- Email: [email protected]
Does loss of heterozygosity in critical genome regions predict a local relapse in patients after laryngectomy?
Mutation Research 600 (2006) 67–76
Does loss of heterozygosity in critical genome regions predict a local relapse in patients after laryngectomy? Katarzyna Szukała a,∗ , Anna Sowi´nska b , Małgorzata Wierzbicka d , Wiesława Biczysko c , Witold Szyfter d , Krzysztof Szyfter a,d a
Institute of Human Genetics, Polish Academy of Sciences, Pozna´n, Poland Department of Biochemistry and Molecular Biology, Medical University of Pozna´n, Pozna´n, Poland c Department of Clinical Pathology, Medical University of Pozna´ n, Pozna´n, Poland Department of Otolaryngology and Laryngeal Oncology, Medical University of Pozna´n, Pozna´n, Poland b
d
Available online 11 July 2006
Abstract Background: Patients, who had an upper aerodigestive tract malignancy, have a high incidence of succeeding tumor development. This has been attributed to the role of “field cancerization” in carcinogenesis. The aim of this study was analysis of loss of heterozygosity (LOH) in the regions frequently lost during the course of head and neck squamous cell carcinomas (HNSCC), especially at early stages, which could answer the clinicians’ question, if LOH analysis has any “predictive” value in relation to tumor occurrence. Material and methods: Sixty-five larynx cancer patients were examined for loss of heterozygosity on 3p, 7q, 8p, 9p and 18q chromosomal arms with the use of 12 microsatellite markers. The material from a single patient consisted of blood, tumor, safe margin and one or two clinically unchanged mucosal samples. During follow up, the material from brush specimens (14 patients) as well as laryngeal swabs (4 patients) was also examined. Results: The highest frequency of LOH was detected for marker D3S1234 in tumor tissues (29%). Analysis of margin samples (b) revealed low LOH frequencies (2–5%) and complete retention of heterozygosity for markers: D3S1234, D7S486, D8S261, D8S264, D9S171 and D18S46. Similarly, for normal appearing mucosa from upper part of larynx (c) frequencies of LOH were low (2–6%), with the complete retention of heterozygosity for markers: D3S1284, D3S1304, D3S1234, D8S264 and D9S1870. We did not detect any LOH in the material of normal appearing mucosa from tracheostoma region (d). During follow up, LOH was detected for eight markers, with the highest incidence for markers D18S46 (six cases), D7S486 (four cases) and D3S1300 (three cases). Conclusions: The data, obtained during this investigation, did not reveal the predictive value of LOH with respect to local relapse occurrence in laryngeal cancer patients. However, time of follow up did not reach 5 years, so that further clinical monitoring should be conducted. © 2006 Elsevier B.V. All rights reserved. Keywords: Larynx cancer; Loss of heterozygosity; Microsatellite markers; Follow up; Local relapse
1. Introduction
∗
Corresponding author at: Institute of Human Genetics, Polish Academy of Science, ul.Strzeszy´nska 32, 60-479 Pozna´n, Poland. Tel.: +48 61 6579214; fax: +48 61 8233235. E-mail address: [email protected] (K. Szukała). 0027-5107/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.mrfmmm.2006.05.027
Head and neck squamous cell carcinomas (HNSCC) constitute a significant medical and social problem, with the occurrence of around 5% of all worldwide-diagnosed malignancies. The difficulties in studying the progression of larynx cancer are connected with its complex
68
K. Szukała et al. / Mutation Research 600 (2006) 67–76
biology, genetics and with a relatively late clinical detection of primary tumors. As the result, local recurrences, lymph node relapses, distant metastases and second primary tumors appear very often [1,2]. Still, in spite of advanced genetic studies and improvements in diagnostics and treatment of HNSCC the survival rates are rather poor, ranging from 40 to 50% [3]. Treatment failures may partially be explained by anatomical predisposition of head and neck region organs—the epithelium of upper respiratory tract, exposed for a prolonged carcinogens activity (in tobacco, alcohol, air, preservatives) is being continuously damaged (the “field cancerization” theory) [4]. Local recurrences can develop within primary tumor resection site, even when surgical margins were histologically tumor-free [5]. These features extorted the investigators to study not only the material derived from tumor, but also from clinically unchanged mucosa resected within the safe margin from the operating field. It is possible that normal-appearing mucosa left after tumor and margin excision may be genetically altered [1,6]. Unfortunately, it means that “field cancerization” process is much more extensive than clinical and histological examinations show [7]. It was estimated, that a cell undertakes an irreversible malignant transformation process after acquiring 6–10 genetic alterations [8]. These alterations concern activation of oncogenes as well as inactivation of tumor suppressor genes, as it was revealed by cytogenetic and molecular studies [9,10]. Loss of heterozygosity (LOH) has been recognized to be an important event in HNSCC, suggesting involvement of the suppressor pathway in the genesis of this disease—LOH is often one of two inactivation steps of tumor suppressor genes (TSGs) [11]. Multiple analyses with the application of microsatellite markers provided the knowledge about chromosomal regions containing critical tumor suppressor genes that are frequently lost or inactivated in HNSCC, like p16 (9p21), p53 (17p13.1), FHIT (3p14.2), VHL (3p25.3), Rb1 (13q14) and ING1 (13q34) [9,10]. Moreover, there are numerous regions lost during HNSCC pathogenesis, where putative TSGs were not identified so far (i.e. 8p arm). Microsatellite analysis appears to have a potential application in predicting cancer risk, as shown by studies conducted previously on oral leukoplakias exhibited LOH at chromosomal regions 3p14 and/or 9p21 to carry higher risk for HNSCC development [12]. In the present study, we investigated LOH in the regions frequently lost during the course of HNSCC, harboring known or putative tumor suppressor loci in the material derived from patients treated surgically for larynx cancer. The material consisted of tumor, safe mar-
gin and samples of normal appearing mucosa taken from distant, opposite points of surgical specimens. The diversity among clinical material should enable a comparison of genetic alterations in the same patient in various localizations (glottis, trachea). Additionally, during follow up, laryngeal swabs from hypopharynx and brush specimens from tracheostoma were collected to study LOH more than 4 years after surgery. The markers were chosen with the emphasis on localization in regions lost “early” during neoplastic transformation [13,14]. This experiment may answer the clinicians’ question, if LOH analysis has any “predictive” value in relation to tumor occurrence. 2. Patients and methods 2.1. Samples collection Altogether 65 larynx cancer patients (62 men and 3 women) were examined. Material from surgical specimens was obtained from the Department of Otolaryngology and Laryngeal Oncology, K. Marcinkowski Medical University, Pozna´n, Poland, and was kept frozen at −80 ◦ C right after the surgery till laboratory studies. The detailed data were described previously [15]. In brief, the material from one patient consisted of four or five samples. In cases of total laryngectomy (46 patients), five samples were taken: • • • •
blood as a reference material; tumor (marked as “a”); safe margin from the operating field (“b”); two samples from distant, opposite points of the surgical specimen—clinically unchanged mucosa from the epiglottis or hypopharynx (proximal point of specimen called “c”) and normal mucosa of the tracheal ring (distal point of specimen called “d”).
In cases of partial laryngectomy (19 patients), four samples were taken: • • • •
blood as a reference material; tumor (marked as “a”); safe margin from the operating field (“b”); normal appearing mucosa from the upper part of the larynx (proximal point of specimen called “c”).
Tumor and safe margin tissues were verified histopathologically by microscopy. Normal mucosa was not verified by histology because of unchanged clinical character and a distance from the tumor exceeding 2 cm. Additionally, during the last follow up visit (June 2005), brush specimens (marked as “z”) were taken from 14 patients after total laryngectomy from tracheostomy and laryngeal swabs (marked as “s”) were taken from hypopharynx from 4 patients after partial laryngectomy.
K. Szukała et al. / Mutation Research 600 (2006) 67–76
69
cycles of 95 ◦ C for 30 s, 54–62 ◦ C for 30 s and 72 ◦ C for 45 s and a final elongation step of 72 ◦ C for 10 min. PCR products were mixed with loading dye and internal sizers 100 and 300 bp (Amersham Pharmacia Biotech), then denaturated for 5 min in 95 ◦ C, quickly chilled on ice and then separated with the use of ALF Express II System (Amersham Pharmacia Biotech). LOH was calculated from the measured peak heights, by dividing the allele ratio in analyzed tissue DNA by the allele ratio in constitutional DNA: [t1 /t2 ]/[n1 /n2 ]. Loss of heterozygosity was defined as a more than 50% reduction in the peak height of the clinical tissue compared to the peak height of corresponding blood tissue.
