Development of a Chromosomal Microarray Test for the Detection of Abnormalities in Formalin-Fixed, Paraffin-Embedded Products of Conception Specimens
2017; Elsevier BV; Volume: 19; Issue: 6 Linguagem: Inglês
10.1016/j.jmoldx.2017.07.001
ISSN1943-7811
Autores Tópico(s)Genetic Syndromes and Imprinting
ResumoTesting the products of conception (POCs) provides information about the cause of fetal loss and helps determine the recurrence risk of future losses and chromosome abnormalities in subsequent pregnancies. Historically, the Mayo Clinic Cytogenetics Laboratory performed targeted fluorescent in situ hybridization (FISH) testing to identify aneuploidy of only certain chromosomes in formalin-fixed, paraffin-embedded (FFPE) POC samples. Chromosomal microarray studies using the Affymetrix OncoScan FFPE Assay can detect copy number changes across the genome. We validated the utility of the OncoScan assay using 25 archival FFPE POC specimens with previous FISH results (five normal, 12 trisomy, six triploidy, two monosomy). Of the five normal samples, four had no clinically relevant findings, and one sample was found to have trisomy 9, which is not detectable by the FISH test. For the 20 samples with abnormal FISH results, the OncoScan assay identified all reported abnormalities along with additional findings. A sample with trisomy 22 was also found to have trisomy 7. Another sample reported as triploidy was found to have four copies of chromosome 16. In conclusion, we verified the performance characteristics of OncoScan on FFPE POC specimens and found it acceptable for clinical use. Additional information was identified in 3 of 25 cases (12%) that would explain the pregnancy loss or provide recurrence risk for the family. Testing the products of conception (POCs) provides information about the cause of fetal loss and helps determine the recurrence risk of future losses and chromosome abnormalities in subsequent pregnancies. Historically, the Mayo Clinic Cytogenetics Laboratory performed targeted fluorescent in situ hybridization (FISH) testing to identify aneuploidy of only certain chromosomes in formalin-fixed, paraffin-embedded (FFPE) POC samples. Chromosomal microarray studies using the Affymetrix OncoScan FFPE Assay can detect copy number changes across the genome. We validated the utility of the OncoScan assay using 25 archival FFPE POC specimens with previous FISH results (five normal, 12 trisomy, six triploidy, two monosomy). Of the five normal samples, four had no clinically relevant findings, and one sample was found to have trisomy 9, which is not detectable by the FISH test. For the 20 samples with abnormal FISH results, the OncoScan assay identified all reported abnormalities along with additional findings. A sample with trisomy 22 was also found to have trisomy 7. Another sample reported as triploidy was found to have four copies of chromosome 16. In conclusion, we verified the performance characteristics of OncoScan on FFPE POC specimens and found it acceptable for clinical use. Additional information was identified in 3 of 25 cases (12%) that would explain the pregnancy loss or provide recurrence risk for the family. In the general reproductive population, miscarriages before 20 weeks' gestation are reported as 10% to 15% of the clinically recognized pregnancies.1Stephenson M.D. Awartani K.A. Robinson W.P. Cytogenetic analysis of miscarriages from couples with recurrent miscarriage: a case-control study.Hum Reprod. 2002; 17: 446-451Crossref PubMed Scopus (378) Google Scholar Of the first trimester spontaneously aborted fetuses, it is estimated that approximately 50% have an underlying chromosomal abnormality, the most common being trisomy, polyploidy, and monosomy X, respectively. Therefore, chromosome studies may provide useful information about the cause of fetal loss. With recurrent miscarriages affecting 3% of couples,1Stephenson M.D. Awartani K.A. Robinson W.P. Cytogenetic analysis of miscarriages from couples with recurrent miscarriage: a case-control study.Hum Reprod. 