INTRODUCTION — Squamous cell carcinoma of the head and neck (SCCHN) is common in several regions of the world. The primary risk factors include tobacco use, alcohol consumption, and human papillomavirus (HPV) infection. The relative prevalence of these risk factors contributes to the variation in the observed distribution of head and neck cancer in different areas of the world. (See "Epidemiology and risk factors for head and neck cancer".)
The chronic exposure of the upper aerodigestive tract to these and other risk factors is thought to produce field cancerization, a process in which patients with cancer or premalignant dysplastic lesions in the oral mucosa are at significant risk for head and neck cancer. The understanding of field cancerization provides the rationale for chemoprevention to decrease the incidence of SCCHN.
The use of chemoprevention to block the progression of precancerous lesions and the potential role of screening are discussed in this topic, focusing primarily on carcinogen-related head and neck cancer.
RATIONALE FOR CHEMOPREVENTION — The rationale for pharmacologic chemoprevention in patients at risk for the development of invasive cancer is based on two factors:
●Field cancerization – Patients with carcinogen-related squamous cell carcinoma of the head and neck (SCCHN) have a predilection for cancer development throughout the oral and oropharyngeal mucosa. Whether this is also true for human papillomavirus (HPV)-associated oropharyngeal tumors is still not well established.
●Multistep carcinogenesis – SCCHNs result from a multistep process with defined intermediate stages, leading to fully transformed, invasive, and metastatic cancer [1].
Standard options for managing oral dysplasia, a precursor of invasive cancer, range from watchful waiting to biopsy, laser surgery, and aggressive resection. However, none of these approaches has a proven clinical benefit. Other approaches to decreasing the incidence of invasive tumor growth in patients with recognizable premalignant lesions have fallen into several categories:
●Avoidance of further carcinogenic exposures (eg, tobacco, alcohol, betel nut) [2]. This strategy is important, but it may not be sufficient since many or most of the critical genetic alterations leading to transformation have already occurred in "early" premalignant lesions. (See "Head and neck squamous cell carcinogenesis: Molecular and genetic alterations".)
●Chemoprevention or pharmacologic intervention to reverse or inhibit carcinogenic progression. Chemoprevention can be considered in two settings: primary chemoprevention in individuals with de novo premalignant lesions and secondary chemoprevention in patients who have undergone potentially curative therapy in order to prevent recurrent disease or second primary tumors.
●Screening individuals known to be at high risk for the development of invasive cancer based on the presence of risk factors or premalignant lesions in a target population.
Field carcinogenesis and chemoprevention — Patients with SCCHN due to environmental carcinogens have a global predilection for the development of cancer throughout the oral, pharyngeal, and laryngeal mucosa. The concept of field carcinogenesis was initially proposed in the 1950s [3]. Subsequent studies confirmed that patients with cancer or premalignant dysplastic lesions in the oral mucosa are at significant risk for the development of head and neck cancer at noncontiguous mucosal sites [4,5]. The presence of such a field defect has precluded the routine use of surgical resection of premalignant disease as a curative approach. (See "Pathology of head and neck neoplasms".)
This field defect is attributed to carcinogenic exposure from risk factors (eg, tobacco and alcohol) throughout the entire oropharyngeal mucosa. (See "Epidemiology and risk factors for head and neck cancer".)
Field carcinogenesis is supported by both epidemiologic and molecular studies:
●Cancer at one site in the head and neck is a strong predictor for the development of a second primary tumor elsewhere [6,7]. Second tumors can either occur at or close to the same time as the primary lesion (synchronous) or six months or more later (metachronous). Synchronous second tumors are often occult lesions in the upper aerodigestive tract discovered during the initial staging evaluation, whereas metachronous second tumors most often involve other sites in the head and neck region or elsewhere in the upper aerodigestive tract (lung, esophagus) [7,8]. (See "Overview of the diagnosis and staging of head and neck cancer", section on 'Incidence of second and multiple primaries'.)
