Gynecologic Oncology 124 (2012) 553–557

Contents lists available at SciVerse ScienceDirect

Gynecologic Oncology

j ourna l homepage: www.e lsev ie r .com/ locate /ygyno

Combined methotrexate–dactinomycin: An effective therapy for low-risk gestationaltrophoblastic neoplasia☆

Lua Eiriksson a,1, Tiffany Wells b,2, Helen Steed a,⁎, Alexandra Schepansky a, Valerie Capstick a,Paul Hoskins b, Judith Pike b, Kenneth Swenerton b

a Division of Gynecologic Oncology, Cross Cancer Institute, 11560 University Avenue, Edmonton, Alberta, Canada, T6G 1Z2b Gynecology Tumour Group, BC Cancer Agency, 600 West 10th Avenue, Vancouver, British Columbia, Canada V5Z 4E6

☆ Prior presentation of paper: Society of GynecologiGeneral Meeting, June 2009, Vancouver, British Columb⁎ Corresponding author at: Cross Cancer Institute, Dep

University Avenue, Edmonton, Alberta, Canada, T6G 1Z2E-mail address: Helen.Steed@albertahealthservices.c

1 Present address: Division of Gynecologic Oncology,Toronto, Ontario, Canada, M5N 2L5.

2 Permanent address: Division of Gynecologic Onc11560 University Avenue, Edmonton, Alberta, Canada, T

0090-8258/$ – see front matter © 2011 Elsevier Inc. Alldoi:10.1016/j.ygyno.2011.10.036

a b s t r a c t

a r t i c l e i n f o

Article history:

Received 14 July 2011Accepted 30 October 2011Available online 9 November 2011

Keywords:Gestational trophoblastic diseaseGestational trophoblastic neoplasmsDrug therapyRemission inductionRecurrenceAdverse effects

Objective. The objective of this study is to examine the outcomes of combined chemotherapy using meth-otrexate and dactinomycin in the management of women with low-risk gestational trophoblastic neoplasia(GTN). The primary outcome is the total number of cycles of chemotherapy required to achieve a normallevel of human chorionic gonadotropin (hCG). The secondary outcome is treatment-related toxicity.

Methods. A retrospective chart review of all patients with GTN treated between 1996–2007 and1991–2007 was performed at the Alberta Cross Cancer Institute and the British Columbia Cancer Agency, re-spectively. Patients with low-risk GTN, treated with 0.6 mg/m2 dactinomycin (days 1 and 2) and methotrex-ate 100 mg/m2 were included. Toxicities were graded using the National Cancer Institute CommonTerminology Criteria for Adverse Events. The number of cycles to achieve normalization of hCG was deter-mined, and multivariate analyses were performed to identify factors associated with treatment duration.

Results. One hundred women were eligible. The average age was 29 years (range 15–46). The median

number of cycles to achieve a normal hCG was 3 (range 1–11). Two patients required second-line treatmentand one patient chose to proceed with hysterectomy. Ninety-eight percent of patients were primarily curedwith this regimen, and 2 were cured with second line treatment. Grade 3 and 4 hematologic toxicities wereexperienced by 12% and 8% of patients, respectively. Grade 2 and 3 stomatitis or mucositis were noted in 44%and 3% of patients, respectively.

Conclusions. Low-risk GTN is reliably and rapidly cured with combined methotrexate–dactinomycin.Toxicity is modest.

© 2011 Elsevier Inc. All rights reserved.

Introduction

Gestational trophoblastic disease is a spectrum of disease process-es characterized by abnormal trophoblastic proliferation. Gestationaltrophoblastic neoplasia (GTN) is a lesion with the potential for localinvasion and metastasis, including invasive moles, gestational chorio-carcinomas and placental site trophoblastic tumors [1]. Hydatidiformmoles are noted in 0.2/1000–9.9/1000 pregnancies according topopulation-based studies [2]. Subsequent post-molar GTN developsin 15–20% of complete moles and approximately 5% of partial moles

c Oncology of Canada, Annualia, Canada.artment of Gynecology, 11560. Fax: +1 780 432 8333.a (H. Steed).M700-610 University Avenue,

ology, Cross Cancer Institute,6G 1Z2.

rights reserved.

following uterine evacuation [2–4]. The majority of GTNs are highlysensitive to chemotherapy.

