Placental mesenchymal dysplasia (PMD) is a rare placental abnormality initially described in the 1990s 1. PMD, previously called “pseudo-partial mole,” is characterized by placentomegaly, dilated and tortuous chorionic vessels, vesicles or cysts on macroscopic examination, and enlarged stem villi with cistern-like formation without trophoblastic proliferation on microscopic examination 2,3. Immunohistochemical assessment of the paternally imprinted, maternally expressed p57 gene usually reveals discordant expression in dysplastic stem villi with positive staining in the cytotrophoblast (CT) and negative staining in stromal cells (SC) 3,4. The cause of PMD is uncertain, but imbalanced gene imprinting in the 11p15.5 region seems to be implicated and androgenetic/biparental mosaicism has been demonstrated in some cases 5,6.
Complete hydatidiform mole (CHM) is a pathology of human pregnancy characterized by absence of embryonic development, edematous avascular villi with cistern formation, budding growth pattern, stromal karyorrhexis, trophoblastic proliferation, and loss of p57 immunohistochemical expression in the CT and SC on microscopic examination 7. CHMs are diandric with an exclusively paternal-derived genome.
The association of CHM with PMD and a live fetus has been reported twice, but there was no clear pathologic description and no confirmation of the diagnosis by molecular genotyping 2,8. Here, we describe the full characterization of a term twin pregnancy with both PMD and CHM. Clinical, ultrasound, macroscopic, microscopic, immunohistochemical, and genetic analyses are reported.
A healthy 34-yr-old woman, gravida 4, para 2 (1 spontaneous first trimester miscarriage and 2 uneventful term pregnancies) was referred at 32.6 weeks of gestation (WG) following the discovery on ultrasound examination of multicystic areas in the placenta. Previous ultrasound scans at 12 and 22 WG showed a singleton intrauterine pregnancy with no apparent abnormality. Maternal serum β-human chorionic gonadotrophin (βhCG) and pregnancy associated plasma protein A levels were 1.58 multiples of the median and 0.70 multiples of the median at 13 WG, respectively. The total hCG assay was 14,748 IU/L at 6.6 WG and 25,690 IU/L at 32.4 WG. Ultrasonography at 36.3 WG (Fig. 1A) revealed a homogeneous hypoechoic zone measuring 102×90×70 mm and occupying one third of the placenta. The remaining two thirds were clearly delimited from the rest of the placenta and had microvacuoles with a Swiss cheese-like appearance scattered throughout. PMD was suspected and a caesarean section was performed at 39.1 WG giving birth to a 2905 g female infant, who was healthy and is developing normally at the time of writing (20 mo of age). There were no signs either of Beckwith-Wiedemann syndrome, or of hemangioma or hepatic tumor. The mother’s hCG levels returned to normal within 10 wk.
Gross placental examination revealed 2 placental masses which separated spontaneously upon handling (Fig. 1B). The main placenta measured 16.5×14×2.5 to 5 cm and weighed 508 g. The chorionic vessels were dilated and tortuous with aneurysmal dilatations. The umbilical cord was 13 cm in length and 10 mm in diameter. Numerous thin-walled cysts were observed on the cut surface across 60% of the placenta; the remaining 40% was macroscopically normal (Fig. 1C). The second mass measured 9×6×2.5 cm and weighed 97 g. It consisted of grape-like vesicles resembling molar tissue. There was no umbilical cord or fetal membrane.
Microscopic examination of the main placenta demonstrated that the parenchymal cystic areas were enlarged edematous stem villi with cistern formation admixed with a few normal terminal villi (Fig. 1D). The SC nuclei were sometimes enlarged and irregular. Thick-walled vessels and focal chorangiosis were observed (Fig. 1E). There was no trophoblastic proliferation. Fetal vascular malperfusion was observed with large vessel thrombi and avascular villi. The grossly normal part of the main placenta showed microscopic features of normal mature placenta. Immunohistochemistry (p57Kip2 Ab-6, Thermoscientific, Fremont, 1:400) revealed p57-discordant expression on dysplastic stem villi with p57-positive nuclear staining in the CT and loss of p57 expression in SC (Fig. 1F), while mature terminal villi had weak staining in both the CT and SC. These features are characteristic of PMD.
