Embed Size (px)
Citation preview
Gynecologic Oncology 112 (2009) 353–357
Contents lists available at ScienceDirect
Gynecologic Oncology
j ourna l homepage: www.e lsev ie r.com/ locate /ygyno
Evaluating methotrexate treatment in patients with low-risk postmolar gestationaltrophoblastic neoplasia☆
Whitfield B. Growdon a,e,f, Adam J. Wolfberg b, Donald P. Goldstein c,d,e,f, Colleen M. Feltmate c,d,e,f,Manuel E. Chinchilla c,f, Ellice S. Lieberman c,f, Ross S. Berkowitz c,d,e,f,⁎a Department of Obstetrics and Gynecology, Division of Gynecologic Oncology, Massachusetts General Hospital, Boston, USAb Department of Obstetrics and Gynecology, Division of Maternal-Fetal Medicine, Tufts- New England Medical Center, Boston, USAc Department of Obstetrics and Gynecology, Division of Gynecologic Oncology, Brigham and Women's Hospital, Boston, USAd New England Trophoblastic Disease Center, Donald P. Goldstein M.D. Tumor Registry, Boston, USAe Dana Farber Cancer Institutue/Harvard Cancer Center, Boston, USAf Harvard Medical School, Boston, USA
☆ This manuscript was presented at the 14th WoTrophoblastic Diseases, Fukuoka, Japan, November 11, 2⁎ Corresponding author. Division of Gynecologic Onc
Brigham and Women's Hospital, ASBI, 75 Francis Street,E-mail address: [email protected] (R.S. Berko
0090-8258/$ – see front matter © 2008 Elsevier Inc. Aldoi:10.1016/j.ygyno.2008.11.003
a b s t r a c t
a r t i c l e i n f oArticle history:
Objective. To identify cli Received 25 August 2008Available online 6 December 2008Keywords:Low risk gestational trophoblastic diseaseSingle agent methotrexatePersistent GTN
nical factors associated with requiring more than a single course of Methotrexate(MTX) to achieve remission amongwomenwith low-risk postmolar gestational trophoblastic neoplasia (GTN).
Methods. We studied 150 womenwith persistent GTN after diagnosis of complete (n=110) or partial mole(n=40) to identify possible predictors of requiring additional treatment after a single treatment ofmethotrexate (MTX). All women had low-risk disease using FIGO and WHO scoring systems.
Results. Seventy women (47%) required additional courses of chemotherapy, of whom 45 (64%) receivedchemotherapy other than MTX. Multivariate analysis revealed that complete mole histology, presence ofmetastasis, single day MTX infusion and any increase in serum beta human chorionic gonadotropin (β-hCG)level 1 week after MTX therapy were independent predictors of requiring additional MTX or alternativechemotherapy. Dilatation and curettage (D+C) within 1 week after the diagnosis of persistence did not affectfuture chemotherapy requirements (pN0.64). Following completemole,β-hCG levels N2000mIU/mL at 1weekpost MTX were associated with a 89% risk of additional cycles chemotherapy including MTX and a 65% risk ofalternative chemotherapy.
Conclusions. Metastatic disease, MTX infusion protocol and complete mole histology were independentlyassociated with the need for additional chemotherapy after an initial course of MTX for women with low riskGTN. D+C at persistence did not alter the chemotherapy requirement. Elevated β-hCG level at 1 week after theinitial course of MTX was also an independent factor predicting the need for additional courses of MTX oralternative chemotherapy.
© 2008 Elsevier Inc. All rights reserved.
Introduction
Gestational trophoblastic disease (GTD) is an entity that encom-passes a spectrum of both benign andmalignant conditions. Hydatidi-form mole accounts for the benign expression of GTD and affectsapproximately 1 in 1000 pregnancies in the United States [1].Management of both partial mole and complete mole is similar:surgical evacuation followed by weekly gonadotropin follow-up untilhCG levels become undetectable. Patients are then followed for 6months to ensure hCG levels do not re-elevate. An increase or aplateau in the hCG level during the follow-up period defines
rld Congress on Gestational007.ology, Department of OBGYN,Boston, MA, 02115, USA.witz).
l rights reserved.
persistent gestational trophoblastic neoplasia (GTN). [2] Whilewomen with complete moles have a 15–29% risk of GTN [3,4], thosewith partial mole have only a 1–5% chance of developing persistentGTN [5–8] Persistent GTN after molar pregnancy is an indication forchemotherapeutic treatment [9].
