91 percent of patients [13]. These data support the use of EMA-CO as a primary treatment for high-risk GTN.

Administration — Treatment with EMA-CO should be administered every two to three weeks. Although a treatment delay or dose reduction may be required due to side effects, these should be avoided as both have been associated with less than optimal outcomes. We routinely employ growth factor support to help minimize hematologic toxicity and to help avoid delays [16].

Treatment should be continued until the hCG level becomes undetectable and remains undetectable for three consecutive weeks. Finally, we administer at least three courses of EMA-CO after the patient achieves undetectable hCG levels as consolidation therapy to reduce the risk of relapse, which is supported by the limited data. (See "Initial management of low-risk gestational trophoblastic neoplasia", section on 'Remission and consolidation treatment'.)

The Charing Cross Group has reported the use of induction low-dose etoposide (100 mg/m²) and cisplatin (20 mg/m²) on days 1 and 2 every seven days in selected patients with high tumor burden. This has almost completely eliminated early mortality from respiratory compromise and hemorrhage. They also report a 94 percent remission rate with EMA-CO by carefully excluding non-gestational tumors using genetic analysis [17].

Patients with brain metastases — For patients with brain metastases, a neurosurgical consult should be obtained prior to treatment. These patients are at risk for complications directly related to their brain metastases or as a consequence of treatment, which may require urgent or emergent treatment (eg, craniotomy for intracerebral bleeding). For these patients, we administer a modification of systemic EMA-CO that uses a higher MTX dose (1000 mg/m² over 24 hours) than what is routinely administered [18]. The higher dose of parenteral MTX allows for adequate levels of MTX within the cerebrospinal fluid (CSF) [18-21]. In addition, these patients should receive dexamethasone to decrease cerebral edema; however, prophylactic anti-epileptic drugs are generally not recommended in most patients, provided there is no history of an antecedent seizure. (See "Seizures in patients with primary and metastatic brain tumors".)

A typical regimen is as follows [21]:

●Etoposide – 100 mg/m² IV over 60 minutes on days 1 and 2

●MTX – 1000 mg/m² IV over 24 hours on day 1

●ActD – 0.5 mg IV bolus on days 1 and 2

●Leucovorin calcium – 30 mg intramuscular (IM) or orally every 12 hours for three days, starting 32 hours after treatment with MTX

●Cyclophosphamide – 600 mg/m² IV on day 8

●Vincristine – 1.0 mg/m² IV on day 8

At these high doses of MTX, significant renal toxicity can occur, typically from crystallization of the MTX in the renal tubules. Renal dysfunction can lead to decreased MTX clearance, which leads to more significant nonrenal toxicities (eg, myelosuppression, mucositis, etc). Adequate hydration and alkalinization of the urine with sodium bicarbonate can help prevent these problems [22].

Finally, these patients should be closely followed during treatment, which can be done using serial imaging.

Role for intrathecal therapy — The need for intrathecal (IT) therapy in these patients is controversial [19,21,23,24], and its use alongside high-dose EMA-CO is based on institutional preferences. If administered, MTX is given as a 12.5 mg dose IT on day 8, followed by leucovorin calcium 15 mg at 24 and 36 hours. However, one study that included 15 patients treated with EMA-CO plus IT MTX reported that 87 percent (13 patients) achieved a sustained remission without the use of whole-brain irradiation [21]. An update of this study reported that 23 (85 percent) of 27 patients with brain metastases achieved complete remission with EMA-CO or EMA-EP (etoposide and cisplatin on day 8) plus IT MTX [25].

Concomitant whole-brain radiation therapy — At the New England Trophoblastic Disease Center, we have administered cranial RT (20 to 30 Gy in 2 Gy daily fractions) concurrently with high-dose MTX chemotherapy. As with the role of IT MTX, the role of cranial RT is also controversial. In addition to shrinking the brain metastases, concomitant cranial irradiation increases the MTX concentration within the central nervous system (CNS) [26] and reduces the risk of cerebral hemorrhage prior to eradication of tumor, and may improve survival [27,28]. However, the use of concurrent MTX and cranial irradiation also increased the likelihood of treatment-related toxicity, especially leukoencephalopathy. (See "Delayed complications of cranial irradiation", section on 'Neurocognitive effects'.)

Cisplatin-containing regimens — In many centers, a modified EMA-CO regimen that incorporates cisplatin is preferentially administered to patients with a risk score ≥12. The most commonly used combination replaces vincristine and cyclophosphamide (used in EMA-CO) with etoposide and cisplatin on day 8 (EMA-EP) (table 3). EMA-EP alone or in combination with surgery induced complete remission in 16 (76 percent) of 21 patients with EMA-CO resistance [5]. Among 143 patients with ultra-high-risk GTN (International Federation of Gynecology and Obstetrics [FIGO] score ≥12), only 97 (67.9 percent) achieved complete remission [29].

