Williams Obstetrics, 24th Edition

CHAPTER 20. Gestational Trophoblastic Disease









Gestational trophoblastic disease (GTD) is the term used to encompass a group of tumors typified by abnormal trophoblast proliferation. Trophoblast produce human chorionic gonadotropin (hCG), thus the measurement of this peptide hormone in serum is essential for GTD diagnosis, management, and surveillance. GTD histologically is divided into hydatidiform moles, which are characterized by the presence of villi, and nonmolar trophoblastic neoplasms, which lack villi.

Hydatidiform moles are excessively edematous immature placentas (Benirschke, 2012). These include the benign complete hydatidiform mole and partial hydatidiform mole and the malignant invasive mole. Invasive mole is deemed malignant due to its marked penetration into and destruction of the myometrium as well as its ability to metastasize.

Nonmolar trophoblastic neoplasms include choriocarcinoma, placental site trophoblastic tumor, and epithelioid trophoblastic tumor. These three are differentiated by the type of trophoblast they contain.

The malignant forms of gestational trophoblastic disease are termed gestational trophoblastic neoplasia (GTN). These include invasive mole, choriocarcinoma, placental site trophoblastic tumor, and epithelioid trophoblastic tumor. Other terms applied to GTN are malignant gestational trophoblastic disease and persistent gestational trophoblastic disease. These malignancies develop weeks or years following any type of pregnancy, but frequently occur after a hydatidiform mole.

Each of the GTN tumor types is histologically distinct and varies in its propensity to invade and metastasize. However, histological confirmation is typically not available. Instead, measurement of serum hCG levels combined with clinical findings—rather than a histological specimen—is used to diagnose and treat this malignancy. Accordingly, GTN is often identified and effectively treated as a group.

In the past, these metastatic tumors had a prohibitively high mortality rate. However, with chemotherapy, currently most tumors are highly curable (Goldstein, 2010). Early-stage GTN is typically cured with single-agent chemotherapy, whereas later-stage disease usually responds to combination chemotherapy.


The classic histological findings of molar pregnancy include villous stromal edema and trophoblast proliferation (Fig. 20-1). The degree of histological changes, karyotypic differences, and the absence or presence of embryonic elements are used to classify them as either complete or partial moles. These two also vary in associated risks for developing medical comorbidities and postevacuation GTN. Of the two, GTN more frequently follows complete hydatidiform mole.


FIGURE 20-1 Complete hydatidiform mole. A. Gross specimen with characteristic vesicles of variable size. (Image contributed by Dr. Brian Levenson.) B. Low-magnification photomicrograph shows generalized edema and cistern formation (black asterisks) within avascular villi. Haphazard trophoblastic hyperplasia is marked by a yellow asterisk on the right. (Image contributed by Dr. Erika Fong.)

A complete mole has abnormal chorionic villi that grossly appear as a mass of clear vesicles. These vary in size and often hang in clusters from thin pedicles. In contrast, a partial molar pregnancy has focal and less advanced hydatidiform changes and contains some fetal tissue. Although both forms of moles usually fill the uterine cavity, they rarely may be tubal or other forms of ectopic pregnancy (Sebire, 2005).

image Epidemiology and Risk Factors

There is an ethnic predisposition to hydatidiform mole, which has increased prevalence in Asians, Hispanics, and American Indians (Drake, 2006; Lee, 2011; Smith, 2006). The incidence in the United States and Europe has been relatively constant at 1 to 2 per 1000 deliveries (Lee, 2011; Lybol, 2011; Salehi, 2011).

The strongest risk factors are age and a history of prior hydatidiform mole. Women at both extremes of reproductive age are most vulnerable. Specifically, adolescents and women aged 36 to 40 years have a twofold risk, but those older than 40 have an almost tenfold risk (Altman, 2008; Sebire, 2002a). For those with a prior complete mole, the risk of another mole is 1.5 percent. With a previous partial mole, the rate is 2.7 percent (Garrett, 2008). After two prior molar pregnancies, Berkowitz and associates (1998) reported that 23 percent of women had a third mole.

image Pathogenesis

With rare exceptions, molar pregnancies arise from chromosomally abnormal fertilizations. Complete moles most often have a diploid chromosomal composition (Table 20-1). These usually are 46,XX and result from androgenesis, meaning both sets of chromosomes are paternal in origin. As shown in Figure 20-2A, an ovum is fertilized by a haploid sperm, which then duplicates its own chromosomes after meiosis. The chromosomes of the ovum are either absent or inactivated. Less commonly, the chromosomal pattern may be 46,XY or 46,XX and due to fertilization by two sperm, that is, dispermic fertilizationor dispermy (Lawler, 1991; Lipata, 2010).

TABLE 20-1. Features of Partial and Complete Hydatidiform Moles



FIGURE 20-2 Typical pathogenesis of complete and partial moles. A. A 46,XX complete mole may be formed if a 23,X-bearing haploid sperm penetrates a 23,X-containing haploid egg whose genes have been “inactivated.” Paternal chromosomes then duplicate to create a 46,XX diploid complement solely of paternal origin. B. A partial mole may be formed if two sperm—either 23,X- or 23,Y-bearing—both fertilize (dispermy) a 23,X-containing haploid egg whose genes have not been inactivated. The resulting fertilized egg is triploid with two chromosome sets being donated by the father (diandry).

