Abstract
Background
Guidelines recommend etoposide, methotrexate, actinomycin D (EMA)/cyclophosphamide, vincristine (CO) as first-line treatment for high-risk gestational trophoblastic neoplasia (GTN). However, the floxuridine, actinomycin D, etoposide and vincristine (FAEV) regimen is commonly used to treat these patients in China. We conducted a randomised controlled trial to compare the efficacies and toxicities of FAEV and EMA/CO.
Methods
Ninety-four patients with GTN were enrolled between May 2015 and April 2019 and randomly assigned to the FAEV or EMA/CO regimen. The rates of complete remission and relapse and the toxicities were compared in August 2021.
Results
Five patients were excluded from the analysis. There were 46 patients in the FAEV group and 43 patients in the EMA/CO group. The complete remission rates following primary treatment were 89.1% and 79.1% (P = 0.193), respectively. The relapse rates were 8.7% and 9.3% (P = 0.604). The apparent incidences of grade 4 myelosuppression were 60.9% and 32.6% (P = 0.008), respectively; however, they became both 32.6% (P = 0.996) after granulocyte colony-stimulating factor support. Other adverse reactions were similar in the two groups. No patient died of disease.
Conclusion
FAEV has comparable efficacy and toxicity to EMA/CO as the primary treatment for high-risk GTN, and may thus be another first-line choice of chemotherapy.
Clinical trial registration
chictr.org.cn: ChiCTR1800017423.
Subject terms: Gynaecological cancer, Chemotherapy
Background
Gestational trophoblastic neoplasia (GTN) comprises a group of malignant tumours that are highly sensitive to chemotherapy. The overall survival (OS) rate of low-risk patients (International Federation of Gynecology and Obstetrics [FIGO] risk factor score <7) is close to 100%, while that of high-risk patients (FIGO score ≥7) is about 85–94% [1–3]. Floxuridine, actinomycin D, etoposide and vincristine (FAEV) and etoposide, methotrexate, actinomycin D (EMA)/cyclophosphamide, vincristine (CO) are commonly used multi-agent chemotherapy regimens for high-risk patients. The FAEV regimen is favoured in China, while EMA/CO is used worldwide, and both have demonstrated good therapeutic effects.
Based on the experience of the Peking Union Medical College Hospital (PUMCH), the FAEV regimen has shown good efficacy [4–7]; however, studies comparing it with the EMA/CO regimen are still lacking. We therefore conducted a prospective randomised controlled study to compare the efficacy and safety of the FAEV and EMA/CO regimens as primary treatment for GTN in patients with high-risk factors. It should be noted that the two groups in this study have included similar proportions of patients with a FIGO score of 5 and 6, which could be treated with single-agent chemotherapy initially. They were included and treated with multi-agent chemotherapy because of an increased risk of resistance to single-agent chemotherapy, which was detailed in the discussion section.
Methods
Study design and conduct
This open-label randomised controlled trial (ChiCTR1800017423) was conducted at PUMCH. The study protocol was approved by the Ethics Committee of PUMCH (ZS-1601). The study was carried out in accordance with the guidelines for Good Clinical Practice and the Declaration of Helsinki. All patients provided signed informed consent. Patients were randomised in a 1:1 ratio to receive FAEV or EMA/CO. Briefly, the patients were enrolled after they passed the inclusion/exclusion criteria and gave consent to the study. They were numbered serially according to the admission order and alternatively assigned to either FAEV or EMA/CO group by the investigator. The consultant who recruited a patient and was responsible for the patient’s subsequent chemotherapy, adjuvant surgery (if needed) and follow-up, did not interfere with the group allocation.
Patients
Eligible patients were newly diagnosed with GTN, with a FIGO risk factor score ≥5. For patients with a FIGO score of 5 and 6, additional inclusion criteria were serum human chorionic gonadotropin (hCG) level >10,000 mIU/mL or clinicopathologic diagnosis of choriocarcinoma. The exclusion criteria were: (1) patients who had received single-agent or multi-agent chemotherapy for GTN; (2) patients with relapse of GTN; (3) patients with non-gestational choriocarcinoma and (4) patients with purely placental site trophoblastic tumour or epithelioid trophoblastic tumour confirmed by pathology.
Procedures
All patients were evaluated comprehensively before receiving chemotherapy to determine the stage and FIGO risk factor score, including medical history, physical examination, blood tests, chest X-ray or computed tomography, and pelvic and abdominal ultrasound or magnetic resonance examination. Cranial computed tomography or magnetic resonance were arranged to detect brain metastases for patients with bulky pulmonary metastasis. Blood tests included whole blood cell count, serum β-hCG level and serum biochemistry. Patients with the massive disease were treated with one to three cycles of actinomycin D and etoposide before FAEV or EMA/CO to decrease tumour burden and prevent sudden tumour collapse with severe bleeding, metabolic acidosis, septicemia and multiple organ failure. Patients then received either FAEV or EMA/CO as primary treatment. The FAEV regimen was as follows [5]; day 1, vincristine 2 mg, intravenous bolus, 3 h before actinomycin D; days 1–5, actinomycin D 0.2 mg/m2/day, intravenous infusion for >30 min; etoposide 100 mg/m2/day, intravenous infusion for >30 min; and floxuridine 800–900 mg/m2/day, intravenous infusion for 6–8 h. The interval between chemotherapy cycles was 17–21 days. Intrathecal injection of methotrexate 15 mg on day 1, 3 and 5 was given for intracranial metastasis before serological remission. The EMA/CO regimen was as follows: [1] EMA day 1: actinomycin D 0.5 mg, intravenous infusion for 1 h; etoposide 100 mg/m2, intravenous infusion for 1 h; methotrexate 100 mg/m2, intravenous bolus; methotrexate 200 mg/m2, intravenous infusion for 12 h; EMA day 2: actinomycin D 0.5 mg, intravenous infusion for 1 h; etoposide 100 mg/m2, intravenous infusion for 1 h; calcium folinate 15 mg, intramuscular injection, every 12 h for four doses, starting 12 h after treatment with methotrexate. CO day 8: vincristine 2 mg, intravenous infusion, 3 h before cyclophosphamide; cyclophosphamide 600 mg/m2, intravenous infusion for 2 h. The next course of treatment was started on day 15. Intrathecal injection of methotrexate 12.5 mg on day 1 and 8 was given for intracranial metastasis before serological remission.
