p53 and Cell Fate: Sensitizing Head and Neck Cancer Stem Cells to Chemotherapy

Abstract

Head and neck cancers are deadly diseases that are diagnosed annually in approximately half a million individuals worldwide. Growing evidence supporting a role for cancer stem cells (CSCs) in the pathobiology of head and neck cancers has led to increasing interest in identifying therapeutics to target these cells. Apart from the canonical tumor-suppressor functions of p53, emerging research supports a significant role for this protein in physiological stem cell and CSC maintenance and reprogramming. Therefore, p53 has become a promising target to sensitize head and neck CSCs to chemotherapy. In this review, we highlight the role of p53 in stem cell maintenance and discuss potential implications of targeting p53 to treat patients with head and neck cancers.

I. INTRODUCTION

The tumor suppressor TP53 is the most commonly mutated gene in cancer. Known as the “guardian of the genome,” it can be considered a master regulator of cell fate. Although p53-null mice are viable, they have early-onset sporadic cancers.¹ In addition, individuals with Li-Fraumeni syndrome, which harbors p53 germ-line mutations, frequently develop cancers.² Given the major impact of p53 in cancer biology, most investigators have focused on the role of p53 as a tumor suppressor. Responses such as apoptosis, cell cycle arrest, and senescence have been the principal focus of p53 studies. Nonetheless, p53 has been found to be involved in normal cellular and developmental processes, including functions in multiple stages of embryonic development.^3,4 Notably, strong evidence shows involvement of p53 in stem cell self-renewal and differentiation.^5,6

Two emerging roles for p53 include its ability to inhibit cancer stem cell (CSC) formation and regulate the CSC state. CSCs share many properties with adult stem cells. For example, they have the ability to self-renew and differentiate into committed progenitor cells with limited self-renewal potential.⁷ CSCs are known to be resistant to standard chemotherapy and radiation therapies and have increased tumorigenic capacity, often playing a part in tumor recurrence and metastasis in vivo.⁸ This review highlights some of the functions of p53 that are relevant to CSC maintenance (e.g., self-renewal and differentiation) and the potential implications of targeting p53 to sensitize CSCs to conventional chemotherapy in head and neck malignancies.

II. p53: GUARDIAN OF THE GENOME

The transcription factor p53 regulates expression of a large number of protein-coding genes and microRNAs (miRNAs) that mediate its downstream response. In its role as guardian of the genome, p53 is found to be mutated in > 50% of sporadic human cancers and thought to be functionally inactivated in a significant portion of remaining cancers.^9–11 It can transiently halt the cell cycle to allow DNA damage repair or irreversibly block proliferation through senescence (or programmed cell death) to eliminate cells with damaged or mutated DNA. Functioning primarily as a transcription factor, this protein is involved in preserving genome integrity.

In normal conditions, p53 is under tight regulation, and its abundance is kept low in nonstressed cells. However, after genotoxic insult, p53 is protected from degradation via several post-translational modifications that allow its rapid accumulation in the cell and subsequent transcriptional activation.¹² Mouse double minute 2 (MDM2) is a major negative regulator of p53 and an E3 ubiquitin ligase that functions cooperatively with MDMX to target p53 for proteasomal degradation.¹³ Levels of p53, MDM2, and MDMX are tightly regulated to sense and respond to stress.¹⁴ This coordinated network is very important in normal homeostasis and development. MDM2-null mice have early embryonic lethality that can be rescued with p53 deletion.¹⁵ Some cancers bypass p53 function by overexpressing MDM2 or MDMX.¹⁶ Responses of p53 on a cell are stress and tissue specific. Below, we explore some of the evidence supporting the role of p53 in stem cell homeostasis.