Partial moles usually have a triploid karyotype—69,XXX, 69,XXY—or much less commonly, 69,XYY. These are each composed of two paternal haploid sets of chromosomes contributed by dispermy and one maternal haploid set (see Fig. 20-2B). Less frequently, a similar haploid egg may be fertilized by an unreduced diploid 46,XY sperm. These triploid zygotes result in some embryonic development, however, it ultimately is a lethal fetal condition. Fetuses that reach advanced ages have severe growth restriction, multiple congenital anomalies, or both.

Twin Pregnancy Comprised of a Normal Fetus and Coexistent Complete Mole

Rarely, in some twin pregnancies, one chromosomally normal fetus is paired with a complete diploid molar pregnancy. These are recognized in only 1 in 22,000 to 100,000 pregnancies (Steller, 1994). It is important that these cases be distinguished from a single partial molar pregnancy with its abnormal associated fetus. Amniocentesis done for fetal karyotyping is used to confirm the diagnosis.

There are a number of unique pregnancy complications with this twin pregnancy. And, many women may choose to terminate the pregnancy, if diagnosed early. In those with continuing pregnancy, survival of the normal fetus is variable and dependent on complications that commonly develop from the molar component. The most worrisome are preeclampsia or hemorrhage, which frequently necessitate preterm delivery. Wee and Jauniaux (2005) reviewed outcomes in 174 women of whom 82 chose termination. Of the remaining 92 pregnancies, 42 percent either miscarried or had a perinatal death; approximately 60 percent delivered preterm; and only 40 percent delivered at term.

Another concern for those continuing their pregnancy is the possible risk for developing subsequent GTN. Sebire and colleagues (2002b) reviewed such twin pregnancies and reported that in those not terminated, 21 percent of mothers subsequently required chemotherapy. But, this was not significantly different from a rate of 16 percent among women who chose termination. Others have reported rates up to 50 percent following continuation (Massardier, 2009). At this time, most data indicate that women with these twin pregnancies are not at greater risk for subsequent neoplasia than those with a singleton complete mole (Niemann, 2007b). Postdelivery surveillance is conducted as for any molar pregnancy and is discussed on page 401.

image Clinical Findings

The clinical presentation of women with a molar pregnancy has changed remarkably over the past several decades because prenatal care is sought much earlier and because sonography is virtually universal. As a result, most molar pregnancies are detected when they are small and before complications ensue (Kerkmeijer, 2009; Mangili, 2008).

Typically, there are usually 1 to 2 months of amenorrhea before discovery. In 41 women with a complete mole diagnosed at a mean of 10 weeks, Gemer and colleagues (2000) reported that 41 percent were asymptomatic and 58 percent had vaginal bleeding. Moreover, only 2 percent had anemia or hyperemesis, and none had other manifestations that in the past were common in these women.

As gestation advances, symptoms generally tend to be more pronounced with complete compared with partial moles (Niemann, 2007a). Untreated molar pregnancies will almost always cause uterine bleeding that varies from spotting to profuse hemorrhage. Bleeding may presage spontaneous molar abortion, but more often, it follows an intermittent course for weeks to months. In more advanced moles with considerable concealed uterine hemorrhage, moderate iron-deficiency anemia develops. Many women have uterine growth that is more rapid than expected. The enlarged uterus has a soft consistency, but typically no fetal heart motion is detected. Nausea and vomiting may become significant. The ovaries contain multiple theca-lutein cysts in 25 to 60 percent of women with a complete mole (Fig. 20-3). These likely result from overstimulation of lutein elements by sometimes massive amounts of hCG. Because theca-lutein cysts regress following pregnancy evacuation, expectant management is preferred. Occasionally a larger cyst may undergo torsion, infarction, and hemorrhage. However, oophorectomy is not performed unless there is extensive infarction that persists after untwisting.


FIGURE 20-3 Sonographic image of an ovary with theca-lutein cysts in a woman with a hydatidiform mole.

The thyrotropin-like effects of hCG frequently cause serum free thyroxine (fT4) levels to be elevated and thyroid-stimulating hormone (TSH) levels to be decreased. Despite this, clinically apparent thyrotoxicosis is unusual and, in our experience, can be mimicked by bleeding and sepsis from infected products. Moreover, the serum free T4 levels rapidly normalize after uterine evacuation. Despite this, a case of presumed “thyroid storm” has been reported (Moskovitz, 2010).

Severe preeclampsia and eclampsia are relatively common with large molar pregnancies. However, these are seldom seen today because of early diagnosis and evacuation. An exception is in the case of a normal fetus coexisting with a complete mole, described earlier. In those cases in which pregnancy is not terminated, severe preeclampsia frequently mandates preterm delivery. The predilection for preeclampsia is explained by the hypoxic trophoblastic mass, which releases antiangiogenic factors that activate endothelial damage (Chap. 40p. 733).

image Diagnosis

Most women initially have amenorrhea that is followed by irregular bleeding that almost always prompts pregnancy testing and sonography. Some women will present with spontaneous passage of molar tissue.

