Evolution of treatment of high-risk metastatic gestational trophoblastic tumors.

Gestational trophoblastic tumors (GTT) encompass a spectrum of neoplastic disorders that arise from placental trophoblastic tissue after abnormal fertilization. There are several systems that are currently in use to classify GTT. Investigators recognized certain clinical factors that appeared to portend a worse prognosis and classified metastatic patients into low-risk groups and high-risk groups. Because single-agent therapy is inadequate for patients with high-risk metastatic tumors, intensive combination chemotherapy is mandatory.

Currently, EMA-CO (etoposide, methotrexate, dactinomycin, cyclophosphamide, and vincristine [Oncovin]) is the preferred regimen. The treatment is generally well tolerated, with no life-threatening toxic effects, and reproductive function is preserved in most of the patients. Unfortunately, about 25–30% of patients with high-risk metastatic disease become refractory and fail to achieve a complete remission. There is no standard salvage chemotherapy regimen with EMA-CO failures; however, almost all salvage regimens are platinum based and paclitaxel is an investigational drug in this setting.

Discussion

Since the introduction of methotrexate for the treatment of GTT more than 50 years ago, the cure rate for these tumors has exceeded 90% overall and deaths are mostly confined to the high-risk metastatic group. Earlier, we have reported the results of our unit in treating all classes of GTT. In the current report, we extend and present our results concerning the treatment of the high-risk metastatic group during the past 13 years.

According to the NIH classification, patients with GTT were classified into nonmetastatic, low-risk metastatic, and high-risk metastatic GTT. This system is very practical and is the basis on which most major US trophoblastic disease centers determine treatment and report results. Initially, the EMA-CO regimen was used at our unit as a salvage treatment after failure of MAC in patients with high-risk metastatic GTT. This combination requires 1 day admission every 2 weeks and therefore is very convenient from the patient's point of view. It will induce reversible alopecia and, to some extent, cumulative myelosuppression. However, recent analysis has confirmed that there is a small but significant increase in second tumors, particularly acute myeloid leukemia.

We observed, although the numbers were small (n= 9, 6 primary and 3 salvage), that the decrease in the hCG level was lesser after omitting the CO arm than after omitting the EMA arm, and knowing that the most active agents are methotrexate, dactinomycin, and etoposide, we decided to omit the CO arm, which by intuition will be less toxic due to the reduction in the total dose given without sacrificing the cure rate. EMA was the first-line therapy in nearly two thirds (46/70) of our patients, achieving primary remission in nearly 70% of patients (32/46) comparable to those reported in the literature. To our knowledge, only reports by Matsui et al. and Soto-Wright et al. showed the efficacy of this regimen. They achieved a primary remission rate of 71–78% in patients treated primarily for high-risk metastatic GTT using a regimen similar to ours. Knowing that our study is retrospective, we think that such therapy should be tested in properly randomized controlled collaborative trials before recommending it widely because of its lesser toxicity.

Adjuvant surgical intervention may be useful in removing resistant tumor foci in selected patients. In our study, one patient had a right lower lung lobectomy due to lung abscess formation complicating pulmonary metastatic involvement. The patient was cured and had successfully conceived and delivered twice following remission and enjoying a normal life 5 years later.

Although the survival of patients with metastatic GTT who were treated with hysterectomy alone is dismal, hysterectomy continues to play an important role when integrated with chemotherapy. The total duration of hospital stay and chemotherapy is shorter, with a lower total dosage in women who underwent primary hysterectomy than in those treated with chemotherapy alone. Therefore, primary hysterectomy may be considered for selected patients with high-risk metastatic GTT with a small extrauterine tumor burden, but it is not as beneficial in this group, and if performed in face of disseminated metastases, it is unlikely to be of any benefit. In the current study, 15 patients (21.43%) had hysterectomy, but unfortunately, it was not of great value in most high-risk metastatic patients; 82.46% (47/57) of patients with high-risk metastatic GTT were cured without having hysterectomy, and one third (5/15) of patients died of disease despite having it.

Of women with metastatic GTT, around 10% will have central nervous system metastases. In some centers, whole brain irradiation has been used in an attempt to prevent intracranial hemorrhage and to treat brain metastases. On the contrary, the Charing Cross group recommended early craniotomy combined with high-dose EMA-CO and intrathecal methotrexate in order to eradicate brain metastases. While, at the King Faisal Hospital and Research Center using PEA-M regimen, they could achieve remission in more than 50% of their patients. Similarly, Ghaemmaghami et al. using EMA-EP with radiotherapy could salvage 56% of their patients. Chemotherapy alone therefore induces a high remission rate in patients with high-risk metastatic GTT and brain involvement. However, at Charing Cross, craniotomy for resection of drug-resistant lesions is justified only rarely and then only for patients who do not have evidence of metastatic disease elsewhere.

In general, craniotomy is reserved for women who require acute decompression of central nervous system hemorrhagic lesions to allow stabilization and institution of therapy. In the current study, one patient had a craniotomy for impending conization and was salvaged by high-dose EMA-CO, while the only patient receiving irradiation died of her disease. Unfortunately, the lady having craniotomy relapsed 8 years later with liver involvement and succumbed to her disease. We could salvage only half (2/4) of our patients with sole brain involvement but none in cases of additional hepatic involvement (0/3).