Various systems have been proposed to separate patients with ges-tational trophoblastic neoplasia (GTN) into risk categories based oncombinations of prognostic factors. The most recent system, as definedby the FIGO Oncology Committee, incorporates patient age, details ofthe antecedent pregnancy, and time to diagnosis, as well as tumor his-tology and size, metastases and treatment history (Table 1) [5].Single-agent chemotherapy is typically employed in the treatment oflow-risk disease (FIGO risk score 0–6) (Table 2), while combinationchemotherapy is usually reserved for patients with high-risk disease(FIGO risk score≥7), or for those with low-risk disease who failsingle-agent chemotherapy.

A unique double-agent protocol was developed at the British Co-lumbia Cancer Agency (BCCA). Methotrexate and dactinomycinwere combined in an effort to improve primary cure rates, aimingfor therapy of short duration and minimal toxicity in young patientsdesiring a rapid return to health and fertility. The BCCA initiatedthis GOTDLR (gyne-oncology trophoblastic disease low risk) protocolin 1991 (Fig. 1). It was then adopted in 1996 by the Cross Cancer

Table 1FIGO 2000 scoring system of gestational trophoblastic neoplasia.⁎

0 1 2 4

Age (years) b40 >40 – –

Antecedent pregnancy Mole Abortion Term –

Interval from indexpregnancy (months)

b4 4–b7 7–b13 >13

Pretreatment serum hCG (IU/L) b103 103–b104 104–b105 >105

Largest tumor size (cm) – 3–b5 >5 –

Site of metastases Lung Kidney/spleen GI/liver BrainNumber of metastases – 1–4 5–8 >8Previous failed chemotherapy – – Single drug ≥2

⁎ Score 0–6=Low Risk. Score>6=High Risk.

554 L. Eiriksson et al. / Gynecologic Oncology 124 (2012) 553–557

Institute (CCI). At both centers GOTDLR remains the first-line treat-ment for low-risk GTN.

The primary objective of this study was to examine the efficacyand toxicity of this regimen. The secondary objective was to comparethese outcomes with those reported for other regimens. While not aprospective trial, and limited by its retrospective nature, this studysought to examine the outcomes of two population-based practicesusing a novel chemotherapeutic approach in the treatment of low-risk GTN.

Methods

This study was designed as a retrospective record review of allwomen with low-risk GTN identified between 1996 and 2006 at theAlberta CCI, and between 1991 and 2006 at the BCCA who were trea-ted with combined methotrexate and dactinomycin. CCI and BCCAdatabases were used to identify eligible patients for whom follow-up results were available.

Gestational trophoblastic disease was diagnosed initially by ultra-sound and uterine curettage. GTN was subsequently diagnosed usingserial quantitative serum human chorionic gonadotropin (hCG) con-centrations. A plateau (decline less than 10% for at least four valuesover three weeks) or a rise (increase of more than 10% in three valuesover two consecutive weeks) was diagnostic. Patients were excludedif they were diagnosed with high-risk GTN (including choriocarci-nomas and placental site trophoblastic tumors), became pregnantduring follow-up prior to resolution of hCG, underwent primary de-finitive surgical management, or experienced spontaneous regression(Fig. 2). While transitioning to the new protocol, 39 low-risk patientsat the CCI received single agent chemotherapy according to physicianpreference and were therefore also excluded from this analysis(Fig. 2).

Patients presenting to the CCI and BCCA with GTN were evaluatedprior to the initiation of chemotherapy with a complete history andphysical examination, chest X-ray and/or computed tomography

Table 2Chemotherapy protocols in the treatment of low risk gestational trophoblasticneoplasia.

