European Journal of Cancer (2014) xxx, xxx– xxx
Available at www.sciencedirect.com
ScienceDirect journal homepage: www.ejcancer.com
BRCA1 gene promoter methylation status in high-grade serous ovarian cancer patients – A study of the tumour Bank ovarian cancer (TOC) and ovarian cancer diagnosis consortium (OVCAD) I. Ruscito a,b,1, D. Dimitrova a,1, I. Vasconcelos a, K. Gellhaus c, T. Schwachula c, F. Bellati b, R. Zeillinger d, P. Benedetti-Panici b, I. Vergote e, S. Mahner f, D. Cacsire-Tong c, N. Concin g, S. Darb-Esfahani h, S. Lambrechts e, J. Sehouli a, S. Olek c, E.I. Braicu a,⇑ a
Department of Gynecology, European Competence Center for Ovarian Cancer, Campus Virchow Klinikum, Charite´ – Universita¨tsmedizin Berlin, Augustenburger Platz 1, D-13353 Berlin, Germany b Department of Gynecology, Obstetrics and Urology, Sapienza University of Rome, Rome, Italy c Ivana Tu¨rbachova Laboratory for Epigenetics, Epiontis GMBH, Berlin, Germany d Department of Obstetrics and Gynecology, Molecular Oncology Group, Comprehensive Cancer Center, Medical University of Vienna, Waehringer Guertel 18-20, A-1090 Vienna, Austria e Division of Gynaecological Oncology, Department of Obstetrics and Gynaecology, Universitaire Ziekenhuizen Leuven, Katholieke Universiteit Leuven, UZ Leuven, Herestraat 49, B-3000 Leuven, Belgium f Department of Gynecology and Gynecologic Oncology, University Medical Center Hamburg-Eppendorf, Martinistraße 52, D-20246 Hamburg, Germany g Department of Gynecology and Obstetrics, Innsbruck Medical University, Anichstrasse 35, 6020 Innsbruck, Austria h Institute of Pathology, Charite Medical University, Berlin, Campus Mitte, Germany Received 25 February 2014; received in revised form 2 May 2014; accepted 5 May 2014
KEYWORDS High grade serous ovarian cancer BRCA1 Methylation Gene promoter Prognosis
Abstract Background: Mutations in BRCA1/2 genes are involved in the pathogenesis of breast and ovarian cancer. Inactivation of these genes can also be mediated by hypermethylation of CpGs in the promoter regions. Aim of this study was to analyse the clinical impact of BRCA1 promoter gene methylation status in a homogenous cohort of high-grade serous ovarian cancer (HGSOC) patients. Methods: The cohort included 257 primary HGSOC patients treated by cytoreduction and platinum-based chemotherapy. DNA was extracted from fresh frozen tissue samples. BRCA1
⇑ Corresponding author: Tel.: +49 30 450664469; fax: +49 30 450564939. 1
E-mail address: [email protected] (E.I. Braicu). Equal contribution.
http://dx.doi.org/10.1016/j.ejca.2014.05.001 0959-8049/Ó 2014 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Ruscito I. et al., BRCA1 gene promoter methylation status in high-grade serous ovarian cancer patients – A study of the tumour Bank ovarian cancer (TOC) and ovarian cancer diagnosis consortium (OVCAD), Eur J Cancer (2014), http://dx.doi.org/10.1016/j.ejca.2014.05.001
2
I. Ruscito et al. / European Journal of Cancer xxx (2014) xxx–xxx
gene promoter methylation rate was assessed using polymerase chain reaction (PCR). Results: 14.8% of patients presented hypermethylation within a selected region of the BRCA1 promoter. The rate of hypermethylation was significantly higher in younger patients (20.8% hypermethylation in the age group 658 years versus 8.7% hypermethylation in the age group >58 years; p = 0.008). Optimal tumour debulking could be reached in 63% of patients, without significant differences in the extent of residual disease with respect to the methylation status. No impact of BRCA1 gene promoter methylation status on progression free- and overallsurvival rates was found. No significant differences within BRCA1 promoter methylation status between primary and metastatic tissue could be observed. These results on BRCA1 promoter methylation status were also confirmed in a subgroup of 107 patients found negative for BRCA1 exon 11 mutations. Conclusions: Our data suggest that BRCA1 methylation determines the earlier onset of HGSOC. Furthermore our study supports the idea that BRCAness is not only due to mutations but also to epigenetic changes in BRCA1 promoter gene. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction
2. Materials and methods
Ovarian cancer is the most lethal malignancy among gynaecological cancers, with over 140,000 deaths per year worldwide [1]. Most of the patients after receiving optimal tumour debulking and platinum based chemotherapy, will develop recurrence, platinum-resistance and will eventually die from the disease [2]. Recent studies showed that different histological subtypes of ovarian cancer are having different clinical behaviours as also different biomolecular features [3]. High grade serous ovarian cancer is the most encountered histological subtype being characterised by advanced International Federation of Gynecology and Obstetrics (FIGO) stage and decreased overall survival despite increased platinum-sensitivity [4]. The urgent need of new therapeutic targets has led researchers to concentrate the efforts in studying new molecular mechanisms which interfere with the integrity of the balance between oncogenes and tumour suppressor genes and therefore lead to personalised treatment of ovarian cancer patients. In particular, a specific family of tumour suppressor genes involved in the repair process of damaged DNA (BReast CAncer genes family, BRCA) has been extensively investigated in the last two decades. If mutated, BRCA1/BRCA2 are associated to an increased risk of breast and ovarian cancer. Furthermore, not only mutations will lead to a deficient DNA repair or oncosuppression, but low levels of ‘healthy’ unmutated gene may cause a functional deficit of the genes. Epigenetic changes might (down) regulate gene expression. In healthy individuals, the regulatory region of the full active BRCA gene is demethylated. Methylation process may cause a downregulation of the protein leading to gene dysfunction [5–12]. The aim of the present study was to analyse the clinical impact of BRCA1 promoter gene methylation status in HGSOC.
