INTRODUCTION — The term gestational trophoblastic neoplasia (GTN) is used when molar and nonmolar pregnancies become malignant. GTN comprises specific histologic entities, including:
●Invasive mole
●Choriocarcinoma
●Placental site trophoblastic tumor (PSTT)
●Epithelioid trophoblastic tumor (ETT)
GTN is a rare disease that ideally should be managed at trophoblastic disease centers where concentration of cases provides clinicians with ample experience, opportunities for research, and improved outcomes [1]. Since most patients are managed locally, however, it is important for clinicians who encounter these patients to familiarize themselves with the latest advances in the field in order to optimize their patients' outcomes.
Prior to the development of effective chemotherapy for GTN, the majority of patients with disease localized to the uterus were cured with hysterectomy, whereas metastatic disease was almost uniformly fatal [2]. With the utilization of sensitive quantitative assays for human chorionic gonadotropin (hCG) and highly effective chemotherapy, most patients with GTN can be cured and their reproductive function preserved, providing they are managed according to well-established guidelines [3,4]. However, patients are at risk for both resistant and recurrent GTN, which will require further therapy. This is especially true if initial treatment is inappropriate [5]. Despite this, patients who experience a recurrence stand a good chance of achieving remission.
Patients whose hCG level re-elevates after becoming undetectable are considered to have recurrent disease. By contrast, patients whose hCG level remains elevated despite treatment are considered to have resistant disease.
This topic will discuss the approach to and treatment of resistant or recurrent GTN. The pathology, epidemiology, clinical manifestations, diagnosis, and management of hydatidiform mole and low-risk and high-risk GTN are discussed elsewhere:
●(See "Gestational trophoblastic disease: Pathology and genetics".)
●(See "Hydatidiform mole: Epidemiology, clinical features, and diagnosis".)
●(See "Hydatidiform mole: Treatment and follow-up".)
●(See "Initial management of low-risk gestational trophoblastic neoplasia".)
●(See "Human chorionic gonadotropin: Biochemistry and measurement in pregnancy and disease".)
ESTIMATING THE RISK OF RESISTANT OR RECURRENT DISEASE — Patients treated for GTN are at varying risks of developing either resistant or recurrent disease. In one study, the reported recurrence rates according to International Federation of Gynecology and Obstetrics (FIGO) staging (table 1) were as follows [6]:
●Stage I (non-metastatic) – 2 percent
●Stages II and III; low-risk GTN – 4 percent
●Stage II to IV; high-risk – 13 percent
At the New England Trophoblastic Disease Center, the reported recurrence rates were 2.9 percent in patients with non-metastatic disease, 8.3 percent in stage II patients, 4.2 percent in stage III patients, and 9.1 percent in patients with FIGO stage IV disease [7]. In a study of 1827 patients with GTN from Peking Union Medical College in Beijing, recurrence was associated with an interval between the antecedent pregnancy and chemotherapy >12 months, and an interval from first chemotherapy to achieving β-human chorionic gonadotropin (hCG) normalization >14 weeks [8].
An increasing total FIGO risk score correlates with a corresponding increase in the likelihood of chemotherapy resistance. However, the majority of the risk factors reflect large initial tumor burden (eg, higher hCG, tumor size, presence and number of metastases), and are not independent prognostic factors [9]. Inadequate primary therapy and noncompliance or loss to follow-up are also important risk factors for resistance [5]. Several studies have also noted that a low uterine artery pulsatility index, as assessed by Doppler ultrasound prior to chemotherapy initiation, is a marker for higher tumor vascularity and suggests an increased risk of methotrexate resistance [10,11].
PRETREATMENT EVALUATION — All patients who develop chemoresistant or recurrent disease require re-staging. In addition, this evaluation should be repeated each time a new course of treatment is indicated. (See "Gestational trophoblastic neoplasia: Epidemiology, clinical features, diagnosis, staging, and risk stratification", section on 'Staging and risk assessment'.)
Patients who recur or develop resistance often have multiorgan involvement, particularly if they were previously treated for high-risk GTN [12]. Therefore, re-imaging with chest, abdominal, and pelvic computed tomography (CT) scans and brain magnetic resonance imaging (MRI) should be performed to help guide treatment options. For patients in whom the diagnosis is questionable, positron emission tomography (PET) scanning may aid in the differential characterization of active disease (which should be PET-positive) from fibrotic tumor nodules [13-15].