While no randomized trials compared EMA-CO with cisplatin-containing regimens, a retrospective study evaluated outcomes among 83 patients treated with cisplatin-containing treatment and 103 patients treated with EMA-CO [30]. Compared with EMA-CO, incorporation of cisplatin was associated with:

●A slightly lower remission rate (85 versus 92 percent, respectively)

●A lower number of cycles to achieve a normal hCG level (three versus five courses)

●More toxicity, including fever, nephropathy, nausea, and diarrhea

FAEV — Treatment with floxuridine, ActD, etoposide, and vincristine (FAEV) has also been reported. In one study of 30 patients with stage IV GTN receiving primary treatment with FAEV, 24 (80 percent) patients achieved complete and sustained remission with tolerable toxicity [31]. Among six patients with resistance, two died from progressive disease and four achieved remission with other regimens.

A review of 17 studies summarizing responses to EMA-CO, EMA-EP, and FAEV in the first-line setting for ultra-high-risk GTN concluded all three regimens had similar complete response rates [32].

APE — Given the activity of cisplatin for GTN, one group reported their experience using ActD, cisplatin, and etoposide (APE) for high-risk GTN that included 59 patients (out of a total of 95) who were treated with this regimen as initial therapy between 1985 and 2013 [33]. Patients with brain metastases or PSTT were excluded from treatment. The overall remission rate for these patients was 95 percent. One patient required treatment discontinuation after four cycles due to grade 2 ototoxicity. The final two had refractory disease to this regimen, but all three entered remission with second-line EMA-CO therapy. The five-year disease-free survival rate was 98 percent.

Surgery — Approximately 50 percent of patients with high-risk, metastatic GTN will require adjuvant surgery to achieve cure, even in the presence of multi-organ involvement [34-47]. As an example, in one study of 50 patients with high-risk, metastatic GTN treated with EMA-CO between 1986 and 2005, among 24 patients who underwent a total of 28 surgical procedures, the cure rate was 87.5 percent. These procedures included hysterectomy, pulmonary resection, uterine wedge resection, small bowel resection, and selective uterine artery embolization [40]. Surgery is generally performed to resect foci of chemotherapy-resistant disease or to control complications such as bleeding or infection.

METASTATIC PSTT OR ETT — Although universally accepted guidelines are not available, patients with metastatic placental site trophoblastic tumor (PSTT) or epithelioid trophoblastic tumor (ETT) should be treated with a combination of surgery and chemotherapy. Multi-agent regimens are usually administered, including etoposide, methotrexate, plus actinomycin D alternating with cyclophosphamide and vincristine (EMA-CO) [48] or EMA plus etoposide and cisplatin (EMA-EP) [49]. There are no prospective data to inform whether one or the other is the preferred regimen. In a retrospective review of cases of 37 women with stage II to IV PSTT treated with floxuridine-based regimens, the initial response rate was 92 percent, with a 20.6 percent relapse rate. Hysterectomy was performed in 28 (78.4 percent) of the patients [50]. While data are more limited, it appears that ETT is even less responsive to chemotherapy than PSTT [51]. (See "Management of resistant or recurrent gestational trophoblastic neoplasia", section on 'High-risk GTN'.)

Unfortunately, patients with metastatic disease have a relatively poor prognosis and a high fatality rate [52-54]. In a 2012 review of the literature that reported on the deaths from GTN, 30 percent had PSTT [54]. In addition, one single-institution series of 18 patients with PSTT reported that five of six who had extrauterine disease ultimately died despite the administration of multi-agent chemotherapy [53]. Survival with PSTT is strongly related to the number of years since the antecedent pregnancy. Papadopoulos et al, reporting on 34 patients, found that while all 27 patients diagnosed within four years of the antecedent pregnancy survived, all seven patients died when the antecedent pregnancy was more than four years [52]. Among 16 patients with stage II to IV ETT, chemotherapy and surgery in 10 patients achieved remission at 1 year in 8 (80 percent) patients [55]. Survival was adversely affected if the interval from the antecedent pregnancy was greater than four years.

As a result of this very poor prognosis, some have advocated the use of high-dose chemotherapy with peripheral blood stem cell support. In an observational study, this aggressive therapy achieved a 45 percent remission rate. Survivors tended to have a low human chorionic gonadotropin (≤12 international units/L) at the time of treatment initiation [56].

In an update to a series including 125 patients with PSTT or ETT, the overall probability of survival at 5 and 10 years following treatment was 80 and 75 percent, respectively [57]. Stage IV disease (hazard ratio [HR] 6.2) and a time interval ≥48 months since the antecedent pregnancy (HR 14.6) remained the significant prognostic factors for death on multivariable analysis. However, with the introduction of cisplatin regimens and high-dose chemotherapy with stem cell support, patients with an interval ≥48 months demonstrated an improved median overall survival compared with the prior report (8.3 versus 2.6 years).