2.1. Sample collection We selected 257 consecutive patients with primary HGSOC subjected to surgical cytoreduction and platinum-based chemotherapy between 2000 and 2011. Further selection criteria were the availability of fresh frozen tumour tissue and a follow-up time of at least 3 years. 207 patients were obtained from the TOC (tumour ovarian cancer) Network. Another 50 patients were obtained from the OVCAD (ovarian cancer diagnosis) project. Five European gynaecologic cancer centres (Berlin, Hamburg, Innsbruck, Leuven and Vienna) prospectively enrolled epithelial ovarian cancer patients into this translational study. The main pathological, surgical and chemotherapy characteristics of the OVCAD patient cohort were published recently [13]. All patients gave their written informed consent before tissue samples were collected. Approval from each local ethics committee was obtained (EK207/ 2003, ML2524, HEK190504, EK366 and EK260). Tumour tissue samples were collected at the time of surgery, immediately frozen in liquid nitrogen within 15 min from the removal and then stored at 80 °C till further analysis. All ovarian cancer tissue samples included underwent histopathological assessment to verify histological subtype and high tissue quality. Only specimens presenting at least 50% of tumour area were included in the BRCA1 promoter methylation status analysis. The majority of tissue samples had approximately 80% of tumour area.
2.2. DNA extraction DNA was extracted from at least 25 mg of fresh frozen tissue specimens using the QIAGEN DNeasy tissue kit (Qiagen GmbH, Hilden, Germany) and following the
Please cite this article in press as: Ruscito I. et al., BRCA1 gene promoter methylation status in high-grade serous ovarian cancer patients – A study of the tumour Bank ovarian cancer (TOC) and ovarian cancer diagnosis consortium (OVCAD), Eur J Cancer (2014), http://dx.doi.org/10.1016/j.ejca.2014.05.001
I. Ruscito et al. / European Journal of Cancer xxx (2014) xxx–xxx
manufacturers’ instructions. Extracted DNA concentration was determined using the spectrophotometer. 2.3. Sodium bisulphite modification and methylationspecific polymerase chain reaction (PCR) The methylation status of CpG islands of BRCA1 promoter gene was assessed by bisulphite modification of DNA and methylation-specific polymerase chain reaction (PCR), as reported by Herman et al. [14]. DNA bisulphite modification was performed using QIAGEN EpiTect Bisulphite Kit (Qiagen GmbH, Hilden, Germany) that enables complete conversion of unmethylated cytosines to uracils and subsequent purification. DNA (2 lg) eluted in a final volume of 20 ll was mixed with 120 ll of the bisulphite reaction components (Bisulphite Mix and DNA protection buffer) and stored at room temperature. Bisulphite DNA conversion was performed using a thermal cycler and modified DNA was purified through repeated centrifugation steps with appropriate washing buffers, using the silica gel-membrane technology. Finally, 20 ll of purified DNA was stored at 20 °C, before PCR assessment. Three microlitres of bisulphite-modified DNA was subjected to PCR amplifications with appropriate primers (0.5 ll), probes (1.5 ll), and 5 ll of Probe Master Mix Solution (Probe Master by RocheÒ Applied Science, RocheÒ Diagnostics GmbH, Mannheim, Germany), in a final volume of 10 ll. The primers used for BRCA1-DNA amplification were as follows: forward, GTCCAAA AAATCTCAACG and reverse, TTTTTTGGTTTTCG TGGTAAC, hydrolysis probe: CACGCCGCGCAATC GC. The reaction was performed using the LightCyclerÒ 480 Real-Time PCR (RocheÒ Diagnostics GmbH, Mannheim, Germany). As normalisers bisulphite-modification of the Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) DNA sequence will be adopted [17]. 2.4. Sequencing of exon 11 of BRCA1 gene in germline DNA
3
A 25 ll-PCR-Reaction was performed using 1 ll leucocyte DNA with concentration 100 ng/ll; 1 ll primermix (forward und reverse primer) with concentration 10 pmol/ll for each primer-pair, 0.5 ll dNTP with concentration 10 pmol/ll, 0.2 ll (1UI) Hot Star Taq DNA Polymerase, 0.75 ll (3%) DMSO and 2.5 ll 10 PCR Buffer. PCR conditions consisted of an initial denaturating step at 95 °C for 15 min., 35 cycles of 95 °C for 20 s, 60 °C for 30 s and 72 °C for 1 min., an extension step at 72 °C for 10 min, and then held at 4 °C for 10 min. Primers being used are shown in Table 6. For separation and visualisation of amplification products agarose gel electrophoresis was used. 5 ll PCR product was mixed with 1 ll 6 Loading Dye (Fermentas) and loaded at 1.5% agarose gel. As reference 3 ll from 100 bp DNA Ready-to-use Leiter (Fermentas) was used. 1 TAE (Tris–acetate–EDTA) buffer was used both as a running buffer and in the preparation of agarose gel. The separation of the DNA-fragments was performed at 150 V for 10 min. 2.5. Clinical data Clinical data of patients included within the study were retrieved both from TOC and OVCAD clinical databases. After primary surgical cytoreduction, all patients underwent platinum-based adjuvant chemotherapy. A validated system, the IMO (Intraoperative Mapping of Ovarian Cancer) – tool [15], was used to prospectively document tumour pattern and residual tumour mass after cytoreductive surgery in all included patients. These data are obtained at the end of each surgery through an interview with the surgeon. Platinum response was assessed according to the interval between the end of platinum-based chemotherapy and relapse of the disease. Platinum resistant and responders were defined as relapse within the first six months or after six months following the last platinum-based chemotherapy cycle, respectively. 2.6. Follow-up
Our mutation analysis was focused on the central region of the BRCA gene represented by Exon 11. The rationale for choosing exon 11 was due to recently published results that showed a correlation between BRCA1 exon 11 mutations and the risk of breast and ovarian cancer [29,30]. Exon 11 covers 60% of the coding region of BRCA1 gene and includes important functional domains. With our screening strategy we expected to detect the great majority of BRCA1 Mutations responsible for ovarian cancer development in our population. A specific PCR with eight primer pairs, which cover completely exon 11 of the BRCA1 gene was designed and established with the support of Epiontis Berlin.