2002; 17: 446-451Crossref PubMed Scopus (378) Google Scholar chromosome studies may also provide information about recurrence risk of future fetal losses and the risk of having subsequent children with chromosome abnormalities. Identification of an unbalanced translocation in the fetus could characterize a balanced translocation in an asymptomatic parent, which would further aid in counseling parents about future risk of pregnancy loss. Conventional cytogenetic techniques such as karyotyping and fluorescent in situ hybridization (FISH) have been used to identify chromosomal abnormalities. Unfortunately, they have fundamental disadvantages such as culture failure and maternal cell contamination. The use of postmortem formalin-fixed, paraffin-embedded (FFPE) samples for genetic analysis can be informative but challenging. For FFPE samples, conventional karyotyping is not possible because of the lack of viable tissue for culture. FISH and chromosomal microarray (CMA) are able to generate data from FFPE samples. Historically, the Mayo Clinic Cytogenetics Laboratory performed targeted FISH testing on FFPE products of conception (POCs) for aneuploidy. This consists of probes for chromosomes X(Xp11.1-q11.1), Y(Yp11.1-q11.1), 13(13q14), 15(15p11.1-q11.1), 16(16q11.2), 18(18p11.1-q11.1), 21(21q22.13-22.2), and 22(22q11.2). The remaining chromosomes are not interrogated. Although the FISH test for interrogating FFPE POC samples has been considered the gold standard, there are disadvantages. The abnormality detection is probe specific, and any other copy number change, outside of the probed area, will not be detected. It will not detect terminal deletions or duplications or unbalanced translocations. In addition, FISH analysis of FFPE specimens is hindered by factors such as poor hybridization, sectioning through cells and only having a partial cell present (truncation, cut artifact), and cell overlap. The presence of maternal cells in the fetal sample can cause false-negative FISH results because of the presence of normal maternal cells.2Bell K.A. Van Deerlin P.G. Haddad B.R. Feinberg R.F. Cytogenetic diagnosis of "normal 46,XX" karyotypes in spontaneous abortions frequently may be misleading.Fertil Steril. 1999; 71: 334-341Abstract Full Text Full Text PDF PubMed Scopus (111) Google Scholar, 3Robberecht C. Schuddinck V. Fryns J.P. Vermeesch J.R. Diagnosis of miscarriages by molecular karyotyping: benefits and pitfalls.Genet Med. 2009; 11: 646-654Abstract Full Text Full Text PDF PubMed Scopus (79) Google Scholar However, the use of single nucleotide polymorphism (SNP) arrays can detect the presence of contamination from maternal cells, effectively reducing the chance of a false-negative result.4Sahoo T. Dzidic N. Strecker M.N. Commander S. Travis M.K. Doherty C. Tyson R.W. Mendoza A.E. Stephenson M. Dise C.A. Benito C.W. Ziadie M.S. Hovanes K. Comprehensive genetic analysis of pregnancy loss by chromosomal microarrays: outcomes, benefits, and challenges.Genet Med. 2016; 19: 83-89Crossref PubMed Scopus (88) Google Scholar, 5Levy B. Sigurjonsson S. Pettersen B. Maisenbacher M.K. Hall M.P. Demko Z. Lathi R.B. Tao R. Aggarwal V. Rabinowitz M. Genomic imbalance in products of conception: single-nucleotide polymorphism chromosomal microarray analysis.Obstet Gynecol. 2014; 124: 202-209Crossref PubMed Scopus (131) Google Scholar A SNP-based CMA is a useful tool to obtain a genome-wide analysis of FFPE POC samples, but the formalin fixation process degrades the DNA. The degraded DNA, short fragment length, and DNA crosslinking can compromise results on other array platforms.6Rowe L.R. Thaker H.M. Opitz J.M. Schiffman J.D. Haddadin Z.M. Erickson L.K. South S.T. Molecular inversion probe array for the genetic evaluation of stillbirth using formalin-fixed, paraffin-embedded tissue.J Mol Diagn. 