Pediatric and Developmental Pathology 6, 69 77, 2002 DOI: 10.

1007/s10024-002-0079-9 2002 Society for Pediatric Pathology

REVIEW

Histopathological Diagnosis of Partial and Complete Hydatidiform Mole in the First Trimester of Pregnancy
NEIL J. SEBIRE,* ROSEMARY A. FISHER, AND HELENE C. REES
Department of Histopathology, Trophoblastic Disease Unit, Charing Cross Hospital, Fulham Palace Road, London W6 8RF, UK
Received July 31, 2002; accepted October 10, 2002; published online December 10, 2002.

ABSTRACT
The diagnosis of molar pregnancy is a continuing diagnostic problem for many practicing histopathologists who are required to examine specimens of products of conception, particularly since changes in gynecological management in recent years have resulted in uterine evacuation at earlier gestations. The aim of this review is to provide practical, up-to-date, diagnostically useful information regarding the histological diagnosis of molar disease in early pregnancy. Pathophysiological issues relevant to molar pregnancies, such as genetic abnormalities, will be briey summarized, but nonhistopathological aspects of molar disease will not be covered in detail in this review. Key words: complete mole, hydatidiform mole, partial mole, trophoblastic disease

INTRODUCTION
Hydatidiform moles (HM) are abnormal conceptions, which occur in about 1 in 500 1000 pregnancies [1]. Pathological and cytogenetic studies have demonstrated that molar pregnancies may be either of two distinct subtypes, complete and partial hydatidiform mole. Complete moles (CM) are usually diploid and androgenetic, pathologically demonstrating minimal embryonal development with hydropic chorionic villi and trophoblastic hyperplasia, while partial moles (PM) are usually pater*Corresponding author, e-mail: [email protected]

nally derived triploid conceptions in which embryonal development occurs in association with trophoblastic hyperplasia [2,3]. The conceptus in molar pregnancies is almost always nonviable and, following diagnosis, molar tissue is evacuated from the uterus by surgical curettage and the patient followed up with serial serum and/or urine human chorionic gonadotropin (hCG) estimations [4]. In the United Kingdom, all cases of hydatidiform mole since 1973 must be registered in a system jointly set up by the Royal College of Obstetricians and Gynecologists and the Department of Health [5]. The aims of registration are to facilitate monitoring of hCG levels following surgical evacuation, to provide optimal management of persistent trophoblastic disease and to provide information for patients and medical staff. Around 1200 patients are registered annually at three centers in London, Shefeld, and Dundee. Following registration, patients undergo hCG monitoring for either 6 months or 2 years depending on their hCG levels at 56 days postevacuation, and all women are requested to notify the center in all subsequent pregnancies for further hCG monitoring, since women with a history of hydatidiform mole are at increased risk of a molar pregnancy in future pregnancies [1].

Clinical importance of accurate diagnosis of partial and complete HM

Persistent gestational trophoblastic disease (pGTD), including overt malignancy such as choriocarcinoma, may

occur following any conception but the risk is signicantly higher following a pregnancy affected by PM or CM. Genetic analysis has now clearly conrmed that choriocarcinoma may arise from either a previous PM or CM, the risk being signicantly greater for CM [6,7]. pGTD in this scenario is usually recognized by either failure of the maternal serum hCG concentration (MShCG) to fall to normal levels, or a rising MShCG. In such cases, chemotherapeutic treatment, usually with methotrexate-based protocols, will lead to complete resolution of disease with hCG levels returning to normal in most cases. In the United Kingdom, about 15% of pregnancies complicated by CM, and 1% of pregnancies complicated by PM, will be affected by pGTD and require chemotherapy [6]. In cases of pGTD in which the index pregnancy was either an unrecognized HM, or following an apparently normal pregnancy, presentation may either be with persistent vaginal bleeding or symptoms/signs of metastatic choriocarcinoma, such as neurological abnormality [6]. Although the majority of cases of such advanced disease also respond well to chemotherapy, the mortality rate is increased compared to pGTD detected at an earlier stage through MShCG screening.

the pathophysiology of HM. The characteristic trophoblastic features of both androgenetic CM and PM are due to the presence of two paternal genomes. Digynic triploids, which have two maternal contributions to the nuclear genome, are not generally associated with molar pathology but instead have an abnormally small placenta and growth-retarded fetus [32]. In conceptions which have no maternal contribution to the nuclear genome, a CM develops, with trophoblastic hyperplasia and little or no fetal development. In PM, the presence of a maternal genome is associated with less trophoblastic hyperplasia and a greater degree of fetal development. Thus overexpression of paternally transcribed genes is likely to play a role in the development of PM, while CM reect both overexpression of paternally transcribed and loss of maternally transcribed genes.