Patients have been followed-up once a month since the surgery till June 2005. 2.2. DNA extraction and LOH analysis DNA was isolated from all samples according to the standard procedures (proteinase K digestion, phenol/chloroform extraction and ethanol precipitation). The DNA elution of brush specimens and laryngeal swabs was conducted with the use of 0.2 M NaOH and 0.04 M Tris–HCl. DNA was PCR amplified with the use of primers, specific for chosen microsatellite markers (dinucleotide repeats). One primer from each primer pair was fluorescently end-labeled with Cy-5. Primers were obtained from http://www.oligo.pl/—DNA Sequencing and Synthesis IBB PAN, Warsaw, Poland. Sequences of all primers, available in Genome Database (http://www.gdbwww.gdb.org/) and Ensembl Genome Browser (http://www.ensembl.org/) are listed in Table 1. PCR reactions were carried out in a total reaction volume of 30 l containing ∼100 ng of genomic DNA, 15 pmol of each primer, 6 nmol of each dNTP (Roche, Mannheim, Germany), 1xPCR buffer [10 mM Tris–HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl2 ], 1U Taq DNA polymerase (Sigma). The mixtures were amplified with a PTC-200 Peltier Thermal Cycler, MJ Research. The PCR profile consisted of initial denaturation step (95 ◦ C for 5 min), followed by 35
2.3. Microsatellite markers Twelve microsatellite markers, used for LOH analysis were mapped to the regions harboring known or putative tumor suppressor genes as follows: 3p13 (D3S1284), 3p14.2 (D3S1300, D3S1234), 3p25.3-p26.1 (D3S1038, D3S1304), 3p21.31 (D3S1568), 7q31.2 (D7S486), 8p22-23.3 (D8S261, D8S264), 9p21 (D9S171, D9S1870) and 18q21.1 (D18S46). These regions were previously shown to be frequently lost in head and neck tumors. The detailed data about chosen microsatellites are listed in Table 1.
Table 1 Data concerning microsatellite markers used in the study Marker
Locus
Gene
Tm (◦ C)
Het (%)
Length of sequence (bp)
Primers sequences
D3S1284
3p13
DUTT1
61
75
155–177
F GCCTTGGGGGTAAATACTCT R GGAATTACAGGCCACTGCTC
D3S1304
3p26.1
VHL
59
80
253–269
F TTCGCTCTTTGATAGGC R ATTTCATTTGTAATTTACTAGCAG
D3S1038
3p25.3
VHL
58
80
115
F TCCAGTAAGAGGCTTCCTAG R AAAGGGGTTCAGGAAACCTG
D3S1300
3p14.2
FHIT
59
83
217–241
F AGCTCACATTCTAGTCAGCCT R GCCAATTCCCCAGATG
D3S1234
3p14.2
FHIT
58
66
99–125
F CCTGTGAGACAAAGCAAGAC R GACATTAGGCACAGGGCTAA
D3S1568
3p21.31
RASSF1
62
87
276–296
F CCATGAACAGAACCTCCCTA R CCGCTGTCCTGCTGTAAG
D7S486
7q31.2
ING3
54
80
114–146
F AAAGGCCAATGGTATATCCC R GCCCAGGTGATTGATAGTGC
D8S261
8p22
?
62
78
128–144
F TGCCACTGTCTTGAAAATCC R TATGGCCCAGCAATGTGTAT
D8S264
8p23.3
?
59
85
121–145
F ACATCTGCGTCGTCTTCATA R CCAACACCTGAGTCAGCATA
D9S171
9p21
p16
59
80
159–177
F AGCTAAGTGAACCTCATCTCTGTCT R ACCCTAGCACTGATGGTATAGTCT
D9S1870
9p21
p16
61
66
179–122
F TGGGTATGGTTTTCTGG R TTGAGGCAGGTCAAATAA
D18S46
18q21.1
MADH4
56
80
129–153
F GAATAGCAGGACCTATCAAAGAGC R CAGATTAAGTGAAAACAGCATATGG
70
K. Szukała et al. / Mutation Research 600 (2006) 67–76
For analyzing brush specimens (z) and laryngeal swabs (s), the same markers were tested, except for two of them: D3S1038 and D3S1234.
3. Results 3.1. LOH analysis in the surgical material In all cases, DNA was successfully amplified. Unfortunately, for two patients (35 and 38), DNA from “c” specimen has been already finished, so that we have no result in sample 35c for markers, D3S1038 and D7S486, as well as we do not have the results in sample 38c for markers D3S1304, D3S1234, D7S486 and D18S46. The schematic representation of LOH distribution in particular tissues for all tested markers is performed in Table 2. The summarized results of microsatellite analysis are shown in Table 3. Only the informative cases (heterozygous at the examined locus) were considered. All cases were informative for at least one marker. The highest number of informative cases was detected for marker D9S1870 (78%), while for markers D3S1234 and D18S46 only 54% of cases were heterozygotes. Considering LOH in particular tissues, only in tumor tissues LOH was observed for all markers. The highest frequency of LOH in tumor tissues was detected for marker D3S1234 (29%) and the lowest frequency was observed for marker D8S264 (9%). LOH in the remaining tissues was detected less frequent. Analysis of margin samples (b) revealed low LOH frequencies (2–5%) and complete retention of heterozygosity for markers: D3S1234, D7S486, D8S261, D8S264, D9S171 and D18S46. The situation was similar for normal appearing mucosa from upper part of larynx (c)—frequencies of LOH were low (2–6%), with the complete retention of heterozygosity for markers: D3S1284, D3S1304, D3S1234, D8S264 and D9S1870. We did not detect any LOH in the material of normal appearing mucosa from tracheostoma region (d). Beside LOH we found microsatellite instability (MSI) in two cases: in case no. 14, we detected MSI in tumor sample (a) for markers D3S1038, D3S1300, D3S1568, D7S486 and D9S171. In margin sample (b), we detected MSI for marker D9S171. In case no. 38, we detected MSI in the normal appearing mucosa sample (c) for marker D9S171. 3.2. Follow up study—LOH analysis in brush specimens and laryngeal swabs The follow up observation period ranged from 3 to 55 months (mean 31 months). For three cases, nos. 23,
25 and 64, data concerning follow up were not available (patients did not appear at any check control). One patient (no. 40) died 6 days after surgery. Seventeen patients died due to distant metastases (brain and thyroid), local relapses, neck recurrences, and other causes. Forty-four patients are still alive, although five of them had a recurrence. The detailed data are presented in Table 4. Two cases, nos. 38 and 84, were not amplified so far. For the rest of cases, we performed PCR amplification for 10 markers; however, because of fast material degradation, we were not able to amplify all samples successfully. These cases are marked as #. The results of LOH analysis in this material are presented in Table 5. LOH was detected for eight markers, with the highest incidence for markers D18S46 (six cases), D7S486 (four cases) and D3S1300 (three cases). LOH was not observed for markers D3S1284 and D8S261. Microsatellite instability was not detected in any case for any marker. Examples of the electropherograms of samples showing LOH and MSI are presented in Fig. 1. 4. Discussion 4.1. LOH analysis Head and neck tumors, including larynx cancer, are a very heterogenous group with numerous chromosomal aberrations, leading to very complex karyotypes [16]. Cytogenetic and molecular studies have revealed many genetic alterations associated with this neoplasm. Loss of heterozygosity studies with the application of microsatellite markers is widely used in tumor genetic investigations [17–22]. This method allows identification of minimal regions of deletions and identification of already known or putative tumor suppressor genes, which losses may promote neoplastic growth. Microsatellite markers are highly polymorphic, easy to analyze and abundant in the whole genome. In patients with HNSCC, high frequencies of LOH on chromosomal arms 3p, 8p, 9p, 13q, 14q, 15q, 17p and 18q have been observed so far [2,23–25]. In the present study, we investigated LOH incidence at 3p13, 3p14.2, 3p21.31, 3p25.3-p26.1, 7q31.2, 8p22, 8p23.3, 9p21 and 18q21.1 chromosomal regions. The highest frequencies of LOH were observed for tumor tissues: losses ranged from 9% for marker D8S264 to 29% for marker D3S1234 (Table 3). The results are lower than observed by other investigators; however, this may be connected with the use of different microsatellite markers—regions of loss may be very small and not
K. Szukała et al. / Mutation Research 600 (2006) 67–76
71
Table 2 The schematic results for all tested markers
Boxes labeled gray mean non-informative/homozygote cases; white boxes without any text mean retention of heterozygosity; LOH, marked for particular tissues (i.e. LOH-a); MSI, microsatellite instability; asterisks mean cases after partial laryngectomy; the rest of the cases were patients after total laryngectomy.
72
K. Szukała et al. / Mutation Research 600 (2006) 67–76
Table 3 Summarized results of microsatellite analysis (frequencies of LOH are shown) Locus symbol
D3S1284 D3S1304 D3S1038 D3S1300 D3S1234 D3S1568 D7S486 D8S261 D8S264 D9S171 D9S1870 D18S46
Number of informative cases (%)
Frequency of LOH in examined samples a – tumor
b – safe margin
c – normal mucosa from upper part of the larynx
d – normal mucosa from tracheostoma region
38/65 (58%) 40/65 (62%) 47/65 (72%) 45/65 (69%) 35/65 (54%) 37/65 (57%) 39/65 (60%) 41/65 (63%) 44/65 (68%) 36/65 (55%) 51/65 (78%) 35/65 (54%)
9/38 (24%) 11/40 (28%) 11/47 (23%) 9/45 (20%) 10/35 (29%) 6/37 (16%) 5/39 (13%) 7/41 (17%) 4/44 (9%) 8/36 (22%) 8/51 (16%) 4/35 (11%)
1/38 (3%) 2/40 (5%) 1/47 (2%) 1/45 (2%) 0 1/37 (3%) 0 0 0 0 2/51 (4%) 0
0 0 2/46 (4%) 1/45 (2%) 0 1/37 (3%) 1/37 (3%) 1/41 (2%) 0 2/36 (6%) 0 1/34 (3%)
0 0 0 0 0 0 0 0 0 0 0 0
Table 4 Patients (numbers) died of recurrences and other causes Recurrences (patient number)
Other causes (patient number)
Upper part of larynx (c)
Tracheostoma (d)
Neck
20, 62
72, 77, 82, 88, 102, 113
68, 78, 79, 101, 121, 129
28, 34, 85, 96, 83, 116, 123, 125
The exceptions (still alive) are underlined.
include chosen markers. The second point is intertumoral and intratumoral heterogeneity—there may exist different subpopulations, carrying different genetic alterations which are subjected to clonal selection [16].
