- Email: [email protected]
Validation of methylation markers for diagnosis of oral cavity cancer
European Journal of Cancer (2015) xxx, xxx– xxx
Available at www.sciencedirect.com
ScienceDirect journal homepage: www.ejcancer.com
Validation of methylation markers for diagnosis of oral cavity cancer L.M.R.B. Arantes a,b, A.C. de Carvalho b, M.E. Melendez b, C.C. Centrone c, J.F. Go´is-Filho d, T.N. Toporcov e, D.N. Caly f, E.H. Tajara a, E.M. Goloni-Bertollo a, A.L. Carvalho b,⇑, GENCAPO g a
Department of Molecular Biology, Sa˜o Jose´ do Rio Preto Medical School (FAMERP), Sa˜o Jose´ do Rio Preto, SP, Brazil Molecular Oncology Research Center, Barretos Cancer Hospital – Pio XII, Barretos, SP, Brazil c Virology Laboratory, Institute of Tropical Medicine of Sa˜o Paulo, University of Sa˜o Paulo, Sa˜o Paulo, SP, Brazil d Head and Neck Surgery Service, Arnaldo Vieira de Carvalho Cancer Institute (ICAVC), Sa˜o Paulo, SP, Brazil e Faculty of Public Health, University of Sa˜o Paulo (FSP), Sa˜o Paulo, SP, Brazil f Laboratory of Clinical Pathology, Helio´polis Hospital, Sa˜o Paulo, SP, Brazil g GENCAPO – Head and Neck Genome Project b
Received 10 December 2014; received in revised form 14 January 2015; accepted 23 January 2015
KEYWORDS OSCC Methylation markers CCNA1 DAPK DCC TIMP3
Abstract Purpose: Activation of proto-oncogenes and inactivation of tumour suppressor genes are the major genetic alterations involved in carcinogenesis. The increase in methylation at the promoter region of a tumour suppressor gene can lead to gene inactivation, selecting cells with proliferative advantage. Thus, promoter hypermethylation is considered a marker in a variety of malignant tumours, including oral cavity. Experimental design: The methylation pattern of eight genes was evaluated in 40 oral cavity squamous cell carcinomas (OSCCs) and 40 saliva samples from healthy individuals by QMSP. Different combinations of genes were also assessed in order to identify gene panels that could better distinguish between OSCC and saliva samples. Results: CCNA1, DAPK, DCC and TIMP3 methylation were highly specific for being found in the OSCC samples. Moreover, the combination of these genes improved detection when compared with single markers, reaching values of 92.5% for sensitivity and specificity (when using the panel CCNA1, DCC, TIMP3). Moreover, DAPK, DCC and TIMP3 were hypermethylated in nearly 90% of clinically T1 and T2 cases. Conclusion: The pursuing of this panel of hypermethylated genes is an important tool for the detection of individuals with OSCC. Moreover, the identification of these markers in early stages of OSCC shows the feasibility of using the panel on saliva as possible biomarkers
⇑ Corresponding author at: Molecular Oncology Research Center, Barretos Cancer Hospital – Pio XII, Rua Antenor Duarte Villela, 1331, Bairro Dr. Paulo, Zip code: 14784-400, Brazil. Tel.: +55 17 3221 6600. E-mail address: [email protected] (A.L. Carvalho).
http://dx.doi.org/10.1016/j.ejca.2015.01.060 0959-8049/Ó 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Arantes L.M.R.B. et al., Validation of methylation markers for diagnosis of oral cavity cancer, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.01.060
2
L.M.R.B. Arantes et al. / European Journal of Cancer xxx (2015) xxx–xxx
for early diagnosis. The lack of association between the methylation status of these genes and clinical characteristics shows that they are able to distinguish OSCC cases irrespective of social and clinical factors (gender, age, human papillomavirus (HPV) status, clinical stage, vascular embolisation and perineural invasion). Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction Head and neck squamous cell carcinoma (HNSCC) is an aggressive malignant tumour type arising from the epithelial mucosal membranes of the upper-aerodigestive tract (oropharynx, hypopharynx and larynx) and the oral cavity, with a worldwide incidence of 685,000 patients annually [1–4]. Oral squamous cell carcinoma (OSCC) being the most common malignancy of the oral cavity poses a significant public health problem due its impact on the speech, mastication, taste, swallowing and aesthetics, with an annual incidence over than 300,000 cases [4]. Despite major progress, the overall survival of patients with oral cancer has slightly improved during the past 20 years, with 5-year survival rates around 60%. The poor survival rate has been ascribed to a high frequency of locoregional recurrences, and the occurrence of new tumours and deaths due to comorbidity; mostly to the fact that the majority of patients present advanced stages of OSCC at the time of diagnosis [5]. Only about one third of the patients present with early-stage disease (Union Internationale Contra le Cancer, UICC, stage I–II), whereas two thirds show already advanced disease (UICC stage III–IV) with poor outcome. The most important prognostic indicator for relapse of OSCC is the presence of metastatic spread to lymph nodes in the neck. In this case, the incidence of distant metastasis can be as high as 50% [6]. The presence of metastasis either regional or distant worsens the prognosis and reduces the survival rate in these patients. This makes it imperative to diagnose the disease at an early stage to facilitate appropriate therapeutic management to reduce the morbidity and mortality associated with this disease. Tobacco and alcohol consumption have been described as the most important risk factors associated with this carcinoma together with high-risk types of human papillomavirus (HPV) [7,8]. The use of molecular markers for tumour detection in body fluids has been explored with the intent to improve screening accuracy and cost-effectiveness. Body fluids can potentially carry whole cells as well as protein, DNA, and RNA that allow for detection of cellular alterations related to cancer. Examples of relevant body fluids used for detection include analysis of sputum for lung cancer diagnosis [9,10], urine for urologic tumours [11,12], saliva for HNSCC [13–16], breast fluid [17], as well as serum or plasma for almost all types of cancer
[18–20]. The detection of DNA methylation in body fluids such as saliva is a non-invasive technique that can easily obtain epithelial cells shed from the mucosal lining of the mouth and throat. Rosas et al. [21] demonstrated for the first time the possibility to detect hypermethylation in saliva. Likewise, Righini et al. [22] evaluated paired tumour and saliva samples collected at diagnosis and identified a panel of six genes with frequencies of hypermethylation of 82% and 78%, respectively. Carvalho et al. [16] were able to confirm an elevated rate of promoter hypermethylation detected in HNSCC patient salivary rinses by using a panel of gene promoters previously described as methylated in HNSCC but not in control subjects, by the same group [23]. Rettori et al. [14] in the analysis of salivary rinse samples taken at diagnosis of HNSCC patients, five genes (CCNA1, DAPK, DCC, MGMT and TIMP3) showed high specificity and sensitivity [24]. Thus, the detection of DNA methylation in body fluids opens the potential to develop biomarkers that can be useful for clinical use. Epigenetic gene silencing is a molecular mechanism of gene silencing through the methylation of its promoter region, and plays a vital role in the development of several types of cancer, including HNSCC [25–27]. In HNSCC, aberrant promoter methylation of CpG islands may affect genes involved in DNA repair [22,28–32], cell-cycle control [22,28–31,33,34], apoptosis [15,22,23,28,30,34,35], cell differentiation [23], cell proliferation [23] and cell adhesion [22,29,36,37]. The search for biomarkers to evaluate and measure the status of normal and pathological processes in cell biology as well as treatment responses is of paramount importance. The pursuing of these biomarkers is important for the identification of individuals in the early stages of cancer, and to stratify patients according to tumour prognosis and response to therapy profiles. Assuming that cancer results from genetic and epigenetic alterations, analysis based on gene-methylation profiles in combination with the pathological diagnosis would be useful in predicting the behaviour of these tumours. This study is a validation of previous studies published in this field. Carvalho et al. (2008) showed an elevated frequency of promoter hypermethylation in HNSCC in a panel of gene promoters previously described as methylated in HNSCC as well as other solid tumours [23]. Moreover, Rettori et al. (2013) found that
Please cite this article in press as: Arantes L.M.R.B. et al., Validation of methylation markers for diagnosis of oral cavity cancer, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.01.060
L.M.R.B. Arantes et al. / European Journal of Cancer xxx (2015) xxx–xxx
CCNA1, DAPK, MGMT, SFRP1 and TIMP3 were able to distinguish HNSCC tumours from control samples with high specificity (>96%) and sensitivity (21–62%) in a different case/control study [14]. We, thus consider that the evaluation of epigenetic changes, such as DNA methylation, would be useful as a tool for diagnosis, surveillance and prognosis of OSCC. 2. Patients and methods 2.1. Ethics statement
3
For controls, genomic DNA was isolated from whole saliva, which was deposited directly into an OG 500 OrageneÒ vial (DNA Genotek Inc, Ontario, Canada) until the 2 ml fill line was reached. DNA was extracted using the OrageneÒ vials according to the manufacturer’s protocol. Vials were incubated at 50 °C for 2 h to lyse the cells and digest nuclear proteins, then 160– 200 lL of OrageneÒ purifier was added to the whole sample before precipitation in ethanol and final resuspension in 400 lL of water. 2.4. Bisulphite treatment
The study protocol was approved by the Committees on Ethics in Research of Barretos Cancer Hospital, Heliopolis Hospital, Arnaldo Vieira de Carvalho Cancer Institute, Clinics Hospital of the Faculty of Medicine (University of Sao Paulo) and by the National Committee on Ethics in Research/CONEP (reference number 1763/05, 18/05/2005). All patients provided their written consent to participate in the study after being informed about the research purposes. 2.2. Patients and samples This study involved tissue specimens from 40 OSCC patients who underwent tumour resection. Only patients diagnosed with primary OSCC, previously untreated, underwent curative intent surgery and presenting with tumours at oral cavity were included in the study. Samples were collected by the Head and Neck Genome Project (GENCAPO), a collaborative consortium of research groups from hospitals and universities in Sao Paulo State, Brazil, whose aim is to develop clinical, genetic and epidemiological analyses of head and neck squamous cell carcinoma. The complete list of GENCAPO members 2014 and affiliations can be found at http://www.gencapo.famerp.br. Tissue samples were snap-frozen in liquid nitrogen within 30 min after resection and stored at 80 °C. All samples were evaluated microscopically for the presence of at least 70% of neoplastic tissue before DNA extraction. For the control group, 40 saliva samples from healthy subjects, matched by gender and age were collected at the Barretos Cancer Hospital (Sao Paulo, Brazil). All subjects responded a survey regarding lifestyle, and risk factors for upper aerodigestive tract malignancies, including alcohol and tobacco consumption. Smoking was defined as use of tobacco, chewable or smoked, for at least 1 year continuously. Alcohol use was defined as intake of more than one alcoholic drink per week, for at least 1 year continuously. 2.3. DNA extraction Genomic DNA was isolated from tissue samples using the TRIzol reagent (Invitrogen, Frederick, MD) following manufacturer’s recommendations.
Sodium-bisulphite conversion of 1 lg of DNA was performed using a kit (EpiTectÒ Bisulphite Kit; Qiagen, Valencia, CA), following the manufacturer’s recommendations. Briefly incubation of the target DNA with sodium bisulphite results in conversion of unmethylated cytosine residues into uracil, leaving the methylated cytosines unchanged. Therefore, bisulphite treatment gives rise to different DNA sequences for methylated and unmethylated DNA. The chemistry of cytosine deamination by sodium bisulphite involves three steps: (1) sulphonation; (2) deamination and (3) desulphonation. 2.5. Target gene selection A total of eight genes were selected for the examination of methylation abnormalities. The panel included genes reported as targets for epigenetic silencing in different human cancers. All genes evaluated in this study present tumour suppressor activities and their silencing could contribute to the carcinogenesis process. Among these genes are CCNA1 [15], DAPK [22,28,30,34] and DCC [23,34,35] which are involved in cell cycle control and apoptosis, CDH1 [22,29,36,38] and TIMP3 [22,29] in cell adhesion, MGMT [22,28–32] in DNA repair, HIC1 [23] in cell differentiation and proliferation, p16 [22,28–31,34] in cell-cycle control, and AIM1 (its function is not yet well understood). It has been shown that the expression of these genes may be affected by aberrant promoter methylation in association with transcription silencing in different types of human malignancies. These genes already have been evaluated in other studies and the proposal is to validate them in a different population. Carvalho et al. (2008) showed an elevated frequency of promoter hypermethylation in HNSCC in a panel of gene promoters previously described as methylated in HNSCC as well as other solid tumours [23]. Moreover, Rettori et al. (2013) found that CCNA1, DAPK, MGMT, SFRP1 and TIMP3 were able to distinguish HNSCC tumours from control samples with high specificity (>96%) and sensitivity (21–62%) in a different case/control study [14]. We, thus consider that the evaluation of epigenetic changes, such as DNA methylation, would be useful as a tool for diagnosis, surveillance and prognosis of OSCC.