●Molecular analyses have often demonstrated that first and second lesions are clonally distinct, ruling out locoregional recurrence as a universal explanation [9]. However, in some cases, physically distinct mucosal tumors may share a common origin, suggesting that genetically abnormal precursor cells have the capacity to migrate extensively within the mucosa [10,11]. Histologically normal mucosa in patients with head and neck cancer may contain evidence of genetically abnormal cells at discrete sites not contiguous with the known cancer [12]. This may in large part explain the failure of surgical excision of grossly evident dysplastic lesions to substantially reduce the risk of cancer in patients with premalignant lesions. However, data from The Cancer Genome Atlas (TCGA) have not addressed these molecular abnormalities in premalignant areas because of the lack of available tissues, which needs to be addressed in future studies [13].
Surrogate markers and potential target agents — The goal of chemoprevention is to decrease the number of cancers that develop in the anatomic area at risk. Direct demonstration of a decreased incidence of cancer requires that very large numbers of patients be followed over long periods of time. Thus, there has been interest in defining alternative surrogate endpoints to both evaluate the efficacy of chemopreventive therapies and to serve as molecular prognostic factors for risk stratification in chemoprevention trials.
The accurate identification of lesions with carcinogenic potential is critical to chemopreventive studies. Grossly evident lesions that may be precancerous include leukoplakia (white plaques) and erythroplakia (red plaques). Standard histologic assessment alone of these lesions lacks adequate specificity and sensitivity [14]. Histologic criteria in noninvasive dysplastic lesions that have been suggested to correlate with progressive carcinogenic potential include the degree of nuclear atypia, failure of organized differentiation, loss of cell-cell contacts, the presence of keratotic bodies, and the cellular DNA content. (See "Pathology of head and neck neoplasms".)
Genetic definitions of progressive premalignant stages are being defined. Histologic progression has been correlated with sequential clonal genomic alterations, including deletions of chromosome arms 3p, 9p, 11q, and 17p [15-17]. A number of tumor suppressor genes thought to be critical in tumorigenesis have been mapped within these regions. (See "Head and neck squamous cell carcinogenesis: Molecular and genetic alterations".)
Many of these genomic alterations have in turn been associated with abnormal expression of a number of proteins that may hold promise as potential intermediate endpoints and molecular prognostic factors for chemoprevention studies. The reversal of molecular alterations that are tightly linked to the development of oral cancer (eg, epidermal growth factor receptor [EGFR] [18], transforming growth factor alpha [TGFA] [19], p53 [20-22], proliferating cell nuclear antigen [PCNA] [23], telomerase [24], and retinoic acid receptor beta [RARB] [25]) might indicate a reduced risk of oral cancer. (See "Epidemiology and risk factors for head and neck cancer".)
In addition, folate deficiency has also been linked to a higher risk of developing SCCHN [26,27]. A deficiency in folate seems to disrupt gene methylation as well as DNA structure. Genes that may be affected include tumor suppressor genes such as p16 and possibly p53 [28].
Genetic polymorphisms in folate metabolism have also been associated with clinical characteristics that may increase the risk of SCCHN, such as advanced age, tobacco and alcohol use, and male sex [29].
Supplemental folic acid may be exerting its chemopreventive role by inhibiting proliferation of mucosal epithelium. This process has been associated with a decreased concentration of EGFR in other tumor types, such as colon cancer [30]. In addition, SCCHNs clearly express folate receptors [31]. Furthermore, a pretreatment elevated intake of dietary folate was suggested as an independent prognostic factor for improved clinical outcomes in SCCHN patients [32]. Folate is therefore an interesting natural compound worth exploring in chemopreventive studies of the head and neck.
