High Incidence of Null-Type Mutations of the TP 53 Gene in Japanese Patients with Head and Neck Squamous Cell Carcinoma

Objective: Molecular targeting therapy has not been generally established in head and neck squamous cell carcinoma (HNSCC) except for cetuximab treatment for targeting epidermal growth factor receptor (EGFR). We analyzed alterations of the TP53, KRAS2, and EGFR genes in Japanese HNSCC to identify subpopulations of tumors potentially susceptible or not susceptible to specific therapy based on their genetic alterations. Materials and Methods: A total of 56 Japanese subjects were included in this study. Genomic DNA of exons 5 9 of the TP53, exons 1 and 2 of the KRAS2, exons 19 22 of the EGFR, and their flanking sequences were amplified by polymerase chain reaction (PCR) followed by direct sequencing. Splicing variants of EGFR were examined by reverse transcription (RT)-PCR. Results: Mutations of the TP53 and KRAS genes were detected in 25 (45%) and 2 (4%) of 56 HNSCC cases, respectively, while neither mutation nor splicing variant of EGFR was observed. The TP53 mutation did not correlate with clinical stages or primary sites of the tumors. The patterns of nucleotide substitutions specific to HNSCC were not observed. However, the incidence of null-type mutations of the TP53, which cannot be detected as abnormal by conventional immunohistochemical (IHC) studies, was significantly higher (10/25; 40%) than that of HNSCC reported in other countries. Conclusion: Frequent TP53 mutations, especially null-type mutations, but infrequent or no alterations of the KRAS and EGFR suggest that the sequencing analysis of the TP53 mutation rather than IHC analysis of p53 provides a potentially useful marker to predict the response of HNSCC to chemotherapy or radiotherapy.


Introduction
Head and neck squamous cell carcinomas (HNSCCs) account for 6% of all cancers worldwide, affecting more than 650,000 patients every year.Most of these cases are classified as the advanced stage III/IV, which is characterized by a poor prognosis with a survival rate of only 40% and little improvement with multi-modality therapy [1] [2].In the past decade, the introduction of molecular targeting therapy had improved the prognosis and resulted in therapeutic gains in patients with HNSCC.Treatment with anti-epidermal growth factor receptor (EGFR) monoclonal antibodies, such as cetuximab, is currently used for HNSCC, and a number of additional drugs are being tested for clinical use.To improve the effectiveness of the molecular targeting therapies, it is important to identify biomarkers related to clinical gains.Presence of the KRAS gene mutation has been established as a useful indicator for predicting poor response to cetuximab treatment in colorectal cancer [3].Since KRAS acts in the downstream cascade of EGFR, its activating mutation would cancel the suppressor effect of the upstream molecule, EGFR, with cetuximab treatment.However, neither screening nor extensive analyses of the KRAS gene mutation have been performed, at least not in Japanese cases of NHSCC.
One successful example of molecular targeting therapy against solid tumors is that of EGFR tyrosine kinase inhibitors, such as gefitinib or erlotinib, for a subset of lung adenocarcinomas, where the presence of the EGFR mutation serves as a very strong indicator of a good response to the therapy.In addition, it is reported that a specific splicing variant of EGFR (vIII) is overexpressed in a subset of tumors and could provide a predictive marker of poor prognosis in glioblastoma patients [4].Only a few studies, however, have reported the incidence of the EGFR mutation or splicing variants of EGFR, at least in Japanese cases of HNSCC.
Another potentially useful biomarker of cancer treatment is the TP53 mutation because it is the most frequent mutation observed in human cancers and is deeply involved in the cellular responses against chemotherapy and radiotherapy.In fact, it has been reported that the presence of TP53 mutations serves as an essential biomarker that correlates well with the efficacy of induction chemotherapy [5]- [7], the local control rate of radiotherapy [8]- [10], and the survival rate of HNSCC patients [11] [12].Mutation of the TP53 is usually tested by immunohistochemical (IHC) analysis because mutant p53 proteins generated by missense mutations show a longer halflife and can be detected by IHC analysis, whereas wild-type p53 proteins cannot be detected by IHC analysis due to their shorter half-lives.On the other hand, null-type mutations of the TP53 cannot be detected by IHC analysis, which might be one reason why the usefulness of the p53 mutation in predicting cancer response to chemotherapy or radiotherapy is limited and controversial [13]- [15].Although a number of studies on the TP53 mutation have been reported for HNSCC from the USA and Europe, only a few reports are available for HNSCC in Japanese and Asian populations.
In this study, we analyzed mutations of the TP53, KRAS, and EGFR genes as well as a splicing variant of EGFR in 56 Japanese cases.A high incidence of null-type mutations of the TP53 gene strongly suggests that sequencing, rather than IHC analysis, is necessary to examine the mutation status of the TP53 gene as a possible biomarker for the treatment of Japanese HNSCC.