Following primary adjuvant chemotherapy, patients were examined every 3 months. A gynaecological and ultrasound examination, together with CA125 serum evaluation, was performed at each patient’s visit. CT and MRI scans were performed when a disease progression or recurrence was suggested by clinical examination or elevated CA125 serum levels. The isolated CA125 serum elevation was not considered as relapse. 2.7. Statistical analysis Statistical analysis was carried out using SPSS version 19.0 software. BRCA1 promoter gene methylation
Please cite this article in press as: Ruscito I. et al., BRCA1 gene promoter methylation status in high-grade serous ovarian cancer patients – A study of the tumour Bank ovarian cancer (TOC) and ovarian cancer diagnosis consortium (OVCAD), Eur J Cancer (2014), http://dx.doi.org/10.1016/j.ejca.2014.05.001
4
I. Ruscito et al. / European Journal of Cancer xxx (2014) xxx–xxx
status was correlated to biological and clinical variables (type of cancer tissue analysed, tumour grading, FIGO stage, presence or absence of ascites, postsurgical residual tumour and platinum response rates) by Fisher’s exact test. Statistical analysis of BRCA1 promoter gene methylation status between different groups was compared using the Mann–Whitney test. Progression-free survival (PFS) and overall survival (OS) were determined with the Kaplan–Meier analysis by the Log-Rank test. For all tests, a probability value p < 0.05 was considered statistically significant. 3. Results 3.1. Cohort 1: entire population Patients’ characteristics including age at diagnosis, FIGO stage, grading, presence or absence of ascites and postsurgical residual tumour are listed in Table 1. Median age at diagnosis was 58 years and over 94% of patients presented an advanced stage disease (FIGO Stage III and IV). All patients included were subjected to standard treatment with primary surgical cytoreduction followed by front line platinum-based chemotherapy. Optimal tumour debulking (no evidence of macroscopic residual tumour) could be reached in 63% of cases. Table 2 shows the different sites of sample collection for the assessment of BRCA1 promoter gene methylation status. Tumour tissue analysis defined a heterogeneous Table 1 Patients characteristics. Parameters
Number of patients (%)
Age at first diagnosis/median (range) Grading GII GIII
58 years (27–93 years) 63 (24.5%) 194 (75.5%)
Table 2 Origin of analysed cancer tissue. Origin of analysed tissue
Number
Ovary Peritoneum Colon Diaphragm Uterus Bowel Bladder/ureter Mesenterium Pelvic wall Douglas Abdominal wall Spleen Sacrum Vaginal cuff Bursa omentalis/pancreas
204 23 7 4 4 3 2 2 2 1 1 1 1 1 1
population with a wide range of BRCA1 promoter gene methylation rate (Table 3). We identified 38 patients (14.8%) whose tumour specimens presented at least 5% of methylation in BRCA1 promoter gene sequence. In this subgroup of patients (defined as Hypermethylated Group) the median age at diagnosis was 54 years, whereas in the Hypomethylated Group (methylation in BRCA1 promoter gene sequence 500 ml
52 (20.2%) 108 (42%) 97 (37.8%)
Residual tumour mass after surgery 0 >0–61 cm >1–P2 cm >2 cm
162 (63%) 73 (28.4%) 11 (4.3%) 11 (4.3%)
P5– < 10% 10–20% 21–30% 31–40% 41–50% 51–60% 61–70% 71–80% 81–100%
BRCA1 promoter methylation status Methylated (P5% methylation of BRCA1 gene promoter) Unmethylated (58 years
16 (29%) 3 (6%)
39 (71%) 49 (94%)
0.002
Origin of cancer tissue Ovary Other
16 (17%) 3 (25%)
79 (83%) 9 (75%)
0.44
Histological grading GII GIII
2 (8%) 17 (20%)
22 (92%) 66 (80%)
0.23
FIGO stage I II III IV
– – 16 (19%) 3 (20%)
5 2 69 (81%) 12 (80%)
0.58 0.46 0.75 0.73
Volume of ascites 6500 ml >500 ml
8 (12.7%) 10 (22.7%)
55 (87.3%) 34 (77.3%)
0.197
55 (81%) 33 (85%)
0.79
86 (81.9%) 2 (100%)
>.05
Residual tumour after surgery (RT) No residual tumour mass 13 (19%) Macroscopical residual 6 (15%) tumour mass Response to platinum based chemotherapy Responders 19 (18.1%) Non-responders 0
Table 6 Primers used for BRCA1 exon 11 sequencing. Primer
Sequence
11.1f 11.1r 11.2f 11.2r 11.3f 11.3r 11.4f 11.4r 11.5f 11.5r 11.6f 11.6r 11.7f 11.7r 11.8f 11.8r
50 -ATATAGCCAGTTGGTTGATTTCC-30 50 -GGAACATCTTCAGTATCTCTAGG-30 50 -GGTAGATCTGAATGCTGATCCC-30 50 -AGGATGAAGGCCTGATGTAGG-30 50 -TAGGAGCATTTGTTACTGAGCC-30 50 -TTCTGCTGTGCCTGACTGGC-30 50 -CCCACCTAATTGTACTGAATTGC-30 50 -ATGCTGCACACTGACTCACAC-30 50 -GGTACTGATTATGGCACTCAGG-30 50 -TTCGTTGCCTCTGAACTGAGATG-30 50 -AAGCCAGTTGATAATGCCAAATG-30 50 -ATTAACAGTCTGAACTACTTCTTC-30 50 -TTTGCAACCTGAGGTCTATAAAC-30 50 -GGTGCTATGCCTAGTAGACTG-30 50 -CTTATCTAGTGAGGATGAAGAGC-30 50 -CACCTTAGGAGGAACATGTTTA-3
