This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Antill, Y. C. Articles by Campbell, I. G. Search for Related Content PubMed PubMed Citation Articles by Antill, Y. C. Articles by Campbell, I. G.

Loss of Heterozygosity Analysis in Ductal Lavage Samples from BRCA1 and BRCA2 Carriers: A Cautionary Tale

¹ Victorian Breast Cancer Research Consortium Cancer Genetics Laboratory, Research Division, ² Department of Haematology and Medical Oncology, ³ Research Division, ⁴ Centre for Biostatistics and Clinical Trials, Peter MacCallum Cancer Centre, and ⁵ The University of Melbourne, Melbourne, Australia

Requests for reprints: Yoland Antill, Department of Haematology and Medical Oncology, Peter MacCallum Cancer Centre, Locked Bag 1, A'Beckett Street, Victoria 8006, Australia. Phone: 61-3-9656-1111; Fax: 61-3-9656-1539. E-mail: Yoland.Antill{at}petermac.org.

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Background: Loss of heterozygosity (LOH) in breast ductal lavage (DL) fluid has been reported to be a potential biomarker of malignant change. Interpretation of LOH is reliant on sufficient quality and quantity of DNA. We investigated LOH of the BRCA1/2 loci in DL samples from BRCA1/2 mutation carriers, while also assessing the effect of DNA quantity.

Methods: DNA yield was estimated using quantitative real-time PCR. Allelic status of DL DNA was determined using fluorescently tagged microsatellite markers with the subject's lymphocytic DNA serving as a control. Samples were scored as consistently heterozygous or as demonstrating LOH if the same result was observed in replicate experiments. Additionally, samples were scored as "discordant LOH" if they initially showed LOH, but in replicate experiments either showed heterozygosity or LOH of the opposite allele.

Results: In 11 BRCA1 carriers, 46 ducts were assessable, and 39 ducts from 14 BRCA2 carriers were assessable. LOH was observed in 17% and 18% of ducts from BRCA1 and BRCA2, respectively. Discordant results were seen in 23 BRCA1 (50%) and 15 BRCA2 (38%) samples. DNA yield was significantly greater in samples that were consistently heterozygous than those that were either discordant or showed LOH in replicate experiments for both BRCA1 (P = 0.003) and BRCA2 (P = 0.003).

Conclusions: DNA quantity is highly variable between DL samples, with low yields likely to detrimentally affect the interpretation of LOH. In conclusion, LOH may not be an adequate method to detect the early stages of malignant change in samples obtained via DL. (Cancer Epidemiol Biomarkers Prev 2006;15(7):1396–8)

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Women who carry a germline mutation in BRCA1 or BRCA2 have significantly increased risks of breast cancer development, and at an earlier age, than for women at population risk (1, 2). The sensitivity of screening is lower in younger women and may be even less effective in BRCA1 and BRCA2 carriers (3). Better strategies for early detection and individual risk assessment clearly remain a priority for this group; in particular, there is a need for specific and sensitive biomarkers indicative of breast cancer risk. Methods of sampling ductal epithelial cells include nipple aspiration, fine needle biopsy, and ductal lavage (DL). DL has the potential benefit of greater yields of intact cells than aspiration, is less invasive than fine needle aspiration (4), and allows repeated sampling of the same ductal system providing the potential for prospective tracking of biomarker changes over time. DL effluent typically contains a variety of cell types including foam cells, macrophages, lymphocytes, and either loose or clusters of epithelial cells (4, 5).

In choosing an appropriate biomarker, some investigators have made use of the fact that extremely high rates of loss of heterozygosity (LOH) are observed at the BRCA1 and BRCA2 loci in breast cancers from BRCA1 or BRCA2 carriers, respectively (6, 7). Isaacs et al. (8) reported LOH in 58% of DL samples taken from breasts unaffected with cancers in BRCA1 carriers. Yonekura et al. (9) reported a rate of 74% to 93% LOH in DL samples from breasts affected with cancer, 60% to 84% in the contralateral, unaffected breast, and up to 67% in women with benign breast disease. Although these data seem promising, on theoretical grounds, the identification of LOH in the context of the high levels of contamination of DL fluid with nonepithelial cells (up to 100%; refs. 4, 5) is perhaps surprising. In samples with a low cellular count, the yield of DNA derived from ductal epithelium is likely to be very low. As a consequence, LOH analysis would be highly susceptible to preferential allele amplification, a well-recognized hazard when undertaking PCR using low genome copy numbers (10). Whether or not this prejudiced the reporting of LOH in previous DL studies (8, 9) is not known. In this study, DL samples obtained from women with germline BRCA1 or BRCA2 mutations were assessed for LOH at 17q.21 (in BRCA1 carriers) and at 13q.12 (in BRCA2 carriers) with the inclusion of specific controls to address the reliability of the LOH assay and to correlate the frequency of LOH with DNA yield.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Subjects
As described elsewhere, women with a known germline mutation in BRCA1 or BRCA2, who had at least one breast unaffected by cancer, were eligible for this study (11). The study was approved by the relevant institutional ethics committees and all participants provided written informed consent. DL, nipple aspiration, and venesection were completed every 6 months from the time of study entry for a maximum of 3 years with clinical follow-up to continue annually for 10 years following the cessation of sample collection.

