OUP user menu

Comparison of Fluorescence In Situ Hybridization, p57 Immunostaining, Flow Cytometry, and Digital Image Analysis for Diagnosing Molar and Nonmolar Products of Conception

Benjamin R. Kipp PhD, Rhett P. Ketterling MD, Trynda N. Oberg MS, CT(ASCP), Margot A. Cousin, Amy M. Plagge SCT(ASCP), CT(IAC), Anne E. Wiktor, Johnita M. Ihrke CT(ASCP), Cecelia H. Meyers, William G. Morice MD, PhD, Kevin C. Halling MD, PhD, Amy C. Clayton MD
DOI: http://dx.doi.org/10.1309/AJCPV7BRDUCX0WAQ 196-204 First published online: 1 February 2010


Pathologic examination of products of conception (POC) is used to differentiate hydropic abortus (HA), partial hydatidiform mole (PM), and complete hydatidiform mole (CM). Histologic classification of POC specimens can be difficult, and ancillary testing is often required for a definitive diagnosis. This study evaluated 66 POC specimens by flow cytometry, digital image analysis, p57 immunohistochemical analysis, and fluorescence in situ hybridization (FISH). The final diagnosis, based on the combined analysis of all test results, included 33 HAs, 24 PMs, and 9 CMs. The p57 immunostain identified 9 CMs that were evaluated as nontriploid by all other techniques. FISH seems to have the best accuracy (100%) for determining whether a specimen contains a triploid chromosome complement. These data suggest that the combination of p57 and FISH seems to be the best ancillary testing strategy to aid pathologists in the appropriate identification of CM, PM, and HA in POC specimens.

Key Words:
  • Gestational trophoblastic disease
  • Hydropic abortus
  • Products of conception
  • Fluorescence in situ hybridization
  • FISH
  • DNA ploidy

Pathologic examination of products of conception (POC) is routinely used to differentiate hydropic abortus (HA), partial hydatidiform mole (PM), and complete hydatidiform mole (CM). The correct classification of POC specimens is clinically important because each of these entities has a different potential for clinical persistence and malignant transformation. Approximately 0.2% to 5% of patients with a PM and 15% to 25% of patients with a CM will develop persistent gestational trophoblastic disease.13 Malignant transformation from persistent gestational trophoblastic disease to choriocarcinoma has been observed in 3% to 5% of patients with a CM.3 Choriocarcinoma has also been observed in patients following a PM, but this is extremely rare.4 As a result, women with a PM or CM must undergo serial monitoring of β-human chorionic gonadotropin levels and are instructed to abstain from pregnancy for an extended period, whereas patients with an HA diagnosis require no hormone surveillance and do not have to refrain from pregnancy for the period that is typically recommended for patients diagnosed with a CM or PM.57

Interpretation of POC specimens requires that pathologists have a good understanding of the morphologic features and genetics of molar pregnancy Table 1. Morphologically, HA specimens have a wide spectrum of different-sized villi, including small sclerotic, normal-sized, and large edematous villi.6 HAs are often associated with karyotypic abnormalities but generally contain a diploid or near-diploid chromosomal composition by ploidy analysis. In contrast, PMs generally have 2 populations of villi (1 with enlarged hydropic villi and 1 of smaller villi), focal trophoblastic hyperplasia, marked scalloping of the villi with trophoblastic inclusions, central cisterns, and, on occasion, identifiable embryonic or fetal tissue. PMs are mostly triploid (69 chromosomes; androgenic triploidy) arising from dispermic fertilization of a haploid ovum (~90%) or fertilization of a haploid ovum with an unreduced diploid sperm (~10%; diandric triploidy). CMs are composed of large villous hydrops, excessive and circumferential trophoblastic proliferation (trophoblast hyperplasia), trophoblastic atypia at the molar implantation site, and villous cavitation. The DNA content of CMs is generally diploid, representing 2 sets of parental chromosomes (diandric diploidy) from dispermic fertilization of an empty egg (~10%; heterozygous) or fertilization of an empty egg by a single sperm that then reduplicates to give a homozygous diploid genome (~90%).2,3,69