Drug Dose Frequency

Methotrexate 30–50 mg/m2

IM or IVWeekly

15–30 mg IM Daily×5 days every 2 weeks0.4 mg/kg IM Daily×5 days every 2 weeks50 mg IM Alternating with folinic acid 15 mg×8 days

every 2 weeks1 mg/kg IM Alternating with folinic acid 0.1 mg/

kg×8 days every 2 weeksDactinomycin 9–13 mcg/kg IV Daily×5 days every 2 weeks

1.25 mg/m2 IV Every 2 weeksAlternating

Methotrexate 20 mg IM Daily×5 days every 2 weeksDactinomycin 500 mcg IV Daily×5 days every 2 weeks

(CT), laboratory investigations (complete blood count, hepatic andrenal function, quantitative hCG), and abdominal and pelvic ultraso-nography as deemed appropriate. “Initial hCG” is defined as the hCGdocumented at the time of diagnosis (and prior to uterine evacua-tion), while “pre-treatment hCG” is the level drawn just prior to theinitiation of chemotherapy. Those classified as low-risk and meetingeligibility criteria (rising hCG post molar pregnancy, pulmonary me-tastases post molar pregnancy, FIGO risk score≤6, excluding risinghCG or pulmonary metastases post normal pregnancy) were offeredGOTDLR (Figs. 1 and 2) (www.bccancer.bc.ca), which included twocycles of chemotherapy following normalization of hCG. Pulmonarymetastases diagnosed on chest X-ray were used for risk-scoring. Allpatients were re-scored at the time of this review, using the FIGO2000 classification, and satisfied the updated criteria for low-riskGTN.

Two hundred and twenty-one patients with GTN were identifiedin these registries. Of these, one hundred patients received GOTDLRas first-line chemotherapy for low-risk GTN (Fig. 2). Seventy-threepatients were treated at the BCCA and twenty-seven patients at theAlberta CCI.

Upon presentation to hospital, patients were pre-hydrated with1000 mL of 2/3 D5W–1/3 NS with 20 mEq potassium chloride and100 mEq sodium bicarbonate/L at 200 mL/h IV. When urine outputwas at least 100 mL/h and urine pH greater than 7, chemotherapywas initiated. Patients received 0.6 mg/m2 dactinomycin on days 1and 2 by IV push, with an overnight hospital stay at the BCCA, versusoutpatient treatment at the CCI. Following dactinomycin on day 1,methotrexate 100 mg/m2 was administered by IV push (10 mg/min)or in 50–100 mL of NS over 20–30 min. This was followed by300 mg/m2 methotrexate by IV in 500 mL NS over four hours. Pa-tients at the BCCA would then receive 1000 mL 2/3 D5W–1/3 NSwith 20 mEq potassium chloride and 100 mEq sodium bicarbonate/Lat 200 mL/h IV for 20 h after the end of the methotrexate infusion.Twenty-four hours after the methotrexate infusion was started, leu-covorin was given orally, 9 doses of 15 mg each, every 6 h. This proto-col was then repeated every 14 days until two cycles post normalhCG.

Serum hCG levels were measured for all patients every two weeksat the time of chemotherapy to assess response. Treatment relatedcomplications, dose reductions or delays, and side effects were docu-mented, and side effects were retrospectively graded according to theNational Cancer Institute (NCI) Common Terminology Criteria for Ad-verse Events. Toxicities were not prospectively collected for the pur-poses of this study, therefore identification and grading of toxicitiesare based on clinic dictations, patient symptom questionnaires andnursing notes. The dose of dactinomycin was reduced to 0.5 mg/m2

in patients with stomatitis, and by 66% in patients with absolute neu-trophil counts (ANC) of 0.7–1.0×109/L. The dose of methotrexate wasreduced by 75%, 70%, and 30–50% in patients with calculated creati-nine clearance of 61–80 mL/min, 51–60 mL/min, and 10–50 mL/min,respectively. Dactinomycin was held for once cycle for an ANC ofb0.7, to be followed by a subsequent dose reduction of 66%. Metho-trexate was avoided for creatinine clearanceb10 mL/min.

All patients were followed for 12 months with serial hCGmonitor-ing (weekly for two months post-normalization then monthly until12 months). Patients were counseled to avoid pregnancy duringtreatment and follow-up, with the choice of contraceptive methodleft to the patient and her primary care physician.

Ethics and institutional review board approval was obtained fromthe CCI in Edmonton, Alberta and the University of British Columbia/BCCA Research Ethics Board in Vancouver, British Columbia for theperformance of this study.