III. CSC HYPOTHESIS

The CSC hypothesis states that a small subset of cancer cells within a tumor have tumor-initiating, self-renewing, and multilineage differentiation potential.¹⁷ This hypothesis postulates that (1) only a small population of cancer cells within a tumor are capable of initiating and propagating tumors in immunodeficient mice; (2) tumors resulting from CSCs contain a mixture of tumorigenic and nontumorigenic cells that recapitulate original tumor heterogeneity; and (3) tumors generated by CSCs can be serially transplanted, demonstrating self-renewal capabilities. These CSCs can be isolated based on the expression of cell-surface markers and/or aldehyde dehydrogenase (ALDH) activity.¹⁸

The first studies supporting the CSC hypothesis showed that tumor tissues have a distinct hierarchy of cells, similar to that observed in adult tissues.¹⁹ Within a tumor, cells can be distinguished by the expression of different cell lineage markers. This hierarchical organization was first evident in leukemic cells and later proven to be true in breast cancer and other solid malignancies.^20–23 Only a small population of cells has tumorigenic potential and stem cell properties of self-renewal and differentiation, but the rest of the tumor tissue is comprised of transit-amplifying and differentiated cells that do not contribute to tumor initiation. Furthermore, CSCs are in a translational phase between epithelial-and mesenchymal-cell stages that might contribute to the plasticity observed in cancer cells.^24,25 Environmental cues can cause CSCs to acquire or loose these stem cell characteristics.^26,27

The frequency and prevalence of CSCs can vary by tumor type and among patients who have the same type of cancer.^28–30 Studies have shown that CSCs can be used as predictive and prognostic markers for patient survival.^28–30 We have not yet identified a CSC marker that is uniquely expressed in these cells. As such, they can only be used to enrich for CSC population rather than isolate a pure population of cells.³¹ This complicates the study of CSCs and presents challenges to the use of these markers for targeted therapeutic approaches. Significant effort is now in place to address these challenges and develop biomarkers that can be used to verify the efficacy of CSC targeted therapies.

IV. p53 IN STEM CELLS

There are several mechanisms by which p53 is thought to regulate both normal stems cells and CSCs.³² In this review, we focus on only two of these. The first involves the effects of p53 on stem cell self-renewal and the second on stem cell differentiation.

A. p53 in Self-Renewal

Stem cells are defined by their ability to self-renew and produce progenitor cells that can ultimately generate multiple cell lineages (i.e., multipotency). Both of these tasks are centered on a single mitotic division event. Whereas, symmetric stem cell division yields two identical daughter cells, asymmetric division produces one stem cell and one proliferative progenitor cell (Fig. 1). When a stem cell undergoes symmetric division, it can yield two identical daughter stem cells or two proliferative progenitor cells. Both of these processes are essential during development and tissue homeostasis to maintain a stem cell pool and produce specialized cells. It has been shown that p53 plays an important role in the balance between self-renewal and differentiation in embryonic and adult stem cells, and this balance is important in cancer development and progression.^5,6

FIG. 1:

Stem cells can undergo both symmetric and asymmetric division events, resulting in expansion, exhaustion, or maintenance of the stem cell pool. When a stem cell undergoes asymmetric division, it can produce an identical daughter cell (self-renewal) and committed progenitor (differentiation). If a stem cell divides symmetrically, two daughter stem cells or two daughter progenitor cells can result. These modes of division are regulated by the stem cell pathways of self-renewal and differentiation.

In mammary stem cells, p53 has been found to regulate polarity of cellular divisions.³³ The absence of p53 promotes self-renewal of these cells, allowing for expansion of the stem cell pool and resulting in unlimited and symmetric self-renewing divisions. Asymmetric divisions can be measured by the distribution of Numb.³⁴ Numb can directly interact with p53 and MDM2 and is thought to regulate p53 signaling by preventing MDM2-mediated ubiquitination.³⁵ Interestingly, p53 knockout mice contain a higher percentage of cells that are capable of forming mammospheres in culture and repopulating the mammary gland in vivo, further supporting a role for p53 in maintaining the stem cell pool.³⁶

Several groups have shown that deregulation of pathways that control self-renewal of normal stem cells can lead to their transformation into cancer cells.^17,37,38 Indeed, p53 regulates the self-renewal of myeloid progenitor cells, transforming them into leukemia-initiating cells that resemble CSCs.³⁷ Conversely, p53 loss promotes CSC-pool expansion.^39,40 Meanwhile, mutant p53 induces CSC marker expression in colorectal cancer.⁴¹ Such findings suggest a strong role for p53 in regulating self-renewal of CSCs.