2013; 15: 466-472Abstract Full Text Full Text PDF PubMed Scopus (10) Google Scholar The Affymetrix OncoScan FFPE assay (Affymetrix, Santa Clara, CA) was chosen because it was designed to give good quality data from FFPE samples. The chemistry of the assay uses molecular inversion probes, which are optimized for the use of degraded and fragmented DNA caused by formalin fixation. It requires a low input of genomic DNA, approximately 80 ng. The OncoScan assay uses >220,000 markers for the detection of copy number changes at about 500 kb genome-wide copy number resolution and regions with absence of heterozygosity (AOH). The SNP-based probes offer genome-wide coverage and aid in the identification of triploidy, complete mole, uniparental isodisomy, and consanguinity. In this study, we test the performance characteristics of the OncoScan assay on FFPE POC specimens and validate the assay for clinical use. Validation was conducted on 25 archived FFPE POC samples with previous FISH testing for aneuploidy detection. These consisted of five specimens that were reported with normal FISH results and 20 specimens that were reported with abnormal FISH results; 12 trisomy samples, six triploidy samples, and two monosomy samples. No samples had known regions of AOH because it is not detectable by FISH testing. Samples with known aberrations near the detection limit of the array were not available in the positive control set. The unused FFPE slides (5 μm thick) from the archived clinical samples were collected for DNA extraction. Five 1-cm2 squares were scraped from the tissue on the slides matching the region indicated as fetal tissue on the hematoxylin and eosin-stained slide that was reviewed by the pathologist. DNA was extracted using the QIAamp DSP DNA FFPE Tissue Kit (Qiagen, Valencia, CA) with some modifications to the Qiagen protocol recommended by Affymetrix for the OncoScan assay. The process starts with increasing the amount of tissue being scraped from the slides. In the modified protocol, 100 μL of 100% ethanol is added after xylene to help pellet the cells. The Qiagen protocol proceeds to centrifugation without the addition of ethanol. The modified protocol includes an overnight digestion versus Qiagen's 1-hour digestion. After the overnight digestion, the lysate is incubated at 90°C to reverse formalin crosslinks. Before Buffer AL is added, the modified protocol states to cool the sample to room temperature and then add RNase A to the sample. After the DNA is purified on the MinElute columns, the centrifuge speed is increased in the modified protocol to aid in drying the membrane completely. The modified protocol also decreases the amount of elution buffer from 200 to 30 μL. These modifications increased our yield of the DNA during OncoScan validation and were adopted for our FFPE extractions. The concentration of the genomic DNA was determined using the Qubit 2.0 Fluorometer (Life Technologies, Grand Island, NY). The FFPE POC samples were hybridized with the OncoScan FFPE assay kit. The OncoScan FFPE Assay Kit (Affymetrix) was then applied to the isolated DNA. The technical details of the molecular inversion probe assay have been previously described.7Wang Y. Cottman M. Schiffman J.D. Molecular inversion probes: a novel microarray technology and its application in cancer research.Cancer Genet. 2012; 205: 341-355Abstract Full Text Full Text PDF PubMed Scopus (88) Google Scholar, 8Ji H. Welch K. Molecular inversion probe assay for allelic quantitation.Methods Mol Biol. 2009; 556: 67-87Crossref PubMed Scopus (11) Google Scholar, 9Wang Y. Carlton V. Karlin-Neumann G. Sapolsky R. Zhang L. Moorhead M. Wang Z.C. Richardson A.L. Warren R. Walther A. Bondy M. Sahin A. Krahe R. Tuna M. Thompson P.A. Spellman P.T. Gray J.W. Mills G.B. Faham M. High quality copy number and genotype data from FFPE samples using Molecular Inversion Probe (MIP) microarrays.BMC Med Genomics. 2009; 2: 8Crossref PubMed Scopus (82) Google Scholar Briefly, a molecular inversion probe targeting a unique SNP or base pair of interest anneals to isolated DNA and circularizes with its complementary nucleotide. Single-stranded material is destroyed, and the circular probe is then released, cleaved, becomes inverted, and is amplified using universal primers. The biotinylated oligo is hybridized overnight to the microarray. Two microarray chips are used per sample, one for the AT incorporation and the other for the GC incorporation. Microarray results were then visualized using the Chromosome Analysis Suite software version 3.1 (Affymetrix, Santa Clara, CA). The quality metrics