LOW-RISK GTN — In general, primary resistance after single-agent chemotherapy occurs in 10 to 30 percent of patients with low-risk GTN. However, this figure rises to between 30 and 50 percent of patients with low-risk GTN in the presence of metastases. Patients who develop resistance to the initial single agent usually will respond to an alternative single agent, and only 5 to 10 percent will require multiagent therapy [16,17].
There have been no randomized trials comparing single-agent with multiagent chemotherapy for patients with low-risk GTN who experience primary resistance to single-agent chemotherapy. However, our clinical experience indicates that a second-line single-agent therapy is effective and induces remission in the vast majority of patients. As such, this is our preferred approach. Other institutions, however, use human chorionic gonadotropin (hCG) level at the time of introduction of second-line treatment to determine whether single- or multiagent chemotherapy should be initiated. For patients with an hCG level under 1000 international units/L, single-agent actinomycin D (ActD) is their standard choice, while for patients with an hCG >1000 international units/L at the time of treatment, etoposide, methotrexate (MTX), ActD, cyclophosphamide, and vincristine (EMA-CO) is chosen. Using this cutoff, approximately 93 percent of patients treated with single-agent ActD successfully completed treatment without requiring additional therapy [18]. Importantly, those who do not achieve remission are cured with additional therapy or surgery. By increasing the cut-off more patients were spared exposure to the more toxic EMA-CO regimen without compromising survival. Of note, some data suggest that patients with higher risk scores (ie, 5 to 6) are at a greater risk of resistant disease compared with those with lower prognostic scores. This was shown in one single-institution study in which only 30 percent of these patients went into a sustained remission with second-line monotherapy [19]. These patients characteristically present with a pretreatment hCG level >100,000 milli-international units/mL and Doppler ultrasound evidence of a large tumor burden in their uterus [19]. For such patients, multiagent chemotherapy is a reasonable option.
In a multicenter retrospective, cohort study evaluating predictors of single-agent resistance (>80 percent risk) in patients with FIGO score 5 or 6 GTN, about 60 percent of patients achieved remission with either first-line or second-line single agents [20]. Almost all patients without metastases or choriocarcinoma (CCA) attained remission with single agents, but 4 percent of these patients with hCG >410,000 IU/L required combination chemotherapy. Among patients with either metastasis or CCA and hCG >150,000 IU/L (14 percent of group), combination chemotherapy was required to attain remission, and all patients with metastatic CCA required combination chemotherapy.
Second-line single-agent chemotherapy — For patients treated with MTX, with or without folinic acid (FA), who develop resistant or recurrent disease, we then administer biweekly bolus ActD. In our experience, more than 70 percent of these patients will achieve sustained remission with second-line ActD and avoid multiagent chemotherapy [21-23]. If the initial treatment was ActD, we proceed with treatment using MTX-FA. These and other single-agent options are discussed separately. (See "Initial management of low-risk gestational trophoblastic neoplasia", section on 'Methotrexate as the preferred option' and "Initial management of low-risk gestational trophoblastic neoplasia", section on 'Alternatives including dactinomycin and other agents'.)
In patients with MTX resistance, single-agent carboplatin has been utilized with mixed results in two studies [24,25]. Complete remission was achieved with carboplatin in 11 of 23 (48 percent) and 17 of 21 (81 percent) patients. Grade 3 or 4 hematologic toxicity was seen with granulocytopenia in 48 and 38 percent and thrombocytopenia in 43 and 29 percent. When available, ActD should be the preferred second-line agent in patients with MTX-resistant, low-risk GTN.
An alternative treatment option for patients with MTX-resistant disease is avelumab. In a phase II, multicenter trial evaluating avelumab (10 mg/kg every two weeks) in 15 patients with MTX-resistant GTN, hCG normalization occurred in eight patients (53 percent) after nine cycles (median) of avelumab during the 25-month (median) follow-up period; no patients relapsed [26]. Subsequent cure for the remaining seven patients was achieved with ActD or combination chemotherapy. Toxicities were consistent both in nature and frequency with those typically seen with immunotherapy.