B. p53 in Differentiation

A correlation between grade of differentiation and presence of p53 mutations has been observed in several malignancies.^42,43 Cancers with p53 mutations or functional p53 inactivation by downstream regulators (e.g., overexpression of Mdm2) correlate with poor-grade and undifferentiated tumors.⁴⁴ In many cancers, histological grade remains one of the best prognostic factors for patient survival.^45,46 This has certainly been the case for most squamous cell carcinomas and salivary gland cancers (SGCs).^47,48

Endogenous p53 was shown to induce differentiation of mouse embryonic stem cells by suppressing Nanog expression after DNA damage.⁴⁹ Nanog is an important gene for embryonic stem cell self-renewal.⁵⁰ Not only does Nanog have a role in embryonic stem cells, it also regulates dedifferentiation of primary p53-deficient mouse astrocytes into CSC-like cells.⁵¹ Tumor-suppressor p53 can regulate differentiation of several cell types,⁵² and its loss can also induce reprogramming of pluripotent stem cells, further supporting a role for p53 in differentiation.

It has been shown that p53 also transcriptionally activates miRNAs responsible for down-regulating stem cell transcription factors⁵³ down-regulates canonical Wnt pathway components by several mechanisms.^54,55 Canonical Wnt signaling was first thought to be essential for maintaining stem cells undifferentiated during development.^56,57 But, the roles of p53 and Wnt in differentiation are cell and context dependent. For example, p53 can induce transcription of Wnt ligands while Wnt signaling is balanced between differentiation and self-renewal pathways.^56,58–60 The role of p53 in regulating cell fate is a balance between stem cell differentiation and self-renewal (Fig. 2). This balance is “hijacked” in cancer to expand the CSC pool. Deregulation of these differentiation and self-renewal pathways can be responsible for the formation of cancer stem-like cells.

FIG. 2:

Tumor suppressor p53 is a key regulator in stem cell fate, balancing self-renewal and differentiation. When p53 is activated in a stem cell, it shifts the balance to asymmetric division. CSCs frequently down-regulate p53 activity, leading to changes in cell polarity that increases symmetric stem cell division and expand the stem cell pool. Activating p53 in CSCs will differentiate these cells, lead to a decreased stem cell pool, and sensitize these otherwise resistant cells to conventional chemotherapies.

V. SENSITIZING CSCs TO CHEMOTHERAPY

Two main approaches can be used to target CSCs. The first, and most studied, is to find agents that can specifically kill these cells.⁶¹ CSCs are resistant to standard chemotherapy agents and radiotherapy used in the clinic.^62–66 For example, breast CSCs (i.e., CD44^high and CD24^low cells) appear to be intrinsically resistant to conventional chemotherapy and radiation therapy.^63–64 Their unique biology allows them to bypass many of the mechanisms used to target cancer cells. For this reason, a CSC-specific treatment is needed to kill these cells. Several clinical trials targeting CSC-specific pathways are currently underway globally.⁵³

The second approach is to find agents that can differentiate CSCs to sensitize them to chemotherapy and radiation therapy. CSCs exploit physiological stem cell pathways to maintain a more undifferentiated state in such a way that they self-renew and give rise to different cells comprising a tumor.⁷ In theory, differentiating CSCs will make them loose their potential to self-renew, therefore becoming short lived. This approach is gaining increasing interest because the idea of sensitizing otherwise resistant cells to standard chemotherapy regimens is an attractive approach for the clinic. Nevertheless, more about the potential to differentiate CSCs must be understood because CSCs in many different tumor types (different from adult stem cells) have not been fully characterized and may present undistinguished differentiation fates.