Serum β-HCG Measurements

With a complete molar pregnancy, serum β-hCG levels are commonly elevated above those expected for gestational age. With more advanced moles, values in the millions are not unusual. Importantly, these high values can lead to erroneous false-negative urine pregnancy test results because of oversaturation of the test assay by excessive β-hCG hormone (Chap. 9p. 169). In these cases, serum β-hCG determinations with or without sample dilution will clarify the conundrum. With a partial mole, β-hCG levels may also be significantly elevated, but more commonly concentrations fall into ranges expected for gestational age.


Although sonographic imaging is the mainstay of trophoblastic disease diagnosis, not all cases are confirmed initially. Sonographically, a complete mole appears as an echogenic uterine mass with numerous anechoic cystic spaces but without a fetus or amnionic sac. The appearance is often described as a “snowstorm” (Figure 20-4). A partial mole has features that include a thickened, multicystic placenta along with a fetus or at least fetal tissue. In early pregnancy, however, these sonographic characteristics are seen in fewer than half of hydatidiform moles (Fowler, 2006). The most common misdiagnosis is incomplete or missed abortion. Occasionally, molar pregnancy may be confused for a multifetal pregnancy or a uterine leiomyoma with cystic degeneration.


FIGURE 20-4 Sonograms of hydatidiform moles. A. Sagittal view of a uterus with a complete hydatidiform mole. The characteristic “snowstorm” appearance is due to an echogenic uterine mass that has numerous anechoic cystic spaces. Notably, a fetus and amnionic sac are absent. B. In this image of a partial hydatidiform mole, the fetus is seen above a multicystic placenta. (Image contributed by Dr. Elysia Moschos.)

Pathological Diagnosis

Surveillance for subsequent neoplasia following molar pregnancy is crucial. Thus, moles must be histologically distinguished from other types of pregnancy failure that have hydropic placental degeneration, which can mimic molar villous changes. Some distinguishing histological characteristics are shown in Table 20-1.

In pregnancies before 10 weeks, classic molar changes may not be apparent because villi may not be enlarged and molar stroma may not yet be edematous and avascular (Paradinas, 1996). In such situations, other techniques are used to differentiate. One takes advantage of the differing ploidy to distinguish partial (triploid) moles from diploid entities. Complete moles and nonmolar pregnancies with hydropic placental degeneration are both diploid.

Another technique involves histological immunostaining to identify the p57KIP2 nuclear protein. Because the gene that expresses p57KIP2 is paternally imprinted, only maternally donated genes are expressed. Because complete moles contain only paternal genetic material, they cannot express this gene; do not produce p57KIP2; and thus, do not pick up this immunostain. In contrast, this nuclear protein is strongly expressed in partial moles and in nonmolar pregnancies with hydropic change (Castrillon, 2001). As a result, the combined use of ploidy analysis and p57KIP2 immunostaining can be used to differentiate: (1) a complete mole (diploid/p57KIP2-negative), (2) a partial mole (triploid/p57KIP2-positive), and spontaneous abortion with hydropic placental degeneration (diploid/p57KIP2-positive) (Merchant, 2005).

image Management

Maternal deaths from molar pregnancies are rare because of early diagnosis, timely evacuation, and vigilant postevacuation surveillance for GTN. Preoperative evaluation attempts to identify known potential complications such as preeclampsia, hyperthyroidism, anemia, electrolyte depletions from hyperemesis, and metastatic disease (Table 20-2) (Lurain, 2010). Most recommend chest x-ray, whereas computed tomography (CT) and magnetic resonance (MR) imaging are not routinely done unless the chest radiograph shows lung lesions or unless there is evidence of other extrauterine disease such as in the brain or liver.

TABLE 20-2. Some Considerations for Management of Hydatidiform Mole



Hemogram; serum β-hCG, creatinine, and hepatic aminotransferase levels

TSH, free T4 levels

Type and Rh; group and screen or crossmatch

Chest radiograph

Consider hygroscopic dilators


Large-bore intravenous catheter(s)

Regional or general anesthesia

Oxytocin (Pitocin): 20 units in 1000 mL RL for continuous infusion

One or more other uterotonic agents may be added as needed:

Methylergonovine (Methergine): 0.2 mg = 1 mL = 1 ampule IM every 2 hr prn

Carboprost tromethamine (PGF2α) (Hemabate): 250 μg = 1 mL = 1 ampule IM every 15–90 min prn

Misoprostol (PGE1) (Cytotec): 200 mg tablets for rectal administration, 800–1000 mg once

Karman cannula—size 10 or 12

Consider sonography machine


Anti-D immune globulin (Rhogam) if Rh D-negative

Initiate effective contraceptiona

Review pathology report

Serum hCG levels: within 48 hours of evacuation, weekly until undetectable, then monthly for 6 months

aIntrauterine devices are not suitable during surveillance.

