OBJECTIVES: This study was designed to examine whether cytokeratin 20 (CK20) is expressed in molar pregnancies and may therefore be used in the diagnosis of gestational trophoblastic disease (GTD). The potential of CK20 expression in predicting the evolution and the prognosis of the different subtypes of GTD was also assessed. METHODS: A total of 48 samples were studied for CK20 expression by RT-PCR methodology. Among these, 24 samples were obtained by curettage of the uterine cavity of patients diagnosed with hydatidiform mole (14 complete moles and 10 partial moles), 4 samples were obtained from choriocarcinoma cell lines (2 JAR and 2 JEG), and 20 samples were of normal trophoblast (control group) obtained from patients that underwent elective termination of pregnancy. RESULTS: Expression of CK20 was identified in all the samples of complete mole (CM), all choriocarcinoma cell lines, and 50% of the patients with partial mole (PM). None of the preparations of normal trophoblastic tissue from the control group expressed the CK20. A significant difference (P < 0.00001) was found in CK20 expression between samples of patients with GTD and control samples. Comparison between CK20 expression in CMs and PMs revealed a significantly more frequent expression of CK20 in CMs (P = 0.006). More than 50% of the patients with PMs that were positive for CK20 had an invasive evolution. CONCLUSIONS: In our opinion, CK20 may assist in distinguishing between molar and normal trophoblastic tissue and may be considered a marker of GTD. In cases in which pathological classification of different subtypes of GTD is in doubt, CK20-positive expression is suggestive for a CM whereas CK20-negative is more indicative for PM.
OBJECTIVES: This study was designed to examine whether cytokeratin 20 (CK20) is expressed in molar pregnancies and may therefore be used in the diagnosis of gestational trophoblastic disease (GTD). The potential of CK20 expression in predicting the evolution and the prognosis of the different subtypes of GTD was also assessed. METHODS: A total of 48 samples were studied for CK20 expression by RT-PCR methodology. Among these, 24 samples were obtained by curettage of the uterine cavity of patients diagnosed with hydatidiform mole (14 complete moles and 10 partial moles), 4 samples were obtained from choriocarcinoma cell lines (2 JAR and 2 JEG), and 20 samples were of normal trophoblast (control group) obtained from patients that underwent elective termination of pregnancy. RESULTS: Expression of CK20 was identified in all the samples of complete mole (CM), all choriocarcinoma cell lines, and 50% of the patients with partial mole (PM). None of the preparations of normal trophoblastic tissue from the control group expressed the CK20. A significant difference (P < 0.00001) was found in CK20 expression between samples of patients with GTD and control samples. Comparison between CK20 expression in CMs and PMs revealed a significantly more frequent expression of CK20 in CMs (P = 0.006). More than 50% of the patients with PMs that were positive for CK20 had an invasive evolution. CONCLUSIONS: In our opinion, CK20 may assist in distinguishing between molar and normal trophoblastic tissue and may be considered a marker of GTD. In cases in which pathological classification of different subtypes of GTD is in doubt, CK20-positive expression is suggestive for a CM whereas CK20-negative is more indicative for PM.