Choice of a chemotherapeutic regimen is based upon thecharacteristics of the presenting GTN. Patients are stratified into lowand high risk GTN by combining the International Federation ofObstetrics and Gynecology (FIGO) stage with the World HealthOrganization (WHO) prognostic scoring system. Low risk GTN isdefined by a FIGO stage I-III with a WHO score ≤6, and high risk isdefined by either FIGO stage IV or any stage with a WHO score ≥7 [2].Single-agent therapy is indicated for low risk GTN, while multi-agentchemotherapies are utilized for high risk GTN.
In the United States, the majority of persistent GTN is low risk andtherefore best treated with single-agent chemotherapy [10,11].Methotrexate is a commonly utilized agent for monotherapy. [12]
354 W.B. Growdon et al. / Gynecologic Oncology 112 (2009) 353–357
Past studies have shown a range of success for different MTX protocolswith 48–92%of patients who receive methotrexate for low risk GTNgoing into remission as defined by 1 year of undetectable hCG levels[13,14]. However, few studies have carefully examined potentialclinical risk factors which may be associated with the requirementfor additional courses of MTX or alternative chemotherapy after initialMTX therapy. The current investigation was undertaken to identifypossible predictors of requiring more than one course of MTX toachieve remission in patients with low risk GTN.
Materials and methods
The Donald P. Goldstein M.D. Trophoblastic Tumor Registry of theNew England Trophoblastic Disease Center includes information on1801 women diagnosed with molar pregnancies on the basis ofpathologic evaluation between 1973 and 2003. Within this group,1443 patients completed all recommended follow up. Of thatpopulation, 199 met the FIGO criteria for GTN and had an hCG plateauof ±10% over the course of at least 3 weeks, or exhibited a rise of β-hCGN10% over the course of 2 weeks. All patients with evidence ofpersistence underwent CXR. Chest CT scan was performed in manypatients with persistence. If chest CT demonstrated evidence ofpulmonary disease, the patient was considered to have stage IIIdisease even if the CXR was normal. Patients with pulmonaryinvolvement, liver function abnormalities or regional symptomsunderwent MRI or CT scan of the brain, abdomen and pelvis. D+C atpersistence was defined as occurring within 7 days of the date that apatient's β-hCG level demonstrated the diagnosis of persistence.Remission was declared if a patient's hCG level became undetectableand remained that way for 1 year. Low risk GTN was defined as a FIGOstage I, II, or III with a WHO score ≤6.
The final study population of 150 patients was identified after theexclusion of 7 patients who were diagnosed with high risk GTN, 8patients who underwent hysterectomy prior to chemotherapy, 17
Table 1Characteristics of women requiring additional (A) courses of chemotherapy and (B) alterna
A. Courses of chemotherapy p valu
1 N1
Presenting B-hCG 364726.4444 321317.2105 0.39
HistologyPartial 31 (77%) 9 (23%)Complete 49 (45%) 61 (55%) b0.00
MTX regimenInfusion 39 (47%) 44 (53%)8-day 41 (61%) 26 (39%) 0.08
D+C at persistenceYes 60 (53%) 53 (47%)No 20 (54%) 17 (46%) 0.92
Metastatic diseaseYes 2 (12%) 14 (88%)No 76 (58%) 56 (42%) b0.00
Average age 27.52 30.00 0.31
Age N35Yes 11 (42%) 15 (58%)No 69 (55%) 55 (45%) 0.22
Prior moleYes 2 (50%) 2 (50%)No 75 (53%) 66 (47%) 0.9
SNDYes 13 (50%) 13 (50%)No 27 (52%) 23 (48%) 0.86
patients who received another chemotherapy other than methotrex-ate as first line, 7 patients with incomplete clinical information and 10patients who met criteria for GTN, but had no chemotherapy given atour center. The two MTX protocols utilized over this time period inthis cohort consisted of a one-day infusion protocol (n=83) and aneight day protocol (n=67). The one-day MTX infusion included a bolusof 100mg/m2 over 1 h, followed by an infusion of 200mg/m2 over thenext 12 h with folinic acid [15]. The eight-day protocol consisted of 4administrations of MTX every other day at 1 mg/kg with 0.1 mg/kgfolinic acid in the intervening days, similar to other protocols reported[16,17]. After the initial course of MTX, no further chemotherapy wasadministered as long as the β-hCG level was progressively falling. Asecond course of chemotherapy was administered if the β-hCG levelplateaued for more than three consecutive weeks, re-elevated or theβ-hCG level did not decline by 1 log within 18 days after thecompletion of the initial treatment. [18] Patients resistant to MTXwere then treated with either actinomycin-D and/or multi-agentregimens (methotrexate, dactinomycin and cyclophosphamide;etoposide, methotrexate, dactinomycin, cyclophosphamide and vin-cristine) depending on the presenting WHO score and metastaticstatus.