View this table:
Table 1

Classifying POC specimens based solely on histologic appearance can be extremely difficult, even for an experienced pathologist.3,10,11 Ancillary testing (ie, immunohistochemical and/or ploidy analysis), in addition to morphologic evaluation, is selectively required for the definitive diagnosis of POC specimens. A variety of ancillary tests have been used to aid pathologists in the interpretation of POC specimens, including flow cytometry (FC),1215 digital image analysis (DIA),1,12,1416 fluorescence in situ hybridization (FISH),1,13 and p57 immunostaining (p57).1720 Although numerous studies have shown that each of these methods is individually useful for POC interpretation, there is no published study that compares all 4 methods on POC tissue from the same patient cohort. The goal of this study was to critically evaluate FISH, p57, FC, and DIA as ancillary tests for the detection of PM, CM, and HA in a series of POC specimens.

Materials and Methods

Patient Population and Specimens

This study included 66 consecutive paraffin-embedded POC specimens that were sent to the Mayo Clinic, Rochester, MN, between November 2008 and January 2009 for evaluation of possible molar pregnancy. Each specimen demonstrated morphologic abnormalities that prompted the pathologist to request ancillary studies or consultation. The age of patients ranged from 18 to 48 years, with mean and median ages of 29 and 28 years, respectively. At the time of this study, p57 immunostaining and FC were orderable tests for the clinical interpretation of POC specimens. Therefore, when available, p57-stained slides and/or FC results were obtained from the clinical files. When p57 and FC testing had not been ordered clinically, the tests were performed on residual tissue. Residual tissue blocks were also used for DIA and FISH testing. A final diagnosis (“gold standard”) was interpreted for all patients at the end of the study using the combined morphologic, DIA, FC, FISH, and p57 results.3 This study was approved by the Mayo Clinic Institutional Review Board.

Pathologist Morphologic Interpretation

Morphologic assessment of H&E-stained tissue slides was done by a board-certified pathologist (A.C.C.) and categorized as HA, PM, or CM based on the aforementioned morphologic criteria for the diagnosis of POC specimens.2,6 All specimens were evaluated without knowledge of ancillary test results or previous clinicopathologic findings. The pathologist also marked areas on the H&E-stained slide that represented areas of villous tissue that were to be evaluated by DIA and FISH.

Digital Image Analysis

DNA content was evaluated using a Nikon COOLSCOPE digital microscope (Nikon Instruments, Melville, NY) and VS Pathology Software Suite program (Bacus Laboratories, Lombard, IL) on 6-μm-thick serial sections cut from the POC paraffin-embedded tissue blocks. Feulgen-stained nuclear images from approximately 200 cells were collected from areas of interest as identified by a pathologist on the corresponding H&E-stained slide. Measurements were obtained primarily from trophoblast nuclei lining the villi while trying to avoid trophoblast nuclei from hyperplastic areas.21 The COOLSCOPE microscope produced a histogram of the collected nuclei based on DNA index. Histograms were interpreted as diploid or triploid Figure 1 by a consensus review from 2 pathologists (K.C.H. and A.C.C.) and 1 DIA cytotechnologist (J.M.I.). Diploid tumors had peak DNA indices (PDI) between 0.90 and 1.10 with a relatively low percentage of nuclei in the S-phase and tetraploid areas. A triploid diagnosis was given to specimens with peak indices between diploid and tetraploid areas (PDI ~1.5) with a relatively low percentage of cells in the diploid or tetraploid areas.