Please cite this article in press as: Arantes L.M.R.B. et al., Validation of methylation markers for diagnosis of oral cavity cancer, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.01.060
4
L.M.R.B. Arantes et al. / European Journal of Cancer xxx (2015) xxx–xxx
2.6. Steps for gene evaluation The first step involved a screening evaluation, designed to select markers that better distinguish tumours from normal samples. Elimination of targets with high rates of promoter hypermethylation in normal control samples facilitated efficient definition of higher value markers. Significance was based on area under the curve (AUC) from receiving operating characteristic analysis, sensitivity, and specificity of that particular gene in differentiating tumour samples (cases) from saliva (controls) for the determination of a marker panel. 2.7. Quantitative methylation-specific PCR The quantitative methylation-specific PCR analyses (Q-MSP) were conducted as previously described (4). Basically, 30 ng of bisulphite-modified DNA was used as template in fluorogenic Q-MSP assays carried out in a final volume of 20 lL in 96-well plates in the 7900HT Fast Real-Time PCR System (Applied Biosystems). PCR was performed in separate wells for each primer/probe set and each sample was run in triplicate. The final reaction mixture contained 3 lL of bisulphitemodified DNA, 1.2 lmol/L of forward and reverse primers, 200 nmol/L of the probe, 0.5 U of platinum Taq polymerase (Invitrogen), 200 lmol/L dNTPs, 16.6 nmol/L ammonium sulphate, 67 mmol/L Trizma, 6.7 mmol/L magnesium chloride, 10 mmol/L mercaptoethanol, 0.1% DMSO, and 1X ROX dye (Invitrogen). PCR was conducted with the following conditions: 95 °C for 2 min, followed by 45 cycles at 95 °C for 15 s and 60 °C for 1 min. Each plate included patient DNA samples, multiple water blanks and serial dilutions (30–0.0003 ng) of a positive control allowing the construction of calibration curves. Leukocyte DNA obtained from a healthy individual was methylated in vitro using SssI methyltransferase (New England Biolabs) to generate methylated DNA at all CpG to be used as positive control. Primers and probes were obtained from the literature and specifically amplify the promoter regions of the eight genes of interest and the internal control gene, ACTB (Supplementary Table S1). The relative DNA methylation level of the eight genes in each sample was determined as a ratio of methylation specific PCR-amplified gene to ACTB and then multiplied by 100 (average value of triplicates of gene of interest divided by the average value of triplicates of ACTB 100). 2.8. HPV analysis (E7 HPV-16 Type-Specific Real Time PCR) Samples were submitted to a type-specific TaqManbased real time qPCR targeting HPV16 E7. qPCR was
performed in an ABI 7300 Real-Time PCR System (Applied Biosystems, Foster City, CA). All samples and controls were run in duplicate. qPCR was performed using oligonucleotide primers and probe as follows: forward (50 GATGAAATAGATGGTCCAGC30 ) and reverse (50 GCTTTGTACGCACAACC-GAAGC30 ) primers HPV16 E7 type-specific (35) and probe (50 6FAM-CAAGCAGAACCGGACAG-MGB-NFQ30 ), in a final volume reaction of 25 lL. Each qPCR contained, 1 TaqMan master mix (Applied Biosystems, Foster City, CA), 400 nM of each forward and reverse primers, 200 nM of fluorogenic TaqMan probe and 5 lL of extracted DNA. The amplification conditions were: 50 °C for 2 min, 95 °C for 10 min, followed by 40 cycles of 95 °C for 15 s, 55 °C for 1 min and 60 °C for 1 min. qPCR run included the following controls: (i) Caski cell line DNA (harbouring 600 copies/cell of HPV16) and (ii) water as negative control. 2.9. Statistical analysis Statistical analysis was performed using the software SPSS 19.0 for Windows. Hypermethylation of each gene was treated as a binary variable (methylated: any value other than zero; versus unmethylated: 0). Proportions of gene methylation were compared between tumour samples (from cases) and saliva samples (from controls) using either Chi-square or Fisher’s exact test, as appropriate. Sensitivity and specificity of each individual gene in detecting OSCC were calculated along with 95% confidence intervals (95% CI) (Table 2a). We evaluated all possible combinations of the selected markers for both tumour and saliva samples, where a positive panel was defined as at least one gene of the panel being methylated (Table 2b). The AUC, an index of predictive power, was also provided. For all analysis we considered statistical significance when p-value <0.05. 3. Results 3.1. Characteristics of the patients The study population consisted of 40 patients with a confirmed diagnostic of OSCC treated from September 2002 through October 2007. Patients were mainly males (90.0%) and Caucasians (71.8%), with ages ranging from 41 to 78 years (median, 54.5 years). Tobacco or alcohol consumption (current or past) was reported by 72.5% and 65.0% of the patients, respectively. With regard to HPV status, 7 (17.5%) of the tumours were HPV-positive. The primary tumour sites were located in the oral tongue (70.0%), in the floor of mouth (20.0%) and in other oral cavity sites (10.0%). Clinical stage was T1/ T2 in 13 cases (32.5%) and T3/T4 in 27 (67.5%); N0 in 11 cases (27.5%) and N+ in 29 (72.5%); and none of the patients had distant metastasis at diagnosis. Vascular embolisation, lymphatic invasion, perineural invasion
Please cite this article in press as: Arantes L.M.R.B. et al., Validation of methylation markers for diagnosis of oral cavity cancer, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.01.060
L.M.R.B. Arantes et al. / European Journal of Cancer xxx (2015) xxx–xxx
and peritumoural inflammatory infiltrate were present in 2 (5.0%), 18 (45.0%), 20 (50.0%) and 20 (51.3%), respectively of OSCC samples (Table 1). The control population was matched by gender and age, and thus comprised mainly males (90.0%) and Cau-
5
casians (90.0%), with ages ranging from 37 to 80 years (median, 53.5 years). Tobacco or alcohol consumption (current or past) was reported by 35.0% and 80.0% of the subjects, respectively. Regarding HPV status on saliva samples, 100% were tested as negative.
Table 1 Clinical and pathological data of the patients enrolled in the study. Characteristic
Age (Mean, median, <45 years 45–65 years >65 years
Patients (%) CCNA1
range) 5 (12.5) 25 (62.5) 10 (25.0)
DAPK
DCC
Negative Positive
P
TIMP3
Negative Positive
P
Negative Positive
P
Negative Positive
P
4 (80.0) 1 (20.0) 18 (72.0) 7 (28.0) 6 (60.0) 4 (40.0)
0.683 0 (0.0) 5 (20.0) 3 (30.0)
5 (100) 0.392 1 (20.0) 4 (80.0) 0.189 1 (20.0) 20 (80.0) 10 (40.0) 15 (60.0) 5 (20.0) 7 (70.0) 1 (10.0) 9 (90.0) 1 (10.0)
4 (80.0) 0.771 20 (80.0) 9 (90.0)
Gender Male Female
36 (90.0) 4 (10.0)
25 (69.4) 11 (30.6) 1.000 7 (19.4) 3 (75.0) 1 (25.0) 1 (25.0)
29 (80.6) 1.000 11 (30.7) 25 (69.4) 1.000 5 (13.9) 3 (75.0) 1 (25.0) 3 (75.0) 2 (50.0)
31 (86.1) 0.134 2 (50.0)
Race Caucasian Other
28 (71.8) 11 (28.2)
19 (67.9) 9 (32.1) 9 (81.8) 2 (18.2)
0.461 7 (25.0) 1 (9.1)
21 (75.0) 0.400 8 (28.6) 10 (90.9) 4 (36.4)
20 (71.4) 0.709 6 (21.4) 7 (63.6) 1 (9.1)
22 (78.6) 0.649 10 (90.9)
Tobacco consumption Current 29 (72.5) Former 10 (25.0) Never 1 (2.5)
23 (79.3) 6 (20.7) 4 (40.0) 6 (60.0) 1 (100) 0 (0.0)
0.052 3 (10.3) 4 (40.0) 1 (100)
26 (89.7) 0.017 9 (31.0) 6 (60.0) 2 (20.0) 0 (0.0) 1 (100)
20 (69.0) 0.244 4 (13.8) 8 (80.0) 2 (20.0) 0 (0.0) 1 (100)
25 (86.2) 0.081 8 (80.0) 0 (0.0)
Alcohol consumption Yes 26 (65.0) No 14 (35.0)
21 (80.8) 5 (19.2) 7 (50.0) 7 (50.0)
0.071 4 (15.4) 4 (28.6)
22 (84.6) 0.416 8 (30.8) 10 (71.4) 4 (28.6)
18 (69.2) 1.000 4 (15.4) 10 (71.4) 3 (21.4)
22 (84.6) 0.679 11 (78.6)
Clinical tumour status T1–T2 13 (32.5) T3–T4 27 (67.5)
10 (76.9) 3 (23.1) 18 (76.7) 9 (33.3)
0.716 2 (15.4) 6 (22.2)
11 (84.6) 1.000 2 (15.4) 11 (84.6) 0.271 2 (15.4) 21 (77.8) 10 (37.0) 17 (63.0) 5 (18.5)
11 (84.6) 1.000 22 (81.5)
Clinical nodal status N0 11 (27.5) N+ 29 (72.5)
10 (90.9) 1 (9.1) 0.124 2 (18.2) 18 (62.1) 11 (37.9) 6 (20.7)
9 (81.8) 1.000 3 (27.3) 23 (79.3) 9 (31.0)
8 (72.7) 1.000 2 (18.2) 20 (69.0) 5 (17.2)
9 (81.8) 1.000 24 (82.8)
Clinical TNM stage* Initial (I/II) 9 (22.5) Advanced (III/IV) 31 (77.5)
8 (88.9) 1 (11.1) 0.233 2 (22.2) 20 (64.5) 11 (35.5) 6 (19.4)
7 (77.8) 1.000 2 (22.2) 7 (77.8) 0.697 2 (22.2) 25 (80.6) 10 (32.3) 21 (67.7) 5 (16.1)
7 (77.8) 0.645 26 (83.9)
Primary tumour site Tongue 28 (70.0) Floor of mouth 8 (20.0) Others 4 (10.0)
19 (67.9) 9 (32.1) 5 (62.5) 3 (37.5) 4 (100) 0 (0.0)
0.370 6 (21.4) 1 (12.5) 1 (25.0)
22 (78.6) 0.827 9 (32.1) 7 (87.5) 2 (25.0) 3 (75.0) 1 (25.0)
19 (67.9) 0.903 5 (17.9) 6 (75.0) 1 (12.5) 3 &5.0) 1 (25.0)
23 (82.1) 0.862 7 (87.5) 3 (75.0)
HPV status Positive Negative
4 (57.1) 3 (42.9) 24 (72.7) 9 (27.3)
0.410 22 (28.6) 5 (71.4) 0.611 1 (14.3) 6 (85.7) 0.652 0 (0.0) 6 (18.2) 27 (81.8) 11 (33.3) 22 (66.7) 7 (21.2)
7 (100.0) 0.317 26 (78.8)
33 (82.5) 7 (17.5)
Vascular embolisation Positive 2 (5.0) Negative 38 (95.0)
2 (100) 0 (0.0) 1.000 1 (50.0) 26 (68.4) 12 (31.6) 7 (18.4)
1 (50.0) 0.364 1 (50.0) 1 (50.0) 0.515 1 (50.0) 31 (81.6) 11 (28.9) 27 (71.7) 6 (15.8)
1 (50.0) 0.323 32 (84.2)
Lymphatic invasion Positive Negative
18 (46.1) 21 (53.9)
11 (61.1) 7 (38.9) 17 (81.0) 4 (19.0)
0.170 4 (22.2) 4 (19.0)
14 (77.8) 1.000 7 (38.9) 17 (81.0) 5 (23.8)
11 (61.1) 0.309 4 (22.2) 16 (76.2) 3 (14.3)
14 (77.8) 0.682 18 (85.7)
Perineural invasion Positive Negative
20 (50.0) 20 (50.0)
13 (65.0) 7 (35.0) 0.490 3 (15.0) 15 (75.0) 5 (25.0.0 5 (25.0)
17 (85.0) 0.695 7 (35.0) 15 (75.0) 5 (25.0)
13 (65.0) 0.490 3 (15.0) 15 (75.0) 4 (20.0)
17 (85.0) 1.000 16 (80.0)
16 (80.0) 1.000 8 (40.0) 15 (78.9) 4 (21.1)
12 (60.0) 0.301 3 (15.0) 15 (78.9) 4 (21.1)
17 (85.0) 0.695 15 (78.9)
Peritumoral inflammatory infiltrate Intense/Moderate 20 (51.3) 14 (70.0) 6 (30.0) Scarce/Absent 19 (48.7) 14 (73.7) 5 (26.3) *
1.000 4 (20.0) 4 (21.1)
Clinical stage according to TNM Classification of Malignant Tumours – 6th ed.