Clinical data from these 150 patients including age, uterine size,gestational age at evacuation, history of prior mole, type of MTXprotocol, molar histology, use of repeat dilatation and curettage (D+C),persistence, presence of metastasis and hCG values at persistence andupon follow up were recorded for statistical analysis. D+C wasperformed on 113 patients in this cohort as therapy for symptomaticuterine bleeding and imaging suggestive of significant molar diseasein the uterine cavity. Most of the patients in our database werefollowed using an hCG assay with a sensitivity of 5mIU/mL, although afew patients cared for prior to 1990 were followed with assayssensitive to 10 mIU/mL. We used chi-square or Fisher's exact test asappropriate for categorical data in our analysis, and t-test statistics forcontinuous variables. β-hCG was grouped into into 3 categories
tive chemotherapies other than MTX
e B. Types of chemotherapy p value
MTX alone Alternate chemotherapy
3 325562.46 377586.7917 0.343
34 (85%) 6 (15%)1 71 (65%) 39 (35%) 0.016
48 (58%) 35 (42%)3 57 (85%) 10 (15%) b0.001
78 (69%) 35 (31%)27 (73%) 10 (27%) 0.64
6 (37%) 10 (63%)1 97 (74%) 35 (26%) 0.003
27.56 28.87 0.31
14 (53%) 12 (47%)91 (73%) 33 (27%) 0.048
3 (75%) 1 (25%)98 (70%) 43 (30%) 0.831
16 (63%) 10 (37%)7 40 (79%) 10 (21%) 0.134
Table 3Frequency table stratified by β-hCG 1 week following initial dose of MTX presenting (A)the risk of requiring additional chemotherapy treatments including MTX and (B) therisk of requiring alternative chemotherapy
A. Risk of additional courses of chemotherapy
hCG(mIU/ml)
All molar pregnancies(percent and 95% CI)
Complete moles(percent and 95% CI)
Partial Moles(percent and 95% CI)
b50 20.4 (10.5–34.3) 23.5 (6.8–49.9) 18.7 (7.2–36.6)(n=49) (n=17) (n=32)
50–99 37.5 (8.2–75.3) 42.9 (9.9– 81.6) 0.0 (0.0– 79.0)(n=8) (n=7) (n=1)
100–199 35.3 (14.2–61.67) 40.0 (16.6–67.3) 0.0 (0.0–65.7)(n=17) (n=15) (n=2)
200–1999 55.3 (40.1–69.8) 53.5 (37.6–68.8) 75.0 (19.4–99.4(n=47) (n=43) (n=4)
≥2000 86.2 (68.3–96.1.6) 89 (71.8–97.7) 0.0 (0.0–79.3)(n=29) (n=28) (n=1)
Total 46.7 (38.5–55.0) 55.5 (45.6–64.8) 22.1 (11.4–38.9)(n=150) (n=110) (n=40)
B. Risk of requiring alternative chemotherapy
hCG(mIU/ml)
All molar pregnancies(percent and 95% CI)
Complete moles(percent and 95% CI)
Partial Moles(percent and 95% CI)
b50 10.2 (3.4–22.3) 5.8 (0.15–28.4) 12.5 (3.5–29.0)(n=49) (n=17) (n=32)
50–99 25.0 (3.2–65.09) 28.6 (3.7– 71.0) 0.0 (0.0– 79.3)(n=8) (n=7) (n=1)
100–199 17.7 (3.8 – 43.4) 20.0 (4.33 – 48.1) 0.0 (0.0–66.0.3)(n=17) (n=15) (n=2)
200–1999 36.2 (22.7–51.5) 35.9 (20.8–51.0) 50.0 (6.76–93.2)(n=47) (n=43) (n=4)
≥2000 62 (42.3–79.3) 64.29 (44.3–81.4) 0.0 (0.0–79.3)(n=29) (n=28) (n=1)
Total 30.0 (22.9–38.1) 35.1 (27.0–45.2) 13.2 (2.4–23.9)(n=150) (n=110) (n=40)
355W.B. Growdon et al. / Gynecologic Oncology 112 (2009) 353–357
(b50 mIU/mL, 50 mIU/mL −1999 mIU/mL and N2000 mIU/mL) forsome analyses as this best modeled the data into triptiles. Multivariateanalysis was performed using logistic regression fromwhich adjustedodds ratios and 95% confidence intervals were calculated. All analyseswere performed using version 8.0 software from SAS Institute, Inc.