One prominent challenge in studying CSCs is that the percentage and markers used to identify these cells in tumors varies by cancer and cancer subtype.⁶⁷ Compelling evidence for the importance of CSCs in tumorigenesis and correlations with patient outcomes has been observed in different tumors.^68–70 Apart from surgery, chemotherapy and ionizing radiation are the most common therapies used to treat cancer, but a good number of cancers do not respond to therapy and some develop resistance over time.⁷¹ Furthermore, most chemotherapy agents are DNA damaging. CSCs can be resistant to conventional chemotherapy through several mechanisms.⁶² It is thought that these cells are intrinsically less susceptible to chemotherapy because they are slow cycling. Because chemotherapy agents are DNA damaging, they are most effective on rapidly proliferating cancer cells. Apart from this, CSCs have ABC transporters that can effectively efflux chemotherapy drugs and are resistant to DNA damaged-induced death, among other potential mechanisms.^72,73

Intravenously injected chemotherapy drugs are capable of penetrating just a few layers of cells in tumor tissues.^74,75 This implies that deeper cell layers in the tumor receive lower doses of these agents. Given the intrinsic resistance of CSCs to chemotherapeutic drugs, subcytotoxic doses received by cells could explain the increase in fraction of CSCs observed when treated with chemotherapy.^76–79 This increase could also be due to the intrinsic nature of these cells to be more resistant to chemotherapeutic agents, thus surviving while the rest of the tumor cells succumb to treatment. A combination of both mechanisms could contribute to this phenomenon. Residual CSCs that survive treatment are thought to be responsible for tumor recurrence and metastasis due to increased cell invasiveness, survival, and tumorigenic potential.⁸

Chemotherapy is capable of inducing trans-differentiation of cancer cells into CSCs, creating a population of chemotherapy-resistant cells that rely on similar mechanisms as those for generation and maintenance of induced pluripotent stem cells (iPSCs).^80,81 A study by Auffinger and colleagues showed that temozolomide induced an increase in the glioma CSC pool, and the increase was a result of a phenotypic shift of the non-CSC pool to a CSC-like state.⁸¹ These authors showed that the newly transdifferentiated cells are more tumorigenic, invasive, and chemoresistant than the original tumor source.

Clinical studies monitoring the prevalence of CSCs before and after chemotherapy treatment have shown that CSCs are resistant to therapy, confirming results observed in in vitro and in preclinical studies.^82,83 As an example, breast cancer patients undergoing neoadjuvant chemotherapy who underwent biopsies before and after treatment showed that cells with CSC markers persisted even as tumor mass regressed.⁶³ CSCs resistant to radiotherapy have also been observed in patients who have received high doses of irradiation.⁸⁴

Studies attempting to exploit similarities between CSCs and iPSCs to sensitize cancers to conventional therapies have been attempted.⁸⁵ Using Nanog as a reporter of CSC formation, Saydaminova and colleagues identified the compound GDM-1515 to be a regulator of histone demethylases that is capable of sensitizing CSCs to cisplatin-induced apoptosis.⁸⁵ This compound was able to inhibit epithelial-mesenchymal transition (EMT) and induction of CSCs by cisplatin. On the other hand, an unbiased screen for small molecules cytotoxic to CSCs revealed salinomycin, a natural compound that can reduce expression of CSC genes and mammary tumor growth and increase epithelial differentiation of mammary gland tumors.⁶² Further studies have proven the effectiveness of GDM-1515 for targeting CSCs in other cancers.^86–89 However, its severe toxicity in normal cells impedes its clinical use.^90,91 Meanwhile, other researchers have shown that targeting pathways that regulate CSC self-renewal can lead to CSC sensitization to chemotherapy.⁹² Bmi-1 is a component of polycomb-repressor complex I and a known regulator of self-renewal. Targeting Bmi-1 results in sensitization of cisplatin-based chemotherapies in CSCs.⁹²

