OF HPV-NEGATIVE ORAL CARCINOMAS

Background/Aim. Oral squamous cell carcinoma (OSCC) is the most common tumor type of head and neck carcinomas, characterized by a high recurrence rate and poor survival. Further elucidation of the function and regulation of the TP53 , a pivotal tumor suppressor gene, would provide advances in predicting the clinical behavior, prognosis and chemotherapy response of OSCC patients. Thus, we investigated the association of TP53 gene mutations with survival and response to cisplatin chemotherapy in HPV-negative OSCC patients. Methods. The potential clinical relevance of TP53 mutations was analyzed in 82 patients with HPV-negative OSCC. All patients underwent radiotherapy and 25 patients received cisplatin chemotherapy. A negative HPV status was determined by type-specific PCR, for high-risk HPV 16, 18, 31, and 33. Targeted sequencing of TP53 exons 4-8 was assessed by Sanger sequencing. Results. Of 82 HPV-negative OSCC patients, 49 (59.79%) had TP53 mutation, and 26 patients (31.7%) carried pathogenic TP53 mutations. Patients with pathogenic TP53 mutations had significantly reduced overall survival, p=0.009. Recurrences status, but not TP53 mutations, was an independent marker of poor survival in our cohort, HR=4.733, [2.027-11.053], p=0.0001. In the subcohort of patients who underwent cisplatin-based chemotherapy, pathogenic TP53 mutations were predictors of poor response to chemotherapy, p=0.026. Conclusions. Our findings indicate that pathogenic TP53 mutations in HPV-negative OSCC tumors could be a prognostic marker of patients’ reduced overall survival. In addition, pathogenic TP53 mutations in HPV-negative OSCC could be a marker of poor chemotherapy response.


Introduction
Oral squamous cell carcinoma (OSCC) is the most common tumor type of head and neck carcinomas, characterized by a high recurrence rate and poor survival. This malignancy is the sixth most common cancer worldwide in men and eighth in women in developed countries, and third most common cancer in men and fourth in women in developing countries (1), which affects approximately 600000 new patients every year worldwide. Oral carcinogenesis is a multi-step process, which encompasses an accumulation of genetic and epigenetic changes that lead to the disruption of the various signalling pathways controlling the cell cycle, proliferation, apoptosis, senescence, and DNA (2). Genetic changes are progressively accumulated and inactivation of tumor suppressor genes, by point mutations, deletions and gene rearrangement, is one of the key changes for malignant transformation. Known etiological factors for developing OSCC are predominantly smoking, alcohol intake and poor oral hygiene. Approximately 20-30% OSCC cases can be associated with tobacco smoking and 7-19% with heavy alcohol drinking, which increases the risk of oral cavity cancer 2-3 times (3). One of the most important advances in oral carcinogenesis in recent decades is evidence of an association of oral cancer and some types of Human Papilloma Virus (HPV) infection, predominantly HPV16 (4).
Gene coding for protein p53 (TP53) is one of the most prominent tumor suppressor genes located on the short arm of chromosome 17 (17p13.1) (5). p53 is a key factor in a signaling pathway that helps the cell to recover from DNA damage (6). Upon a DNA damage, the wild type p53 arrests cell cycle in the G1 phase prohibiting transition to S phase until the damage is repaired. Additionally, throughout the retinoblastoma tumor suppressor pathway, p53 can direct cells to a state of permanent cell cycle arrest, or induce pro-apoptotic genes cellular senescence (7), . TP53 is one of the most frequently mutated human genes in more than 50% of cancers. Germline TP53 mutations cause Li-Fraumeni syndrome, a rare autosomal, hereditary disorder predisposing to sarcoma, breast cancer, leukemia, adrenal gland carcinoma (8).
