Clinical Impact of TP53 Mutations in Patients with Head and Neck Cancer Who Were Treated with Targeted Therapies or Immunotherapy

Article information

Cancer Res Treat. 2025;57(3):709-719

Publication date (electronic) : 2024 December 23

doi : https://doi.org/10.4143/crt.2024.836

Eun Joo Kang¹, Shinwon Hwang², Yun-Gyoo Lee³, Jong-Kwon Choi⁴, Seong Hoon Shin⁵, Yoon Hee Choi⁶, Keun-Wook Lee⁷, Hyun Woo Lee⁸, Min Kyoung Kim⁹, Seung Taek Lim¹⁰, Hwan Jung Yun¹¹, Sang-Gon Park¹², Sangwoo Kim¹³, Sung-Bae Kim^,14, Hye Ryun Kim^,15

¹Division of Hemato-oncology, Department of Internal Medicine, Korea University Guro Hospital, Korea University College of Medicine, Seoul, Korea

²Department of Dermatology, Yonsei University College of Medicine, Seoul, Korea

³Department of Internal Medicine, Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, Seoul, Korea

⁴Division of Hematology and Medical Oncology, Department of Internal Medicine, Konyang University Hospital, Daejeon, Korea

⁵Department of Internal Medicine, Kosin University Gospel Hospital, Busan, Korea

⁶Division of Hematology-Oncology, Department of Internal Medicine, Dongnam Institute of Radiological and Medical Sciences, Busan, Korea

⁷Department of Internal Medicine, Seoul National University Bundang Hospital, Seoul National University College of Medicine, Seongnam, Korea

⁸Department of Hematology-Oncology, Ajou University School of Medicine, Suwon, Korea

⁹Division of Hematology-Oncology, Department of Internal Medicine, Yeungnam University Hospital, Yeungnam University College of Medicine, Daegu, Korea

¹⁰Department of Hematology and Medical Oncology, Wonju Severance Christianity Hospital, Wonju, Korea

¹¹Department of Internal Medicine, Chungnam National University Hospital, Daejeon, Korea

¹²Department of Hemato-oncology, Chosun University Hospital, Gwangju, Korea

¹³Department of Biomedical Systems Informatics and Graduate School of Medical Science, Yonsei University College of Medicine, Seoul, Korea

¹⁴Department of Oncology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea

¹⁵Divison of Medical Oncology, Department of Internal Medicine, Yonsei Cancer Center, Yonsei University College of Medicine, Seoul, Korea

Correspondence: Sung-Bae Kim, Department of Oncology, Asan Medical Center, University of Ulsan College of Medicine, 88 Olympic-ro 43 gil, Songpa-gu, Seoul 05505, Korea Tel: 82-2-3010-3217 E-mail: sbkim3@amc.seoul.kr

Co-correspondence: Hye Ryun Kim, Division of Medical Oncology, Department of Internal Medicine, Yonsei Cancer Center, Yonsei University College of Medicine, 50-1 Yonse-ro, Seodaemun-gu, Seoul 03722, Korea Tel: 82-2-2228-8130 E-mail: NOBELG@yuhs.ac

*Eun Joo Kang and Shinwon Hwang contributed equally to this work.

Received 2024 August 28; Accepted 2024 December 21.

Abstract

Introduction

The tumor suppressor p53 (TP53) gene, a crucial tumor suppressor gene that has been implicated in various cancers, encodes the p53 protein. This protein controls cell division and death at a low level in normal tissue [1,2]. However, mutations in the TP53 gene due to numerous endogenous or exogenous stressors can cause a loss or gain of function of the wild-type TP53, which promotes the proliferation, invasion, and metastasis of cancer cells. The TP53 mutations are frequently detected in head and neck squamous cell carcinomas (HNSCCs) and are associated with shorter survival times and resistance to chemotherapy or radiotherapy [3].

Genomic sequencing techniques are being increasingly integrated into routine diagnostic processes for precision medicine. Thus, clinicians can now readily identify patients with HNSCC who harbor the TP53 mutations via next-generation sequencing (NGS) in a clinical setting. Concurrently, with the advancement of genomic alteration detection and availability of various targeted agents as well as immunooncology drugs, several patients with metastatic and advanced HNSCC can now receive individualized treatments. Therefore, there is a pressing need to thoroughly investigate the clinical significance of TP53 mutations in patients with HNSCC who are being treated with novel drugs in the era of precision medicine.