GENETIC DIAGNOSIS OF HM
Although diagnosis of molar pregnancy is usually possible based on morphology alone, poor sampling, necrosis, and earlier uterine evacuations can make a pathological diagnosis difcult [33,34]. In such cases, other techniques may be required to make a rm diagnosis. Particularly useful are those techniques that can be applied to xed material. Assessment of ploidy using in situ hybridization or, more usually, ow cytometry, may be used for distinguishing diploid CM from triploid PM [34 38]. However, examination of ploidy will not distinguish between CM and diploid nonmolar products of conception. More recently, the products of imprinted genes have shown useful in the differential diagnosis of PM and CM. For example p57KIP2, the product of the cyclin dependent kinase inhibitor CDKN1C, shows high levels of expression in normal human placenta [39] but is repressed in the cytotrophoblast of CM [40]. Immunohistochemical staining with antibody to p57KIP2 has been shown to be a reliable discriminator between CM and PM [41]. Since CM are the only type of conceptus to show repression of p57KIP2, this technique also permits a distinction between CM and nonmolar products of conception. However, none of these techniques distinguishes between molar and nonmolar triploids, monospermic and dispermic CM, or androgenetic and biparental CM. To distinguish between these entities, the parental origin of the nuclear chromosomes needs to be determined. DNA polymorphisms, in particular microsatellite polymorphisms [42,43], are highly informative genetic markers and, when used in conjunction with the polymerase chain reaction (PCR), can provide useful information regarding the parental origin of a sample even from small amounts of xed tissue from parafn blocks [44 48]. These types of genetic studies have been used to conrm a diagnosis of CM or PM, and to differentiate between a twin pregnancy with CM and coexistent normal fetus from a PM. Analysis of DNA polymorphisms

Genetic studies in HM
CM have been shown to have a diploid 46,XX or 46,XY karyotype while PM are usually triploid [8 10]. The majority of CM are androgenetic in origin [11,12], all 46 chromosomes in the molar tissue being paternally derived. Most CM are also monospermic and arise following fertilization of an anucleate egg by a haploid sperm which undergoes endoreduplication [1315], while approximately 20% are dispermic [16,17], arising by fertilization of an anucleate egg by two sperm [18]. Despite the fact that the nuclear genome is androgenetic, the mitochondrial DNA in CM, as in a normal conceptus, is maternally derived [19,20]. CM which are diploid but biparental, rather than androgenetic, in origin, although rare, are now well recognized [17,2124]. These unusual CM tend to be associated with recurrent HM or families with several affected individuals [2224] and are thought to represent a familial form of CM. PM, which are triploid, usually arise as a result of fertilization of an ovum by two sperm [25,26] although fertilization of an egg by a single diploid sperm cannot be excluded [27]. Occasional triploid or tetraploid CM have been reported as have tetraploid PM [28]. Again, these usually have an excess of paternal contributions, polyploid CM being androgenetic [28] while tetraploid PM have one maternal and three paternal contributions to the genome [29,30]. A small number of genes are transcribed only from the maternally or paternally inherited allele, the allele inherited from the other parent being imprinted or silent [31]. This phenomenon of genomic imprinting underlies

70

D. RAKHEJA

ET AL.

can be used to differentiate between monospermic (homozygous) CM and dispermic (heterozygous), although the clinical signicance of this distinction has yet to be established. Most recently, molecular genetic diagnosis has been used to differentiate the rare diploid, biparental CM from the more usual androgenetic CM [2224]. It is potentially clinically important to identify diploid, biparental CM, because patients with these have a particularly high risk of subsequent CM; in vitro fertilization techniques, used to avoid subsequent PM or androgenetic CM in patients with recurrent molar pregnancies [49], may not be successful for this group of patients.

pathological criteria are necessary, the discussion of which is the aim of this article.

Histopathological diagnostic features

There are several important questions which the histopathologist must address when examining products of conception in the context of possible molar disease. The rst is to decide whether the chorionic villi show any morphological abnormalities other than those associated with fetal demise. If there are abnormal villous features, do these indicate molar change or other nonmolar abnormal conception such as fetal aneuploidy or mesenchymal dysplasia? Finally, if the features are those of HM, do they indicate CM or PM? Even the rst of these questions may sometimes be difcult to answer in clinical practice since the tissue examined may be limited, either only a single block being submitted or possibly several blocks which contain primarily decidua with only a few scattered chorionic villi present. In such circumstances, although only scanty villi may occasionally be adequate for the diagnosis of CM or PM, it must be accepted that it may be impossible to exclude a PM on histopathological grounds alone on the basis of limited material. It is not possible to be dogmatic regarding the number of chorionic villi required for a sample to be adequate, because in some cases, particularly those affected by CM, a condent diagnosis may be made on the basis of only a small number of villi, whereas in other cases, greater number of villi may be present but these may show much more subtle changes, the interpretation of which requires a more global assessment of the pregnancy. In order to minimize such problems, any case of suspected molar disease, either clinically, sonographically, or on the basis of routine histopathological examination of products of conception, should have all tissue submitted for histopathological examination; if referral to a tertiary center is required, all such blocks/slides should be examined. Once appropriate criteria are used for the diagnosis of HM, even in products from the early rst trimester, the majority of cases of CM or PM in which adequate tissue is submitted for examination will be correctly identied. Some cases of possible PM, particularly those in whom there is limited material available, may require ancillary investigations such as karyotyping or ow cytometry in order to reach a denitive diagnosis. In the United Kingdom, since this number represents only a small proportion of all cases referred for suspected CM or PM ( 4%) [56], in such cases a report may be issued stating the reasons for uncertainty and the most likely diagnosis, following which the patient is registered for MShCG follow up. In cases of nonmolar HA, the MShCG falls rapidly and, once normal levels are achieved, follow-up is curtailed. This policy will result in a small number of nonmolar cases receiving a short MShCG follow-up but minimizes the number of cases of PM which are not