MONITORING DURING TREATMENT — As with women who are treated for low-risk GTN, all women with high-risk GTN should be monitored with serial measurements of serum human chorionic gonadotropin (hCG) at the start of treatment and then at weekly intervals during therapy. This and other considerations for patients during treatment are discussed separately. (See "Initial management of low-risk gestational trophoblastic neoplasia", section on 'Monitoring during treatment'.)

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 [58-60].

Definition of remission — Remission is achieved when the quantitative hCG level becomes undetectable for three consecutive weeks. Given the sensitivity of this tumor marker, no imaging is required if levels are consistent with remission because abnormalities on imaging can persist despite the attainment of undetectable hCG levels, representing fibrosis rather than active tumor.

Persistent or progressive disease — The French Trophoblastic Disease Reference Center in Lyon defines chemotherapy resistance as an increase or a plateau in two consecutive hCG values over a two-week interval [61]. As described above, other generally accepted criteria include detection of new metastases [62]. For patients who appear to have persistent disease, surgery may be a curative option. The approach to patients who experience disease progression is discussed separately. (See "Management of resistant or recurrent gestational trophoblastic neoplasia".)

PROGNOSIS — The overall cure rate for patients with high-risk GTN (stage II to III) ranges between 90 and 100 percent [63]. At the New England Trophoblastic Disease Center (NETDC), we treated 93 patients between July 1965 and December 2018, of whom 75 percent had a sustained complete remission with primary chemotherapy (table 4). As depicted in the table, of those who developed resistance, second-line therapy ultimately resulted in sustained remission in 85 percent of patients. Overall, 90 (97 percent) of 93 patients with high-risk stage II and III GTN achieved complete remission.

For patients with International Federation of Gynecology and Obstetrics (FIGO) stage IV high-risk GTN, the prognosis is not as good, though complete remission can be expected in 60 to 70 percent of patients [63]. Much of this success is due to the use of combination chemotherapy. After combination chemotherapy was introduced as primary treatment in this group, the survival rate dramatically improved. Since the introduction of intensive primary combination chemotherapy with adjunctive surgery and radiation therapy, survival at the NETDC increased from 30 to 78 percent (table 5) [34].

Prognostic factors — For patients with high-risk GTN, long-term survival appears to be associated with variables related to disease extent [64]:

●In the presence of liver metastases, only 27 percent were alive

●If brain metastases were present, the survival rate was 70 percent

●If both were present, only 10 percent were alive

Data from the Charing Cross Hospital indicate that survival with hepatic metastases appears to have improved over time. Excluding early deaths (within four weeks of presentation) and two deaths unrelated to GTN, the cause-specific survival in 25 patients was 68 percent [65,66].

Other poor prognostic factors have also been reported. A retrospective study including 10 GTN Reference Centers in Brazil confirmed the negative prognostic impact of brain and liver metastases, but also identified additional risk factors for death from high-risk GTN including increased time between the end of antecedent pregnancy and the initiation of chemotherapy (relative risk [RR] 4.1), chemoresistance (RR 3.2), and initial treatment outside the Reference Center (RR 28.3) [67].

POST-TREATMENT SURVEILLANCE — After remission is achieved, serum human chorionic gonadotropin (hCG) should be measured monthly until there have been undetectable hCG levels for 12 months [68,69]. In a population-based cohort study from the United Kingdom, 86 percent of recurrences in patients with high-risk GTN occurred within one year of treatment cessation and 95 percent occurred within two years [70]. These data support discontinuing routine monitoring after one year. If a patient develops a rising hCG after remission, it is important to consider alternative sources for this elevation. These would include a new conception, which could be new molar pregnancy or a new intrauterine pregnancy, or pituitary hCG, which is more common if the ovaries experience significant damage following intensive combination chemotherapy [71].

Assays should measure hCG at least down to 5 milli-international units/mL and should measure not only intact hCG but also hCG fragments and subunits. When assays measure hCG to very low levels (<1 milli-international units/mL), it is particularly important to suppress pituitary production of luteinizing hormone and hCG with strong, oral contraceptives to facilitate reliable monitoring of disease status. (See "Human chorionic gonadotropin: Biochemistry and measurement in pregnancy and disease".)

The follow-up of patients treated for high-risk GTN is similar to that of women treated for low-risk GTN and includes considerations of contraception and the timing of subsequent pregnancies. These issues are discussed separately. (See "Initial management of low-risk gestational trophoblastic neoplasia", section on 'Posttreatment surveillance' and "Initial management of low-risk gestational trophoblastic neoplasia", section on 'Posttreatment issues'.)