In South Korea, a large-scale clinical trial, the Translational Biomarker-driven Umbrella Project for Head and Neck Squamous Cell Carcinoma (TRIUMPH), was conducted. In this multicenter, prospective, and umbrella trial, treatment regimens were allocated patients with recurrent and/or metastatic HNSCC on the basis of their individual NGS results. The clinical trial included the use of alpelisib, poziotinib, nintedanib, abemaciclib and durvalumab with or without tremelimumab. The TRIUMPH trial demonstrated the feasibility and applicability of NGS-based genomic profiling in precision medicine in patients with HNSCC. Using the database of the TRIUMPH trial, we aimed to further analyze the clinical characteristics of patients harboring the TP53 mutations by utilizing the genomic profiling results of the participants in the TRIUMPH trial. Additionally, we evaluated the difference in the efficacy and prognostic implications of various drug treatments, by stratifying the patients according to the presence or absence of TP53 mutations.

Materials and Methods

Results

1. Characteristics of the patients

In the TRIUMPH trial, 179 patients met the eligibility criteria and were assigned to one of the five groups according to the genomic results. Groups 1, 2, 3, 4, and 5 comprise of 45, 17, 10, 34, and 73 patients, respectively (Fig. 1). The baseline characteristics of patients according to the presence of a TP53 mutation are described in Table 1. Of the 179 patients, 116 (64.8%) had TP53 mutations. The median age of the participants in the TP53 MT group and TP53 WT group was 64 and 61 years, respectively. The TP53 MT group consisted of a higher proportion of patients aged ≥ 75 years than the TP53 WT group (9.5% vs. 1.7%, p=0.045). TP53 MT was observed more commonly in the oral cavity (25.1%), hypopharynx (16.2%), and larynx (13.4%). However, the TP53 WT was more prevalent in the oropharynx (15.6%, p < 0.001). Furthermore, the proportion of HPV-positive tumors was higher in the TP53 WT group than in the TP53 MT group (15.1% vs. 7.3%, p < 0.001). The TP53 MT group had more underweight patients with a body mass index (BMI) of < 18.5 than the TP53 WT group (19.6% vs. 4.5%, p=0.029). In the treatment groups 2, 3, and 4, more patients exhibited TP53 MT than TP53 WT.

Fig. 1.

Consort diagram. CDK, cyclin dependent kinase; CDKN2A, cyclin dependent kinase inhibitor 2A; EGFR, epidermal growth factor receptor; FGFR, fibroblast growth factor receptor; HNSCC, head and neck squamous cell carcinoma; IHC, immunohistochemistry; NGS, next-generation sequencing; PIK3CA, phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha.

Table 1.

Characteristics of the patients in the TP53 mutant and TP53 wild-type groups

2. TP53 mutations in HNSCC

Three patients were filtered out because of the lack of a variant in the oncoplot of the total population (Fig. 2A). The TP53 gene harbored the highest number of mutations in the HNSCC samples. Moreover, missense mutations were the most common TP53 mutation in HNSCC (50.8%). These mutations were mainly located in the DNA binding domain of p53 (Fig. 2B). The other mutations observed were nonsense mutations (16.4%), frameshift deletions (12.5%), frameshift insertions (8.6%), and splice site mutations (8.6%) (Fig. 2A and B).

Fig. 2.

Genomic landscape and profile of tumor suppressor p53 (TP53) mutations. (A) Oncoplot of total population. (B) Lollipop plot of TP53 mutation. The percentage of each gene mutation in (A) was calculated based on 176 samples, after filtering out three samples due to the absence of variants. HPV, human papillomavirus; NA, not available; TMB, tumor mutational burden.