Other nutritional compounds of possible interest in chemoprevention include zinc. A mild deficiency of zinc was observed in 25 percent of the normal healthy volunteers and 48 percent of the head and neck cancer subjects, raising the possibility that zinc deficiency might contribute to the development of SCCHN. This has been explained by the possible dysfunction of type 1 T helper (TH1) cells, as evidenced by decreased production of interleukin 2 (IL-2) and interferon gamma in zinc-deficient head and neck cancer patients [33]. There is, however, no clear evidence for a clinical benefit from zinc supplementation in the chemoprevention of SCCHN.
Despite the interest in selenium as a possible chemopreventive agent [34], there is insufficient evidence to suggest a benefit for selenium supplementation in the prevention or prognosis of SCCHN.
Studies have suggested that soy supplementation has anticancer properties, and soy has been shown to inhibit carcinogenesis in animal models [35,36]. In addition, Bowman-Birk inhibitor is a protease inhibitor derived from soybeans that has demonstrated chemopreventive activity and has been tested in patients with oral leukoplakia with evidence of clinical benefit in 31 percent of patients [37]. There is, however, no conclusive evidence to support the use of genistein, a soy derivative, to prevent SCCHN or improve its prognosis. Studies investigating the properties of genistein as an antiinflammatory and anticarcinogenic agent in SCCHN are underway.
At least some data suggest that expression of podoplanin, a lymphatic endothelial marker whose biologic roles are incompletely understood, might be a powerful predictive factor for oral cancer development in patients with leukoplakia [38]. These data were derived from a randomized chemoprevention trial comparing 13-cis-retinoic acid, beta-carotene plus retinyl palmitate, and retinyl palmitate alone in 163 patients with oral leukoplakia. High levels of podoplanin expression by immunohistochemistry were associated with a significantly higher incidence of oral cancer (five-year risk 37 versus 8 percent in those with podoplanin-negative lesions). Risk could be further stratified by combining histology (the presence or absence of dysplasia) with podoplanin expression.
Which factors are useful predictors of carcinogenic progression and which molecular abnormalities best predict reduction in subsequent cancer risk when normalized remain unclear. In addition, a complete histologic response in premalignant lesions may be seen following administration of a chemopreventive agent in patients despite the persistence of chromosomal abnormalities [14].
SCREENING — We recommend oral cavity cancer screening for squamous cell carcinoma of the head and neck (SCCHN) in all patients, including high-risk groups of alcohol and tobacco users. In a randomized clinical trial, regular screening reduced mortality from oral cavity cancer by approximately 30 percent, likely due to detection of early-stage disease [39,40].
Oral cancer occurs in sites that are generally accessible by physical examination. Routine mouth examinations prevent SCCHN by detecting premalignant or early malignant lesions in the general population and in high-risk individuals (alcohol and tobacco users) [41-43]. In addition to screening, information and education on the health risks of tobacco should be thoroughly disseminated.
Population-based studies have demonstrated that a primary care screening strategy for early detection of head and neck cancer reduces overall mortality in the general population and can reach a large number of at-risk individuals [39,40]. In resource-limited countries, screening high-risk individuals could be an effective prevention strategy, although it is not known whether it is cost-effective.
The challenge with conducting screening studies is in identifying the high-risk group, the appropriate intervention or screening test, the duration of screening initiative, and the health outcome of interest. The quality of a screening program is affected by participation rate, test performance (sensitivity and specificity), frequency of screening, participant compliance, and duration of follow-up of positive test results.