Concerning the safe margin and clinically unchanged mucosa samples, we observed a low incidence of loss of heterozygosity (<6%). The highest incidence of LOH was detected for marker D9S171 (6%) in normal appear-
Table 5 The schematic results of LOH during follow up
Boxes labeled gray mean non-informative/homozygote cases; white boxes mean retention of heterozygosity; # means lack of PCR product; asterisks mean cases after partial laryngectomy; the rest of the cases were patients after total laryngectomy.
K. Szukała et al. / Mutation Research 600 (2006) 67–76
73
Fig. 1. The examples of electropherograms. (A) Case no. 14 for marker D9S171; arrow indicates microsatellite instability (MSI) in tumor tissue (14a). (B) Case no. 101 for marker D3S1284; arrows indicate LOH in tumor (a) and margin (b) samples.
ing mucosa from upper part of larynx (point c). We did not find any LOH for any marker in the material of normal mucosa from tracheostoma region (d). The results are shown in Table 3. Analyzing tissues from the safe part of the excision specimens, we cannot forget that losses on 3p and 9p arms are supposed to be early events during head and neck carcinogenesis [26–28]. As it was shown by Mao et al. [12] the risk of cancer development on the basis of premalignant lesion is about half-fold higher when chromosome instability appears in two key regions: 3p14 and 9p21. In our study, we cannot consider these regions as hallmarks of cancer risk development—the frequencies of LOH were very low in these chromosomal regions. Moreover, patients who have developed a relapse in the supraglottic region or in tracheostoma did not reveal LOH in “c” or “d” samples. We also cannot point out the significance of other studied markers. It may suggest that other regions were lost during the course of larynx cancer and that there may be other important tumor suppressor genes localized there. The 8p arm was selected based on the previous studies providing evidence for multiple TSGs in laryngeal cancer. On 8p arm, at least three regions of frequent deletions were identified: 8p21, 8p22-p23 and 8p23.3 [7,10,29]. Deletions were observed most frequently in band 8p23 [2,30]. In this arm, several genes of suppressor activity are located; however, so far there is no proof for a direct involvement of these loci in the larynx pathogenesis. On the other hand, Bockm¨uhl et al. [31] showed a significant association between allelic loss on 8p23 and a shorter disease-free interval with respect to metastases development. In our study LOH frequencies
ranged from 9 to 17% for both markers for tumor tissue and we detected LOH in only one case of normal appearing mucosa (Table 3). In the group of patients with recurrences, we found LOH in tumor sample only in two cases: no. 20 (D8S264) and no. 62 (D8S261). Both of them had a relapse in the upper part of larynx. We did not detect LOH in these markers in a group of patients with a relapse in tracheostoma (Table 2). We suppose that other markers on 8p arm could be of higher significance in microsatellite analysis. In the present study, we also tested two markers which were not connected with early losses so far. The first marker is located in the region 7q31.2 (D7S486). LOH on 7q was rarely analyzed, however frequent allelic losses in band q31 were detected in almost one-half of studied informative cases. In this region, ING3 gene is located—loss of ING3 expression was related to higher mortality in the group of tongue and larynx cancer patients [32]. In our study, LOH in this region was detected in 5 out of 39 (13%) tumor samples and in 1 out of 37 (3%) normal mucosa samples from the upper part of the larynx, however what is more important we detected LOH in this marker in 4 cases analyzed during follow up study (Table 5). It suggests that loss in band 7q31.2 could be a “weak” alteration, which succumbed to the selection pressure in transformed cells, but remained in non-transformed cells, which did not develop into tumor. The second marker is located in region 18q21.1 (D18S46). As it was previously described, LOH in this chromosomal region is believed to be connected with advanced tumor stage, aggressive tumor behavior and
74
K. Szukała et al. / Mutation Research 600 (2006) 67–76
poor survival in HNSCC patients [33]. The objective was to check if LOH frequency in tumor samples in this region would be higher than in others, supposed to be lost at early stages. We did not observe such effect—LOH in this region was less frequent (11% in tumor tissue and 3% in normal mucosa). It is clear that this study should be extended for more markers; however, we cannot exclude a similar assumption as for marker D7S486—with selection pressure in tumor samples. This hypothesis is supported by the fact that in the follow up group we detected LOH in this marker in six cases—but two out of these six cases (nos. 30 and 124) had LOH also detected in tumor samples (Tables 2 and 5). On the basis of these results, we cannot point out any marker which could be a good predictor of local recurrence risk. We have observed that two patients (nos. 20 and 62) who died because of relapse in upper part of larynx had five and seven losses in tumor samples for different markers, while five patients who died because of relapse in tracheostoma region (nos. 72, 77, 82, 88 and 113) revealed maximum two losses in tumor samples. Moreover, three out of these five cases had no LOH at any marker tested. This may suggest that field cancerization process is more extended in upper part of aerodigestive tract than in lower parts, near tracheostoma region. This may also be supported by the finding that we detect no LOH at any marker tested in mucosal samples taken from tracheostoma region. On the other hand, there is a group of patients—alive, with no evidence of recurrence who had multiple losses at multiple markers (i.e. nos. 14, 38 and 124). In cases nos. 14 and 38, additionally microsatellite instability (appearing of additional allele, not observed in reference material) was detected. Follow up period lasted 39–52 months. This group displayed an overall genetic instability but so far we have no certainty if they develop a recurrence in the future. Maybe these patients have to acquire other genetic alterations to develop second tumor.
pattern more than 4 years after completion of surgical treatment. The low number of monitored cases is a result of laryngeal cancer patients’ lifestyle—continued habits, no regular appearance to the check controls. In 8 out of 10 microsatellite markers, we detected LOH in at least one case (D3S1304, D3S1568, D8S264, D9S171 and D9S1870). For three markers (D3S1300, D7S486 and D18S46), we found LOH in three, four and six cases, respectively. These results suggest that patients may be in a group of increased risk of malignant process occurrence, although in the clinical examination there was no suspicion of relapse. However, we suppose that these alterations may be “weak” and not influence tumor development process. Out of these 18 losses, only two cases—nos. 30 and 124, for marker D18S46 had LOH detected in tumor tissue (30a and 124a). The remaining 16 losses were not observed previously, what may suggest that these changes appeared already after surgery as the result of continuous exposures to carcinogens. It is also likely that appearing of new alterations may be a result of the heterogeneity of larynx. 5. Conclusions Preliminary data, obtained during this investigation, did not reveal the predictive value of LOH with respect to local relapse occurrence in laryngeal cancer patients. However, time of follow up did not reach 5 years so far, so next recurrences cannot be ruled out. We also cannot exclude the possibility that postoperative radiation has removed all changed or tumor cells left behind during surgery and those new alterations, observed during follow up, appeared already after surgery. It also seems that the number of analyzed markers should be increased: first, because of larynx cancers heterogeneity, resulting in existence of multiple subclones “carrying” different genetic alterations, and second, to improve informativeness. With respect to these results, further clinical monitoring of patients should be conducted.