Descriptive statistics were obtained for clinical parameters includ-ing age, tumor stage, initial (i.e. prior to uterine evacuation) and pre-treatment (i.e. prior to chemotherapy) levels of hCG, molar histologyand cytogenetics (complete vs. partial mole), cycles to normal hCG,

noitartsinimdAesoDgurD

Dactinomycin 0.6 mg/m 2 IV on days 1 and 2 IV push

100 mg/m 2 IV on day 1 (after

dactinomycin)

IV push Methotrexate

300 mg/m 2 IV infusion (after

IV push methotrexate)

IV in 500 mL NS over 4 hours

Leucovorin 15 mg po q6hrs x 9 doses Start 24 hours after start of

methotrexate infusion

Fig. 1. British Columbia Cancer Agency and Alberta Cross Cancer Institute Chemotherapy protocol for low-risk gestational trophoblastic neoplasia. Repeat every 14 days until 2 cyclespost normal human chorionic gonadotropin.No monitoring of methotrexate levels.

555L. Eiriksson et al. / Gynecologic Oncology 124 (2012) 553–557

and toxicities. Means were obtained for continuous variables and fre-quencies (%) were obtained for categorical variables. Univariate andmultivariate analyses were used to identify factors associated withthe number of cycles required to achieve remission. Logistic regres-sion was used to identify factors associated with treatment-relatedtoxicities. A p-value of b0.1 was used for statistical significance onunivariate analysis only, while a p-value of b0.05 was used for allother statistical significance. SAS® 9.1.3 (Cary, NC: SAS InstituteInc.) was used to conduct all statistical analyses.

Results

The average age was 29, ranging between 15 and 46 years. Onepatient had a previous molar pregnancy and one patient had a previ-ous pregnancy complicated by triploidy. Antecedent pregnancies in-cluded molar pregnancies (n=97), ectopic pregnancies (n=2), anda therapeutic abortion (n=1). Of the 100 cases, 75 were establishedto be complete moles and 9 were partial moles. Twenty-three pa-tients had metastatic disease, all but one involving the lungs. The pa-tient with extra-pulmonary metastasis had disease in the adnexa. The

GTN(n = 221)

High Risk GTN(n = 27)

Incomplete/NotApplicable (n = 50)

Choriocarcinoma(n = 16)

Placental Site Tropho-blastic Tumour (n = 7)

High Risk GTN(n = 4)

Trea

Prtr

Ch

Spono

Fig. 2. Consort

median pre-treatment hCG was 5 400 mIU/mL, although a wide rangeof values was seen, from 6.7 to 636 000 mIU/mL.

The median number of cycles required to achieve normalization ofhCG was 3 (Fig. 3) and the mean was 3.4. Sixty percent of patientsachieved a negative hCG with three or fewer cycles, while seventy-nine percent had achieved a negative hCG by cycle 4. The medianand mean total number of cycles received were 5 and 5.3, respective-ly. There were 2 patients who required second-line chemotherapydue to hCG non-normalization. Subsequent cure was achieved withthe GOTDHR (MACE) protocol (methotrexate, dactinomycin, cisplat-in, etoposide; www.bccancer.bc.ca). One patient elected to undergohysterectomy following three cycles of treatment. One patient recurred8 months after hCG normalization, following 4 cycles of chemotherapy.Dilatation and curettage was performed with a pre-evacuation hCG of67 000. Pathology revealed a complete mole. Spontaneous regressionof hCG took place without further treatment.

Twelve and eight patients experienced grade 3 or grade 4 hemato-logic toxicities, respectively. Hematologic toxicity resulted in 9 dosereductions and 7 treatment delays, typically of one week duration.Grade 2 and grade 3 stomatitis and mucositis were noted in 44 and3 patients, respectively. This resulted in 19 dose reductions but no

Low Risk GTN (n = 144)

tment at AlternateSite (n = 7)

egnancy duringeatment (n = 3)

arts unavailable(n = 17)

taneous regressionf hCG (n = 23)

Primary hysterectomy(n = 4)

Surgical resection (n = 1)

Alternate low-risk treatment (n = 39)

Combined chemo-therapy (n = 100)

diagram.