3. Mutational profiles of the TP53 MT and TP53 WT groups

The TP53 MT group had significantly more patients with a high tumor mutational burden (TMB; ≥ 10 mutations/mb) than the TP53 WT group (44.1% vs. 11.7%, p < 0.001). Although the microsatellite instability distribution was not significantly different between the two groups (p=0.069), more unstable cases were observed in the TP53 MT group than in the TP53 WT group (21.9% vs. 7.8%). Several comutations, including mutations of the EGFR (11.2% vs. 2.2%, p=0.03), CDKN2A (22.3% vs. 2.8%, p=0.001), CCND1 (21.8% vs. 1.1%, p < 0.001), NFE2L2 (11.7% vs. 1.7%, p=0.012), and ARAF (14.0% vs. 1.7%, p=0.003) genes, were more prevalent in the TP53 MT group than in the TP53 WT group. However, the FBXW7 mutation was more common in the TP53 WT group than in the TP53 MT group (6.7% vs. 1.7%, p < 0.001) (Table 2).

Table 2.

Mutational profiles in the TP53 mutant and TP53 wild-type groups

4. Survival in the TP53 MT and TP53 WT groups following targeted treatments and immunotherapy

The median OS was significantly shorter in the TP53 MT group than in the TP53 WT group (11.1 vs. 28.8 months, p=0.005) (Fig. 3A). The median PFS was also shorter in the TP53 MT group than in the TP53 WT group (1.7 vs. 3.8 months, p=0.002). In addition, we compared the survival between patients who were treated with targeted therapies (groups 1-4) and those who were treated immunotherapy (group 5). In 105 patients who were treated with targeted therapies, the OS was not statistically significantly different between those with TP53 MT or TP53 WT (11.2 vs. 15.2 months, p=0.242) (Fig. 3B). However, among the patients treated with targeted therapies, the PFS was shorter in patients with TP53 MT than in those with TP53 WT (2.5 vs. 7.3 months, p=0.009). Among the 73 patients in group 5, the median OS was shorter in patients with TP53 MT than in those with TP53 WT (8.1 vs. 33.0 months, p=0.001) (Fig. 3C). However, there was no significant difference in PFS between the 73 patients in group 5 who exhibited TP53 MT or TP53 WT. Thereafter, we analyzed the survival of patients in the immunotherapy group 2, which consisted of patients in group 5 and patients who crossed over to group 5 from groups 1-4 after failure of these treatments. Among the 109 patients in the immunotherapy group 2, the median OS was shorter in patients with TP53 MT than in those with TP53 WT (3.3 vs. 11.6 months, p=0.002) (Fig. 3D). However, the median PFS of this population was not statistically different between those with TP53 MT and those with TP53 WT (1.6 vs. 3.6 months, p=0.057).

Fig. 3.

Overall survival (OS) of patients according to the tumor suppressor p53 (TP53) mutations. (A) All participants (groups 1-5). (B) Targeted therapy group (groups 1-4). (C) Immunotherapy group (group 5). (D) Immunotherapy group 2 (group 5+participants who crossed over to group 5 from groups 1-4). CI, confidence interval.

The univariate analysis revealed that the presence of EGFR mutations or amplifications, which were allocated to the group 2, and cell cycle pathway-related gene alterations, including CDK4/6 mutation, CCND1 amplification, or CDKN2A mutations, which were allocated to group 4, were significant predictors of a poor prognosis. In addition, we conducted univariate analyses for various clinical factors including HPV status, smoking, BMI, Charlson Comorbidity Index, different age criteria, sex, site of primary tumor location, and the number of prior lines of systemic treatment. However, none of these clinical factors showed significant differences in OS. Therefore, we included genomic factors as covariates for the multivariate analysis, which showed statistical significance in univariate analysis. The multivariate analysis revealed that a TP53 mutation was an independent poor prognostic factor (HR, 1.61; p=0.049) (Table 3).

Table 3.