This studywas approved by the Human Research Committee at ourinstitution.
Results
Out of 1419 molar pregnancies identified with complete follow upinformation, 199 developed GTN (14%), with a complete molepersistence rate of 15.0% (153/1029) and a partial mole persistencerate of 5.6% (22/390). Twenty-four additional patients with GTN afterpartial molewere identified but not included in the persistence rate astheywere referred to our institution after the diagnosis of GTN.Withinthis cohort, low-risk GTN accounted for 95% (n=145/153) of GTN aftercomplete mole, and 98% (n=40/46) of GTN after partial mole. Our finalstudy populationwas limited to the 150 womenwho receivedMTX forprimary therapy of GTN. This group represented 75% (n=110) ofcomplete and 89% (n=40) of the partial mole cases respectively. Allpatients who initially received MTX therapy ultimately achievedremission with 70 women (47%) requiring additional courses ofchemotherapy including MTX, and 45 of these women (30%) requiringalternate types of chemotherapy after any amount of MTX.
Within this low risk group of 150 women who initially receivedMTX, clinical factors such asmedianβ-hCG level at presentation, historyof prior mole, uterine size greater than dates, and the use of D+C atpersistence did not affect whether or not a patient required additionalMTX doses or alternate types of chemotherapy (Table 1). Maternal ageN35 years did confer a modest risk of requiring alternate chemotherapyregimens, though age was not a significant risk factor for additionalcycles of MTX chemotherapy. Median β-hCG levels for GTN followingpartial mole were significantly lower at both persistence and 1 weekafter chemotherapy compared to GTN following complete mole(pb0.002, data not shown). Complete molar histology and metastaticdisease were associated with an elevated risk of patients requiringadditional courses of MTX and alternate chemotherapy to achieveremission. The infusion MTX regimen initially used was also asignificant factor associated with the need for alternate chemotherapy.
Multivariate analysis was used to quantify the risk of requiringadditional cycles of any chemotherapy and the risk of alternate typesof chemotherapy after initial MTX (Table 2). At persistence, the use of
Table 2Logistic regression of variables that affect the risk of requiring (A) additional courses ofany chemotherapy including MTX and (B) additional courses with alternativechemotherapy
Multivariate analysis of variables affecting the risk of requiring additionalchemotherapy treatments
n OR
Use of infusion protocol 83 3.0 (95% CI 1.5–6.4)Complete mole 110 6.7 (95% CI 2.1–19.2)Elevated BHCGN2000 mIU/ml at persistence 54 1.2 (95% CI 0.2–6.7)Elevated BHCGN2000 mIU/ml week after MTX 29 13.84 (95% CI 3.4–57.0)Lack of D+C at persistence 37 1.4 (95% CI 0.6–3.2)Increase in HCG 1 week after MTX 22 12.5 (95% CI 3.2–49.3)
Multivariate analysis of variables affecting the risk of requiring alternate types ofchemotherapy
n OR
Use of infusion protocol 83 6.6 (95% CI 2.7 – 15.7)Complete mole 110 5.46 (95% CI 1.6 – 18.9)Elevated BHCGN2000 mIU/ml at persistence 54 1.2 (95% CI 0.2 – 9.2)Elevated BHCGN2000 mIU/ml week after MTX 29 9.1(95% CI 2.1 – 40.8)Lack of D+C at persistence 37 1.3 (95% CI 0.5 – 3.3)Increase in HCG 1 week after MTX 22 7.1 (95% CI 2.3 – 22.1)
the one-day infusionMTX protocol as well as complete mole histologywere independent prognostic factors predicting increased risk of bothfurther chemotherapy MTX cycles and need for alternate chemother-apy regimens. Notably, D+C performed at persistence did not affectfuture requirements for MTX or other chemotherapy. Logisticregression indicated that a β-hCG level N2000 mIU/mL 1 week afterinitial MTX was a powerful independent predictor of whether or not apatient would require additional cycles of any chemotherapy includ-ing MTX (OR 13.8) and alternate types (OR 9.1) of chemotherapy. Thislogistic regressionmodel also investigates whether or not any increasein β-hCG level following initial MTX is associated with an increase inthe chemotherapy requirements. Following initial MTX therapy,15% ofpatients (22/150) had β-hCG values that increased over 1 week. Of thepatients that required additional cycles of any chemotherapy, 27% hadβ-hCG levels that increased compared to 3% (pb0.001) of those thatwent into remission after one dose of MTX. The same analysis foralternate types of chemotherapy revealed that 40% of those requiringalternate types of chemotherapy, other than MTX, had β-hCG levelsthat increased compared to 9% (pb0.005) of those who went intoremission with MTX therapy alone. Multivariate analysis demon-strated that any increase in β-hCG over the first week was anindependent prognostic factor associated with a 12-fold risk ofadditional cycles of chemotherapy, and a 7-fold risk of requiringalternative agents other than MTX (Table 2).
