ETTs and PSTTs are rare gestational trophoblastic neoplasms, with respective differentiation toward chorionic-type ITs and implantation-site ITs . The non-neoplastic counterpart of an ETT is a PSN, whereas that of a PSTT is an exaggerated placental-site reaction (EPSR). The pathogenesis of ETTs and PSTTs remains obscure. A recent study demonstrated a lack of a biologic link between an EPSR and PSTTs . However, atypical PSNs with morphologic features intermediate between typical PSNs and ETTs were described [4, 9]. Transformation of a PSN through an atypical PSN to a malignant ETT was also reported . In a clinicopathologic study of 14 cases of ETTs, two cases were intimately associated with PSNs . Those findings suggest that PSNs have the potential to develop into ETTs. The transition from a PSN to a coexisting ETT and PSTT in the current case further supports the concept. Moreover, a similar distribution of ETTs and PSNs in the uterus also implies a link between them. Approximately half of ETT cases and 40% of PSN cases are located in the lower uterine segment and/or upper endocervix [3, 4, 9, 11, 12]. A recent study showed a lack of a Y-chromosome complement in >80% of cases of ETTs . Analysis of the Y allele in PSNs may be helpful to further clarify the relationship between ETTs and PSNs.
Normal placental-site tissue is expelled from the uterus within several weeks postpartum . Factors contributing to the retention of ITs that lead to PSN formation remain unclear. Much evidence supports surgical interventions, including CSs and therapeutic abortions, being related to PSN formation. Many patients with PSNs have a history of a therapeutic abortion and/or CS [4, 11]. Fistulous tracts of the lower uterine segment, coated by ITs resembling PSNs, were described in patients who had received a CS [14–17]. The endometrium damaged by a CS or therapeutic abortion might not undergo cyclic shedding during subsequent menstrual cycles. Perturbation of endometrial shedding may contribute to PSN formation . The present case showed distribution of PSNs at the CS site, which supports the relationship between PSN formation and surgical interventions.
Interestingly, our case showed mixed features of an ETT and PSTT. The frequency of ETTs combined with other gestational trophoblastic neoplasms is not low. In fact, the histologic, ultrastructural, and immunohistochemical features of ETTs were first described in lungs of patients with a choriocarcinoma following intensive chemotherapy [18, 19]. The largest clinicopathologic study of ETTs, that included 14 patients, showed three cases combined with a choriocarcinoma and/or a PSTT . Only one case of a mixed ETT and PSTT without a choriocarcinoma component was found in the literature . The 15-year-old patient had a history of a preceding complete mole 1 year before the diagnosis of the mixed ETT and PSTT. Abnormal vaginal bleeding was the presenting symptom. The serum β-HCG level was not available. The tumor was limited to the uterus. She underwent a hysterectomy without postoperative chemotherapy and had no evidence of tumor recurrence or metastasis 84 months later. The two cases of a mixed ETT and PSTT, including ours, had clinical presentations similar to those of a pure PSTT or ETT [3, 7, 20, 21]. With a pure PSTT or ETT, abnormal vaginal bleeding is the most common presenting symptom, and serum β-HCG levels are elevated in approximately 80% of patients. Because only two cases of a mixed ETT and PSTT were reported, it is difficult to establish the prognostic factors of the tumor. These two cases, both of which had a tumor limited to the uterus, like most patients with a stage I pure ETT or PSTT, had a favorable prognosis after a hysterectomy. The clinical behavior of a mixed ETT and PSTT seems similar to that of a pure ETT or PSTT, but more cases are required to elucidate its prognosis. Long-term follow-up is essential for patients with a mixed ETT and PSTT, because patients with a stage I pure ETT or PSTT occasionally have local recurrence or distant metastasis during the follow-up period [3, 7, 20, 21]. As the β-HCG level was not elevated in 20% of patients with a PSTT or ETT, glypican 3 was recently suggested to be a potential serum tumor marker of gestational trophoblastic neoplasms . The role of glypican 3 as a tumor marker needs further clinical evaluation.
The hybrid features of the present tumor were consistent with the proposed model of pathogenesis of gestational trophoblastic neoplasms [2, 4]. In this model, transformed trophoblastic stem cells retain differentiation plasticity. Stem cells are able to differentiate toward cytotrophoblast, syncytiotrophoblast, chorionic-type ITs and implantation-site ITs. This model also explains the sensitivity of trophoblastic neoplasms to chemotherapy. Choriocarcinomas, the most primitive trophoblastic tumors, are sensitive to chemotherapy. Conversely, ETTs and PSTTs are not responsive to chemotherapy due to their more-differentiated natures. Because of the differentiation plasticity of neoplastic trophoblasts, it is clinically critical for pathologists to thoroughly examine pathologic specimens, particularly small curettage samples, to identify components of trophoblastic neoplasms: ETTs, PSTTs, choriocarcinomas, or their combinations. Therapeutic approaches to gestational trophoblastic tumors rather depend on the differentiation of neoplastic trophoblasts. An immunohistochemical panel including cytokeratin 18, HLA-G, HSD3B1, β-HCG, β-catenin, hPL, CD146, p63, and Ki-67 can be of great help in establishing a correct diagnosis .