Amplification of TOP2A and HER-2 genes in breast cancers occurring in patients harbouring BRCA1 germline mutations

Abstract

BRCA1 associated tumours are found to express an oestrogen receptor negative “basal epithelial-like” phenotype. In contrast to ER negative tumours in general, such tumours rarely harbour amplification of the HER-2 gene. However, little is known about TOP2A gene amplification status in BRCA1-associated tumours. Such information may be of importance to therapy, as amplification of TOP2A has been associated with dose-dependent sensitivity to anthracycline therapy in breast cancer.

We examined 40 breast carcinomas from BRCA1 mutation carriers and 40 sporadic breast carcinomas matched for age, tumour diameter and histological grade for HER-2 and TOP2A amplification status using fluorescence in situ hybridisation (FISH). Co-amplification of TOP2A and HER-2 was found in four of the mutation carriers and in three of the controls. While six tumours in the control group harboured HER-2 amplifications with normal TOP2A, this occurred in three of the BRCA1 associated tumours only. In contrast, three of the BRCA1-associated tumours but none of the controls harboured TOP2A amplification despite normal HER-2 status. Our findings have potential therapeutic implications. HER-2 assessment is routinely used to select breast cancer patients for trastuzumab but also dose-intensive anthracycline therapy. Our data suggest that BRCA1-associated breast cancers also need to be tested for TOP2A amplification.

Women with germline mutations in BRCA1 have an increased risk of developing breast and ovarian cancer [1]. BRCA1 associated breast cancers have been reported to carry a poor prognosis [2–5], expressing a panel of biological parameters generally associated with poor outcome, including histological high grade, ER negativity, high proliferation indices (Ki-67,) and a “basal epithelial-like” phenotype [6–9].

While the poor prognosis emphasize the need for adjuvant therapy of BRCA1 associated tumours, we lack information whether the biology of these tumours may convey sensitivity to chemotherapeutic agents different from sporadic breast cancers.

Amplification of the HER-2 gene has been described in 18–35% of all primary breast cancers independent of stage [10], [11]. While HER-2 amplification or protein overexpression signals sensitivity to trastuzumab [12], it has also been associated with a dose-dependent sensitivity to anthracyline therapy, advocating higher doses in HER2 positive tumours [13]. Recent evidence suggests this may be due to co-amplification of TOP2A [14], [15], the gene coding for topoisomerase II alpha (Topo-IIα). TopoIIα is a major target for anthracyclines [16], and in vitro studies have shown topoisomerase IIα inhibitors, such as the anthracyclines, to be more effective in tumours with high topoisomerase IIα expression caused by TOP2A gene amplification [17–19].

While amplification of the TOP2A gene is found in 5 to 19% of sporadic breast cancers [14], [15], [18], [20], little is known about the incidence of TOP2A amplification in BRCA1 associated breast cancer.

The aim of the current study was to investigate the prevalence of alterations in TOP2A and HER-2 gene copy number in BRCA1 associated cancer using FISH.

Materials and methods

Forty patients with BRCA1 associated breast cancer diagnosed between 1994 and 2004 were identified from the archives of Section of Genetic Counselling at the Norwegian Radium Hospital, Oslo. Formalin-fixed, paraffin-embedded biopsy material from these patients was retrieved from the archives of Department of Pathology and Medical Genetics, St. Olav's Hospital, Trondheim University Hospital and the Department of Pathology, Stavanger University Hospital. Each BRCA1 case was matched for age, tumour diameter and histological grade with cases of sporadic breast cancers from the archives of St. Olav's Hospital.

The histological slides were reviewed to confirm the diagnosis and histological grading was done according to the guidelines recommended by the Norwegian Breast Cancer Group (NBCG) which are based on the Bloom-Richardson breast cancer grading system modified by Elston and Ellis [21].

ER status was based on routine assessment using the 1D5 (DakoCytomation) or 6F11 (Novocastra) antibodies.

FISH was done on sections cut at 4 µm from representative areas of the tumours. Locus-specific probes for the HER-2 and TOP2A genes were used together with a centromere 17 control (LSI Triple Probe Mix SpectrumAqua – SpectrumGreen – SpectrumOrange, Vysis Inc., Downers Grove, IL, USA). Absolute copy numbers and the relative copy number ratios were determined for each gene (ratio between mean number of signals for TOP2A or HER2 and the mean number of chromosome 17 centromere signals).

After pre-treatment, the slides were dehydrated (70%, 80%, and 100% ethanol, two minutes in each) and allowed to air-dry. An EasiSeal (Hybaid Ltd., Ashford, UK) frame was placed on the slide and 10 µl chromosome 17 specific centromere probe/TOP2A and HER-2 locus specific probe was applied to the sections which were then denatured at 72°C for 5 min and hybridised at 37°C overnight in a Perkin-Elmer In Situ Thermal Cycler (Applied Biosystems Division, CA, USA). Post-hybridisation washes were done (2 min in Stringent Wash Buffer 0.4×SSC/0,3% NP-40, pH 7.0 at 73°C followed by 1 min in Wash Buffer 2×SSC/0.1% NP-40, pH 7.0 at room temperature). The slides were then dehydrated (70%, 85% and 100% ethanol) and dried for 5 min at 60°C before 10–15 µl fluorescence mounting medium containing DAPI (Blue fluorescence, Vysis Inc., Downers Grove, IL, USA.) counter-stain was applied. After cover-slipping, the sections were examined at×400 magnification and photographed using Nikon E600 fluorescence microscope (Nikon, Tokyo, Japan) using Cytovision software (Applied Imaging, Newcastle upon Tyne, England).