Changes in clinical presentation in relation to diagnostic pathological features

Classically, HM presented as second trimester vaginal bleeding, large for dates uterus, and spontaneous abortion with passage of vesicles per vaginum. Such pregnancies may also be complicated by early onset preeclampsia and features of hyperthyroidism [50,51]. With changes in gynecological investigation and management, this classical presentation of pregnancies affected by HM is now highly unusual [52,53]. In Europe, the majority of clinically recognized pregnancies undergo a routine rst-trimester ultrasound examination at which time fetal number and viability are conrmed. Furthermore, almost all pregnancies presenting with rst-trimester vaginal bleeding are investigated using transvaginal ultrasound examination. These changes in practice have resulted in the majority of molar pregnancies being evacuated in the late rst trimester, either suspected as molar disease sonographically or, more commonly, simply diagnosed sonographically as a nonviable pregnancy (missed abortion) [54]. In a study of 194 consecutive cases of suspected PM or CM referred to the Charing Cross Trophoblastic Disease Centre, the sonographic diagnosis was that of a missed abortion/anembryonic pregnancy with no prenatal suspicion of molar pregnancy in 131 (67%), referral being purely on the basis of routine histological examination of products of conception. In 63 cases, however, ultrasound examination did suggest molar pregnancy; in 84% of these, the diagnosis of molar pregnancy was correct, the remainder being nonmolar hydropic abortions (HA), sonographically mimicking HM [55]. Overall, 58% of CM had a sonographic diagnosis of molar pregnancy compared to only 17% of partial moles, conrming that the majority of HM now present as missed abortion/anembryonic pregnancy, highlighting the importance of routine histological examination of products of conception to diagnose gestational trophoblastic disease (GTD). Furthermore, since diagnosis/evacuation is now usually in the rst trimester of pregnancy, the classic pathological features previously described in cases persisting to the second trimester are no longer appropriate [52], and updated

CHIARI I MALFORMATION

AND

CYSTIC FIBROSIS

71

followed, and limits the workload and cost for laboratories performing ancillary molecular diagnostic techniques.

COMPLETE HYDATIDIFORM MOLE (CM)

Classical histopathological features
The entire specimen consists of large, vesicular chorionic villi with extensive central cistern formation and marked circumferential trophoblastic hyperplasia. No fetal parts, fetal blood vessels, or nucleated fetal red blood cells (RBCs) are present.

orid interstitial extravillus trophoblast invasion, sometimes resulting in interstitial hemorrhage [58] (Fig. 5). Immunohistochemical staining may aid the diagnosis in some difcult cases (see section below).

PARTIAL HYDATIDIFORM MOLE (PM)

Classic histopathological features
Specimen comprises a mixture of two distinct populations of villi, some morphologically normal, others demonstrating an irregular outline, villous edema, cistern formation, and trophoblast hyperplasia. Amnion and fetal parts can be present.

First-trimester histopathological features

Numerous chorionic villi are usually present and may show an overall variation in size consistent with immature intermediate villi from early gestation. Scattered larger, hydropic villi may be present but hydrops is usually not marked in early pregnancy [56]. Similarly, scattered villi with central cistern formation may be seen but this also is not a constant feature, particularly in cases with limited material. The villi show a highly characteristic budding architecture (Figs. 1, 2), with focally marked mucoid/myxoid degeneration of the villous stroma, which may be enhanced using periodic acid-Schiff (PAS) staining. Within the stroma of well-preserved villi, abundant nuclear karyorrhectic debris is present; note that such debris may often be present within fetal vessels in nonmolar failed pregnancies, only stromal debris is signicant for the diagnosis of CM. Villi from cases of CM may often contain blood vessels in early pregnancy, usually collapsed, but very occasionally containing RBCs (Fig. 3). With advancing gestation, the frequency of identiable vessels decreases [56]. Similarly, microscopic fragments of amnion may also be present in some rare cases of CM [57]. Presence of identiable fetal parts, signicant amounts of amnion or large numbers of nucleated RBCs essentially excludes the diagnosis of CM, although the possibility of a twin pregnancy with a CM and coexisting nonmolar conception must always be considered since such pregnancies require MShCG follow-up in the same manner as singleton CM. Finally, and most importantly, the diagnosis of molar disease requires the presence of abnormal trophoblast proliferation, recognized as more than two layers of trophoblast, often with nuclear pleomorphism, forming circumferential masses. Even in early CM, there are usually areas of denite circumferential trophoblast hyperplasia (Fig. 4), and sheets of pleomorphic extravillous trophoblast may also be present. These fragments do not indicate the presence of choriocarcinoma and appear to have little prognostic signicance other than aiding the diagnosis of CM [3]. Extravillous trophoblast invasion also appears disorganized in pregnancies affected by CM, many showing a prominent implantation site with absence of the normal endovascular trophoblast population but