Microscopic examination of the detached vesicular mass showed large avascular villi with cistern formation and trophoblastic proliferation (Fig. 1G), some of which were necrotic. There were no terminal villi. Immunohistochemistry demonstrated loss of p57 expression in both the CT and SC (Fig. 1H) consistent with the diagnosis of CHM.
Multiplex microsatellite DNA genotyping was performed on formalin-fixed paraffin-embedded tissue from macrodissected decidua, normal-appearing terminal villi, PMD areas, and CHM areas (Fig. 2). A panel of 16 STR markers was studied, using the AmpFlSTR Identifier Plus kit (Applied Biosystems, Carlsbad, CA) according to the manufacturer’s instructions. Normal-appearing villi were diploid biparental (MP1). The PMD areas had an excess of paternal genome in all informative loci with a paternal to maternal allele ratio ranging from 2:1 to 3:1. There were no maternal alleles in any of the informative loci in the CHM regions. The presence of a single allele in CHM is indicative of a diandric monoallelic genome. The paternal allele was similar in normal-appearing villi, and in the PMD and CHM areas, suggesting that the paternal contribution came from a single sperm (MP1 for normal-appearing villi and PMD, and P1P1 for the CHM).
To confirm the diploid genome of the normal, dysplastic and molar villi, fluorescent in situ hybridization was performed on formalin-fixed paraffin-embedded tissue sections. Interphase nuclei were hybridized with probes for the centromeric region of chromosomes 3, 7, and 17 (ZytoVision, Bremerhaven, Germany). This analysis confirmed the diploidy and demonstrated the presence of 2 signals in both the CT and SC analyzed from different regions of the normal, molar and mesenchymal dysplasia areas [see Figure, Supplemental Digital Content 1, https://links.lww.com/IJGP/A124, which demonstrates diploidy in the SC (white arrow) and in the CT (black arrow head), fluorescent in situ hybridization with CEP 3, 7, and 17].
We report herein an extremely rare case of combined CHM and PMD. Both are characterized by lacunar ultrasound images. The rule in CHM is that there is no fetal development; however, an embryo/fetus can be observed in some rare cases of CHM 9, especially in twin pregnancies consisting of CHM with a coexisting fetus 10–12. Although CHMs are usually diagnosed during the first trimester on ultrasound screening, our case was detected late, at 32.6 WG. Giorgione et al. 11 reported a detection rate of 92% at 17±2.7 GW, and Massardier et al. 12 a detection rate of 71% at the end of the first or in the second trimester in their series of CHMs with a coexisting fetus. In a recent study 13, abnormal features were observed on ultrasound examination in 75% of PMD placentas in the first trimester and in 90% of cases in the second trimester. This confirms that as in our patient, cases of CHM and PMD are sometimes missed during routine first or second trimester ultrasound scans.
Similarly, maternal serum hCG levels were not elevated in our patient. This is usual in PMD where serum hCG concentrations are normal in the majority of cases and it is more common for α-fetoprotein levels to be elevated 14. In CHM, hCG levels are usually >100,000 IU/L. However, this is rarely reported in CHM with a coexisting fetus. This may be because molar villi often undergo necrosis as the pregnancy progresses. Weekly postpartum hCG measurements are required following CHM because of the risk of progression to gestational trophoblastic neoplasia (GTN). According to some studies, this complication is more frequent in CHM with a coexisting fetus than in classic CHM 11,12. Although her postpartum hCG levels decreased slowly, our patient did not progress to GTN according to FIGO criteria.