Data from various screening initiatives in industrialized and resource-limited countries are as follows:
●In India, the Kerala Oral Cancer Screening Trial demonstrated that visual screening of the oral cavity over almost a decade reduced mortality from oral cancer by up to 29 percent, especially in high-risk groups of tobacco and/or alcohol users and despite expected limitations in compliance with screening and subsequent treatment referral. This trial evaluated over 190,000 healthy patients, age 35 years or older without a prior history of oral cancer, in 13 population clusters (municipal administrative units). Patients were randomly assigned either to three rounds of oral visual inspection by trained health workers at three-year intervals (seven clusters of 96,516 patients) or to standard care (six clusters of 95,354 patients) [39,40]. Patients who screened positive for disease were subsequently referred for clinical examination, biopsy, and treatment. At up to nine years of total follow-up, compared with standard care, screening reduced oral cancer mortality by 27 percent in all patients (HR 0.73, 95% CI 0.54-0.98) and by 29 percent in ever-tobacco and/or ever-alcohol users (HR 0.71, 95% CI 0.506-0.995) [40]. The absolute efficacy of screening was also greatest in those predicted to be at the highest risk of developing oral cancer at seven years. The number of screens administered was consistent with expected real-life compliance (3 screens planned versus 1.76 screens received). Similarly, the compliance rate of screen-positive patients with expert referral was also estimated at approximately 67 percent.
●In Cuba, the Oral Cancer Case Finding Program (OCCFP) was based on health education and thorough examination of the oral cavity by "stomatologists" in specialized stomatology centers [41]. Between 1983 and 1990, over 10 million patients were examined, 30,478 were referred for detected abnormalities, and 8259 complied with their referral. In the last group, 2367 leukoplakias, 853 other precancerous lesions, and 708 malignant neoplasms were diagnosed. The potential efficacy of this program is evident by an increased incidence of stage I oral cancers (from 22 to 48 percent) and the reduced incidence of more-advanced (stages II to IV) cancers (from 77 to 52 percent). However, these health outcome measures are subject to the potential influences of both "length bias" and "lead time bias." A reduction in population mortality would be a more appropriate surrogate measure of the effectiveness of the screening method.
●In a study in the United States, screening for squamous cell carcinomas of the oral cavity, pharynx, and larynx with symptom assessment and systematic inspection of the oral mucosa was performed by primary care practitioners at health care sites serving residents of urban Boston [43]. Of the 4611 tobacco users older than 40 years who were screened, 313 had specific criteria for otolaryngologic referral. The prevalence of oral mucosal lesions and persistent hoarseness was 13 and 11 percent, respectively. Abnormal findings were seen in over 70 percent of referred patients, and cancer was diagnosed in 3 percent.
●In studies involving over 250,000 subjects in Sri Lanka, primary health care workers detected a total of 44 new oral cancers. However, mortality was not reduced in these areas, possibly due to poor compliance following initial screening [42].
Future screening strategies based on detection of molecular abnormalities (eg, promoter hypermethylation) in body fluids such as serum and saliva may hold some promise, but this technology remains investigational [44-47].
CHEMOPREVENTION TRIALS — No chemopreventive agents have been shown to decrease the incidence of squamous cell carcinoma of the head and neck (SCCHN) in adequately powered randomized clinical trials.
Patients suitable for a chemoprevention trial may have a documented precancerous lesion or have a known risk factor for developing malignancy without having a premalignant lesion. A third group of patients are those with previously treated malignancy who are at a significant risk for developing a second primary tumor [48].
So far, chemopreventive trials in SCCHN have been conducted in patients with known premalignant lesions or in those at risk for a second primary tumor. With the observed sharp increase in the incidence of human papilloma virus (HPV)-related oropharynx cancer, a future generation of chemopreventive trials may focus on subjects with a high risk for developing HPV-related oropharyngeal cancer based on known HPV infection, lifestyle, environmental exposure, or a combination of these factors. The challenges facing chemoprevention in this group are the lack of a clearly defined precancerous lesion and the lack of clear risk stratification that would clearly identify the subjects who might benefit [49,50].
Most of the completed chemoprevention trials for premalignant lesions of the oral cavity and oropharynx have studied either naturally occurring compounds (vitamin A, vitamin E, beta-carotene) or synthetically derived retinoids. In addition, early phase studies have examined epidermal growth factor receptor (EGFR) and cyclooxygenase 2 (COX-2) inhibitors as single agents or in combinations [51-53].