PFS and OS were 24 and 56 months, respectively, with no significant difference registered between both groups (Fig. 2a and b). 4. Discussion Ovarian cancer is globally recognised as the most lethal malignancy of the female genital tract [1] and approximately 15% of cases are characterised by mutations in BRCA1/2 genes DNA sequence. Recently, a large pooled analysis of 26 observational studies on ovarian cancer patients’ survival, including 1213 epithelial ovarian cancer (EOC) patients, revealed that having a germline mutation in BRCA1/2 genes is associated with an improved 5-year OS, especially in BRCA2 mutated patients group [16]. The remaining 85% of ovarian cancer cases, with no mutations in BRCA1/2 genes DNA sequence, belongs to the so called ‘BRCA-negative’ tumours. It has been recently observed that up to 50% of women with high-grade serous epithelial ovarian cancer show functional loss of proteins involved in the homologous recombination pathway of DNA repair and, as a consequence, their tumours phenotypically behave like BRCA1/2 mutant cancers, even in the absence of a BRCA1/2 mutation. This phenomenon has been recently termed ‘BRCAness’ and it is characterised by DNA homologous recombination deficiency [18–20]. Another potential mechanism through which the cell can lose the homologous recombination pathway of DNA repair, thereby resembling the behaviour of a BRCA-mutated cell, is an epigenetic inactivation of the BRCA gene without alterations of its DNA sequence. This epigenetic phenomenon is characterised by the methylation of CpG islands in the BRCA1 gene promoter [21], which allows the binding between CpG islands and specific protein of the Histone Deacetylase Complex (HDAC), thus permitting the transition from eucromatinic to heterocromatinic status of the DNA
Please cite this article in press as: Ruscito I. et al., BRCA1 gene promoter methylation status in high-grade serous ovarian cancer patients – A study of the tumour Bank ovarian cancer (TOC) and ovarian cancer diagnosis consortium (OVCAD), Eur J Cancer (2014), http://dx.doi.org/10.1016/j.ejca.2014.05.001
I. Ruscito et al. / European Journal of Cancer xxx (2014) xxx–xxx
7
Fig. 2. Cohort 2: survival rates depending on BRCA1 promoter methylation status.
sequence. This event leads to the impossibility of BRCA1 gene transcription and to its consequent inactivation, thus increasing risk of DNA damage susceptibility and cancerogenesis. Previous studies in heterogenous ovarian cancer cohorts have reported a broad variability in the rates of BRCA1 promoter methylation, ranging from 5% to 40% [4–11]. Several authors agree in assigning this variability to the different regions analysed and to the different techniques employed in assessing BRCA1 promoter methylation status (e.g. Methylation-Specific PCR, MSP; Quantitative Multiplex Methylation-Specific PCR, QM.MSP; Methylation-Specific Multiplex Ligation-dependent Probe Amplification, MS-MLPA) [22,23]. The recent results of The Cancer Genome Atlas (TCGA) project confirmed germline and somatic mutations of the breast cancer susceptibility genes 1 and 2 (ca. 25%) and BRCA1 gene promoter methylation (11.5%) as second most frequent genomic and epigenomic alterations in HGSOC tumours, following p53 mutations [3].Within the TCGA study, BRCA1 and 2 mutations are linked with a better prognosis explained by a higher sensitivity to DNA-damaging agents, such as platinum-compounds. The aim of our study was to assess the BRCA1 gene promoter methylation status in primary high-grade serous ovarian cancer (HGSOC) tissue and to evaluate its clinical impact. The major strength points of our study are: the accurate patient and sample selection, which lead to the definition of a homogeneous ovarian cancer population. In this study, the incidence of BRCA1 promoter hypermethylation (14.8%) is in accordance with previous studies employing the same assessment technique [8]. The limitation of our study is the lack of information about somatic and genomic mutation of BRCA1 and BRCA2 genes. In 118 patients we had available data regarding the mutation status in BRCA1 exon 11. Exon 11 represents 60% of the BRCA1 gene. From these patients 11 of them presented mutations by gene sequencing. We performed
the same analysis in the subgroup of BRCA1 negative patients. The results were similar as in the whole collective of patients. Except age at first diagnosis (p = 0.022) no other significant correlation with clinical factors or with OS and PFS have been found. The cut-off of 5% for defining methylation status represents the lowest quintile. Our 5% methylation cut-off is in accordance with data from one recent publication by Hansmann et al. [31], who used a 6% cut-off value for BRCA1 methylation status. The authors considered this cut-off value optimal, due to the fact that an increased methylation level can be due to an increased rate of single aberrant methylated CpG sites in overall hypomethylated alleles. They considered it plausible to assume that single CpG errors without functional implications account for most promoter methylation values
Available at www.sciencedirect.com
ScienceDirect journal homepage: www.ejcancer.com
BRCA1 gene promoter methylation status in high-grade serous ovarian cancer patients – A study of the tumour Bank ovarian cancer (TOC) and ovarian cancer diagnosis consortium (OVCAD) I. Ruscito a,b,1, D. Dimitrova a,1, I. Vasconcelos a, K. Gellhaus c, T. Schwachula c, F. Bellati b, R. Zeillinger d, P. Benedetti-Panici b, I. Vergote e, S. Mahner f, D. Cacsire-Tong c, N. Concin g, S. Darb-Esfahani h, S. Lambrechts e, J. Sehouli a, S. Olek c, E.I. Braicu a,⇑ a