GTN = gestational trophoblastic neoplasia; hCG = human chorionic gonadotropin; IM = intramuscular; PG = prostaglandin; RL = Ringer lactate; T4 = thyroxine; TSH = thyroid-stimulating hormone.

Termination of Molar Pregnancy

Regardless of uterine size, molar evacuation by suction curettage is usually the preferred treatment. Preoperative cervical dilatation with an osmotic agent is recommended if the cervix is minimally dilated. Intraoperative bleeding can be greater with molar pregnancy than with a comparably sized uterus containing nonmolar products. Thus with large moles, adequate anesthesia, sufficient intravenous access, and blood-banking support is imperative. The cervix is mechanically dilated to allow insertion of a 10- to 14-mm suction curette. As evacuation is begun, oxytocin is infused to limit bleeding. Intraoperative sonography is recommended to help ensure that the uterine cavity has been emptied. When the myometrium has contracted, thorough but gentle curettage with a sharp large-loop Sims curette is performed. If bleeding continues despite uterine evacuation and oxytocin infusion, other uterotonic agents shown in Table 20-2 are given. In some cases, pelvic arterial embolization or hysterectomy may be necessary (Tse, 2007). Profuse hemorrhage and surgical methods that may be useful for its management are discussed in Chapter 41 (p. 818).

It is invariable that some degree of trophoblastic deportation into the pelvic venous system takes place during molar evacuation (Hankins, 1987). With large molar pregnancies, the volume of tissue may be sufficient to produce clinically apparent respiratory insufficiency, pulmonary edema, or even embolism. In our earlier experiences with very large moles, these and their chest x-ray manifestations clear rapidly without specific treatment. However, fatalities have been described (Delmis, 2000). Because of deportation, there is concern that trophoblastic tissue will thrive within the lung parenchyma to cause persistent disease or even overt malignancy. Fortunately, there is no evidence that this is a major problem.

Following curettage, anti-D immunoglobulin (Rhogam) is given to Rh D-negative women because fetal tissues with a partial mole may include red cells with D-antigen (Chap. 15p. 311). Those with suspected complete mole are similarly treated because a definitive diagnosis of complete versus partial mole may not be confirmed until pathological evaluation of the evacuated products.

Following evacuation, the long-term prognosis for women with a hydatidiform mole is not improved with prophylactic chemotherapy (Goldstein, 1995). Moreover, chemotherapy toxicity—including death—may be significant, and thus it is not recommended routinely by the American College of Obstetricians and Gynecologists (2012).

Methods other than suction curettage may be considered for select cases. Hysterectomy with ovarian preservation may be preferable for women who have completed childbearing. Of women aged 40 and older, approximately a third will subsequently develop GTN, and hysterectomy markedly reduces this likelihood (Hanna, 2010). Theca-lutein cysts seen at the time of hysterectomy do not require removal, and they spontaneously regress following molar termination. Some recommend aspiration of larger cysts to minimize pain and torsion risk. In contrast, labor induction or hysterotomy is seldom used for molar evacuation in the United States. Both will likely increase blood loss and theoretically may increase the incidence of persistent trophoblastic disease (American College of Obstetricians and Gynecologists, 2012).

Postevacuation Surveillance

Close biochemical surveillance for persistent gestational neoplasia should follow hydatidiform mole evacuation. Concurrently, reliable contraception is imperative to avoid confusion caused by rising β-hCG levels from a new pregnancy. Most recommend either combination hormonal contraception or injectable medroxyprogesterone acetate. The latter is particularly useful if there is poor compliance. Intrauterine devices are not used until β-hCG levels are undetectable because of the risk of uterine perforation if there is an invasive mole. Finally, barrier and other methods are not recommended because of their relatively high failure rates.

Biochemical surveillance is by serial measurements of serum β-hCG to detect persistent or renewed trophoblastic proliferation. The initial β-hCG level is obtained within 48 hours after evacuation. This serves as the baseline, which is compared with β-hCG quantification done thereafter every 1 to 2 weeks until levels progressively decline to become undetectable.

The median time for such resolution is 7 weeks for partial moles and 9 weeks for complete moles. Once β-hCG is undetectable, this is confirmed with monthly determinations for another 6 months. After this, surveillance is discontinued and pregnancy allowed. Because such intensive monitoring has a high noncompliance rate, a truncated approach has been studied, and it may be unnecessary to verify undetectable β-hCG levels for 6 months. Specifically, it was shown that no woman with a partial or complete mole whose serum β-hCG level became undetectable subsequently developed neoplasia (Lavie, 2005; Wolfberg, 2004). Importantly, during the time during which β-hCG levels are monitored, either increasing or persistently plateaued levels mandate evaluation for trophoblastic neoplasia. If the woman has not become pregnant, then these levels signify increasing trophoblastic proliferation that is most likely malignant.

