Does This Woman Have Gestational Trophoblastic Disease?
Does This Woman Have Gestational Trophoblastic Disease?
A 36-year-old woman (gravida 2, para 1-1-0-2) had a positive pregnancy test 1 week before she began to have heavy vaginal bleeding with passage of tissue. She brought in a sample of the passed tissue to her primary care physician, which was read at an outside institution as "gestational trophoblastic disease, highly suspicious for choriocarcinoma."
The patient was referred to the University of Colorado Hospital for further evaluation and treatment. Review of her histology slides was not convincing for choriocarcinoma, and the university team favored a diagnosis of early abortus. Since this would make a huge clinical difference in the management of this patient, these physicians decided to send the slides for an extramural consultation to the Johns Hopkins Hospital, where Drs. Russell Vang, Brigitte Ronnett, and Robert Kurman reviewed the slides.
Further work-up was performed at the University of Colorado Hospital. A pelvic ultrasound showed no uterine or endometrial abnormality. A computed tomography (CT) scan of the pelvis, abdomen, and chest revealed no evidence of metastatic disease. When the patient presented to the University of Colorado Hospital, the quantitative beta-human chorionic gonadotropin (HCG) was 3 mIU/mL (normal < 5 mIU/mL) and creatinine was 0.7 mg/mL. Two weeks later, a repeat beta-HCG was < 1 mIU/mL.
Dr. Ali Akalin: What did the microscopic examination of the slides reveal?
Drs. Meenakshi Singh and Russell Vang: The review of the histology slides revealed predominantly decidual tissue with exaggerated placental site and a small focus of trophoblastic tissue composed of cytotrophoblast and syncytiotrophoblast with mild atypia (Figure 1). However, no necrosis or tissue invasion was identified. No villi were seen.
Dr. Akalin: What is your differential diagnosis in this case?
Dr. Singh: The differential diagnosis includes early placenta, choriocarcinoma, hydatiform mole, placental site trophoblastic tumor, and exaggerated placental site.
The characteristic histologic finding in hydatiform mole is enlarged, hydropic villi and trophoblastic proliferation. The fact that no villi are seen in this case argues against a molar disease although the absence of villi on the tissues examined does not completely rule out a molar disease since the villi may have all been passed with the abortus, leaving only implantation site trophoblast, cytotrophoblast, and syncytiotrophoblast. Therefore, appropriate clinical, radiologic, and laboratory follow-up is required to completely rule out this possibility.
The histologic findings in exaggerated placental site and its neo-plastic counterpart, placental site trophoblastic tumor, are similar and would show infiltrative proliferation of monomorphic intermediate-type trophoblastic cells between myometrial smooth muscle bundles without destruction, invasion into blood vessel walls with endothelial permeation, and abundant extracellular, eosinophilic, fibrinoid deposition (Figure 2A and 2B). Therefore, exaggerated placental site and placental site trophoblastic tumor may be confused with one another in a biopsy or curettage specimen.
The features that would favor exaggerated placental site include an absence of mass—forming lesion in the uterus on macroscopic examination and imaging studies, low or absent mitotic activity, intermediate trophoblastic cells admixed evenly with multinucleate trophoblastic cells and the presence of normal chorionic villi. However, the mere absence of chorionic villi in small tissue samples should not exclude the diagnosis of exaggerated placental site when all the other features support this diagnosis, as sampling error may account for this finding.
The absence of villi and the presence of an intimate admixture of cytotrophoblast and syncytiotrophoblast with a dimorphic arrangement and mild cytologic atypia in trophoblastic cells suggest the possibility of choriocarcinoma in this case. However, similar microscopic findings can also be seen in abortus material from very early gestation, as the earliest chorionic villi do not form until day 13 of pregnancy. In addition, although one would expect the presence of villi after day 13, the villi might have been lost with the abortus, leaving only residual implantation site trophoblastic cells, cytotrophoblast, and syncytiotrophoblast in small tissue samples brought in to the clinician.
The histologic requirement for a diagnosis of choriocarcinoma is the presence of tissue invasion. Therefore, appropriate clinical, radiologic, and laboratory follow-up may be required to confidently distinguish between choriocarcinoma and early abortus in small tissue samples. Choriocarcinoma usually manifests as single or multiple well-circumscribed hemorrhagic nodular lesions that may extend deeply into the myometrium in the uterus.
Unlike choriocarcinoma, no mass within the myometrium is seen in early placenta on macroscopic examination and imaging studies. The serial determination of serum beta-HCG is also very important in differentiating choriocarcinoma from early placenta. In choriocarcinoma, one would see very high levels of beta-HCG for the estimated gestational age and a gradual rise during follow-up, whereas, in early placenta, the serum beta-HCG levels would be at the expected reference range for the estimated gestational age and would show a gradual decline to undetectable levels on follow-up, as observed in our case.