Department of Gynecology, European Competence Center for Ovarian Cancer, Campus Virchow Klinikum, Charite´ – Universita¨tsmedizin Berlin, Augustenburger Platz 1, D-13353 Berlin, Germany b Department of Gynecology, Obstetrics and Urology, Sapienza University of Rome, Rome, Italy c Ivana Tu¨rbachova Laboratory for Epigenetics, Epiontis GMBH, Berlin, Germany d Department of Obstetrics and Gynecology, Molecular Oncology Group, Comprehensive Cancer Center, Medical University of Vienna, Waehringer Guertel 18-20, A-1090 Vienna, Austria e Division of Gynaecological Oncology, Department of Obstetrics and Gynaecology, Universitaire Ziekenhuizen Leuven, Katholieke Universiteit Leuven, UZ Leuven, Herestraat 49, B-3000 Leuven, Belgium f Department of Gynecology and Gynecologic Oncology, University Medical Center Hamburg-Eppendorf, Martinistraße 52, D-20246 Hamburg, Germany g Department of Gynecology and Obstetrics, Innsbruck Medical University, Anichstrasse 35, 6020 Innsbruck, Austria h Institute of Pathology, Charite Medical University, Berlin, Campus Mitte, Germany Received 25 February 2014; received in revised form 2 May 2014; accepted 5 May 2014
KEYWORDS High grade serous ovarian cancer BRCA1 Methylation Gene promoter Prognosis
Abstract Background: Mutations in BRCA1/2 genes are involved in the pathogenesis of breast and ovarian cancer. Inactivation of these genes can also be mediated by hypermethylation of CpGs in the promoter regions. Aim of this study was to analyse the clinical impact of BRCA1 promoter gene methylation status in a homogenous cohort of high-grade serous ovarian cancer (HGSOC) patients. Methods: The cohort included 257 primary HGSOC patients treated by cytoreduction and platinum-based chemotherapy. DNA was extracted from fresh frozen tissue samples. BRCA1
⇑ Corresponding author: Tel.: +49 30 450664469; fax: +49 30 450564939. 1
E-mail address: [email protected] (E.I. Braicu). Equal contribution.
http://dx.doi.org/10.1016/j.ejca.2014.05.001 0959-8049/Ó 2014 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Ruscito I. et al., BRCA1 gene promoter methylation status in high-grade serous ovarian cancer patients – A study of the tumour Bank ovarian cancer (TOC) and ovarian cancer diagnosis consortium (OVCAD), Eur J Cancer (2014), http://dx.doi.org/10.1016/j.ejca.2014.05.001
2
I. Ruscito et al. / European Journal of Cancer xxx (2014) xxx–xxx
gene promoter methylation rate was assessed using polymerase chain reaction (PCR). Results: 14.8% of patients presented hypermethylation within a selected region of the BRCA1 promoter. The rate of hypermethylation was significantly higher in younger patients (20.8% hypermethylation in the age group 658 years versus 8.7% hypermethylation in the age group >58 years; p = 0.008). Optimal tumour debulking could be reached in 63% of patients, without significant differences in the extent of residual disease with respect to the methylation status. No impact of BRCA1 gene promoter methylation status on progression free- and overallsurvival rates was found. No significant differences within BRCA1 promoter methylation status between primary and metastatic tissue could be observed. These results on BRCA1 promoter methylation status were also confirmed in a subgroup of 107 patients found negative for BRCA1 exon 11 mutations. Conclusions: Our data suggest that BRCA1 methylation determines the earlier onset of HGSOC. Furthermore our study supports the idea that BRCAness is not only due to mutations but also to epigenetic changes in BRCA1 promoter gene. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction
2. Materials and methods
Ovarian cancer is the most lethal malignancy among gynaecological cancers, with over 140,000 deaths per year worldwide [1]. Most of the patients after receiving optimal tumour debulking and platinum based chemotherapy, will develop recurrence, platinum-resistance and will eventually die from the disease [2]. Recent studies showed that different histological subtypes of ovarian cancer are having different clinical behaviours as also different biomolecular features [3]. High grade serous ovarian cancer is the most encountered histological subtype being characterised by advanced International Federation of Gynecology and Obstetrics (FIGO) stage and decreased overall survival despite increased platinum-sensitivity [4]. The urgent need of new therapeutic targets has led researchers to concentrate the efforts in studying new molecular mechanisms which interfere with the integrity of the balance between oncogenes and tumour suppressor genes and therefore lead to personalised treatment of ovarian cancer patients. In particular, a specific family of tumour suppressor genes involved in the repair process of damaged DNA (BReast CAncer genes family, BRCA) has been extensively investigated in the last two decades. If mutated, BRCA1/BRCA2 are associated to an increased risk of breast and ovarian cancer. Furthermore, not only mutations will lead to a deficient DNA repair or oncosuppression, but low levels of ‘healthy’ unmutated gene may cause a functional deficit of the genes. Epigenetic changes might (down) regulate gene expression. In healthy individuals, the regulatory region of the full active BRCA gene is demethylated. Methylation process may cause a downregulation of the protein leading to gene dysfunction [5–12]. The aim of the present study was to analyse the clinical impact of BRCA1 promoter gene methylation status in HGSOC.