4.2. Follow up study Recently, the number of studies conducted on body fluids, as well as the material from brush cytology has increased [34–38]. El-Naggar et al. observed a high incidence of LOH (49%) in saliva samples (cytological examination revealed no malignant cells), compared to 86% of LOH obtained for tumor samples [39]. Rizos et al. found significant incidence of LOH on 8p, 9p and 17q arms in cytological specimens of the larynx [35]. In this study, we used brush specimens as well as laryngeal swabs from 16 patients to investigate LOH
Acknowledgement This work was partially supported by a grant from the Foundation ORLEN-Dar Serca (ORLEN-Gift of A Heart). References [1] M.P. Tabor, R.H. Brakenhoff, V.M.M. van Houten, J.A. Kummer, M.H.J. Snel, P.J.F. Snijders, G.B. Snow, C.R. Leemans, B.J.M. Braakhuis, Persistence of genetically altered fields in head and
K. Szukała et al. / Mutation Research 600 (2006) 67–76
[2]
[3]
[4]
[5]
[6]
[7]
[8] [9] [10]
[11] [12]
[13]
[14]
[15]
[16]
[17]
[18]
neck cancer patients: biological and clinical implications, Clin. Cancer Res. 7 (2001) 1523–1532. L.B. Beder, M. Gunduz, M. Ouchida, K. Fukushima, E. Gunduz, S. Ito, A. Sakai, N. Nagai, K. Nishizaki, K. Shimizu, Genomewide analyses on loss of heterozygosity in head and neck squamous cell carcinomas, Lab. Invest. 83 (2003) 99–105. J.S. Cooper, T.F. Pajak, P. Rubin, L. Tupchong, L.W. Brady, S.A. Leibel, G.E. Laramore, V.A. Marcial, L.W. Davis, J.D. Cos, Second malignancies in patients who had head and neck cancer: incidence, effect on survival, and implications based on the ROTG experience, Int. J. Radiat. Onc. Biol. Phys. 1 (1989) 449–456. D. Slaughter, H. Southwick, W. Smejkal, Field cancerization in oral stratified squamous epithelium: clinical implications of multicentric origin, Cancer 6 (1953) 953–971. C.R. Leemans, R. Tiwari, J.J. Nauta, I. van der Waal, G.B. Snow, Recurrence at the primary site in head and neck cancer and the significance of neck lymph node metastases as a prognostic factor, Cancer 73 (1994) 187–190. W.M. Lydiatt, P.E. Anderson, T. Bazzana, M. Casale, C.J. Hughes, A.G. Huvos, D.D. Lydiatt, S.P. Schantz, Molecular support for field cancerization in the head and neck, Cancer 82 (1998) 1376–1380. M. Partridge, S. Peteromichelakis, E. Phillips, G.G. Emilion, R.P. A’Hern, J.D. Langdon, A case-control study confirms that microsatellite assay can identify patients at risk of developing oral squamous cell carcinoma within a field of cancerization, Cancer Res. 60 (2000) 3893–3898. D. Sidransky, Molecular genetics of head and neck cancer, Curr. Opin. Oncol. 7 (1995) 229–233. M.A. Nagai, Genetic alterations in head and neck squamous cell carcinomas, Braz. J. Med. Biol. Res. 32 (1999) 897–904. S.M. Gollin, Chromosomal alterations in squamous cell carcinomas of the head and neck: window to the biology of disease, Head Neck (2001) 238–252. P. Devilee, A.M. Cleton-Jansen, C.J. Cornelisse, Ever since Knudson, Trends Genet. 17 (2001) 569–573. L. Mao, J.S. Lee, Y.H. Fan, J.Y. Ro, J.G. Batsakis, S. Lippman, W. Hittleman, W.K. Hong, Frequent microsatellite alterations at chromosome 9p21 and 3p14 in oral premalignant lesions and their value in cancer risk assessment, Nat. Med. (1996) 682. J. Califano, P. van der Riet, W. Westra, H. Nawroz, G. Clayman, S. Piantadosi, R. Corio, D. Lee, B. Greenberg, W. Koch, D. Sidransky, Genetic progression model for head and neck cancer: implications for field cancerization, Cancer Res. 56 (1996) 2488–2492. J. Califano, W.H. Westra, G. Meininger, R. Corio, W.M. Koch, D. Sidransky, Genetic progression and clonal relationship of recurrent premalignant head and neck lesions, Clin. Cancer Res. 6 (2000) 347–352. K. Szukala, J. Brieger, K. Bruch, W. Biczysko, M. Wierzbicka, W. Szyfter, K. Szyfter, Loss of heterozygosity on chromosome arm 13q in larynx cancer patients: analysis of tumor, margin and clinically unchanged mucosa, Med. Sci. Monit. 10 (2004) 233–240. Ch. Jin, Y. Jin, J. Wennerberg, J. Akervall, M. Dictor, F. Mertens, Karyotypic heterogeneity and clonal evolution in squamous cell carcinomas of the head and neck, Cancer Genet. Cytogen. 132 (2002) 85–96. S. Lai, W.F. Benedict, S.A. Silver, A.K. El-Naggar, Loss of retinoblastoma gene function and heterozygosity at the RB locus in renal cortical neoplasms, Hum. Pathol. 28 (1997) 693–697. A. Goel, C.N. Arnold, D. Nied´zwiecki, D.K. Chang, L. Ricciardiello, J.M. Carethers, J.M. Dowell, L. Wasserman, C. Compton,
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
75
R.J. Mayer, M.M. Bertagnolli, C.R. Boland, Characterization of sporadic colon cancer by patterns of genomic instability, Cancer Res. 63 (2003) 1608–1614. C. Dotzenrath, B.T. Teh, F. Farnebo, K. Cupisti, A. Svensson, A. Toell, P. Goretzki, C. Larsson, Allelic loss of the retinoblastoma tumor suppressor gene: a marker for aggressive parathyroid tumors? J. Clin. Endocrinol. Metab. 81 (1996) 3194–3196. N.A. Thomas, P.J. Neville, S.W. Baxter, I.G. Campbell, Genetic analysis of benign ovarian tumors, Int. J. Cancer 105 (2003) 499–505. G. Eiriksdottir, G. Johannesdottir, S. Ingvarsson, I.B. Bjornsdottir, J.G. Jonasson, B.A. Agnarsson, J. Hallgrimsson, J. Gudmundsson, V. Egilsson, H. Sigurdsson, R.B. Barkardottir, Mapping loss of heterozygosity at chromosome 13q: loss at 13q12-q13 is associated with breast tumour progression and poor prognosis, Eur. J. Cancer 34 (1998) 2076–2081. J.T. Dong, J.C. Boyd, H.F. Frierson, Loss of heterozygosity at 13q14 and 13q21 in high grade, high stage prostate cancer, Prostate 49 (2001) 166–171. K.W. Ah-See, T.G. Cooke, I.R. Pickford, D. Soutar, A. Balmain, An allelotype of squamous carcinoma of the head and neck using microsatellite markers, Cancer Res. 54 (1994) 1617–1621. H. Nawroz, P. van der Riet, R.H. Hruban, W. Koch, J.M. Ruppert, D. Sidransky, Allelotype of head and neck squamous cell carcinoma, Cancer Res. 54 (1994) 1152–1155. H. Rowley, A. Jones, D. Spandidos, J. Field, Definition of a tumor suppressor gene locus on the short arm of chromosome 3 in squamous cell carcinoma of the head and neck by means of microsatellite markers, Arch. Otolaryngol. Head Neck Surg. 122 (1996) 497–501. P. van der Riet, H. Nawroz, R.H. Hruban, R. Corio, K. Tokino, W. Koch, D. Sidransky, Frequent loss of chromosome 9p21-22 early in head and neck cancer progression, Cancer Res. 54 (1994) 1156–1158. L. Roz, C.L. Wu, S. Porter, C. Scully, P. Speight, A. Read, P. Sloan, N. Thakker, Allelic imbalance on chromosome 3p in oral dysplastic lesions: an early event in oral carcinogenesis, Cancer Res. 56 (1996) 1228–1231. Z. Guo, K. Yamaguchi, M. Sanchez-Cespedes, W.H. Westra, W.M. Koch, D. Sidransky, Allelic losses in OraTest-directed biopsies of patients with prior upper aerodigestive tract malignancy, Clin. Cancer Res. 7 (2001) 1963–1968. M. Partridge, G.G. Emilion, S. Pateromichelakis, E. Phillips, J.D. Langdon, Location of candidate tumour suppressor gene loci at chromosomes 3p, 8p and 9p for oral squamous cell carcinomas, Int. J. Cancer 83 (1999) 318–325. J.B. Sunwoo, P.C. Sun, V.K. Gupta, A.P. Schmidt, S. El-Mofty, S.B. Scholnick, Localization of a putative tumor suppressor gene in the sub-telomeric region of chromosome 8p, Oncogene 18 (1999) 2651–2655. U. Bockm¨uhl, Ch.S. Ishwad, R.E. Ferrell, S.M. Gollin, Association of 8p23 deletions with poor survival in head and neck cancer, Otolaryngol. Head Neck Surg. 124 (2001) 451–455. M. Gunduz, M. Ouchida, K. Fukushima, S. Ito, Y. Jitsumori, T. Nakashima, N. Nagai, K. Nishizaki, K. Shimizu, Allelic loss and reduced expression of the ING3, a candidate tumor suppressor gene at 7q31, in human head and neck cancers, Oncogene 21 (2002) 4462–4470. S. Takebayashi, A. Hickson, T. Ogawa, K.Y. Jung, H. Mineta, Y. Ueda, R. Grenman, S.G. Fisher, T.E. Carey, Loss of chromosome arm 18q with tumor progression in head and neck squamous cancer, Genes Chromosomes Cancer 41 (2004) 145–154.
76
K. Szukała et al. / Mutation Research 600 (2006) 67–76
[34] J.A. Veltman, A.H.N. Hopman, F.J. Bot, F.C.S. Ramaekers, J.J. Manni, Detection of chromosomal aberrations in cytologic brush specimens from head and neck squamous cell carcinoma, Cancer 81 (1997) 309–314. [35] E. Rizos, G. Sourvinos, D.A. Spandidos, Loss of heterozygosity at 8p, 9p and 17q in laryngeal cytological specimens, Oral Oncol. 34 (1998) 519–523. [36] M.F. Spafford, W.M. Koch, A.L. Reed, J.A. Califano, L.H. Xu, C.F. Eisenberger, L. Yip, P.L. Leong, L. Wu, S.X. Liu, C. Jeronimo, W.H. Westra, D. Sidransky, Detection of head and neck squamous cell carcinoma among exfoliated oral mucosal cells by microsatellite analysis, Clin. Cancer Res. 7 (2001) 607–612. [37] S. Lopes Bittencourt Rosas, W. Koch, M. da Gloria da Costa Carvalho, L. Wu, J. Califano, W. Westra, J. Jen, D. Sidransky,
Promoter hypermethylation patterns of p16, O6 -methylguanineDNA-methyltransferase, and death-associated protein kinase in tumors and saliva of head and neck cancer patients, Cancer Res. 61 (2001) 939–942. [38] S. Temam, J. Benard, C. Dugas, M. Trassard, E. Gormally, J.C. Soria, S. Faivre, B. Luboinski, P. Marandas, P. Hainaut, G. Lenoir, L. Mao, F. Janot, Molecular detection of early-stage laryngopharyngeal squamous cell carcinoma, Clin. Cancer Res. 11 (2005) 2547–2551. [39] A.K. El-Naggar, L. Mao, G. Staerkel, M.M. Coombes, S.L. Tucker, M.A. Luna, G.L. Clayman, S. Lippman, H. Goepfert, Genetic heterogeneity in saliva from patients with oral squamous carcinomas, J. Mol. Diagn. 3 (2001) 164–170.