Flow Cytometry

DNA analysis was performed using a FACSCalibur system (BD Biosciences, Franklin Lakes, NJ) as previously described.22 Briefly, nuclei from two 50-μm paraffin-embedded sections were extracted by dewaxing with Histo-Clear (National Diagnostics, Atlanta, GA) followed by rehydration in a series of 100%, 95%, and 70% ethanol. The specimen was then placed in a trypsin solution (0.0113 g of trypsin in 375 mL of Stock DNA buffer) for 10 minutes, ribonuclease A (Worthington Biochemical, Lakewood, NJ) for 30 minutes, and propidium iodide stain (Sigma-Aldrich, St Louis, MO) for a minimum of 30 minutes. The intensity of DNA-bound dye was measured, and a DNA histogram was generated. From this histogram, the presence of a triploid peak (PDI ~1.5) was interpreted as “triploid population detected,” whereas all other specimens were interpreted as “no triploid population detected” Figure 2.

p57 Immunohistochemical Staining

Formalin-fixed, paraffin-embedded POC tissue sections (4 μm) were stained with the DAKO EnVision+ System (DAKO, Carpinteria, CA) using purified mouse monoclonal antibodies (clone 57P06 at 1:1,000 dilution; NeoMarkers, Fremont, CA) against the p57 protein. Interpretation of p57 staining was performed by a pathologist (A.C.C.) based on the presence or absence of stain in villous stromal cells, cytotro-phoblasts, intermediate trophoblasts, and maternal decidua. Specimens were interpreted as “positive” for p57 staining when there was distinct nuclear staining of villous stromal cells and cytotrophoblasts. The p57 stain was interpreted as “negative” when there was no distinct staining or limited nuclear staining (<10%) of villous stromal cells and cytotrophoblasts and staining of intermediate trophoblasts and/or maternal decidua, which served as the positive internal control for these specimens Image 1.3,19

Figure 1

Representative examples of diploid (A; observed in hydropic abortuses and complete moles) and triploid (B; observed in partial moles) histograms produced by digital image analysis.

Figure 2

Representative examples of diploid (A; observed in hydropic abortuses and complete moles) and triploid (B; observed in partial moles) histograms produced by flow cytometry.

Fluorescence In Situ Hybridization

Processing of POC specimens for FISH analysis began by placing 5-μm paraffin tissue sections in a 90°C oven for 15 minutes. The tissue sections were then deparaffinized in xylene (twice for 15 minutes each time) and dehydrated in 100% ethanol (twice for 5 minutes each time). Once air dried, the slides were immersed in 10 mmol/L citric acid (pH 6.0) and run on high power in a humidified microwave for 10 minutes. The slides were immediately transferred into 2× saline sodium citrate (SSC) at 37°C for 5 minutes. The samples were then digested with pepsin using Digest-All 3 (Invitrogen, Carlsbad, CA) at 37°C for 20 minutes, placed in room temperature phosphate-buffered saline for 2 minutes, and dehydrated in 75%, 85%, and 100% ethanol solutions for 2 minutes each.

Three separate FISH probe sets were used in this study. The FISH probe sets targeted chromosomes X, Y, and 18 (probe set 1), chromosomes 13 and 21 (probe set 2), and chromosomes 15, 16, and 22 (probe set 3) Image 2. Ten microliters of each of the 3 FISH probe sets (Abbott Molecular, Des Plaines, IL) was applied to separate slides with the tissue area of interest as identified by a pathologist on the corresponding H&E-stained slide. Probe and target DNA were codenatured and hybridized on a ThermoBrite Denaturation/Hybridization System (Abbott Molecular) set at a denaturation temperature of 80°C (5 minutes) and a hybridization temperature of 37°C (minimum 8 hours). After hybridization, unbound probe was removed by washing in 2× SSC/0.1% NP-40 at 72°C for 2 minutes. Ten microliters of 4′-6-diamidino-2-phenylindole (DAPI) I counterstain (Abbott Molecular) was applied, and the slides were coverslipped. FISH signals were enumerated using a Leica fluorescent microscope (Leica Microsystems, Wetzlar, Germany) equipped with SpectrumOrange, SpectrumGreen and SpectrumAqua filters (Abbott Molecular).