First-trimester histopathological features

Villi of varying sizes are usually present but these are of variable size and often demonstrate a spectrum rather than two distinct populations. Similar to the ndings with CM in the rst trimester, villous hydrops may be patchy and mild with only occasional scattered cisterns. The larger, hydropic villi may demonstrate an abnormal, irregular dentate outline with cleft formation, and round or oval pseudoinclusions are often present in some villous cross-sections (Fig. 6). Amnion or fetal parts may be present and the chorionic villi usually demonstrate blood vessels often containing nucleated fetal red blood cells. In some cases, especially with advancing gestation, angiomatoid change may be seen focally, manifest as dilated thin-walled vessels within enlarged villi [56] (Fig. 7). Abnormal, nonpolar trophoblast hyperplasia is present, but this is almost always focal and less marked than in CM at this gestation, usually being multifocal rather than truly circumferential and sometimes demonstrating a lace-like or vacuolated pattern. The villous stroma will show a mixture of hydropic change and stromal brosis in different villi, which is a useful feature since CM rarely exhibit significant stromal brosis. The stroma of PM in well-preserved villi does not exhibit signicant stromal karyorrhectic debris (Fig. 8). The implantation site in cases of PM is usually unremarkable with endovascular trophoblast population identiable.

NONMOLAR CONCEPTIONS (HYDROPIC ABORTIONS [HA])

First-trimester histopathological features
Villi of varying sizes may be present but there are usually no villi with marked enlargement. The majority of villi usually show hydropic change but with no well-formed cistern formation, and some smaller villi may be brotic. Most villi are hypovascular, some may show empty, collapsed vessels, others appear avascular. Occasional trophoblastic pseudoinclusions may be identied, including single cell inclusions, and scattered villi may show an irregular outline, particularly in cases of fetal nonmolar aneuploidy. Some villi may show normal polar cytotro-

72

D. RAKHEJA

ET AL.

Figure 1. First trimester complete mole (CM), demonstrating characteristic budding appearance of the villi. Much of the trophoblast has been displaced from this villus. Figure 2. First trimester CM, demonstrating characteristic budding appearance of the villi and mild trophoblast hyperplasia. Figure 3. First trimester CM, demonstrating presence of collapsed villous vessels. In products of conception from CM obtained in the rst trimester, villous vessels are often identied. Figure 4. First trimester CM, demonstrating characteristic abnormal circumferential trophoblast hyperplasia. Figure 5. Implantation site fragments from a pregnancy affected by CM, at 8 wk gestation, demonstrating presence of orid interstitial extravillous trophoblast invasion but lack of normal endovascular trophoblast plugging, with resultant interstitial hemorrhage. Figure 6. First trimester partial mole (PM), demonstrating chorionic villi with an irregular dentate outline and in which trophoblastic pseudoinclusions are present.

Figure 7. First trimester PM, demonstrating chorionic villi with villous vessels exhibiting angiomatoid change. This appearance is more common with advancing gestational age. Figure 8. First trimester PM, demonstrating a well-preserved chorionic villus with a prominent villous vessel but no stromal karyorrhectic debris. Figure 9. First trimester nonmolar hydropic abortion, demonstrating chorionic villi with polar trophoblastic cell columns but no abnormal trophoblast hyperplasia. Figure 10. Products of conception from a rst trimester failed pregnancy demonstrating presence of folded gestation sac which should not be confused with cistern formation. Figure 11. First trimester chorionic villi from a PM, demonstrating positive nuclear immunostaining with p57KIP2, an imprinted gene product. Such cytotrophoblast and villus stromal staining is absent in cases of complete hydatidiform mole.

phoblastic cell column formation and occasionally small trophoblast buds are present, but there is no abnormal trophoblast hyperplasia (Fig. 9). The implantation site is

usually unremarkable, although in some cases may show features related to an underlying condition predisposing to the pregnancy loss.

CHIARI I MALFORMATION

AND

CYSTIC FIBROSIS

73

Table 1. Main diagnostic histopathological features and differences between complete mole, partial mole, and nonmolar hydropic abortion, in products of conception evacuated in the rst trimestera
CM Villous size Villous outline Villous stroma Villous hydrops Cistern formation Stromal karyorrhectic debris Villous vessels Nucleated red blood cells Trophoblast pseudoinclusions Trophoblast hyperplasia Extravillous trophoblast Implantation site Fetal parts/amnion Varied Budding Mucoid/myxoid Varied Varied Present Absent/collapsed Very rare Present Present Pleomorphic Floridd Absent PM Varied Irregular Fibrotic Varied Varied Absent Present Present Present Present
c

HA Uniform Smoothb Hydropic Present None Absent Absent/collapsed Present Rareb Polar Absent Normal Present

Present Normal Present

CM, complete mole; PM, partial mole; HA, nonmolar hydropic abortion. a The features are oversimplied in order to act as a guide only; please refer to the text for further discussion of individual features. b Cases of nonmolar fetal aneuploidy may demonstrate irregular, hydropic villi with pseudoinclusions, but no abnormal trophoblast hyperplasia is present. c Abnormal trophoblast hyperplasia may be focal and demonstrate a lace-like pattern around the villus. d Extravillous interstitial trophoblast invasion may appear orid but there is reduced normal endovascular trophoblast plugging and interstitial hemorrhage may be seen.