IV. p53 IN HEAD AND NECK CANCER

Head and neck cancers, arising in mucosal surfaces of the oral cavity, nasopharynx, oropharynx, hypo-pharynx, larynx, paranasal sinuses, nasal cavity, and salivary glands, constitute ~4% of all cancers worldwide. Among those, more than 90% are head and neck squamous cell carcinomas (HNSCCs). HNSCC is the sixth most common cancer worldwide, occurring in more than 550,000 individuals and resulting in greater than 380,000 deaths each year.⁹³ Despite advancement in treatment modalities, there has been limited improvement in patient survival during the last three decades.

HNSCC can be stratified based on human papilloma virus (HPV) infection status; HPV-positive HNSCC carries a better survival rate than HPV-negative cancers.^94–98 Accumulating epidemiological, molecular, and clinical evidence supports HPV-positive cancer to be a distinct subtype of head and neck cancers.^99–101 Although the overall incidence of head and neck cancers has declined slightly in industrialized countries, there has been an increase in the incidence of HPV-positive cancers of the oropharynx.^102–105

Mutations in TP53 are the most frequent genomic alterations in HNSCC and have been correlated with poor patient survival.^95,106–108 The Cancer Genome Atlas (TCGA) analysis of 279 HNSCC patients showed TP53 mutations in 70.4% of tumors.⁹⁵ TP53 mutations are thought to be an early event in HNSCC carcinogenesis, because they have been found in premalignant lesions, and their incidence is associated with cancer progression.¹⁰⁹ Although TP53 is frequently mutated in HPV-negative tumors, it is typically wild type in HPV-positive tumors.^95,110,111 It has been shown that the HPV16 E6 protein in HPV-positive tumors can bind and target the p53 protein for proteasome degradation, resulting in functional inactivation of p53 signaling.¹¹² These results indicate that p53 signaling is key in both HPV-negative and -positive HNSCC.

Although TP53 is commonly mutated in HNSCC, it is rarely mutated in SGCs when compared to other neoplasms.¹¹³ SGCs are rare malignant tumors that account for ~6% of all head and neck cancers. The most recent classification of the World Health Organization showed that 22 types of malignant salivary gland tumors exist.¹¹⁴ Their histological and clinical diversity results in diagnostic and management challenges for clinicians.¹¹⁵ Primary therapy commonly involves radical surgery and/or radiation therapy, with conventional chemotherapy as a palliative aim for recurrent or metastatic disease. The limited clinical trial data on systemic therapeutic approaches for SGCs is a problem that has largely been a result of challenges in recruiting enough patients (due to the rarity of these tumors) or lack of patient stratification by tumor type. SGCs can originate in any major or minor salivary gland. SGC subtype and gland location are prognostic factors of patient survival. Although little is known about the pathogenesis of SGCs, several recurrent chromosomal translocations have been identified in some of the most common SGC subtypes. These common translocations are reviewed elsewhere.¹¹⁶

Mucoepidermoid carcinoma (MEC) and adenoid cystic carcinoma (ACC) are the two most common malignant SGCs.¹¹⁴ They are characterized by recurrent chromosomal translocations that are thought to play an important part in tumorigenesis.¹¹⁶ Despite these findings, the role of these translocations in tumor progression is not fully understood. Limited genomic studies of SGCs have hindered our understanding of the pathobiology of these cancers and identification of molecular targets for targeted therapies.

