Utilization of Chromogenic In Situ Hybridization to ...

AJCP / ORIGINAL ARTICLE

Utilization of Chromogenic In Situ Hybridization to Assess Ploidy in the Diagnosis of Hydatidiform Mole John P. Kunesh, MD, Jacqueline G. Kunesh, Rachel Jade Jorgensen, Catrina D. Corral, and John D. Blakey, MD From the Department of Pathology, Torrance Memorial Medical Center, Torrance, CA.

Am J Clin Pathol July 2016;146:125-131 DOI: 10.1093/AJCP/AQW095

ABSTRACT Objectives: Ploidy assessment is often required for the diagnosis of partial molar pregnancy. While fluorescence in situ hybridization has been shown to be effective, it is not available in many laboratories. We validated chromogenic in situ hybridization (CISH) for this purpose. Methods: CISH using probes to chromosomes 17 and 10 was performed on 20 POC cases with known cytogenetics to establish a reference percentage. This was then used to classify a randomized set of abnormal and normal cases. Results: An abnormal CISH cutoff of greater than 7% was established. All abnormal cases (six triploid and three tetraploid), 11 “normal” (46, XX or XY or undetectable abnormalities), and one trisomy 10 were all correctly classified by the assay. Conclusions: CISH is a useful ancillary technique for the diagnosis of molar pregnancy. Its greater accessibility and ability to score even rare placental tissue in a background of maternal tissue offer advantages over other methods.

© American Society for Clinical Pathology, 2016. All rights reserved. Downloaded For from https://academic.oup.com/ajcp/article-abstract/146/1/125/2836654 permissions, please e-mail: [email protected] by guest 125 on 17 May 2018

Upon completion of this activity you will be able to: • discuss the genetics of molar pregnancy with emphasis on the relationship to diagnosis. • describe how ancillary testing may be used in the diagnosis of hydatidiform mole. • compare the advantages of using chromogenic in situ hybridization in diagnosing molar pregnancy to that of other methods. The ASCP is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The ASCP designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 CreditTM per article. Physicians should claim only the credit commensurate with the extent of their participation in the activity. 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. Exam is located at www.ascp.org/ajcpcme.

Histologic review of the products of conception specimen is a frequent requirement of most surgical pathology practices. This review requires not only the confirmation of the presence of chorionic villi or fetal parts to exclude ectopic pregnancy but also the exclusion of hydatidiform mole (both partial and complete). Accurate diagnosis of hydatidiform mole is essential because of its serious health implications, particularly the risks of gestational trophoblastic neoplasia with recurrent disease and potential for metastases. Persistence or transformation into malignant disease requiring chemotherapy has been estimated to occur in 0.5% to 4% of partial moles and 15% to 25% of complete moles.1,2 Although the risk of metastatic disease following partial mole is thought to be very low, genetic analysis on metastatic trophoblastic disease after molar pregnancy has

Am J Clin Pathol 2016;146:125-131 125 DOI: 10.1093/ajcp/aqw095

CME/SAM

Key Words: Obstetric and perinatal; Molecular diagnostics; Gynecological; Partial mole; CISH; Hydatidiform mole; Molar pregnancy; In situ hybridization