0

5

10

15

20

25

30

1 2 3 4 5 6 7 8 9 10 11

Fig. 3. Cycles to normal hCG.

Table 3Association between patient characteristics and number of cycles requiredto achieve normal hCG.

Significance (p value)

Univariate analysisInitial hCG 0.078Pre-treatment hCG b0.001Metastases 0.031Tumor histology⁎ 0.421Age 0.273

Multivariate analysisInitial hCG 0.035Pre-treatment hCG b0.001Metastases 0.122

⁎ Complete vs. partial mole

556 L. Eiriksson et al. / Gynecologic Oncology 124 (2012) 553–557

treatment delays. The presence of alopecia was not consistentlyrecorded for all patients, but was not found to be frequent or severe.The majority of nausea (40%) was grade 1 and no grade 3 or 4 nausea/emesis was noted (Fig. 4).

Using univariate analysis, initial hCG (p=0.078), pre-treatmenthCG (pb0.001) and the presence of metastases (p=0.031) were sig-nificantly associated with the number of cycles required to achievenormalization of hCG. On multivariate analysis, only initial hCG(p=0.035) and pre-treatment hCG (pb0.001) reached significance,as shown in Table 3.

Discussion

Multiple chemotherapy regimens are available for the treatmentof low-risk GTN (Table 2). At least 14 have been published in the En-glish literature in the past twenty years [6]. Most protocols use eithersingle-agent methotrexate or single-agent dactinomycin, althoughtreatments with fluorouracil or etoposide are also described. Typical-ly, the choice of regimen depends on the region, institution, and expe-rience or familiarity with the chemotherapeutic agents and regimens.Few randomized trials have been performed to date, with the major-ity of data obtained from retrospective analyses of single-institutionexperiences [6]. Although the ultimate cure rates with chemotherapyapproach 100%, the most efficacious and least toxic regimen has yet tobe identified. Recent data suggest single-agent regimens may requiremultiple cycles of treatment, but achieve modest cure rates, often ne-cessitating the use of additional alternative drugs. Cure rates withfirst-line methotrexate are reported to be between 50 and 75%[7–12], while cure rates with first-line dactinomycin range between61 and 90% [7–9,12,13]. Even the most recent randomized phase IIItrial, demonstrating the superiority of pulsed dactinomycin overweekly methotrexate, showed suboptimal cure rates of 70% and

0

5

10

15

20

25

30

35

40

45

50

%

grade 1 grade 2 grade 3 grade 4

hematologic

stomatitis

alopecia

nausea/vomiting

fatigue

abdominal pain

Fig. 4. Toxicities.

53%, respectively, using single-agent chemotherapy for low-riskGTN [14]. At the Alberta CCI and BCCA, the combination chemothera-py protocol for the treatment of low-risk GTN administers dactino-mycin followed by methotrexate and folinic acid as delineated inFig. 1. The rationale for this protocol was the anecdotal poor responserates with single-agent chemotherapy, sometimes requiring an ex-tended duration of treatment, imposing a longer period of disabilityin a group of women often anxious to become pregnant again soon.This combined chemotherapy protocol is beneficial if it indeed has ashorter duration of treatment. As low-risk GTN is a curable disease,treatment goals include high cure rates, short duration of treatment,and minimal toxicity.

This study highlights the efficacy of double-agent chemotherapy.A 98% primary cure rate was achieved, with a median of 3 cycles re-quired to reach normal quantitative hCG values. On multivariate anal-ysis it was noted that initial and pre-treatment hCG levels werepredictive of the length of treatment required. This may prove use-ful in providing patients with an individualized estimate of thenumber of chemotherapy cycles necessary. Although a 12% inci-dence of grade 3 and 4 hematologic toxicities was encountered, re-quiring dose reductions or brief treatment delays, there were notoxicity-related hospital admissions. Grade 2 and 3 stomatitis andmucositis were noted, but required only dose reductions, withouttreatment delays.