Surgical Materials
A total of 56 Japanese subjects (48 men and 8 women; median age: 65 years old; age range: 24 -87 years old) provided informed, written consent in accordance with the ethics board of the University of Tokyo (ID: 20-39-1016) and were included in this study.All subjects were diagnosed as having HNSCCs and were operated on at the Department of Otolaryngology-Head and Neck Surgery at the University of Tokyo from April, 2008, to June, 2009.Primary sites were the oral cavity in 21 cases, the oropharynx in 8 cases, the hypopharynx in 12 cases, the larynx in 13 cases, the nasal cavity in 1 case, and the auditory canal in 1 case (Table 1).The clinical stages were I, II, III, and IV in 4, 9, 15, and 28 cases, respectively, including 12 recurrent cases (Stage III in 4 cases and Stage IV in 8 cases).The majority of subjects in this cohort (71%; 35 of 49 patients) were heavy smokers (≥25 pack years).Specimens were collected at surgical resection and were quickly frozen in liquid nitrogen and stored at −80˚C or preserved in Allprotect Tissue Reagent (Qiagen, Alameda, CA, USA).

Sequencing Analysis
Genomic DNA was extracted using the AllPrep Mini Kit (Qiagen).Exons 1 and 2 of the KRAS, exons 19 -22 of the EGFR, exons 5 -8 of the TP53 genes, and their flanking sequences were amplified by polymerase chain reaction (PCR) using KOD FX polymerase (Toyobo, Osaka, Japan).Primers used for PCR are described in Table S1.The PCR mixtures contained 10 µl of 2× PCR buffer, 4 µl of 2 mM dNTPs, 0.4 unit of KOD FX, 0.6 µl of forward and reverse primers (each 10 µM), 40 ng of template DNA, and distilled water to make a reaction mixture of 20 µl.PCR cycles were as follows: denaturing at 94˚C for 2 min followed by 35 cycles of denaturing at 98˚C for 5 sec, annealing at 57˚C for 10 sec, and extension at 68˚C for 10 sec, with a final extension at 68˚C for 7 min.PCR products were separated by electrophoresis on agarose gel, and fragments were cut from the gel and purified by Wizard SV Gel and PCR Clean-Up System (Promega, Madison, WI, USA).Sequencing was conducted by reactions with Applied Biosystems' BigDye 3.1 kit (Applied Biosystems, Foster City, CA, USA) on a 3130xl DNA sequencer.Sequence results were analyzed using Lasergene (Dnastar, Madison, WI, USA), and statistical analysis was done using SPSS software.The significance level was set at P < 0.05.

Detection of Splicing Variants
Total cellular RNA were extracted using the FastTrack 2.0 Kit (Invitrogen, Carlsbad, CA) and reverse-transcribed by Superscript II (Invitrogen, Carlsbad, CA) as described previously [16].A 1037-bp cDNA fragment of EGFR corresponding to exons 1 -8 and a possible 236-bpfragment of EGFRvIII were amplified by PCR using a pair of primers, EGFRvIIIOneStepF and EGFRvIIIOneStepR as described previously [17], while a 452-bp fragment of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was amplified as an internal control in the same reaction.Primers used for PCR are described in Table S1.The PCR product was subjected to electrophoresis on agarose gel as described previously [16].

Analysis of the KRAS and EGFR Mutations as Well as Splicing Variants of EGFR
Sequencing analysis of exons 1 and 2 and their flanking regions of the KRAS gene identified 2 activating mutations (3.6%) out of 56 HNSCC cases (Figure 1(a), Figure 1(b)).The G to A transition at codon 14 (valine to isoleucine) is an uncommon but an already reported activating mutation of the KRAS gene.Interestingly, both cases are oropharyngeal HNSCCs (2 of 8; 25%); one is in clinical Stage III and the other is in IV.In contrast, none of the other 48 HNSCCs shows any KRAS mutation, suggesting that the KRAS mutation could not provide a practical marker for predicting response to cetuximab treatment in Japanese HNSCCs except for oropharyngeal squamous cell carcinomas (SCCs).
On the other hand, sequencing analysis of exons 19 -22 and their flanking regions of the EGFR gene did not show any mutations.Moreover, reverse transcription (RT)-PCR analysis covering the sequences corresponding to exons 1 -8 did not show any extra products, suggesting that the splicing variant III, which is reported to be present in glioblastomas and to be associated with poor prognosis for glioblastoma patients, is not expressed in HNSCC.

