Oncology

Gestational trophoblastic diseases

Jump to Section

Gestational trophoblastic disease

What every physician needs to know:

Gestational trophoblastic disease (GTD)

GTD is a spectrum of diseases that affect women primarily of reproductive age. GTD entities are derived from placental origin, ranging from abnormal products of conception with malignant potential (partial and complete hydatidiform moles) to malignant gestational trophoblastic neoplasia (GTN). GTN also comprises a range of neoplasms, including gestational choriocarcinomas, invasive moles, and more rarely, placental site trophoblastic tumors or epithelioid trophoblastic tumors. Most of these conditions are associated with an active or identifiable antecedent pregnancy and secrete abnormal levels of human chorionic gonadotropin (hCG).

Hydatidiform moles are identified in approximately 1:1000 pregnancies while gestational choriocarcinoma complicates 1:20,000-40,000 pregnancies in the USA. Approximately 1500 women are treated annually in the USA for GTN.

Hydatidiform mole

Patients with hydatidiform moles are usually treated by evacuation of the pregnancy and are followed with serial serum hCG levels until the hCG values normalize, using assays with threshold values of at least under 5 IU/L. Patients are usually monitored after normal values are obtained before an attempt at another pregnancy is encouraged. If hCG values do not fall appropriately during surveillance, they are diagnosed as having postmolar GTN.

A complete mole is a result of fertilization of an empty ovum by two sperm or a duplication of a single sperm that duplicates, resulting in a 46 XX or 46 XY karyotype.

A partial mole arises from fertilization of a haploid ovum by two sperm or duplication of a single sperm, resulting in a triploid karyotype (69 XXY, 69 XXX, 69 XYY).

Gestational trophoblastic neoplasia

Women with postmolar GTN are usually promptly identified through hCG surveillance and can be cured with relatively simple chemotherapy regimens, most often without hysterectomy. Patients with nonmolar GTN may present many months after a pregnancy with symptoms of vaginal bleeding from uterine tumors or symptoms from distant metastases.

In sharp contrast to other solid tumors, the majority of patients with advanced stage GTN can be cured with aggressive treatment. Most patients treated for GTN can preserve fertility. Women of reproductive age who present with metastatic malignancies from an unknown primary site, CNS hemorrhages or pulmonary lesions should be screened with a serum hCG; a significant elevation of hCG would raise suspicion of GTN.

Are you sure your patient has gestational trophoblastic disease? What should you expect to find?

Hydatidiform mole

Patients with hydatidiform moles are usually identified as having an active pregnancy event. Most frequently, patients with partial moles present with a positive pregnancy test and have a "missed spontaneous abortion" identified by an ultrasound that is obtained because of a uterus that is smaller than expected for gestational age, or hCG values are lower than expected for gestational age.

The diagnosis is made by histological evaluation of the products of gestation after suction dilatation and curettage. Patients with complete moles may present with a similar clinical scenario, but many will present with vaginal bleeding, uterus enlarged greater than expected dates, hCG higher than expected for gestational age, and will have an ultrasound demonstrating an absent fetus and a heterogeneous mass of fluid filled vesicles filling the uterus (Figure 1).

Figure 1.

Pelvic ultrasound of complete mole demonstrating uterine tumor comprised of multiple vesicles.

Histology from suction dilatation and evacuation will confirm the diagnosis. Very rarely, patients with a normal twin fetus and a mole will be identified by ultrasound. These pregnancies require rigorous evaluation and management by a team of specialists experienced in ultrasound, high-risk maternal/fetal medicine and management of GTD.

Gestational trophoblastic neoplasia

Postmolar gestational trophoblastic neoplasia

After evacuation of a molar pregnancy and histologic identification as a partial or complete mole, patients are followed with serial weekly serum quantitative beta hCG assays. Approximately 2.5-5% of patients with partial, and approximately 20% of patients with complete, moles will be identified as having postmolar GTN.

Usually, the diagnosis is made by a rising hCG value, presence of identifiable metastases, or histological diagnosis of gestational choriocarcinoma in uterine curettings. Because most patients are diagnosed on the basis of serum hCG values and treated without hysterectomy, a precise histologic diagnosis is usually not obtained.

Based on studies before the advent of effective chemotherapy, approximately one-third of postmolar GTN was gestational choriocarcinoma, with the remaining two-thirds of cases having invasive or proliferative moles. Postmolar GTN constitutes approximately two-thirds of all cases treated for GTN.

Although practitioners may use other criteria for initiating treatment in individual patients, current FIGO Criteria for the diagnosis of postmolar GTN includes:

  • A significant (>10%) rise in serum hCG values sustained over 2 weeks duration, confirmed by 3 values obtained at weekly intervals (x, x + 7 days, x + 14 days).

  • A sustained plateau (< 10% drop) in serum hCG values obtained over 3 weeks duration, confirmed by 4 values obtained at weekly intervals (X, x + 7 days, x + 14 days, x + 21 days).

  • Persistent hCG for more than 6 months after molar evacuation.

  • Presence of metastatic disease on physical examination or radiographic evaluation.

  • Histological diagnosis of choriocarcinoma, placental site trophoblastic tumor, or epithelioid trophoblastic tumor.

Nonmolar gestational trophoblastic neoplasia

Patients with gestational choriocarcinoma, placental site trophoblastic tumors, and epithelioid trophoblastic tumors will sometimes have the diagnosis made on the basis of histology from uterine curettings, hysterectomy, or biopsy of metastatic lesions (Figure 2). Metastases of choriocarcinoma are often highly vascular, with a tendency for hemorrhage with biopsy. Therefore, a presumptive diagnosis of gestational choriocarcinoma is often made when there is metastatic disease of unidentified primary site in a woman of reproductive age who has high levels of serum hCG.

Figure 2.

Multiple pulmonary metastases of GTN diagnosed by chest x-ray in a patient with high-risk disease.

In contrast, the intermediate trophoblast cells in placental site trophoblastic tumors secrete predominantly free beta hCG, which is not usually detected by conventional hCG assays, and epithelioid trophoblastic tumors usually secrete low levels of hCG. Thus, these two forms of GTN are usually diagnosed by histology.

Gestational choriocarcinoma accounts for approximately 98-99% of nonmolar GTN. While 50-65% of the cases of choricarcinoma are sequelae of molar GTN, the remainder are roughly equally divided among cases that follow a normal pregnancy and those diagnosed after a spontaneous miscarriage, ectopic pregnancy, or other nonmolar pregnancy event. Hematogenously disseminated metastases and rapid doubling times are hallmarks of choriocarcinoma. Gestational choriocarcinoma is usually extremely sensitive to chemotherapy, including single-agent methotrexate and actinomycin-D.

Placental site trophoblastic tumors may follow any type of pregnancy. In addition to differences in amount of hCG production, these tumors are less aggressive than choriocarcinoma. Local invasion of the uterus is the most common stage at diagnosis, and lymphatic metastasis are more frequently encountered than in choriocarcinoma. In contrast to choriocarcinoma, placental site trophoblastic tumors are usually resistant to single-agent methotrexate and actinomycin-D.

Epithelioid trophoblastic tumors are the least common form of GTN. Although they are often confined to the uterus, they may metastasize widely and are often resistant to chemotherapy.

Beware of other conditions that can mimic gestational trophoblastic disease:

Early pregnancy

Early pregnancy ultrasound images may be non-diagnostic in spontaneous miscarriages, thus most partial and many complete moles are identified by histology after evacuation. Twin and other multiple gestations may produce hCG levels that are abnormally high for gestational age; early diagnosis should not be dependent upon hCG value alone.

Patients should use effective contraception during hCG level surveillance after molar evacuation or treatment of GTN, so that hCG from a normal pregnancy does not result in the false diagnosis of GTN and result in inappropriate treatment in a desired pregnancy.

Placental bleeding can cause transient elevations of serum hCG. Rarely, a retroplacental hematoma complicating an otherwise normal pregnancy can produce both an abnormally elevated hCG and an ultrasound that suggests a twin concurrent molar gestation.

Phantom hCG

The term "phantom hCG" refers to conditions when serum assays for hCG are falsely elevated, usually because of substances that cross-react with the monoclonal antibodies that are used for the capture and/or labeling antibodies in the assay techniques. Lab errors in patient serum identification or assay contamination may also case a false positive hCG. In most cases, hCG values are relatively low, however, values over 500 have been reported, and are relatively static.

The most frequently documented case of phantom hCG are heterophile antibodies, such as the human anti-murine antibody, that cross-react with the murine monoclonal antibodies used in currently available hCG assays.