The main diagnostic histopathological features and differences between CM, PM, and HA in products of conception evacuated in the rst trimester are summarized in Table 1.

PRACTICAL ISSUES IN HISTOPATHOLOGICAL EVALUATION OF PRODUCTS OF CONCEPTION

Limited material
As with all aspects of histopathology, sampling adequacy is of signicant importance with regard to diagnostic accuracy. In many cases of molar pregnancy, there may have been partial spontaneous uterine evacuation prior to curettage, or the diagnosis was not suspected by the referring clinician, therefore only limited material may be submitted for examination. In most cases, all material should be processed and examined, and all villi evaluated. It must be recognized that in cases with only few villi present, it may be impossible on histopathological grounds alone to reach a rm diagnosis; management must be pragmatic in such cases, if denitive ancillary investigations are not available.

of a CM is now known to be incorrect when examining products of conception in the rst trimester [56], although this is a rare nding; it remains true that large numbers of nucleated red cells in villous vessels should make the diagnosis of CM questioned. In a previous study involving histological examination of 3180 molar conceptions, there were 60 cases (1.8%) with clear histological features of CM and a diploid chromosomal complement on ow cytometry, in which either fetal nucleated red cells or fragments of amnion were present in the sample [59]. In some of the cases, further examination revealed the presence of a CM with possible twin, but in the remaining cases, there was no evidence of a twin and nucleated fetal red blood cells were present within the vessels of the molar villi. Therefore, it should be remembered that the nding of apparent fetal tissue with molar villi or nucleated red cells in fetal vessels is not sufcient to classify the case as PM, since it may represent either a twin pregnancy with CM, or abortive fetal development in CM. Such rare diploid moles with amnion or nucleated red cells present have histological appearances and prognosis otherwise similar to those of classic early CM.

Fetal vessels and nucleated red blood cells

The dogma that presence of fetal nucleated red blood cells or fetal vessels within villi excludes the possibility

Trophoblast hyperplasia
Abnormal trophoblast hyperplasia is a requirement for the diagnosis of molar pregnancy. In most cases of CM,

74

D. RAKHEJA

ET AL.

the degree of hyperplasia is more marked compared to PM, although this in itself should not be used as a distinguishing criterion, especially since villous trophoblast may be dislodged during evacuation and observed as extravillous trophoblast sheets. The character of the proliferating trophoblast may vary from fairly monomorphic sheets of apparent cytotrophoblast to a delicate lace-like branching pattern of vacuolated trophoblast, particularly seen in some cases of PM. Furthermore, following surgical evacuation of retained products of conception (ERPC), particularly with PM, excessive trophoblast may be stripped from the villi resulting in only scanty villi remaining in which circumferential trophoblast hyperplasia can be identied. Similarly, although not diagnostic per se, the presence of large amounts of extravillous trophoblast fragments should alert the pathologist to the possible presence of molar change, particularly if marked trophoblast nuclear pleomorphism is present. A mimic of trophoblast hyperplasia may be seen where there is perivillous brin deposition present. This can result in separation of the syncytiotrophoblast from the villous core with secondary cytotrophoblast proliferation. For the purposes of diagnosis of molar disease, trophoblastic hyperplasia should not be based solely on villi in which perivillous brin deposition is present.

change. Therefore, in cases with sufcient material for assessment in which some features of possible PM are identied but where there is no nonpolar trophoblast hyperplasia, the presence of nonandrogenetic triploid, fetal aneuploidy should be considered.

Pseudopartial mole/stem villous hydrops

A relatively recently described condition, known as placental mesenchymal dysplasia (PMD), pseudopartial mole or stem villous hydrops, may be diagnostically confused with PM since there may be marked villous hydropic change in association with a raised MShCG during the pregnancy [60]. However, in cases of PMD, the fetus is usually normally formed and the pregnancy progresses to the third trimester. Histologically, there is marked stem villous hydrops, often with associated peripheral chorioangiomatoid change, but no trophoblastic hyperplasia is present [61]. The fetus and placenta demonstrate a diploid karyotype and, in most cases, it appears that the infant is phenotypically normal, although there is an association between PMD and Beckwith-Wiedemann syndrome. Suspicion of this condition should be raised in any case of apparent PM associated with a phenotypically normal fetus in the third trimester.

Villous hydrops
In the older literature, villous hydrops was the most striking feature of products of conception affected by molar disease. As previously stated, in rst trimester specimens, the degree of villous hydrops may be much less marked and patchy, with both CM and PM. Enlarged hydropic villi in molar pregnancies usually show an irregular outline and other features of GTD, whereas in cases of HA, the enlarged villi are usually more circular in cross section with hypocellular villous cores and trophoblast attenuation. Occasionally, fragments of gestation sac may become folded with the appearance of a central cell-free area, which may be misinterpreted as a villous cistern. Gestation sac demonstrates a denser collagenous stroma, and no trophoblast hyperplasia is usually present (Fig. 10).