Kunesh et al / USE

OF

CISH

TO

ASSESS PLOIDY

IN

DIAGNOSIS

OF

HYDATIDIFORM MOLE

confirmed that partial moles have the capacity to metastasize.3 The recommended follow-up after the diagnosis of a molar pregnancy requires serial monitoring of serum human chorionic gonadotropin, as well as abstention from pregnancy during the monitoring period.1,2,4 Because of this, an equivocal or uncertain diagnosis is often unacceptable to the patient and treating physician. Accurate and definitive diagnosis is necessary to provide appropriate patient care. Although some cases are histologically straightforward, histologic overlap between hydropic changes in nonmolar pregnancies and hydatidiform mole is a known problem, particularly in early moles.3,5,6 Diagnostic accuracy of partial moles based on morphology alone is particularly problematic.5,7-9 This has led to the necessity of ancillary testing for diagnosis. Diagnosis of complete moles had been greatly facilitated by immunohistochemical staining for p57kip2 (p57), the product of a paternally imprinted, maternally expressed gene lacking expression in the paternal genomic component.5,6,10-13 Because the genome of the complete mole is strictly paternal, this stain lacks expression in complete moles. It is, however, of no use in distinguishing partial moles from nonmolar pregnancies. The mainstay of partial mole diagnosis lies in ploidy assessment. Partial moles are predominately triploid (although tetraploid partial moles and complete moles are also known occurrences).14 Because specimens for conventional cytogenetics are not always collected, and even if collected do not always yield results, ancillary tests able to be performed on formalin-fixed, paraffin-embedded tissue are necessary. Published effective tests on paraffin-embedded tissue for this purpose include DNA ploidy analysis by flow cytometry,15-18 digital image analysis,15,17-19 and fluorescence in situ hybridization (FISH).5,16,20 FISH, in particular, has been shown to be a very effective technique in comparison to the others.5 Single-nucleotide polymorphism microarray technology is also possible on paraffin-embedded material; however, the technique is expensive and not effective in identifying tetraploid chromosome complements. While FISH is very effective, this technology is not available in many surgical pathology practices. Exclusion of molar pregnancies (particularly partial mole) thus has become an outsourced “send-out” to reference laboratories. Depending on the setting, these may be a frequent specimen, often coming from the emergency department, with a variety of insurance payers and, sometimes, unknown primary care physicians. Thus, there is a need to develop a relatively cost-effective way to assess ploidy in these specimens that can be performed in most surgical pathology laboratories. Chromogenic in situ hybridization (CISH) is quickly becoming available in most surgical pathology laboratories. Its ability to be performed on automated platforms and read

126 Am J Clin Pathol 2016;146:125-131 Downloaded 126 from https://academic.oup.com/ajcp/article-abstract/146/1/125/2836654 DOI: 10.1093/ajcp/aqw095 by guest on 17 May 2018

on bright-field microscopes has opened this technology to many surgical pathology settings. Evaluation of human epidermal growth factor receptor 2 (Her2) gene amplification status in breast carcinoma has familiarized many practices to this technology, and it has been validated for other genes typically assessed by FISH, such as epidermal growth factor receptor (EGFR).21 Using this as a guideline, we have validated chromosome 17 and chromosome 10 CISH using commercially available probes on an automated staining platform to assess ploidy in POC specimens.

Materials and Methods Patient Specimens All POC specimens with known cytogenetics over a 2-year period from June 2013 to June 2015 at a single institution (Torrance Memorial Medical Center) were obtained. Cases with “no results” were removed from the sample. The remainder made up the study set, with the addition of two cases with unknown cytogenetics but diagnosed as partial hydatidiform mole after histologic consultation with a recognized expert in the field at an academic medical center. CISH Probes to chromosomes 17 and 10 labeled with DIG indicator (Ventana Medical Systems, Tucson, AZ) were chosen because of their commercial availability, pathologists’ familiarization with the method, and the relative infrequency of trisomy 17 and 10 in spontaneous abortions.22-24 Sections (4 lm) from paraffin-embedded tissue samples were baked at 72 C and deparaffinized using EZ prep solution. Slides were washed at 36 C with CC2, a citrate buffer at pH 6, for 52 minutes. ISH Protease 3 was applied for 24 minutes. Pretreatments were followed by a reaction buffer rinse. Following DNA denaturation at 86 C, the probe was applied. A stringency wash of SSC10 consisting of sodium citrate and sodium chloride was performed to wash off any unbound probe. The Ventana ultraView red ISH DIG detection kit (Ventana Medical Systems) was used to detect the labeled probes (24 minutes for both mouse anti-DIG antibody and alkaline phosphatase–conjugated goat anti-mouse antibody). The red visual signal in this kit was obtained by chromogenic substrates napthol and fast red reacting with alkaline phosphatase. After a reaction buffer rinse, Liquid Coverslip (LCS) was applied to ensure an aqueous environment prior to coverslipping. After a mild detergent bath to remove the LCS, slides were dried for 30 minutes in a 65 C oven and coverslipped. Test cases were selected and the assays for chromosomes 10 and 17 were optimized for each probe. CISH with both probes was performed on all cases in the study. CISH

© American Society for Clinical Pathology


was repeated once for all cases considered too weak to score or those cases with interfering background staining.

