Don’t Just Go with the Flow:
STR Analysis of Molar Pregnancy
Geiersbach KG, Baker C, Samowitz WS
Introduction
Materials and Methods
Case 1
Molar pregnancy is defined as a nonnon-viable pregnancy with
trophoblastic proliferation and an imbalance in the number of
haploid sets of paternal and maternal chromosomes. Molar
pregnancies occur in about 1/1000 to 1/2000 pregnancies in the
western hemisphere, and are more common in some Asian
countries. Molar pregnancies typically abort in the first
trimester, and they are associated with a risk for persistent
trophoblastic disease. The most concerning trophoblastic disease
is choriocarcinoma, which can occur in up to 3% of women with
a history of complete molar pregnancy. Choriocarcinoma is an
aggressive malignancy, but it is highly responsive to
chemotherapy.
Complete molar pregnancy is characterized by florid
trophoblastic proliferation in the context of greatly reduced or
absent embryonic tissue. Clinically, complete molar pregnancy
is characterized by greatly elevated hCG levels and by the
absence of an embryo on ultrasonography. Complete moles are
nearly always diploid and are considered complete uniparental
disomy because they consist solely of paternal DNA (excluding
mitochondrial DNA). Complete moles are most commonly
associated with fertilization of an “empty”
empty” ovum (egg) lacking
nuclear DNA by one spermatozoon (sperm) that subsequently
duplicates its DNA. Complete mole may also result from
fertilization of an empty egg by two sperm. Since complete
moles are diploid, cytogenetic studies and flow cytometry are of
limited diagnostic utility. Currently, many laboratories use
histopathologic features and clinical history to diagnose
complete molar pregnancy. Immunohistochemical staining for
p57 has also been utilized in some centers, with variable
success.
Partial molar pregnancy is characterized by less florid
trophoblastic proliferation, often in the context of an
incompletely formed embryo or fetus. Ploidy analysis by flow
cytometry can distinguish triploid populations. However,
triploidy is a common occurrence (about 1 in 100 conceptions),
and only a third of these cases are truly partial moles. STR
analysis can identify triploidy and can also identify the parental
parental
origin of the extra haploid set of chromosomes. Maternally
derived triploid conceptions do not result in partial molar
pregnancies. A little over half of paternally derived triploid
conceptions result in partial molar pregnancies.
The histopathologic features of products of conception from very
early molar pregnancies may overlap considerably with those
from nonnon-molar pregnancies. NonNon-molar pregnancies may
exhibit features of molar pregnancies, including hydropic villi.
Genetic studies of short tandem repeat (STR) polymorphisms
offer a robust and reliable method for definitively identifying
complete molar pregnancy (see Table 1). In addition, STR
analysis may be performed on blood, fresh tissue, or formalin
fixed paraffin embedded tissue. Since the latter is often the only
only
specimen available, STR analysis has become a useful ancillary
technique for pathologic diagnosis of complete molar pregnancy.
Case 3
*
Case 4
Case 2
mother (decidua).
Interpretation
STR result
Mechanism
FIGURE 1: Histologic appearance of complete mole (cases 1 and 2), partial mole (case 3),
and hydropic change (case 4).
Case
Case
Case
Case
1:
2:
3:
4:
Case 2
One case had a triallelic pattern at D13S317, and was
demonstrated to have 3 copies of RB1 by FISH, consistent with
trisomy 13 (Figures 3 and 4). Complete moles were
characterized by an increased S phase population by ploidy
analysis, but this flow cytometry pattern was also observed in
two partial moles and one non-molar pregnancy sample.
Case 3
Conclusions
Case 4
REFERENCES:
Pathology Diagnosis
STR interpretation
Flow Cytometry
complete mole
complete mole (monospermy) CASE 1
diploid
complete mole
complete mole (dispermy) CASE 2
diploid
complete mole
complete mole (monospermy)
diploid
complete mole
complete mole (monospermy)
diploid
suspicious for partial mole
non-molar pregnancy
diploid
suggestive of partial mole
non-molar pregnancy
diploid
partial mole
partial mole (dispermy)
triploid
partial mole
partial mole (dispermy) CASE 3
triploid
Normal fertilization
partial mole
non-molar pregnancy
diploid
Fertilization of haploid egg by 2 sperm†
partial mole
partial mole (dispermy)
triploid
Fertilization of haploid egg by 1 diploid sperm‡
partial mole
non-molar pregnancy
diploid
suspicious for partial mole
partial mole (dispermy)
triploid
Complete mole
Single allele per locus; non-identity
Fertilization of empty egg by diploid sperm†
suspicious for partial mole
non-molar pregnancy
diploid
Complete mole
Two alleles per locus; non-identity
Fertilization of empty egg by 2 sperm
hydropic change
failed; unable to isolate decidua
diploid
Maternally derived triploidy
3 alleles or 1:2 ratio (pat:mat)
Fertilization of diploid egg by 1 sperm ‡
hydropic change
non-molar pregnancy CASE 4
diploid
Trisomy / copy number variant
3 alleles at 1 locus; 1:1 ratio at other loci
Nondisjunction error in meiosis
hydropic change
non-molar pregnancy
diploid
hydropic change
non-molar pregnancy
diploid
hydropic change
non-molar pregnancy
diploid
hydropic change
non-molar pregnancy
diploid
hydropic change
non-molar pregnancy
diploid
Alleles identical at all loci
Microdissection failure
Results
STR analysis was successfully performed on 19 of the 20 cases
(Table 2). In one sample of villous hydropic change, decidua
could not be isolated from the sample. Maternal blood was not
available to complete STR analysis on this case. 4 cases
histologically classified as complete mole and 7 cases
histologically classified as hydropic change were confirmed by
STR analysis. 5 out of 9 cases histologically classified as
partial mole or suspicious for partial mole were re-classified as
non-molar pregnancy by STR analysis; the remaining 4 cases
histologically classified as partial mole were confirmed by STR
analysis. Flow cytometry results were concordant with STR
analysis results for all cases of hydropic change or partial mole,
but flow cytometry did not identify complete moles.