Chemotherapy was changed to an alternative multi-agent regimen if chemoresistance occurred or if toxicity precluded an adequate dose or frequency of treatment. The salvage regimens included EMA/CO (for FAEV group), FAEV (for EMA/CO group), EMA/etoposide, cisplatin (EP), paclitaxel, etoposide/paclitaxel, cisplatin (TE/TP), floxuridine, etoposide and vincristine (FEV), and floxuridine, actinomycin D and vincristine (FAV), or immune checkpoint inhibitors if appropriate. Surgical resection of the tumour or uterus was performed if necessary, including total hysterectomy, uterine tumour excision, lung wedge resection or lobectomy and resection of abdominal, pelvic or intracranial metastases. Radiation therapy was administered in addition to chemotherapy and surgery in the event of intracranial chemotherapy-resistant metastasis.
Serum β-hCG levels were measured once a week within 1 month of treatment cessation, and then measured every 2 weeks for 2 months, monthly for half a year, and with gradually decreasing frequency (every 2–6 months) up to 5 years, after which levels were monitored once a year. If the patient became pregnant during the follow-up period, β-hCG levels were measured monthly within 1 year after the end of the pregnancy. The deadline for the follow-up of this report was August 2021.
Endpoints
Efficacy
Complete remission (CR) was achieved when the serum β-hCG level decreased to normal and remained at <5 mIU/mL for 4 continuous weeks. An additional 3–4 courses of consolidating chemotherapy were given to patients with a FIGO score > 6 and 2–3 courses to patients with a FIGO score of 5 and 6 after β-hCG levels normalised. Chemoresistance was defined if serum β-hCG levels plateaued or increased after two consecutive courses of chemotherapy, in which case salvage chemotherapy was administered. A diagnosis of relapse was made when serum β-hCG levels increased again following CR at >1 month after the cessation of chemotherapy, in the absence of a new pregnancy. The primary endpoint was the CR rate of the initial regimen. The secondary endpoints included the chemotherapy duration needed to achieve CR and relapse rate.
Toxicity
Whole blood cell counts were evaluated every other day during treatment and serum biochemistry was evaluated weekly. Side effects of chemotherapy, including myelosuppression, liver and renal damage, oral ulcers, vomiting and diarrhoea, were recorded to evaluate the safety of chemotherapy. Toxicity was graded according to the Common Toxicity Criteria for Adverse Events (5th edition) developed by the National Cancer Institute of the United States. Patients receiving EMA/CO also received granulocyte colony-stimulating factor (G-CSF) proactively in all treatment intervals, while patients receiving FAEV were administered G-CSF in the event of grade 3–4 or febrile neutropenia. Full-dose chemotherapy was continued on schedule if myelosuppression or other side effects were relieved. The safety endpoint was the rate at which patients changed to other regimens because of chemotherapy intolerance.
Additional follow-up information on post-chemotherapy pregnancy and delivery was collected to evaluate the influence of the regimens on fertility.
Sample size and statistical analysis
CR rate of first-line FAEV regimen in patients with high-risk GTN as intervention group was estimated as 90% in accordance with our clinical practice and that of EMA/CO as reference group was reported to be around 80% (62–94%) [2, 3, 8–12]. Using non-inferiority test for the difference between two proportions, we estimated that two study groups that each included 30 patients would result in the study having 80% power to detect a non-inferiority difference of –15% with 95% certainty (PASS version 11, NCSS, LLC., Kaysville, Utah, USA).
Continuous variables with a normal distribution were described as mean ± standard deviation and analysed by t-tests. The homogeneity of variance was confirmed by Levene’s test. Non-normally distributed continuous variables were described as median with interquartile range and analysed using the Wilcoxon–Mann–Whitney U test. Categorical variables were compared using Pearson’s χ2 or Fisher’s exact test, and multiple groups of categorical variables were compared using the likelihood ratio test. Statistical analyses were carried out using SPSS version 23.0 (IBM Corp, Armonk, NY, USA). All tests were two-sided. Differences were considered to be significant if the P-value was <0.05.
Results
Patient characteristics
A total of nighty-four patients were enrolled between May 2015 and April 2019 and were randomly assigned to receive FAEV (n = 47) or EMA/CO (n = 47). The assignment of subjects and their outcomes are shown in Fig. 1. Five patients (one in FAEV group and four in EMA/CO group) were excluded from analysis because of non-gestational choriocarcinoma or pathologically confirmed placental site trophoblastic tumour (see Fig. 1). The baseline demographic characteristics and commonly assumed prognostic factors were similar in both groups (Table 1).