Immunohistochemistry
The basis for the diagnosis of CM and PM remains detailed examination of tissue morphology on routine hematoxylin and eosin (H&E) stained sections. Additional special stains may be carried out to highlight certain aspects of the microanatomy, such as trichrome stains for collagen, but these have not been shown to provide useful additional diagnostic information. Immunohistochemical methods may be used to identify trophoblast and dene the hormonal prole of the tissue but, in contrast to the case with gestational trophoblastic tumors, in molar pregnancy this does not provide clues to the differential diagnoses [52]. Immunomarkers of cell proliferation, such as PCNA and Ki67, have also been examined in this context, and similarly have been reported to be of little practical use in differentiating between HA, CM, and PM [62,63]. Recently, the expression of cell cycle control proteins has been examined, and E2F-1 and cyclin E reported to be upregulated in molar tissue [64]. The most useful immunohistochemical marker, however, appears to be related to the fact that molar pregnancies are associated with imprinting abnormalities. p57KIP2 is a gene expressed predominantly from the maternal allele in most tissues; p57KIP2 expression in cytotrophoblast and villous mesenchyme is markedly reduced or absent in CM compared with strong expression in both PM and HA [41] and may, therefore, be useful in the rare cases in which the main differential diagnosis is between CM and PM, rather than PM and HA (Fig. 11). At present, immunohisto-

Trophoblast pseudoinclusions
Cases of both CM and PM usually demonstrate villous trophoblastic pseudoinclusions from an early gestation. In CM the inclusions are irregular in shape and scattered, whereas in PM, they are often more regular, round or oval in shape, and more widespread. The presence of villous pseudoinclusions should raise the possibility of molar disease but may indicate other pregnancy pathologies, particularly nonmolar fetal aneuploidy. Both gynegenetic triploidy and a range of fetal trisomies may be associated with the presence of villous pseudoinclusions, irregular and abnormally shaped villi, and mild hydropic

CHIARI I MALFORMATION

AND

CYSTIC FIBROSIS

75

chemical techniques remain potentially useful adjuncts to morphological diagnosis in molar disease.

17.

CONCLUSIONS
Changing clinical management of early pregnancy complications has resulted in the majority of complete and partial hydatidiform moles being evacuated in the rst or early second trimester, at which time the histopathological diagnostic criteria may be different from those classically described at later gestations. Recognition of these specic features allows a condent histopathological diagnosis to be made in most cases. The use of ancillary techniques, such as ow cytometry, immunohistochemistry, and molecular studies, will usually facilitate the diagnosis in cases where the morphological features are inconclusive.

18. 19.

20.

21.

22.

REFERENCES
1. Kim S. Epidemiology. In: Hancock BW, Newlands ES, Berkowitz RS, eds. Gestational Trophoblastic Disease. London: Chapman & Hall, 1997;2742. Fisher RA. Genetics. In: Hancock BW, Newlands ES, Berkowitz RS, eds. Gestational Trophoblastic Disease. London: Chapman & Hall, 1997;526. Paradinas FJ. Pathology. In: Hancock BW, Newlands ES, Berkowitz RS, eds. Gestational Trophoblastic Disease. London: Chapman & Hall, 1997;4376. Goldstein DP, Berkowitz RS. Current management of complete and partial molar pregnancy. J Reprod Med 1994;39:139146. Bagshawe KD, Dent J. Hydatidiform mole in England and Wales, 19731983. Lancet 1986;2:673677. Newlands ES. Presentation and management of persistent gestational trophoblastic disease and gestational trophoblastic tumors in the UK. In: Hancock BW, Newlands ES, Berkowitz RS, eds. Gestational Trophoblastic Disease. London: Chapman & Hall, 1997;143156. Seckl MJ, Fisher RA, Salerno G, et al. Choriocarcinoma and partial hydatidiform moles. Lancet 2000;356:3639. Vassilakos P, Kajii T. Hydatidiform mole: two entities. Lancet 1976;1:259. Vassilakos P, Riotton G, Kajii T. Hydatidiform mole: two entities. A morphological and cytogenetic study with some clinical considerations. Am J Obstet Gynecol 1977;127: 167170. Szulman AE, Surti U. The syndromes of hydatidiform mole. I. Cytogenetic and morphologic correlations. Am J Obstet Gynecol 1978;131:665671. Kajii T, Ohama K. Androgenetic origin of hydatidiform mole. Nature 1977;268:633634. Wake N, Takagi N, Sasaki M. Androgenesis as a cause of hydatidiform mole. J Natl Cancer Inst 1978;60:5157. Lawler SD, Pickthall VJ, Fisher RA, Povey S, Evans MW, Szulman AE. Genetic studies of complete and partial hydatidiform moles. Lancet 1979;2:580. Jacobs PA, Wilson CM, Sprenkle JA, Rosenshein NB, Migeon BR. Mechanism of origin of complete hydatidiform moles. Nature 1980;286:714716. Lawler SD, Povey S, Fisher RA, Pickthall VJ. Genetic studies on hydatidiform moles. II. The origin of complete moles. Ann Hum Genet 1982;46:209222. Fisher RA, Povey S, Jeffreys AJ, Martin CA, Patel I, Lawler SD. Frequency of heterozygous complete hydatidiform

23.