Table 1 Chromosome Complement of 52 Cases With Known Cytogenetics

Validation Study for Chromosomes 17 and 10

Cytogenetics

No. (%) of Cases

For phase 1 of the study (reference cutoff determination), 20 cases with normal cytogenetics or cytogenetically abnormal cases whose abnormalities were not detectable by the CISH assay (eg, trisomies, 45X) were selected. Both probes were then run on the 20 cases for determination of the reference cutoff. Four readers recorded a 50-cell count on each slide (total 200 cells per slide), and the number of copies of red signal was determined for each cell. The cutoff was then established with the b inverse function in Microsoft Excel with a confidence level of 95%, a method using a binomial statistical formula to project the upper bound of the 95th percentile that has been published in models of validation of analogous FISH probes.25,26 For phase 2 of the study, the cases studied were nine known triploid or tetraploid cases, one trisomy 10 case, 11 normal (46,XX or 46,XY) cases, and two cases without known cytogenetics but considered morphologically consistent with partial mole by academic pathology consultation. Initially, 50 cells were scored for each case for copy number of both chromosomes 17 and 10. If the count of greater than two signals was between three and six, an additional 50-cell count was performed.

46,XX or 46,XY Trisomy 16 Trisomy 22 69,XXX 69,XXY 92,XXYY 92,XXXX 45,X Trisomy 21 Trisomy 8 Trisomy 15 Trisomy 10 Trisomy 4 Trisomy 20 Trisomy 9 46,XX add 12(q24.3) 96,XXXX,þ15,þ15,þ21,þ21[13]/ 48,XXX,þ15,þ21[7]

19 (37) 5 (10) 5 (10) 3 (6) 3 (6) 2 (4) 1 (2) 2 (4) 2 (4) 2 (4) 2 (4) 1 (2) 1 (2) 1 (2) 1 (2) 1 (2) 1 (2)

Results Of the 84 cases originally sent for cytogenetics, 32 (38%) had no analysis possible. Details of the cytogenetic abnormalities of the remainder are in Table 1 . In summary, there were six triploid, three tetraploid, 19 normal, and 26 with assorted chromosome abnormalities. The most common of these were trisomy 22 (n ¼ 5) and trisomy 16 (n ¼ 5). There were two cases each of 45,X, trisomy 21, trisomy 8, and trisomy 15. Single occurrences included trisomy 14, trisomy 10, trisomy 4, and trisomy 20. Two cases had a complex karyotype (96,XXXX,þ15,þ15,þ21, þ21[13]/48,XXX,þ15,þ21[7] and 46,XX add(12)(q24.3)). In the reference range determination phase of the study Table 2 , the number of cells with greater than two signals in the 200 cell counts of the “normal” cases ranged from zero to seven for chromosome 17 and zero to eight for chromosome 10. Using the b inverse function with a 95% confidence level, a value of 6.8% (rounded up 7%) was obtained for the greater of the two. An abnormal result was thus defined when cells with greater than two signals exceeded this percentage of 7%. In phase 2 of the study Table 3 , the known triploid and tetraploid cases were studied along with additional normal

© American Society for Clinical Pathology Downloaded 127 from https://academic.oup.com/ajcp/article-abstract/146/1/125/2836654 by guest on 17 May 2018

Table 2 Reference Range Determination: Cells With More Than Two Signals (200 Cell Count) Case No.

Karyotype

Chromosome 10, No. (%)

Chromosome 17, No. (%)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

46,XX 46,XX 46,XX add(12)q24.3 47,XY,116 46,XX 47,XX,þ14 45,X 47,XX,þ22 47,XX,þ21 47,XY,þ8 46,XX 46,XY 47,XX,þ16 47,XY,þ22 47,XY,þ16 46,XX 47,XY,þ22 47,XX,þ22 47,XX,þ22 47,XX,þ16

8 (4) 3 (1.5) 2 (1) 2 (1) 0 (0.5) 7 (3.5) 0 (0) 6 (3) 3 (1.5) 3 (1.5) 1 (0.5) 4 (2) 0 (0) 1 (0.5) 1 (0.5) 4 (2) 0 (0) 2 (1) 0 (0) 2 (1)

4 (2) 1 (0.5) 5 (2.5) 1 (0.5) 7 (3.5) 2 (1) 1 (0.5) 0 (0) 2 (1) 3 (1.5) 4 (2) 6 (3) 2 (1) 5 (2.5) 2 (1) 4 (2) 2 (1) 0 (0) 0 (0) 3 (1.5)

cases, the trisomy 10 case, and the two cases diagnosed by outside, expert consultation. Initial counts on 50 cells were performed for each probe. When an abnormal signal number was detected in three to six cells, an additional 50-cell count was performed for a total of 100 cells (necessary in two cases). For the known triploid and tetraploid cases, all CISH counts were abnormal. For chromosome 17, the number of cells with abnormal signal from this group ranged from 32% to 72%. For chromosome 10, the number of cells with abnormal signal ranged from 34% to 66%. All of the