A false-positive hCG caused by heterophile antibodies will not exhibit an appropriate change with serial dilutions of serum, and will often register widely different hCG values when different assays are used. It will be eliminated when antibodies are "stripped" from the patient's serum and will not produce a positive urine hCG test even when serum hCG levels are elevated above the threshold of the urinary hCG assay because the heterophile antibodies are not secreted in the urine.

The commercial assay most responsible for cases of "phantom hCG" reported in the late 1990s and early 2000s has been removed from the market, and most other commercial assays have been re-formulated to avoid this problem.

Quiescent gestational trophoblastic disease

Some patients who have been followed after evacuation of a mole or treatment of GTN will have persistent low levels of hCG that do not respond to chemotherapy. Rarely, this can be encountered after a pregnancy that has not been complicated by GTD. The majority of these patients have very low (< 27%) to undetectable proportions of hyperglycosylated hCG relative to normal hCG.

Hyperglycosylated hCG is the form of hCG that is manufactured by invasive trophoblast cells, and if the hyperglycosylated portion is less than 40% of the total hCG, chemotherapy is unlikely to be successful.

In the majority of patients with very low (< 27%) hyperglycosylated levels hCG production will ultimately cease with no further sequelae. Patients with low levels of hyperglycosylated hCG should be monitored closely, however, as up to one quarter of the patients will exhibit rising total hCG values and convert to malignant behavior.

Pituitary hCG

The pituitary gland normally produces low levels of normal hCG during ovulation and the luteal phase of the ovulatory cycle. In contrast to placental hCG, pituitary hCG is sulfated, but can be detected by commercial assays especially in patients after oophorectomy, or in postmenopausal patients. Familial clusters of "inheritable hCG syndrome" with detectable tonic levels of pituitary hCG have been reported.

Secretion of hCG by non-gestational tumors

Gonadal and nongonadal germ cell tumors, including non-gestational choriocarcinomas and dysgerminomas, may secrete normal or hyperglycosylated hCG. Identification of an ovarian tumor or a midline mass in a woman with high levels of hCG should raise the possibility of a germ cell tumor.

Detection of free beta chain hCG in the serum and beta-chain fragments of hCG in the urine have been frequently observed in patients with a variety of gynecologic and non-gynecologic malignancies, but it is rare for other tumor types to secrete significant levels of normal hCG that would be detected by conventional hCG assays.

Which individuals are most at risk for developing gestational trophoblastic disease:

The epidemiology of GTD is poorly defined. This is partially caused by studies that combined patients with molar gestations and postmolar GTN with patients who had nonmolar GTN, predominantly gestational choriocarcinoma, in case-control studies.

Many of the older studies used hospital-based cohorts rather than population-based cohorts of patients for reporting case-controlled studies, which probably inflated the incidence of molar pregnancies and GTN because of referral bias.

Hydatidiform mole

Prior mole increases the risk of hydatidiform moles in subsequent pregnancies, and is the strongest risk factor. Patients with a prior mole have a 1-2% risk of a second mole, with the risk rising to 10% after a second, and 30% after a third mole. Some families have increased risk of moles in pregnancies.

The incidence of moles is increased in women over 18, and increases progressively with advancing age of 35 years, and over. Increasing paternal age is also associated with a slight increased risk for moles. While ethnicity has been emphasized in the past, dietary factors, including reduced amounts of beta carotene and animal fat intake may account in part for this difference.

Prior pregnancy wastage (ie, pregnancy loss), ABO blood type (Type A), and exposure to environmental toxins have weaker associations with increased risk.

Choriocarcinoma

Complete hydatidiform moles are the highest risk factor for choriocarcinoma. Up to 3-9% of patients with complete moles will have choriocarcinoma, with a much lower risk following partial moles. Increasing maternal age is also correlated with an increased risk of choriocarcinoma and GTN after a mole. Other factors have much weaker associations with the development of choriocarcinoma.

What laboratory and imaging studies should you order to characterize this patient's tumor (i.e., stage, grade, Ct/MRI vs PET/CT, cellular and molecular markers, immunophenotyping, etc.) How should you interpret the results and use them to establish prognosis and plan initial therapy?

Hydatidiform mole

Patients with a suspected hydatidiform mole should undergo pelvic ultrasound which may provide the radiographic definitive diagnosis, and will provide an estimate of the volume of the uterine contents.

If typical findings of a uterus filled with a mass of vesicular structures is seen (Figure 1), the patient should be prepared for a uterine suction dilatation and evacuation (D&E) or, in rare patients who desire sterilization, hysterectomy.

Baseline laboratory evaluation should include:

  • Serum hCG for baseline.

  • Complete blood count.

  • Electrolyte and renal functions.

  • Thyroid function tests if the hCG is more than 100,000.

High elevations of hCG may result in secondary hyperthyroidism because of homology between the alpha subunit of hCG and thyroid stimulating hormone. Physical manifestations of hyperthyroidism should be treated prior to surgery. Patients are often anemic from vaginal bleeding and may require transfusion during or after surgery.

A post-evacuation chest x-ray should be performed within 48 hours to screen for trophoblastic embolization that can lead to metastatic disease.

As noted above, patients are followed with serial hCG values after evacuation to diagnose postmolar GTN. When hCG values normalize after molar evacuation, continued surveillance is usually recommended for at least 6 months, although the incidence of postmolar GTN drops to under 0.4% when 2 or more normal hCG values have been recorded.

Gestational trophoblastic neoplasia

Patients diagnosed with GTN should undergo physical examination and radiographic evaluation to establish stage and also undergo a risk factor assessment to establish a risk score that will be used to determine the intensity of initial treatment.

Studies for staging should include:

  • Chest x-ray (Figure 2).

  • Pelvic examination (Figure 3) and ultrasound.

  • Computed tomography (CT) of chest, abdomen and pelvis.

  • Magnetic resonance imaging (MRI) or contrasted CT scans of the brain (Figure 4).

Figure 3.

The bluish suburethral mass is a vaginal metastasie of GTN. These lesions are highly vascular and biopsy should be avoided

Figure 4.

Two enhancing cerebral metastases with surrounding edema in a patient with high-risk metastatic GTN.

A baseline (pretherapy) hCG is important for assigning risk score. Complete blood count, renal functions, electrolyte panel, and liver function tests are important for monitoring during chemotherapy. As in hydatidiform moles, patients with high hCG values should have thyroid function tests evaluated.

Staging

Staging is based on anatomic sites of disease involvement, but it has long been recognized that there is considerable variability in the prognosis of patients with GTN based on other clinical factors, especially among patients with Stage III disease. Table I states the staging of GTN by the International Federation of Gynecology and Obstetrics (FIGO).

In 2000, FIGO accepted the recommendation that a modification of the World Health Organization (WHO) risk score be used to triage patients into low-risk and high-risk groups. Changes are as follows:

  • Factors incorporated into the risk score, these include age, antecedent pregnancy, interval from antecedent pregnancy, pre-therapy hCG level, largest tumor size (including intrauterine tumor), site of metastatic involvement, number of metastases, and prior therapy.

  • Each factor is given an individual factor score ranging from 0 to 4.

  • The sum of the individual factor scores is the total risk score; patients with a risk factor score of 6 and under are low-risk, while those with risk-factor score of 7 and higher are high-risk.

  • Pulmonary lesions detected only by CT scan do not affect outcome.

Clinical studies have shown excellent correlation between survival and FIGO risk score (see Table II). In general, patients in the low-risk group have an excellent prognosis, approaching 100% survival when treated initially with single-agent chemotherapy.

Table II.

FIGO risk score for gestational trophoblastic diseases.
Factor/Score 0 1 2 4
Age (years) ≤ 39 > 40
Antecedent pregnancy Hydatidiform mole Abortion Term pregnancy
Interval from index pregnancy (mo) ≤ 4 4 - 6 7 - 12 > 12
Pretreatment hCG (IU/L) < 1,000 1,000 - 10,000 ≥ 10,000 - 100,000 > 100,000
Largest tumor size including uterus (cm) 3 - 4 ≥ 5
Size of metastases Spleen Kidney Gastrointestinal Tract Brain Liver
Number of metastases identified 0 1-4 5-8 >8
Previous failed chemotherapy drugs Single drug 2 or more

Patients in the high-risk group have a much worse outcome when treated initially with single-agent regimens, but have an excellent outcome (overall survival > 80-90%) when treated with multiagent chemotherapy and multimodality therapy at centers that have experience treating GTN.

Both the patient's stage and risk score should be recorded.

What therapies should you initiate immediately i.e., emergently?