Please cite this article in press as: Arantes L.M.R.B. et al., Validation of methylation markers for diagnosis of oral cavity cancer, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.01.060
6
L.M.R.B. Arantes et al. / European Journal of Cancer xxx (2015) xxx–xxx
3.2. Promoter hypermethylation of AIM1, CCNA1, CDH1, DAPK, DCC, HIC1, MGMT, and TIMP3 The promoter methylation status of eight genes was tested in tumour samples from 40 OSCC patients. Promoter methylation was found in 17.5% at AIM1, 30.0% at CCNA1, 95.0% at CDH1, 80.0% at DAPK, 70.0% at DCC, 95.0% at HIC1, 30.0% at MGMT and 82.5% at TIMP3, of OSCC cases. The methylation status of the same genes was evaluated in 40 normal saliva from controls, and the specificity levels varied from 0% to 100%. These analyses revealed that hypermethylation of AIM1 and MGMT was not frequent in head and neck tumour, thus showing low sensitivity, while CDH1 and HIC1 were frequently methylated in saliva controls, indicating a low specificity (Table 2a, Figs. 1 and 2). 3.3. Promoter hypermethylation of the panel Excluding genes with low sensitivity and/or specificity such as AIM1, CDH1, HIC1 and MGMT (AUC < 0.6), several combinations were performed as a panel, where a positive panel was defined as at least one gene being methylated in the sample. In the remaining four genes: CCNA1, DAPK, DCC and TIMP3 (AUC > 0.6), the hypermethylation detection rate in tumours varied from 77.5% to 95.0% of sensitivity and specificity rates ranged from 75.0% to 97.5%, depending on the tested panel (Table 2b and Fig. 2). These four genes had highest value for the full panel: 95.0% of sensitivity and AUC = 0.925. 3.4. Association between aberrant methylation and patient characteristics The methylation pattern of the four selected genes was analysed for potential associations with clinical and pathological characteristics of OSCC patients, including age, gender, race, tobacco use, alcohol consumption, primary tumour site, T stage, N stage, vascular embolisation, lymphatic invasion, perineural invasion and peritumoural inflammatory infiltrate. Regarding DAPK, 89.7% of patients harbouring hypermethylation were
current smokers (p = 0.017). No significant associations were observed between the methylation profile of DCC and TIMP3 and clinical-pathological characteristics of OSCC patients (Table 1). 4. Discussion The progression of cancer may be associated with changes in DNA methylation that can be used as biomarkers for diagnosis, treatment selection and prognosis. Several studies indicate that abnormal DNA methylation may occur early in cancer development, including HNSCC [13,16,23,29,38]. Activation of proto-oncogenes and inactivation of tumour suppressor genes are the major genetic alterations involved in carcinogenesis. In addition, epigenetic changes (altered patterns of gene expression that do not affect the primary sequence of the DNA) can alter the expression of genes important in the development of a variety of tumours [39]. The use of aberrant promoter hypermethylation detection markers of tumour-specific cells in different materials from OSCC such as body fluids, exfoliated cells and solid tumours, has been proposed. The present study evaluated a population of OSCC patients and matched controls to determine the ability of Q-MSP to detect promoter methylation of specific genes that could distinguish these groups. The sensitivity of the Q-MSP technique allows the detection of the presence of methylated alleles in a background of normal at a threshold of 1/1000–1/10,000 [23]. Therefore, this strategy allowed us to define highly tumour-specific methylated genes that were rarely or never present in saliva controls. It is noteworthy that the control population in the current study can be considered as at high risk for OSCC with the majority of them being males, with median age 55 years and with reported regular consumption of tobacco and alcohol. Given that the association between clinical and demographic data and the gene methylation patterns of the genes evaluated did not show relevant statistically significance, we can hypothesise that these factors do not significantly
Table 2a Comparison of hypermethylation detection on tumour samples (Oral squamous cell carcinoma (OSCC) patients) and saliva (controls). Individual gene evaluation
Case (n)
Control (n)
Sensitivity % (95% confidence interval (CI))
Specificity % (95% CI)
Area under the curve (AUC)
TIMP3 DCC DAPK CCNA1 AIM1 MGMT CDH1 HIC1
40 40 40 40 40 40 40 40
40 40 40 40 40 40 40 40
82.5 70.0 80.0 30.0 17.5 30.0 95.0 95.0
100.0 (0.91–1.00) 95.0 (0.83–0.98) 82.5 (0.68–0.91) 100.0 (0.91–1.00) 100.0 (0.91–1.00) 85.0 (0.70–0.92) 20.0 (0.10–0.34) 0.0 (0.00–0.08)
0.913 0.825 0.813 0.638 0.587 0.575 0.575 0.475
(0.68–0.91) (0.54–0.81) (0.65–0.89) (0.18–0.45) (0.08–0.31) (0.18–0.45) (0.83–0.98) (0.83–0.98)
Please cite this article in press as: Arantes L.M.R.B. et al., Validation of methylation markers for diagnosis of oral cavity cancer, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.01.060
L.M.R.B. Arantes et al. / European Journal of Cancer xxx (2015) xxx–xxx
7
Fig. 1. Methylation rates of AIM1, CCNA1, CDH1, DAPK, DCC, HIC1, MGMT and TIMP3 in oral squamous cell carcinoma (OSCC) tissue samples and saliva from controls. Each symbol represents a different sample. X-axis, proportion of methylated cases/tested cases for each sample type; Y-axis, quantity of hypermethylation (gene of interest/ACTB 100).
change the specificity and sensitivity levels of these genes in distinguishing cases from controls. Finally, the high sensitivity of Q-MSP may have allowed the detection of low quantity methylation even in a subset of healthy controls. The search for biomarkers has as its main goal to evaluate and measure the status of normal
and pathological biological processes as well as pharmacological responses to certain treatments. The searching for these biomarkers is an important step for the identification of individuals in the early stages of head and neck cancer for its diagnostic and prognostic relevance. Therefore, assuming that cancer results from genetic and epigenetic changes, analyses based on the
Please cite this article in press as: Arantes L.M.R.B. et al., Validation of methylation markers for diagnosis of oral cavity cancer, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.01.060
8
L.M.R.B. Arantes et al. / European Journal of Cancer xxx (2015) xxx–xxx
Fig. 2. ROC curve for CCNAI, DAPK, DCC and TIMP3 and selected panels from oral squamous cell carcinoma (OSCC) tissue samples and saliva from controls.
Table 2b Comparison of best combination of genes for hypermethylation detection on tumour samples (oral squamous cell carcinoma (OSCC) patients) and saliva (controls). Panel
Case (n)
Control (n)
Sensitivity % (95% confidence interval (CI))
Specificity % (95% CI)
Area under the curve (AUC)
CCNA1 + DCC + TIMP3 DCC + TIMP3 CCNA1 + TIMP3 DAPK + DCC CCNA1 + DAPK + DCC + TIMP3 CCNA1 + DAPK + DCC DAPK + DCC + TIMP3 CCNA1 + DCC DAPK + TIMP3 CCNA1 + DAPK + TIMP3 CCNA1 + DAPK
40 40 40 40 40 40 40 40 40 40 40
40 40 40 40 40 40 40 40 40 40 40
92.5 90.0 85.0 95.0 95.0 95.0 92.5 77.5 85.0 87.5 85.0
92.5 95.0 97.5 77.5 75.0 75.0 77.5 90.0 82.5 80.0 80.0
0.925 0.925 0.913 0.862 0.850 0.850 0.850 0.838 0.838 0.837 0.825
methylation profile in combination with the pathological diagnosis would be useful in predicting the behaviour of these head and neck cancers. We were able to show the possibility of using single genes for tumour detection; however, the use of a com-
(0.80–0.97) (0.76–0.96) (0.70–0.92) (0.8–0.98) (0.83–0.98) (0.83–0.98) (0.80–0.97) (0.77–0.87) (0.70–0.92) (0.73–0.94) (0.70–0.92)
(0.80–0.97) (0.83–0.98) (0.87–0.99) (0.62–0.87) (0.59–0.85) (0.59–0.85) (0.62–0.87) (0.76–0.96) (0.68–0.91) (0.65–0.89) (0.65–0.89)
bination of genes (CCNA1, DCC, TIMP3 and DAPK) in a panel provided improvement in sensitivity rates. Since variations in the patterns of promoter hypermethylation in individual tumours may occur depending on alterations on specific molecular pathways, the use
Please cite this article in press as: Arantes L.M.R.B. et al., Validation of methylation markers for diagnosis of oral cavity cancer, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.01.060
L.M.R.B. Arantes et al. / European Journal of Cancer xxx (2015) xxx–xxx
of multiple genes may provide greater applicability and coverage for different tumours when compared with a single gene for general detection, corroborating to results obtained by Carvalho et al. [23] and Rettori et al. [14] in different cohorts. From the initial screening of eight genes, four genes were selected as part of a panel to distinguish OSCC patients and healthy controls. The possible combination of those four genes was able to provide a sensitivity ranging from 77.5% to 95.0% and a specificity ranging from 75.0% to 97.5%. Appropriately matched normal controls are of paramount importance for the adequate assessment of the utility of promoter hypermethylation in OSCC. In addition, the presence of promoter hypermethylation of tumour suppressor genes observed in control populations can happen as a random or a physiologic event associated with tissue-specificity or related to age or environmental carcinogenic exposures. These factors significantly affect the selection of a healthy control group, as limited size control groups comprised of young donors, with minimal tobacco and ethanol exposure can bias reporting of falsely elevated specificity for candidate genes [23]. In general, we were able to validate a panel for OSCC detection with high specificity and high sensitivity. Such characteristics would be highly useful as a method for diagnostic and screening, other panels were defined as harbouring high specificity but accompanied by a low sensitivity, as well as high sensitivity and low specificity which may have potential as prognostic markers during surveillance after an OSCC treatment or in high-risk populations. The lack of association between the methylation status of these markers and clinical characteristics at diagnosis shows that they are able to distinguish tumour cases irrespective of gender, age, HPV status, clinical stage, vascular embolisation, lymphatic invasion, perineural invasion and peritumoral inflammatory infiltrate. Moreover, DAPK, DCC and TIMP3 were found hypermethylated in nearly 90% of clinically T1 and T2 cases, showing the feasibility of using these markers in early diagnosis.