Does loss of heterozygosity in critical genome regions predict a local relapse in patients after laryngectomy? Katarzyna Szukała a,∗ , Anna Sowi´nska b , Małgorzata Wierzbicka d , Wiesława Biczysko c , Witold Szyfter d , Krzysztof Szyfter a,d a
Institute of Human Genetics, Polish Academy of Sciences, Pozna´n, Poland Department of Biochemistry and Molecular Biology, Medical University of Pozna´n, Pozna´n, Poland c Department of Clinical Pathology, Medical University of Pozna´ n, Pozna´n, Poland Department of Otolaryngology and Laryngeal Oncology, Medical University of Pozna´n, Pozna´n, Poland b
d
Available online 11 July 2006
Abstract Background: Patients, who had an upper aerodigestive tract malignancy, have a high incidence of succeeding tumor development. This has been attributed to the role of “field cancerization” in carcinogenesis. The aim of this study was analysis of loss of heterozygosity (LOH) in the regions frequently lost during the course of head and neck squamous cell carcinomas (HNSCC), especially at early stages, which could answer the clinicians’ question, if LOH analysis has any “predictive” value in relation to tumor occurrence. Material and methods: Sixty-five larynx cancer patients were examined for loss of heterozygosity on 3p, 7q, 8p, 9p and 18q chromosomal arms with the use of 12 microsatellite markers. The material from a single patient consisted of blood, tumor, safe margin and one or two clinically unchanged mucosal samples. During follow up, the material from brush specimens (14 patients) as well as laryngeal swabs (4 patients) was also examined. Results: The highest frequency of LOH was detected for marker D3S1234 in tumor tissues (29%). Analysis of margin samples (b) revealed low LOH frequencies (2–5%) and complete retention of heterozygosity for markers: D3S1234, D7S486, D8S261, D8S264, D9S171 and D18S46. Similarly, for normal appearing mucosa from upper part of larynx (c) frequencies of LOH were low (2–6%), with the complete retention of heterozygosity for markers: D3S1284, D3S1304, D3S1234, D8S264 and D9S1870. We did not detect any LOH in the material of normal appearing mucosa from tracheostoma region (d). During follow up, LOH was detected for eight markers, with the highest incidence for markers D18S46 (six cases), D7S486 (four cases) and D3S1300 (three cases). Conclusions: The data, obtained during this investigation, did not reveal the predictive value of LOH with respect to local relapse occurrence in laryngeal cancer patients. However, time of follow up did not reach 5 years, so that further clinical monitoring should be conducted. © 2006 Elsevier B.V. All rights reserved. Keywords: Larynx cancer; Loss of heterozygosity; Microsatellite markers; Follow up; Local relapse
1. Introduction
∗
Corresponding author at: Institute of Human Genetics, Polish Academy of Science, ul.Strzeszy´nska 32, 60-479 Pozna´n, Poland. Tel.: +48 61 6579214; fax: +48 61 8233235. E-mail address: [email protected] (K. Szukała). 0027-5107/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.mrfmmm.2006.05.027
Head and neck squamous cell carcinomas (HNSCC) constitute a significant medical and social problem, with the occurrence of around 5% of all worldwide-diagnosed malignancies. The difficulties in studying the progression of larynx cancer are connected with its complex
68
K. Szukała et al. / Mutation Research 600 (2006) 67–76
biology, genetics and with a relatively late clinical detection of primary tumors. As the result, local recurrences, lymph node relapses, distant metastases and second primary tumors appear very often [1,2]. Still, in spite of advanced genetic studies and improvements in diagnostics and treatment of HNSCC the survival rates are rather poor, ranging from 40 to 50% [3]. Treatment failures may partially be explained by anatomical predisposition of head and neck region organs—the epithelium of upper respiratory tract, exposed for a prolonged carcinogens activity (in tobacco, alcohol, air, preservatives) is being continuously damaged (the “field cancerization” theory) [4]. Local recurrences can develop within primary tumor resection site, even when surgical margins were histologically tumor-free [5]. These features extorted the investigators to study not only the material derived from tumor, but also from clinically unchanged mucosa resected within the safe margin from the operating field. It is possible that normal-appearing mucosa left after tumor and margin excision may be genetically altered [1,6]. Unfortunately, it means that “field cancerization” process is much more extensive than clinical and histological examinations show [7]. It was estimated, that a cell undertakes an irreversible malignant transformation process after acquiring 6–10 genetic alterations [8]. These alterations concern activation of oncogenes as well as inactivation of tumor suppressor genes, as it was revealed by cytogenetic and molecular studies [9,10]. Loss of heterozygosity (LOH) has been recognized to be an important event in HNSCC, suggesting involvement of the suppressor pathway in the genesis of this disease—LOH is often one of two inactivation steps of tumor suppressor genes (TSGs) [11]. Multiple analyses with the application of microsatellite markers provided the knowledge about chromosomal regions containing critical tumor suppressor genes that are frequently lost or inactivated in HNSCC, like p16 (9p21), p53 (17p13.1), FHIT (3p14.2), VHL (3p25.3), Rb1 (13q14) and ING1 (13q34) [9,10]. Moreover, there are numerous regions lost during HNSCC pathogenesis, where putative TSGs were not identified so far (i.e. 8p arm). Microsatellite analysis appears to have a potential application in predicting cancer risk, as shown by studies conducted previously on oral leukoplakias exhibited LOH at chromosomal regions 3p14 and/or 9p21 to carry higher risk for HNSCC development [12]. In the present study, we investigated LOH in the regions frequently lost during the course of HNSCC, harboring known or putative tumor suppressor loci in the material derived from patients treated surgically for larynx cancer. The material consisted of tumor, safe mar-
gin and samples of normal appearing mucosa taken from distant, opposite points of surgical specimens. The diversity among clinical material should enable a comparison of genetic alterations in the same patient in various localizations (glottis, trachea). Additionally, during follow up, laryngeal swabs from hypopharynx and brush specimens from tracheostoma were collected to study LOH more than 4 years after surgery. The markers were chosen with the emphasis on localization in regions lost “early” during neoplastic transformation [13,14]. This experiment may answer the clinicians’ question, if LOH analysis has any “predictive” value in relation to tumor occurrence. 2. Patients and methods 2.1. Samples collection Altogether 65 larynx cancer patients (62 men and 3 women) were examined. Material from surgical specimens was obtained from the Department of Otolaryngology and Laryngeal Oncology, K. Marcinkowski Medical University, Pozna´n, Poland, and was kept frozen at −80 ◦ C right after the surgery till laboratory studies. The detailed data were described previously [15]. In brief, the material from one patient consisted of four or five samples. In cases of total laryngectomy (46 patients), five samples were taken: • • • •
blood as a reference material; tumor (marked as “a”); safe margin from the operating field (“b”); two samples from distant, opposite points of the surgical specimen—clinically unchanged mucosa from the epiglottis or hypopharynx (proximal point of specimen called “c”) and normal mucosa of the tracheal ring (distal point of specimen called “d”).
In cases of partial laryngectomy (19 patients), four samples were taken: • • • •
blood as a reference material; tumor (marked as “a”); safe margin from the operating field (“b”); normal appearing mucosa from the upper part of the larynx (proximal point of specimen called “c”).
Tumor and safe margin tissues were verified histopathologically by microscopy. Normal mucosa was not verified by histology because of unchanged clinical character and a distance from the tumor exceeding 2 cm. Additionally, during the last follow up visit (June 2005), brush specimens (marked as “z”) were taken from 14 patients after total laryngectomy from tracheostomy and laryngeal swabs (marked as “s”) were taken from hypopharynx from 4 patients after partial laryngectomy.
K. Szukała et al. / Mutation Research 600 (2006) 67–76
69
cycles of 95 ◦ C for 30 s, 54–62 ◦ C for 30 s and 72 ◦ C for 45 s and a final elongation step of 72 ◦ C for 10 min. PCR products were mixed with loading dye and internal sizers 100 and 300 bp (Amersham Pharmacia Biotech), then denaturated for 5 min in 95 ◦ C, quickly chilled on ice and then separated with the use of ALF Express II System (Amersham Pharmacia Biotech). LOH was calculated from the measured peak heights, by dividing the allele ratio in analyzed tissue DNA by the allele ratio in constitutional DNA: [t1 /t2 ]/[n1 /n2 ]. Loss of heterozygosity was defined as a more than 50% reduction in the peak height of the clinical tissue compared to the peak height of corresponding blood tissue.
Patients have been followed-up once a month since the surgery till June 2005. 2.2. DNA extraction and LOH analysis DNA was isolated from all samples according to the standard procedures (proteinase K digestion, phenol/chloroform extraction and ethanol precipitation). The DNA elution of brush specimens and laryngeal swabs was conducted with the use of 0.2 M NaOH and 0.04 M Tris–HCl. DNA was PCR amplified with the use of primers, specific for chosen microsatellite markers (dinucleotide repeats). One primer from each primer pair was fluorescently end-labeled with Cy-5. Primers were obtained from http://www.oligo.pl/—DNA Sequencing and Synthesis IBB PAN, Warsaw, Poland. Sequences of all primers, available in Genome Database (http://www.gdbwww.gdb.org/) and Ensembl Genome Browser (http://www.ensembl.org/) are listed in Table 1. PCR reactions were carried out in a total reaction volume of 30 l containing ∼100 ng of genomic DNA, 15 pmol of each primer, 6 nmol of each dNTP (Roche, Mannheim, Germany), 1xPCR buffer [10 mM Tris–HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl2 ], 1U Taq DNA polymerase (Sigma). The mixtures were amplified with a PTC-200 Peltier Thermal Cycler, MJ Research. The PCR profile consisted of initial denaturation step (95 ◦ C for 5 min), followed by 35
2.3. Microsatellite markers Twelve microsatellite markers, used for LOH analysis were mapped to the regions harboring known or putative tumor suppressor genes as follows: 3p13 (D3S1284), 3p14.2 (D3S1300, D3S1234), 3p25.3-p26.1 (D3S1038, D3S1304), 3p21.31 (D3S1568), 7q31.2 (D7S486), 8p22-23.3 (D8S261, D8S264), 9p21 (D9S171, D9S1870) and 18q21.1 (D18S46). These regions were previously shown to be frequently lost in head and neck tumors. The detailed data about chosen microsatellites are listed in Table 1.