Hydatidiform mole

Patients should be stabilized in anticipation of D&E, or in rare patients who desire sterilization, hysterectomy. In severely anemic patients, transfusions should be given to increase hct if less than 25%. Beta blockers, such as metoprolol, may be required to control tachycardia and hypertension associated with secondary hyperthyroidism.

If the uterus is under 8 weeks gestational size, D&E may be performed in an ambulatory surgical center, but patients with larger uterine size should undergo evacuation in an operating room with facilities for exploratory laparotomy if uterine perforation or severe hemorrhage is encountered during evacuation.

Patients with uterine enlargement of more than 12-14 weeks size are at risk for post-operative respiratory distress or trophoblastic embolization; baseline oxygen saturation or arterial blood gas measurements are advisable. Blood type and screening should be performed on all patients and Rhogam should be administered following evacuation if Rho negative.

Gestational trophoblastic neoplasia

Patients with GTN should undergo staging with risk assessment and stabilization for chemotherapy. Secondary hyperthyroidism, if present, should be treated. Chemotherapy should be administered as soon as possible because of the potential for rapid progression of choriocarcinoma. Rare patients with high-risk disease and large tumor burden will require ventilatory and intensive care support at presentation; it is critical that support be given during the initial phases of treatment even if widely metastatic disease is present because of the relatively good survival rates (e.g.: >75% for patients with brain metastases) in patients who can receive adequate therapy.

What should the initial definitive therapy for the cancer be?

Hydatidiform mole

Molar evacuation

Evacuation of the hydatidiform mole is the initial step in management of a molar pregnancy. In most patients, uterine D&E is most appropriate. The cervix should be dilated and a 9 to 12mm suction cannula is used to remove the uterine contents if a complete mole is present.

Intravenous pitocin infusion at 20U/L is begun after cervical dilatation to prevent uterine contractions against an undilated cervix. Gentle sharp curettage is usually performed after suction curettage to ensure complete evacuation.

Sharp curettage must be performed with caution, however, because of the risk of uterine perforation in moles, and the rare potential to form intra-uterine senechiae that can lead to amenorrhea and infertility (Asherman's syndrome). Ergotamines may be given following suction D&E to promote uterine involution.

Patients will be followed with weekly hCG values to ensure complete regression of the mole and postmolar GTN will be diagnosed in approximately 20% of patients after D&E, as described above.

Rarely, older patients may wish concurrent evacuation and sterilization. Women over 35 years of age are at an increased risk for postmolar GTN. If the patient is clinically stable, abdominal or laparoscopic hysterectomy is an acceptable choice for evacuation in these women.

Normal ovaries can be retained because adnexal metastases are rare. The risk of postmolar GTN is reduced from approximately 20% to less than 5% compared to D&E, at the expense of a major operative procedure. Because of the risk of postmolar GTN, hCG monitoring is still indicated after hysterectomy.

Other methods of uterine evacuation of moles, such as abdominal hysterotomy, sharp curettage, and prostaglandin induction of labor are not recommended because of an increased operative risk of these procedures and increased risk for postmolar GTN when compared with suction D&E.

Prophylactic chemotherapy after evacuation

Randomized studies have shown that a single course of methotrexate with leukovorin rescue or actinomycin D will significantly reduce the risk of postmolar GTN in patients with high-risk moles from 50% to less than 15%, with very little toxicity.

The low incidence of postmolar GTN in patients with low-risk moles, however, is not affected. Patients who developed GTN after treatment with methotrexate appeared to have an increased risk for methotrexate resistance, although this was not suggested in the trial that used actinomycin D as the treatment arm.

This treatment strategy is rarely used in the USA because of unnecessary exposure to chemotherapy for many patients with high risk moles, availability of hCG surveillance, and excellent outcome for patients with postmolar GTN when they are treated at the time of rising hCG.

Prophylactic chemotherapy is usually reserved for patients with high-risk moles who are predicted to be noncompliant with hCG surveillance, but given the risk of post-treatment GTN hCG follow-up is still recommended after prophylactic chemotherapy.

Gestational trophoblastic neoplasia

Low-risk gestational trophoblastic neoplasia

Patients with nonmetastatic and metastatic GTN who have FIGO risk scores of 6 and over have an excellent outcome when initially treated with single agent regimens, usually consisting of methotrexate or actinomycin D.

The majority of these patients have postmolar GTN diagnosed on the basis of hCG criteria, and many do not have gestational choriocarcinoma. In most series, survival approaches 100%, over 85% retain fertility, and less than 10% of patients require combination therapy.

Patients are monitored during therapy with weekly hCG levels and at least biweekly CBC, renal functions and electrolytes, and liver function tests.

Treatment is continued as long as hCG levels are declining more than 10% each 14 day cycle. Resistance is usually defined as a rise in hCG of more than 10%, that is sustained over 14 days, a plateau of approximately 10% sustained over 21 days, or the appearance of new metastases.

Patients who fail a single agent regimen are re-assessed with FIGO risk score, and if they remain in the low risk group are treated with an alternative drug regimen; patients with a high-risk score are treated with a multiagent regimen as if they have high-risk GTN. Usually 1 to 2 cycles of chemotherapy are given after normalization of hCG, with recurrence rates under 5% after primary treatment of low-risk GTN.

Chemotherapy for low-risk gestational trophoblastic neoplasia

A variety of single agent regimens have been used to treat patients with low risk GTN. Because of the relative rarity of GTN, only a few prospective chemotherapy trials have been conducted through cooperative groups, and only two randomized chemotherapy trials have been reported in this group of patients to date.

While etoposide and 5-flurouracil regimens have been used in Southeast Asia and China, the majority of studies in the USA and other international centers have evaluated methotrexate (MTX) and actinomycin D (ACT) regimens. The results of the most frequently used regimens are summarized and discussed in Table III.

Table III.

Regimens and response rates.
Regimen Complete response rate range
MTX 30-50mg/m2 IM recycled evey 7 days 70-80% Nonmetastatic GTN
48-80% Mixed low risk GTN
MTX 0.4mg/kg IM X 5 days recycled every 14 days 90-94% Nonmetastatic GTN
60-90% Mixed low risk GTN
MTX 1mg/kg days 1, 3, 5, 7 Leukovorin Factor 0.1mg/kg days 2, 4, 6, 8 recycled every 14 days 70-75% Nonmetastatic GTN
60-70% Mixed low risk GTN
ACT 1.25mg/m2 IV bolus recycled every 14 days 70-93% Mixed low risk
ACT 300mcg/m2 IV X 5 days recycled every 14 days 85-90% Mixed low risk

In the early 1990s, many centers in the US adopted weekly MTX as their standard regimen for low-risk GTN. Subsequently, biweekly ACT was shown to have higher complete response rates, but also substantially higher gastrointestinal toxicity. To date, no randomized trials have compared other multidose MTX or ACT regimens with the biweekly ACT pulse regimen.

Salvage regimens for patients failing primary therapy have been poorly detailed, with most investigators reporting a change to the alternative single agent, without detailing regimens, toxicity, or response rates. Some patients who are treated initially with weekly MTX or biweekly ACT bolus can be salvaged with multidose regimens of the same agent.

Because of the concern for future fertility and chronic toxicity in patients who have a good chance for cure, it is reasonable to consider this strategy in selected patients who have a low FIGO risk score after failure of both of the single dose MTX and ACT regimens.

Surgery in low-risk gestational trophoblastic neoplasia

Because of the high success rate for primary chemotherapy in low-risk GTN, and dismal results using surgery alone in the treatment of nonmetastatic GTN before effective chemotherapy was developed, surgical procedures are rarely indicated in the management of these patients.

The use of a second D&E to induce remission has mixed results in the literature, with complications leading to hysterectomy in up to 8% of patients. In the US, a second D&E is usually used only in patients who have low-risk GTN with symptomatic uterine bleeding and who need to be stabilized before initiation of chemotherapy.

Hysterectomy is most often used as a salvage procedure after failed primary or secondary chemotherapy. However, several studies have demonstrated that primary hysterectomy reduces the amount of chemotherapy required to treat patients with low risk nonmetastatic and selected patients with low risk metastatic GTN. Both abdominal and laparoscopic hysterectomy have been used, with no significant increase in complications when they are performed during chemotherapy for GTN.

Because of the high recurrence rates reported for women with nonmetastatic GTN treated by hysterectomy alone in the pre-chemotherapy era, single agent chemotherapy should be given to all patients undergoing hysterectomy during treatment for GTN.

High-risk gestational trophoblastic neoplasia

Aggressive multiagent chemotherapy is indicated as the primary modality for treatment of patients with high-risk GTN. While long term remission rates of more than 80% are cited in the literature, it should be recognized that these reflect the experience of centers that specialize in the treatment of GTN.