Progression after second-line single-agent therapy — Patients who have resistant or recurrent disease despite second-line single-agent chemotherapy should be treated with combination chemotherapy. As in patients with high-risk GTN, our regimen of choice is EMA-CO. (See 'Later-line therapy' below and "Initial management of high-risk gestational trophoblastic neoplasia".)
For patients who do not respond to initial combination chemotherapy, alternative regimens can be offered. However, other modalities, including surgery, may be reasonably pursued. (See 'Later-line therapy' below and 'Surgery' below.)
HIGH-RISK GTN — Despite the success of primary therapy with etoposide, methotrexate (MTX), and actinomycin D (ActD) alternating with cyclophosphamide plus vincristine (EMA-CO), roughly 30 to 40 percent of patients with high-risk metastatic GTN will have an incomplete response to first-line therapy or will relapse from remission and will need additional multiagent chemotherapy with or without other treatment modalities [12,27-29]. Risk factors for resistant or recurrent disease include the presence of multiple metastases to sites other than the lung and vagina, and inadequate first-line chemotherapy. (See "Initial management of high-risk gestational trophoblastic neoplasia", section on 'EMA-CO'.)
Second-line combination therapy — Although there are no universally accepted evidence-based guidelines for second-line treatment for patients who develop resistant or recurrent disease on first-line multiagent chemotherapy, our approach is as follows:
●For patients previously treated with EMA-CO, we administer the multiagent chemotherapy combination consisting of EMA followed by etoposide plus cisplatin (EMA-EP). (See "Initial management of high-risk gestational trophoblastic neoplasia", section on 'Cisplatin-containing regimens'.)
Neutropenia may become an issue for patients receiving EMA-CO followed by EMA-EP, and in our clinical experience, prolonged neutropenia develops by the second or third cycle. Therefore, growth factor support (ie, filgrastim) is usually required [30]. In addition to chemotherapy, these patients may benefit from surgical excision of localized, persistent tumor [31-33]. Further discussion of the role of surgery is covered below. (See 'Surgery' below.)
In one report of 21 patients who had previously received EMA-CO, EMA-EP (alone or in combination with surgery) induced remission in 16 (76 percent) [34]. In a separate report that included 49 patients, the complete remission rate was 82 percent [33].
●Patients who were not treated with EMA-CO in the first-line setting should undergo treatment with this combination in the second-line setting. (See "Initial management of high-risk gestational trophoblastic neoplasia", section on 'EMA-CO'.)
LATER-LINE THERAPY — For patients with resistant or recurrent disease despite two prior combination regimens, a number of alternative regimens can be administered. However, none of these regimens have been used in a sufficient number of patients to identify one as being the optimal choice.
It is important to re-stage patients prior to the initiation of a new regimen so that the extent of disease may be appropriately characterized and to evaluate for a potential role for surgical treatment. Case reports of patients treated with these later-line regimens do not indicate that one is preferred over another.
Multiagent therapy
TE-TP — Paclitaxel and etoposide alternating weekly with paclitaxel and cisplatin (TE-TP) has had encouraging results in heavily pretreated patients. It appears to be well tolerated as well, though the experience remains quite limited.
●In one report of two patients, both had a complete remission to this regimen [35].
●In a separate series of 16 patients (including six previously treated with cisplatin), 19 and 31 percent had a complete and partial remission, respectively [36].
Etoposide has been linked to an increased risk of secondary tumors such as leukemia, melanoma, colon cancer, and breast cancer [37]. However, a subsequent report from the same institution on a larger number of patients suggested that the risk of secondary tumors may be more limited than originally described [38].
PC — The combination of paclitaxel and carboplatin (PC) was shown in one prospective study of 65 patients to produce remissions in approximately 60 percent of patients at a median follow-up of 30 months and may be an appropriate option for those wishing to avoid the possible increased risk for secondary tumors associated with etoposide-containing regimens [39].
BEP — The combination of bleomycin, etoposide, and cisplatin (BEP) is an available regimen, widely used to treat germ cell tumors of the ovary and testicle. In one report of 16 patients with EMA-CO-resistant disease, 11 patients (69 percent) had a complete response, and nine (56 percent) survived [33].