Vitamin A, vitamin E, and beta-carotene — The interest in the antioxidants vitamin A, vitamin E, and beta-carotene as chemopreventive agents is due to their apparent efficacy in inducing regression of oral leukoplakia. Although these agents have been studied relatively extensively, both for primary and secondary chemoprevention, none has an established role.
Primary chemoprevention — Reports on the utility of vitamin A to treat oral leukoplakia date back to 1957, when a 90 percent response rate was described in 20 patients [54]. Subsequent trials supported the activity of vitamin A in oral leukoplakia in populations likely to be deficient in vitamin A and at risk based on tobacco or betel nut chewing [55,56]. The applicability of these results to disease in relatively developed nations where vitamin deficiency is rare is unclear. High response rates have also been described with the administration of beta-carotene, a precursor of vitamin A, to patients with oral leukoplakia [57-59]. In one series of 24 patients, for example, there were 15 partial and 2 complete responses with no significant toxicity [57].
One double-blind placebo-controlled trial has been performed [60]. This study randomized 160 people in India with oral precancerous lesions to vitamin A (300,000 international units per week), beta-carotene (360 mg per week), or placebo over a 12-month period. The complete regression rates were 10 percent with placebo, 52 percent with vitamin A, and 33 percent with beta-carotene. No major toxicities were observed. One-half of the responders to beta-carotene and two-thirds of those to vitamin A relapsed after the supplements were discontinued.
Alpha-tocopherol (vitamin E) has also been evaluated for the primary prevention of upper aerodigestive malignancies in patients at risk for head and neck cancer [61,62]. These two placebo-controlled studies, which randomly assigned patients to beta-carotene, alpha-tocopherol, both, or neither, did not provide any evidence to support a protective effect from these agents.
These observations provided the rationale for a pivotal randomized placebo-controlled study of isotretinoin (1 to 2 mg/kg per day) in 44 patients with leukoplakia [63]. This trial included pretreatment and posttreatment histologic evaluation in all patients, permitting analysis of the response among truly dysplastic lesions. After three months of therapy, both the overall clinical response rates (67 versus 10 percent) and the response rates among patients with biopsy-proven dysplasia (54 versus 10 percent) were significantly higher in the actively treated patients.
However, toxicity was particularly severe in patients dosed at the higher dose of isotretinoin (2 mg/kg per day), with 47 percent of patients requiring dose reduction. Patients treated at the lower dose level (1 mg/kg per day) had less toxicity and did not require dose reduction, although most noted xeroderma and 29 percent developed conjunctivitis. Furthermore, over 50 percent of responders relapsed within three months of treatment cessation.
In a second trial, 70 patients with leukoplakia underwent induction with high-dose isotretinoin (1.5 mg/kg per day for three months) and were then randomized to "maintenance" with low-dose isotretinoin (0.5 mg/kg per day for nine months) or beta-carotene (30 mg/day) [64]. Of the 53 evaluable patients, a response to maintenance therapy or continuation of stable lesions was significantly more common with isotretinoin (92 versus 45 percent, p<0.001). Side effects were generally mild but were more prominent with isotretinoin.
Secondary chemoprevention — Beta-carotene and alpha-tocopherol (vitamin E) have been evaluated as secondary chemopreventive agents in patients with head and neck cancer [65]. No evidence of benefit has been observed, while a possible detrimental effect of supplementation with beta-carotene plus alpha-tocopherol was suggested in a secondary prevention trial in which 540 patients undergoing radiation therapy for stage I or II head and neck cancer were randomly assigned to both agents or placebo for three years starting the first day of radiation [65]. Beta-carotene supplementation was discontinued after 156 patients were enrolled because of ethical concerns. After a median follow-up of 52 months, alpha-tocopherol-supplemented patients had a significantly higher rate of tumor recurrence or second primary cancer (hazard ratio [HR] 1.86, 95% CI 1.27-2.72) during the supplementation period, which decreased thereafter.