DL
Lavage was done using the FirstCyte DL catheter (Cytyc Corporation, Boxborough, MA) as previously described (4) with minor modifications to operator techniques (11). Cannulation was attempted on both nipple aspirate fluid–producing and any additional "dry" ducts detected by gentle probing, with a maximum of three ducts per visit. If nipple aspiration fluid was produced, attempts were made to cannulate the specific nipple aspirate fluid–producing ducts.

Sample Processing
Between 10 and 12 mL of DL washings were recovered from each duct with a 2 to 5 mL aliquot dispensed into 30 mL Cytolyt solution (Cytyc Corporation) for cytologic processing. The remaining fluid was processed within 2 hours into pellets of 10,000 cells before snap-freezing and storage at –80°C.

DNA Isolation and Quantitation
DNA was extracted from DL cellular pellets using the DNAeasy Tissue Kit (Qiagen, Valencia, CA) and lymphocytic DNA was processed using the Mini-blood kit (Qiagen). The DNA yield was estimated for each sample using quantitative real-time PCR as described previously (12). Quantities of DNA were estimated by plotting against a standard curve of known dilutions of DNA derived from normal lymphocytes.

LOH
Microsatellite markers within and/or flanking BRCA1 (17q21; D17S800, D17S1322, D17S855, and D17S579) and BRCA2 (13q12.3; D13S221, D13S260, D13S171, and D13S267) were used to assess LOH. The forward primer was labeled with a 5'-fluorescent dye (FAM or HEX). LOH was assessed by electrophoresis on a 6% denaturing polyacrylamide gel by comparing the relative intensities of the alleles derived from lavage fluid with the allele intensities from the same subject's lymphocytic DNA. Because of the heterogeneous nature of cellular content in DL samples, and the likelihood of normal cell "contamination," LOH of an individual marker was scored if the intensity of an allele in the lavage sample showed a >20% reduction when compared with the subject's lymphocytic DNA. All LOH analyses were done at least twice. A call of LOH was made if a sample showed loss with at least two markers at that locus. Samples were scored as showing concordant heterozygosity if no loss was shown on any occasion and as having concordant LOH if loss of the same allele was repeatable in replicate assays. A sample was scored as discordant for LOH if they initially showed a >20% reduction in allele intensity, but in replicate experiments showed either no LOH (heterozygous) or LOH of the opposite allele.

As an additional control, LOH was assessed in a subset of samples using two microsatellite markers (D1S123 and D12S385), recognized to be infrequently lost in breast cancer.

Statistical Analysis
The Kruskal-Wallis test was used to compare the amount of amplifiable DNA among samples that were heterozygous, discordant, or showed LOH. Fisher's exact test was used to compare the proportion of samples with atypia (mild, severe, or both) between heterozygous and LOH groups. All tests were two-sided. Statistical analyses were done using StatXact 6.0 software (Cytel Software Corporation, Cambridge, MA).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Subjects
Thirty-four women (16 BRCA1 and 18 BRCA2 mutation carriers), with a median age of 42.5 years, had at least one lavage done and were eligible for this study (Table 1 ). Nine were excluded from analysis (five BRCA1 and four BRCA2); two were not informative for at least two microsatellite markers, six had insufficient amounts of DNA for amplification, and one had no matching lymphocytic DNA. In total, 46 ducts from 27 DL dates in 11 BRCA1 carriers, and 39 ducts from 28 dates in 14 BRCA2 carriers were studied.

LOH
DL samples from BRCA1 and BRCA2 carriers were screened with up to four 17q21 and four 13q21 microsatellite markers, respectively. Among BRCA1 carriers, 15 ducts were heterozygous, whereas 8 showed LOH. Among BRCA2 carriers, 17 ducts were heterozygous, whereas 7 showed LOH (Table 2 ). Notably, a further 23 (50%) samples from BRCA1 carriers and 15 (38%) from BRCA2 carriers had results that were discordant. The DNA yield was significantly higher in samples that were heterozygous compared with those that were discordant or that showed LOH (P = 0.003; Table 3 ).