Uterine arteriovenous malformation — Patients treated for GTN that invaded the myometrium (both low-risk and high-risk) are at risk of developing a uterine artery malformation (also known as an arteriovenous fistula), which can persist for months or years after remission is achieved. A fistula may remain asymptomatic and undiagnosed or may present with menorrhagia. The presence of an arteriovenous malformation has also been associated with recurrent miscarriage.

The diagnosis is made with the use of color Doppler transabdominal or endovaginal ultrasonography. Selective pelvic arteriography clearly identifies the abnormal vascular malformation and helps in evaluating possible treatment modalities. If symptomatic, treatment consists of hysterectomy or, when preservation of reproductive function is desired, treatment consists of selective uterine artery embolization [72-74]. When uterine artery embolization is not available, treatment with medroxyprogesterone acetate and tranexamic acid controlled hemorrhage in 14 of 16 (87.5 percent) patients with class I to II hemorrhage and in 6 of 11 (54.5 percent) patients with class III to IV hemorrhage (American College of Surgeons Classification of Hemorrhage) [75].

DIAGNOSIS OF RECURRENT OR RESISTANT DISEASE — Patients whose human chorionic gonadotropin (hCG) level re-elevates after becoming undetectable for three consecutive weeks are considered to have recurrent disease. By contrast, patients whose hCG level remains elevated despite treatment are considered to have resistant disease. Despite the success of combination chemotherapy, patients treated for high-risk GTN have an 8 to 10 percent risk of recurrence, dependent on stage and risk score [76]. This is often detected by a re-elevation in the quantitative serum hCG levels after three consecutive weeks of undetectable quantitative hCG levels. The approach to patients with recurrent or resistant disease is discussed separately. (See "Management of resistant or recurrent gestational trophoblastic neoplasia".)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Gestational trophoblastic disease".)

SUMMARY AND RECOMMENDATIONS

●Gestational trophoblastic neoplasia (GTN) is characterized by the International Federation of Gynecology and Obstetrics (FIGO) stage and the World Health Organization (WHO) risk score (table 1). Patients are defined as having high-risk GTN if they have stage IV disease or stage II to III disease with a risk score >6. (See 'Definition of high-risk disease' above.)

●The WHO Prognostic Scoring System is not applicable to patients with placental site trophoblastic tumor (PSTT) or epithelioid trophoblastic tumor (ETT). Therefore, patients with these subtypes of GTN are not categorized as either low risk or high risk. They should be characterized by their FIGO stage. (See 'Definition of high-risk disease' above.)

●GTN is uniquely sensitive to chemotherapy, which is the major treatment modality for patients with high-risk disease. The exception to this is women with PSTT or ETT, in which case, the primary treatment may be a combination of surgery and chemotherapy, primarily because PSTT and ETT are relatively resistant to chemotherapy as compared with choriocarcinoma and invasive mole. (See 'Approach to treatment' above.)

●For patients with high-risk GTN, we recommend multi-agent chemotherapy rather than single-agent therapy (Grade 1B). We suggest a combination of etoposide, methotrexate (MTX), plus actinomycin D (ActD) alternating with cyclophosphamide and vincristine (EMA-CO) (Grade 2C). (See 'EMA-CO' above.)

●For patients with high-risk GTN and brain metastases, a neurosurgical consult should be obtained prior to treatment. We suggest EMA-CO as the primary systemic treatment, using a higher MTX dose (1000 mg/m² over 24 hours) than what is routinely administered otherwise (Grade 2C). We also suggest cranial radiation therapy (RT) for these patients (Grade 2C). However, for patients who do not wish to proceed with cranial RT, we suggest additional chemotherapy using intrathecal MTX (Grade 2C). (See 'Patients with brain metastases' above.)

●Approximately 50 percent of patients with high-risk, metastatic GTN will require adjuvant surgery to achieve cure, even in the presence of multi-organ involvement. (See 'Surgery' above.)

●Although universally accepted guidelines are not available, for patients with PSTT or ETT, we suggest a combination of surgery and chemotherapy (Grade 2C). Multi-agent regimens are usually administered, including EMA-CO or EMA plus etoposide and cisplatin (EMA-EP). There are no prospective data to inform whether one or the other is the preferred regimen, and a choice between them is based on institutional preferences. (See 'Metastatic PSTT or ETT' above.)

●As with women who are treated for low-risk GTN, all women with high-risk GTN should be monitored with serial measurements of serum human chorionic gonadotropin (hCG) at the start of treatment and then at weekly intervals during therapy. (See 'Monitoring during treatment' above.)

●The overall cure rate for patients with high-risk GTN (stage II to III) ranges between 95 and 100 percent. The extent of disease is a prognostic factor among these patients. (See 'Prognosis' above.)

●After remission is achieved, serum hCG should be measured monthly until monitoring has shown one year of normal hCG levels. (See 'Post-treatment surveillance' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Donald Peter Goldstein, MD, who contributed to an earlier version of this topic review.