Several studies have reported worse survival for patients with high-risk disease who receive initial treatment outside of specialized centers; therefore, it is important to consult with a physician experienced in the management of this relatively rare malignancy during treatment.

Patients with high-risk GTN will more often have a prior abortion or term pregnancy and are more likely to have a histologic diagnosis of choriorcarcinoma than individuals in the low-risk group. They often have a larger tumor burden, with metastases to high-risk anatomic sites that puts the patient at risk for bleeding complications from metastases to the lungs, liver, or brain.

Deaths from high-risk GTN follow a biphasic pattern, with early deaths in the first 2 to 3 weeks of treatment caused by acute complications related to tumor distribution and late deaths related to chemoresistant disease. Multimodality therapy is often required to successfully treat these patients.

It is critical to evaluate patients early in the course of treatment for potential complications from metastatic involvement and provide supportive care during the early phase of therapy.

First-line chemotherapy for high-risk gestational trophoblastic neoplasia

It was originally observed that patients with high-risk factors rarely had complete responses to single agent MTX or ACT regimens. Combination chemotherapy with MTX, ACT, and either chlorambucil or cytoxan (MAC) was initially developed in the 1960s as a salvage regimen for patients who failed single agent therapy, but it was soon recognized that patients with high-risk disease who were initially treated with MAC had a much higher response rate than patients who received MAC as second or third-line therapy.

MAC regimen for GTN:

  • Methotrexate (MTX) 15mg, daily (days 1-5), fixed dose, IM days X 5 days

  • ACT 8-10mcg/kg IV X 5 days

  • Chlorambucil 8-10mg PO or cyclphosphamide 3mg/kg IV X 5 days

  • Repeat cycles every 14-21 days

Variations of the MAC regimen were used throughout the 1980s, but have largely been replaced by the alternating weekly etoposide-MTX-ACT/cyclophosphamide-vincristine (EMA/CO) regimen as the standard initial treatment for high-risk GTN.

Although randomized trials have not compared the two regimens, it appears that EMA/CO has less acute toxicity than MAC regimens and has a higher response rate among patients with very high-risk scores. Some clinicians will still use MAC regimens for patients with relatively low-risk scores in order to avoid the leukemogenic potential of etoposide.

EMA/CO Regimen for high-risk GTN:

EMA

  • Day 1 - ACT 500mcg IV bolus, etoposide 100mg/m2 IV, MTX 100mg/m2 IV bolus and 200mg/m2 12 hr IV infusion.

  • Day 2 - ACT 500mcg IV bolus, etoposide 100mg/m2 IV, Leukovorin factor 15mg PO or IV q 6 hours X 4 doses.

CO

  • Day 8 - Vincristine 0.8-1mg/m2 IV bolus (max 2 mg), Cyclophosphamide 600mg/m2 IV

Repeat cycles every 14 days.

Patients with brain metastases (Figure 4) should be evaluated for craniotomy with resection of isolated metastases or radiation to control hemorrhage from multiple metastases early in the course of treatment. Likewise, patients with hepatic metastases may benefit from tumor embolization, hepatic radiation or surgical resection early in treatment.

In patients who are unstable, or have an extensive metastatic burden and hCG levels of over 500,000 IU/L, an initial treatment with MTX infusion/leukovorin rescue or etoposide 100mg/m2 +/- cisplatin 50-75 mg/m2 is sometimes given to avoid catastrophic hemorrhage from high-risk sites of metastases, with EMA/CO initiated as soon as possible.

Patients are monitored with hCG values every week and with appropriate monitoring of CBC, renal functions, electrolytes and liver function tests. It is not uncommon to observe a rise in hCG approximately 7 days after the initial treatment, but usually, the hCG will begin to respond by day 14 of the first cycle of EMA/CO.

If a patient has responded to EMA/CO and is clinically stable with good tolerance of treatment, therapy can be administered as an overnight hospitalization for days 1 and 2 of EMA, with outpatient infusion of the CO portion of treatment.

Because hCG is such a sensitive tumor marker for this disease, repeated imaging is used only for clinical indications or after remission is established, as a baseline. It has been estimated that threshold hCG levels detect a tumor burden of more than 10,000 to 100,000 cells. For this reason at least 3 cycles of chemotherapy should be given after hCG values normalize.

Complete response rates ranging up to 85% have been reported for first-line therapy with EMA/CO in high-risk GTN. Patients with high-risk GTN who receive maintenance chemotherapy have recurrence rates of less than12.5%, compared to 25% among patients whose chemotherapy is stopped when hCG normalizes.

It is important to avoid treatment delays because gestational choriocarcinoma can exhibit extremely rapid growth with doubling of serum hCG values at less than 14 day intervals in some cases. Acute grade 4 hematologic toxicity is observed in more than 10% of patients treated with EMA/CO, but patients will often develop low-grade neutropenia or thrombocytopenia during therapy. Cytokine growth factor support with granulocyte stimulating factor given on non-treatment days may be needed to allow prompt recycling.

If hematologic toxicity is limiting scheduled retreatment and the patient is responding to EMA/CO, the CO portion of the regimen can be dropped and the patient treated with EMA cycles. Of note, small series of patients treated primarily with EMA have similar response rates to patients treated with EMA/CO, but randomized trials have not been performed.

Others have substituted an etoposide/cisplatin doublet for CO and reported similar response rates at the expense of increased grade 4 hematologic toxicity. Because the collective experience is greatest with EMA/CO, most investigators continue to use this as their preferred first-line regimen for high-risk GTN.

Remission is defined as 3 successive weekly hCG values that are below threshold. As mentioned above, viable trophoblast cells may remain after a normal hCG value. For this reason, maintenance chemotherapy is recommended beyond normal hCG levels. When 1 to 2 cycles are given in low-risk GTN, recurrence rates are under 5%, while recurrence rates drop from 25% to below 12.5% in high-risk patients who receive at least 3 maintenance cycles.

Second-line chemotherapy for high-risk gestational trophoblastic neoplasia

A relatively small series of patients with chemo-refractory GTN previously exposed to EMA/CO have been reported. Many of the reports include patients who underwent hysterectomy or metastectomy during salvage therapy, which reduces the ability to judge the activity of the chemotherapy regimen.

One option is the EMA-EP regimen, in which cisplatin-etoposide is substituted for the cyclophosphosphamide and vincristine in EMA-CO:

EMA

Day 1 -

  • ACT: 500mcg IV bolus

  • Etoposide: 100mg/m2 IV

  • MTX: 100mg/m2 IV bolus and 200mg/m2 12 hr IV infusion

Day 2 -

  • ACT: 500mcg IV bolus

  • Etoposide: 100mg/m2 IV

  • Leukovorin factor: 15mg PO or IV q 6 hours X 4 doses

EP

Day 8 -

  • Etoposide: 100mg/m2 IV

  • Cisplatin: 80mg/m2 IV

Repeat cycles every 14 days.

Some investigators have eliminated the day 2 etoposide and ACT doses to reduce toxicity, while others have maintained the original EMA schedule. Approximately 95% of patients with persistent low-level plateaus of hCG during EMA/CO therapy will respond to EP/EMA, while responses in patients with rising hCG values are 75-80%, and only 43% among patients with recurrence after exposure to EMA/CO.

Modified germ cell regimens using vincristine-bleomycin-cisplatin (VBP), bleomycin-etoposide-cisplatin (BEP), and ifosfamide-carboplatin-etoposide (ICE) are most successful treating patients who have not been exposed to EMA/CO.

Cumulative hematologic toxicity often limits treatment. Partial responses are more frequently reported than complete responses. However, complete responses have been reported in up to 60% of patients who had these regimens incorporated into salvage therapy of drug-resistant GTN.

Anecdotal reports have indicated complete responses to single agent paclitaxel and paclitaxel/platinum regimens in patients with chemorefractory GTN.

Patients with drug-resistant GTN can be extremely difficult to treat because of accrued toxicity in addition to resistance to effective chemotherapeutic agents. In this scenario, each patient must be carefully evaluated to determine whether surgery or radiation therapy will be beneficial.

Surgery and radiation therapy in high-risk gestational trophoblastic neoplasia

At least a third to a half of patients with high-risk GTN will require multimodality treatment as part of their therapy, to treat complications of disease or treatment and allow initiation or continuation of therapy, as planned prophylaxis of complications, or to treat an isolated focus of disease with hysterectomy or metastectomy.

Primary hysterectomy does not appear to be beneficial in this group of patients. Unlike with low-risk disease, patients in the high-risk group often have a large metastatic tumor burden and primary hysterectomy has a negligible effect on the amount of treatment needed to control disease. Primary metastectomy is usually performed to prevent or treat hemorrhage from metastases and allow stabilization early in the course of treatment.