Fig. 1. Flowchart of subject assignment and treatment outcomes.
PSTT placental site trophoblastic tumour, NGC non-gestational choriocarcinoma, CR complete remission, PR partial remission, including chemoresistance and chemotherapy intolerance.
Table 1.
Baseline clinical characteristics of patients.
| Clinical characteristics | FAEV (N = 46) | EMA/CO (N = 43) | P |
|---|---|---|---|
| Age, year, mean ± SD | 35.52 ± 8.12 | 33.47 ± 7.94 | 0.231 |
| Gravidity, median (IQR) | 3 (2–4) | 3 (2–4) | 0.239 |
| Parity, median (IQR) | 1 (1–2) | 1 (1–2) | 0.182 |
| Body mass index, kg/m2, mean ± SD | 23.26 ± 3.66 | 23.54 ± 4.15 | 0.730 |
| Body surface area, m2, mean ± SD | 1.62 ± 0.17 | 1.64 ± 0.16 | 0.644 |
| GTN diagnosis classification, n (%) | 0.211 | ||
| CC | 39 (84.8) | 37 (86.0) | |
| CC + ETT | 1 (2.2) | 0 | |
| CC + PSTT | 2 (4.3) | 0 | |
| IM | 4 (8.7) | 6 (14.0) | |
| FIGO stage, n (%) | 0.057 | ||
| I | 13 (28.3) | 10 (23.3) | |
| II | 0 | 2 (4.7) | |
| III | 23 (50.0) | 28 (65.1) | |
| IV | 10 (21.7) | 3 (7.0) | |
| FIGO risk factor score, mean ± SD | 9.37 ± 3.72 | 8.58 ± 2.72 | 0.260 |
| FIGO score classification, n (%) | 0.757 | ||
| 5–6 | 12 (26.1) | 10 (23.3) | |
| ≥7 | 34 (73.9) | 33 (76.7) | |
| Antecedent pregnancy, n (%) | 0.483 | ||
| Mole | 6 (13.0) | 8 (18.6) | |
| Abortion | 15 (32.6) | 17 (39.5) | |
| Term | 25 (54.3) | 18 (41.9) | |
| Interval from index pregnancy, months, n (%) | 0.208 | ||
| <4 | 19 (41.3) | 16 (37.2) | |
| 4–<7 | 6 (13.0) | 6 (14.0) | |
| 7–<13 | 0 | 3 (7.0) | |
| ≥13 | 21 (45.7) | 18 (41.9) | |
| Pre-treatment hCG, mIU/mL, median (IQR) | 134,638 (33,290–322,539) | 113,959 (29,851–266,994) | 0.440 |
| Largest tumour size including uterus, cm, mean ± SD | 5.05 ± 2.65 | 5.2 ± 1.79 | 0.771 |
SD standard deviation, IQR interquartile range, GTN gestational trophoblastic neoplasia, IM invasive mole, CC choriocarcinoma, ETT epithelioid trophoblastic tumour, PSTT placental site trophoblastic tumour, FIGO International Federation of Gynecology and Obstetrics, hCG human chorionic gonadotropin.
Efficacy outcomes
The efficacy outcomes of FAEV and EMA/CO are listed in Table 2. There was no significant difference in the CR rates of the initial regimen in all patients between the FAEV (41/46, 89.1%) and EMA/CO groups (34/43, 79.1%) (P = 0.193), and no difference in the subset of high-risk patients (FIGO score >6) (85.3% vs. 78.8%, P = 0.487). All patients achieved CR after salvage treatment. Adjuvant surgeries (excluding curettage and surgeries after relapse) were also similar in both groups.
Table 2.
Treatment and efficacy outcomes.
| Outcomes | FAEV (N = 46) | EMA/CO (N = 43) | P |
|---|---|---|---|
| CR rate, n (%) | |||
| CR rate of initial regimen | 41/46 (89.1) | 34/43 (79.1) | 0.193 |
| CR rate of initial regimen in patients with a FIGO score >6 | 29/34 (85.3) | 26/33 (78.8) | 0.487 |
| CR rate after salvage chemotherapy | 46/46 (100) | 43/43 (100) | |
| Chemotherapy duration, weeks, mean ± SD | |||
| Chemotherapy weeks before reaching CR | 13.00 ± 6.73 | 11.34 ± 4.84 | 0.187 |
| Consolidating chemotherapy weeks | 9.51 ± 3.06 | 7.22 ± 2.52 | <0.001 |
| Total chemotherapy weeks | 22.51 ± 7.47 | 18.56 ± 5.24 | 0.005 |
| Surgery (excluding curettage and those after relapse), n (%) | 0.877 | ||
| Extra-uterine surgery ± uterine surgerya | 8 (17.4) | 7 (16.3) | |
| Uterine surgerya only | 20 (43.5) | 21 (48.8) | |
| None | 18 (39.1) | 15 (34.9) | |
| Follow-up | |||
| Follow-up period, months, median (IQR) | 46.1 (33.1–54.6) | 48.7 (42.2–55.9) | 0.067 |
| Lost-to-follow-up rate, n (%) | 2/46 (4.3) | 2/43 (4.7) | 1.000 |
| Relapse rate, n (%) | 4/46 (8.7) | 4/43 (9.3) | 0.604 |
| Relapse rate in patients with a FIGO score >6 | 2/34 (5.9) | 4/33 (12.1) | 0.427 |
| CR rate after relapse, n (%) | 4/4 (100) | 4/4 (100) | |
| Dead of disease, n (%) | 0/43 (1 suicide) | 0/41 | |
CR complete remission, SD standard deviation, IQR interquartile range.
aUterine surgery included hysterectomy and excision of uterine disease.