Retinoid therapy has been evaluated as a secondary chemopreventive after surgery or radiation for localized head and neck cancer to prevent recurrence of the original primary or development of a second primary [65-71]. These trials have failed to demonstrate any consistent evidence of clinical benefit.
As an example, in the largest trial, 1190 patients were randomly assigned to isotretinoin 30 mg daily or placebo after treatment for stage I or II head and neck cancer [70]. Isotretinoin provided no significant benefit in terms of time to develop a second primary tumor, rate of second primary tumors, overall survival, or recurrence-free survival. Only one much smaller trial, which used relatively high doses of isotretinoin and included only 103 patients, suggested a decrease in the incidence of secondary cancers [68].
Synthetic retinoids — The most extensively studied compounds for premalignant oropharyngeal disease are the synthetic retinoids [66,67,72-74]. These studies examined the utility of three retinoids: isotretinoin (13-cis-retinoic acid), all-trans-retinoic acid, and etretinate; all were active, with overall response rates of 59 to 92 percent. Despite these observed responses, a major limitation to the agents is the retinoid-related toxicity. The newer generations of receptor-selective retinoids may help in overcoming this limitation [75].
Mechanism of action — The mechanism by which retinoids affect premalignant lesions has not been fully elucidated. Many premalignant dysplastic lesions (60 percent in one series compared with none in normal specimens) express very low levels of the nuclear retinoid receptor beta (RAR beta), a component of several bipartite transcription factors responsive to retinoic acid [76]. This deficiency is reversed with retinoid therapy, which may reestablish many aspects of normal growth and differentiation in the aberrantly proliferating premalignant clone [76,77]. In one study of patients with premalignant lesions, the highest response rate (70 percent) was noted in patients with low baseline expression of the tumor suppressor p53 protein and upregulation of RAR beta messenger RNA after isotretinoin [77]. Retinoids also have antiangiogenic activity, which may directly or indirectly contribute to their antineoplastic activity [78].
Cyclooxygenase 2 inhibitors
Nonsteroidal antiinflammatory drug therapy — Intake of aspirin and other nonsteroidal antiinflammatory drugs (NSAIDs) prevents the development of intraepithelial neoplasia in the colon and rectum. The mechanism of this chemopreventive effect is thought to be inhibition COX-2 and diminished synthesis of prostaglandins, which are overproduced in colorectal tumors and have been implicated in tumorigenesis. (See "NSAIDs (including aspirin): Role in prevention of colorectal cancer".)
A similar rationale has been applied to the study of NSAIDs as chemopreventive agents for head and neck cancer, since both premalignant and malignant lesions of the oral cavity overexpress COX-2 [79,80]. In addition, administration of NSAIDs to laboratory animals prevents oral cancer development [81,82].
Results from randomized trials are not available, and the results of case-control and cohort studies are conflicting:
●A possible protective effect of aspirin use on the development of head and neck cancer was suggested in a questionnaire-based case-control study of 529 head and neck cancer patients and 529 hospital-based controls matched for age, sex, and smoking status [83]. Compared with never users, aspirin use was associated with a significant 25 percent reduction in the risk of head and neck cancer, and there was a trend toward greater risk reduction with increasing duration of use. When analyzed by smoking and alcohol exposure levels, participants exposed to heavy smoking and alcohol use did not appear to benefit from the protective effect of aspirin.
●A population-based cohort study from Denmark failed to find evidence that NSAIDs protect against oral cancer [84]. However, the study did not control for tobacco or alcohol use and considered only prescription use of NSAIDs.
●A case-control study reported a protective effect of NSAID intake on the risk of head and neck cancer in tobacco users [85] but was later retracted because of concerns over data fabrication [86].
Long-term ingestion of NSAIDs may be limited by toxicity. A novel approach, topical application, was used in a double-blind trial in which 57 patients with oropharyngeal leukoplakia were randomly assigned (in a 2:1 ratio) to a 30 second, twice daily rinse with ketorolac (0.1 percent solution) or placebo [87]. Compared with controls, the treated group did not have a greater objective decrease in the size of the leukoplakia lesions (which were bidimensionally measured at days 0 and 90) or in their histologic appearance.