5. Conclusion The search for biomarkers is important for the identification of individuals with OSCC and ideally in the early stages decreasing morbidity and increasing survival rates. Body fluids such as saliva, which can be obtained by non-invasive techniques and can be used to evaluate gene methylation profiles, are potential sources for the development of biomarkers for diagnosis, surveillance and prognosis. The combination of hypermethylation pattern for CCNA1, DCC, TIMP3 and DAPK genes improved detection when compared with single markers, reaching values of 92.5% for sensitivity and specificity (when using the panel CCNA1,
9
DCC, TIMP3). Moreover, DAPK, DCC and TIMP3 are hypermethylated in nearly 90% of clinically T1 and T2 cases, showing the feasibility of using these markers in early diagnosis. Conflict of interest statement None declared. Acknowledgments The authors would like to thank the members of the GENCAPO (Head and Neck Genome) Project and Fundac¸a˜o de Amparo a` Pesquisa do Estado de Sa˜o Paulo (FAPESP). This study was supported by FAPESP (10/51168-0). L.M.R.B.A. is recipient of scholarship from Fundac¸a˜o de Amparo a` Pesquisa do Estado de Sa˜o Paulo (FAPESP; grant number 2011/02864-7). A.L.C. has a National Counsel of Technological and Scientific Development Scholarship (CNPq). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10. 1016/j.ejca.2015.01.060. References [1] Kim L, King T, Agulnik M. Head and neck cancer: changing epidemiology and public health implications. Oncology 2010;24(10):915–9, 924. [2] Marcu LG, Yeoh E. A review of risk factors and genetic alterations in head and neck carcinogenesis and implications for current and future approaches to treatment. J Cancer Res Clin Oncol 2009;135(10):1303–14. [3] Ragin CC, Modugno F, Gollin SM. The epidemiology and risk factors of head and neck cancer: a focus on human papillomavirus. J Dent Res 2007;86(2):104–14. [4] Wang Q, Gao P, Wang X, Duan Y. Investigation and identification of potential biomarkers in human saliva for the early diagnosis of oral squamous cell carcinoma. Clin Chim Acta 2014;427:79–85. [5] Montebugnoli L, Gissi DB, Flamminio F, Gentile L, Dallera V, Leonardi E, et al. Clinicopathologic parameters related to recurrence and locoregional metastasis in 180 oral squamous cell carcinomas. Int J Surg Pathol 2014;22(1):55–62. [6] Grobe A, Blessmann M, Hanken H, Friedrich RE, Schon G, Wikner J, et al. Prognostic relevance of circulating tumor cells in blood and disseminated tumor cells in bone marrow of patients with squamous cell carcinoma of the oral cavity. Clin Cancer Res 2014;20(2):425–33. [7] Leemans CR, Braakhuis BJ, Brakenhoff RH. The molecular biology of head and neck cancer. Nat Rev Cancer 2011;11(1):9–22. [8] Olivieri EH, da Silva SD, Mendonca FF, Urata YN, Vidal DO, Faria Mde A. CYP1A2*1C, CYP2E1*5B, and GSTM1 polymorphisms are predictors of risk and poor outcome in head and neck squamous cell carcinoma patients. Oral Oncol 2009;45(9):e73–9. [9] Markopoulou S, Nikolaidis G, Liloglou T. DNA methylation biomarkers in biological fluids for early detection of respiratory tract cancer. Clin Chem Lab Med 2012;50(10):1723–31.
Please cite this article in press as: Arantes L.M.R.B. et al., Validation of methylation markers for diagnosis of oral cavity cancer, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.01.060
10
L.M.R.B. Arantes et al. / European Journal of Cancer xxx (2015) xxx–xxx
[10] Leng S, Liu Y, Thomas CL, Gauderman WJ, Picchi MA, Bruse SE, et al. Native American ancestry affects the risk for gene methylation in the lungs of Hispanic smokers from New Mexico. Am J Respir Crit Care Med 2013;188(9):1110–6. [11] Garcia-Baquero R, Puerta P, Beltran M, Alvarez M, Sacristan R, Alvarez-Ossorio JL, et al. Methylation of a novel panel of tumor suppressor genes in urine moves forward noninvasive diagnosis and prognosis of bladder cancer: a 2-center prospective study. J Urol 2013;190(2):723–30. [12] Zuiverloon TC, Beukers W, van der Keur KA, Nieuweboer AJ, Reinert T, Dyrskjot L, et al. Combinations of urinary biomarkers for surveillance of patients with incident nonmuscle invasive bladder cancer: the European FP7 UROMOL project. J Urol 2013;189(5):1945–51. [13] Matthews AM, Kaur H, Dodd M, D’Souza J, Liloglou T, Shaw RJ, et al. Saliva collection methods for DNA biomarker analysis in oral cancer patients. Br J Oral Maxillofac Surg 2013;51(5):394–8. [14] Rettori MM, de Carvalho AC, Bomfim Longo AL, de Oliveira CZ, Kowalski LP, Carvalho AL, et al. Prognostic significance of TIMP3 hypermethylation in post-treatment salivary rinse from head and neck squamous cell carcinoma patients. Carcinogenesis 2013;34(1):20–7. [15] Sun W, Zaboli D, Liu Y, Arnaoutakis D, Khan T, Wang H, et al. Comparison of promoter hypermethylation pattern in salivary rinses collected with and without an exfoliating brush from patients with HNSCC. PLoS One 2012;7(3):e33642. [16] Carvalho AL, Henrique R, Jeronimo C, Nayak CS, Reddy AN, Hoque MO, et al. Detection of promoter hypermethylation in salivary rinses as a biomarker for head and neck squamous cell carcinoma surveillance. Clin Cancer Res 2011;17(14):4782–9. [17] Twelves D, Nerurkar A, Osin P, Dexter T, Ward A, Gui GP, et al. DNA promoter hypermethylation profiles in breast duct fluid. Breast Cancer Res Treat 2013;139(2):341–50. [18] Muraoka T, Soh J, Toyooka S, Aoe K, Fujimoto N, Hashida S, et al. The degree of microRNA-34b/c methylation in serumcirculating DNA is associated with malignant pleural mesothelioma. Lung Cancer 2013;82(3):485–90. [19] Zhao ZH, Fan YC, Yang Y, Wang K. Association between Ras association domain family 1A promoter methylation and hepatocellular carcinoma: a meta-analysis. World J Gastroenterol 2013;19(41):7189–96. [20] Huang YT, Li FF, Ke C, Li Z, Li ZT, Zou XF, et al. PTPRO promoter methylation is predictive of poorer outcome for HER2positive breast cancer: indication for personalized therapy. J Transl Med 2013;11:245. [21] Rosas SL, Koch W, da Costa Carvalho MG, Wu L, Califano J, Westra W, et al. Promoter hypermethylation patterns of p16, O6methylguanine-DNA-methyltransferase, and death-associated protein kinase in tumors and saliva of head and neck cancer patients. Cancer Res 2001;61(3):939–42. [22] Righini CA, de Fraipont F, Timsit JF, Faure C, Brambilla E, Reyt E, et al. Tumor-specific methylation in saliva: a promising biomarker for early detection of head and neck cancer recurrence. Clin Cancer Res 2007;13(4):1179–85. [23] Carvalho AL, Jeronimo C, Kim MM, Henrique R, Zhang Z, Hoque MO, et al. Evaluation of promoter hypermethylation detection in body fluids as a screening/diagnosis tool for head and neck squamous cell carcinoma. Clin Cancer Res 2008;14(1):97–107. [24] Arantes LM, de Carvalho AC, Melendez ME, Carvalho AL, Goloni-Bertollo EM. Methylation as a biomarker for head and neck cancer. Oral Oncol 2014;50(6):587–92.
[25] Brinkmann O, Kastratovic DA, Dimitrijevic MV, Konstantinovic VS, Jelovac DB, Antic J, et al. Oral squamous cell carcinoma detection by salivary biomarkers in a Serbian population. Oral Oncol 2011;47(1):51–5. [26] Park NJ, Zhou H, Elashoff D, Henson BS, Kastratovic DA, Abemayor E, et al. Salivary microRNA: discovery, characterization, and clinical utility for oral cancer detection. Clin Cancer Res 2009;15(17):5473–7. [27] Li Y, Zhou X, St John MA, Wong DT. RNA profiling of cell-free saliva using microarray technology. J Dent Res 2004;83(3):199–203. [28] Pierini S, Jordanov SH, Mitkova AV, Chalakov IJ, Melnicharov MB, Kunev KV, et al. Promoter hypermethylation of CDKN2A, MGMT, MLH1, and DAPK genes in laryngeal squamous cell carcinoma and their associations with clinical profiles of the patients. Head Neck 2013;36(8):1103–8. [29] Shaw RJ, Hall GL, Lowe D, Bowers NL, Liloglou T, Field JK, et al. CpG island methylation phenotype (CIMP) in oral cancer: associated with a marked inflammatory response and less aggressive tumour biology. Oral Oncol 2007;43(9):878–86. [30] Dikshit RP, Gillio-Tos A, Brennan P, De Marco L, Fiano V, Martinez-Penuela JM, et al. Hypermethylation, risk factors, clinical characteristics, and survival in 235 patients with laryngeal and hypopharyngeal cancers. Cancer 2007;110(8):1745–51. [31] Kato K, Hara A, Kuno T, Mori H, Yamashita T, Toida M, et al. Aberrant promoter hypermethylation of p16 and MGMT genes in oral squamous cell carcinomas and the surrounding normal mucosa. J Cancer Res Clin Oncol 2006;132(11):735–43. [32] Zuo C, Ai L, Ratliff P, Suen JY, Hanna E, Brent TP, et al. O6methylguanine-DNA methyltransferase gene: epigenetic silencing and prognostic value in head and neck squamous cell carcinoma. Cancer Epidemiol Biomarkers Prev 2004;13(6):967–75. [33] Demokan S, Chuang A, Suoglu Y, Ulusan M, Yalniz Z, Califano JA, et al. Promoter methylation and loss of p16(INK4a) gene expression in head and neck cancer. Head Neck 2012;34(10):1470–5. [34] Hasegawa M, Nelson HH, Peters E, Ringstrom E, Posner M, Kelsey KT. Patterns of gene promoter methylation in squamous cell cancer of the head and neck. Oncogene 2002;21(27):4231–6. [35] Brait M, Loyo M, Rosenbaum E, Ostrow KL, Markova A, Papagerakis S, et al. Correlation between BRAF mutation and promoter methylation of TIMP3, RARbeta2 and RASSF1A in thyroid cancer. Epigenetics 2012;7(7):710–9. [36] Marsit CJ, Posner MR, McClean MD, Kelsey KT. Hypermethylation of E-cadherin is an independent predictor of improved survival in head and neck squamous cell carcinoma. Cancer 2008;113(7):1566–71. [37] de Moraes RV, Oliveira DT, Landman G, de Carvalho F, Caballero O, Nonogaki S, et al. E-cadherin abnormalities resulting from CPG methylation promoter in metastatic and nonmetastatic oral cancer. Head Neck 2008;30(1):85–92. [38] Shaw RJ, Liloglou T, Rogers SN, Brown JS, Vaughan ED, Lowe D, et al. Promoter methylation of P16, RARbeta, E-cadherin, cyclin A1 and cytoglobin in oral cancer: quantitative evaluation using pyrosequencing. Br J Cancer 2006;94(4):561–8. [39] Sweeney C, Boucher KM, Samowitz WS, Wolff RK, Albertsen H, Curtin K, et al. Oncogenetic tree model of somatic mutations and DNA methylation in colon tumors. Genes Chromosomes Cancer 2009;48(1):1–9.