The FISH signal patterns for each FISH probe set were independently enumerated in 100 total nuclei by 2 technologists on each of the 3 slides in the areas of interest as determined by the pathologist (A.C.C.) on the H&E-stained slide. The normal cutoff for false-positive signal patterns for each probe set was calculated from the results of the normal diploid samples using the β-inverse function.23 The normal cutoff for individual trisomies, individual monosomies, and triploidy (3 sex chromosome signals and 3 signals for each autosome) was calculated to be 25% or less. A normal cutoff of 25% or less should eliminate truncation artifacts caused by sectioning of paraffin tissue. In addition, samples with fewer than 50% of the cells exhibiting an abnormal signal pattern consistent with trisomy or monosomy required correlation of the FISH results with a review of the H&E-stained slide by a pathologist (R.P.K.) for evaluation of a mixture of maternal and fetal cells. While POC may contain degenerated tissues with associated coagulative necrosis, we did not encounter any cases in this study for which FISH did not yield a viable result for the probe sets tested.

Image 1

Representative examples of products of conception specimens demonstrating diffuse p57 staining (A, ×100; observed in hydropic abortuses and partial moles) and loss of p57 staining in villous tissue with staining of intermediate trophoblasts (B, ×100; observed in complete moles).


The final diagnosis for the 66 POC cases, based on the pathologist’s morphologic review and the combined analysis of all test results, included 33 HAs (50%), 24 PMs (36%), and 9 CMs (14%) Table 2. Of the 33 samples identified as HA, 32 (97%) had a combined result of positive p57 staining, diploid DIA interpretation, nontriploid FC interpretation, and FISH results that were disomic or neardisomic (all consistent with an HA diagnosis). The FISH results from these 32 HA specimens included disomy (n = 21 [66%]), trisomy 16 (n = 4 [13%]), trisomy 13 (n = 2 [6%]), monosomy X (n = 2 [6%]), trisomy 21 (n = 1 [3%]), monosomy X with trisomy 21 (n = 1 [3%]), and mosaic X/XXX (n = 1 [3%]; Image 2). One HA specimen had conflicting morphologic and ancillary test results, including a “favor HA” morphologic interpretation, positive p57 staining, triploid DIA result, nontriploid FC result, and tetraploid FISH result. This patient was categorized as having an HA because p57 was positive (reducing the probability of CM), 2 of the 3 tests assessing for ploidy had nontriploid results (reducing the probability of PM), and the pathologist’s morphologic impression was HA.

For 24 specimens, the final diagnosis was PM. In 21 of these samples, the expected ancillary test results of diffuse p57 staining, triploid DIA, triploid FC, and triploid FISH interpretations were obtained (Table 2). One specimen within this group was interpreted by FISH to have 3 copies of all autosomal chromosomes and an additional sex chromosome with a pattern of XXYY (Image 2). There were 3 PM specimens with nonconcordant test results. In these 3 specimens, FC was interpreted as nontriploid, which was inconsistent with the DIA and FISH results (which were interpreted as triploid; Table 2). One specimen within this nonconcordant group had a triploid chromosomal complement by FISH, with the exception that there were 4 copies of chromosome 21 (Image 2).

For 9 specimens, the final diagnosis was CM. In 4 of the 9 specimens, the expected ancillary test results of p57 staining loss, diploid DIA, nontriploid FC, and diploid FISH diagnosis were observed. In the remaining 5 CM specimens (56%) with nonconcordant ancillary test results, DIA incorrectly identified specimens as triploid (which is most consistent with a PM), whereas the FC, FISH, and p57 results were all consistent with a diagnosis of CM (Table 2).

Image 2

Representative images of fluorescence in situ hybridization results using the 3 different probe sets in the specimens for identification of hydropic abortuses, complete moles, and partial moles.