Patients with brain metastases are at high risk for intracranial hemorrhage and neurological decompensation early in the course of therapy. Options for management include concurrent whole brain radiation to approximately 25-30 Gy delivered during initiation of chemotherapy or craniotomy for resection of one or two surgically accessible metastases involving non-critical structures.

In patients with isolated lesions involving critical central nervous system structures, intensity modulated radiation or cyberknife radiation to the individual lesions may be employed.

Patients undergoing whole brain radiation should be treated with lower doses of MTX, because of the risk of long-term neurological complications from concurrent radiation and chemotherapy, while those undergoing excision or focal radiation to isolated lesions should be treated with MTX at doses of more than 500 mg/m2/course with prolonged leukovorin rescue as described above.

This results in therapeutic methotrexate levels of methotrexate in the cerebrospinal fluid, and theoretically reduces the risk of central nervous system failure. Overall survival for patients presenting with brain metastasis is approximately 75% using either approach.

Liver metastases are the highest risk site of metastatic GTN, with overall survival rates of only 25% to less than 50%. Liver lesions are rarely isolated at presentation, making surgical resection a rarely useful primary procedure.

Selective embolization of liver metastases or irradiation of the liver to 20 Gy may be used to prevent or treat acute hemorrhagic complications of liver lesions. Whole liver radiation at this level does not substantially increase the risk of chemotherapy toxicity. Retrospective analyses of case series suggest that chemotherapy using etoposide-based regimens are superior to MAC chemotherapy.

Splenectomy, pulmonary wedge resection, segmental bowel resection and nephrectomy have been used to treat hemorrhage from metastases in individual patients during the treatment of individual patients with metastases to these sites.

Radiation to unilateral renal or vaginal metastases may be also be required. It should be emphasized, however, that chemotherapy is the mainstay for treatment of patients with high risk GTN and uncomplicated lesions at these sites will usually respond readily to chemotherapy without surgery or radiation.

Planned hysterectomy or metastectomy may be useful in highly selected patients with drug-resistant disease. Patients with limited disease burden elsewhere, relatively low (< 1,000 IU/L) hCG, and options for effective chemotherapy regimens after surgery are the best candidates for resections of this type, which may remove focal drug-resistant clones of tumor, or remove tumor at sites that are relatively inaccessible for therapeutic doses of chemotherapy because of fibrosis and poor vascularity. Surgical extirpation is usually performed during chemotherapy and chemotherapy is usually continued after surgery.

Radiographic evidence of pulmonary metastases often persists for months or years after successful treatment of metastatic GTN. Because hCG is usually such a sensitive tumor marker for this disease, elective pulmonary resection of persistent lesions is not recommended in a patient who is in remission with normal serum hCG values.

Placental site trophoblastic tumor (PSTT)

PSTT is a rare form of GTN with distinct histologic features, hormone secretion profile, and clinical characteristics when compared to choriocarcinomas. These tumors are comprised of neoplastic intermediate trophoblastic cells. Usually low levels of intact hCG are secreted in comparison to tumor burden. In contrast to choriocarcinomas, PSTT is not usually responsive to single agent MTX or ACT or the multiagent MAC chemotherapy regimens. The majority of patients present with disease localized to the uterus.

Retrospective case series suggest that hysterectomy should be incorporated into the primary treatment of PSTT, with survival rates of 90% in Stage I disease, dropping to approximately 50% for stages II-IV. Increasing interval from prior pregnancy more than 4 years appears to be a poor prognostic feature. EMA/CO or EP/EMA are the most active chemotherapy regimens for this disease, while BEP has produced anecdotal responses in chemorefractory disease.

Epithelioid trophoblastic tumor

More rare than PSTT, these tumors are also derived from intermediate trophoblast cells. Approximately one third have hydatidiform mole as the antecedent pregnancy, and intervals from prior pregnancy event to diagnosis are often longer than observed in PSTT or choriocarcinoma. Serum hCG levels are elevated in the majority of patients, but usually at lower levels than in choriocarcinoma.

Hysterectomy is the mainstay for treatment of localized disease. Metastectomy should be considered in cases with isolated metastases, because chemotherapy has been reported only anecdotally, with conflicting results using EMA/CO-type regimens for treating disseminated disease.

What other therapies are helpful for reducing complications?

Hydatidiform Mole

Many patients with complete mole present with or develop anemia from uterine hemorrhage and may require transfusion support. The possibility of hemorrhage during uterine evacuation should be anticipated, and patients should have at least a type and screen or cross-matched blood and a large-bore IV line during the procedure.

Beta-blockers may be useful in controlling manifestations of hyperthyroidism around uterine evacuation. Very rare patients will develop pregnancy induced hypertension complicating a mole and may require treatment with intravenous magnesium sulfate to prevent eclampsia. Rhogam (Rho immunoglobulin) should be administered to Rho-negative patients to prevent sensitization in future pregnancies.

Post-operative respiratory compromise may develop in up to 20% of patients after evacuation of moles with uterine enlargement more than 12-14 weeks gestational size. Causes include Acute Respiratory Distress Syndrome, pulmonary edema from fluid overload, high-output cardiac failure caused by severe anemia or hyperthyroidism, pulmonary embolism, or trophoblastic embolization. This will occasionally require ventilator support in addition to treatment of the underlying cause.

Effective contraception is encouraged during hCG surveillance after molar evacuation. In the absence of contraindications, oral contraceptives are preferred and do not increase the risk of postmolar GTN.

Low-risk gestational trophoblastic neoplasia

The majority of patients with low-risk GTN receive chemotherapy regimens that have a relatively low incidence of severe toxicity. Conventional anti-emetics are usually sufficient to prevent emetogenic toxicity from weekly MTX or biweekly ACT pulsed therapy, but occasional patients will require 5-HT3 receptor antagonists, such as ondansetron, to control nausea. Severe myelosuppression is very rarely observed.

Hysterectomy should be considered a component of the primary treatment of low-risk GTN in patients who desire sterilization.

High-risk gestational trophoblastic neoplasia

Because of the aggressive, rapidly recycled regimens used to treat high-risk GTN, indwelling central venous access, such as a port or peripheral indwelling central catheter, should be considered. Urinary alkilynization with intravenous sodium bicarbonate should be used in patients receiving MTX infusions at doses more than 500mg/m2, and methotrexate levels should be followed to gauge the duration of leukovorin rescue in these patients if there is an elevated creatinine clearance.

Filigrastim support is often required to allow recycling of chemotherapy during EMA/CO or EP/EMA regimens. Although radiation therapy, hysterectomy, and metastectomy are rarely incorporated into primary therapy of high-risk GTN, they may play a role in stabilization during initial treatment or as part of salvage treatment. Because of the favorable survival of patients with even disseminated GTN, intensive care unit support with ventilatory support during the initial phases of treatment should be aggressively used if indicated.

Patients treated for GTN have increased levels of depression and anxiety, with decreased self-esteem, and may require appropriate counseling support.

What should you tell the patient and the family about prognosis?

Hydatidiform Mole

Patients with hydatidiform mole have an excellent prognosis. Eighty percent will undergo spontaneous remission after molar evacuation, and patients with postmolar GTN have an excellent chance of cure if they are promptly diagnosed during hCG surveillance.

The risk of postmolar GTN decreases sharply after hCG values normalize, especially if normal hCG values have been recorded for several months. Effective contraception is recommended during hCG surveillance to prevent interference of monitoring by the hCG associated with an intercurrent pregnancy.

Among patients who wish to preserve child-bearing capacity, future pregnancies do not appear to have increased complications, other than a 1 to 2% risk of a second mole. If a patient has a second mole, the risk of a third mole increases to approximately 10%, and the risk of repetitive moles rises thereafter.

However, even after 3 or 4 consecutive moles, the chance of a normal pregnancy is greater than the chance of a subsequent mole. Patients should undergo an early obstetrical ultrasound in subsequent pregnancies to evaluate the placenta for a mole, and should be screened with a serum hCG 6 to 8 weeks after delivery to ensure that the hCG normalizes.

Low-risk gestational trophoblastic neoplasia

Patients in the low-risk group can be reassured that long-term cure rates approach 100%. More than 90% of patients wishing to preserve fertility can achieve remission without undergoing hysterectomy and can avoid multiagent therapy that might result in premature ovarian failure or increase their risk of secondary malignancies.

Patients who are treated with single agent regimens for GTN have apparent normal fertility rates, and other than the 1 to 2% increased risk of a second mole, do not have an increased incidence of pregnancy-related complications. Furthermore, there does not appear to be an increase in congenital abnormalities among the offspring of women treated for GTN with single agent MTX or ACT regimens.