Table 1 Data concerning microsatellite markers used in the study Marker
Locus
Gene
Tm (◦ C)
Het (%)
Length of sequence (bp)
Primers sequences
D3S1284
3p13
DUTT1
61
75
155–177
F GCCTTGGGGGTAAATACTCT R GGAATTACAGGCCACTGCTC
D3S1304
3p26.1
VHL
59
80
253–269
F TTCGCTCTTTGATAGGC R ATTTCATTTGTAATTTACTAGCAG
D3S1038
3p25.3
VHL
58
80
115
F TCCAGTAAGAGGCTTCCTAG R AAAGGGGTTCAGGAAACCTG
D3S1300
3p14.2
FHIT
59
83
217–241
F AGCTCACATTCTAGTCAGCCT R GCCAATTCCCCAGATG
D3S1234
3p14.2
FHIT
58
66
99–125
F CCTGTGAGACAAAGCAAGAC R GACATTAGGCACAGGGCTAA
D3S1568
3p21.31
RASSF1
62
87
276–296
F CCATGAACAGAACCTCCCTA R CCGCTGTCCTGCTGTAAG
D7S486
7q31.2
ING3
54
80
114–146
F AAAGGCCAATGGTATATCCC R GCCCAGGTGATTGATAGTGC
D8S261
8p22
?
62
78
128–144
F TGCCACTGTCTTGAAAATCC R TATGGCCCAGCAATGTGTAT
D8S264
8p23.3
?
59
85
121–145
F ACATCTGCGTCGTCTTCATA R CCAACACCTGAGTCAGCATA
D9S171
9p21
p16
59
80
159–177
F AGCTAAGTGAACCTCATCTCTGTCT R ACCCTAGCACTGATGGTATAGTCT
D9S1870
9p21
p16
61
66
179–122
F TGGGTATGGTTTTCTGG R TTGAGGCAGGTCAAATAA
D18S46
18q21.1
MADH4
56
80
129–153
F GAATAGCAGGACCTATCAAAGAGC R CAGATTAAGTGAAAACAGCATATGG
70
K. Szukała et al. / Mutation Research 600 (2006) 67–76
For analyzing brush specimens (z) and laryngeal swabs (s), the same markers were tested, except for two of them: D3S1038 and D3S1234.
3. Results 3.1. LOH analysis in the surgical material In all cases, DNA was successfully amplified. Unfortunately, for two patients (35 and 38), DNA from “c” specimen has been already finished, so that we have no result in sample 35c for markers, D3S1038 and D7S486, as well as we do not have the results in sample 38c for markers D3S1304, D3S1234, D7S486 and D18S46. The schematic representation of LOH distribution in particular tissues for all tested markers is performed in Table 2. The summarized results of microsatellite analysis are shown in Table 3. Only the informative cases (heterozygous at the examined locus) were considered. All cases were informative for at least one marker. The highest number of informative cases was detected for marker D9S1870 (78%), while for markers D3S1234 and D18S46 only 54% of cases were heterozygotes. Considering LOH in particular tissues, only in tumor tissues LOH was observed for all markers. The highest frequency of LOH in tumor tissues was detected for marker D3S1234 (29%) and the lowest frequency was observed for marker D8S264 (9%). LOH in the remaining tissues was detected less frequent. Analysis of margin samples (b) revealed low LOH frequencies (2–5%) and complete retention of heterozygosity for markers: D3S1234, D7S486, D8S261, D8S264, D9S171 and D18S46. The situation was similar for normal appearing mucosa from upper part of larynx (c)—frequencies of LOH were low (2–6%), with the complete retention of heterozygosity for markers: D3S1284, D3S1304, D3S1234, D8S264 and D9S1870. We did not detect any LOH in the material of normal appearing mucosa from tracheostoma region (d). Beside LOH we found microsatellite instability (MSI) in two cases: in case no. 14, we detected MSI in tumor sample (a) for markers D3S1038, D3S1300, D3S1568, D7S486 and D9S171. In margin sample (b), we detected MSI for marker D9S171. In case no. 38, we detected MSI in the normal appearing mucosa sample (c) for marker D9S171. 3.2. Follow up study—LOH analysis in brush specimens and laryngeal swabs The follow up observation period ranged from 3 to 55 months (mean 31 months). For three cases, nos. 23,
25 and 64, data concerning follow up were not available (patients did not appear at any check control). One patient (no. 40) died 6 days after surgery. Seventeen patients died due to distant metastases (brain and thyroid), local relapses, neck recurrences, and other causes. Forty-four patients are still alive, although five of them had a recurrence. The detailed data are presented in Table 4. Two cases, nos. 38 and 84, were not amplified so far. For the rest of cases, we performed PCR amplification for 10 markers; however, because of fast material degradation, we were not able to amplify all samples successfully. These cases are marked as #. The results of LOH analysis in this material are presented in Table 5. LOH was detected for eight markers, with the highest incidence for markers D18S46 (six cases), D7S486 (four cases) and D3S1300 (three cases). LOH was not observed for markers D3S1284 and D8S261. Microsatellite instability was not detected in any case for any marker. Examples of the electropherograms of samples showing LOH and MSI are presented in Fig. 1. 4. Discussion 4.1. LOH analysis Head and neck tumors, including larynx cancer, are a very heterogenous group with numerous chromosomal aberrations, leading to very complex karyotypes [16]. Cytogenetic and molecular studies have revealed many genetic alterations associated with this neoplasm. Loss of heterozygosity studies with the application of microsatellite markers is widely used in tumor genetic investigations [17–22]. This method allows identification of minimal regions of deletions and identification of already known or putative tumor suppressor genes, which losses may promote neoplastic growth. Microsatellite markers are highly polymorphic, easy to analyze and abundant in the whole genome. In patients with HNSCC, high frequencies of LOH on chromosomal arms 3p, 8p, 9p, 13q, 14q, 15q, 17p and 18q have been observed so far [2,23–25]. In the present study, we investigated LOH incidence at 3p13, 3p14.2, 3p21.31, 3p25.3-p26.1, 7q31.2, 8p22, 8p23.3, 9p21 and 18q21.1 chromosomal regions. The highest frequencies of LOH were observed for tumor tissues: losses ranged from 9% for marker D8S264 to 29% for marker D3S1234 (Table 3). The results are lower than observed by other investigators; however, this may be connected with the use of different microsatellite markers—regions of loss may be very small and not
K. Szukała et al. / Mutation Research 600 (2006) 67–76
71
Table 2 The schematic results for all tested markers
Boxes labeled gray mean non-informative/homozygote cases; white boxes without any text mean retention of heterozygosity; LOH, marked for particular tissues (i.e. LOH-a); MSI, microsatellite instability; asterisks mean cases after partial laryngectomy; the rest of the cases were patients after total laryngectomy.
72
K. Szukała et al. / Mutation Research 600 (2006) 67–76
Table 3 Summarized results of microsatellite analysis (frequencies of LOH are shown) Locus symbol
D3S1284 D3S1304 D3S1038 D3S1300 D3S1234 D3S1568 D7S486 D8S261 D8S264 D9S171 D9S1870 D18S46
Number of informative cases (%)
Frequency of LOH in examined samples a – tumor
b – safe margin
c – normal mucosa from upper part of the larynx
d – normal mucosa from tracheostoma region
38/65 (58%) 40/65 (62%) 47/65 (72%) 45/65 (69%) 35/65 (54%) 37/65 (57%) 39/65 (60%) 41/65 (63%) 44/65 (68%) 36/65 (55%) 51/65 (78%) 35/65 (54%)
9/38 (24%) 11/40 (28%) 11/47 (23%) 9/45 (20%) 10/35 (29%) 6/37 (16%) 5/39 (13%) 7/41 (17%) 4/44 (9%) 8/36 (22%) 8/51 (16%) 4/35 (11%)
1/38 (3%) 2/40 (5%) 1/47 (2%) 1/45 (2%) 0 1/37 (3%) 0 0 0 0 2/51 (4%) 0
0 0 2/46 (4%) 1/45 (2%) 0 1/37 (3%) 1/37 (3%) 1/41 (2%) 0 2/36 (6%) 0 1/34 (3%)
0 0 0 0 0 0 0 0 0 0 0 0
Table 4 Patients (numbers) died of recurrences and other causes Recurrences (patient number)
Other causes (patient number)
Upper part of larynx (c)
Tracheostoma (d)
Neck
20, 62
72, 77, 82, 88, 102, 113
68, 78, 79, 101, 121, 129
28, 34, 85, 96, 83, 116, 123, 125
The exceptions (still alive) are underlined.
include chosen markers. The second point is intertumoral and intratumoral heterogeneity—there may exist different subpopulations, carrying different genetic alterations which are subjected to clonal selection [16].