Pathologist accuracy for assessing individual specimens morphologically, without knowledge of the ancillary test results, is summarized in Table 3. The pathologist diagnosis was considered “correct” when the final diagnosis was favored by the pathologist or when the final diagnosis was represented in the pathologist’s differential diagnoses (eg, pathologist interpretation was “favor complete mole” or “partial mole vs complete mole” and the final diagnosis was “complete mole”). Of the 66 pathologist’s interpretations, 47 (71%) consisted of a “favored” (eg, favor CM) diagnosis, whereas 19 specimens (29%) resulted in a differential diagnosis (eg, PM vs CM). As Table 3 shows, the pathologist correctly classified the POC final result in 94% of HA specimens (31/33), 75% of PMs (18/24), and 89% of CMs (8/9). The overall accuracy of all ancillary tests is summarized in Table 4.

View this table:
Table 2
View this table:
Table 3


The findings from this study confirm that the histologic evaluation of POC specimens remains challenging for pathologists and that ancillary tests are often required for accurate interpretation of these specimens. Numerous publications have described the poor intraobserver and interobserver agreement among pathologists for diagnosing POC specimens based solely on histologic criteria.10,2426 With the advent of more sensitive β-human chorionic gonadotropin tests and the use of early ultrasonographic examination, moles that were generally diagnosed in the second trimester of pregnancy are now being identified in the first trimester. The earlier clinical detection of these specimens makes pathologic interpretation even more challenging because many of the histologic characteristics that separate molar and nonmolar diagnoses are more difficult to discern in first-trimester specimens.9

The pathologists in our study correctly classified 94% of HA specimens, 75% of PMs, and 89% of CMs when the final diagnosis was favored by the pathologist or when the final diagnosis was represented in the differential diagnoses. However, the lack of a definitive diagnosis by the pathologist is of limited or no value to a clinician who must select the optimal clinical follow-up for a patient based on the pathologic diagnosis. Table 3 shows that the pathologist was only able to definitively establish the correct diagnosis in 82% of HA specimens (27/33), 33% of PMs (8/24), and 67% of CMs (6/9). These results, along with data from other studies,10,2426 solidify the importance of incorporating ancillary testing algorithms for evaluating histologically abnormal POC specimens.

FC and DIA are commonly used ancillary tests to aid pathologists in the classification of POC specimens, specifically PMs.1216,21,27 Because HA and CM specimens characteristically consist of diploid or near-diploid chromosomal complements, FC and DIA cannot reliably distinguish HAs from CMs. However, a triploid interpretation by FC or DIA is considered diagnostic of a PM because most PMs consist of an androgenic triploid chromosomal complement. FC is a well-accepted ancillary technique because it can analyze the ploidy status of a large number of cells quickly. One limitation of FC is that it analyzes all cells within a tissue section, including maternal cells, if present. A PM specimen with a large proportion of maternal cells will reduce the percentage of triploid cells present on a histogram, making it difficult to definitively classify the specimen as triploid. Actively dividing cells from a CM or HA that are going through the DNA synthesis phase (S phase) of the cell cycle can also have increased DNA content and can resemble the DNA ploidy histogram pattern of a specimen with a small fraction of triploid cells with maternal contamination. On occasion, it can be difficult to distinguish a true triploid cell population from an actively dividing diploid cell population by FC, and this can lead to a false-negative diploid result in a patient with a PM.12

In our study, there were 3 PM specimens that were interpreted as triploid by DIA and FISH and diploid by FC. These false-negative results could be due to extensive maternal contamination or flow histogram misinterpretation (as detailed above). Rereview of the H&E-stained slides and histograms for these 3 specimens demonstrated that in 1 case, more than 50% of the tissue analyzed by FC was maternal tissue, which appears to have diluted the fetal villous cells to a degree that the histogram appeared diploid. In the remaining 2 cases, more than 75% of the analyzed tissue was fetal tissue, so maternal contamination was likely not the cause of the diploid result. A review of the FC histograms for these 2 cases demonstrated that a considerable proportion of the cells were in the triploid range, suggesting that these specimens, albeit subjectively because there was not a clearly definable triploid peak, could have been interpreted as triploid.

