Cytogenetics of trophoblasts from complete hydatidiform moles

The role of deoxyribonucleic acid image cytometric and interphase cytogenetic analyses in the differential diagnosis, prognosis, and clinical follow-up of hydatidiform moles. A report from the Central Molar Registration in The Netherlands

OBJECTIVES

To assess the value of deoxyribonucleic acid ploidy in the differential diagnosis and clinical follow-up of hydatidiform moles, the histopathologic features, deoxyribonucleic acid ploidy, and clinical follow-up were compared in 347 cases: 143 complete moles, 52 partial moles, and 152 abortions, of which 56 cases were hydropic abortions with histologic features of triploidy but lacked trophoblastic hyperplasia.

STUDY DESIGN

In all cases deoxyribonucleic acid image cytometry was performed, and in 85 of these cases interphase cytogenetics was also performed.

RESULTS

With use of deoxyribonucleic acid image cytometry and interphase cytogenetics, a bimodal polyploid deoxyribonucleic acid pattern was present in 97% of complete moles, 27% of partial moles, and 4% of abortions. All these cases of partial mole were reclassified to complete mole on the basis of this deoxyribonucleic acid pattern and the histopathologic features in spite of the presence of fetal blood cells, amnion, or yolk sac. Deoxyribonucleic acid triploidy was found in 95% of the remaining partial moles, in 77% of hydropic abortions with histologic features of triploidy, and in 14% of the remaining abortions. Reliable differentiation between deoxyribonucleic acid triploid partial moles and hydropic abortions with histologic features of triploidy was not possible on basis of the histopathologic features (trophoblastic hyperplasia) or 3.5c exceeding rates. Deoxyribonucleic acid diploidy was found in 1% of complete moles, 23% of hydropic abortions with features of triploidy, and 78% of the remaining abortions. Deoxyribonucleic acid tetraploidy was rarely found (1% of complete moles, 2% of partial moles, 1% of abortions). Persistent gestational trophoblastic disease developed in 33% of the bimodal deoxyribonucleic acid polyploid cases (all complete moles), in 1% of the diploid cases (concerning one of the two diploid complete moles), and in 1% of the triploid cases (partial moles).

CONCLUSION

Deoxyribonucleic acid analysis is essential in the diagnosis of hydatidiform moles to decide on clinical follow-up.

Sister chromatid exchange and chromosome breakage in complete hydatidiform moles

Women with complete hydatidiform moles (CHM) are at a 10% risk for developing persistent trophoblastic disease or choriocarcinoma. We studied sister chromatid exchange (SCE) as a prognostic indicator for malignancy in peripheral blood lymphocytes (PBL) from women with CHM and their husbands, but found no differences from normal control couples. SCE levels in cultured tissue derived from 11 CHM (avg. 7.9) and 2 choriocarcinomas (avg. 6.8) were not significantly different from those of 8 normal skin fibroblast cultures (avg. 7.8). These same tissues were then examined for chromosome breakage which was significantly higher for CHM (0.48/cell) and choriocarcinoma (0.87/cell) than normal fibroblasts (0.33/cell). Chromosome breaks occurred at 50-60% known fragile sites and at 50-55% of cancer breakpoints. Whereas SCE was only associated with 13% of breaks in the three tissues, half of these were at known fragile sites. Our results suggest that SCE is not an indicator of malignancy in PBL or cultured cells from CHM or choriocarcinoma and that the level of SCE is not elevated in CHM or choriocarcinoma. However, our results confirm the increased breakage seen in the latter two tissues which may represent general DNA instability predisposing to choriocarcinoma and its accompanying chromosomal rearrangements.

Spontaneous chromosome fragility in chorionic villus cells

Human fragile sites are only very rarely expressed spontaneously. In this paper we report the presence of non-random spontaneous chromosome lesions (CL) in chorionic villus samples and their coincidence with fragile site (FS) bands. The average number of CL was about 9% both in RPMI-1640 and in Chang media. To determine any possible influence of external factors other than culture media, the results were grouped according to age of gestation. No differences were observed among the different groups. A total of 101 chromosome lesions could be precisely identified by sequential Leishman Staining/Wright G-banding; 76.2% of them coincided with FS-bands. The most affected region was at 1q12-1q21.1 (15.8% of total CL); other FS with a clustering of breakpoints in our study were 1p36, 1q44, 2q37, 3p24, 3q27, 10q22 and 16q23. These results suggest that spontaneous expression of some FS could be a characteristic of embryonic tissues.

Chromosomal rearrangement in choriocarcinoma cell lines

Malignant trophoblastic cells from four choriocarcinoma cell lines were evaluated in detail using Q, G, and C banding at various passages. The modal chromosome numbers for BeWo, DoSmi, ElFa, and Jar were 73, 71, 77, and 72, respectively. All the four tumor cell lines exhibited extensive chromosomal rearrangements with several consistent marker chromosomes in each. The majority of these markers have not been previously recognized in this malignancy. Rearrangements of chromosomes 1, 7, 9, 10, and 12 were noted in all four cell lines, but abnormalities of chromosomes 1 and 12 were not consistently present in ElFa and Jar, respectively. Telomeric associations were observed in two cell lines involving chromosomes 11 and 21 as well as chromosomes 3 and 12, resulting in two consistent marker chromosomes. A total of 86 breakpoints were involved in the consistent rearrangements observed in all four cell lines. Most of these breakpoints were located on chromosomes 1, 3, 9, 13, 12, 7, and 21, in order of frequency.