Since surrogate markers of exposure in the oral cavity (eg, prostaglandin E2 inhibition) were not studied, it is difficult to ascertain whether this negative result reflects ineffective delivery of the chemopreventive agent to the oral epithelial cells or an incorrect hypothesis.
A large body of laboratory investigations supports the use of COX-2 inhibitors in treating several cancers [88-90], including SCCHN [91-94]. COX-2 plays a role in the carcinogenesis of epithelial tumors by influencing key processes, such as apoptosis, angiogenesis, and proliferation, as well as invasion and metastasis [89]. In SCCHN the expression of COX-2 is noted by western blot analysis in the mucosa of subjects at different stages of carcinogenesis, unlike in the normal mucosa [95]. COX-2 is also expressed in early and intermediate stages of carcinogenesis with progressively increased expression throughout all of these stages, further supporting the role it may play in the carcinogenesis process and lending support to the rationale for using COX-2 inhibitors in SCCHN chemoprevention [94].
The evidence from earlier studies that NSAIDs may delay the growth of head and neck tumor cell lines [96-98] has led to the use of COX-2 inhibitors as single agents in patients with dysplasia [99].
The use of COX-2 inhibitors was, however, associated with a small but significant increase in the incidence of cardiovascular complications in some placebo-controlled trials [100,101].
Epidermal growth factor receptor inhibition — EGFR is highly expressed in SCCHN [18,102] as well as a wide variety of other tumors (eg, colon, non-small cell lung, breast, kidney, head and neck, bladder, and ovarian cancers) [103-105]. However, data are limited on the efficacy of EGFR inhibitors such as cetuximab [106] or erlotinib [107].
Erlotinib did not demonstrate efficacy as a chemoprevention agent for SCCHN in one phase III randomized trial [107]. In the chemoprevention study Erlotinib Prevention of Oral Cancer (EPOC), 150 patients with documented oral leukoplakia (ie, oral premalignant lesions) and high-risk loss of heterozygosity (LOH) profiles were randomly assigned to either erlotinib or placebo for one year, followed by observation for the development of cancer. Patients with a high-risk LOH profile included either those without cancer but with LOH in 3p and/or 9p and one other specific chromosomal locus, or those with curatively treated oral cancer and LOH in 3p and/or 9p. After a median follow-up of three years, the cancer-free survival rates were similar between those treated with erlotinib and those treated with placebo (74 versus 70 percent, HR 1.27, 95% CI 0.68-2.38).
Molecular profiling of premalignant oral cavity lesions remains an active area of investigation. In the erlotinib chemoprevention study discussed above, LOH was validated as a marker of oral cancer risk and was associated with increased EGFR copy numbers [107]. Another separate analysis demonstrated an association with the development of oral cancer in oral precancers with select molecular features (eg, high tumor mutation burden, TP53 mutations, and a specific 12-gene RNA signature) [108].
Combined inhibition of epidermal growth factor receptor and cyclooxygenase 2 — The interaction between the EGFR and COX-2 pathways has been extensively investigated in preclinical as well as clinical studies, with evidence that targeting these two pathways can synergistically or additively reverse the progression of SCCHN [109,110].
These observations provided the rationale for using the combination of EGFR and COX-2 inhibitors in SCCHN chemoprevention trials.
In one such study, the maximum tolerated dose of erlotinib in combination with celecoxib at 400 mg twice daily was 50 mg/day; skin rash was the main toxicity. Even though the small number of patients limited the ability to draw conclusions, the reversal of dysplasia to a lower grade was reported to be 63 percent (complete response 43 percent, partial response 14 percent, stable disease 29 percent, disease progression 14 percent) and response to therapy was correlated with downregulation of EGFR and phospho-ERK in serial biopsies, suggesting the possible utility of these markers as surrogate endpoints for response [51,52].