Signal counting and ratio calculation

The numbers of CEP17, TOP2A and HER-2 signals in each of 60 nuclei were counted and an informative result was defined as an abnormal result for either chromosome 17, TOP2A or HER-2 in 60% or more of the nuclei. Press et al. have suggested that 20 cells are sufficient for the assessment of chromosome amplification [11]. We chose to assess at least 60 cells since the number recommended by the manufacturer for assessment of HER2-gene amplification is 20 to 40 nuclei, (DakoCytomation HER2 FISH pharm Dx, Interpretation Guide). The ratios between TOP2A and chromosome 17 and between HER-2 and chromosome 17 were calculated. The cut-off level for amplification was set at a gene: chromosome ratio of >2.0 as described in the literature for HER2 [22]. For deletions, ratios <1.0 were considered abnormal [22].

Hybridization efficiency

Only areas of invasive tumour where >90% of tumour cell nuclei showed successful hybridization for all three probes were assessed.

The study was approved by the Regional Committee for Medical Research Ethics and by the Ministry of Health and Care Services who gave their consent to the establishment of the biobank.

Results

Grade, tumour size and oestrogen receptor status for the BRCA1 associated cancers and the control tumours are depicted in Table I.

Table I.  Characteristics of cases and controls.

		pT1 gr.2	pT1 gr.3	pT2 gr.2	pT2 gr.3
BRCA1	ER +	3	4	1	0
n = 40	ER −	1	23	0	8
Controls	ER +	4	15	1	2
n = 40	ER −	0	12	0	6

Co-amplification of HER-2 and TOP2A was found in three of the mutation carriers and in three of the sporadic breast cancers (Figure 1a). In contrast, HER-2 amplification alone was found in three of the mutation carriers, compared to six among the sporadic breast cancers (Figure 1b). Importantly, TOP2A amplification with normal HER-2 status, a phenomenon rarely seen in sporadic breast cancers, was found in three of the mutation carriers, but in none of sporadic breast cancers. Possible deletion of one or both genes was seen in one of the mutation carriers and in two of the sporadic cancer cases. A summary of the findings for the mutation carriers and for the sporadic cancers is given in Table II.

Figure 1a.  FISH analysis of cancer cells showing HER-2 amplification without TOP2A amplification and four CEP 17. Red signals = TOP2A, green signals = HER-2, aqua signals = CEP 17.

Figure 1b.  FISH analysis of cancer cells showing HER-2 and TOP2A coamplification. There are three CEP 17 (aqua) and greater than ten signals for both HER-2 (green) and TOP2A (red) genes. Genes to chromosome ratio >3.

Table II.  HER-2/TOP2A findings.

		HER-2/TOP2A coamp	HER-2 amp/TOP2A norm	HER-2 norm/TOP2A amp	HER-2/TOP2A norm	HER-2 norm/TOP2A del	HER-2/TOP2A del
BRCA1	ER +	1	1	2	3	1	0
n = 40	ER −	2	2	1	27	0	0
Controls	ER +	0	3	0	13	1	1
n = 40	ER −	3	3	0	16	0	0

Discussion

In this study, we observed HER-2 amplification, three alone and three together with TOP2A amplification, among six out of 40 BRCA1 associated tumours. Among the forty sporadic cancers, nine showed HER-2 amplification, six without and three with co-amplification of TOP2A. These results are in concordance with other studies which have also shown that BRCA1 associated cancers, despite being ER negative, are associated with a lower level of HER-2 amplification and HER-2 protein overexpression compared to sporadic breast cancer [6], [23], consistent with “basal-like” tumour subtype [8], [9].

In contrast, we found six cases (6/40) of TOP2A amplification among the BRCA1 associated tumours, but only three (3/40) in the sporadic cancers. Thus, TOP2A gene amplification without HER-2 gene amplification was observed in three cases of BRCA1 associated cancers but in none of the sporadic cancers, consistent with previous observations that TOP2A gene amplification without amplification of the HER-2 gene is a rare event in sporadic breast cancer [15], [18], [20], [24].

Our findings may have clinical implications. HER-2 amplification signals dose-related sensitivity to anthracycline therapy [13], and patients whose tumours are HER-2 amplified are generally treated with anthracyclines at higher doses [13], [25]. Merging evidence suggest the association between HER-2 amplification and anthracycline sensitivity to be due to co-amplification of TOP2A [14], [15]. The results of Tanner et al. suggests that TOP2A amplification may be a predictor of efficacy of dose-escalated FEC in HER-2 amplified high-risk breast cancer [26].Thus, if our finding that BRCA1 associated tumours may harbour TOP2A amplification without HER-2 amplification is confirmed by others, this should advocate TOP2A measurements to be performed on a routine basis in tumours occurring in patients harbouring BRCA1 mutations.

Due to the fact that topoisomerase IIa is a target for anthracycline therapy [16], deletion of the gene could be a cause of anthracycline resistance. The possible TOP2A gene deletion observed in this study in one of the BRCA1 associated and in two of the sporadic cancers, does not suggest this to be a particular problem with respect to therapy of BRCA1 associated breast cancers. There is no consensus in the literature regarding cut-off levels for determination of gene deletion. Nuclear truncation in histological sections may lead to an unduly low gene: chromosome ratio [27]. In light of this we chose to state that gene deletion might be present in cases with ratios <1.0.

In conclusion, despite the small sample size, our study suggests HER-2 gene amplification to be infrequent among BRCA1 associated tumours. In contrast, TOP2A amplifications seem to occur more frequently in BRCA1 associated breast cancers than in sporadic breast cancers. The finding that TOP2A amplification may occur independent of HER-2 amplification, suggests that this parameter should be determined on a routine basis in BRCA1 associated tumours.

This work was supported by the Cancer Fund at St.Olav's Hospital, Trondheim University Hospital, Trondheim.