Given the high frequency of discordant allelotype calls and the possibility that this might be due to the low yield of DNA, we examined the effect of DNA concentration on the reproducibility of the genotyping using limiting dilutions of normal lymphocyte DNA known to be heterozygous for the microsatellite marker D17S1322. Reactions containing DNA quantities in the range of 10 to 1,000 pg were amplified. The ratio of the alleles in PCR reactions containing 100, 200, 500, and 1,000 pg were constant, consistent with a heterozygous genotype. However, the relative ratios of the alleles in the reactions containing 50, 20, and 10 pg of DNA (16, 6, and 3 genome equivalents, respectively) were consistent with LOH, and furthermore, in repeat experiments, some samples showed loss of the opposite allele.

To determine whether the LOH observed was specific for breast malignancy, we assessed for LOH using two markers located on loci not renowned for frequent LOH in breast cancer. A total of 13 ducts from five subjects (three BRCA1 and two BRCA2 carriers) were used to assess for LOH in D1S123 and in 16 samples from four women (two BRCA1 and two BRCA2 carriers) for marker D12S385. LOH was seen in 8 of 13 (61.5%) ducts for D1S123 and for 8 of 16 (50%) samples from ducts for D12S385. Samples that showed LOH had a mean DNA yield of 22.4 ng for D1S123, and 14.8 ng for D12S385 compared with samples that were heterozygous, which had mean yields of 42.8 and 161.4 ng, respectively.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
DL is a minimally invasive, novel means of obtaining epithelial and other cell types directly from the potential site of breast tumor development, making it an attractive method for the detection of the phenotypic and genomic changes that are hypothesized to precede established breast cancer. Two previous studies suggest that the assessment for LOH is possible in cells obtained via DL (8, 9), however, neither study addressed possible confounding factors which may have adversely influenced the reliability of genotyping. DL fluid contains a variety of different cell types including foam cells, macrophages, lymphocytes, and only a small proportion of these may be of epithelial origin (4, 5), which are the most likely source of genetically aberrant premalignant cells. Because the detection of LOH requires relatively pure clonal cell populations, it is likely that even if the DL fluid contains premalignant cells harboring a region of LOH, it would be masked by the overwhelming number of normal cells. Furthermore, even if such a population of cells were present in a large proportion of the DL sample, the low DNA yield would compromise the reliability of the genotyping data. Allele dropout due to preferential allele amplification, which could simulate the appearance of LOH, is a well-recognized phenomenon when amplifying from low concentrations of DNA template (10).

Previous studies did not quantitate the DNA recovered from DL samples and were not able to assess the extent to which preferential allele amplification influenced the accuracy of genotyping (8, 9). In our study, we used a real-time quantitative PCR assay to accurately determine DNA yield in all DL samples. Despite using best practice DL and DNA extraction protocols, the median total DNA yield per duct was only 16.6 ng for BRCA1 carriers and 9 ng for BRCA2 carriers with 45% of samples yielding <10 ng. Consequently, most PCR reactions contained <16 genome equivalents (assuming 3 pg per haploid genome). In light of this information, the discordant genotyping observed for some samples is not surprising. The median DNA yield was lower for ducts that were scored as LOH or discordant than for ducts that were consistently heterozygous (P = 0.003), supporting the notion that DNA quantity rather than true genetic aberration was the principal cause of the LOH. Furthermore, LOH was observed at even higher frequencies on chromosome 1 (61%) and chromosome 12 (50%), two regions not commonly associated with LOH in breast disease. Our data suggests that much of the LOH observed is artifact rather than true LOH. Although it is unclear to what extent this was an issue with previous studies, it is interesting to note that in the study published by Isaacs et al., three of the four subjects whose samples showed LOH had inadequate epithelial cell numbers for cytologic evaluation (8), indicating that the DNA yield for these samples may have been very low.

Future Potential for Identifying Biomarkers Using DL
Other molecular approaches that circumvent the issues of low DNA yield and mixed cell populations, such as methylation analysis, may be a more appropriate use of DL effluent. Several studies have reported feasibility with the detection of 1 copy of methylated DNA in 1,000 unmethylated copies of genomic DNA (13-15) or in DL samples with as few as five epithelial cells (14).

Conclusions
The amount of amplifiable DNA obtained from DL specimens is highly variable and low DNA yield is likely to have a detrimental effect on the interpretation of LOH studies. Although we detected LOH, the strong correlation between low DNA yield, and contradictory LOH results in replicated experiments, suggest that much of the LOH detected may be PCR artifacts. In our experience, LOH analysis is an unreliable method for the investigation of early biomarkers of malignancy using DL specimens.

	Acknowledgments

 
We thank Professor Joseph Sambrook for his advice and for reviewing this manuscript, Dr. Geoff Lindeman and Associate Professor Judy Kirk for their assistance in sample collection, Dr. William Murray for cytology review, Dr. Juliana Di Iulio for data management, and Cytyc Corporation for assistance with provision of the FirstCyte ductal lavage catheters.

	Footnotes

 
Grant support: NHMRC program grant "Beyond BRCA1/2".