Kunesh et al / USE

OF

CISH

TO

ASSESS PLOIDY

IN

DIAGNOSIS

OF

HYDATIDIFORM MOLE

Table 3 Ploidy Determination Case No. Karyotype

% of Cells With Abnormal Signal 3, % of Cells With Abnormal Signal 3, 4 (3 1 4) Chromosome 10 4 (3 1 4) Chromosome 17 Result

1 2 3 4 5 6 7 8 9 10

34, 32 (66) 0, 0 (0) 48, 2 (50) 40, 0 (40) 4, 0 (4) 2, 0 (2) 4, 0 (4) 0, 0 (0) 24, 22 (46) 44, 0 (44)

40, 32 (72) 2, 0 (2) 40, 0 (40) 38, 0 (38) 2, 0 (2) 2, 0 (2) 0, 0 (0) 0, 0 (0) 22, 10 (32) 36, 0 (36)

Abnormal, suggest tetraploid Diploid Abnormal (triploid) Abnormal (triploid) Diploid Diploid Diploid Diploid Abnormal, suggest tetraploid Abnormal (triploid)

32, 2 (34) 30, 4 (34) 4, 0 (4) 4, 0 (4)

36, 0 (36) 34, 0 (34) 2, 0 (2) 2, 0 (2)

Abnormal (triploid) Abnormal (triploid) Diploid Diploid

5, 0 (5) 2, 0 (2) 42, 2 (44) 4, 0 (4) 22, 16 (38) 52, 6 (58) 2, 0 (2) 0, 0 (0) 42, 2 (44)

2, 0 (2) 2, 0 (2) 42, 4 (46) 2, 0 (2) 24, 10 (34) 56, 8 (64) 0, 0 (0) 0, 0 (0) 0, 0 (0)

Diploid Diploid Abnormal (triploid) Diploid Abnormal (suggest tetraploid) Abnormal (triploid) Diploid Diploid Abnormal (trisomy 10)

11 12 13 14 15 16 17 18 19 20 21 22 23

92,XXYY 46,XX 69,XXY 69,XXX 45,X 46,XY 46,XX 46,XX 92,XXYY No karyotype morphologic consultation 69,XXX 69,XXX 46,XX No karyotype morphologic consultation 46,XY 46,XY 69,XXY 47,XX,þ8 92,XXXX 69,XXY 47,XY,þ15 46,XY 47,XY,þ10

nontriploid or tetraploid cases had less than 7% abnormal signal (highest 5%). Of the two cases without known cytogenetics but thought to be partial moles by outside morphologic consultation, one was found to be abnormal by the study, but one was normal.

Discussion Diagnosis of molar pregnancies can be very difficult morphologically due to significant overlap with hydropic abortion.3,5-9 The determination is important, however, because the diagnosis of mole portends risk of gestational trophoblastic neoplasia and metastatic trophoblastic disease13 and thus requires close clinical follow-up with abstention from pregnancy during the monitoring period.1,2,4 This study shows that CISH is a viable technique for assessing ploidy in POC specimens and excluding partial hydatidiform mole. When used in conjunction with p57 immunohistochemistry, molar pregnancies can be excluded when morphology is ambiguous. The ability to perform the assay on automated platforms and read it on bright-field microscopes opens this test to a wider number of practices than FISH. Chromosome 17 was selected due to the familiarity with this probe in HER2 testing and the infrequency of trisomy 17.22-24,27 Because of the possibility of trisomy 17, another infrequent trisomy chromosome (chromosome 10)