The chemotherapy duration before reaching CR was similar in the two groups (P = 0.187). The consolidating chemotherapy duration and total chemotherapy duration were both longer in the FAEV compared with the EMA/CO group (P < 0.001, P = 0.005, respectively) (see Table 2).
Three patients in the FAEV group (3/46, 6.5%) and eight in the EMA/CO group (8/43, 18.6%) developed chemoresistance and received salvage chemotherapy (P = 0.083). The diagnoses, primary regimens, salvage regimens, adjuvant surgeries and outcomes are listed in Table 3. Most salvage treatments involved switching between FAEV and EMA/CO, but other regimens, such as FEV or FAV, were given to patients who had grade 2 liver damage when using EMA/CO (see Cases 1 and 6 in EMA/CO group in Table 3).
Table 3.
Patients with chemoresistance to primary regimens.
| Case | Diagnosis | Primary treatment | Salvage treatment | Adjuvant surgeries | Outcome |
|---|---|---|---|---|---|
| FAEV group | |||||
| 1 | CC III:9 | FAEV*4 | EMA/CO*7 | HYS | CR |
| 2 | CC IV:11 | FAEV*5 | EMA/CO*5+TE/TC*4 | LL, nephrectomy, HYS | CR |
| 3 | CC IV:20 | FAEV*7 | EMA/CO*5 | Partial hepatectomy | CR-RL-CR |
| EMA/CO group | |||||
| 1 | CC III:9 | EMA/CO*7 | FEV*3 | LL | CR-LF |
| 2 | CC III:8 | EMA/CO*7 | FAEV*4 | CR | |
| 3 | CC III:9 | EMA/CO*8 | FAEV*3 | HYS | CR |
| 4 | CC IV:16 | EMA/CO*4+AE*2+MTX it*2 | FAEV*8 | CR | |
| 5 | CC III:12 | EMA/CO*4+AE*1 | FAEV*4 | HYS | CR |
| 6 | CC I:11 | EMA/CO*6 | FAV*3 | CR | |
| 7 | CC III:10 | EMA/CO*5 | FAEV*5 | LL | CR |
| 8 | CC III:5 | EMA/CO*2 | EMA/EP*3 | HYS | CR |
Case 2 in the FAEV group used carboplatin instead of cisplatin in TE/TP regimen because of unilateral renal dysfunction owing to metastasis.
CC III:9 diagnosed with choriocarcinoma, stage III, FIGO risk factor score 9, CR complete remission, RL relapse, LF lost to follow-up, HYS hysterectomy, LL lung lobectomy, FAEV floxuridine, actinomycin D, etoposide and vincristine, EMA/CO etoposide, methotrexate, actinomycin D/cyclophosphamide, vincristine, TE/TC paclitaxel, etoposide/paclitaxel, carboplatin, FEV floxuridine, etoposide and vincristine, AE actinomycin D, etoposide, FAV floxuridine, actinomycin D and vincristine, MTX it methotrexate intrathecal injection (used for intracranial metastasis), EMA/EP etoposide, methotrexate, actinomycin D/etoposide, cisplatin.
The two groups were followed up at similar times. The relapse rates were similar in all patients between the FAEV (4/46, 8.7%) and EMA/CO groups (4/43, 9.3%) (P = 0.604), and were also similar in the subset of high-risk patients (5.9% vs. 12.1%, P = 0.427, see Table 2). All relapsed patients (4/4 in FAEV and 4/4 in EMA/CO) achieved CR after salvage treatment (see Table 4). Three patients (Cases 3 and 4 in FAEV group and Case 2 in EMA/CO group; see Table 4) had second relapses and achieved CR after treatment. One patient in the FAEV group committed suicide 20 months after the termination of chemotherapy, with no evidence of disease recurrence. At the time of analysis, four patients (2 FAEV, 2 EMA/CO) had been lost to follow-up and no patients had died of disease in either group (see Table 2).
Table 4.
Patients with relapse.