CHM is characterized by an exclusive paternal contribution to the genome due to fertilization of an empty oocyte by 2 sperms (dispermy) or by 1 sperm followed by endoreduplication (monospermy). This diandry leads to stromal overgrowth and trophoblast hyperplasia and precludes embryonic development. CHM progresses to GTN in 9% to 20% of cases 7. PMD is a pathology of the villous mesenchyme (stroma and vessels) in which embryonic development is maintained, leading to a normal fetus in the majority of cases. However, 25% of cases are associated with Beckwith-Wiedemann syndrome, and fetal or neonatal tumors such as hemangioma or mesenchymal hamartomas, predominantly involving the liver, have been reported 3,5,6. Our case had morphologic and immunohistochemical features of both these entities. However, given the rarity of this association and the implications for clinical management and prognosis of the patient and the child, we confirmed the diagnosis by genetic analysis (Fig. 2).
Twin pregnancy with CHM and a coexisting fetus is a rare condition affecting 1 in 22,000 to 1 in 100,000 pregnancies and is associated with complications such as fetal death, vaginal bleeding, pre-eclampsia, preterm delivery, and risk of GTN. In 2 recent studies of twin pregnancies with CHM and a coexisting fetus, Giorgione et al. 11 found that only 3/13 cases (23%) lasted until the third trimester and Massardier et al. 12 just 1/14 cases (7%). The combination of PMD with a normal placenta in a twin pregnancy is also very rare. Cases of dichorionic or monochorionic twin pregnancies associating 1 normal placenta and 1 placenta with PMD features have been reported 5,15,16. Although the association of p57-discordant villi (similar to PMD findings) and CHM are sometimes encountered on p57 immunostaining performed on first trimester products of conception 17, second or third trimester twin pregnancies similar to the present case featuring both PMD and CHM have only been described twice 2,8. Paradinas et al. 2 reported the presence of a twin pregnancy with a complete mole in 2 of their series of 15 “pseudo-partial” moles, both ending with stillbirths at 17 GW and 27 GW, but details of the pathologic examinations were not reported. McCowan and Becroft reported a female livebirth at 27 GW with Beckwith-Wiedemann features and a separate cystic mass on macroscopic examination 8. Its microscopic features were consistent with CHM and PMD, but p57 immunohistochemistry was not performed.
In keeping with previous molecular genotyping analyses of PMD tissues 4, the paternal to maternal allele ratio in the present case was close to 3:1. This genotype pattern, in combination with p57 immunostaining and fluorescent in situ hybridization analysis, is suggestive of androgenetic/biparental mosaicism with a biparental genome in the trophoblast and an androgenetic genome in SC 7. On the other hand. the CHM component was purely androgenetic. As our results also show that the same haploid paternal set contributed to all cell lineages (MP1 for normal-appearing villi, MP1/P1P1 in PMD tissue, and P1P1 for the CHM), a likely explanation is that the oocyte was fertilized by a single haploid sperm. Figure 3 illustrates the proposed mechanism for the pathogenesis of this rare case, whereby fertilization of 1 oocyte by 1 sperm followed by 2 abnormal duplications of the paternal genome gives rise to 2 cell populations, 1 androgenetic and 1 biparental, which are then unevenly distributed in the fetal and placental tissues, resulting in confined placental mosaicism. However, we cannot formally rule out the possibility that mosaicism may not have been confined to the placenta and that androgenetic tissues were also present in the fetus, as previously described 18,19, warranting postnatal follow-up of the infant.
In conclusion, this case illustrates the importance of pathologic evaluation of placentas that appear lacunar on ultrasound examination. Pathologists should be aware that, although the presence of a fetus is usually in favor of either PMD or CHM with a coexisting fetus, attentive macroscopic examination, extensive sampling, and microscopic examination with p57 immunohistochemistry are necessary to avoid missing a possible association of both CHM and PMD. This rare condition then requires molecular genotyping to confirm the diagnosis and appropriately manage the patient and the child.