2.

24.

3.

25.

4.

26.

5. 6.

27.

7. 8. 9.

28.

29.

30.

10.

11. 12. 13.

31. 32.

33. 34.

14.

15.

35.

16.

moles, estimated by locus-specic minisatellite and Y chromosome-specic probes. Hum Genet 1989;8:259263. Kovaks BW, Shahbahrami B, Tast DE, Curtin JP. Molecular genetic analysis of complete hydatidiform moles. Cancer Genet Cytogenet 1991;54:143152. Ohama K, Kajii T, Okamoto E, et al. Dispermic origin of XY hydatidiform moles. Nature 1981;29:551552. Wallace DC, Surti U, Adams CW, Szulman AE. Complete moles have paternal chromosomes but maternal mitochondrial DNA. Hum Genet 1982;61:145147. Edwards YH, Jeremiah SJ, McMillan SL, Povey S, Fisher RA, Lawler SD. Complete hydatidiform moles combine maternal mitochondria with a paternal nuclear genome. Ann Hum Genet 1984;48:119127. Jacobs PA, Hunt PA, Matsuura JS, Wilson CC, Szulman AE. Complete and hydatidiform mole in Hawaii: cytogenetics, morphology and epidemiology. Br J Obstet Gynaecol 1982;89:258266. Helwani MN, Seoud M, Zahed L, Zaatari G, Khalil A, Slim R. A familial case of recurrent hydatidiform molar pregnancies with biparental genomic contribution. Hum Genet 1999;105:112115. Fisher RA, Khatoon R, Paradinas FJ, Roberts AP, Newlands ES. Repetitive complete hydatidiform mole can be biparental in origin and either male or female. Hum Reprod 2000;15:594598. Sensi A, Gualandi F, Pittalis MC, et al. Mole maker phenotype: possible narrowing of the candidate region. Eur J Hum Genet 2000;8:641644. Lawler SD, Fisher RA, Pickthall VJ, Povey S, Evans MW. Genetic studies on hydatidiform moles. I. The origin of partial moles. Cancer Genet Cytogenet 1982;5:309320. Jacobs PA, Szulman AE, Funkhouser J, Matsuura JS, Wilson CC. Human triploidy: relationship between parental origin of the additional haploid complement and development of partial hydatidiform mole. Ann Hum Genet 1982; 46:223231. Zaragoza MV, Surti U, Redline RW, Millie E, Chakravarti A, Hassold TJ. Parental origin and phenotype of triploidy in spontaneous abortions: predominance of diandry and association with the partial hydatidiform mole. Am J Hum Genet 2000;66:18071820. Vejerslev LO, Fisher RA, Surti U, Wake N. Hydatidiform mole: cytogenetically unusual cases and their implications for the present classication. Am J Obstet Gynecol 1987; 157:180184. Sheppard DM, Fisher RA, Lawler SD, Povey S. Tetraploid conceptus with three paternal contributions. Hum Genet 1982;62:371374. Surti U, Szulman AE, Wagner K, Leppert M, OBrien SJ. Tetraploid partial hydatidiform moles: two cases with a triple paternal contribution and a 92,XXXY karyotype. Hum Genet 1986;72:1521. Reik W, Walter J. Genomic imprinting: parental inuence on the genome. Nat Rev Genet 2001;2:2132. McFadden DE, Kwong LC, Yam IY, Langlois S. Parental origin of triploidy in human fetuses: evidence for genomic imprinting. Hum Genet 1993;92:465469. Paradinas FJ. The histological diagnosis of hydatidiform moles. Curr Diagn Pathol 1994;1:2431. Genest DR. Partial hydatidiform mole: clinicopathological features, differential diagnosis, ploidy and molecular studies, and gold standards for diagnosis. Int J Gynecol Pathol 2001;20:315322. Van de Kaa CA, Hanselaar AGJM, Hopman AHN, et al. DNA cytometric and interphase cytogenetic analyses of parafn-embedded hydatidiform moles and hydropic abortions. J Pathol 1993;170:229238.