There are a number of risk factors for developing trophoblastic neoplasia following molar evacuation. Most important, complete moles have a 15 to 20 percent incidence of malignant sequelae, compared with 1 to 5 percent following partial moles. Surprisingly, with much earlier recognition and evacuation of molar pregnancies, the risk for neoplasia has not been lowered (Schorge, 2000). Other risk factors are older age, β-hCG levels > 100,000 mIU/mL, uterine size that is large-for-gestational age, theca-lutein cysts > 6 cm, and slow decline in β-hCG levels (Berkowitz, 2009; Kang, 2012; Wolfberg, 2005). Although not routine, postevacuation uterine sonographic surveillance showing myometrial nodules or hypervascularity may be a predictor of subsequent neoplasia (Garavaglia, 2009).


This group includes invasive mole, choriocarcinoma, placental site trophoblastic tumor, and epithelioid trophoblastic tumor. These tumors almost always develop with or follow some form of recognized pregnancy. Half follow hydatidiform mole, a fourth follow miscarriage or tubal pregnancy, and another fourth develop after a preterm or term pregnancy (Goldstein, 2012). Although these four tumor types are histologically distinct, they are usually diagnosed solely by persistently elevated serum β-hCG levels because tissue is frequently not available for pathological study. Criteria for the diagnosis of postmolar gestational trophoblastic neoplasia are shown in Table 20-3.

TABLE 20-3. Criteria for Diagnosis of Gestational Trophoblastic Neoplasia

1. Plateau of serum β-hCG level (± 10 percent) for four measurements during a period of 3 weeks or longer—days 1, 7, 14, 21

2. Rise of serum β-hCG level > 10 percent during three weekly consecutive measurements or longer, during a period of 2 weeks or more—days 1, 7, 14

3. Serum β-hCG level remains detectable for 6 months or more

4. Histological criteria for choriocarcinoma

image Clinical Findings

These placental tumors are characterized clinically by their aggressive invasion into the myometrium and propensity to metastasize. The most common finding with gestational trophoblastic neoplasms is irregular bleeding associated with uterine subinvolution. The bleeding may be continuous or intermittent, with sudden and sometimes massive hemorrhage. Myometrial perforation from trophoblastic growth may cause intraperitoneal hemorrhage. In some women, lower genital tract metastases are evident, whereas in others there are only distant metastases with no trace of a uterine tumor.

image Diagnosis, Staging, and Prognostic Scoring

Consideration for the possibility of gestational trophoblastic neoplasia is the most important factor in its recognition. Unusually persistent bleeding after any type of pregnancy should prompt measurement of serum β-hCG levels and consideration for diagnostic curettage. Uterine size is assessed along with careful examination for lower genital tract metastases, which usually appear as bluish vascular masses (Cagayan, 2010). Tissue diagnosis is unnecessary, thus biopsy is not required and may cause significant bleeding.

Once the diagnosis is verified, in addition to a baseline serum β-hCG level and hemogram, a search for local disease and metastases includes tests of liver and renal function, transvaginal sonography, chest CT scan or radiograph, and brain and abdominopelvic CT scan or MR imaging. Less commonly, positron-emission tomographic (PET) scanning and cerebrospinal fluid β-hCG level determination are used to identify metastases (Lurain, 2011).

Gestational trophoblastic neoplasia is staged clinically using the system of the International Federation of Gynecology and Obstetrics (FIGO) (2009). This includes a modification of the World Health Organization (1983) prognostic index score, with which scores of 0 to 4 are given for each of the categories shown in Table 20-4. Women with WHO scores of 0 to 6 are considered to have low-risk disease, whereas those with a score ≥ 7 are considered in the high-risk group.

TABLE 20-4. International Federation of Gynecology and Obstetrics (FIGO) Staging and Diagnostic Scoring System for Gestational Trophoblastic Neoplasia


image Histological Classification

Again, it is stressed that the diagnosis of trophoblastic neoplasias is usually made by persistently elevated serum β-hCG levels without confirmation by pathological tissue study. Clinical staging is arrived at without regard for histological findings, even if available. Still, there are distinct histological types that are described next.

Invasive Mole

These are the most common trophoblastic neoplasms that follow hydatidiform moles, and almost all invasive moles arise from partial or complete moles (Sebire, 2005). Previously known as chorioadenoma destruens, invasive mole is characterized by extensive tissue invasion by trophoblast and whole villi. There is penetration deep into the myometrium, sometimes with involvement of the peritoneum, adjacent parametrium, or vaginal vault. Although locally aggressive, invasive moles are less prone to metastasize such as with choriocarcinoma.

Gestational Choriocarcinoma

This is the most common type of trophoblastic neoplasm to follow a term pregnancy or a miscarriage, and only a third of cases follow a molar gestation (Soper, 2006). Choriocarcinoma is composed of cells reminiscent of early cytotrophoblast and syncytiotrophoblast, however, it contains no villi. This rapidly growing tumor invades both myometrium and blood vessels to create hemorrhage and necrosis. Myometrial tumor may spread outward and become visible on the uterine surface as dark, irregular nodules. Metastases often develop early and are generally blood-borne (Fig. 20-5). The most common sites are the lungs and vagina, but tumor may be metastatic to the vulva, kidneys, liver, ovaries, brain, and bowel. Choriocarcinomas are commonly accompanied by ovarian theca-lutein cysts.