2.1. Sample collection We selected 257 consecutive patients with primary HGSOC subjected to surgical cytoreduction and platinum-based chemotherapy between 2000 and 2011. Further selection criteria were the availability of fresh frozen tumour tissue and a follow-up time of at least 3 years. 207 patients were obtained from the TOC (tumour ovarian cancer) Network. Another 50 patients were obtained from the OVCAD (ovarian cancer diagnosis) project. Five European gynaecologic cancer centres (Berlin, Hamburg, Innsbruck, Leuven and Vienna) prospectively enrolled epithelial ovarian cancer patients into this translational study. The main pathological, surgical and chemotherapy characteristics of the OVCAD patient cohort were published recently [13]. All patients gave their written informed consent before tissue samples were collected. Approval from each local ethics committee was obtained (EK207/ 2003, ML2524, HEK190504, EK366 and EK260). Tumour tissue samples were collected at the time of surgery, immediately frozen in liquid nitrogen within 15 min from the removal and then stored at 80 °C till further analysis. All ovarian cancer tissue samples included underwent histopathological assessment to verify histological subtype and high tissue quality. Only specimens presenting at least 50% of tumour area were included in the BRCA1 promoter methylation status analysis. The majority of tissue samples had approximately 80% of tumour area.
2.2. DNA extraction DNA was extracted from at least 25 mg of fresh frozen tissue specimens using the QIAGEN DNeasy tissue kit (Qiagen GmbH, Hilden, Germany) and following the
Please cite this article in press as: Ruscito I. et al., BRCA1 gene promoter methylation status in high-grade serous ovarian cancer patients – A study of the tumour Bank ovarian cancer (TOC) and ovarian cancer diagnosis consortium (OVCAD), Eur J Cancer (2014), http://dx.doi.org/10.1016/j.ejca.2014.05.001
I. Ruscito et al. / European Journal of Cancer xxx (2014) xxx–xxx
manufacturers’ instructions. Extracted DNA concentration was determined using the spectrophotometer. 2.3. Sodium bisulphite modification and methylationspecific polymerase chain reaction (PCR) The methylation status of CpG islands of BRCA1 promoter gene was assessed by bisulphite modification of DNA and methylation-specific polymerase chain reaction (PCR), as reported by Herman et al. [14]. DNA bisulphite modification was performed using QIAGEN EpiTect Bisulphite Kit (Qiagen GmbH, Hilden, Germany) that enables complete conversion of unmethylated cytosines to uracils and subsequent purification. DNA (2 lg) eluted in a final volume of 20 ll was mixed with 120 ll of the bisulphite reaction components (Bisulphite Mix and DNA protection buffer) and stored at room temperature. Bisulphite DNA conversion was performed using a thermal cycler and modified DNA was purified through repeated centrifugation steps with appropriate washing buffers, using the silica gel-membrane technology. Finally, 20 ll of purified DNA was stored at 20 °C, before PCR assessment. Three microlitres of bisulphite-modified DNA was subjected to PCR amplifications with appropriate primers (0.5 ll), probes (1.5 ll), and 5 ll of Probe Master Mix Solution (Probe Master by RocheÒ Applied Science, RocheÒ Diagnostics GmbH, Mannheim, Germany), in a final volume of 10 ll. The primers used for BRCA1-DNA amplification were as follows: forward, GTCCAAA AAATCTCAACG and reverse, TTTTTTGGTTTTCG TGGTAAC, hydrolysis probe: CACGCCGCGCAATC GC. The reaction was performed using the LightCyclerÒ 480 Real-Time PCR (RocheÒ Diagnostics GmbH, Mannheim, Germany). As normalisers bisulphite-modification of the Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) DNA sequence will be adopted [17]. 2.4. Sequencing of exon 11 of BRCA1 gene in germline DNA
3
A 25 ll-PCR-Reaction was performed using 1 ll leucocyte DNA with concentration 100 ng/ll; 1 ll primermix (forward und reverse primer) with concentration 10 pmol/ll for each primer-pair, 0.5 ll dNTP with concentration 10 pmol/ll, 0.2 ll (1UI) Hot Star Taq DNA Polymerase, 0.75 ll (3%) DMSO and 2.5 ll 10 PCR Buffer. PCR conditions consisted of an initial denaturating step at 95 °C for 15 min., 35 cycles of 95 °C for 20 s, 60 °C for 30 s and 72 °C for 1 min., an extension step at 72 °C for 10 min, and then held at 4 °C for 10 min. Primers being used are shown in Table 6. For separation and visualisation of amplification products agarose gel electrophoresis was used. 5 ll PCR product was mixed with 1 ll 6 Loading Dye (Fermentas) and loaded at 1.5% agarose gel. As reference 3 ll from 100 bp DNA Ready-to-use Leiter (Fermentas) was used. 1 TAE (Tris–acetate–EDTA) buffer was used both as a running buffer and in the preparation of agarose gel. The separation of the DNA-fragments was performed at 150 V for 10 min. 2.5. Clinical data Clinical data of patients included within the study were retrieved both from TOC and OVCAD clinical databases. After primary surgical cytoreduction, all patients underwent platinum-based adjuvant chemotherapy. A validated system, the IMO (Intraoperative Mapping of Ovarian Cancer) – tool [15], was used to prospectively document tumour pattern and residual tumour mass after cytoreductive surgery in all included patients. These data are obtained at the end of each surgery through an interview with the surgeon. Platinum response was assessed according to the interval between the end of platinum-based chemotherapy and relapse of the disease. Platinum resistant and responders were defined as relapse within the first six months or after six months following the last platinum-based chemotherapy cycle, respectively. 2.6. Follow-up
Our mutation analysis was focused on the central region of the BRCA gene represented by Exon 11. The rationale for choosing exon 11 was due to recently published results that showed a correlation between BRCA1 exon 11 mutations and the risk of breast and ovarian cancer [29,30]. Exon 11 covers 60% of the coding region of BRCA1 gene and includes important functional domains. With our screening strategy we expected to detect the great majority of BRCA1 Mutations responsible for ovarian cancer development in our population. A specific PCR with eight primer pairs, which cover completely exon 11 of the BRCA1 gene was designed and established with the support of Epiontis Berlin.