| Case | Primary diagnosis | Primary treatment | Primary surgeries | PFS, months | Medical treatment of relapse | Adjuvant therapy | Outcome |
|---|---|---|---|---|---|---|---|
| FAEV group | |||||||
| 1 | CC I:7 | FAEV*7 | 3 | FAEV*3+EMA/CO*5 | HYS | CR | |
| 2 | CC IV:20 | FAEV*7 + EMA/CO*5 | Partial hepatectomy | 2 | ICI*13 | LW | CR |
| 3 | CC + ETT III:5 | FAEV*5 | 3 | EMA/CO*4+FAEV*2, (RL) ICI*10 | HYS, LL | CR-RL-CR | |
| 4 | CC III:5 | FAEV*7 | 8 | EMA/CO*4, (RL) EMA/CO*4+AE*1 | Resection of uterine tumour, (RL) LW | CR-RL-CR | |
| EMA/CO group | |||||||
| 1 | CC III:9 | EMA/CO*10 | LL, HYS | 9 | FAEV*2+MTX it*2+AE*2 | Resection of intracranial metastasis, RT | CR |
| 2 | CC III:11 | EMA/CO*5 | HYS | 22 | FAEV*6, (RL) FAEV*3+ICI*4 | LL | CR-RL-CR |
| 3 | CC I:9 | EMA/CO*8 | 5 | FAEV*3 | HYS | CR | |
| 4 | CC IV:13 | EMA/CO*8+MTX it*6 | 2 | FAEV*4+MTX it*3 | LL, RT | CR | |
PFS progression-free survival, CC I:7 diagnosed with choriocarcinoma, stage I, FIGO risk factor score 7, ETT epithelioid trophoblastic tumour, CR complete remission, RL relapse, RT radiation therapy, HYS hysterectomy, LW lung wedge resection, LL lung lobectomy, FAEV floxuridine, actinomycin D, etoposide and vincristine, EMA/CO etoposide, methotrexate, actinomycin D/cyclophosphamide, vincristine, AE actinomycin D, etoposide, MTX it methotrexate intrathecal injection (used for intracranial metastasis), ICI immune checkpoint inhibitor.
Toxicity outcomes
Severe adverse reactions during treatment with the primary regimens are listed in Table 5. Myelosuppression, mainly neutropenia, was the most common adverse reaction in both groups. Grade 4 myelosuppression and neutropenia were significantly more frequent in the FAEV group (P = 0.008 and P = 0.001, respectively). However, these two side effects in FAEV group both reduced and became similar to those in EMA/CO group (P = 0.996 and P = 0.466, respectively) after patients in FAEV group received G-CSF support. The occurrences of other grade 4 adverse reactions, including thrombocytopenia, anaemia, liver damage, renal damage, vomiting, diarrhoea and oral ulcers, were similar in both groups.
Table 5.
Grade 4 adverse reactions occurring during treatment with primary regimens.
| Adverse reactions with primary regimens | FAEV (N = 46) n (%) | EMA/CO (N = 43) n (%) | P |
|---|---|---|---|
| Myelosuppression | 28 (60.9) | 14 (32.6) | 0.008 |
| Myelosuppression after G-CSF support | 15 (32.6) | 14 (32.6) | 0.996 |
| Neutropenia | 28 (60.9) | 11 (25.6) | 0.001 |
| Neutropenia after G-CSF support | 15 (32.6) | 11 (25.6) | 0.466 |
| Thrombocytopenia | 3 (6.5) | 1 (2.3) | 0.617 |
| Anaemia | 1 (2.2) | 4 (9.3) | 0.193 |
| Elevation of alanine aminotransferase | 0 | 1 (2.3) | 0.483 |
| Elevation of bilirubin | 0 | 0 | |
| Elevation of serum creatinine | 0 | 0 | |
| Vomit | 0 | 0 | |
| Diarrhoea | 0 | 0 | |
| Oral ulcer | 0 | 0 | |
| Patients altered to second-line treatment due to chemotherapy intolerance | 2 (4.3) | 1 (2.3) | 0.526 |
G-CSF granulocyte colony-stimulating factor.
Most of the side effects were tolerated after active treatment. However, two patients (4.3%) in the FAEV group had grade 4 neutropenia, which delayed their chemotherapy and they were switched to EMA/CO. One patient (2.3%) in the EMA/CO group had intolerable liver damage and switched to FAV. The difference between the groups was not significant (P = 0.526) and all three patients achieved CR after changing the regimen and undergoing adjuvant surgeries.
Fertility
We evaluated the impact of chemotherapy on fertility during follow-up as an exploratory endpoint. Among the fertility-spared patients, seven (7/22, 31.8%) patients in the FAEV group and three (3/16, 18.8%) in the EMA/CO group stopped using contraception after chemotherapy. Eight and three pregnancies were reported in the FAEV and EMA/CO groups, respectively, resulting in two term deliveries, two premature deliveries, one miscarriage and three ongoing pregnancies in the FAEV group, and two-term deliveries and one ongoing pregnancy in the EMA/CO group. Four patients in the FAEV group (4/7, 57.1%) and two (2/3, 66.7%) in the EMA/CO group had live births at the time of analysis (P > 0.95). The interval from treatment cessation to pregnancy was similar in the FAEV (24.4 ± 14.3 months) and EMA/CO groups (38.0 ± 18.2 months) (P = 0.220).
Discussion
GTN is a malignant tumour that is highly sensitive to chemotherapy, with many different chemotherapy regimens producing favourable results. There is currently no unified primary treatment plan in clinical practice, and different primary treatment schemes are adopted worldwide. Factors influencing the choice of treatment include the availability of chemotherapeutic agents, the accuracy of blood β-hCG determination, doctors’ proficiency in using different chemotherapy regimens and their experience in dealing with the side effects of different chemotherapeutic agents.