Given the predictive value of β-hCG levels after initial MTXtherapy, a stratified model was created to describe the risk ofadditional therapies by level of post chemotherapy β-hCG (Table 3).The risk of requiring additional treatments and alternative che-motherapy increased with elevated β-hCG at 1 week after MTXtherapy. For all patients, β-hCG levels from 200–1999 mIU/mL 1 weekafter the initial course of MTX were associated with a 55% risk ofneeding additional MTX cycles, and a 36% risk of needing alternativechemotherapy. For GTN following complete mole, β-hCG levels
Table 4Logistic regression examining (A) the risk of requiring additional cycles ofchemotherapy and (B) the risk of requiring alternative chemotherapy for patientswith GTN after complete mole
Multivariate analysis examing risk of additional cycles of chemotherapy
n OR at persistence
Use of infusion protocol 52 2.7 (95% CI 1.2–6.3)Metastatic disease 15 13.5 (95% CI 1.7–111.1)Elevated BHCGN2000 mIU/ml at persistence 51 1.6 (95% CI 0.7–3.7)Lack of D+C at persistence 23 1.2 (95% CI 0.4–3.4)
Multivariate analysis examing risk of alternate chemotherapy
n OR at persistence
Use of infusion protocol 52 5.8 (95% CI 2.3–14.5)Metastatic disease 15 4.4 (95% CI 1.2–16.4)Elevated BHCGN2000 mIU/ml at persistence 51 1.9 (95% CI 0.8–4.6)Lack of D+C at persistence 23 1.1 (95% CI 0.3–3.5)
356 W.B. Growdon et al. / Gynecologic Oncology 112 (2009) 353–357
≥2000 mIU/mL at 1 week post MTX were associated with an 89% riskof additional cycles of MTX and a 64% risk of alternative types ofchemotherapy to achieve remission.
Patients with low-risk metastatic disease accounted for 11% (16/150) of the cohort, with 15 of these patients having had a completemole prior to GTN. A multivariate analysis limited to women withcomplete mole revealed that metastatic disease was independentlyassociated with a 13-fold increased risk of requiring additional cyclesof chemotherapy, and 4-fold risk of requiring alternate chemotherapyat the time of persistent disease (Table 4).
Discussion
In this investigation, we sought to identify prognostic factorsassociated with increased requirements for chemotherapy, both anyadditional chemotherapy and alternative chemotherapy, in womenwith low-risk GTN initially treated with two different MTX regimens.These data demonstrate that molar histology prior to GTN, metastaticdisease, type of MTX regimen and β-hCG level 1 week after one ofthese two MTX regimens were predictive of whether or not a patientwith low risk GTN required additional cycles of MTX and alternativechemotherapy.