Concerning the safe margin and clinically unchanged mucosa samples, we observed a low incidence of loss of heterozygosity (<6%). The highest incidence of LOH was detected for marker D9S171 (6%) in normal appear-
Table 5 The schematic results of LOH during follow up
Boxes labeled gray mean non-informative/homozygote cases; white boxes mean retention of heterozygosity; # means lack of PCR product; asterisks mean cases after partial laryngectomy; the rest of the cases were patients after total laryngectomy.
K. Szukała et al. / Mutation Research 600 (2006) 67–76
73
Fig. 1. The examples of electropherograms. (A) Case no. 14 for marker D9S171; arrow indicates microsatellite instability (MSI) in tumor tissue (14a). (B) Case no. 101 for marker D3S1284; arrows indicate LOH in tumor (a) and margin (b) samples.
ing mucosa from upper part of larynx (point c). We did not find any LOH for any marker in the material of normal mucosa from tracheostoma region (d). The results are shown in Table 3. Analyzing tissues from the safe part of the excision specimens, we cannot forget that losses on 3p and 9p arms are supposed to be early events during head and neck carcinogenesis [26–28]. As it was shown by Mao et al. [12] the risk of cancer development on the basis of premalignant lesion is about half-fold higher when chromosome instability appears in two key regions: 3p14 and 9p21. In our study, we cannot consider these regions as hallmarks of cancer risk development—the frequencies of LOH were very low in these chromosomal regions. Moreover, patients who have developed a relapse in the supraglottic region or in tracheostoma did not reveal LOH in “c” or “d” samples. We also cannot point out the significance of other studied markers. It may suggest that other regions were lost during the course of larynx cancer and that there may be other important tumor suppressor genes localized there. The 8p arm was selected based on the previous studies providing evidence for multiple TSGs in laryngeal cancer. On 8p arm, at least three regions of frequent deletions were identified: 8p21, 8p22-p23 and 8p23.3 [7,10,29]. Deletions were observed most frequently in band 8p23 [2,30]. In this arm, several genes of suppressor activity are located; however, so far there is no proof for a direct involvement of these loci in the larynx pathogenesis. On the other hand, Bockm¨uhl et al. [31] showed a significant association between allelic loss on 8p23 and a shorter disease-free interval with respect to metastases development. In our study LOH frequencies
ranged from 9 to 17% for both markers for tumor tissue and we detected LOH in only one case of normal appearing mucosa (Table 3). In the group of patients with recurrences, we found LOH in tumor sample only in two cases: no. 20 (D8S264) and no. 62 (D8S261). Both of them had a relapse in the upper part of larynx. We did not detect LOH in these markers in a group of patients with a relapse in tracheostoma (Table 2). We suppose that other markers on 8p arm could be of higher significance in microsatellite analysis. In the present study, we also tested two markers which were not connected with early losses so far. The first marker is located in the region 7q31.2 (D7S486). LOH on 7q was rarely analyzed, however frequent allelic losses in band q31 were detected in almost one-half of studied informative cases. In this region, ING3 gene is located—loss of ING3 expression was related to higher mortality in the group of tongue and larynx cancer patients [32]. In our study, LOH in this region was detected in 5 out of 39 (13%) tumor samples and in 1 out of 37 (3%) normal mucosa samples from the upper part of the larynx, however what is more important we detected LOH in this marker in 4 cases analyzed during follow up study (Table 5). It suggests that loss in band 7q31.2 could be a “weak” alteration, which succumbed to the selection pressure in transformed cells, but remained in non-transformed cells, which did not develop into tumor. The second marker is located in region 18q21.1 (D18S46). As it was previously described, LOH in this chromosomal region is believed to be connected with advanced tumor stage, aggressive tumor behavior and
74
K. Szukała et al. / Mutation Research 600 (2006) 67–76
poor survival in HNSCC patients [33]. The objective was to check if LOH frequency in tumor samples in this region would be higher than in others, supposed to be lost at early stages. We did not observe such effect—LOH in this region was less frequent (11% in tumor tissue and 3% in normal mucosa). It is clear that this study should be extended for more markers; however, we cannot exclude a similar assumption as for marker D7S486—with selection pressure in tumor samples. This hypothesis is supported by the fact that in the follow up group we detected LOH in this marker in six cases—but two out of these six cases (nos. 30 and 124) had LOH also detected in tumor samples (Tables 2 and 5). On the basis of these results, we cannot point out any marker which could be a good predictor of local recurrence risk. We have observed that two patients (nos. 20 and 62) who died because of relapse in upper part of larynx had five and seven losses in tumor samples for different markers, while five patients who died because of relapse in tracheostoma region (nos. 72, 77, 82, 88 and 113) revealed maximum two losses in tumor samples. Moreover, three out of these five cases had no LOH at any marker tested. This may suggest that field cancerization process is more extended in upper part of aerodigestive tract than in lower parts, near tracheostoma region. This may also be supported by the finding that we detect no LOH at any marker tested in mucosal samples taken from tracheostoma region. On the other hand, there is a group of patients—alive, with no evidence of recurrence who had multiple losses at multiple markers (i.e. nos. 14, 38 and 124). In cases nos. 14 and 38, additionally microsatellite instability (appearing of additional allele, not observed in reference material) was detected. Follow up period lasted 39–52 months. This group displayed an overall genetic instability but so far we have no certainty if they develop a recurrence in the future. Maybe these patients have to acquire other genetic alterations to develop second tumor.
pattern more than 4 years after completion of surgical treatment. The low number of monitored cases is a result of laryngeal cancer patients’ lifestyle—continued habits, no regular appearance to the check controls. In 8 out of 10 microsatellite markers, we detected LOH in at least one case (D3S1304, D3S1568, D8S264, D9S171 and D9S1870). For three markers (D3S1300, D7S486 and D18S46), we found LOH in three, four and six cases, respectively. These results suggest that patients may be in a group of increased risk of malignant process occurrence, although in the clinical examination there was no suspicion of relapse. However, we suppose that these alterations may be “weak” and not influence tumor development process. Out of these 18 losses, only two cases—nos. 30 and 124, for marker D18S46 had LOH detected in tumor tissue (30a and 124a). The remaining 16 losses were not observed previously, what may suggest that these changes appeared already after surgery as the result of continuous exposures to carcinogens. It is also likely that appearing of new alterations may be a result of the heterogeneity of larynx. 5. Conclusions Preliminary data, obtained during this investigation, did not reveal the predictive value of LOH with respect to local relapse occurrence in laryngeal cancer patients. However, time of follow up did not reach 5 years so far, so next recurrences cannot be ruled out. We also cannot exclude the possibility that postoperative radiation has removed all changed or tumor cells left behind during surgery and those new alterations, observed during follow up, appeared already after surgery. It also seems that the number of analyzed markers should be increased: first, because of larynx cancers heterogeneity, resulting in existence of multiple subclones “carrying” different genetic alterations, and second, to improve informativeness. With respect to these results, further clinical monitoring of patients should be conducted.