Please cite this article in press as: Arantes L.M.R.B. et al., Validation of methylation markers for diagnosis of oral cavity cancer, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.01.060
Available at www.sciencedirect.com
ScienceDirect journal homepage: www.ejcancer.com
Validation of methylation markers for diagnosis of oral cavity cancer L.M.R.B. Arantes a,b, A.C. de Carvalho b, M.E. Melendez b, C.C. Centrone c, J.F. Go´is-Filho d, T.N. Toporcov e, D.N. Caly f, E.H. Tajara a, E.M. Goloni-Bertollo a, A.L. Carvalho b,⇑, GENCAPO g a
Department of Molecular Biology, Sa˜o Jose´ do Rio Preto Medical School (FAMERP), Sa˜o Jose´ do Rio Preto, SP, Brazil Molecular Oncology Research Center, Barretos Cancer Hospital – Pio XII, Barretos, SP, Brazil c Virology Laboratory, Institute of Tropical Medicine of Sa˜o Paulo, University of Sa˜o Paulo, Sa˜o Paulo, SP, Brazil d Head and Neck Surgery Service, Arnaldo Vieira de Carvalho Cancer Institute (ICAVC), Sa˜o Paulo, SP, Brazil e Faculty of Public Health, University of Sa˜o Paulo (FSP), Sa˜o Paulo, SP, Brazil f Laboratory of Clinical Pathology, Helio´polis Hospital, Sa˜o Paulo, SP, Brazil g GENCAPO – Head and Neck Genome Project b
Received 10 December 2014; received in revised form 14 January 2015; accepted 23 January 2015
KEYWORDS OSCC Methylation markers CCNA1 DAPK DCC TIMP3
Abstract Purpose: Activation of proto-oncogenes and inactivation of tumour suppressor genes are the major genetic alterations involved in carcinogenesis. The increase in methylation at the promoter region of a tumour suppressor gene can lead to gene inactivation, selecting cells with proliferative advantage. Thus, promoter hypermethylation is considered a marker in a variety of malignant tumours, including oral cavity. Experimental design: The methylation pattern of eight genes was evaluated in 40 oral cavity squamous cell carcinomas (OSCCs) and 40 saliva samples from healthy individuals by QMSP. Different combinations of genes were also assessed in order to identify gene panels that could better distinguish between OSCC and saliva samples. Results: CCNA1, DAPK, DCC and TIMP3 methylation were highly specific for being found in the OSCC samples. Moreover, the combination of these genes improved detection when compared with single markers, reaching values of 92.5% for sensitivity and specificity (when using the panel CCNA1, DCC, TIMP3). Moreover, DAPK, DCC and TIMP3 were hypermethylated in nearly 90% of clinically T1 and T2 cases. Conclusion: The pursuing of this panel of hypermethylated genes is an important tool for the detection of individuals with OSCC. Moreover, the identification of these markers in early stages of OSCC shows the feasibility of using the panel on saliva as possible biomarkers
⇑ Corresponding author at: Molecular Oncology Research Center, Barretos Cancer Hospital – Pio XII, Rua Antenor Duarte Villela, 1331, Bairro Dr. Paulo, Zip code: 14784-400, Brazil. Tel.: +55 17 3221 6600. E-mail address: [email protected] (A.L. Carvalho).
http://dx.doi.org/10.1016/j.ejca.2015.01.060 0959-8049/Ó 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Arantes L.M.R.B. et al., Validation of methylation markers for diagnosis of oral cavity cancer, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.01.060
2
L.M.R.B. Arantes et al. / European Journal of Cancer xxx (2015) xxx–xxx
for early diagnosis. The lack of association between the methylation status of these genes and clinical characteristics shows that they are able to distinguish OSCC cases irrespective of social and clinical factors (gender, age, human papillomavirus (HPV) status, clinical stage, vascular embolisation and perineural invasion). Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction Head and neck squamous cell carcinoma (HNSCC) is an aggressive malignant tumour type arising from the epithelial mucosal membranes of the upper-aerodigestive tract (oropharynx, hypopharynx and larynx) and the oral cavity, with a worldwide incidence of 685,000 patients annually [1–4]. Oral squamous cell carcinoma (OSCC) being the most common malignancy of the oral cavity poses a significant public health problem due its impact on the speech, mastication, taste, swallowing and aesthetics, with an annual incidence over than 300,000 cases [4]. Despite major progress, the overall survival of patients with oral cancer has slightly improved during the past 20 years, with 5-year survival rates around 60%. The poor survival rate has been ascribed to a high frequency of locoregional recurrences, and the occurrence of new tumours and deaths due to comorbidity; mostly to the fact that the majority of patients present advanced stages of OSCC at the time of diagnosis [5]. Only about one third of the patients present with early-stage disease (Union Internationale Contra le Cancer, UICC, stage I–II), whereas two thirds show already advanced disease (UICC stage III–IV) with poor outcome. The most important prognostic indicator for relapse of OSCC is the presence of metastatic spread to lymph nodes in the neck. In this case, the incidence of distant metastasis can be as high as 50% [6]. The presence of metastasis either regional or distant worsens the prognosis and reduces the survival rate in these patients. This makes it imperative to diagnose the disease at an early stage to facilitate appropriate therapeutic management to reduce the morbidity and mortality associated with this disease. Tobacco and alcohol consumption have been described as the most important risk factors associated with this carcinoma together with high-risk types of human papillomavirus (HPV) [7,8]. The use of molecular markers for tumour detection in body fluids has been explored with the intent to improve screening accuracy and cost-effectiveness. Body fluids can potentially carry whole cells as well as protein, DNA, and RNA that allow for detection of cellular alterations related to cancer. Examples of relevant body fluids used for detection include analysis of sputum for lung cancer diagnosis [9,10], urine for urologic tumours [11,12], saliva for HNSCC [13–16], breast fluid [17], as well as serum or plasma for almost all types of cancer
[18–20]. The detection of DNA methylation in body fluids such as saliva is a non-invasive technique that can easily obtain epithelial cells shed from the mucosal lining of the mouth and throat. Rosas et al. [21] demonstrated for the first time the possibility to detect hypermethylation in saliva. Likewise, Righini et al. [22] evaluated paired tumour and saliva samples collected at diagnosis and identified a panel of six genes with frequencies of hypermethylation of 82% and 78%, respectively. Carvalho et al. [16] were able to confirm an elevated rate of promoter hypermethylation detected in HNSCC patient salivary rinses by using a panel of gene promoters previously described as methylated in HNSCC but not in control subjects, by the same group [23]. Rettori et al. [14] in the analysis of salivary rinse samples taken at diagnosis of HNSCC patients, five genes (CCNA1, DAPK, DCC, MGMT and TIMP3) showed high specificity and sensitivity [24]. Thus, the detection of DNA methylation in body fluids opens the potential to develop biomarkers that can be useful for clinical use. Epigenetic gene silencing is a molecular mechanism of gene silencing through the methylation of its promoter region, and plays a vital role in the development of several types of cancer, including HNSCC [25–27]. In HNSCC, aberrant promoter methylation of CpG islands may affect genes involved in DNA repair [22,28–32], cell-cycle control [22,28–31,33,34], apoptosis [15,22,23,28,30,34,35], cell differentiation [23], cell proliferation [23] and cell adhesion [22,29,36,37]. The search for biomarkers to evaluate and measure the status of normal and pathological processes in cell biology as well as treatment responses is of paramount importance. The pursuing of these biomarkers is important for the identification of individuals in the early stages of cancer, and to stratify patients according to tumour prognosis and response to therapy profiles. Assuming that cancer results from genetic and epigenetic alterations, analysis based on gene-methylation profiles in combination with the pathological diagnosis would be useful in predicting the behaviour of these tumours. This study is a validation of previous studies published in this field. Carvalho et al. (2008) showed an elevated frequency of promoter hypermethylation in HNSCC in a panel of gene promoters previously described as methylated in HNSCC as well as other solid tumours [23]. Moreover, Rettori et al. (2013) found that
Please cite this article in press as: Arantes L.M.R.B. et al., Validation of methylation markers for diagnosis of oral cavity cancer, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.01.060
L.M.R.B. Arantes et al. / European Journal of Cancer xxx (2015) xxx–xxx
CCNA1, DAPK, MGMT, SFRP1 and TIMP3 were able to distinguish HNSCC tumours from control samples with high specificity (>96%) and sensitivity (21–62%) in a different case/control study [14]. We, thus consider that the evaluation of epigenetic changes, such as DNA methylation, would be useful as a tool for diagnosis, surveillance and prognosis of OSCC. 2. Patients and methods 2.1. Ethics statement
3
For controls, genomic DNA was isolated from whole saliva, which was deposited directly into an OG 500 OrageneÒ vial (DNA Genotek Inc, Ontario, Canada) until the 2 ml fill line was reached. DNA was extracted using the OrageneÒ vials according to the manufacturer’s protocol. Vials were incubated at 50 °C for 2 h to lyse the cells and digest nuclear proteins, then 160– 200 lL of OrageneÒ purifier was added to the whole sample before precipitation in ethanol and final resuspension in 400 lL of water. 2.4. Bisulphite treatment
The study protocol was approved by the Committees on Ethics in Research of Barretos Cancer Hospital, Heliopolis Hospital, Arnaldo Vieira de Carvalho Cancer Institute, Clinics Hospital of the Faculty of Medicine (University of Sao Paulo) and by the National Committee on Ethics in Research/CONEP (reference number 1763/05, 18/05/2005). All patients provided their written consent to participate in the study after being informed about the research purposes. 2.2. Patients and samples This study involved tissue specimens from 40 OSCC patients who underwent tumour resection. Only patients diagnosed with primary OSCC, previously untreated, underwent curative intent surgery and presenting with tumours at oral cavity were included in the study. Samples were collected by the Head and Neck Genome Project (GENCAPO), a collaborative consortium of research groups from hospitals and universities in Sao Paulo State, Brazil, whose aim is to develop clinical, genetic and epidemiological analyses of head and neck squamous cell carcinoma. The complete list of GENCAPO members 2014 and affiliations can be found at http://www.gencapo.famerp.br. Tissue samples were snap-frozen in liquid nitrogen within 30 min after resection and stored at 80 °C. All samples were evaluated microscopically for the presence of at least 70% of neoplastic tissue before DNA extraction. For the control group, 40 saliva samples from healthy subjects, matched by gender and age were collected at the Barretos Cancer Hospital (Sao Paulo, Brazil). All subjects responded a survey regarding lifestyle, and risk factors for upper aerodigestive tract malignancies, including alcohol and tobacco consumption. Smoking was defined as use of tobacco, chewable or smoked, for at least 1 year continuously. Alcohol use was defined as intake of more than one alcoholic drink per week, for at least 1 year continuously. 2.3. DNA extraction Genomic DNA was isolated from tissue samples using the TRIzol reagent (Invitrogen, Frederick, MD) following manufacturer’s recommendations.
Sodium-bisulphite conversion of 1 lg of DNA was performed using a kit (EpiTectÒ Bisulphite Kit; Qiagen, Valencia, CA), following the manufacturer’s recommendations. Briefly incubation of the target DNA with sodium bisulphite results in conversion of unmethylated cytosine residues into uracil, leaving the methylated cytosines unchanged. Therefore, bisulphite treatment gives rise to different DNA sequences for methylated and unmethylated DNA. The chemistry of cytosine deamination by sodium bisulphite involves three steps: (1) sulphonation; (2) deamination and (3) desulphonation. 2.5. Target gene selection A total of eight genes were selected for the examination of methylation abnormalities. The panel included genes reported as targets for epigenetic silencing in different human cancers. All genes evaluated in this study present tumour suppressor activities and their silencing could contribute to the carcinogenesis process. Among these genes are CCNA1 [15], DAPK [22,28,30,34] and DCC [23,34,35] which are involved in cell cycle control and apoptosis, CDH1 [22,29,36,38] and TIMP3 [22,29] in cell adhesion, MGMT [22,28–32] in DNA repair, HIC1 [23] in cell differentiation and proliferation, p16 [22,28–31,34] in cell-cycle control, and AIM1 (its function is not yet well understood). It has been shown that the expression of these genes may be affected by aberrant promoter methylation in association with transcription silencing in different types of human malignancies. These genes already have been evaluated in other studies and the proposal is to validate them in a different population. Carvalho et al. (2008) showed an elevated frequency of promoter hypermethylation in HNSCC in a panel of gene promoters previously described as methylated in HNSCC as well as other solid tumours [23]. Moreover, Rettori et al. (2013) found that CCNA1, DAPK, MGMT, SFRP1 and TIMP3 were able to distinguish HNSCC tumours from control samples with high specificity (>96%) and sensitivity (21–62%) in a different case/control study [14]. We, thus consider that the evaluation of epigenetic changes, such as DNA methylation, would be useful as a tool for diagnosis, surveillance and prognosis of OSCC.