High Incidence of Null-Type Mutations of the TP53 Gene in Japanese HNSCCs
Since most of the TP53 mutations were found in its hotspot corresponding to exons 5 -9, we amplified the genomic DNA fragments of exons 5 -9 as well as their flanking sequences by PCR followed by direct sequencing.Representative results are shown in Figure 2, in which a missense mutation in codon 134 from TTT (phenylalanine) to TGT (cysteine) in exon 5 (Figure 2(a)), a nonsense mutation in codon 242 from TGC (cysteine) to TGA (stop) in exon 7 (Figure 2(b)), and a nonsense mutation in codon 306 from CGA (arginine) to TGA (stop) in exon 8 of the TP53 gene (Figure 2(c)) are demonstrated.Similar analysis identified the TP53 mutations in 25 (44.6%) of 56 primary HNSCC cases.Clinicopathological analyses of HNSCC-carrying mutations demonstrate that the TP53 mutation does not significantly correlate with the clinical stages of HNSCC tumors (Table 1).It also does not correlate with the primary sites of the tumors, although the incidence in oropharyngeal HNSCC (2 of 8; 25%) is relatively low.Regarding the patterns of nucleotide substitutions, the G:C to A:T substitution is the most frequent change (36%), followed by the G:C to T:A (24%) and A:T to G:C (20%) substitutions (Table S2).

Discussion
In the treatment of both solid tumors and lymphoma/leukemias, the importance of molecular targeting therapies   in association with specific genetic markers is growing.However, it is not well established in HNSCC, except for cetuximab treatment.In the present study, we examined mutations of the KRAS, EGFR, and TP53 genes as possible genetic markers for molecular targeting therapy of HNSCC.Specific mutations in exons 19 -22 of the EGFR gene are potent genetic markers for selecting a subset of lung adenocarcinomas that will respond well to treatment with EGFR tyrosine kinase inhibitors (TKIs), such as gefitinib and erlotinib.However, we failed to detect any EGFR mutations in 56 HNSCC cases, supporting the current findings that EGFR TKI treatment is not effective against HNSCC.