MEC is the most common malignant salivary gland tumor, accounting for 30%−35% of all salivary tumors.^117,118 MEC, a heterogeneous cancer that consists of mucin-producing, epidermoid, and intermediate cells, is commonly classified into low-, intermediate-, and high-grade disease. High-grade MEC has a prominent epithelial/solid component and low-grade MEC a prominent cystic/mucous component. High-grade MEC has the worst patient prognosis and frequent regional and distant metastasis.⁴⁸ More than 80% of MEC cases contain a t(11;19)(q21–22;p13) translocation that results in fusion between MAML2 and CRCT1 or CRCT3 genes.¹¹⁹ With this fusion, the Notch-binding domain of MAML2 is replaced by the cyclic adenosine monophosphate (cAMP) response element-binding (CREB) domain of CRCT1/3.¹²⁰ The molecular and pathological consequences of this fusion are still in the process of being elucidated, but it is thought the CREB dysregulation, mediated by this fusion, participates in tumorigenesis.¹²¹ It has been suggested that CRCT1-MAML fusion is an early event in MEC pathogenesis that can serve as a biomarker of MEC. A subgroup of fusion-negative high-grade tumors are suspected of being a different class of carcinomas and only fusion-positive tumors should be classified as MEC.¹²² Reports correlating mutational status with positive patient outcomes must be reevaluated, because of increasing evidence confirming fusion-negative tumors to be different classification of SGCs.^123,124

Varying reports of p53 mutational status in MEC samples ^125,126 have been limited by small sample sizes and unbiased detection of p53 mutations. Whole-exome sequencing of a cohort of 18 patients found p53 to be mutated in 30% of all MEC cases,¹¹⁹ and p53 mutations were only found in intermediate- and high-grade tumors. Comprehensive genomic profiling of 48 MEC patient samples further confirmed p53 to be a common genetic alteration in MEC that was found in ~40% of the cases analyzed and prevalent in higher disease grade samples.¹²⁷ Similar results were observed in a previous study using immunohistochemistry, in which aberrant expression of p53 was detected in higher histological graded tumors.¹²⁶ The p53 mutational status and its implications in MEC pathobiology must be further evaluated to determine frequency of these genetic alterations and to understand their implications in disease progression.

VII. HEAD AND NECK CSCs

A. CSCs in HNSCC

Increasing evidence supports a role for CSCs in the pathogenesis and progression of HNSCC. CSCs in HNSCC were first described in CD44^high cells by Prince and colleagues.¹²⁸ CD44, a type I transmembrane glycoprotein involved in cell-cell interactions, adhesion, and migration, can act as a receptor for hyaluronic acid and other extracellular matrix proteins. In these experiments, cells expressing high levels of CD44 had increased tumorigenic potential over cells that expressed low levels.¹²⁸ Moreover, tumors found to be generated from CD44^high cells could reproduce original tumor heterogeneity, as observed by the presence of both CD44^high and CD44^low cells. Gene expression analysis showed that CD44^high cells differentially expressed the Bmi-1 gene, a self-renewal protein. Both self-renewal and differentiation potential were shown in this study of CD44^high cells, supporting a CSC hypothesis in HNSCC.

Similar experiments using ALDH showed that ALDH-positive cells had higher tumorigenic potential than ALDH-negative cells.^129,130 ALDH is an enzyme involved in oxidation of aldehydes, cellular detoxification, retinoic acid metabolism, and protection from reactive oxygen species, among other important cellular pathways.¹³¹ Clay and colleagues found that as few as 500 ALDH^high cells were sufficient to form tumors in mice, and these tumors replicated original cell heterogeneity.¹²⁹ Notably, ALDH-positive cells isolated from patient samples were radioresistant and initiated tumors in mice.¹³² Krishnamurthy and colleagues went on to show that combined ALDH^high and CD44^hgh expression further enhanced the ability to identify CSCs and tumors generated with ALDH^hgh and CD44^hgh cells resembled original patient histology.¹³³ Moreover, using serial transplantations in vivo, the authors showed that these cells had increased Bmi-1 expression and self-renewal abilities. Further studies using these markers continue to prove the existence and role of these CSCs in the pathogenesis of HNSCC.