TP53 mutations are early alterations during oral carcinogenesis and more than 25000 mutations have been discovered so far (9). Most of them, 70%, are missense mutations in the coding regions (10), where approximately 30% of mutations occur in exon 7 and exon 8, known as mutation hot spots. These exons code for the DNA binding domain, preventing the p53 binding to the promoter of target genes (11). Common TP53 codon 72 gene polymorphism in this domain produces two functional variants of p53 (p53P (Proline) and p53A (Arginine), which could reduce its ability to mediate apoptosis and cell cycle arrest, and therefore could affect the survival and chemotherapy response (12). A number of studies reported inconsistent findings regarding whether the TP53 mutations and codon 72 polymorphism influence survival and chemotherapy response in OSCC patients (13).
Inactivation of wild type p53 can also be achieved throughout human papillomavirus (HPV) E6 protein (11). HPV-positive oropharyngeal cancer cells have a different molecular profile from HPV-negative oropharyngeal cancer cells, where HPV-negative oropharyngeal cancers have more frequent loss of heterozygosity of 3p, 9p or 17p chromosomal regions (14). HPVnegative oropharyngeal cancers have at least two times more mutations compared to HPVpositive tumors (15) (16), and worse outcome (17), indicating the necessity of molecular characterization of p53 in HPV-negative OSCC.
Further elucidation of the function and regulation of the p53 in HPV-negative OSCC would provide advances in predicting the clinical behavior, prognosis and patients' chemotherapy response. Finding the potential markers that could predict tumor response to chemotherapy, developing new strategies with therapeutics targeting different pathways that will override the resistance and tumor molecular profiling would provide an individualized approach to the treatment modalities of OSCC patients.

Materials and Methods
The current study was approved by the  (20). Missense, stop-gain, in-frame insertions/deletions, frameshift and splice site TP53 mutations with pathogenic and likely pathogenic significance, and criteria provided by multiple or single submitters, reviewed by expert panels or given in practice guidelines, were classified as pathogenic mutations. On the other hand likely benign, protective or with uncertain significance were classified as non-pathogenic mutations.
Statistical analysis. Obtained data were analyzed by SPSS 20.0 software (IBM Inc., Chicago, IL, USA). Contingency tables were assessed by χ 2 -test or Fisher's exact test. Overall survival was calculated from the date of diagnosis until death from any cause. Kaplan-Meier survival curves were compared using the log-rank test. Cox proportional hazard regression analysis was performed to estimate the hazard ratios (HR), with 95% confidence interval (95% CI).
Variables found significant in the univariate analysis, including those with significance level below 20%, were subsequently analyzed in multivariate Cox's regression. The Cox model was performed using the forward stepwise method, that removed variables with p<0.1. The associations were considered as significant when p values were less than 0.05.

Results
Association of p53 gene mutations, pathogenic p53 mutations and polymorphism p72 with demographic and clinicopathological features of HPV-negative OSCC patients.
Eighty-two HPV-negative OSCC samples were screened for TP53 mutations in exons 4 -8, and mutations were found in a total of 49 patients (59.8%). TP53 mutations were classified as pathogenic and non-pathogenic, as previously suggested (19) (20). The list of detected TP53 mutations and their classification according to clinical significance, assessed by ClinVar database and Simple ClinVar web server, is given in Supplement Table 1. Pathogenic p53 mutations were detected in 26 of 82 (31.7%) of OSCC patients.
The association of TP53 gene mutations, pathogenic TP53 mutations and polymorphism p72 with demographic and clinicopathological features are presented in Table 1. No association was found between TP53 mutations or pathogenic TP53 mutations and sex, or smoking.