Multivariate analysis of prognostic genetic alterations

Discussion

TP53 gene mutation is frequently reported in most human cancers. Since its initial discovery in 1979, numerous studies have been conducted to identify the role of TP53 mutation in cancer biology [11]. Both germline and somatic TP53 mutations affect cancer development. Germline TP53 mutations contribute to the development of Li-Fraumeni syndrome, which is characterized by the early onset of multiple cancers [12]. According to the National Cancer Institute–sponsored TP53 database, which was originally established in 1994 by the International Agency for Research on Cancer and the World Health Organization, the frequency of somatic mutations of the TP53 gene ranges from 20% to 90% in various sporadic cancers [13,14]. The incidence of TP53 mutations in HNSCC has been primarily reported in Western populations, and it varies widely across different studies. In The Cancer Genome Atlas (TCGA) analysis of 279 patients with HNSCC, TP53 mutations were reported in 41% of the tumors tested [15]. In larger cohorts of HNSCC from publicly available datasets such as cBioPortal [16], American Association for Cancer Research Project GENIE [17], and Caris Life Sciences (Phoenix), the incidence of TP53 mutations is 68% [18]. Among the Asian countries, the incidence of TP53 mutations was reportedly 38.8% in a Japanese cohort (n=283) and 47% in a Chinese cohort (n=89) [19,20]. In our study, TP53 mutations were reported more frequently (64.8%). Considering these results, the frequency of TP53 mutations in Korean HNSCC appears to be relatively higher than that in other Asian countries and similar to that in Western populations.

In our study, the TP53 mutations were associated with aggressive clinical characteristics such as HPV-negativity, advanced age, and oral cavity origin. TP53 mutations were also significantly associated with poor survival in the Korean population. Moreover, the multivariate analysis revealed that patients harboring TP53 mutations demonstrated a 1.61-fold increase in risk of death.

The inverse relationship between the presence of HPV DNA and TP53 mutations has been documented in several reports. In 2015, the report by TCGA presented the difference in carcinogenesis between HPV-related HNSCCs and non-HPV–related HNSCCs via genomic sequencing [15]. Furthermore, it stated that accumulation of tumor mutations and chromosomal gains and losses were more in non-HPV–related HNSCCs than HPV-related HNSCCs. Moreover, the most prevalent mutation in the non-HPV–related HNSCCs was TP53 mutation. Therefore, the prevalence of TP53 mutations might be lower in larynx or hypopharynx cancers, which are weakly related to HPV, than oropharynx cancers. Additionally, the prevalence of TP53 mutations might be higher in older adults, who have a higher incidence of larynx or hypopharynx cancers, than younger adults. Our findings on the differences in characteristics between the TP53 MT an TP53 WT groups are consistent with those of previous studies.

In our study, the prevalence of TP53 mutations differed according to the BMI. TP53 mutations were more common in patients with a lower BMI, particularly underweight patients with a BMI of < 18.5, than in those with a higher BMI. A lower BMI in patients with HNSCCs may be attributed to the weight loss associated with the destruction of structures involved in chewing and swallowing in the hypopharynx or oral cavity. The higher incidence of TP53 mutations observed in these patients may be related to the fact that patients with hypopharyngeal or oral cavity cancer are more likely to have a low BMI. Furthermore, BMI is regarded as a surrogate marker of the nutritional status of patients with cancer, which affects the OS [21,22]. Previous studies have demonstrated that a higher BMI in HNSCC is associated with better survival outcomes, while being underweight is associated with a higher risk of death [23,24]. Comprehensively, these clinical characteristics may explain the poor prognosis of the TP53 MT group.

Our study findings demonstrated that the PFS associated with the treatment option significantly impacts the OS. The median PFS of the entire cohort was shorter in the TP53 MT group than in the TP53 WT group, and this difference was also observed among patients who were administered targeted agents. However, the median PFS did not differ between the TP53 MT and TP53 WT groups among patients who were allocated to the group without any druggable genomic targets. This finding indicates that TP53 mutation is not a predictor of treatment response of immunotherapy, at least in patients without druggable genetic alterations. However, TP53 MT was associated with a significantly poor OS, especially in patients in the immunotherapy group. Thus, TP53 mutations is a powerful prognostic marker in patients without druggable genomic alterations. Our study findings also demonstrated that EGFR gene aberrations and cell cycle pathway gene alterations, including CDK4/6 mutation, CCND1 amplification, or CDKN2A mutations, were associated with a poor survival. However, the multivariate analysis revealed that only TP53 was independently associated with poor survival. Therefore, a TP53 mutation, which plays a major role in the various steps and multiple pathways of carcinogenesis, is the most powerful prognostic factor in HNSCC.