for the OncoScan array are recommended by Affymetrix; ndSnpQC >26.0, mapd <0.3, and ndWaviness 1 Mb and duplications >2 Mb; however, smaller changes with pathogenic potential may also be reported, whereas larger changes that are well-documented benign variants were not reported. AOH involving chromosomes known to have uniparental isodisomy–associated phenotypes and genome-wide AOH involving >10% of the autosomal DNA complement were also reported. In this study, 25 archival FFPE POC samples that had undergone previous FISH testing were analyzed (Table 1). For the five samples with normal FISH results; four samples had no clinically relevant findings by microarray. One of the samples with normal FISH results (sample 6) was found to have trisomy 9 by microarray (Figure 1). The FISH assay does not currently have a probe for chromosome 9; therefore, trisomy 9 was not detectable with the FISH assay. For the 20 samples with previously reported abnormalities by the FISH test, all reported aneuploidies were identified with the OncoScan FFPE array. Additional abnormalities were identified in two samples. Sample 14, identified as having trisomy 22 by FISH, was also found to have trisomy 7 by microarray (Figure 2). Sample 17 was identified as a triploidy sample by the FISH test (Figure 3). The OncoScan array detected triploidy in this sample but also identified an additional copy of chromosome 16 (four copies of chromosome 16). The array results for this sample showed the necessity for having SNP probes on the array. The copy number probes alone were not able to detect triploidy and could only identify this abnormality as a trisomy 16. SNP probes aided in the identification of triploidy and the additional chromosome 16 (tetrasomy 16). It was unexpected that the FISH assay missed the chromosome 16 abnormality because the assay has a probe on chromosome 16. The FISH slides for this sample were re-reviewed, and no clear evidence of tetrasomy 16 was observed. This case illustrated a disadvantage of FISH, which is the difficulty to score four copies of FISH signal, especially in the background of triploidy. No samples tested had previously known regions of AOH. AOH was detected but did not meet the reporting criteria.Table 1Chromosomal Microarray and Previous FISH Results of Archived FFPE POC Samples Used for the Validation of the Affymetrix Oncoscan ArraySampleCMA resultsPrevious FISH resultsConcordant1Trisomy 16Trisomy 16Yes2Triploid XXXTriploid XXXYes3Trisomy 18Trisomy 18Yes4NormalNormalYes5Triploid XXYTriploid XXYYes6Trisomy 9NormalSee Discussion7Trisomy 16Trisomy 16Yes8Trisomy 13Trisomy 13Yes9Trisomy 18Trisomy 18Yes10Trisomy 21Trisomy 21Yes11Triploid XYYTriploid XYYYes12Trisomy 13Trisomy 13Yes13NormalNormalYes14Trisomy 7, Trisomy 22Trisomy 22See Discussion15Trisomy 22Trisomy 22Yes16Trisomy 15Trisomy 15Yes17Triploid XYY, Tetrasomy 16Triploid XYYSee Discussion18Triploid XXXTriploid XXXYes19NormalNormalYes20Triploid XXYTriploid XXYYes21NormalNormalYes22Monosomy XMonosomy XYes23Trisomy 15Trisomy 15Yes24Monosomy XMonosomy XYes25Trisomy 21Trisomy 21YesCMA, chromosomal microarray; FISH, fluorescence in situ hybridization; FFPE, formalin-fixed, paraffin-embedded; POC, product of conception. Open table in a new tab Figure 2Whole-genome view of the copy number detection (log2 ratio; top plot) and the single nucleotide polymorphism detection (allele difference; bottom plot) for sample 14, which was a female fetus with trisomies 7 and 22.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Figure 3Whole-genome view of sample 17 showing triploidy on single nucleotide polymorphism (SNP) detection (allele difference; bottom plot) and an additional copy of chromosome 16 on both copy number detection (log2 ratio; top plot) and SNP detection (bottom plot).View Large Image Figure ViewerDownload Hi-res image Download (PPT) CMA, chromosomal microarray; FISH, fluorescence in situ hybridization; FFPE, formalin-fixed, paraffin-embedded; POC, product of conception. It is estimated that approximately 5% to 10.5% of all pregnancies result in spontaneous abortions caused by chromosomal abnormalities.10Ogasawara M. Aoki K. Okada S. Suzumori K. Embryonic karyotype of abortuses in relation to the number of previous miscarriages.Fertil Steril. 