The 2000 FIGO staging and classification classified a risk score ≤6 as low risks, a score >6 as high risk, and a FIGO score >12 as ultra-high risk. The FIGO Cancer Report 2021 on GTN suggested that high-risk patients should receive first-line treatment with multi-agent chemotherapy regimens while low-risk patients could use single-agent regimens [1]. However, among low-risk patients, a higher risk score (5–6) and a clinicopathologic diagnosis of choriocarcinoma are both associated with an increased risk of resistance to single-agent chemotherapy [1, 13, 14]. These patients only have about a 35% chance of being cured with a methotrexate regimen [14, 15]. A recent multicenter retrospective cohort study suggested that for patients with GTN presenting with a FIGO score of 5 or 6, primary multi-agent chemotherapy should only be given to patients with metastatic disease and choriocarcinoma [16]. The FIGO Cancer Report 2021 on GTN allowed the use of multi-agent chemotherapy in these otherwise low-risk patients [1]. The patients with a FIGO score of 5 and 6 in the current study presented with a high pre-treatment hCG level or diagnosis of choriocarcinoma and received primary multi-agent regimens. There was no difference in the proportion of these patients between the two groups (26.1% vs. 23.3%, P = 0.757).
Efficacy
Previous studies reported CR rates of first-line EMA/CO of 62–94% in patients with high-risk GTN, and OS rates of 71–100% after a median follow-up period of 12–72 months [2, 3, 8–10]. The CR rate of EMA/CO in China under similar conditions was 77.8–82.7% [11, 12] and the OS rate was 87.0%, with an average follow-up period of 56 months [11].
The FIGO guidelines currently recommend the FAEV regimen as salvage therapy after chemoresistance to EMA/CO or as the first-line choice for ultra-high-risk patients [1]. However, there are currently fewer clinical data for FAEV than for EMA/CO. Data from PUMCH showed CR rates of 69.6% for FAEV as primary treatment in high-risk GTN patients [6] and 80% in stage IV GTN patients [7], while FAEV salvage treatment could achieve CR rates of 60.4–85.9% in patients with resistant or relapsed GTN [4, 5].
In the present study, the CR rates of FAEV during primary treatment were 89.1% overall and 85.3% in patients with a risk score >6, and the corresponding rates for EMA/CO were 79.1% and 78.8%, respectively. These results are consistent with previously published data [2, 6–12]. There were no significant differences in CR rates between the two primary treatment groups, with CR rates of 100% in both groups after salvage therapy. There was also no difference in relapse rates between the two groups after follow-up for about 4 years in all subjects (P = 0.604) or high-risk subset (P = 0.427), and all patients with one or two relapse(s) achieved CR after salvage treatment. These results indicate that the FAEV regimen had similar efficacy to EMA/CO over both the short- and long-terms.
The median number of treatment courses in the EMA/CO group was nine and the median treatment period was 17.3 weeks, which was consistent with recent data from the UK (9 courses [IQR 6–9], 17.6 weeks [IQR 13.–20.7]) [17]. There was no difference in the chemotherapy duration before reaching CR between the two groups (P = 0.187). But the consolidating chemotherapy duration and total treatment duration were both longers in the FAEV group (P < 0.001, P = 0.005, respectively). This may have been because the FAEV cycle was longer than EMA/CO cycle. The FIGO guideline suggests 2–3 courses of consolidating chemotherapy for low-risk GTN and 4 courses for ultra-high-risk GTN [1]. As per the recommendation on the number of courses, in this study, patients in both groups received a median of 3 courses of consolidation chemotherapy.
Toxicity
The major side effects of FAEV and EMA/CO both included hematologic toxicity, gastrointestinal reactions, and reversible alopecia. The incidences of FAEV-related grade 4 neutropenia and thrombocytopenia at PUMCH were 16.5–26.7% and 1.1–6.7%, respectively [5, 7]. Data from Turkey and the UK showed that 9.1–15.0% of patients who received EMA/CO developed grade 4 neutropenia [17, 18].
Hematologic toxicity was the most common adverse reaction in both groups in the present study. The only significant difference between the two groups was in grade 4 neutropenia, which was more common in the FAEV group (P = 0.001). The difference in grade 4 myelosuppression was influenced by neutropenia. G-CSF support was employed to help minimise hematologic toxicity and to avoid delays or dosage reduction in both groups. The treatment interval for the EMA/CO regimen was so short that neutropenia without G-CSF would be likely to delay the dosage, so G-CSF was given proactively in the EMA/CO group. However, in the FAEV group, G-CSF was only used after the development of grade 3–4 neutropenia. Once the patients in FAEV group started with G-CSF support, grade 4 neutropenia and myelosuppression were reduced and became similar to that in EMA/CO group (P = 0.466, P = 0.996, respectively). Furthermore, the difference in the occurrence of neutropenia did not lead to a corresponding difference in changes to the regimens because of intolerance (P = 0.526). Both regimens were well tolerated considering the active treatment of side effects.
Fertility
Fertility-sparing treatment of GTN has been advocated for over 60 years, and most reports have shown favourable results [19, 20]. A retrospective analysis of 464 patients at PUMCH who were treated with only floxuridine-based regimens revealed a livebirth rate of 72.2% over a median follow-up period of 85 months [19]. A meta-analysis evaluating reproductive outcomes after chemotherapy for GTN revealed a pregnancy rate of 86.7% [20]. A systematic review showed that pregnancies after EMA/CO had a chance of 57–78% ending with live healthy babies [21]. The fertility results in the current study appeared similar in the two groups, but the sample size of this subset was too small for statistical analysis.