was validated to be performed if chromosome 17 is abnormal. Although trisomy 10 is considered infrequent, we did have one case in our study. Most normal (nontriploid or tetraploid) cases had at least a few cells with greater than two signals. These could be due to overlapping nuclei. Although nonoverlapping nuclei were a scoring criteria in the assessment, this can sometimes be very difficult to detect. Restricting the count to villous stromal cells, when possible, helps to minimize this problem. When the red ISH signal is particularly strong, the signal can also become “smudgy.” In some cases, this can make it difficult to determine how many ISH signals are present. Both cases requiring an additional 50-cell count had particularly strong staining. Counting only clear, separate, unambiguous signals can help minimize this problem. During the study, mitotic figures were detected by H&E in both cytotrophoblasts and villous stromal cells Image 1 . Mitotically active cells could also contribute to greater than two signals in a normal nucleus. All three tetraploid cases produced results suggesting a tetraploid complement (the counts all had at least 10% of cells with four signals). Although this study was not designed to distinguish tetraploid from triploid, and not enough cases of each were believed to be present to make this determination, the results suggest that this is possible. As both partial moles and complete moles may sometimes be tetraploid,14 and the test is intended to be used in conjunction with p57 immunohistochemistry, determining



accessible technique that is possible in a wider range of pathology practices than other currently available methods. Corresponding author: John P. Kunesh, MD, Dept of Pathology, Torrance Memorial Medical Center, 3330 Lomita Blvd, Torrance, CA 90505; [email protected].

References

Image 1 Mitotic activity in stromal and trophoblastic cells (H&E, 600).

diploid vs nondiploid (triploid, tetraploid) is the important determination in excluding mole. One of the advantages of this technique is the ability to easily distinguish maternal tissue from fetal or placental tissue. This is a reported problem with FISH assays and thought to be a potential cause of reduced sensitivity of FISH for this type of ploidy assessment.5,20 The bright-field microscopy aspect of CISH makes it possible to score even very rare chorionic villi in a background of abundant maternal tissue Image 2 . Several of the stains had to be repeated due to weak signal, and one had diffuse red speckling background staining. The causes of these problems went unidentified, but repeating the stain alleviated the problems. It is interesting that of the two cases thought to be compatible with partial mole after morphologic consultative review and p57 staining only, one was determined to be diploid and, thus, normal by this study. Although a diploid partial mole could explain this, it is controversial whether diploid partial mole exists, and this is unlikely.20,28 More likely, this case exemplifies the difficulties in morphologic assessment, even by experts, and the importance of using more definitive ancillary testing. This study shows that CISH using an automated platform and bright-field microscopy is a viable method for assessing ploidy in products of conception. When used in conjunction with p57 immunohistochemical staining, these specimens can be definitively assessed without the need of an outsourced test. Bright-field microscopy makes it possible to assess even very rare villi with prominent background maternal tissue, making it potentially advantageous over FISH, flow cytometry, or other techniques available on paraffin-embedded tissue. It is an


1. Duggan P, Leung Y, Neesham D, et al. RANZCOG statement for the management of gestational trophobic disease. C-Gyn. 2013;31:1-18. 2. Cavaliere A, Ermito S, Dinatale A, et al. Management of molar pregnancy. J Prenat Med. 2009;3:15-17. 3. Cheung ANY, Lai CYL, Chan KYK, et al. Metastatic trophoblastic disease after an initial diagnosis of partial hydatidiform mole. Cancer. 2004;100:1411-1415. 4. Soper JT, Mutch DG, Schink JC; American College of Obstetricians and Gynecologists. Diagnosis and treatment of gestational trophoblastic disease. ACOG Practice Bulletin No. 53. Gynecol Oncol. 2004;93:575-585. 5. Kipp BR, Ketterling RP, Oberg TN, et al. Comparison of fluorescence in situ hybridization p57 Immunostaining, flow cytometry, and digital image analysis for diagnosing molar and nonmolar products of conception. Am J Clin Pathol. 2010;133:196-204. 6. Maggiori MS, Peres LC. Morphological, immunohistochemical and chromosome in situ hybridization in the differential diagnosis of hydatidiform mole and hydropic abortion. Eur J Obstet Gynecol Reprod Biol. 2007;135:170-176. 7. Wells M. The pathology of gestational trophoblastic disease: recent advances. Pathology. 2007;38:88-96. 8. Fukunaga M, Katabuchi H, Nagasaka T, et al. Interobserver and intraobserver variability in the diagnosis of hydatidiform mole. Am J Surg Pathol. 2005;29:942-947. 9. Howat AJ, Beck S, Fox H, et al. Can histopathologists reliably diagnose molar pregnancy? J Clin Pathol. 1993;46:599-602. 10. Crisp H, Burton JL, Stewert R, et al. Refining the diagnosis of hydatidiform mole: image ploidy analysis and p57KIP2 immunohistochemistry. Histopathology. 2003;43:363-373. 11. Jun SY, Ro JY, Kim KR. P57Kip2 is useful in the classification and differential diagnosis of complete and partial hydatidiform moles. Histopathology. 2003;44:17-25. 12. Merchant SH, Amin MB, Viswantha DS, et al. p57kip2 immunohistochemistry in early molar pregnancies: emphasis on its complementary role in the differential diagnosis of hyropic abortuses. Hum Pathol. 2005;36:180-186. 13. Romuguera RL, Rodriguez MM, Bruce JH, et al. Molar gestations and hydropic abortions differentiated by p57 immunostaining. Fetal Pediatr Pathol. 2004;23:181-190. 14. Lage JM, Weinberg DS, Yavner DL, et al. The biology of tetraploid hydatidiform moles; histopathology, cytogenetics, and flow cytometry. Hum Pathol. 1989;20:419-425. 15. Topalovski M, Hankin RC, Michael C, et al. Ploidy analysis of products of conception by image and flow cytometry with cytogenetic correlation. Am J Clin Pathol. 1995;103:409-414.