In our study, a high TMB or microsatellite instability was seen more frequently in the TP53 MT group than in the TP53 WT group. In a report by Klinakis and Rampias, TP53 mutations were associated with a high TMB, particularly in metastatic HNSCC [25]. Furthermore, data from the Caris Life Sciences cohort demonstrate that tumors harboring TP53 mutations exhibit a significantly higher average TMB than tumors lacking TP53 mutations [18]. Because TP53 has multiple functions that attribute to its interaction with numerous target genes involved in DNA damage, cell cycle, cell metabolism, and apoptosis, TP53 mutations are naturally correlated with genetic instability or higher TMB [26]. In this study, we also found that comutations in CDKN2A or CCND1 were higher in the TP53 MT group than in the TP53 WT group. The normal functioning of the TP53 gene is negatively regulated by the MDM2 gene, which encodes an E3 ubiquitin ligase, the MDM2 protein [27,28]. CDKN2A is also involved in stabilizing TP53. The p14ARF protein, which is encoded by CDKN2A, reportedly binds to MDM2 and stabilizes TP53 by suppressing its E3 ubiquitin ligase function [29]. Consistent with our study findings, previous studies have demonstrated the low frequency of tumors with CDKN2A alterations in the absence of an accompanying TP53 mutation [18]. However, a high incidence of FBWX7 co-mutation was found in our study’s TP53 WT group. FBWX7 is a well-established tumor suppressor gene that is inactivated by mutations in numerous cancers [30]. FBWX7 acts as a ubiquitin E3 ligase and regulates the p53 protein level. In response to DNA damage, p53 accumulates and induces FBXW7 expression. The FBXW7 expression subsequently promotes p53 degradation, keeping the p53 level in check. Gong et al. [31] reported that FBXW7 inactivation induces significant cellular senescence in TP53 WT cells, but not in TP53 MT or normal cells. Furthermore, they determined that the simultaneous p53 inactivation abrogated the induced cellular senescence [31]. This finding is similar to our study finding that FBXW7 inactivation triggers tumorigenesis predominantly in the TP53 WT cells.

Our study results validate TP53 as a prognostic factor and provide clinically relevant insights for the application of emerging targeted therapies in the treatment of patients with recurrent or metastatic head and neck cancer. Currently, various targeted therapies are under development and clinical trials. These targeted therapies are likely to be utilized as NGS-guided treatments in the near future. In such cases, the presence of TP53 mutations may influence the choice of targeted therapy in patients with TP53 MT may be associated with a poor PFS. Moreover, current clinical trial data suggest minimal efficacy of immunotherapies following failure of platinum-based therapies [32]. Thus, targeted therapies rather than immunotherapy can be prioritized in patients with targetable genetic alterations and TP53 WT after failure of platinum-based therapies. However, in patients with TP53 mutations and no other actionable genetic alterations, treatment with the immune checkpoint inhibitor is unlikely to increase survival, potentially guiding the avoidance of over-treatments.

The limitations of this study include the small sample sizes within each targeted therapy group, which may have restricted the analysis and interpretation of PFS differences according to the TP53 mutational status. Further studies on its use as a predictive or prognostic marker of TP53 mutation is required with larger sample sizes in each targetable gene alteration group. Furthermore, immune checkpoint inhibitors were only administered to patients without targetable driver mutations in the TRIUMPH study. Therefore, the prognostic role of TP53 mutations in an entire patient population receiving immune checkpoint inhibitor remains unclear. However, our study has some notable strengths. The analyses were performed by utilizing a large-scale genomic dataset obtained from the nation-wide Korean patients. Future studies should analyze a larger patient cohort and explore other potential markers or therapeutic strategies to improve the poor prognosis associated with TP53 mutations.

In conclusion, TP53 mutation is an independent poor prognostic factor in patients with advanced HNSCC. Aggressive clinical characteristics, which are more common in patients with TP53 mutations and poor response to targeted therapies, might be the cause for the shorter OS. Furthermore, patients with TP53 MTs exhibited poor prognosis even with immunotherapy. Considering the high frequency of TP53 mutations in patients with HNSCC, additional investigations should be performed to personalize strategies targeted at TP53 MTs in this population.

Article information Continued