ICE — The combination of ifosfamide, etoposide, and cisplatin (ICE) is often administered to men with recurrent or chemorefractory testicular cancer. Its use in patients with recurrent or resistant GTN is limited. In one experience that included six patients, four had a complete remission, and three ultimately survived [33].
PVB — Cisplatin, vinblastine, and bleomycin (PVB) has only been reported in small case series of patients with high-risk, drug-resistant GTN. The complete remission rates range from 18 to 62 percent [40-42].
FU plus ActD — High-dose fluorouracil (FU) in conjunction with actinomycin D (ActD) is commonly used in Asia, though the experience with this regimen is limited. For example, one series of 11 patients reported a complete remission rate of 82 percent [43].
FUDR — In one series, floxuridine (FUDR) in combination with ActD, etoposide, and vindesine was reported to induce remission in all 21 patients, all of whom had evidence of chemotherapy-resistant disease [44]. Floxuridine-based regimens have also been reported to be highly effective in placental site trophoblastic tumor (PSTT), which is relatively resistant to chemotherapy [45,46]. In one study of 37 patients with stage II to IV PSTT treated with floxuridine-based regimens, 34 (92 percent) had an initial complete response and seven (20.6 percent) patients relapsed [46]. FUDR with ActD, etoposide, and vincristine (FAEV) was administered to 91 patients with chemotherapy-resistant or relapsed GTN and 55 (60 percent) patients achieved a complete remission, 29 had no response, and 7 had intolerable toxicity [47].
Single-agent therapy — Very limited data suggest that single-agent treatment may provide benefit in patients with previously treated recurrent, advanced, or metastatic GTN. These include:
●Paclitaxel – Case reports indicate that paclitaxel may be an active agent following prior treatment. However, the experience with single-agent paclitaxel therapy remains quite limited [48,49].
●Capecitabine – A case report indicates that capecitabine may also be active in the management of relapsed high-risk GTN [50].
●Pegylated liposomal doxorubicin – Pegylated liposomal doxorubicin was reported to induce complete remission in two patients with chemotherapy-resistant high-risk GTN with brain metastases and having received multiple prior lines of chemotherapy [51].
Investigational therapies — New investigational therapies for GTN include immune checkpoint inhibitors and molecularly targeted agents. The rationale for immunotherapy in this disease is histologic evidence that GTNs strongly express programmed cell death ligand 1 (PD-L1) [52-54].
Due to the rarity of these diseases, prospective clinical trials are difficult to pursue, however, particularly for recurrent disease. Hence, further evidence of potential new therapies comes from small case series. In one such report, investigators administered the immune checkpoint inhibitor pembrolizumab to four patients with resistant GTN [55]. Three of the four patients in the case series had a complete response, and all were durable at 24, 15, and 5 months after stopping treatment. Of note, all patients had tumors with >90 percent programmed cell death protein 1 (PD-1) expression. In addition, all three patients expressed tumor-infiltrating lymphocytes, which were not present in the patient who did not respond to treatment. Additional case reports have reported similar results [56,57].
These provocative data point to the potential role of immune checkpoint inhibitors for resistant GTN and warrant further evaluation.
Antiangiogenic agents may also be active in the treatment of drug-resistant GTN. A 36-year-old patient with metastatic choriocarcinoma refractory to multiple lines of combination chemotherapy including stem cell transplant achieved a durable complete remission after receiving an anti-endoglin monoclonal antibody and bevacizumab [58]. She remains in complete remission off therapy for greater than four years.
SURGERY — For patients with recurrent or resistant GTN, a surgical approach may be curative. However, decisions regarding surgery must be individualized based upon the clinical scenario. The efficacy, feasibility, and type of surgery vary depending upon the site and extent of metastases.
Uterine disease — Hysterectomy should be avoided if possible, especially for patients who desire future childbearing. However, it may be indicated after chemotherapy, especially for heavy bleeding, large bulky intrauterine disease, or in the presence of sepsis [59]. In one report including 134 patients with GTN, a hysterectomy was performed in 13 (9 percent) patients for profuse bleeding and 31 (24 percent) for uterine perforation, respectively [60].