Similar to patients with moles, hCG surveillance with contraception to avoid intercurrent pregnancy is recommended. Pregnancy can be encouraged after 12 months of normal hCG values. Early obstetrical ultrasound and chest x-ray with uterine shielding are encouraged during subsequent pregnancies. Similar to moles, hCG values should be evaluated after delivery.

High-risk gestational trophoblastic neoplasia

Even patients in the high risk group have a relatively good prognosis for survival, with overall survival rates exceeding 85 to 90%% and 75% survival rates reported for patients presenting with brain metastases. Optimal survival is achieved when patients with high-risk GTN are managed by a physician or center with experience treating patients with this disease. Aggressive multiagent regimens and coordination of multimodality treatment may be required and acute toxicity is more likely than among patients with low-risk GTN.

After remission, hCG surveillance with effective contraception is recommended for at least one year. The prospects for future fertility may be compromised, compared to patients with low-risk GTN, but successful pregnancies after EMA/CO therapy have been reported. Patients with GTN who are treated with etoposide-containing regimens have earlier menopause by only a few years compared to those treated with single agent MTX or ACT.

Exposure to etoposide also increases the risk of secondary malignancies, compared to patients treated with MTX or ACT or to the general population. During long-term follow-up, increases in leukemia and colorectal cancers were the most frequent second malignancies. Patients treated with EMA/CO have an overall risk of leukemia of approximately 0.5%, but this rises to more than 2% when exposed to more than six cycles of EMA/CO.

Patients should be aware of this increased risk and the need for annual complete blood counts and appropriate colorectal screening after treatment.

What if scenarios.

Pre-therapy hCG definition

It should be emphasized that the FIGO risk score and other systems that segregate patients into high and low-risk groups use the pre-therapy hCG value to assess risk. This refers to the hCG value determined at the time GTN is diagnosed, rather than an hCG obtained at the time of a mole evacuation.

Incomplete radiographic staging

The majority of patients diagnosed with postmolar GTN have relatively low hCG values and will have a low risk of metastatic disease if a chest X-ray is normal and the uterus does not contain a large volume of disease on ultrasound.

However, at least 10% of patients with high-risk sites of metastatic GTN are reported to have had negative chest X-rays at initial staging. Up to 40% of GTN patients with negative chest X-rays have small pulmonary metastases detected on chest CT scans. To avoid missing a potentially curable site of high-risk metastasis, it is recommended that patients with nonmolar GTN, and those with molar GTN having hCG levels more than 1,000 IU/L, undergo complete radiographic evaluation before initiating therapy.

Use of combination therapy in low-risk gestational trophoblastic neoplasia refractory to initial chemotherapy

Less than 10% of patients with low-risk GTN will need combination chemotherapy. Because of the leukemogenic potential with etoposide-based regimens, it is recommended that patients with low risk GTN be re-evalutated with a FIGO risk score if they fail initial single agent chemotherapy.

If they have a risk score less than 6, they can be safely treated with an alternative single agent regimen, with a good chance of success. Furthermore, patients treated with a single-dose MTX or ACT regimen may respond to a multi-dose regimen using the same agent, thus avoiding multiagent chemotherapy or hysterectomy in most of these patients.

Delays in chemotherapy recycling

Because of the potential for rapid growth of gestational choriocarcinoma, initial cycles of chemotherapy should be administered aggressively. If EMA/CO cannot be administered on a 15 day schedule because of hematologic toxicity, filgrastim support or dropping the CO portion of the chemotherapy should be considered.

Maintenance chemotherapy

As noted above, maintenance chemotherapy beyond normalization of the hCG value reduces the incidence of recurrent GTN. This is especially true in patients with high-risk GTN. Even if the patient has had significant toxicity from prior therapy, every effort should be made to give 1 to 2 cycles of maintenance chemotherapy in low-risk disease and at least 3 cycles in high-risk disease.

Follow-up surveillance and therapy/management of recurrences.

Recurrent gestational trophoblastic neoplasia

Remission is monitored with at least monthly hCG levels for one year, when the risk of recurrence falls below 1%. During this interval, it is critical that patients prevent an intercurrent pregnancy because elevated hCG from a normal pregnancy would mask tumor hCG.

In high-risk patients, additional hCG monitoring is usually performed for an additional year at 3-6 month intervals. In a series reported from Duke, over 95% of recurrences occurred within 2 years of treatment, and all occurred within 36 months of treatment. Overall, 68% of patients were successfully treated for long term remissions, with a significant improvement in outcome among patients treated after 1978.

Patients diagnosed with a recurrence should be carefully evaluated for metastatic disease and risk score; the majority will require multiagent chemotherapy. In many patients however, metastatic disease is often limited and hysterectomy or metastectomy was incorporated into treatment of many of the Duke patients who were successfully treated for recurrent disease.

Pathophysiology

Hydatidiform mole

Hydatidiform moles are chromosomally abnormal gestations caused by faulty fertilization of the oocyte. Partial and complete moles have distinct cytogenetic origins, histologic features, clinical presentation, and different risks for postmolar GTN.

Partial moles usually have a complete trisomy with two X chromosomes, comprising a haploid set of maternal and a diploid set of paternal chromosomal complement. 69, XYY partial moles have not been observed. More than 90% of partial moles have 69, XXX karyotype suggesting that unispermic fertilization with duplication of the spermic chromosomes is the primary mechanism for development.

In contrast, complete moles have a diploid karyotype, with all chromosomal material derived from the paternal genome. Similar to partial moles, 95% have a 46 XX karyotype, suggesting that dispermic fertilization is a relatively rare cause of hydatidiform mole.

In partial moles, the chorionic villi usually have subtle edema and focal trophoblastic proliferation, with histologic evidence of fetal tissues. Usually this consists of nucleated fetal red blood cells, but sometimes the fetus will persist much later into the gestation.

Complete moles, in contrast, have diffuse edema of the villi with central "clearing" and diffuse trophoblastic proliferation. Patients who undergo evacuation of a complete mole in the early first trimester often have less exuberant trophoblastic proliferation identified histologically. In complete moles, the fetus does not develop in an identifiable form; reabsorption occurs before development of the circulatory system and vessels with fetal rbcs are not identified. Serum hCG levels reflect the relative bulk of trophoblastic tissue; therefore partial moles tend to have lower hCG levels at diagnosis than complete moles.

Clinically, partial moles tend to present at an earlier gestational age at diagnosis than complete moles. The majority of partial moles are identified clinically and with ultrasound as "missed abortions" because of the subtle villous edema. Uterine size is typically less than expected for gestational dates.

In the past complete moles typically presented with bleeding, uterine enlargement, gestational dates, with absent fetal heart tones, and with clinically significant vaginal bleeding. Up to 25% would have medical complications related to the mole, such as secondary hyperthyroidism, respiratory distress, pregnancy induced hypertension, anemia, or theca lutein ovarian cysts caused by ovarian hyperstimulation by elevated levels of hCG.

With the advent of liberalized indications for early obstetrical ultrasounds, up to 60% of moles are currently diagnosed clinically as "missed abortions" at an earlier gestational age, with a concomitant decrease in the incidence of medical complications related to the mole.

Postmolar GTN is diagnosed in up to 20% of patients with complete moles and less than 5% in patients with partial moles. Patients with hCG values above 100,000 IU/L, uterine enlargement of over 12-14 weeks gestational size, theca lutein ovarian cysts, aged over 35 years, or medical complications of a mole, have a risk of postmolar GTN of over 30% that increases to over 50% if more than one factor is present. However, earlier diagnosis of complete moles has not had an impact on the incidence of postmolar GTN.

Currently, more than two thirds of patients with postmolar GTN had no risk factors identified at the time of molar evacuation. This emphasizes the need for close hCG monitoring after the evacuation of hydatidiform mole.

Gestational trophoblastic neoplasia

Normal trophoblast seeks out maternal endometrial spiral arterioles to provide the placenta and fetus with oxygen and nutrients during pregnancy. When the underlying endometrium is abnormally thin, as during implantation of a normal placenta over a prior cesarian section scar, there can be invasion directly into the myometrium producing placenta accreta.

Trophoblastic cells can also be harvested from the maternal circulation during normal pregnancies. This natural tendency for local invasion is exaggerated in GTN lesions, and the trigger for this behavior appears to be the autocrine hyperglycosylated form of hCG which promotes invasion and inhibits apoptosis.