Single-agent MTX therapy for low-risk GTN has been associatedwith a 65–90% remission rate, which is comparable to our 70%remission rate with MTX alone. [19,20] The two types of MTXregimens employed at our institution, single day infusion and 8-day,have previously been examined in prior low risk GTN cohorts withsignificant ranges in reported single agent efficacy, with a trendtowards improved single agent remission rates with the 8-dayregimen [13,15,17,20]. No studies to our knowledge have evaluatedthe two regimens together in retrospective, prospective or rando-mized studies. To understand the difference in efficacy between theseMTX regimens, this study evaluates the 8-day and infusion MTXregimens in a large cohort utilizing multivariate analysis. Wecontrolled for other variables, such as β-hCG, pre-GTNmolar histologyand the use of D+C as an adjunctive therapy. Our data suggest thatpatients who receive the infusion protocol were 3 and 6 times morelikely to require more courses of MTX chemotherapy and alternateregimens respectively.
Few studies have examined the impact of histology on the need foraddition chemotherapy for women with post-molar, low risk GTN[21]. Our results suggest that GTN after complete mole is 5–7 timesmore likely to require additional cycles of MTX or alternate types ofchemotherapy compared to GTN after partial mole. This effect isindependent of the type of MTX protocol utilized, the β-hCG level atthe time of persistence (before treatment) and the usage of D+C,suggesting either an innate biological difference or a variablesensitivity to MTX. These data suggest that GTN after complete mole
is not only more common, but also more refractory to MTX comparedto GTN after partial mole.
The logistic regression examining the predictive role of β-hCG1 week after MTX suggests it is an independent prognostic factorassociated with increased risk of requiring additional as well asalternate chemotherapy (Table 2). We found that when the entirecohort is examined as one group, the risk of requiring any additionalMTX chemotherapy cycles and alternative chemotherapy increases ina stepwise fashion with increasing levels of β-hCG 1 week after initialMTX therapy (Table 3). When the 110 patients with GTN aftercomplete mole are examined, the data indicate that as the β-hCG1 week after MTX increases, so does the risk of needing furtherchemotherapy. One can observe an 89% risk of requiring additionalMTX chemotherapy doses, and a 65% risk of alternate types ofchemotherapy when the β-hCG is N2000 mIU/ml after 1 week, as itwas in 25% of this cohort.
Previous studies have indicated that post-evacuation β-hCG valuescan be used to predict the risk of persistence for both complete andpartial moles [21,22], so it is not surprising that these postchemotherapy β-hCG values would have a similar predictive value.Others have recently used these post-chemotherapy β-hCG values toconstruct a serum normogram showing the decay of β-hCG for low-risk patients who went into remission with single agent MTXdemonstrating predictive value early in follow up. [23] We notedthat in some instances, the β-hCG value 1 week after initial MTXincreased, andmultivariate analysis demonstrated that any increase inβ-hCG conferred a 12- and 7-fold risk of requiring additionalchemotherapy treatment and alternative chemotherapy respectively.Clinically, our data suggest that post-chemotherapy β-hCG levels canbe used to identify a sub-population at increased risk of requiringmore chemotherapy. These levels provide a powerful tool to counselpatients about their likelihood of requiring additional chemotherapy.For patients with a high chance of needing more treatment,particularly those in geographic locations where resources for followup are limited, clinicians could utilize this risk stratification system tooffer more up front chemotherapy.
The presence of metastasis was also a predictor of the need foradditional MTX courses and alternative chemotherapy. Sixteenpatients presented with low risk metastatic disease with pulmonaryinvolvement. Because only one patient had metastasis after partialmole, multivariate analysis was limited to GTN after complete mole.Metastatic disease was a strong independent predictor of requiringadditional cycles (OR=13) and alternative chemotherapy (OR=4). Ofthe 14 patients with low risk metastatic GTN who required additionalcycles of chemotherapy, 10 subsequently required alternate che-motherapy indicating that metastatic disease is likely to be lesssensitive to MTX, and that more upfront chemotherapy should beemployed for this sub-population.
Controversy exists as to whether or not a D+C performed at thetime of persistence will shorten a patient's follow up course or affect apatient's need for chemotherapy [24]. Some investigators havereported that GTN can regress without any chemotherapy after re-evacuation [25,26]. Univariate and multivariate analyses of our datasuggests that D+C had no effect on whether or not a patient neededadditional chemotherapy treatments or alternative chemotherapy.