FIGURE 20-5 Metastatic choriocarcinoma. A. Chest radiograph demonstrates widespread metastatic lesions. B. Autopsy specimen with multiple hemorrhagic hepatic metastases. (Images contributed by Dr. Michael Conner.)

Placental Site Trophoblastic Tumor (PSST)

This uncommon tumor arises from implantation site-intermediate trophoblast. These tumors have associated serum β-hCG levels that may be only modestly elevated, but they produce variant forms of hCG, and identification of a high proportion of free β-hCG (> 30 percent) is considered diagnostic. Treatment of placental site trophoblastic tumor by hysterectomy is preferred because these locally invasive tumors are usually resistant to chemotherapy (Baergen, 2006). For higher-risk stage I and for later stages, adjuvant multidrug chemotherapy is also given (Schmid, 2009).

Epithelioid Trophoblastic Tumor

This rare tumor develops from chorionic-type intermediate trophoblast. Grossly, the tumor grows in a nodular fashion. Primary treatment is hysterectomy because this tumor is relatively resistant to chemotherapy. Approximately a fourth of women with this neoplasm will have metastatic disease, and they are given combination chemotherapy (Morgan, 2008).

image Treatment

Women with gestational trophoblastic neoplasia are best managed by oncologists. Chemotherapy is usually the primary treatment, and repeat evacuation is not recommended by most because of risks for uterine perforation, bleeding, infection, or intrauterine adhesion formation. In a few, suction curettage may be necessary if there is bleeding or a substantial amount of retained molar tissue. Although controversial, some also consider a second uterine evacuation to be an initial therapeutic option in some cases of GTN following molar evacuation in an effort to avoid or minimize chemotherapy (Pezeshki, 2004; van Trommel, 2005). Moreover, in specific cases, hysterectomy may be primary or adjuvant treatment (Clark, 2010).

Single-agent chemotherapy protocols are usually sufficient for nonmetastatic or low-risk metastatic neoplasia (Horowitz, 2009). In their review of 108 women with low-risk disease, Abrão and colleagues (2008) reported that monotherapy protocols with either methotrexate or actinomycin D were equally effective compared with a regimen containing both. In general, methotrexate is less toxic than actinomycin D (Chan, 2006; Seckl, 2010). Regimens are repeated until serum β-hCG levels are undetectable.

Combination chemotherapy is given for high-risk disease, and reported cure rates approximate 90 percent (Lurain, 2010). A number of regimens have been used with success. One is EMA-CO, which includes etoposide, methotrexate, actinomycin D, cyclophosphamide, and oncovin (vincristine). Adjuvant surgical and radiotherapy may also be employed (Hanna, 2010).

With either low- or high-risk disease, once serum β-hCG levels are undetectable, serosurveillance is continued for 1 year. During this time, effective contraception is crucial to avoid any teratogenic effects of chemotherapy to the fetus as well as to mitigate confusion from rising β-hCG levels caused by superimposed pregnancy.

A small number of women during surveillance, despite no evidence of metastases, will be found to have very low β-hCG levels that plateau. This phenomenon is referred to as quiescent hCG and presumably is caused by dormant trophoblast. Close observation without therapy is recommended, and 20 percent will eventually have recurrent active and progressive trophoblastic neoplasia (Khanlian, 2003).


Women with prior gestational trophoblastic disease or successfully treated neoplasia usually do not have impaired fertility, and their pregnancy outcomes are usually normal (Tse, 2012). The primary concern in these women is their 2-percent risk for developing trophoblastic disease in a subsequent pregnancy (Garrett, 2008). Sonographic evaluation is recommended in early pregnancy, and subsequently if indicated. At delivery, the placenta or products of conception are sent for pathological evaluation, and a serum β-hCG level is measured 6 weeks postpartum.


Abrão RA, de Andrade JM, Tiezzi DG, et al: Treatment for low-risk gestational trophoblastic disease: comparison of single-agent methotrexate, dactinomycin and combination regimens. Gynecol Oncol 108:149, 2008

Altman AD, Bently B, Murray S, et al: Maternal age-related rate of gestational trophoblastic disease. Obstet Gynecol 112:244, 2008

American College of Obstetricians and Gynecologists: Diagnosis and treatment of gestational trophoblastic disease. Practice Bulletin No. 53, June 2004, Reaffirmed 2012

Baergen RN, Rutgers JL, Young RH, et al: Placental site trophoblastic tumor: a study of 55 cases and review of the literature emphasizing factors of prognostic significance. Gynecol Oncol 100:511, 2006

Benirschke K, Burton GJ, Baergen RN (eds): Molar pregnancies. In Pathology of the Human Placenta, 6th ed. New York, Springer, 2012, p 687

Berkowitz RS, Goldstein DP: Current management of gestational trophoblastic diseases. Gynecol Oncol 112(3):654, 2009