In normal placentas, this is produced in the extravillous cytotrophoblasts, while normal hCG, which supports luteal function, is produced by the syncytiotrophoblasts. The proportion of hyperglycosylated hCG drastically falls from approximately half of placental production early in a normal pregnancy to only 1% in the second or third trimesters. Invasive and metastatic forms of GTD secrete an increased proportion of hyperglycosylated hCG compared to noninvasive lesions, which is roughly proportional to the total hCG production by these lesions.

Invasive moles histologically invade directly into the myometrium and can produce metastatic deposits of typical molar villi. The myometrial invasion can lead to uterine rupture and hemorrhage, vaginal metastases can cause bleeding, and pulmonary metastases can produce hemoptysis. Despite the locally aggressive nature of complete moles, however, most primary and secondary tumors of invasive moles have limited longevity unless there is transformation to gestational choriocarcinoma. Disseminated metastasis beyond the pulmonary system is rarely documented.

In contrast to invasive moles, gestational choriocarcinoma, PSTT, and epithelioid trophoblastic tumors represent true malignant transformation with the ability to produce sustained metastatic behavior.

Choriocarcinoma is the most common form of nonmolar GTN and has a characteristic dimorphic cell population: plump polygonal cytotrophoblasts and multinucleate syncytiotrophoblasts. Chorionic villi are not identified. The syncytotrophoblast cells secrete abundant levels of intact molecules of alpha and beta hCG into the serum. Hyperglycosylated hCG, abnormal forms of "nicked" or free chains of hCG may be produced by individual tumors, but in the majority of patients, standard quantitative serum assays of total beta hCG provide a reliable correlation to disease volume and may be used to follow patients during therapy and remission.

These lesions initially invade into the myometrial venous circulation, with metastases directly to the lungs and/or pelvis. Vaginal metastases are not common, but present as soft bluish nodules or tumors underlying the vaginal mucosa. These are highly vascular and should not be biopsied because most will respond promptly to chemotherapy. The lung is the most frequent site of metastasis identified in patients with GTN.

Furthermore, except in rare patients with significant right-to-left vascular shunts, the vast majority of patients with metastasis to other organs have radiographic evidence of pulmonary metastasis. Brain, liver, kidney, and bowel are most frequent sites of nonpulmonary distant metastases, but involvement of almost any organ has been reported, including the placenta and even the fetus when choriocarcinoma has been diagnosed during pregnancy. Tumor half-life, as indirectly measured by increasing serum hCG levels, can be very rapid with doubling of serum hCG within 7 to 14 days among some patients with choriocarcinoma.

PSTT can follow any type of pregnancy event. Histologically, the tumors are comprised of pleomorphic intermediate trophoblast cells that often stain positively for human placental lactogen (HPL) and infiltrate into the myometrium. Intact hCG is secreted at much lower levels than in gestational choriocarcinoma and HPL is rarely useful as a serum tumor marker. Serum free beta subunit levels of hCG may be elevated and serve as a tumor marker in the majority of patients with PSTT. Lymphatic metastasis occurs more frequently than in choriocarcinoma.

Epithelioid trophoblastic tumors are comprised of relatively well-differentiated intermediate trophoblast cells. They tend to form nodular masses with local expansion and often bulky involvement of the uterus. Hyaline necrosis is frequently associated with the tumor nodules and can be mistakenly identified as keratin, leading to a false histological diagnosis of squamous cell carcinoma. Serum hCG is most often identified, but usually at much lower levels than in patients with choriocarcinoma. These tumors tend to metastasize later in their natural history than gestational choriocarcinomas.

What other clinical manifestations may help me to diagnose gestational trophoblastic disease?

History

Hydatidiform mole

Patients with molar pregnancies may present with symptoms of vaginal bleeding or passage of tissue, hyperemesis of pregnancy, symptoms of secondary hyperthyroidism, or may be aware of rapid uterine enlargement.

Gestational trophoblastic neoplasia

Vaginal bleeding after molar evacuation should raise concern for postmolar GTN. Irregular bleeding after a normal pregnancy should prompt evaluation with a serum hCG. Some patients with nonmolar GTN will give a history of repeated evaluations for a "miscarriage" with elevated hCG values that are never followed to normal values.

Hemoptysis or neurological symptoms or after pregnancy in the absence of pre-eclampsia should raise the concern for metastatic GTN. The overwhelming majority of patients with high-risk metastases from GTN have symptoms related to the site of extrapulmonary disease including headaches or seizures with brain metastases, abdominal pain with liver or spleen metastases, hematuria with renal metastases, and gastrointestinal bleeding with intestinal metastases.

Physical exam

Hydatidiform mole

Uterine enlargement beyond dates, especially with uterine enlargement more than 12 to 14 weeks with absent fetal heart tones is a classical finding in complete moles, especially when combined with bleeding during early pregnancy. Pregnancy-induced hypertension with proteinuria occurring before 20 weeks gestation is unusual but almost pathognomonic of a hydatidiform mole.

Gestational trophoblastic neoplasia

The majority of patients have lesions that are detected through hCG testing or radiographic evaluation. The uterus is often enlarged on pelvic examination. A soft, bluish cyst involving vaginal or central vulvar mucosa may indicate metastatic GTN. These are very vascular lesions and should not be biopsied without considering the diagnosis of GTN. Patients with GTN should have a neurological screening examination, abdominal examination for hepatosplenomegaly, and testing for hematuria and occult fecal blood.

What other additional laboratory studies may be ordered?

Limitations of convention serum hCG assays

Most clinicians have been trained to think of hCG as a single hormone that serves as a very sensitive tumor marker for GTN (and some other nongestational germ cell tumors). The sensitive polyclonal hCG radioimmunometric assays developed in the early 1970s yielded a value that reflected the sum of all major forms of the hCG molecule, including normal intact hCG, hyperglycosylated hCG, free beta subunit, and other "nicked" fragments of hCG.

In the 1980s and 1990s, monoclonal antibody technology was developed that led to the currently used immunometric sandwich assays incorporating dye or chemoluminescent or fluroimmunometric assays that are widely used today.

Most commercially available assays use nonhuman (e.g. murine, goat or rat) monoclonal antibodies directed against epitopes on the alpha hCG chain as the "capture" or immobilizing antibody and a monoclonal antibody directed against epitopes on the beta hCG chain as the labeling antibody. Quantification of these complexes is proportional to the hCG levels in the serum. These assays have been optimized to measure normal intact hCG in the serum of patients with normal pregnancies.

Unfortunately, most of these do not recognize hyperglycosylated hCG, and by design do not measure free beta chain hCG. Therefore, the "total" hCG measured by the assay may not correlate with the hCG forms that are secreted by an individual patient's GTD. Fortunately, the total hCG measured by conventional assays is usually proportional to the sum of the hyperglycosylated hCG secreted by invasive hCG, and in most cases will be an effective tumor marker.

This defect in the design of conventional assays must be recognized, however, in cases of GTD when the hCG value does not seem to correlate with the clinical course of the patient. Assays using capture and labeling monoclonal antibodies directed against different epitopes of the beta chain of hCG appear to be more effective in detecting hyperglycosylated hCG, in addition to detecting most forms of "nicked" hCG and free beta chain hCG.

Furthermore, some laboratories specializing in prenatal screening may offer an assay for hyperglycosylated hCG, which is used in some centers for screening patients for Down syndrome and other abnormalities of pregnancy.

What's the evidence?

Alazzam, M, Tidy, J, Hancock, BW. "First-line chemotherapy in low risk gestational trophoblastic neoplasia". Cochrane Database Syst Rev. vol. 1. 2009. pp. CD007102.

Cole, LA, Butler, SA, Khanlian, SA. "Gestational trophoblastic diseases: 2". Hyperglycosylated hCG as a reliable marker of active neoplasia. Gynecol Oncol. vol. 102. 2006. pp. 151-9.

Cole, LA, Khanlian, SA, Muller, CY. "Gestational trophoblastic diseases: 3". Human chorionic gonadotropin free beta-subunit, a reliable marker of placental site trophoblastic tumor. Gynecol Oncol. vol. 102. 2006. pp. 160-4.

Cole, LA. "New discoveries on the biology and detection of human chorionic gonadotropin". Reprod Biol Endocrinol. vol. 7. 2009. pp. 8.

Cole, LA. "Biological functions of hCG and hCG-related molecules". Reprod Biol Endocrinol. vol. 8. 2010. pp. 102.

Covens, AL, Filiaci, VL, Burger, RA. "Phase II trial of pulse dactinomycin as salvage therapy for failed low-risk gestational trophoblastic neoplasia: a gynecologic oncology group study". Cancer. vol. 107. 2006. pp. 1280-6.

Edge, SB, Byrd, DR, Compton, CC. AJCC Cancer Staging Manual (ed 7). Springer SBM, LLC. 2009.