The current study has several limitations. Firstly, the alternative assignment of patients according to the enrollment sequence was not an exact randomisation method. Therefore, we tested the baseline clinical characteristics of patients in the two groups and found no significant difference. Secondly, primary multi-agent chemotherapy for GTN patients with a FIGO score 5–6 might be viewed as radical. Some data suggested primary single-agent chemotherapy [16]. We enrolled selected patients with other risk factors based on our centre’s experience that this subgroup of patients had a high possibility to develop resistance to single-agent chemotherapy, which was also supported by the FIGO guideline and data from other centres [1, 13]. Thirdly, the consolidation chemotherapy complied with FIGO guideline in course numbers, resulting in a longer consolidation chemotherapy period in FAEV group. However, it had no influence on the CR rate of the primary regimen in the present study. In the end, the recruitment was slow and lasted for 4 years, which might cause time-related bias. Most patients with GTN referred to PUMCH had received previous chemotherapy at a local hospital, thus precluding their inclusion in the present study.
In conclusion, our results suggest that the FAEV regimen favoured in China has comparable efficacy and toxicity to the EMA/CO regimen, which is widely used internationally. The FAEV regimen may thus be used as another first-line treatment for patients with GTN, especially high-risk patients. Further studies with longer follow-ups are needed to clarify the impacts of FAEV on disease relapse and fertility. Multicenter prospective randomised controlled trials are needed to gather a larger sample and acquire more data to guide the optimal selection of chemotherapy regimens in patients with high-risk GTN.
Supplementary information
Acknowledgements
We thank Susan Furness, PhD, from Liwen Bianji (Edanz) (www.liwenbianji.cn/) for editing the English text of a draft of this manuscript.
Author contributions
MJ and SJ participated in data collection, data interpretation, statistical analyses and wrote the original draft. JZ, XW, FF, TR, JY and YX participated in the patient enrollment, investigation and data curation. JZ and YX conceived the study, and participated in its design, supervision, data interpretation, analysis and manuscript revision. All authors read and approved the final manuscript.
Funding
This work was supported by grants from the National Natural Science Foundation of China (grant no. 81971475) and the National Key Technology R&D Program of China (grant no. 22019YFC1005204).
Data availability
The data that support the findings of this study are available from the authors upon reasonable request and with permission of Peking Union Medical College Hospital.
Competing interests
The authors declare no competing interests.
Ethics approval and consent to participate
This open-label randomised controlled trial (ChiCTR1800017423) was conducted at PUMCH. The study protocol was approved by the Ethics Committee of PUMCH (ZS-1601). The study was carried out in accordance with the guidelines for Good Clinical Practice and the Declaration of Helsinki. All patients provided signed informed consent.
Consent for publication
No relevant identifiable patient data.
Footnotes
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
These authors contributed equally: Mingliang Ji, Shiyang Jiang.
Contributor Information
Jun Zhao, Email: zhaojun@pumch.cams.cn.
Yang Xiang, Email: xiangy@pumch.cn.
Supplementary information
The online version contains supplementary material available at 10.1038/s41416-022-01809-3.
References
- 1.Ngan HYS, Seckl MJ, Berkowitz RS, Xiang Y, Golfier F, Sekharan PK, et al. Diagnosis and management of gestational trophoblastic disease: 2021 update. Int J Gynaecol Obstet. 2021;155:86–93. doi: 10.1002/ijgo.13877. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Braga A, Elias KM, Horowitz NS, Berkowitz RS. Treatment of high-risk gestational trophoblastic neoplasia and chemoresistance/relapsed disease. Best Pract Res Clin Obstet Gynaecol. 2021;74:81–96. doi: 10.1016/j.bpobgyn.2021.01.005. [DOI] [PubMed] [Google Scholar]
- 3.Alifrangis C, Agarwal R, Short D, Fisher RA, Sebire NJ, Harvey R, et al. EMA/CO for high-risk gestational trophoblastic neoplasia: good outcomes with induction low-dose etoposide-cisplatin and genetic analysis. J Clin Oncol. 2013;31:280–6. doi: 10.1200/JCO.2012.43.1817. [DOI] [PubMed] [Google Scholar]
- 4.Wan XR, Xiang Y, Yang XY, Wu Y, Liu N, Chen L, et al. Efficacy of FAEV regimen in treatment of high-risk drug-resistant gestational trophoblastic tumor. Zhonghua Fu Chan Ke Za Zhi. 2006;41:88–90. [PubMed] [Google Scholar]
- 5.Feng F, Xiang Y, Wan X, Geng S, Wang T. Salvage combination chemotherapy with floxuridine, dactinomycin, etoposide, and vincristine (FAEV) for patients with relapsed/chemoresistant gestational trophoblastic neoplasia. Ann Oncol. 2011;22:1588–94. doi: 10.1093/annonc/mdq649. [DOI] [PubMed] [Google Scholar]