Berkowitz RS, Im SS, Bernstein MR, et al: Gestational trophoblastic disease. Subsequent pregnancy outcome, including repeat molar pregnancy. J Reprod Med 43:81, 1998

Cagayan MS: Vaginal metastases complicating gestational trophoblastic neoplasia. J Reprod Med 55(5–6):229, 2010

Castrillon DH, Sun D, Weremowicz S, et al: Discrimination of complete hydatidiform mole from its mimics by immunohistochemistry of the paternally imprinted gene product p57KIP2. Am J Surg Pathol 25(10):1225, 2001

Chan KK, Huang Y, Tam KF, et al: Single-dose methotrexate regimen in the treatment of low-risk gestational trophoblastic neoplasia. Am J Obstet Gynecol 195:1282, 2006

Clark RM, Nevadunsky NS, Ghosh S, et al: The evolving role of hysterectomy in gestational trophoblastic neoplasia at the New England Trophoblastic Disease Center. J Reprod Med 55(5–6):194, 2010

Delmis J, Pfeifer D, Ivanisecvic M, et al: Sudden death from trophoblastic embolism in pregnancy. Eur J Obstet Gynecol Reprod Biol 92:225, 2000

Drake RD, Rao GG, McIntire DD, et al: Gestational trophoblastic disease among Hispanic women: a 21-year hospital-based study. Gynecol Oncol 103:81, 2006

FIGO Committee on Gynecologic Oncology: Current FIGO staging for cancer of the vagina, fallopian tube, ovary, and gestational trophoblastic neoplasia. Int J Gynaecol Obstet 105:3, 2009

Fowler DJ, Lindsay I, Seckl MJ, et al: Routine pre-evacuation ultrasound diagnosis of hydatidiform mole: experience of more than 1000 cases from a regional referral center. Ultrasound Obstet Gynecol 27(1):56, 2006

Garavaglia E, Gentile C, Cavoretto P, et al: Ultrasound imaging after evacuation as an adjunct to beta-hCG monitoring in posthydatidiform molar gestational trophoblastic neoplasia. Am J Obstet Gynecol 200(4):417.e1, 2009

Garrett LA, Garner EIO, Feltmate CM, et al: Subsequent pregnancy outcomes in patients with molar pregnancy and persistent gestational trophoblastic neoplasia. J Reprod Med 53:481, 2008

Gemer O, Segal S, Kopmar A, et al: The current clinical presentation of complete molar pregnancy. Arch Gynecol Obstet 264(1):33, 2000

Goldstein DP: Gestational trophoblastic neoplasia: where we came from, where we stand today, where we are heading. Keynote address. J Reprod Med 55(5–6):184, 2010

Goldstein DP, Berkowitz RS: Current management of gestational trophoblastic neoplasia. Hematol Oncol Clin North Am 26(1):111, 2012

Goldstein DP, Berkowitz RS: Prophylactic chemotherapy of complete molar pregnancy. Semin Oncol 22:157, 1995

Hankins GD, Wendel GD, Snyder RR, et al: Trophoblastic embolization during molar evacuation: central hemodynamic observations. Obstet Gynecol 63:368, 1987

Hanna RK, Soper JT: The role of surgery and radiation therapy in the management of gestational trophoblastic disease. Oncologist 15(6):593, 2010

Horowitz NS, Goldstein DP, Berkowitz RS: Management of gestational trophoblastic neoplasia. Semin Oncol 36(2):181, 2009

Kang WD, Choi HS, Kim SM: Prediction of persistent gestational trophoblastic neoplasia: the role of hCG level and ratio in 2 weeks after evacuation of complete mole. Gynecol Oncol 124(2):250, 2012

Kerkmeijer LG, Massuger LF, Ten Kate-Booij MJ, et al: Earlier diagnosis and serum human chorionic gonadotropin regression in complete hydatidiform moles. Obstet Gynecol 113:326, 2009

Khanlian SA, Smith HO, Cole LA: Persistent low levels of human chorionic gonadotropin: a premalignant gestational trophoblastic disease. Am J Obstet Gynecol 188(5):1254, 2003

Lavie I, Rao GG, Castrillon DH, et al: Duration of human chorionic gonadotropin surveillance for partial hydatidiform moles. Am J Obstet Gynecol 192:1362, 2005

Lawler SD, Fisher RA, Dent J: A prospective genetic study of complete and partial hydatidiform moles. Am J Obstet Gynecol 164:1270, 1991

Lee C, Smith HO, Kim SJ: Epidemiology. In Hancock BW, Seckl MJ, Berkowitz RS, et al (eds): Gestational trophoblastic disease, 3rd ed. London, International Society for the Study of Trophoblastic Disease, 2011, p 57. Available at: http://www.isstd.org/index.html. Accessed January 17, 2012

Lipata F, Parkash V, Talmor M, et al: Precise DNA genotyping diagnosis of hydatidiform mole. Obstet Gynecol 115(4):784, 2010

Lurain JR: Gestational trophoblastic disease I: epidemiology, pathology, clinical presentation and diagnosis of gestational trophoblastic disease, and management of hydatidiform mole. Am J Obstet Gynecol 203(6):531, 2010