1 shared allele per locus
Contamination
DNA was amplified with the AmpFlSTR Identifiler kit (Applied
Biosystems) and subsequently analyzed with GeneMapper
(Applied Biosystems). Ploidy analysis by flow cytometry was
performed on all 20 samples. Fluorescence in situ hybridization
(FISH) for the RB1 locus on chromosome 13q14 was employed
to investigate a triallelic pattern in one non-molar pregnancy
sample (Case 4).
Case 1
3 alleles or 2:1 ratio (pat:mat*)
* pat = paternal; mat = maternal
† most common mechanism ‡ diploid zygote resulting from meiosis II error or duplication of haploid content (homozygous)
(case 1, left) and partial mole (case 3, right). Blue asterisks denote
diploid population (G0-G1 and G2-M). Red asterisks denote triploid
population (G0-G1 and G2-M).
Monoallelic pattern with non-identity: complete mole (monospermy).
Biallelic pattern with non-identity: complete mole (dispermy).
Triallelic pattern with 2:1 paternal: maternal contribution: partial mole (dispermy).
Biallelic pattern showing normal Mendelian inheritance.
Normal Mendelian inheritance
2:1 ratio at all loci (pat:mat)
*
FIGURE 3: STR plots. Villi (conceptus) on top; decidua (mother) below.
Partial mole
Partial mole
** *
FIGURE 2: Example flow cytometry histograms for complete mole
TABLE 2: Results
TABLE 1: Interpretation of STR results of conceptus (villi) in comparison to
*
20 cases of complete mole (n=4), partial mole (n=9), and villous
hydropic change (n=7) were selected from archived formalin
fixed, paraffin embedded products of conception (Table 2).
Areas of villi and decidua were microdissected from aniline
blue stained slides and extracted overnight with proteinase K.
1. Hayashi K et al. Hydatidiform mole in the United States (1970
(1970-1977): a
statistical and theoretical analysis. Am J Epidemiol. 1982;115:671982;115:67-77.
Short tandem repeat (STR) analysis of formalin-fixed, paraffin
embedded products of conception shows high sensitivity and
specificity in classifying complete and partial molar
pregnancies. In addition, STR analysis reveals the different
genetic mechanisms causing molar pregnancies.
Ploidy
analysis by flow cytometry does not detect complete molar
pregnancies, and histologic classification of molar pregnancy is
not as reliable as STR analysis. Autosomal trisomies and rare
mutations of STR loci may be detected by STR analysis, but
these entities do not result in misclassification with the use of
a carefully constructed reporting algorithm. STR analysis is
highly specific and is therefore a useful method for evaluating
the possibility of molar pregnancy and the need for followfollow-up.
2. Palmer JR. Advances in the epidemiology of gestational trophoblastic
trophoblastic disease.
J Reprod Med. 1994;39:1551994;39:155-62.
3. Petignat P, Billieux MH, Blouin JL, Dahoun S, Vassilakos P. Is genetic
analysis useful in the routine management of hydatidiform mole? Hum
Reprod. 2003;18:2432003;18:243-9.
FIGURE 4: Fluorescence in situ
hybridization (FISH) of trophoblastic cell
showing 3 signals at RB1 locus (13q14,
green) and 2 signals at D21S259,
D21S341, and D21S342 (21q22.13q22.2, red), consistent with trisomy 13
(Case 4). Decidua showed a normal
diploid pattern at RB1.
4. Pfeifer JD. Molecular Genetic Testing in Surgical Pathology. New York:
Lippincott Williams and Wilkins, 2006.
5. Seckl MJ et al. Choriocarcinoma and partial hydatidiform moles. Lancet.
2000;356:362000;356:36-9.
6. Zaragoza MV et al. Parental origin and phenotype of triploidy in spontaneous
abortions: predominance of diandry and association with the partial
partial
hydatidiform mole. Am J Hum Genet. 2000;66:18072000;66:1807-20.
Acknowledgements
The authors wish to thank Carl Wittwer and Rae Lynn Wies
(Flow Cytometry Laboratory, ARUP Laboratories), Maria
Bettinson (Fragment Analysis Laboratory, ARUP), and Karl
Voelkerding and Maria Erali (Institute for Advanced
Technology, ARUP) for facilitating this project.