Pathogenic TP53 mutations were significantly associated with age, p=0.005, and high alcohol intake, p=0.009, Table 1. Locally advanced tumors did not have a statistically higher TP53 mutation rate, or pathogenic TP53 mutations, compared to early-stage OSCC. Mutations in exon 4 of p53 gene were significantly associated with histological and nuclear grade (p=0.012 and p=0.032, respectively), while mutations in exon 7 were associated with smoking status, p=0.017.  In the subgroup of 25 patients who received chemotherapy in our cohort, when all p53 mutations were taken into account, p53 mutation status was not associated with chemotherapy response, p=0.641, Figure 2A. However, overall survival in patients who had received cisplatin chemotherapy was significantly shorter for those with pathogenic p53 mutations compared to patients with wild type p53, p=0.026, Figure 2B. Non-pathogenic TP53 mutations in patients who had received cisplatin chemotherapy were not related to OS in our cohort (p=0.453, log-rank test). These findings indicate that the response to chemotherapy was associated with the type of p53 mutation, and that the patients with pathogenic p53 mutations were resistant to platinum-based chemotherapy, as opposed to the patients with wild type p53. The Cox Regression analysis demonstrated that the high alcohol intake, stage, tumor size, nodal status and recurrences are highly associated with hazard risk,   Our findings indicate that HPV-negative OSCC patients with pathogenic p53 mutations had a significantly lower survival rate. In the subcohort of patients who underwent cisplatin-based chemotherapy, overall survival was significantly shorter for those with pathogenic p53 mutations than those with wild type TP53. In contrast, when all TP53 mutations were taken into account, TP53 mutation status was not associated with overall survival. These findings indicate that the overall survival and the resistance to platinum-based chemotherapy in OSCC could be associated with the type of TP53 mutation, and that pathogenic TP53 mutations are a significant predictor of poor overall survival as opposed to benign or likely-benign mutations.
Based on the p53 mechanism of action, as one of the key cell cycle regulators after DNA damage, a number of trials investigate the association between TP53 mutation and survival, as well as radio-and chemotherapy-response. Our findings of the high incidence of TP53 mutations in HPV-negative OSCC are in accordance with previous studies, where TP53 is mutated in approximately 50% of HNSCC cases (1). Mutations in the DNA-binding domain of TP53 may influence individual responsiveness to chemotherapy via its ability to mediate apoptosis and cell cycle arrest (12). The most frequent genetic change in our study was TP53 codon 72 polymorphism, but it was not associated with prognosis or chemotherapy response. Mutated TP53 was previously associated with shorter overall survival and poor radio and chemotherapy response, which indicates its potential as a marker for a clinical course in OSCC patients. In the study of locally advanced oral cancer patients, which received cisplatin chemotherapy, patients carrying the high-risk p53 mutations had reduced cisplatin sensitivity and 10 times greater risk for residual disease compared to patients with low-risk mutations (39).
Lower response to cisplatin-based chemotherapy in patients with TP53-mutated tumors (32), suggested the potential clinical use of p53-based therapeutics in restoring the p53 function. As a result of p53 adenoviral mono-therapy or in combination with radio and chemotherapy, tumor regression was observed (40). OSCC patients carrying TP53 mutations had a 2.7 times higher risk for cisplatin and 5-Fu based therapy resistance, compared to patients with functional p53 (33). In addition, a strong connection between nonfunctional p53 and low response rate to cisplatin-based neoadjuvant chemotherapy was demonstrated in OSCC patients (34). Another potentially promising approach is treatment with small molecules that reactivate mutated p53, using PRIMA-1 (P53 Reactivation and Induction of Massive Apoptosis) as a single agent and in combination with standard chemotherapy (41). PRIMA -1 therapy is more active in cell lines containing mutant p53 than wild type p53 cells, and results in the increased expression of p53-target genes p21, Bax, Puma and Noxa (41). Another p53 reactivating molecule RITA (Reactivation of the p53 and Induction of Tumor cell Apoptosis) induces p53 accumulation and reactivation, promotes apoptosis via p21, BAX and caspase-3 upregulation, and induces growth inhibition in OSCC cells in vitro and in vivo (42).
In conclusion, our findings indicate that pathogenic TP53 mutations in HPV-negative OSCC tumors could be a prognostic marker of patients reduced overall survival. In addition, HPV-