4.2. Follow up study Recently, the number of studies conducted on body fluids, as well as the material from brush cytology has increased [34–38]. El-Naggar et al. observed a high incidence of LOH (49%) in saliva samples (cytological examination revealed no malignant cells), compared to 86% of LOH obtained for tumor samples [39]. Rizos et al. found significant incidence of LOH on 8p, 9p and 17q arms in cytological specimens of the larynx [35]. In this study, we used brush specimens as well as laryngeal swabs from 16 patients to investigate LOH
Acknowledgement This work was partially supported by a grant from the Foundation ORLEN-Dar Serca (ORLEN-Gift of A Heart). References [1] M.P. Tabor, R.H. Brakenhoff, V.M.M. van Houten, J.A. Kummer, M.H.J. Snel, P.J.F. Snijders, G.B. Snow, C.R. Leemans, B.J.M. Braakhuis, Persistence of genetically altered fields in head and
K. Szukała et al. / Mutation Research 600 (2006) 67–76
[2]
[3]
[4]
[5]
[6]
[7]
[8] [9] [10]
[11] [12]
[13]
[14]
[15]
[16]
[17]
[18]
neck cancer patients: biological and clinical implications, Clin. Cancer Res. 7 (2001) 1523–1532. L.B. Beder, M. Gunduz, M. Ouchida, K. Fukushima, E. Gunduz, S. Ito, A. Sakai, N. Nagai, K. Nishizaki, K. Shimizu, Genomewide analyses on loss of heterozygosity in head and neck squamous cell carcinomas, Lab. Invest. 83 (2003) 99–105. J.S. Cooper, T.F. Pajak, P. Rubin, L. Tupchong, L.W. Brady, S.A. Leibel, G.E. Laramore, V.A. Marcial, L.W. Davis, J.D. Cos, Second malignancies in patients who had head and neck cancer: incidence, effect on survival, and implications based on the ROTG experience, Int. J. Radiat. Onc. Biol. Phys. 1 (1989) 449–456. D. Slaughter, H. Southwick, W. Smejkal, Field cancerization in oral stratified squamous epithelium: clinical implications of multicentric origin, Cancer 6 (1953) 953–971. C.R. Leemans, R. Tiwari, J.J. Nauta, I. van der Waal, G.B. Snow, Recurrence at the primary site in head and neck cancer and the significance of neck lymph node metastases as a prognostic factor, Cancer 73 (1994) 187–190. W.M. Lydiatt, P.E. Anderson, T. Bazzana, M. Casale, C.J. Hughes, A.G. Huvos, D.D. Lydiatt, S.P. Schantz, Molecular support for field cancerization in the head and neck, Cancer 82 (1998) 1376–1380. M. Partridge, S. Peteromichelakis, E. Phillips, G.G. Emilion, R.P. A’Hern, J.D. Langdon, A case-control study confirms that microsatellite assay can identify patients at risk of developing oral squamous cell carcinoma within a field of cancerization, Cancer Res. 60 (2000) 3893–3898. D. Sidransky, Molecular genetics of head and neck cancer, Curr. Opin. Oncol. 7 (1995) 229–233. M.A. Nagai, Genetic alterations in head and neck squamous cell carcinomas, Braz. J. Med. Biol. Res. 32 (1999) 897–904. S.M. Gollin, Chromosomal alterations in squamous cell carcinomas of the head and neck: window to the biology of disease, Head Neck (2001) 238–252. P. Devilee, A.M. Cleton-Jansen, C.J. Cornelisse, Ever since Knudson, Trends Genet. 17 (2001) 569–573. L. Mao, J.S. Lee, Y.H. Fan, J.Y. Ro, J.G. Batsakis, S. Lippman, W. Hittleman, W.K. Hong, Frequent microsatellite alterations at chromosome 9p21 and 3p14 in oral premalignant lesions and their value in cancer risk assessment, Nat. Med. (1996) 682. J. Califano, P. van der Riet, W. Westra, H. Nawroz, G. Clayman, S. Piantadosi, R. Corio, D. Lee, B. Greenberg, W. Koch, D. Sidransky, Genetic progression model for head and neck cancer: implications for field cancerization, Cancer Res. 56 (1996) 2488–2492. J. Califano, W.H. Westra, G. Meininger, R. Corio, W.M. Koch, D. Sidransky, Genetic progression and clonal relationship of recurrent premalignant head and neck lesions, Clin. Cancer Res. 6 (2000) 347–352. K. Szukala, J. Brieger, K. Bruch, W. Biczysko, M. Wierzbicka, W. Szyfter, K. Szyfter, Loss of heterozygosity on chromosome arm 13q in larynx cancer patients: analysis of tumor, margin and clinically unchanged mucosa, Med. Sci. Monit. 10 (2004) 233–240. Ch. Jin, Y. Jin, J. Wennerberg, J. Akervall, M. Dictor, F. Mertens, Karyotypic heterogeneity and clonal evolution in squamous cell carcinomas of the head and neck, Cancer Genet. Cytogen. 132 (2002) 85–96. S. Lai, W.F. Benedict, S.A. Silver, A.K. El-Naggar, Loss of retinoblastoma gene function and heterozygosity at the RB locus in renal cortical neoplasms, Hum. Pathol. 28 (1997) 693–697. A. Goel, C.N. Arnold, D. Nied´zwiecki, D.K. Chang, L. Ricciardiello, J.M. Carethers, J.M. Dowell, L. Wasserman, C. Compton,
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
75
R.J. Mayer, M.M. Bertagnolli, C.R. Boland, Characterization of sporadic colon cancer by patterns of genomic instability, Cancer Res. 63 (2003) 1608–1614. C. Dotzenrath, B.T. Teh, F. Farnebo, K. Cupisti, A. Svensson, A. Toell, P. Goretzki, C. Larsson, Allelic loss of the retinoblastoma tumor suppressor gene: a marker for aggressive parathyroid tumors? J. Clin. Endocrinol. Metab. 81 (1996) 3194–3196. N.A. Thomas, P.J. Neville, S.W. Baxter, I.G. Campbell, Genetic analysis of benign ovarian tumors, Int. J. Cancer 105 (2003) 499–505. G. Eiriksdottir, G. Johannesdottir, S. Ingvarsson, I.B. Bjornsdottir, J.G. Jonasson, B.A. Agnarsson, J. Hallgrimsson, J. Gudmundsson, V. Egilsson, H. Sigurdsson, R.B. Barkardottir, Mapping loss of heterozygosity at chromosome 13q: loss at 13q12-q13 is associated with breast tumour progression and poor prognosis, Eur. J. Cancer 34 (1998) 2076–2081. J.T. Dong, J.C. Boyd, H.F. Frierson, Loss of heterozygosity at 13q14 and 13q21 in high grade, high stage prostate cancer, Prostate 49 (2001) 166–171. K.W. Ah-See, T.G. Cooke, I.R. Pickford, D. Soutar, A. Balmain, An allelotype of squamous carcinoma of the head and neck using microsatellite markers, Cancer Res. 54 (1994) 1617–1621. H. Nawroz, P. van der Riet, R.H. Hruban, W. Koch, J.M. Ruppert, D. Sidransky, Allelotype of head and neck squamous cell carcinoma, Cancer Res. 54 (1994) 1152–1155. H. Rowley, A. Jones, D. Spandidos, J. Field, Definition of a tumor suppressor gene locus on the short arm of chromosome 3 in squamous cell carcinoma of the head and neck by means of microsatellite markers, Arch. Otolaryngol. Head Neck Surg. 122 (1996) 497–501. P. van der Riet, H. Nawroz, R.H. Hruban, R. Corio, K. Tokino, W. Koch, D. Sidransky, Frequent loss of chromosome 9p21-22 early in head and neck cancer progression, Cancer Res. 54 (1994) 1156–1158. L. Roz, C.L. Wu, S. Porter, C. Scully, P. Speight, A. Read, P. Sloan, N. Thakker, Allelic imbalance on chromosome 3p in oral dysplastic lesions: an early event in oral carcinogenesis, Cancer Res. 56 (1996) 1228–1231. Z. Guo, K. Yamaguchi, M. Sanchez-Cespedes, W.H. Westra, W.M. Koch, D. Sidransky, Allelic losses in OraTest-directed biopsies of patients with prior upper aerodigestive tract malignancy, Clin. Cancer Res. 7 (2001) 1963–1968. M. Partridge, G.G. Emilion, S. Pateromichelakis, E. Phillips, J.D. Langdon, Location of candidate tumour suppressor gene loci at chromosomes 3p, 8p and 9p for oral squamous cell carcinomas, Int. J. Cancer 83 (1999) 318–325. J.B. Sunwoo, P.C. Sun, V.K. Gupta, A.P. Schmidt, S. El-Mofty, S.B. Scholnick, Localization of a putative tumor suppressor gene in the sub-telomeric region of chromosome 8p, Oncogene 18 (1999) 2651–2655. U. Bockm¨uhl, Ch.S. Ishwad, R.E. Ferrell, S.M. Gollin, Association of 8p23 deletions with poor survival in head and neck cancer, Otolaryngol. Head Neck Surg. 124 (2001) 451–455. M. Gunduz, M. Ouchida, K. Fukushima, S. Ito, Y. Jitsumori, T. Nakashima, N. Nagai, K. Nishizaki, K. Shimizu, Allelic loss and reduced expression of the ING3, a candidate tumor suppressor gene at 7q31, in human head and neck cancers, Oncogene 21 (2002) 4462–4470. S. Takebayashi, A. Hickson, T. Ogawa, K.Y. Jung, H. Mineta, Y. Ueda, R. Grenman, S.G. Fisher, T.E. Carey, Loss of chromosome arm 18q with tumor progression in head and neck squamous cancer, Genes Chromosomes Cancer 41 (2004) 145–154.
76
K. Szukała et al. / Mutation Research 600 (2006) 67–76
[34] J.A. Veltman, A.H.N. Hopman, F.J. Bot, F.C.S. Ramaekers, J.J. Manni, Detection of chromosomal aberrations in cytologic brush specimens from head and neck squamous cell carcinoma, Cancer 81 (1997) 309–314. [35] E. Rizos, G. Sourvinos, D.A. Spandidos, Loss of heterozygosity at 8p, 9p and 17q in laryngeal cytological specimens, Oral Oncol. 34 (1998) 519–523. [36] M.F. Spafford, W.M. Koch, A.L. Reed, J.A. Califano, L.H. Xu, C.F. Eisenberger, L. Yip, P.L. Leong, L. Wu, S.X. Liu, C. Jeronimo, W.H. Westra, D. Sidransky, Detection of head and neck squamous cell carcinoma among exfoliated oral mucosal cells by microsatellite analysis, Clin. Cancer Res. 7 (2001) 607–612. [37] S. Lopes Bittencourt Rosas, W. Koch, M. da Gloria da Costa Carvalho, L. Wu, J. Califano, W. Westra, J. Jen, D. Sidransky,
Promoter hypermethylation patterns of p16, O6 -methylguanineDNA-methyltransferase, and death-associated protein kinase in tumors and saliva of head and neck cancer patients, Cancer Res. 61 (2001) 939–942. [38] S. Temam, J. Benard, C. Dugas, M. Trassard, E. Gormally, J.C. Soria, S. Faivre, B. Luboinski, P. Marandas, P. Hainaut, G. Lenoir, L. Mao, F. Janot, Molecular detection of early-stage laryngopharyngeal squamous cell carcinoma, Clin. Cancer Res. 11 (2005) 2547–2551. [39] A.K. El-Naggar, L. Mao, G. Staerkel, M.M. Coombes, S.L. Tucker, M.A. Luna, G.L. Clayman, S. Lippman, H. Goepfert, Genetic heterogeneity in saliva from patients with oral squamous carcinomas, J. Mol. Diagn. 3 (2001) 164–170.