Hysterectomy may also be performed to manage chemotherapy resistance to reduce tumor burden [61]. At our center, hysterectomy was performed in 19 patients for drug resistance, and 16 (84 percent) of the patients attained a complete and sustained remission [62]. Similarly in another study, hysterectomy was performed in 18 patients with chemotherapy-resistant uterine tumors and 15 (83 percent) patients achieved complete remission [63].
For patients undergoing a hysterectomy, it is important to note that the removal of the ovaries is usually not indicated, even in the presence of theca lutein cysts. In our experience, metastatic disease involving the ovary is rare. Symptomatic ovarian theca lutein cysts can be aspirated if indicated, but otherwise will regress spontaneously after treatment.
Patients with GTN undergoing minimally invasive compared with open hysterectomy appear to have similar oncologic outcomes [64]. However, hospital length of stay and blood loss are less with minimally invasive approaches.
Bowel, splenic, and renal metastases — Intra-abdominal metastases may require resection, as they are often less chemosensitive and may persist despite chemotherapy [65,66]. In addition, these metastases may cause complications, particularly hemorrhage [65,67,68]. Some data suggest that clinical parameters may predict the outcome following resection:
●In one series of 33 patients who underwent various surgical procedures for resistant or recurrent GTN, successful outcome was achieved mostly in patients with the following characteristics: a single preoperative disease site, underwent salvage surgery within one year of initial diagnosis, non-choriocarcinoma histology, and/or a total World Health Organization (WHO) score of <8 [69].
●In another report of 61 patients with chemoresistant GTN undergoing surgical salvage, the clinical factors that predicted failure were age over 35, human chorionic gonadotropin (hCG) >10,000 milli-international units/mL, antecedent nonmolar pregnancy, and metastases outside the lungs [65].
Hepatic metastases — Management of persistent liver metastases can be a particularly difficult and challenging problem. Hepatic resection or selective embolization may be used in select cases to control bleeding or excise resistant tumor [70-72]. Because of their hypervascular nature, biopsy should not be performed because of the potential for life-threatening hemorrhage. In a series from Charing Cross Hospital, after excluding early deaths, the cause-specific survival was 68 percent in patients with liver metastases [73,74].
Pulmonary metastases — Resection of chemoresistant pulmonary disease can be curative. In one report of 15 patients with chemoresistant lung metastases, 14 achieved remission [75]. Positive prognostic indicators include normalization of hCG within one to two weeks after resection of a pulmonary metastasis [76], and a separate series reported that successful outcomes were possible even after multiple resections [77]. When evaluating patients for thoracotomy, it is important to remember that fibrotic lung nodules may mimic areas of active disease.
Criteria that predict favorable outcome include [75]:
●Absence of other systemic metastases
●Presence of a unilateral solitary nodule
●No uterine involvement
●Serum hCG <1500 milli-international units/mL
Prior to surgery, other sites of metastases should be excluded. It is particularly important to determine that the nodule contains viable tumor, since nonviable fibrotic nodules may persist indefinitely after tumor regression [66]. One study suggested that a residual lung lesion on imaging after completion of GTN treatment was not associated with increased relapse rate [14]. Positron emission tomography combined with computerized tomography (PET/CT) scans have been used to identify viable tumor before surgery [13,78]. Since most resistant pulmonary nodules contain choriocarcinoma, it is not unusual for patients to require additional chemotherapy to treat occult disease. However, additional therapy may not be necessary if the hCG level declines promptly and remains undetectable during follow-up. Several investigators have reported that undetectable hCG levels one or two weeks after resection of a solitary nodule is highly predictive of a favorable outcome [79-81], and residual lung lesions after normalization of hCG levels do not alter the prognosis of GTN [82].
Brain metastases — Craniotomy can play a vital role in the management of GTN cerebral metastases. It is indicated for the resection of peripheral, solitary, drug-resistant lesions and can be life-saving, especially in the context of a neurologic emergency due to intracranial hemorrhage or increased intracranial pressure [83-85]. Resection of intracranial disease can also result in a remission. This was shown in a series of patients who underwent craniotomy for drug-resistant disease [86]. Following craniotomy, five of seven patients had a complete remission. In addition to surgery, radiation, either stereotactic or whole brain, can be employed. Furthermore, the dose of methotrexate in the setting of brain metastasis needs to be increased to 1 g/m2. In one series of 27 patients with brain metastases, 85 percent of patients were long-term survivors [85]. These patients received neurosurgery, when indicated, increased MTX dosing (1 g/m2), and intrathecal MTX (12.5 mg), but did not undergo any radiotherapy.