76

D. RAKHEJA

ET AL.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

Fisher RA, Lawler SD, Ormerod MG, Imrie P, Povey S. Flow cytometry used to distinguish between complete and partial hydatidiform moles. Placenta 1987;8:249256. Hemming JD, Quirke P, Womack C, Wells M, Elston CW. Diagnosis of molar pregnancy and persistent trophoblastic disease by ow cytometry. J Clin Pathol 1987;40:615620. Lage JM, Driscoll SG, Yavner DL, Olivier AP, Mark SD, Weinberg DS. Hydatidiform moles; application of ow cytometry in diagnosis. Am Clin Pathol 1988;89:596600. Lee MH, Reynisdottir I, Massague J. Cloning of p57KIP2, a cyclin-dependent kinase inhibitor with unique domain structure and tissue distribution. Genes Dev 1995;9:639 649. Chilosi M, Piazzola E, Lestani M, et al. Differential expression of p57kip2, a maternally imprinted cdk inhibitor, in normal human placenta and gestational trophoblastic disease. Lab Invest 1998;78:269276. Castrillon DH, Sun D, Weremowicz S, Fisher RA, Crum CP, Genest DR. Discrimination of complete hydatidiform mole from its mimics by immunohistochemistry of the paternally imprinted gene product p57KIP2. Am J Surg Pathol 2001;25:12251230. Weber JL, May PE. Abundant class of human DNA polymorphisms which can be typed using the polymerase chain reaction. Am J Hum Genet 1989;44:388396. Reed PW, Davies JL, Copeman JB, et al. Chromosomespecic microsatellite sets for uorescence-based, semiautomated genome mapping. Nat Genet 1994;7:390395. Fisher RA, Newlands ES. Rapid diagnosis and classication of hydatidiform moles using the polymerase chain reaction. Am J Obstet Gynecol 1993;168:563569. Lane SA, Taylor GR, Ozols B, Quirke P. Diagnosis of complete molar pregnancy by microsatellites in archival material. J Clin Pathol 1993;46:346348. Bell KA, Van Deerlin V, Addya K, Clevenger CV, Van Deerlin PG, Leonard DG. Molecular genetic testing from parafn-embedded tissue distinguishes nonmolar hydropic abortion from hydatidiform mole. Mol Diagn 1999; 4:1119. Amr MF, Fisher RA, Foskett MA, Paradinas FJ. Triplet pregnancy with hydatidiform mole. Int J Gynecol Cancer 2000;10:7681. Weaver DT, Fisher RA, Newlands ES, Paradinas FJ. Amniotic tissue in complete moles can be androgenetic. J Pathol 2000;191:6770. Reubinoff BE, Lewin A, Verner M, Safran A, Schenker JG, Abeliovich D. Intracytoplasmic sperm injection combined with preimplantation genetic diagnosis for the prevention of recurrent gestational trophoblastic disease. Hum Reprod 1997;12:805808. Evans AC, Soper JT, Hammond CB. Clinical features of molar pregnancies and gestational trophoblastic tumors. In: Hancock BW, Newlands ES, Berkowitz RS, eds. Gestational Trophoblastic Disease. London: Chapman & Hall, 1997;109125.

51.

52. 53.

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

64.

Berkowitz RS, Goldstein DP. Presentation and management of molar pregnancy. In: Hancock BW, Newlands ES, Berkowitz RS, eds. Gestational Trophoblastic Disease. London: Chapman & Hall, 1997;127142. Paradinas FJ. The histological diagnosis of hydatidiform moles. Curr Diagn Pathol 1994;1:2431. Paradinas FJ, Browne P, Fisher RA, Foskett M, Bagshawe KD, Newlands E. A clinical, histopathological and ow cytometric study of 149 complete moles, 146 partial moles and 107 non-molar hydropic abortions. Histopathology 1996;28:101110. Berkowitz RS, Goldstein DP. The diagnosis of molar pregnancy by ultrasound: a continuing challenge. Ultrasound Obstet Gynecol 1997;9:45. Sebire NJ, Rees H, Paradinas FJ, Seckl MJ, Newlands ES. The diagnostic implications of routine ultrasound examination in histologically conrmed early molar pregnancies. Ultrasound Obstet Gynecol 2001;18:662665. Paradinas FJ. The diagnosis and prognosis of molar pregnancy: the experience of the National Referral Centre in London. Int J Gynaecol Obstet 1998;60:Suppl 1:S57S64. Fisher RA, Paradinas FJ, Soteriou BA, Foskett M, Newlands ES. Diploid hydatidiform moles with fetal red blood cells in molar villi. 2Genetics. J Pathol 1997;181:189195. Sebire NJ, Rees H, Paradinas FJ, et al. Extravillus endovascular implantation site trophoblast invasion is abnormal in complete versus partial molar pregnancies. Placenta 2001;22:725728. Paradinas FJ, Fisher RA, Browne P, Newlands ES. Diploid hydatidiform moles with fetal red blood cells in molar villi. 1Pathology, incidence, and prognosis. J Pathol 1997; 181:183188. Jauniaux E, Nicolaides KH, Hustin J. Perinatal features associated with placental mesenchymal dysplasia. Placenta 1997;18:701706. 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. 15 cases and a review of the literature. Histopathology 2001;39:447454. Cheung ANY, Ngan HYS, Chen WZ, et al. The signicance of the proliferating cell nuclear antigen in human trophoblastic disease: an immunohistochemical study. Histopathology 1993;22:565568. Suresh UR, Hale RJ, Fox H, Buckley CH. Use of proliferation cell nuclear antigen immunoreactivity for distinguishing hydropic abortions from partial hydatidiform moles. J Clin Pathol 1993;46:4850. Olvera M, Harris S, Amezcua CA, et al. Immunohistochemical expression of cell cycle proteins E2F-1, Cdk-2, Cyclin E, p27 (kip1) and Ki-67 in normal placenta and gestational trophoblastic disease. Mod Pathol 2001;14:1036 1042.

CHIARI I MALFORMATION

AND

CYSTIC FIBROSIS

77