Please cite this article in press as: Arantes L.M.R.B. et al., Validation of methylation markers for diagnosis of oral cavity cancer, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.01.060
4
L.M.R.B. Arantes et al. / European Journal of Cancer xxx (2015) xxx–xxx
2.6. Steps for gene evaluation The first step involved a screening evaluation, designed to select markers that better distinguish tumours from normal samples. Elimination of targets with high rates of promoter hypermethylation in normal control samples facilitated efficient definition of higher value markers. Significance was based on area under the curve (AUC) from receiving operating characteristic analysis, sensitivity, and specificity of that particular gene in differentiating tumour samples (cases) from saliva (controls) for the determination of a marker panel. 2.7. Quantitative methylation-specific PCR The quantitative methylation-specific PCR analyses (Q-MSP) were conducted as previously described (4). Basically, 30 ng of bisulphite-modified DNA was used as template in fluorogenic Q-MSP assays carried out in a final volume of 20 lL in 96-well plates in the 7900HT Fast Real-Time PCR System (Applied Biosystems). PCR was performed in separate wells for each primer/probe set and each sample was run in triplicate. The final reaction mixture contained 3 lL of bisulphitemodified DNA, 1.2 lmol/L of forward and reverse primers, 200 nmol/L of the probe, 0.5 U of platinum Taq polymerase (Invitrogen), 200 lmol/L dNTPs, 16.6 nmol/L ammonium sulphate, 67 mmol/L Trizma, 6.7 mmol/L magnesium chloride, 10 mmol/L mercaptoethanol, 0.1% DMSO, and 1X ROX dye (Invitrogen). PCR was conducted with the following conditions: 95 °C for 2 min, followed by 45 cycles at 95 °C for 15 s and 60 °C for 1 min. Each plate included patient DNA samples, multiple water blanks and serial dilutions (30–0.0003 ng) of a positive control allowing the construction of calibration curves. Leukocyte DNA obtained from a healthy individual was methylated in vitro using SssI methyltransferase (New England Biolabs) to generate methylated DNA at all CpG to be used as positive control. Primers and probes were obtained from the literature and specifically amplify the promoter regions of the eight genes of interest and the internal control gene, ACTB (Supplementary Table S1). The relative DNA methylation level of the eight genes in each sample was determined as a ratio of methylation specific PCR-amplified gene to ACTB and then multiplied by 100 (average value of triplicates of gene of interest divided by the average value of triplicates of ACTB 100). 2.8. HPV analysis (E7 HPV-16 Type-Specific Real Time PCR) Samples were submitted to a type-specific TaqManbased real time qPCR targeting HPV16 E7. qPCR was
performed in an ABI 7300 Real-Time PCR System (Applied Biosystems, Foster City, CA). All samples and controls were run in duplicate. qPCR was performed using oligonucleotide primers and probe as follows: forward (50 GATGAAATAGATGGTCCAGC30 ) and reverse (50 GCTTTGTACGCACAACC-GAAGC30 ) primers HPV16 E7 type-specific (35) and probe (50 6FAM-CAAGCAGAACCGGACAG-MGB-NFQ30 ), in a final volume reaction of 25 lL. Each qPCR contained, 1 TaqMan master mix (Applied Biosystems, Foster City, CA), 400 nM of each forward and reverse primers, 200 nM of fluorogenic TaqMan probe and 5 lL of extracted DNA. The amplification conditions were: 50 °C for 2 min, 95 °C for 10 min, followed by 40 cycles of 95 °C for 15 s, 55 °C for 1 min and 60 °C for 1 min. qPCR run included the following controls: (i) Caski cell line DNA (harbouring 600 copies/cell of HPV16) and (ii) water as negative control. 2.9. Statistical analysis Statistical analysis was performed using the software SPSS 19.0 for Windows. Hypermethylation of each gene was treated as a binary variable (methylated: any value other than zero; versus unmethylated: 0). Proportions of gene methylation were compared between tumour samples (from cases) and saliva samples (from controls) using either Chi-square or Fisher’s exact test, as appropriate. Sensitivity and specificity of each individual gene in detecting OSCC were calculated along with 95% confidence intervals (95% CI) (Table 2a). We evaluated all possible combinations of the selected markers for both tumour and saliva samples, where a positive panel was defined as at least one gene of the panel being methylated (Table 2b). The AUC, an index of predictive power, was also provided. For all analysis we considered statistical significance when p-value <0.05. 3. Results 3.1. Characteristics of the patients The study population consisted of 40 patients with a confirmed diagnostic of OSCC treated from September 2002 through October 2007. Patients were mainly males (90.0%) and Caucasians (71.8%), with ages ranging from 41 to 78 years (median, 54.5 years). Tobacco or alcohol consumption (current or past) was reported by 72.5% and 65.0% of the patients, respectively. With regard to HPV status, 7 (17.5%) of the tumours were HPV-positive. The primary tumour sites were located in the oral tongue (70.0%), in the floor of mouth (20.0%) and in other oral cavity sites (10.0%). Clinical stage was T1/ T2 in 13 cases (32.5%) and T3/T4 in 27 (67.5%); N0 in 11 cases (27.5%) and N+ in 29 (72.5%); and none of the patients had distant metastasis at diagnosis. Vascular embolisation, lymphatic invasion, perineural invasion
Please cite this article in press as: Arantes L.M.R.B. et al., Validation of methylation markers for diagnosis of oral cavity cancer, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.01.060
L.M.R.B. Arantes et al. / European Journal of Cancer xxx (2015) xxx–xxx
and peritumoural inflammatory infiltrate were present in 2 (5.0%), 18 (45.0%), 20 (50.0%) and 20 (51.3%), respectively of OSCC samples (Table 1). The control population was matched by gender and age, and thus comprised mainly males (90.0%) and Cau-
5
casians (90.0%), with ages ranging from 37 to 80 years (median, 53.5 years). Tobacco or alcohol consumption (current or past) was reported by 35.0% and 80.0% of the subjects, respectively. Regarding HPV status on saliva samples, 100% were tested as negative.
Table 1 Clinical and pathological data of the patients enrolled in the study. Characteristic
Age (Mean, median, <45 years 45–65 years >65 years
Patients (%) CCNA1
range) 5 (12.5) 25 (62.5) 10 (25.0)
DAPK
DCC
Negative Positive
P
TIMP3
Negative Positive
P
Negative Positive
P
Negative Positive
P
4 (80.0) 1 (20.0) 18 (72.0) 7 (28.0) 6 (60.0) 4 (40.0)
0.683 0 (0.0) 5 (20.0) 3 (30.0)
5 (100) 0.392 1 (20.0) 4 (80.0) 0.189 1 (20.0) 20 (80.0) 10 (40.0) 15 (60.0) 5 (20.0) 7 (70.0) 1 (10.0) 9 (90.0) 1 (10.0)
4 (80.0) 0.771 20 (80.0) 9 (90.0)
Gender Male Female
36 (90.0) 4 (10.0)
25 (69.4) 11 (30.6) 1.000 7 (19.4) 3 (75.0) 1 (25.0) 1 (25.0)
29 (80.6) 1.000 11 (30.7) 25 (69.4) 1.000 5 (13.9) 3 (75.0) 1 (25.0) 3 (75.0) 2 (50.0)
31 (86.1) 0.134 2 (50.0)
Race Caucasian Other
28 (71.8) 11 (28.2)
19 (67.9) 9 (32.1) 9 (81.8) 2 (18.2)
0.461 7 (25.0) 1 (9.1)
21 (75.0) 0.400 8 (28.6) 10 (90.9) 4 (36.4)
20 (71.4) 0.709 6 (21.4) 7 (63.6) 1 (9.1)
22 (78.6) 0.649 10 (90.9)
Tobacco consumption Current 29 (72.5) Former 10 (25.0) Never 1 (2.5)
23 (79.3) 6 (20.7) 4 (40.0) 6 (60.0) 1 (100) 0 (0.0)
0.052 3 (10.3) 4 (40.0) 1 (100)
26 (89.7) 0.017 9 (31.0) 6 (60.0) 2 (20.0) 0 (0.0) 1 (100)
20 (69.0) 0.244 4 (13.8) 8 (80.0) 2 (20.0) 0 (0.0) 1 (100)
25 (86.2) 0.081 8 (80.0) 0 (0.0)
Alcohol consumption Yes 26 (65.0) No 14 (35.0)
21 (80.8) 5 (19.2) 7 (50.0) 7 (50.0)
0.071 4 (15.4) 4 (28.6)
22 (84.6) 0.416 8 (30.8) 10 (71.4) 4 (28.6)
18 (69.2) 1.000 4 (15.4) 10 (71.4) 3 (21.4)
22 (84.6) 0.679 11 (78.6)
Clinical tumour status T1–T2 13 (32.5) T3–T4 27 (67.5)
10 (76.9) 3 (23.1) 18 (76.7) 9 (33.3)
0.716 2 (15.4) 6 (22.2)
11 (84.6) 1.000 2 (15.4) 11 (84.6) 0.271 2 (15.4) 21 (77.8) 10 (37.0) 17 (63.0) 5 (18.5)
11 (84.6) 1.000 22 (81.5)
Clinical nodal status N0 11 (27.5) N+ 29 (72.5)
10 (90.9) 1 (9.1) 0.124 2 (18.2) 18 (62.1) 11 (37.9) 6 (20.7)
9 (81.8) 1.000 3 (27.3) 23 (79.3) 9 (31.0)
8 (72.7) 1.000 2 (18.2) 20 (69.0) 5 (17.2)
9 (81.8) 1.000 24 (82.8)
Clinical TNM stage* Initial (I/II) 9 (22.5) Advanced (III/IV) 31 (77.5)
8 (88.9) 1 (11.1) 0.233 2 (22.2) 20 (64.5) 11 (35.5) 6 (19.4)
7 (77.8) 1.000 2 (22.2) 7 (77.8) 0.697 2 (22.2) 25 (80.6) 10 (32.3) 21 (67.7) 5 (16.1)
7 (77.8) 0.645 26 (83.9)
Primary tumour site Tongue 28 (70.0) Floor of mouth 8 (20.0) Others 4 (10.0)
19 (67.9) 9 (32.1) 5 (62.5) 3 (37.5) 4 (100) 0 (0.0)
0.370 6 (21.4) 1 (12.5) 1 (25.0)
22 (78.6) 0.827 9 (32.1) 7 (87.5) 2 (25.0) 3 (75.0) 1 (25.0)
19 (67.9) 0.903 5 (17.9) 6 (75.0) 1 (12.5) 3 &5.0) 1 (25.0)
23 (82.1) 0.862 7 (87.5) 3 (75.0)
HPV status Positive Negative
4 (57.1) 3 (42.9) 24 (72.7) 9 (27.3)
0.410 22 (28.6) 5 (71.4) 0.611 1 (14.3) 6 (85.7) 0.652 0 (0.0) 6 (18.2) 27 (81.8) 11 (33.3) 22 (66.7) 7 (21.2)
7 (100.0) 0.317 26 (78.8)
33 (82.5) 7 (17.5)
Vascular embolisation Positive 2 (5.0) Negative 38 (95.0)
2 (100) 0 (0.0) 1.000 1 (50.0) 26 (68.4) 12 (31.6) 7 (18.4)
1 (50.0) 0.364 1 (50.0) 1 (50.0) 0.515 1 (50.0) 31 (81.6) 11 (28.9) 27 (71.7) 6 (15.8)
1 (50.0) 0.323 32 (84.2)
Lymphatic invasion Positive Negative
18 (46.1) 21 (53.9)
11 (61.1) 7 (38.9) 17 (81.0) 4 (19.0)
0.170 4 (22.2) 4 (19.0)
14 (77.8) 1.000 7 (38.9) 17 (81.0) 5 (23.8)
11 (61.1) 0.309 4 (22.2) 16 (76.2) 3 (14.3)
14 (77.8) 0.682 18 (85.7)
Perineural invasion Positive Negative
20 (50.0) 20 (50.0)
13 (65.0) 7 (35.0) 0.490 3 (15.0) 15 (75.0) 5 (25.0.0 5 (25.0)
17 (85.0) 0.695 7 (35.0) 15 (75.0) 5 (25.0)
13 (65.0) 0.490 3 (15.0) 15 (75.0) 4 (20.0)
17 (85.0) 1.000 16 (80.0)
16 (80.0) 1.000 8 (40.0) 15 (78.9) 4 (21.1)
12 (60.0) 0.301 3 (15.0) 15 (78.9) 4 (21.1)
17 (85.0) 0.695 15 (78.9)
Peritumoral inflammatory infiltrate Intense/Moderate 20 (51.3) 14 (70.0) 6 (30.0) Scarce/Absent 19 (48.7) 14 (73.7) 5 (26.3) *
1.000 4 (20.0) 4 (21.1)
Clinical stage according to TNM Classification of Malignant Tumours – 6th ed.