- 6.Wang T, Feng FZ, Xiang Y, Wan XR, Ren T. Combination chemotherapy regimen with floxuridine, dactinomycin, etoposide, and vincristine as primary treatment for gestational trophoblastic neoplasia. Zhongguo Yi Xue Ke Xue Yuan Xue Bao. 2014;36:300–304. doi: 10.3881/j.issn.1000-503X.2014.03.013. [DOI] [PubMed] [Google Scholar]
- 7.Yang J, Xiang Y, Wan X, Feng F, Ren T. Primary treatment of stage IV gestational trophoblastic neoplasia with floxuridine, dactinomycin, etoposide and vincristine (FAEV): a report based on our 10-year clinical experiences. Gynecol Oncol. 2016;143:68–2. doi: 10.1016/j.ygyno.2016.07.099. [DOI] [PubMed] [Google Scholar]
- 8.Deng L, Zhang J, Wu T, Lawrie TA. Combination chemotherapy for primary treatment of high-risk gestational trophoblastic tumour. Cochrane Database Syst Rev. 2013;CD005196. 10.1002/14651858.CD005196.pub4. [DOI] [PMC free article] [PubMed]
- 9.Brown J, Naumann RW, Seckl MJ, Schink J. 15years of progress in gestational trophoblastic disease: scoring, standardization, and salvage. Gynecol Oncol. 2017;144:200–207. doi: 10.1016/j.ygyno.2016.08.330. [DOI] [PubMed] [Google Scholar]
- 10.Jareemit N, Horowitz NS, Goldstein DP, Berkowitz RS, Elias KM. EMA vs EMACO in the treatment of gestational trophoblastic neoplasia. Gynecol Oncol. 2020;158:99–104. doi: 10.1016/j.ygyno.2020.04.699. [DOI] [PubMed] [Google Scholar]
- 11.Lu WG, Ye F, Shen YM, Fu YF, Chen HZ, Wan XY, et al. EMA-CO chemotherapy for high-risk gestational trophoblastic neoplasia: a clinical analysis of 54 patients. Int J Gynecol Cancer. 2008;18:357–62. doi: 10.1111/j.1525-1438.2007.00999.x. [DOI] [PubMed] [Google Scholar]
- 12.Liu W, Zhao W, Zhang YQ, Huang XF. Curative effects and influenced factors of EMA-CO as an initial regimen for the treatment of high-risk gestational trophoblastic neoplasia. Zhonghua Yi Xue Za Zhi. 2018;98:3896–99. doi: 10.3760/cma.j.issn.0376-2491.2018.47.017. [DOI] [PubMed] [Google Scholar]
- 13.Lurain JR. Gestational trophoblastic disease II: classification and management of gestational trophoblastic neoplasia. Am J Obstet Gynecol. 2011;204:11–18. doi: 10.1016/j.ajog.2010.06.072. [DOI] [PubMed] [Google Scholar]
- 14.Braga A, Mora P, de Melo AC, Nogueira-Rodrigues A, Amim-Junior J, Rezende-Filho J, et al. Challenges in the diagnosis and treatment of gestational trophoblastic neoplasia worldwide. World J Clin Oncol. 2019;10:28–37. doi: 10.5306/wjco.v10.i2.28. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Sita-Lumsden A, Short D, Lindsay I, Sebire NJ, Adjogatse D, Seckl MJ, et al. Treatment outcomes for 618 women with gestational trophoblastic tumours following a molar pregnancy at the Charing Cross Hospital, 2000-2009. Br J Cancer. 2012;107:1810–14. doi: 10.1038/bjc.2012.462. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Braga A, Paiva G, Ghorani E, Freitas F, Velarde LGC, Kaur B, et al. Predictors for single-agent resistance in FIGO score 5 or 6 gestational trophoblastic neoplasia: a multicentre, retrospective, cohort study. Lancet Oncol. 2021;22:1188–98. doi: 10.1016/S1470-2045(21)00262-X. [DOI] [PubMed] [Google Scholar]
- 17.Singh K, Gillett S, Ireson J, Hills A, Tidy JA, Coleman RE, et al. M-EA (methotrexate, etoposide, dactinomycin) and EMA-CO (methotrexate, etoposide, dactinomycin / cyclophosphamide, vincristine) regimens as first-line treatment of high-risk gestational trophoblastic neoplasia. Int J Cancer. 2021;148:2335–44. doi: 10.1002/ijc.33403. [DOI] [PubMed] [Google Scholar]
- 18.Turan T, Karacay O, Tulunay G, Boran N, Koc S, Bozok S, et al. Results with EMA/CO (etoposide, methotrexate, actinomycin D, cyclophosphamide, vincristine) chemotherapy in gestational trophoblastic neoplasia. Int J Gynecol Cancer. 2006;16:1432–8. doi: 10.1136/ijgc-00009577-200605000-00074. [DOI] [PubMed] [Google Scholar]
- 19.Jiang F, Yang K, Wan XR, Xiang Y, Feng FZ, Ren T, et al. Reproductive outcomes after floxuridine-based regimens for gestational trophoblastic neoplasia: a retrospective cohort study in a national referral center in China. Gynecol Oncol. 2020;159:464–9. doi: 10.1016/j.ygyno.2020.08.018. [DOI] [PubMed] [Google Scholar]
- 20.Tranoulis A, Georgiou D, Sayasneh A, Tidy J. Gestational trophoblastic neoplasia: a meta-analysis evaluating reproductive and obstetrical outcomes after administration of chemotherapy. Int J Gynecol Cancer. 2019;29:1021–31. doi: 10.1136/ijgc-2019-000604. [DOI] [PubMed] [Google Scholar]
- 21.Joneborg U, Coopmans L, van Trommel N, Seckl M, Lok CAR. Fertility and pregnancy outcome in gestational trophoblastic disease. Int J Gynecol Cancer. 2021;31:399–411. doi: 10.1136/ijgc-2020-001784. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are available from the authors upon reasonable request and with permission of Peking Union Medical College Hospital.