Evans, AC, Soper, JT, Clarke-Pearson, DL. "Gestational trophoblastic disease metastatic to the central nervous system". Gynecol Oncol. vol. 59. 1995. pp. 226-30.

"Staging classifications and clinical practice guidelines of gynaecologic cancers". Int J Gynecol Obstet. vol. 70. 2000. pp. 207-312.

Growden, WB, Wolfberg, AJ, Feltmate. "Postevacuation hCG levels and risk of gestational trophoblastic neoplasia among women with partial molar pregnancies". J Reprod Med. vol. 51. 2006. pp. 871-4.

Hancock, BW, Welch, EM, Gillespie, AM. "A retrospective comparison of current and proposed staging and scoring systems for persitent gestational trophoblastic disease". Int J Gynecol Cancer. vol. 10. 2000. pp. 318-22.

Hanna, RK, Soper, JT. "The role of surgery and radiation therapy in the management of gestational trophoblastic disease". Oncologist. vol. 15. 2010. pp. 593-600.

Homesley, HD. "Single-agent therapy for nonmetastatic and low-risk gestational trophoblastic disease". J Reprod Med. vol. 43. 1998. pp. 69-74.

Hunter, F, Raymond, E, Christensen, C. "Efficacy of the metastatic survey in the staging of gestational trophoblastic disease". Cancer. vol. 65. 1990. pp. 1647-50.

Jordan, S, Randall, LM, Karamurzin, Y. "Differentiating squamous cell carcinoma of the cervix and epithelioid trophoblastic tumor". Int J Gynecol Cancer. vol. 21. 2011. pp. 918-22.

Kang, WD, Choi, HS, Kim, SM. "Weekly methotrexate (50 mg/m(2)) without dose escalation as a primary regimen for low-risk gestational trophoblastic neoplasia". Gynecol Oncol. vol. 117. 2010. pp. 477-80.

Kim, DS, Moon, H, Kim, KT. "Effects of prophylactic chemotherapy for persistent tophoblastic disease in patients with complete hydatidiform mole". Obstet Gynecol. vol. 67. 1986. pp. 690-4.

Kohorn, EI. "The new FIGO 2000 staging and risk factor scoring system for gestational trophoblastic disease: description and critical assessment". Int J Gyecol Cancer. vol. 11. 2001. pp. 73-7.

Kohorn, EI. "Is lack of response to single-agent chemotherapy in gestational trophoblastic disease associated with dose scheduling or chemotherapy resistance?". Gynecol Oncol. vol. 85. 2002. pp. 36-9.

Lan, C, Li, Y, He, J. "Placental site trophoblastic tumor: lymphatic spread and possible target markers". Gynecol Oncol. vol. 116. 2010. pp. 430-7.

Lertkhachonsuk, AA, Israngura, N, Wilailak, S. "Actinomycin D versus methotrexate-folic acid as the treatment of stage I, low-risk gestational trophoblastic neoplasia: a randomized controlled trial". Int J Gynecol Cancer. vol. 19. 2009. pp. 985-8.

Limpongsanurak, S. "Prophylactic actinomycin D for high-risk complete hydatidiform mole". J Reprod Med. vol. 46. 2001. pp. 110-6.

Lurain, JR, Nejad, B. "Secondary chemotherapy for high-risk gestational trophoblastic neoplasia". Gynecol Oncol. vol. 97. 2005. pp. 618-23.

Lurain, JR, Singh, DK, Schink, JC. "Management of metastatic high-risk gestational trophoblastic neoplasia: FIGO stages II-IV: risk factor score > or = 7". J Reprod Med. vol. 55. 2010. pp. 199-207.

Lurain, JR. "Gestational trophoblastic disease I: epidemiology, pathology, clinical presentation and clinical presentation and diagnosis of gestational trophoblastic disease and management of hydatidiform mole". Am J Obstet Gynecol. vol. 203. 2010. pp. 531-9.

Lurain, JR. "Gestational trophoblastic disease II: classification and management of gestational trophoblastic neoplasia". Am J Obstet Gynecol. vol. 204. 2011. pp. 11-8.

Mutch, DG, Soper, JT, Baker, ME. "Role of computed axial tomography of the chest in staging patients with nonmetastatic gestational trophoblastic disease". Obstet Gynecol. vol. 68. 1986. pp. 348-52.

Mutch, DG, Soper, JT, Babcock, CJ. "Recurrent gestational trophoblastic disease". Experience of the Southeastern Regional Trophoblastic Disease Center. Cancer. vol. 66. 1990. pp. 978-82.

Newlands, ES, Bagshawe, KD, Gegent, RHJ. "Results with the EMA/CO (etoposide, methotrexate, actinomcin D, cyclophosphamide, vincristine) regimen in high risk gestational trophoblastic tumors, 1979 to 1989". Br J Obstet Gynaecol. vol. 98. 1991. pp. 550-7.

Newlands, ES, Mulholland, PJ, Holden, L. "Etoposide and cisplatin/etopside, methrotrexate and actinomycin D (EMA) chemotherapy for patients with high-risk gestational tumors refractory to EMA/cyclophosphamide and vincristine chemotherapy and patients presenting with metastatic placental site trophoblastic tumors". J Clin Oncol. vol. 18. 2000. pp. 854-9.

Osborne, RJ, Filliaci, V, Schink, JC. "Phase III trial of weekly methotrexate or pulsed dactinomycin for low-risk gestational trophoblastic neoplasia: a gynecologic oncology group study". J Clin Oncol. vol. 29. 2011. pp. 825-31.

Palmer, JA. "Advances in the epidemiology of gestational trophoblastic disease". J Reprod Med. vol. 39. 1994. pp. 155-62.

Rustin, GJS, Newlands, ES, Begent, RHJ. "Weekly alternating etoposide, methotrexate, and actinomycin/vincristine and cyclophosphamide chemotherapy for the treatment of CNS metastases of choriocarcinoma". J Clin Oncol. vol. 7. 1989. pp. 900-3.

Schmid, P, Nagai, Y, Agarwal, R. "Prognostic markers and long-term outcome of placental-site trophoblastic tumours: a retrospective observational study". Lancet. vol. 374. 2009. pp. 48-55.

Seckl, MJ, Sebire, NJ, Berkowitz, RS. "gestational trophoblastic disease". Lancet. vol. 376. 2010. pp. 717-29.

Shen, X, Xiang, Y, Guo, L. "Analysis of clinicopathological prognostic factors in 9 patients with epithelioid trophoblastic tumor". Int J Gynecol Cancer. vol. 21. pp. 1124-30.

Smith, HO. "Gestational trophoblastic disease epidemiology and trends". Clin Obstet Gynecol. vol. 46. 2003. pp. 541-56.

Soper, JT, Clarke-Pearson, DL, Berchuck, DL. "5-day methotrexate for women with metastatic gestational trophoblastic disease". Gynecol Oncol. vol. 54. 1994. pp. 76-9.

Soper, JT, Evans, AC, Rodriguez, GL. "Wtoposid-platin combination chemotherapy for chemorefractory gestational trophoblastic disease". Gynecol Oncol. vol. 56. 1995. pp. 421-4.

Soper, JT, Mutch, DG, Schink, JC. "American College of Obstetricians and Gynecologists: Diagnosis and treatment of gestational trophoblastic disease: ACOG Practice Bulletin No. 53". Gynecol Oncol. vol. 93. 2004. pp. 575-85.

Soper, JT. "Gestational trophoblastic disease". Obstet Gynecol. vol. 108. 2006. pp. 176-87.

Soper, JT, Spillman, M, Sampson, JH. "High-risk gestational trophoblastic neoplasia with brain metastases: individualized multidisciplinary therapy in the management of four patients". Gynecol Oncol. vol. 104. 2007. pp. 691-4.

Tidy, JA, Gillespie, AM, Bright, N. "Gestational trophoblastic disease: a study of mode of evacuation and subsequent need for treatment with chemotherapy". Gynecol Oncol. vol. 78. 2000. pp. 309-12.

Wolfberg, AJ, Berkowitz, RS, Goldstein, DP. "Postevacuation hCG levels and risk of gestationa trophoblastic neoplasia in women with complete molar pregnancy". Obstet Gynecol. vol. 106. 2005. pp. 548-52.

Yarandi, F, Eftekar, Z, Shojaei, H. "Pule methotrexate versus pulse actinomycin D in the treatment of low-risk gestational trophoblastic neoplasia". Int J Gynaecol Obstet. vol. 103. 2008. pp. 33-7.

Loading links....
You must be a registered member of Psychiatry Advisor to post a comment.

Sign Up for Free e-newsletters