Many different chemotherapy regimens involving MTX anddactinomycin have been employed as primary single-agent treatmentfor post-molar GTNwith excellent and reasonably comparable results.[27,28] We believe the choice of an optimal regimen is influenced bymany factors including remission rate, toxicity, cost and logistics ofadministration. No study has prospectively compared all the commonsingle-agent regimens using MTX and dactinomycin in GTN. A recentGynecologic Oncology Group phase III randomized clinical trialcompared single-agent MTX at 30 mg/mm2 IM given weekly topulsed IV dactinomycin 1.25 mg/mm2 every 2 weeks. The preliminaryresults indicated a higher response rate with fewer chemotherapy
357W.B. Growdon et al. / Gynecologic Oncology 112 (2009) 353–357
cycles amongst the patients receiving dactinomycin compared to theutilized MTX regimen. [29] Despite this recent evidence, it remainsdifficult to suggest what is unquestionably the best single-agentregimen for primary therapy in low risk GTN. At our center, we arecurrently choosing the 8-day MTX regimen as our preferred singleagent therapy for this disease.
Despite the efforts to conduct multivariate analysis and control forconfounding variables, this retrospective study is subject to all thebiases inherent in observational studies. This data only applies topatients initially treated with one of these two MTX regimens. One ofthe other notable limitations involves the GTN following partial molegroup. The relatively small number of women in this group in ourstudy (N=40) limited our ability to perform multivariate analyses andsubgroup analyses according to post-MTX β-hCG level. The incidenceof GTN after partial mole has been reported to be 1–5% [5–8] withmetastatic disease being particularly uncommon. While we have alimited number of patients with GTN following partial mole, ourexperience represents one of the larger reports in the literature. Thislimited our ability to determine whether post-MTX β-hCG is a goodpredictor of subsequent chemotherapy requirements in those withGTN following partial mole.
Despite these limitations, our data do identify several importantpatient characteristics that predispose women to require additionalMTX cycles and alternative chemotherapy after initial MTX for lowrisk GTN. These factors include the use of the MTX infusion protocol,antecedent complete molar pregnancy, metastatic disease, and β-hCG levels N2,000 mIU/mL at 1 week after MTX therapy. We nowadvise our patients manifesting these risk factors at the outset oftreatment that they are more likely to require additional MTXchemotherapy and potentially alternative chemotherapy to achieveremission.Though D+C may aid in removing large intrauterine tumorvolume at persistence, our data suggest that D+C does not alterfuture chemotherapy requirements.
Conflict of interest statementNone of the authors has any conflict of interest regarding the publication of this originalresearch.
References
[1] Smith HO. Gestational trophoblastic disease epidemiology and trends. Clin ObstetGynecol 2003;46:541–56.
[2] Kohorn EI. Negotiating a staging and risk factor scoring system for gestationaltrophoblastic neoplasia. A progress report. J Reprod Med 2002;47:445–50.
[3] Wolfberg AJ, Feltmate C, Goldstein DP, Berkowitz RS, Lieberman E. Low risk ofrelapse after achieving undetectable HCG levels in women with complete molarpregnancy. Obstet Gynecol 2004;104:551–4.
[4] Berkowitz RS, Goldstein DP. Chorionic tumors. N Engl J Med 1996;335:1740–8.[5] Goto S, Yamada A, Ishizuka T, Tomoda Y. Development of postmolar trophoblastic
disease after partial molar pregnancy. Gynecol Oncol 1993;48:165–70.[6] Bagshawe KD, Lawler SD, Paradinas FJ, Dent J, Brown P, Boxer GM. Gestational
trophoblastic tumours following initial diagnosis of partial hydatidiform mole.Lancet 1990;335:1074–6.
[7] Feltmate CM, Growdon WB, Wolfberg AJ, Goldstein DP, Genest DR, Chinchilla ME,et al. Clinical characteristics of persistent gestational trophoblastic neoplasia afterpartial hydatidiform molar pregnancy. J Reprod Med 2006;51:902–6.
[8] Hancock BW, Nazir K, Everard JE. Persistent gestational trophoblastic neoplasiaafter partial hydatidiform mole incidence and outcome. J Reprod Med 2006;51:764–6.
[9] Berkowitz RS, Goldstein DP, Bernstein MR. Management of nonmetastatictrophoblastic tumors. J Reprod Med 1981;26:219–22.
[10] Soper JT. Gestational trophoblastic disease. Obstet Gynecol 2006;108:176–87.[11] Lurain JR, Brewer JI, Torok EE, Halpern B. Gestational trophoblastic disease:
treatment results at the brewer trophoblastic disease center. Obstet Gynecol 1982;60:354–60.