2000; 73: 300-304Abstract Full Text Full Text PDF PubMed Scopus (403) Google Scholar Identification of chromosomal abnormalities in these fetuses can be helpful for various reasons. This can help patients understand why they experienced a pregnancy loss and help them overcome the grief associated with a loss and the anxiety for subsequent pregnancies.11Stephenson M.D. Frequency of factors associated with habitual abortion in 197 couples.Fertil Steril. 1996; 66: 24-29Abstract Full Text PDF PubMed Google Scholar Diagnosing a chromosomal abnormality can provide recurrence risk for the family. Although certain abnormalities (monosomy X, triploidy) do not have an increased recurrence risk and are not associated with maternal age effect, abnormalities such as trisomies have a slightly increased risk in comparison with the usual maternal age-associated risk.12Warburton D. Dallaire L. Thangavelu M. Ross L. Levin B. Kline J. Trisomy recurrence: a reconsideration based on North American data.Am J Hum Genet. 2004; 75: 376-385Abstract Full Text Full Text PDF PubMed Scopus (126) Google Scholar, 13Lindor N.M. Ney J.A. Gaffey T.A. Jenkins R.B. Thibodeau S.N. Dewald G.W. A genetic review of complete and partial hydatidiform moles and nonmolar triploidy.Mayo Clin Proc. 1992; 67: 791-799Abstract Full Text Full Text PDF PubMed Scopus (51) Google Scholar Identification of a familial chromosome rearrangement as a reason for the loss would prevent unnecessary testing to identify the cause of additional losses and would also provide the opportunity for prenatal and pre-implantation diagnosis.14Wou K. Hyun Y. Chitayat D. Vlasschaert M. Chong K. Wasim S. Keating S. Shannon P. Kolomiet E. Analysis of tissue from products of conception and perinatal losses using QF-PCR and microarray: a three-year retrospective study resulting in an efficient protocol.Eur J Med Genet. 2016; 59: 417-424Crossref PubMed Scopus (12) Google Scholar This would also help other family members to perform testing before planning for a pregnancy. To perform testing on the POC tissue, a fresh specimen may not always be available; therefore, FFPE tissue can be used for testing. Historically, the only testing method available for FFPE POC tissue has been FISH. However, FISH testing is limited to targeted chromosomes with common aneuploidies, thereby having a limited diagnostic value. CMA testing allows for increased diagnostic utility; however, many microarray platforms are not able to overcome the challenges of using FFPE samples. Because the OncoScan FFPE assay is optimized for using FFPE samples, it is preferable for testing FFPE POC samples. The OncoScan FFPE assay is able to use degraded DNA and output SNP data that can be used in detecting copy number changes and absence of heterozygosity. The study by Rowe et al6Rowe L.R. Thaker H.M. Opitz J.M. Schiffman J.D. Haddadin Z.M. Erickson L.K. South S.T. Molecular inversion probe array for the genetic evaluation of stillbirth using formalin-fixed, paraffin-embedded tissue.J Mol Diagn. 2013; 15: 466-472Abstract Full Text Full Text PDF PubMed Scopus (10) Google Scholar was the first to use this assay on archival FFPE fetal autopsy samples and identified chromosomal abnormalities that were not detected by conventional cytogenetics. Since then, there have only been a few studies using CMA on FFPE POC samples. Kudesia et al15Kudesia R. Li M. Smith J. Patel A. Williams Z. Rescue karyotyping: a case series of array-based comparative genomic hybridization evaluation of archival conceptual tissue.Reprod Biol Endocrinol. 2014; 12: 19Crossref PubMed Scopus (20) Google Scholar ran 20 samples on a non-SNP platform with an 80% success rate, and, despite the poor quality metrics, a higher yield of abnormalities was detected compared with traditional methods. Maslow et al16Maslow B.S. Budinetz T. Sueldo C. Anspach E. Engmann L. Benadiva C. Nulsen III, J.C. Single-nucleotide polymorphism-microarray ploidy analysis of paraffin-embedded products of conception in recurrent pregnancy loss evaluations.Obstet Gynecol. 