Lurain JR: Gestational trophoblastic disease II: classification and management of gestational trophoblastic neoplasia. Am J Obstet Gynecol 204(1):11, 2011

Lybol C, Thomas CM, Bulten J: et al: Increase in the incidence of gestational trophoblastic disease in The Netherlands. Gynecol Oncol 121(2):334, 2011

Mangili G, Garavaglia E, Cavoretto P, et al: Clinical presentation of hydatidiform mole in northern Italy: has it changed in the last 20 years? Am J Obstet Gynecol 198(3):302.e1, 2008

Massardier J, Golfner F, Journet D, et al: Twin pregnancy with complete hydatidiform mole and coexistent fetus obstetrical and oncological outcomes in a series of 14 cases. Eur J Obstet Gynecol Reprod Biol 143:84, 2009

Merchant SH, Amin MB, Viswanatha DS, et al: p57KIP2 immunohistochemistry in early molar pregnancies: emphasis on its complementary role in the differential diagnosis of hydropic abortuses. Hum Pathol 36:180, 2005

Morgan JM, Lurain JR: Gestational trophoblastic neoplasia: an update. Curr Oncol Rep 10(6):497, 2008

Moskovitz JB, Bond MC: Molar pregnancy-induced thyroid storm. J Emerg Med 38(5):e71, 2010

Niemann I, Petersen LK, Hansen ES, et al: Differences in current clinical features of diploid and triploid hydatidiform mole. BJOG 114:1273, 2007a

Niemann I, Sunde L, Petersen LK: Evaluation of the risk of persistent trophoblastic disease after twin pregnancy with diploid hydatidiform mole and coexisting normal fetus. Am J Obstet Gynecol 197:45.e1, 2007b

Paradinas FJ, Browne P, Fisher RA, et al: A clinical, histopathological and flow cytometric study of 149 complete moles, 146 partial moles and 107 non-molar hydropic abortions. Histopathology 28(2):101, 1996

Pezeshki M, Hancock BW, Silcocks P: The role of repeat uterine evacuation in the management of persistent gestational trophoblastic disease. Gynecol Oncol 95(3):423, 2004

Salehi S, Eloranta S, Johansson AL: Reporting and incidence trends of hydatidiform mole in Sweden 1973–2004. Acta Oncol 50(3):367, 2011

Schmid P, Nagai Y, Agarwal R: Prognostic markers and long-term outcome of placental-site trophoblastic tumours: a retrospective observational study. Lancet 374(9683):48, 2009

Schorge JO, Goldstein DP, Bernstein MR, et al: Recent advances in gestational trophoblastic disease. J Reprod Med 45:692, 2000

Sebire NJ, Foskett M, Fisher RA, et al: Risk of partial and complete hydatidiform molar pregnancy in relation to maternal age. Br J Obstet Gynaecol 109:99, 2002a

Sebire NJ, Foskett M, Parainas FJ, et al: Outcome of twin pregnancies with complete hydatidiform mole and healthy co-twin. Lancet 359:2165, 2002b

Sebire NJ, Lindsay I, Fisher RA: Overdiagnosis of complete and partial hydatidiform mole in tubal ectopic pregnancies. Int J Gynecol Pathol 24(3):260, 2005

Seckl MJ, Sebire NJ, Berkowitz RS: Gestational trophoblastic disease. Lancet 376(9742):717, 2010

Smith HO, Wiggins C, Verschraegen CF, et al: Changing trends in gestational trophoblastic disease. J Reprod Med 51:777, 2006

Soper JT: Gestational trophoblastic disease. Obstet Gynecol 108:176, 2006

Steller MA, Genest DR, Bernstein MR, et al: Clinical features of multiple conception with partial or complete molar pregnancy and coexisting fetuses. J Reprod Med 39(3):147, 1994

Tse KY, Chan KK, Tam KF: 20-year experience of managing profuse bleeding in gestational trophoblastic disease. J Reprod Med (5):397, 2007

Tse KY, Ngan HY: Gestational trophoblastic disease. Best Pract Res Clin Obstet Gynaecol 26(3):357, 2012

van Trommel NE, Massuger LF, Verheijen RH, et al: The curative effect of a second curettage in persistent trophoblastic disease: a retrospective cohort survey. Gynecol Oncol 99:6, 2005

Wee L, Jauniaux E: Prenatal diagnosis and management of twin pregnancies complicated by a co-existing molar pregnancy. Prenat Diagn 25(9):772, 2005

Wolfberg AJ, Berkowitz RS, Goldstein DP: Postevacuation hCG levels and risk of gestational trophoblastic neoplasia in women with complete molar pregnancy. Obstet Gynecol 106(3):548, 2005

Wolfberg AJ, Feltmate C, Goldstein DP, et al: Low risk of relapse after achieving undetectable hCG levels in women with complete molar pregnancy. Obstet Gynecol 104:551, 2004

World Health Organization Scientific Group: Gestational trophoblastic disease. WHO Tech Rep Ser 692:1, 1983