Hepatic involvement is reported in around 5–10% of patients with metastatic disease and denotes a very ominous sign, with more than two thirds of patients dying from disease in our experience (9/13). Unfortunately, the optimal management of such patients has not been established, with radiation used in some patients to minimize the risk of fatal intraabdominal hemorrhage; also, selective lesion chemoembolization, hepatic artery embolization, and partial hepatectomy have been described to control hepatic hemorrhage.

It is known that primary resistance occurs in about 25–30% of high-risk metastatic patients, and a good proportion of these could be salvaged by second- and third-line combination chemotherapy. Newlands et al. included cisplatinum in the relapse schedule for patients who failed on the EMA-CO regimen and the combination of EMA-EP and salvage surgery and have achieved remission in approximately 70% of patients in this category. Recently, they updated their results, reporting an overall 88% remission rate (30/34) in patients with high-risk metastatic GTT who developed resistance to EMA-CO with EMA-EP regimen. However, they stressed that the EMA-EP schedule is moderately toxic due to the administration of weekly chemotherapy that combines a renally excreted drug, methotrexate, with a potentially nephrotoxic drug, cisplatinum. The myelosuppression is cumulative in some patients, necessitating the administration of granulocyte colony–stimulating factors, with delays in chemotherapy. With the EMA-CO regimen and second-line regimens based on cisplatinum, bleomycin, etoposide, or vinblastine, an expected 85% remission rate has been reported in the high-risk metastatic GTT patients.

Similarly, Xiang et al. reported similar findings, with more than 75% response rate. However, in the current study, we were able to salvage only less than two thirds (63.6%) of the relapsed patients (14/22). Second-line treatment salvaged 10 out of 22 patients (45.45%), and third-line treatment salvaged four out of ten patients (40%). It is to be noted that originally, out of the 14 patients salvaged, 3 had received MAC and were cured by EMA-CO. If we exclude those receiving MAC, this leaves us with 19 patients with 57.9% salvage rate. This is in agreement with the results reported by John I. Brewer Trophoblastic Disease Center where they have salvaged only 61.5% (16/22) of their relapsed patients. On the other hand, Matsui et al. reported a high salvage rate, reaching more than 80% using 5-fluorouracil–containing regimens, and Wan et al. confirmed these findings in their report.

It is believed that initiating EMA-EP as a first-line therapy in extremely high-risk groups as those with brain and/or liver involvement will decrease the rate of relapse and resistance, an approach we are adopting currently. We consider resistance to cisplatinum a very grave sign. Unfortunately, we do not have the facility to give high-dose therapy with stem cell support, and we have used third-line treatment mostly in the form of weekly paclitaxel and cisplatinum, with two out of four patients surviving. In vitro studies and several anecdotal reports showed activity and remission achieved by paclitaxel in such situations. Other drugs showing activity include 5-fluorouracil, campothecins, and gemcitabine, but further testing is needed. It is possible that high-dose combination chemotherapy with stem cell support will prove effective against the few tumors that prove resistant. Innovative experimental approaches, like the use of monoclonal antibody–directed therapy, in which a substantially greater dose of a cytotoxic agent is delivered at the tumor site, will add to the armamentarium against chemoresistant GTT.

Confirming our first report, the current study showed that four (50%) of our eight patients who died due to disease had low levels of hCG (less than 100 mIU/mL) at the time of their death, denoting that in long-duration disease and under the pressure of intensive prolonged chemotherapy, selective growth and evolution of cell clones with different characteristics and decreased hCG production occur. Accordingly, the importance of hCG levels in monitoring such patients would be of little value and response should be assessed using other methods.

Using univariate and multivariate analyses, our study found the presence of brain and/or liver metastases to be the worst prognostic sign, followed by the type of antecedent pregnancy and resistance to combination chemotherapy, which is in agreement with other studies. No significance was found for patients older than 39 years or duration of the disease more than 4 months from the antecedent pregnancy. Although these results are not in total agreement with the Charing Cross experience or the new FIGO staging, we would like to acknowledge the small number in our retrospective series (total 70 patients), and the fact that the proposed prognostic factors have been chosen on the basis of univariate analysis on patients treated by simple regimens, and each factor was assumed to act independently, neglecting that many are closely interrelated. The data do not suggest that one system is superior to any of the scoring systems; we chose the NIH classification, despite some of its limitations, due to its practicality, simplicity, and reliability.

Conclusions

Although relapse is not always ominous and effective second-line treatment is available, it is still far from optimal. Hopefully, the increased availability of hCG monitoring will increase early diagnosis in patients with persistent GTT and thereby reduce the occurrence of brain and/or liver involvement. While tremendous advances have been made in the treatment of GTT, we must not become complacent but intensify our efforts to improve the outcome in high-risk metastatic GTT. It is hoped that in these extremely high-risk categories, elaboration of more intensive and innovative combination chemotherapy with stem cell support will achieve superior results to those currently reported. With continued collaboration between different centers and development of new drugs, we can expect better improvement in the outcome of this uncommon interesting disease.