G A T G T N T T G C T T C C T G M A T G G G C T A A G T G A G G
On the other hand, the KRAS gene mutation is known to serve as a marker for a subpopulation of colorectal cancer that is resistant to molecular targeting therapy by anti-EGFR antibodies, such as cetuximab, because the activated KRAS mutant acts in the downstream cascade of EGFR.Although several studies have demonstrated that the clinical gain with cetuximab treatment alone is limited for HNSCC patients, we failed to detect a high incidence of the KRAS mutation in Japanese HNSCC.It is noteworthy, however, that a relatively high incidence (2 of 8; 25%) of the KRAS mutation was detected in oropharyngeal HNSCC in this study, because previous studies reported that the KRAS gene was mutated in none of 71, none of 12 or only 4 of 123 (totally, 4 of 204; 2%) oropharyngeal HNSCCs [17]- [19].Further studies focusing on the oropharyngeal HNSCCs would be required to assess the response to cetuximab according the presence or absence of the KRAS mutation as well as the human papilloma virus (HPV) infection.
The TP53 mutation is not a genetic marker for molecular targeting therapy.However, it is well known that it provides a good predictive marker for resistance to chemotherapy and radiotherapy and for poor prognoses in various cancers, including HNSCCs [5]- [12].In this study, we found the TP53 mutation in 25 (45%) of 56 Japanese HNSCCs.This incidence is comparable to those reported previously [7] [9]- [11] (Table 2) and shows no correlation to the clinical stages of the tumors (Table 1).Regarding the sites of the tumors, it is noteworthy that the incidence of the TP53 mutation is relatively low (25%) in oropharyngeal HNSCC.Since infection with the human papillomavirus (HPV) is specifically observed in this type of tumor, the p53 function of a portion of the oropharyngeal HNSCCs is inactivated by protein interaction with the HPV E6 protein instead of by the gene mutation.The patterns of the nucleotide substitutions in our study are essentially the same as those of previous studies on HNSCC or those summarized by the International Agency for Research on Cancer (IARC) in the TP53 database [20] [21].The relatively high incidence of G:C to T:A transversions in HNSCC appears to be caused by tobacco carcinogens, including benzopyrene.
On the other hand, when we consider the mutation spectrum, the incidence of null-type mutations, especially that of nonsense mutations, is clearly higher than that of previous reports or the TP53 database of the IARC for HNSCC [7] [9]- [11] [22].The relatively high incidence of null-type mutations of the TP53 in HNSCC (48% in our study and 32% of HNSCCs in the IARC database) in comparison with that of all tumor types in the IARC database (18%) suggests that loss of the p53 function by null-mutation would be beneficial in HNSCC.It is noteworthy that among 56 HNSCCs examined, 12 were derived from recurrent tumors, in which only 2 (17%) tumors showed the TP53 mutation with nonsense mutations, suggesting that the therapeutic intervention might not play a role in the high incidence of null-type TP53 mutations in HNSCC.On the other hand, a markedly high incidence of nonsense mutations (40%) in this study as compared with that in the IARC TP53 database for HNSCC (13%) suggests that Japanese HNSCC might develop through some unique etiological background as compared with HNSCC cases from Europe or the USA.Further studies would be required to determine the etiologic factors as well as the significance of the TP53 mutation in Japanese HNSCCs.
At present, mutant p53 is screened by IHC staining of tumors in many hospitals and laboratories because aberrant p53 proteins with missense mutations show a longer half-life than do wild-type p53 proteins, and, therefore, only mutant p53 with missense mutations, but not wild-type p53, can be detected by IHC analysis.It should be noted, however, that p53 proteins with null-type mutations cannot be detected by IHC analysis due to the nonsense-mediated RNA decay and the resultant decreased amount of the protein with truncation.Therefore, our findings demonstrate that the sequencing analysis of the TP53 gene, rather than IHC analysis, is essential for assessing the mutational state of the TP53 in Japanese HNSCCs for the genetic marker to predict the response to chemotherapy, local control of radiotherapy, and the overall survival rate.A systematic approach using nextgeneration sequencing of the surgically resected primary tumors would be beneficial for determining the mutational state of the relevant genes and for establishing a gene-based medicine for HNSCC.The limitation of this study, however, is that we have not systematically examined the clinical features of the HNSCC cases with or without the TP53 mutations.Further study would be necessary to assess the significance of the TP53 mutation in predicting the clinical outcome of 56 cases of HNSCCs.

Conclusion
To assess the genetic markers for molecular targeting and gene-based therapy, mutations of the EGFR, KRAS, and TP53 genes were examined in 56 Japanese HNSCCs.No mutation of the EGFR gene supports the fact that gefitinib is not effective against HNSCC.A low incidence of the KRAS mutation (3.6%) demonstrates that the KRAS mutation is not an effective marker for predicting response to cetuximab treatment, which shows a limited clinical gain in HNSCC.On the other hand, a high incidence of the TP53 mutation (45%) suggests that screening for the TP53 mutation could provide a potentially useful marker for predicting responsiveness to chemotherapy, the rate of local control in radiotherapy, or the overall survival with HNSCC, although further studies would be required to assess the significance of the TP53 mutation in predicting the clinical features of HNSCC in individual cases.On the basis of a high incidence of null-type mutations in Japanese HNSCCs in the present study, we propose that sequencing analysis, instead of IHC analysis, is essential for establishing the TP53 mutation as a genetic marker for the optimal treatment of HNSCC.

Figure 1 .Figure 2 .
Figure 1.Mutation of the KRAS gene in HNSCC.(a) The glycine codon (GGT) at position 12 is mutated to the asparagine codon (GAT) in an oropharyngeal SCC with T2N1M0.(b) The valine codon (GTA) at position 14 is mutated to the isoleucine codon (ATA) in an oropharyngeal SCC with T3N2bM0.

Table 1 .
Clinical stages and primary sites of HNSCC tumors with mutation of the TP53 and KRAS genes.

Table 2 .
Mutation spectra of the TP53 gene in HNSCC.
*Null type mutation includes nonsense mutation, frameshift mutation, and splicing mutation.