1. Moscoso G, Jauniaux E, Hustin J. Placental vascular anomaly with diffuse mesenchymal stem villous hyperplasia: a new clinico-pathological entity? Pathol Res Pract 1991;187:324–8.
2. Paradinas FJ, Sebire NJ, Fisher RA, et al. Pseudo-partial moles: placental stem vessel hydrops and the association with Beckwith-Wiedemann syndrome and complete moles. Histopathology 2001;39:447–54.
3. Faye-Petersen OM, Kapur RP. Placental mesenchymal dysplasia. Surg Pathol Clin 2013;6:127–51.
4. Armes JE, McGown I, Williams M, et al. The placenta in Beckwith-Wiedemann syndrome: genotype-phenotype associations, excessive extravillous trophoblast and placental mesenchymal dysplasia. Pathology 2012;44:519–27.
5. Kaiser-Rogers KA, McFadden DE, Livasy CA, et al. Androgenetic/biparental mosaicism causes placental mesenchymal dysplasia. J Med Genet 2006;43:187–92.
6. Robinson WP, Lauzon JL, Innes AM, et al. Origin and outcome of pregnancies affected by androgenetic/biparental chimerism. Hum Reprod 2007;22:1114–22.
7. Ronnett BM. Hydatidiform moles: ancillary techniques to refine diagnosis. Arch Pathol Lab Med 2018;142:1485–502.
8. McCowan LME, Becroft DMO. Beckwith-Wiedemann syndrome, placental abnormalities, and gestational proteinuric hypertension. Obstet Gynecol 1993;83:813–7.
9. Baergen RN, Kelly T, McGinniss MJ, et al. Complete hydatidiform mole with a coexistent embryo. Hum Pathol 1996;27:731–4.
10. Suksai M, Suwanrath C, Kor-Anantakul O, et al. Complete hydatidiform mole with co-existing fetus: predictors of live birth. Eur J Obstet Gynecol Reprod Biol 2017;212:1–8.
11. Giorgione V, Cavoretto P, Cormio G, et al. Prenatal diagnosis of twin pregnancies with complete hydatidiform mole and coexistent normal fetus: a series of 13 cases. Gynecol Obstet Invest 2017;82:404–9.
12. Massardier J, Golfier F, Journet D, et al. Twin pregnancy with complete hydatidiform mole and coexistent fetus: obstetrical and oncological outcomes in a series of 14 cases. Eur J Obstet Gynecol Reprod Biol 2009;143:84–7.
13. Gavanier D, Allias F, Huissoud C, et al. Ultrasound findings and clinical outcomes in 23 cases of placental mesenchymal dysplasia. J Reprod Med 2017;62:366–75.
14. Guenot C, Kingdom J, De Rham M, et al. Placental mesenchymal dysplasia: an underdiagnosed placental pathology with various clinical outcomes. Eur J Obstet Gynecol Reprod Biol 2019;234:155–64.
15. Surti U, Hill LM, Dunn J, et al. Twin pregnancy with a chimeric androgenetic and biparental placenta in one twin displaying placental mesenchymal dysplasia phenotype. Prenat Diagn 2005;25:1048–56.
16. Gheysen W, Strybol D, Moerman P, et al. Discordance for placental mesenchymal dysplasia in a monochorionic diamniotic twin pregnancy: a case report. Clin Case Rep 2018;6:1557–60.
17. Gaillot-Durand L, Patrier S, Aziza J, et al. p57-discordant villi in hydropic products of conception: a clinicopathological study of 70 cases. Hum Pathol 2020;101:18–30.
18. Repnikova E, Roberts J, Kats A, et al. Biparental/androgenetic mosaicism in a male with features of overgrowth and placental mesenchymal dysplasia. Clin Genet 2018;94:564–8.
19. Reed RC, Beischel L, Schoof J, et al. Androgenetic/biparental mosaicism in an infant with hepatic mesenchymal hamartoma and placental mesenchymal dysplasia. Pediatr Dev Pathol 2008;11:377–83.