CD44 expression has been correlated with worse disease grade and prognosis for pharyngeal and laryngeal cancers.^134,135 Moreover, CD44 concentration in peripheral blood has the potential to serve as a prognostic and diagnostic tool for HNSCC patients.¹³⁵ CD44-positive cells in HNSCC were found to have an EMT phenotype, overexpress PD-L1, and were not as immunogenic as CD44-negative cells.¹³⁶ ALDH1 isoenzyme expression has also been correlated with decreased overall patient survival.¹³⁷ Clinical studies have shown that ALDH1A1 expression correlates with poor tumor differentiation and poor patient prognosis.¹³⁸ A recent study showed that NCT-501, an ALDH1A1 inhibitor, is able to sensitize cisplatin-resistant HNSCC cells to cisplatin and decrease CSC markers, self-renewal, and tumorigenic potential.¹³⁹ ALDH1A1 is immunogenic in HNSCC and an effective target for CD8⁺ T-cell-mediated immune response.^140,141 A recent transcriptome analysis of TCGA data for five different cancers including 520 HNSCC tumors versus 40 normal samples reviewed differential expression of 19 ALDH isoforms and found that differential expression of each isoform was similar among all cancers.¹⁴² Although some ALDH isoforms are up-regulated, others are down-regulated in these cancers. Finally, it was shown that differential expression of most isoforms correlated with cancer prognosis.

B. Salivary Gland CSCs

Recent studies have also identified ALDH⁺/CD44^high as markers for CSCs in salivary MEC and ACC.^143,144 Similar to what was shown with HNSCC, serial transplantation and in vitro sphere assays demonstrated self-renewal and differentiation potential of these cells in both MEC and ACC. A recent study found that ALDH and CD44 markers enriched for CSCs in ACC and MEC patient-derived xenografts, validating the use of such markers in salivary gland CSC research.¹⁴⁵ Three patient-derived xenografts of a high-grade MEC were generated from successive relapse surgeries (~9 mo) in the same patient. Further genomic characterization of these tumors showed increased mutational burden and stem cell marker expression, with decreased expression of tumor suppressors such as TP53 during disease progression. Additionally, increased sphere-forming abilities and a larger fraction of ALDH⁺/CD44^high cells in later relapses were shown.

VIII. CONCLUSION

Given the increasing evidence supporting a significant role for head and neck CSCs in tumor progression, we postulate that targeting pathways involved in CSC maintenance can potentially sensitize resistant cancers to conventional therapies and result in better eradication of tumor cells. In this review, we highlight some of the current evidence showing the importance of p53 signaling in CSC maintenance. We know that p53 has a prominent role in self-renewal and differentiation of normal and tumorigenic stem cells.³² Aberrant p53 signaling or inactivation in either HPV-positive or -negative tumors correlates with tumor progression in head and neck cancers. Moreover, p53 status in head and neck cancers has been associated with poor patient prognosis.¹⁰⁹

MDM2 inhibitors can activate endogenous wildtype p53¹⁴⁶ and reduce tumor volume, rate of recurrence, and CSC population in SGC.^147,148 They can also sensitize HNSCC to chemotherapy.¹⁴⁹ MDM2 inhibitors are in Phase I and II clinical trials for the treatment of several malignancies.^150,151 Other methods for activating p53 include reactivation of wild-type p53 function in mutant p53 and use of viral vectors to deliver wild-type p53 in p53-deficient cells.¹⁵² A better understanding of the mechanisms by which p53 induces differentiation of CSCs in head and neck tumors could inform a new therapeutic paradigm for cancer. Cancer patients may benefit from p53-induced differentiation of CSCs that can sensitize these cells to conventional chemotherapy.

ABBREVIATIONS:

ACC

adenoid cystic carcinoma

ALDH

aldehyde dehydrogenase

CSC

cancer stem cell

HNSCC

head and neck squamous cell carcinoma

HPV

human papilloma virus

iPSC

induced pluripotent stem cell

MEC

mucoepidermoid carcinoma

SGC

salivary gland carcinoma