Please cite this article in press as: Arantes L.M.R.B. et al., Validation of methylation markers for diagnosis of oral cavity cancer, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.01.060
6
L.M.R.B. Arantes et al. / European Journal of Cancer xxx (2015) xxx–xxx
3.2. Promoter hypermethylation of AIM1, CCNA1, CDH1, DAPK, DCC, HIC1, MGMT, and TIMP3 The promoter methylation status of eight genes was tested in tumour samples from 40 OSCC patients. Promoter methylation was found in 17.5% at AIM1, 30.0% at CCNA1, 95.0% at CDH1, 80.0% at DAPK, 70.0% at DCC, 95.0% at HIC1, 30.0% at MGMT and 82.5% at TIMP3, of OSCC cases. The methylation status of the same genes was evaluated in 40 normal saliva from controls, and the specificity levels varied from 0% to 100%. These analyses revealed that hypermethylation of AIM1 and MGMT was not frequent in head and neck tumour, thus showing low sensitivity, while CDH1 and HIC1 were frequently methylated in saliva controls, indicating a low specificity (Table 2a, Figs. 1 and 2). 3.3. Promoter hypermethylation of the panel Excluding genes with low sensitivity and/or specificity such as AIM1, CDH1, HIC1 and MGMT (AUC < 0.6), several combinations were performed as a panel, where a positive panel was defined as at least one gene being methylated in the sample. In the remaining four genes: CCNA1, DAPK, DCC and TIMP3 (AUC > 0.6), the hypermethylation detection rate in tumours varied from 77.5% to 95.0% of sensitivity and specificity rates ranged from 75.0% to 97.5%, depending on the tested panel (Table 2b and Fig. 2). These four genes had highest value for the full panel: 95.0% of sensitivity and AUC = 0.925. 3.4. Association between aberrant methylation and patient characteristics The methylation pattern of the four selected genes was analysed for potential associations with clinical and pathological characteristics of OSCC patients, including age, gender, race, tobacco use, alcohol consumption, primary tumour site, T stage, N stage, vascular embolisation, lymphatic invasion, perineural invasion and peritumoural inflammatory infiltrate. Regarding DAPK, 89.7% of patients harbouring hypermethylation were
current smokers (p = 0.017). No significant associations were observed between the methylation profile of DCC and TIMP3 and clinical-pathological characteristics of OSCC patients (Table 1). 4. Discussion The progression of cancer may be associated with changes in DNA methylation that can be used as biomarkers for diagnosis, treatment selection and prognosis. Several studies indicate that abnormal DNA methylation may occur early in cancer development, including HNSCC [13,16,23,29,38]. Activation of proto-oncogenes and inactivation of tumour suppressor genes are the major genetic alterations involved in carcinogenesis. In addition, epigenetic changes (altered patterns of gene expression that do not affect the primary sequence of the DNA) can alter the expression of genes important in the development of a variety of tumours [39]. The use of aberrant promoter hypermethylation detection markers of tumour-specific cells in different materials from OSCC such as body fluids, exfoliated cells and solid tumours, has been proposed. The present study evaluated a population of OSCC patients and matched controls to determine the ability of Q-MSP to detect promoter methylation of specific genes that could distinguish these groups. The sensitivity of the Q-MSP technique allows the detection of the presence of methylated alleles in a background of normal at a threshold of 1/1000–1/10,000 [23]. Therefore, this strategy allowed us to define highly tumour-specific methylated genes that were rarely or never present in saliva controls. It is noteworthy that the control population in the current study can be considered as at high risk for OSCC with the majority of them being males, with median age 55 years and with reported regular consumption of tobacco and alcohol. Given that the association between clinical and demographic data and the gene methylation patterns of the genes evaluated did not show relevant statistically significance, we can hypothesise that these factors do not significantly
Table 2a Comparison of hypermethylation detection on tumour samples (Oral squamous cell carcinoma (OSCC) patients) and saliva (controls). Individual gene evaluation
Case (n)
Control (n)
Sensitivity % (95% confidence interval (CI))
Specificity % (95% CI)
Area under the curve (AUC)
TIMP3 DCC DAPK CCNA1 AIM1 MGMT CDH1 HIC1
40 40 40 40 40 40 40 40
40 40 40 40 40 40 40 40
82.5 70.0 80.0 30.0 17.5 30.0 95.0 95.0
100.0 (0.91–1.00) 95.0 (0.83–0.98) 82.5 (0.68–0.91) 100.0 (0.91–1.00) 100.0 (0.91–1.00) 85.0 (0.70–0.92) 20.0 (0.10–0.34) 0.0 (0.00–0.08)
0.913 0.825 0.813 0.638 0.587 0.575 0.575 0.475
(0.68–0.91) (0.54–0.81) (0.65–0.89) (0.18–0.45) (0.08–0.31) (0.18–0.45) (0.83–0.98) (0.83–0.98)
Please cite this article in press as: Arantes L.M.R.B. et al., Validation of methylation markers for diagnosis of oral cavity cancer, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.01.060
L.M.R.B. Arantes et al. / European Journal of Cancer xxx (2015) xxx–xxx
7
Fig. 1. Methylation rates of AIM1, CCNA1, CDH1, DAPK, DCC, HIC1, MGMT and TIMP3 in oral squamous cell carcinoma (OSCC) tissue samples and saliva from controls. Each symbol represents a different sample. X-axis, proportion of methylated cases/tested cases for each sample type; Y-axis, quantity of hypermethylation (gene of interest/ACTB 100).
change the specificity and sensitivity levels of these genes in distinguishing cases from controls. Finally, the high sensitivity of Q-MSP may have allowed the detection of low quantity methylation even in a subset of healthy controls. The search for biomarkers has as its main goal to evaluate and measure the status of normal
and pathological biological processes as well as pharmacological responses to certain treatments. The searching for these biomarkers is an important step for the identification of individuals in the early stages of head and neck cancer for its diagnostic and prognostic relevance. Therefore, assuming that cancer results from genetic and epigenetic changes, analyses based on the
Please cite this article in press as: Arantes L.M.R.B. et al., Validation of methylation markers for diagnosis of oral cavity cancer, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.01.060
8
L.M.R.B. Arantes et al. / European Journal of Cancer xxx (2015) xxx–xxx
Fig. 2. ROC curve for CCNAI, DAPK, DCC and TIMP3 and selected panels from oral squamous cell carcinoma (OSCC) tissue samples and saliva from controls.
Table 2b Comparison of best combination of genes for hypermethylation detection on tumour samples (oral squamous cell carcinoma (OSCC) patients) and saliva (controls). Panel
Case (n)
Control (n)
Sensitivity % (95% confidence interval (CI))
Specificity % (95% CI)
Area under the curve (AUC)
CCNA1 + DCC + TIMP3 DCC + TIMP3 CCNA1 + TIMP3 DAPK + DCC CCNA1 + DAPK + DCC + TIMP3 CCNA1 + DAPK + DCC DAPK + DCC + TIMP3 CCNA1 + DCC DAPK + TIMP3 CCNA1 + DAPK + TIMP3 CCNA1 + DAPK
40 40 40 40 40 40 40 40 40 40 40
40 40 40 40 40 40 40 40 40 40 40
92.5 90.0 85.0 95.0 95.0 95.0 92.5 77.5 85.0 87.5 85.0
92.5 95.0 97.5 77.5 75.0 75.0 77.5 90.0 82.5 80.0 80.0
0.925 0.925 0.913 0.862 0.850 0.850 0.850 0.838 0.838 0.837 0.825
methylation profile in combination with the pathological diagnosis would be useful in predicting the behaviour of these head and neck cancers. We were able to show the possibility of using single genes for tumour detection; however, the use of a com-
(0.80–0.97) (0.76–0.96) (0.70–0.92) (0.8–0.98) (0.83–0.98) (0.83–0.98) (0.80–0.97) (0.77–0.87) (0.70–0.92) (0.73–0.94) (0.70–0.92)
(0.80–0.97) (0.83–0.98) (0.87–0.99) (0.62–0.87) (0.59–0.85) (0.59–0.85) (0.62–0.87) (0.76–0.96) (0.68–0.91) (0.65–0.89) (0.65–0.89)
bination of genes (CCNA1, DCC, TIMP3 and DAPK) in a panel provided improvement in sensitivity rates. Since variations in the patterns of promoter hypermethylation in individual tumours may occur depending on alterations on specific molecular pathways, the use
Please cite this article in press as: Arantes L.M.R.B. et al., Validation of methylation markers for diagnosis of oral cavity cancer, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.01.060
L.M.R.B. Arantes et al. / European Journal of Cancer xxx (2015) xxx–xxx
of multiple genes may provide greater applicability and coverage for different tumours when compared with a single gene for general detection, corroborating to results obtained by Carvalho et al. [23] and Rettori et al. [14] in different cohorts. From the initial screening of eight genes, four genes were selected as part of a panel to distinguish OSCC patients and healthy controls. The possible combination of those four genes was able to provide a sensitivity ranging from 77.5% to 95.0% and a specificity ranging from 75.0% to 97.5%. Appropriately matched normal controls are of paramount importance for the adequate assessment of the utility of promoter hypermethylation in OSCC. In addition, the presence of promoter hypermethylation of tumour suppressor genes observed in control populations can happen as a random or a physiologic event associated with tissue-specificity or related to age or environmental carcinogenic exposures. These factors significantly affect the selection of a healthy control group, as limited size control groups comprised of young donors, with minimal tobacco and ethanol exposure can bias reporting of falsely elevated specificity for candidate genes [23]. In general, we were able to validate a panel for OSCC detection with high specificity and high sensitivity. Such characteristics would be highly useful as a method for diagnostic and screening, other panels were defined as harbouring high specificity but accompanied by a low sensitivity, as well as high sensitivity and low specificity which may have potential as prognostic markers during surveillance after an OSCC treatment or in high-risk populations. The lack of association between the methylation status of these markers and clinical characteristics at diagnosis shows that they are able to distinguish tumour cases irrespective of gender, age, HPV status, clinical stage, vascular embolisation, lymphatic invasion, perineural invasion and peritumoral inflammatory infiltrate. Moreover, DAPK, DCC and TIMP3 were found hypermethylated in nearly 90% of clinically T1 and T2 cases, showing the feasibility of using these markers in early diagnosis.