Hospital admission may be seen as an advantage in this two-daytreatment, since guaranteed intravenous hydration allows a higherdose of methotrexate to be given without undue toxicity. Given ade-quate oral fluid intake, however, this protocol can be administered asan outpatient. While overall cure rates in the treatment of GTN maybe high, the necessity for second-line treatment when first-linesingle-agents fail is not only costly to the health care system, butalso carries psychological implications for the patient. Although toxic-ities may be increased using combined methotrexate and dactinomy-cin, the issues of mucositis/stomatitis and myelosuppression tend toresolve with dose reductions, typically of dactinomycin, with pre-served high cure rates. The toxicities encountered would have to beevaluated in relation to patient satisfaction with the short durationof treatment. The majority of these women are young and healthyand toxicities are tolerated and accepted.

Ours is not the first report of combined chemotherapy in the treat-ment of low-risk GTN. Other groups have reported on other combina-tions of chemotherapy. A double-agent regimen, alternating a five-daytreatment of methotrexate with a five-day treatment of dactinomycinevery two weeks, was first reported in 1975, with response rates of 79to 100% and a median of 3 to 5 cycles required to achieve remission[15]. In 1989 Rose described 9 patients who received alternatingmetho-trexate and dactinomycinwith a 100% response rate and amedian offivecycles to achieve remission [16]. More recently, a retrospective compar-ison reviewed 24 patients who received combined methotrexate and

557L. Eiriksson et al. / Gynecologic Oncology 124 (2012) 553–557

dactinomycin versus single agent methotrexate and single agent dacti-nomycin [7]. Remission rates were 79%, 69%, and 71%, respectively(p=0.7) with similar durations of treatment and number of cycles re-quired to achieve remission (median of 3 cycles). With this protocol,however, 63% of patients receiving combinedmethotrexate and dactino-mycin experienced side effects, compared to 29% and 19% with singleagent methotrexate and dactinomycin, respectively (p=0.0003). Al-though the specific toxic effectswere not listed, grade 3 and 4 side effectswere documented in 8.3% of patients receiving combined treatment ver-sus 4.7% and 0% receiving single agent methotrexate and dactinomycin,respectively. Of note, this study was not randomized, folinic acid rescuewas not administered with this protocol, a five-day treatment protocolof methotrexate, dactinomycin, or both, was used, and patient baselinecharacteristics were unequal prior to treatment.

Compared to these previous studies, this paper reports the largestseries in published literature on double-agent chemotherapy in thetreatment of low-risk GTN, and the first published on this particularprotocol. Although retrospective, this study analyzed two population-based cohorts, thereby diminishing the effect of selection bias. Patientswith FIGO risk score 0–6were included, as in previously published stud-ies of chemotherapy in low-risk GTN [7–8,12,14] with a comparableproportion of patients with FIGO risk score 5–6 [14]. At both sitesthere were no competing trials, such that all eligible patients at theBCCA and the majority of eligible patients at the Alberta CCI were of-fered this treatment protocol. It is therefore anticipated that this datashould be generalizable to diverse patient groups. Initial cure rateswere excellent (98%) and treatment duration was short. Although tox-icities may be somewhat higher than noted in single-agent protocols,this regimen offers women a consistent and predictable treatment out-come, with reliable and rapid primary cure, despite its transient side ef-fects. At the individual level, single-agent chemotherapy may bepreferred. However, when failures are considered, the duration andcomplexity of treatment increases. Therefore, a two-drug regimenmay be more effective at the population level, particularly when thereis no reliable way to predict which patients will fail first-line treatment.

The results of this large, population-based experience are impres-sive, and consideration should be made for the adoption of this proto-col, with careful monitoring of patient outcomes. The observedeffectiveness of combined chemotherapy should encourage the de-velopment of alternative two-drug regimens of predictably less toxic-ity, deliverable in a fully ambulatory setting. The substantial activityof many individual chemotherapy agents offers considerable roomfor such development, and the potential to avoid toxicities such as al-opecia and mucositis. Protocols with lesser toxicities may permitmore convenient single-day schedules while maintaining primarycure rates. Although currently available protocols in the treatmentof low-risk gestational trophoblastic neoplasia are successful inachieving cure, this should not discourage the pursuit of improvedstrategies. In future evaluations of such protocols it may be interest-

ing to review pre- and post-treatment patient questionnaires to de-termine patient preferences.