2015; 126: 175-181Crossref PubMed Scopus (18) Google Scholar ran 62 samples on a SNP platform with a 71% success rate and highlighted the importance of SNP probes in the identification of maternal cell contamination. The most recent study by Sahoo et al4Sahoo T. Dzidic N. Strecker M.N. Commander S. Travis M.K. Doherty C. Tyson R.W. Mendoza A.E. Stephenson M. Dise C.A. Benito C.W. Ziadie M.S. Hovanes K. Comprehensive genetic analysis of pregnancy loss by chromosomal microarrays: outcomes, benefits, and challenges.Genet Med. 2016; 19: 83-89Crossref PubMed Scopus (88) Google Scholar had 1832 FFPE samples, and the researchers were able to achieve an 86.4% success rate with most common failure being insufficient fetal tissue. They also found that maternal cell contamination was higher in FFPE samples and that 5.6% of their samples would not have been reportable with conventional cytogenetics. In this study, we have validated the Oncoscan FFPE assay for routine clinical use on FFPE POC specimens. We were able to detect all previously reported abnormalities in our cohort of samples. In addition, three samples (of 25; 12%) were found to have previously undetected abnormalities that would have helped explain pregnancy loss or provide recurrence risk for future pregnancies (samples 6, 14, and 17). Sample 6 had a normal FISH result, but the CMA identified trisomy 9. This would indicate an increased recurrence risk in comparison with the usual maternal age-associated risk.11Stephenson M.D. Frequency of factors associated with habitual abortion in 197 couples.Fertil Steril. 1996; 66: 24-29Abstract Full Text PDF PubMed Google Scholar Although sample 14 had a trisomy 22 result, CMA detected both trisomy 22 and trisomy 7. Double trisomy is rare, accounting for 0.21% to 2.8% of karyotyped miscarriages17Diego-Alvarez D. Ramos-Corrales C. Garcia-Hoyos M. Bustamante-Aragones A. Cantalapiedra D. Diaz-Recasens J. Vallespin-Garcia E. Ayuso C. Lorda-Sanchez I. Double trisomy in spontaneous miscarriages: cytogenetic and molecular approach.Hum Reprod. 2006; 21: 958-966Crossref PubMed Scopus (57) Google Scholar and is thought to come about from two nondisjunction events during meiosis.18Subramaniyam S. Pulijaal V.R. Mathew S. Double and multiple chromosomal aneuploidies in spontaneous abortions: a single institutional experience.J Hum Reprod Sci. 2014; 7: 262-268Crossref PubMed Scopus (19) Google Scholar This case received an increased risk of recurrence from the FISH result of trisomy 22, and the identification of a double trisomy is thought to carry the same maternal age-related risk of recurrence as a single trisomy. It has been noted, however, that double trisomy is found more frequently with increased maternal age (34.1 ± 5.7 years),19Reddy K.S. Double trisomy in spontaneous abortions.Hum Genet. 1997; 101: 339-345Crossref PubMed Scopus (56) Google Scholar, 20Segawa T. Kuroda T. Kato K. Kuroda M. Omi K. Miyauchi O. Watanabe Y. Okubo T. Osada H. Teramoto S. Cytogenetic analysis of the retained products of conception after missed abortion following blastocyst transfer: a retrospective, large-scale, single-centre study.Reprod Biomed Online. 2017; 34: 203-210Abstract Full Text Full Text PDF PubMed Scopus (17) Google Scholar and our patient was 38 years old. Sample 17 had a triploid result by FISH which would not indicate an increased recurrence risk; however, CMA also identified an additional copy of chromosome 16. This finding would indicate an increased recurrence risk of a trisomic conception for the patient.21Waltman L.A. Eckel-Passow J.E. Sharma R.G. Van Dyke D.L. Advanced maternal age in polyploidy with concurrent aneuploidy.Am J Med Genet A. 2013; 161A: 1200-1202Crossref PubMed Scopus (4) Google Scholar The OncoScan FFPE assay is a superior method over FISH studies on FFPE POC samples. The ability to interrogate the whole genome versus a targeted assay increases the diagnostic yield for these samples. In addition, the SNP probes allow for better diagnosis of triploid samples and aid in the diagnosis of complete mole, uniparental isodisomy, and consanguinity, further increasing the diagnostic yield.
Referência(s)