Following primary adjuvant chemotherapy, patients were examined every 3 months. A gynaecological and ultrasound examination, together with CA125 serum evaluation, was performed at each patient’s visit. CT and MRI scans were performed when a disease progression or recurrence was suggested by clinical examination or elevated CA125 serum levels. The isolated CA125 serum elevation was not considered as relapse. 2.7. Statistical analysis Statistical analysis was carried out using SPSS version 19.0 software. BRCA1 promoter gene methylation
Please cite this article in press as: Ruscito I. et al., BRCA1 gene promoter methylation status in high-grade serous ovarian cancer patients – A study of the tumour Bank ovarian cancer (TOC) and ovarian cancer diagnosis consortium (OVCAD), Eur J Cancer (2014), http://dx.doi.org/10.1016/j.ejca.2014.05.001
4
I. Ruscito et al. / European Journal of Cancer xxx (2014) xxx–xxx
status was correlated to biological and clinical variables (type of cancer tissue analysed, tumour grading, FIGO stage, presence or absence of ascites, postsurgical residual tumour and platinum response rates) by Fisher’s exact test. Statistical analysis of BRCA1 promoter gene methylation status between different groups was compared using the Mann–Whitney test. Progression-free survival (PFS) and overall survival (OS) were determined with the Kaplan–Meier analysis by the Log-Rank test. For all tests, a probability value p < 0.05 was considered statistically significant. 3. Results 3.1. Cohort 1: entire population Patients’ characteristics including age at diagnosis, FIGO stage, grading, presence or absence of ascites and postsurgical residual tumour are listed in Table 1. Median age at diagnosis was 58 years and over 94% of patients presented an advanced stage disease (FIGO Stage III and IV). All patients included were subjected to standard treatment with primary surgical cytoreduction followed by front line platinum-based chemotherapy. Optimal tumour debulking (no evidence of macroscopic residual tumour) could be reached in 63% of cases. Table 2 shows the different sites of sample collection for the assessment of BRCA1 promoter gene methylation status. Tumour tissue analysis defined a heterogeneous Table 1 Patients characteristics. Parameters
Number of patients (%)
Age at first diagnosis/median (range) Grading GII GIII
58 years (27–93 years) 63 (24.5%) 194 (75.5%)
Table 2 Origin of analysed cancer tissue. Origin of analysed tissue
Number
Ovary Peritoneum Colon Diaphragm Uterus Bowel Bladder/ureter Mesenterium Pelvic wall Douglas Abdominal wall Spleen Sacrum Vaginal cuff Bursa omentalis/pancreas
204 23 7 4 4 3 2 2 2 1 1 1 1 1 1
population with a wide range of BRCA1 promoter gene methylation rate (Table 3). We identified 38 patients (14.8%) whose tumour specimens presented at least 5% of methylation in BRCA1 promoter gene sequence. In this subgroup of patients (defined as Hypermethylated Group) the median age at diagnosis was 54 years, whereas in the Hypomethylated Group (methylation in BRCA1 promoter gene sequence 500 ml
52 (20.2%) 108 (42%) 97 (37.8%)
Residual tumour mass after surgery 0 >0–61 cm >1–P2 cm >2 cm
162 (63%) 73 (28.4%) 11 (4.3%) 11 (4.3%)
P5– < 10% 10–20% 21–30% 31–40% 41–50% 51–60% 61–70% 71–80% 81–100%
BRCA1 promoter methylation status Methylated (P5% methylation of BRCA1 gene promoter) Unmethylated (58 years
16 (29%) 3 (6%)
39 (71%) 49 (94%)
0.002
Origin of cancer tissue Ovary Other
16 (17%) 3 (25%)
79 (83%) 9 (75%)
0.44
Histological grading GII GIII
2 (8%) 17 (20%)
22 (92%) 66 (80%)
0.23
FIGO stage I II III IV
– – 16 (19%) 3 (20%)
5 2 69 (81%) 12 (80%)
0.58 0.46 0.75 0.73
Volume of ascites 6500 ml >500 ml
8 (12.7%) 10 (22.7%)
55 (87.3%) 34 (77.3%)
0.197
55 (81%) 33 (85%)
0.79
86 (81.9%) 2 (100%)
>.05
Residual tumour after surgery (RT) No residual tumour mass 13 (19%) Macroscopical residual 6 (15%) tumour mass Response to platinum based chemotherapy Responders 19 (18.1%) Non-responders 0
Table 6 Primers used for BRCA1 exon 11 sequencing. Primer
Sequence
11.1f 11.1r 11.2f 11.2r 11.3f 11.3r 11.4f 11.4r 11.5f 11.5r 11.6f 11.6r 11.7f 11.7r 11.8f 11.8r
50 -ATATAGCCAGTTGGTTGATTTCC-30 50 -GGAACATCTTCAGTATCTCTAGG-30 50 -GGTAGATCTGAATGCTGATCCC-30 50 -AGGATGAAGGCCTGATGTAGG-30 50 -TAGGAGCATTTGTTACTGAGCC-30 50 -TTCTGCTGTGCCTGACTGGC-30 50 -CCCACCTAATTGTACTGAATTGC-30 50 -ATGCTGCACACTGACTCACAC-30 50 -GGTACTGATTATGGCACTCAGG-30 50 -TTCGTTGCCTCTGAACTGAGATG-30 50 -AAGCCAGTTGATAATGCCAAATG-30 50 -ATTAACAGTCTGAACTACTTCTTC-30 50 -TTTGCAACCTGAGGTCTATAAAC-30 50 -GGTGCTATGCCTAGTAGACTG-30 50 -CTTATCTAGTGAGGATGAAGAGC-30 50 -CACCTTAGGAGGAACATGTTTA-3