One benefit of performing DIA instead of FC is that it allows for visualization of the cells that are being analyzed, allowing the user to select the nonmaternal cells (ie, trophoblasts) for ploidy interpretation. This reduces the chance of a false diploid result in patients with a PM. However, DIA has limitations as well. There were 6 DIA results in our study that did not correspond to the final diagnosis. More specifically, 5 CM and 1 HA specimen were interpreted as triploid by DIA, whereas FC and FISH correctly interpreted them as nontriploid. It is interesting that each of these 6 specimens was noted as being difficult to classify at the time of the original DIA histogram interpretation. On rereview of the DIA histograms for these 6 cases, it was evident that there was a considerable percentage of cells in the triploid area without a clearly defined triploid peak. The cells in the nondiploid area most likely represented active trophoblastic cells that were going through the S phase of the cell cycle, which as stated before, is difficult to discern from true triploid cells on a histogram.

FISH, a more recently described technique for assessing POC specimens, was also evaluated in this study. Similar to previously published findings,25,28 FISH accurately (100%) determined whether specimens were triploid (representing a PM) or nontriploid (representing an HA or a CM). FISH has advantages over FC and DIA in that it allows for visualization of chromosomal abnormalities in individual villous cells. In addition, FISH has the ability to identify specific chromosomal anomalies, such as trisomy 13, trisomy 16, trisomy 21, and mosaicism, and is able to determine the sex chromosome complement (eg, XX, XY, XXYY; Table 2). The ability to detect specific chromosomal abnormalities is dependent on the number of different FISH probes used. However, the cost of the FISH assay is also dependent on the number of FISH probes used.

View this table:
Table 4

Our study evaluated 3 different FISH probe sets comprising 8 different chromosomal targets (chromosomes 13, 15, 16, 18, 21, 22, X, and Y) to detect common cytogenetic anomalies associated with pregnancy loss. For samples with a triploid result, all probe sets demonstrated trisomies of the autosomes by FISH, corroborating the PM result. In addition, 11 of the 33 HA specimens were abnormal by FISH, demonstrating isolated trisomy 13, 16, or 21 (7 total cases); monosomy X (3 cases); and tetraploidy in all cells (1 case; Table 2). The identification of these chromosomal abnormalities adds to the usefulness of a FISH approach for evaluation of HA specimens as these abnormal results yield the chromosomal cause for the miscarriage. However, while the use of 3 FISH probe sets can result in additional information for the miscarriage, a determination as to whether the sample is triploid or non-triploid can be obtained using only 1 FISH probe set, and this would reduce the cost of the assay for the patient and be less tedious to perform in the clinical laboratory.

Before this study, p57Kip (p57) immunohistochemical staining was shown to be a reliable assay for classifying CMs.18,19,28 The p57 gene, a tumor suppressor gene located on chromosome 11p15.5, is paternally imprinted and expressed only from the maternal allele. As a result, CMs demonstrate complete loss of p57 immunostaining because they lack a maternal genome. HA and PM, on the other hand, demonstrate diffuse p57 staining because they contain maternal and paternal genomes.19 We identified 9 CM specimens in this study by demonstrating loss of the p57 stain; all HA and PM specimens demonstrated diffuse p57 staining. While FISH and FC identified each CM as diploid, DIA incorrectly classified 5 of the 9 CMs as triploid (Table 2). Because CMs cannot typically be conclusively identified by morphologic examination alone (Table 3), these results indicate the DIA results should be interpreted with caution when suspecting a CM. In addition, rare tetraploid or mosaic tetraploid/diploid CMs can occur. Depending on the individual laboratory, the results by FC, DIA, or FISH may be classified as nontriploid, tetraploid, or diploid for these specimens, all suggesting a diagnosis of HA or CM. However, morphologic suspicions of a CM with loss of p57 staining will allow one to confidently make the diagnosis of a CM in these specimens.