Kunesh et al / USE

OF

CISH

TO

ASSESS PLOIDY

IN

DIAGNOSIS

OF

HYDATIDIFORM MOLE

A

B

C

D

Image 2 A, Bright-field microscopy enables identification of even rare chorionic villi (arrow) in a background of maternal tissue (40). B, Diploid POC specimen with two signals (red dots) in most cells (600). C, Triploid mole with three signals in many cells (600). D, Tetraploid mole with four signals in many cells (600).

16. Cheville JC, Grenier T, Robinson RA, et al. Ploidy analysis by flow cytometry and fluorescence in situ hybridization in hydropic placentas and gestational trophoblastic disease. Hum Pathol. 1995;26:753-757. 17. Barclay ID, Dabbagh L, Babiac J, et al. DNA analysis (ploidy) of molar pregnancies with image analysis or paraffin tissue sections. Am J Surg Pathol. 1993;100: 451-455. 18. Berezowsky J, Zbieranowski I, Demers J, et al. DNA ploidy of hydatidiform moles and nonmolar conceptuses: a study using flow and tissue section image cytometry. Mod Pathol. 1995;8:775-781. 19. Gschwendtner A, Neher A, Kreczy A, et al. DNA ploidy of hydatidiform mole: image analysis and p57KIP2 immunohistochemistry. Histopathology. 2003;43:363-373.


20. LeGallo RD, Stelow EB, Nilsa CR, et al. Diagnosis of hydatidiform moles using p57 Immunohistochemistry and HER2 fluorescent in situ hybridization. Am J Clin Pathol. 2008;129:749-755. 21. Sholl LM, Iafrate AJ, Chou YP, et al. Validation of chromogenic in situ hybridization for detection of EGFR copy number amplification in nonsmall cell lung carcinoma. Mod Pathol. 2007;20:1028-1035. 22. Wang BT, Chong TP, Boyar FZ, et al. Abnormalities in spontaneous abortions detected by G-banding and chromosomal microarray array (CMA) at a national reference laboratory. Mol Cytogenet. 2014;7:33-39. 23. Arakaki DT, Waxman SH. Chromosome abnormalities in early spontaneous abortions. J Med Genet. 1970;7: 118-124.



24. Menasha J, Levy B, Hirschhorn K, et al. Incidence and spectrum of chromosome abnormalities in spontaneous abortions: new insights from a 12-year study. Genet Med. 2005;7:251-263. 25. Saxe DF, Persons DL, Wolff DJ, et al. Validation of fluorescence in situ hybridization using an analyte-specific reagent for detection of abnormalities involving the mixed lineage leukemia gene. Arch Pathol Lab Med. 2012;136:47-52. 26. Wiktor AE, Van Dyke DL, Stupca PJ, et al. Preclinical validation of fluorescence in situ hybridization assays for clinical practice. Genet Med. 2006;8:16-23.


27. Kim JW, Lee WS, Yoon TK, et al. Chromosomal abnormalities in spontaneous abortion after assisted reproductive treatment. BMC Med Genet. 2010;11:153-159. 28. Genest DR, Ruiz RE, Weremowicz S, et al. Do nontriploid partial moles exist? A histologic and flow cytometric reevaluation of nontriploid specimens. J Reprod Med. 2002;47:363-368.