IS THERE A ROLE FOR AUTOLOGOUS HEMATOPOIETIC STEM CELL TRANSPLANT? — The use of high-dose chemotherapy and autologous hematopoietic stem cell transplantation (HDSCT) is not clear, and the experience with this is mainly limited to case reports [87-89]. For example, in one study of 11 patients with previously treated disease, only two temporary complete responses were reported [89]. However, in an updated review, 32 patients were treated with HDSCT, and 7 patients achieved complete sustained remission and 6 additional patients attained remission with additional treatment [90]. Overall, 13 of 32 (41 percent) patients achieved complete remission after HDSCT. The patients that survived had human chorionic gonadotropin values ≤12 IU/L before HDSCT. Although encouraging, the role of HDSCT in GTN has not yet been clearly defined. However, the use of autologous hematopoietic stem cell transplant, which is associated with profound immunosuppression, should be avoided during the COVID-19 pandemic [91].
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
●Prior to the development of effective chemotherapy for gestational trophoblastic neoplasia (GTN), metastatic disease was almost uniformly fatal. With the utilization of sensitive quantitative assays for human chorionic gonadotropin (hCG) and highly effective chemotherapy, patients with resistant disease and those who experience a recurrence stand a good chance of achieving remission. (See 'Introduction' above.)
●Patients treated for GTN are at varying risks of developing either resistant or recurrent disease. In one study, the reported recurrence rates were 2, 4, and 13 percent for stage I (non-metastatic), stage II to III (low-risk), and stage III to IV (high-risk) disease. (See 'Estimating the risk of resistant or recurrent disease' above.)
●Patients who recur or develop resistance often have multiorgan involvement, particularly if they were previously treated for high-risk GTN. Therefore, re-imaging with chest, abdominal, and pelvic CT scans and brain MRI should be performed. (See 'Pretreatment evaluation' above.)
●For patients with low-risk GTN who are resistant to or recur after prior single-agent treatment, we recommend an alternative single agent rather than multiagent chemotherapy (Grade 1B). For example, patients who are resistant to or recur after prior treatment with methotrexate (MTX) with or without folinic acid (FA) should receive biweekly bolus actinomycin D (ActD). (See 'Second-line single-agent chemotherapy' above.)
●For patients previously diagnosed with low-risk GTN who develop resistant or recurrent disease despite two prior regimens using single agents, we administer combination chemotherapy. We suggest etoposide, MTX, ActD, cyclophosphamide, and vincristine (EMA-CO) (Grade 2C). (See 'Progression after second-line single-agent therapy' above.)
●For patients with high-risk GTN who have resistant or recurrent disease after first-line combination chemotherapy, our approach is as follows (see 'Second-line combination therapy' above):
•For patients previously treated with EMA-CO, we suggest EMA followed by etoposide plus cisplatin (EMA-EP) (Grade 2C). (See "Initial management of high-risk gestational trophoblastic neoplasia", section on 'Cisplatin-containing regimens'.)
•For patients not treated with EMA-CO in the first-line setting, we suggest EMA-CO (Grade 2C).
●For patients who develop resistant or recurrent disease despite two lines of multiagent chemotherapy, alternative combination regimens are available. In the absence of prospective data, there is no one preferred regimen. It is important to re-stage patients prior to the initiation of a new regimen so that the extent of disease may be appropriately characterized and to evaluate for a potential role for surgical treatment. (See 'Later-line therapy' above.)
●For patients with recurrent or resistant GTN, a surgical approach may be curative. However, decisions regarding surgery must be individualized based upon the clinical scenario. The efficacy, feasibility, and type of surgery vary depending upon the site and extent of metastases. (See 'Surgery' above.)
ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Donald Peter Goldstein, MD, who contributed to an earlier version of this topic review.
Do you want to add Medilib to your home screen?