PFS and OS were 24 and 56 months, respectively, with no significant difference registered between both groups (Fig. 2a and b). 4. Discussion Ovarian cancer is globally recognised as the most lethal malignancy of the female genital tract [1] and approximately 15% of cases are characterised by mutations in BRCA1/2 genes DNA sequence. Recently, a large pooled analysis of 26 observational studies on ovarian cancer patients’ survival, including 1213 epithelial ovarian cancer (EOC) patients, revealed that having a germline mutation in BRCA1/2 genes is associated with an improved 5-year OS, especially in BRCA2 mutated patients group [16]. The remaining 85% of ovarian cancer cases, with no mutations in BRCA1/2 genes DNA sequence, belongs to the so called ‘BRCA-negative’ tumours. It has been recently observed that up to 50% of women with high-grade serous epithelial ovarian cancer show functional loss of proteins involved in the homologous recombination pathway of DNA repair and, as a consequence, their tumours phenotypically behave like BRCA1/2 mutant cancers, even in the absence of a BRCA1/2 mutation. This phenomenon has been recently termed ‘BRCAness’ and it is characterised by DNA homologous recombination deficiency [18–20]. Another potential mechanism through which the cell can lose the homologous recombination pathway of DNA repair, thereby resembling the behaviour of a BRCA-mutated cell, is an epigenetic inactivation of the BRCA gene without alterations of its DNA sequence. This epigenetic phenomenon is characterised by the methylation of CpG islands in the BRCA1 gene promoter [21], which allows the binding between CpG islands and specific protein of the Histone Deacetylase Complex (HDAC), thus permitting the transition from eucromatinic to heterocromatinic status of the DNA
Please cite this article in press as: Ruscito I. et al., BRCA1 gene promoter methylation status in high-grade serous ovarian cancer patients – A study of the tumour Bank ovarian cancer (TOC) and ovarian cancer diagnosis consortium (OVCAD), Eur J Cancer (2014), http://dx.doi.org/10.1016/j.ejca.2014.05.001
I. Ruscito et al. / European Journal of Cancer xxx (2014) xxx–xxx
7
Fig. 2. Cohort 2: survival rates depending on BRCA1 promoter methylation status.
sequence. This event leads to the impossibility of BRCA1 gene transcription and to its consequent inactivation, thus increasing risk of DNA damage susceptibility and cancerogenesis. Previous studies in heterogenous ovarian cancer cohorts have reported a broad variability in the rates of BRCA1 promoter methylation, ranging from 5% to 40% [4–11]. Several authors agree in assigning this variability to the different regions analysed and to the different techniques employed in assessing BRCA1 promoter methylation status (e.g. Methylation-Specific PCR, MSP; Quantitative Multiplex Methylation-Specific PCR, QM.MSP; Methylation-Specific Multiplex Ligation-dependent Probe Amplification, MS-MLPA) [22,23]. The recent results of The Cancer Genome Atlas (TCGA) project confirmed germline and somatic mutations of the breast cancer susceptibility genes 1 and 2 (ca. 25%) and BRCA1 gene promoter methylation (11.5%) as second most frequent genomic and epigenomic alterations in HGSOC tumours, following p53 mutations [3].Within the TCGA study, BRCA1 and 2 mutations are linked with a better prognosis explained by a higher sensitivity to DNA-damaging agents, such as platinum-compounds. The aim of our study was to assess the BRCA1 gene promoter methylation status in primary high-grade serous ovarian cancer (HGSOC) tissue and to evaluate its clinical impact. The major strength points of our study are: the accurate patient and sample selection, which lead to the definition of a homogeneous ovarian cancer population. In this study, the incidence of BRCA1 promoter hypermethylation (14.8%) is in accordance with previous studies employing the same assessment technique [8]. The limitation of our study is the lack of information about somatic and genomic mutation of BRCA1 and BRCA2 genes. In 118 patients we had available data regarding the mutation status in BRCA1 exon 11. Exon 11 represents 60% of the BRCA1 gene. From these patients 11 of them presented mutations by gene sequencing. We performed
the same analysis in the subgroup of BRCA1 negative patients. The results were similar as in the whole collective of patients. Except age at first diagnosis (p = 0.022) no other significant correlation with clinical factors or with OS and PFS have been found. The cut-off of 5% for defining methylation status represents the lowest quintile. Our 5% methylation cut-off is in accordance with data from one recent publication by Hansmann et al. [31], who used a 6% cut-off value for BRCA1 methylation status. The authors considered this cut-off value optimal, due to the fact that an increased methylation level can be due to an increased rate of single aberrant methylated CpG sites in overall hypomethylated alleles. They considered it plausible to assume that single CpG errors without functional implications account for most promoter methylation values