5. Conclusion The search for biomarkers is important for the identification of individuals with OSCC and ideally in the early stages decreasing morbidity and increasing survival rates. Body fluids such as saliva, which can be obtained by non-invasive techniques and can be used to evaluate gene methylation profiles, are potential sources for the development of biomarkers for diagnosis, surveillance and prognosis. The combination of hypermethylation pattern for CCNA1, DCC, TIMP3 and DAPK genes improved detection when compared with single markers, reaching values of 92.5% for sensitivity and specificity (when using the panel CCNA1,
9
DCC, TIMP3). Moreover, DAPK, DCC and TIMP3 are hypermethylated in nearly 90% of clinically T1 and T2 cases, showing the feasibility of using these markers in early diagnosis. Conflict of interest statement None declared. Acknowledgments The authors would like to thank the members of the GENCAPO (Head and Neck Genome) Project and Fundac¸a˜o de Amparo a` Pesquisa do Estado de Sa˜o Paulo (FAPESP). This study was supported by FAPESP (10/51168-0). L.M.R.B.A. is recipient of scholarship from Fundac¸a˜o de Amparo a` Pesquisa do Estado de Sa˜o Paulo (FAPESP; grant number 2011/02864-7). A.L.C. has a National Counsel of Technological and Scientific Development Scholarship (CNPq). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10. 1016/j.ejca.2015.01.060. References [1] Kim L, King T, Agulnik M. Head and neck cancer: changing epidemiology and public health implications. Oncology 2010;24(10):915–9, 924. [2] Marcu LG, Yeoh E. A review of risk factors and genetic alterations in head and neck carcinogenesis and implications for current and future approaches to treatment. J Cancer Res Clin Oncol 2009;135(10):1303–14. [3] Ragin CC, Modugno F, Gollin SM. The epidemiology and risk factors of head and neck cancer: a focus on human papillomavirus. J Dent Res 2007;86(2):104–14. [4] Wang Q, Gao P, Wang X, Duan Y. Investigation and identification of potential biomarkers in human saliva for the early diagnosis of oral squamous cell carcinoma. Clin Chim Acta 2014;427:79–85. [5] Montebugnoli L, Gissi DB, Flamminio F, Gentile L, Dallera V, Leonardi E, et al. Clinicopathologic parameters related to recurrence and locoregional metastasis in 180 oral squamous cell carcinomas. Int J Surg Pathol 2014;22(1):55–62. [6] Grobe A, Blessmann M, Hanken H, Friedrich RE, Schon G, Wikner J, et al. Prognostic relevance of circulating tumor cells in blood and disseminated tumor cells in bone marrow of patients with squamous cell carcinoma of the oral cavity. Clin Cancer Res 2014;20(2):425–33. [7] Leemans CR, Braakhuis BJ, Brakenhoff RH. The molecular biology of head and neck cancer. Nat Rev Cancer 2011;11(1):9–22. [8] Olivieri EH, da Silva SD, Mendonca FF, Urata YN, Vidal DO, Faria Mde A. CYP1A2*1C, CYP2E1*5B, and GSTM1 polymorphisms are predictors of risk and poor outcome in head and neck squamous cell carcinoma patients. Oral Oncol 2009;45(9):e73–9. [9] Markopoulou S, Nikolaidis G, Liloglou T. DNA methylation biomarkers in biological fluids for early detection of respiratory tract cancer. Clin Chem Lab Med 2012;50(10):1723–31.
Please cite this article in press as: Arantes L.M.R.B. et al., Validation of methylation markers for diagnosis of oral cavity cancer, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.01.060
10
L.M.R.B. Arantes et al. / European Journal of Cancer xxx (2015) xxx–xxx
[10] Leng S, Liu Y, Thomas CL, Gauderman WJ, Picchi MA, Bruse SE, et al. Native American ancestry affects the risk for gene methylation in the lungs of Hispanic smokers from New Mexico. Am J Respir Crit Care Med 2013;188(9):1110–6. [11] Garcia-Baquero R, Puerta P, Beltran M, Alvarez M, Sacristan R, Alvarez-Ossorio JL, et al. Methylation of a novel panel of tumor suppressor genes in urine moves forward noninvasive diagnosis and prognosis of bladder cancer: a 2-center prospective study. J Urol 2013;190(2):723–30. [12] Zuiverloon TC, Beukers W, van der Keur KA, Nieuweboer AJ, Reinert T, Dyrskjot L, et al. Combinations of urinary biomarkers for surveillance of patients with incident nonmuscle invasive bladder cancer: the European FP7 UROMOL project. J Urol 2013;189(5):1945–51. [13] Matthews AM, Kaur H, Dodd M, D’Souza J, Liloglou T, Shaw RJ, et al. Saliva collection methods for DNA biomarker analysis in oral cancer patients. Br J Oral Maxillofac Surg 2013;51(5):394–8. [14] Rettori MM, de Carvalho AC, Bomfim Longo AL, de Oliveira CZ, Kowalski LP, Carvalho AL, et al. Prognostic significance of TIMP3 hypermethylation in post-treatment salivary rinse from head and neck squamous cell carcinoma patients. Carcinogenesis 2013;34(1):20–7. [15] Sun W, Zaboli D, Liu Y, Arnaoutakis D, Khan T, Wang H, et al. Comparison of promoter hypermethylation pattern in salivary rinses collected with and without an exfoliating brush from patients with HNSCC. PLoS One 2012;7(3):e33642. [16] Carvalho AL, Henrique R, Jeronimo C, Nayak CS, Reddy AN, Hoque MO, et al. Detection of promoter hypermethylation in salivary rinses as a biomarker for head and neck squamous cell carcinoma surveillance. Clin Cancer Res 2011;17(14):4782–9. [17] Twelves D, Nerurkar A, Osin P, Dexter T, Ward A, Gui GP, et al. DNA promoter hypermethylation profiles in breast duct fluid. Breast Cancer Res Treat 2013;139(2):341–50. [18] Muraoka T, Soh J, Toyooka S, Aoe K, Fujimoto N, Hashida S, et al. The degree of microRNA-34b/c methylation in serumcirculating DNA is associated with malignant pleural mesothelioma. Lung Cancer 2013;82(3):485–90. [19] Zhao ZH, Fan YC, Yang Y, Wang K. Association between Ras association domain family 1A promoter methylation and hepatocellular carcinoma: a meta-analysis. World J Gastroenterol 2013;19(41):7189–96. [20] Huang YT, Li FF, Ke C, Li Z, Li ZT, Zou XF, et al. PTPRO promoter methylation is predictive of poorer outcome for HER2positive breast cancer: indication for personalized therapy. J Transl Med 2013;11:245. [21] Rosas SL, Koch W, da Costa Carvalho MG, Wu L, Califano J, Westra W, et al. Promoter hypermethylation patterns of p16, O6methylguanine-DNA-methyltransferase, and death-associated protein kinase in tumors and saliva of head and neck cancer patients. Cancer Res 2001;61(3):939–42. [22] Righini CA, de Fraipont F, Timsit JF, Faure C, Brambilla E, Reyt E, et al. Tumor-specific methylation in saliva: a promising biomarker for early detection of head and neck cancer recurrence. Clin Cancer Res 2007;13(4):1179–85. [23] Carvalho AL, Jeronimo C, Kim MM, Henrique R, Zhang Z, Hoque MO, et al. Evaluation of promoter hypermethylation detection in body fluids as a screening/diagnosis tool for head and neck squamous cell carcinoma. Clin Cancer Res 2008;14(1):97–107. [24] Arantes LM, de Carvalho AC, Melendez ME, Carvalho AL, Goloni-Bertollo EM. Methylation as a biomarker for head and neck cancer. Oral Oncol 2014;50(6):587–92.
[25] Brinkmann O, Kastratovic DA, Dimitrijevic MV, Konstantinovic VS, Jelovac DB, Antic J, et al. Oral squamous cell carcinoma detection by salivary biomarkers in a Serbian population. Oral Oncol 2011;47(1):51–5. [26] Park NJ, Zhou H, Elashoff D, Henson BS, Kastratovic DA, Abemayor E, et al. Salivary microRNA: discovery, characterization, and clinical utility for oral cancer detection. Clin Cancer Res 2009;15(17):5473–7. [27] Li Y, Zhou X, St John MA, Wong DT. RNA profiling of cell-free saliva using microarray technology. J Dent Res 2004;83(3):199–203. [28] Pierini S, Jordanov SH, Mitkova AV, Chalakov IJ, Melnicharov MB, Kunev KV, et al. Promoter hypermethylation of CDKN2A, MGMT, MLH1, and DAPK genes in laryngeal squamous cell carcinoma and their associations with clinical profiles of the patients. Head Neck 2013;36(8):1103–8. [29] Shaw RJ, Hall GL, Lowe D, Bowers NL, Liloglou T, Field JK, et al. CpG island methylation phenotype (CIMP) in oral cancer: associated with a marked inflammatory response and less aggressive tumour biology. Oral Oncol 2007;43(9):878–86. [30] Dikshit RP, Gillio-Tos A, Brennan P, De Marco L, Fiano V, Martinez-Penuela JM, et al. Hypermethylation, risk factors, clinical characteristics, and survival in 235 patients with laryngeal and hypopharyngeal cancers. Cancer 2007;110(8):1745–51. [31] Kato K, Hara A, Kuno T, Mori H, Yamashita T, Toida M, et al. Aberrant promoter hypermethylation of p16 and MGMT genes in oral squamous cell carcinomas and the surrounding normal mucosa. J Cancer Res Clin Oncol 2006;132(11):735–43. [32] Zuo C, Ai L, Ratliff P, Suen JY, Hanna E, Brent TP, et al. O6methylguanine-DNA methyltransferase gene: epigenetic silencing and prognostic value in head and neck squamous cell carcinoma. Cancer Epidemiol Biomarkers Prev 2004;13(6):967–75. [33] Demokan S, Chuang A, Suoglu Y, Ulusan M, Yalniz Z, Califano JA, et al. Promoter methylation and loss of p16(INK4a) gene expression in head and neck cancer. Head Neck 2012;34(10):1470–5. [34] Hasegawa M, Nelson HH, Peters E, Ringstrom E, Posner M, Kelsey KT. Patterns of gene promoter methylation in squamous cell cancer of the head and neck. Oncogene 2002;21(27):4231–6. [35] Brait M, Loyo M, Rosenbaum E, Ostrow KL, Markova A, Papagerakis S, et al. Correlation between BRAF mutation and promoter methylation of TIMP3, RARbeta2 and RASSF1A in thyroid cancer. Epigenetics 2012;7(7):710–9. [36] Marsit CJ, Posner MR, McClean MD, Kelsey KT. Hypermethylation of E-cadherin is an independent predictor of improved survival in head and neck squamous cell carcinoma. Cancer 2008;113(7):1566–71. [37] de Moraes RV, Oliveira DT, Landman G, de Carvalho F, Caballero O, Nonogaki S, et al. E-cadherin abnormalities resulting from CPG methylation promoter in metastatic and nonmetastatic oral cancer. Head Neck 2008;30(1):85–92. [38] Shaw RJ, Liloglou T, Rogers SN, Brown JS, Vaughan ED, Lowe D, et al. Promoter methylation of P16, RARbeta, E-cadherin, cyclin A1 and cytoglobin in oral cancer: quantitative evaluation using pyrosequencing. Br J Cancer 2006;94(4):561–8. [39] Sweeney C, Boucher KM, Samowitz WS, Wolff RK, Albertsen H, Curtin K, et al. Oncogenetic tree model of somatic mutations and DNA methylation in colon tumors. Genes Chromosomes Cancer 2009;48(1):1–9.
Please cite this article in press as: Arantes L.M.R.B. et al., Validation of methylation markers for diagnosis of oral cavity cancer, Eur J Cancer (2015), http://dx.doi.org/10.1016/j.ejca.2015.01.060