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Note: Dr. Yoland Antill is the current recipient of the Goodman Fielder National Breast Cancer Foundation Scholarship.

Kelly-Anne Phillips is the Cancer Council Victoria, Dr. John Colebatch Clinical Research Fellow.

All participants involved in this study have given informed written consent with the research protocol having been approved by the ethics committees of each participating site.

Received 12/26/05; revised 3/ 9/06; accepted 5/ 2/06.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Antoniou A, Pharoah PD, Narod S, et al. Average risks of breast and ovarian cancer associated with BRCA1 or BRCA2 mutations detected in case series unselected for family history: a combined analysis of 22 studies. Am J Hum Genet 2003;72:1117–30.[CrossRef][Medline]
2. Ford D, Easton DF, Stratton M, et al. Genetic heterogeneity and penetrance analysis of the BRCA1 and BRCA2 genes in breast cancer families. The Breast Cancer Linkage Consortium. Am J Hum Genet 1998;62:676–89.[CrossRef][Medline]
3. Brekelmans CT, Seynaeve C, Bartels CC, et al. Effectiveness of breast cancer surveillance in BRCA1/2 gene mutation carriers and women with high familial risk. J Clin Oncol 2001;19:924–30.[Abstract/Free Full Text]
4. Dooley WC, Ljung BM, Veronesi U, et al. Ductal lavage for detection of cellular atypia in women at high risk for breast cancer. J Natl Cancer Inst 2001;93:1624–32.[Abstract/Free Full Text]
5. King BL, Crisi GM, Tsai SC, Haffty BG, Phillips RF, Rimm DL. Immunocytochemical analysis of breast cells obtained by ductal lavage. Cancer 2002;96:244–9.[Medline]
6. Osorio A, de la Hoya M, Rodriguez-Lopez R, et al. Loss of heterozygosity analysis at the BRCA loci in tumor samples from patients with familial breast cancer. Int J Cancer 2002;99:305–9.[CrossRef][Medline]
7. Esteller M, Silva JM, Dominguez G, et al. Promoter hypermethylation and BRCA1 inactivation in sporadic breast and ovarian tumors. J Natl Cancer Inst 2000;92:564–9.[Abstract/Free Full Text]
8. Isaacs C, Cavalli LR, Cohen Y, et al. Detection of LOH, mitochondrial DNA. alterations in ductal lavage and nipple aspirate fluids from high-risk patients. Breast Cancer Res Treat 2004;84:99–105.[CrossRef][Medline]
9. Yonekura Y, Yamamoto D, Okugawa H, Tanaka K, Kamiyama Y. Loss of heterozygosity in ductal lavage for breast tumor and the contralateral breast. Oncol Rep 2005;13:739–43.[Medline]
10. Tomlinson IP, Lambros MB, Roylance RR. Loss of heterozygosity analysis: practically and conceptually flawed? Genes Chromosomes Cancer 2002;34:349–53.[CrossRef][Medline]
11. Mitchell G, Antill YC, Murray W, et al. Nipple aspiration and ductal lavage in women with a germline BRCA1 or BRCA2 mutation. Breast Cancer Res 2005;7:R1122–31.[Medline]
12. Campbell IG, Russell SE, Choong DY, et al. Mutation of the PIK3CA gene in ovarian and breast cancer. Cancer Res 2004;64:7678–81.[Abstract/Free Full Text]
13. Evron E, Dooley WC, Umbricht CB, et al. Detection of breast cancer cells in ductal lavage fluid by methylation-specific PCR. Lancet 2001;357:1335–6.[CrossRef][Medline]
14. Fackler MJ, McVeigh M, Mehrotra J, et al. Quantitative multiplex methylation-specific PCR assay for the detection of promoter hypermethylation in multiple genes in breast cancer. Cancer Res 2004;64:4442–52.[Abstract/Free Full Text]
15. Krassenstein R, Sauter E, Dulaimi E, et al. Detection of breast cancer in nipple aspirate fluid by CpG island hypermethylation. Clin Cancer Res 2004;10:28–32.[Abstract/Free Full Text]

This article has been cited by other articles:

				I. Locke, Z. Kote-Jarai, E. Bancroft, S. Bullock, S. Jugurnauth, P. Osin, A. Nerurkar, L. Izatt, G. Pichert, G. P. H. Gui, et al. Loss of Heterozygosity at the BRCA1 and BRCA2 Loci Detected in Ductal Lavage Fluid from BRCA Gene Mutation Carriers and Controls. Cancer Epidemiol. Biomarkers Prev., July 1, 2006; 15(7): 1399 - 1402. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Antill, Y. C. Articles by Campbell, I. G. Search for Related Content PubMed PubMed Citation Articles by Antill, Y. C. Articles by Campbell, I. G.