Conflict of interest statementThe authors had no financial or personal relationships with other people or organiza-tions that could inappropriately influence or bias their work.

Acknowledgments

1. Sunita Ghosh, PhD, PStat, Assistant Clinical Professor, Depart-ment of Oncology, University of Alberta, Senior Research Biostatistician,Alberta Health Services — Cancer Care

2. Medical Records, CCI and BCCA3. Department of Pathology, CCI and BCCA

References

[1] Soper JT. Gestational trophoblastic disease. Obstet Gynecol 2006;108(1):176–87.[2] Smith HO. Gestational trophoblastic disease epidemiology and trends. Clin Obstet

Gynecol 2003;46(3):541–56.[3] Wolfberg AJ, Feltmate C, Goldstein DP, Berkowitz RS, Lieberman E. Low risk of re-

lapse after achieving undetectable HCG levels in women with complete molarpregnancy. Obstet Gynecol 2004;104(3):551–4.

[4] Goto S, Yamada A, Ishizuka T, Tomoda Y. Development of postmolar trophoblasticdisease after partial molar pregnancy. Gynecol Oncol 1993;48(2):165–70.

[5] Kohorn EI. The new FIGO 2000 staging and risk factor scoring system for gesta-tional trophoblastic disease: description and critical assessment. Int J GynecolCancer 2001;11(1):73–7.

[6] Alazzam M, Tidy J, Hancock BW, Osborne R. First line chemotherapy in low riskgestational trophoblastic neoplasia. Cochrane Database Syst Rev Jan 21 2009;1:CD007102.

[7] Abrão RA, de Andrade JM, Tiezzi DG, Marana HR, Reis FJ Candido dos, Clagnan WS.Treatment of low-risk gestational trophoblastic disease: comparison of single-agent methotrexate, dactinomycin and combination regimens. Gynecol Oncol2008;108(1):149–53.

[8] Gilani MM, Yarandi F, Eftekhar Z, Hanjani P. Comparison of pulse methotrexateand pulse dactinomycin in the treatment of low-risk gestational trophoblasticneoplasia. Aust N Z J Obstet Gynaecol 2005;45(2):161–4.

[9] Lertkhachonsuk AA, Israngura N, Wilailak S, Tangtrakul S. Actinomycin d versusmethotrexate–folinic acid as the treatment of stage I, low-risk gestational tropho-blastic neoplasia: a randomized controlled trial. Int J Gynecol Cancer 2009;19(5):985–8.

[10] Smith EB, Weed Jr JC, Tyrey L, Hammond CB. Treatment of nonmetastatic gesta-tional trophoblastic disease: results of methotrexate alone versus methotrexate-folinic acid. Am J Obstet Gynecol 1982;144(1):88–92.

[11] Wong LC, Choo YC, Ma HK. Methotrexate and citrovorum factor rescue in gesta-tional trophoblastic disease. Am J Obstet Gynecol 1985;152(1):59–62.

[12] Yarandi F, Eftekhar A, Shojaei H, Kanani S, Sharifi A, Hanjani P. Pulse methotrexateversus pulse actinomycin D in the treatment of low-risk gestational trophoblasticneoplasia. Int J Gynaecol Obstet 2008;103(1):33–7.

[13] Kohorn EI. Decision making for chemotherapy administration in patients withlow risk gestational trophoblastic neoplasia. Int J Gynecol Cancer 1996;6(4):279–85.

[14] Osborne RJ, Filiaci V, Schink JC, et al. Phase III trial of weekly methotrexate orpulsed dactinomycin for low-risk gestational trophoblastic neoplasia: a gyneco-logic oncology group study. J Clin Oncol 2011;29(7):825–31.

[15] Smith JP. Chemotherapy in gynecologic cancer. Clin Obstet Gynecol 1975;18(4):109–24.

[16] Rose PG, Piver MS. Alternating methotrexate and dactinomycin in nonmetastaticgestational trophoblastic disease. J Surg Oncol 1989;41(3):148–52.