The combined findings of our study strongly suggest that the combination of FISH and p57 immunostaining results provide the most accurate identification of CM, PM, and HA specimens. The correct algorithm that each laboratory should use to perform POC testing is not a “one size fits all” and is dependent on numerous factors including the following: (1) availability of testing platforms (ie, some clinical laboratories may already have a flow cytometer but not a fluorescence microscope); (2) consultant expertise with specific testing platforms; (3) availability of technical personnel to perform specific testing methods; (4) overall cost of performing each of these assays; and (5) pathologist and clinician judgment as to the best algorithm for their practice. The goal of this study was not to persuade clinical laboratories to adopt a certain testing algorithm, but was to scientifically compare the different methods so that individual laboratories could use these data to help define the optimal testing algorithm for their institutions. These data may also be beneficial for smaller clinical practices that are trying to determine the best laboratories to send their POC specimens to because most of the reference laboratories that provide POC testing use one or more of the methods described in this study.

One potential limitation of this study is that it did not evaluate a more recently described technique for POC interpretation, molecular genotyping. Recent data suggest that molecular genotyping may have advantages over the methods evaluated in this study because genotyping can distinguish maternal from paternal alleles, which can then be used to discern androgenic diploidy, androgenic triploidy, and biparental diploidy, which are characteristic of CMs, PMs, and HAs, respectively.3,29 One limitation of genotyping is that maternal decidual tissue that is free of fetal tissue (or a separate sample of maternal-only tissue) must be present for comparison of villous and paternal alleles. Further studies are needed comparing molecular genotyping with p57, FC, DIA, and/or FISH to better understand the potential clinical role of molecular genotyping for POC analysis.

Classification of abnormal POC specimens based solely on histologic evaluation is often difficult, requiring one or more ancillary tests to definitively distinguish PM, CM, and HA specimens. FISH seems to have the best accuracy for determining whether a specimen contains a triploid chromosome complement consistent with a PM and also provides valuable information regarding the cause of pregnancy loss for a subset of HA specimens. FC and DIA demonstrate limitations in their accuracy for identifying PM specimens but could be considered adequate substitutes for FISH for diagnosing PMs in laboratories that do not perform FISH testing. p57 is a robust and well-accepted immunostain that can reliably diagnose CMs. Therefore, the combination of p57 staining along with FISH seems to be the best algorithm to reliably distinguish HAs, PMs, and CMs. Further evaluation of molecular genotyping is needed to determine if this new method offers additional advantages over p57, DIA, FC, and FISH testing for the interpretation of POC specimens.


Upon completion of this activity you will be able to:

  • describe the morphologic and genetic characteristics of hydropic abortus, partial mole, and complete mole.

  • describe expected ancillary test results (ie, ploidy analysis by digital image analysis, fluorescence in situ hybridization, or flow cytometry; and p57 immunohistochemistry results) for hydropic abortus, partial mole, and complete mole specimens.

  • outline the appropriate strategy for additional testing, and determine which cases should be further investigated when working up placental tissue with atypical histologic features.

The ASCP is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The ASCP designates this educational activity for a maximum of 1 AMA PRA Category 1 Credit™ per article. This activity qualifies as an American Board of Pathology Maintenance of Certification Part II Self-Assessment Module.

The authors of this article and the planning committee members and staff have no relevant financial relationships with commercial interests to disclose.

Questions appear on p 336. Exam is located at www.ascp.org/ajcpcme.


  • * Drs Ketterling and Clayton contributed equally as senior authors.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
View Abstract