[12] Kohorn EI. Single-agent chemotherapy for nonmetastatic gestational trophoblasticneoplasia. Perspectives for the 21st century after three decades of use. J ReprodMed 1991;36:49–55.
[13] McNeish IA, Strickland S, Holden L, Rustin GJ, Foskett M, Seckl MJ, et al. Low-riskpersistent gestational trophoblastic disease: outcome after initial treatment withlow-dose methotrexate and folinic acid from 1992 to 2000. J Clin Oncol 2002;20:1838–44.
[14] Lurain JR, Elfstrand EP. Single-agent methotrexate chemotherapy for the treatmentof nonmetastatic gestational trophoblastic tumors. Am J Obstet Gynecol 1995;172:574–9.
[15] Garrett AP, Garner EO, Goldstein DP, Berkowitz RS. ethotrexate infusion and folinicacid as primary therapy for nonmetastatic and low-risk metastatic gestationaltrophoblastic tumors. 15 years of experience. J Reprod Med 2002;47:355–62.
[16] Soper JT, Clarke-Pearson DL, Berchuck A, Rodriguez G, Hammond CB. 5-daymethotrexate for women with metastatic gestational trophoblastic disease.Gynecol Oncol 1994;54:76–9.
[17] Berkowitz RS, Goldstein DP, Bernstein MR. Ten year's experience with methotrex-ate and folinic acid as primary therapy for gestational trophoblastic disease.Gynecol Oncol 1986;23:111–8.
[18] Berkowitz RS, Goldstein DP. Methotrexate with citrovorum factor rescue fornonmetastatic gestational trophoblastic neoplasms. Obstet Gynecol 1979;54:725–8.
[19] Matsui H, Suzuka K, Yamazawa K, Tanaka N, Mitsuhashi A, Seki K, et al. Relapse rateof patients with low-risk gestational trophoblastic tumor initially treated withsingle-agent chemotherapy. Gynecol Oncol 2005;96:616–20.
[20] Wong LC, Ngan HY, Cheng DK, Ng TY.Methotrexate infusion in low-risk gestationaltrophoblastic disease. Am J Obstet Gynecol 2000;183:1579–82.
[21] Growdon WB, Wolfberg AJ, Feltmate CM, Goldstein DP, Genest DR, Chinchilla ME,et al. Postevacuation hCG levels and risk of gestational trophoblastic neoplasiaamong women with partial molar pregnancies. J Reprod Med 2006;51:871–4.
[22] Wolfberg AJ, Berkowitz RS, Goldstein DP, Feltmate C, Lieberman E. PostevacuationhCG levels and risk of gestational trophoblastic neoplasia inwomenwith completemolar pregnancy. Obstet Gynecol 2005;106:548–52.
[23] van Trommel NE, Massuger LF, Schijf CP, ten Kate-Booij MJ, Sweep FC, Thomas CM.Early identification of resistance to first-line single-agent methotrexate in patientswith persistent trophoblastic disease. J Clin Oncol 2006;24:52–8.
[24] Goldstein DP, Garner EI, Feltmate CM, Berkowitz RS. The role of repeat uterineevacuation in the management of persistent gestational trophoblastic disease.Gynecol Oncol 2004;95:421–2.
[25] van Trommel NE, Massuger LF, Verheijen RH, Sweep FC, Thomas CM. The curativeeffect of a second curettage in persistent trophoblastic disease: a retrospectivecohort survey. Gynecol Oncol 2005;99:6–13.
[26] Pezeshki M, Hancock BW, Silcocks P, Everard JE, Coleman J, Gillespie AM, et al. Therole of repeat uterine evacuation in the management of persistent gestationaltrophoblastic disease. Gynecol Oncol 2004;95:423–9.
[27] Osborne R, Gerulath A. What is the best regimen for low-risk gestationaltrophoblastic neoplasia? A review. J Reprod Med 2004;49:602–16.
[28] Hoekstra AV, Lurain JR, Rademaker AW, Schink JC. Gestational trophoblasticneoplasia: treatment outcomes. Obstet Gynecol 2008;112:251–8.
[29] Osborne R, Filliaci V, Schink J, Mannel R, Provencher D, Alvarez-Secord A, et al. Arandomized phase III trial comparing weekly parenteral methotrexate and“pulsed” dactinomycin as primary management for low-risk gestational tropho-blastic neoplasia: a gynecologic oncology group study. Gynecol Oncol 2008;108:S1.