Since tumorigenesis has been associated with the activation of the mammalian (mechanistic) target of rapamycin (mTOR) pathway, there has been interest in exploring mTOR inhibitors as possible effective chemopreventive agents in SCCHN [111]. Interestingly, the combined chemopreventive effect of celecoxib and erlotinib seemed to exert some modulation on the mTOR pathway [52]. Though worth exploring, mTOR inhibitors have not reached clinical applications in SCCHN chemoprevention.
Natural compounds — An array of natural compounds, such as green tea extract, curcumins, resveratrol, luteolin, pomegranate juice, and soybeans, is under investigation in head and neck chemoprevention trials. These compounds contain high levels of polyphenols, which have antioxidant properties and inhibit carcinogenesis mediated by downstream signaling pathways [37,112-114]. Dietary intake of certain nutrients (vitamin C, carotene, vitamin E) has also been inversely linked to the incidence of oral and pharyngeal cancers [115].
There is limited evidence for the efficacy of increased consumption of green tea and its associated derivatives, either alone or in combination with targeted agents (eg, erlotinib) for chemoprevention of SCCHN. Data are as follows:
●Green tea, which contains polyphenols, may have anticancer properties based on preclinical studies and an early phase clinical trial [116-119]. As an example, the combination of green tea polyphenon E (PPE) 200 mg three times a day and erlotinib 100 mg daily was evaluated in a phase 1 trial in patients with advanced premalignant lesions of the head and neck; a majority had severe dysplasia. In this study, the pathologic complete response rate was 47 percent and the regimen was well tolerated, suggesting its possible application in further clinical trials [120].
●Another main ingredient found in green tea extracts, epigallocatechin gallate (EGCG), has also been linked to modulation of the signaling pathways that play a key role in the carcinogenesis process. EGCG also seems to have a synergistic or additive effect when combined with other natural compounds, such as lycopene [121,122]. In preclinical studies, EGCG also demonstrated synergistic activity on cancer cells in combination with the EGFR tyrosine kinase inhibitor erlotinib [123]. Further clinical studies are needed to validate these results.
Immunotherapy — Checkpoint inhibitor immunotherapy may have a possible role in primary chemoprevention of SCCHN, and further clinical trials are needed. The potential efficacy of programmed cell death receptor 1 (PD-1)/programmed cell death ligand 1 (PD-L1) checkpoint inhibitors for primary chemoprevention is suggested by the existence of an initial immune-privileged site after initial viral HPV infection [124] and the established activity of immunotherapy in those with advanced or metastatic HNSCC, which is discussed separately. (See "Treatment of metastatic and recurrent head and neck cancer".)
HPV-related oropharyngeal carcinoma — Human papillomavirus (HPV) 16 is an important risk factor for the development of oropharyngeal cancer and is responsible for the increased incidence of this disease. (See "Epidemiology, staging, and clinical presentation of human papillomavirus associated head and neck cancer", section on 'Epidemiology'.)
The E6 and E7 HPV viral proteins in HPV 16 are responsible for promoting cell cycle progression and viral DNA replication in differentiated normal mucosal epithelial cells. While E6 protein binds to p53 tumor suppressor protein and induces its degradation, the E7 protein ubiquitinates the retinoblastoma tumor suppressor protein. This process leads to driving the mucosal cells into S phase and induces cellular DNA synthesis [125,126].
Studies using oral rinses and polymerase chain reaction (PCR) for HPV DNA have identified 7 percent of the adult population with active oral HPV infection, but only approximately 3 percent of males and 1 percent of females have high-risk HPV infection [127], and this group could form a target for chemoprevention.
Prevention of HPV-related oropharyngeal carcinoma — The improvement in overall survival for patients with SCCHN has been unfortunately linked specifically to the increased incidence of human papillomavirus (HPV)-related oropharyngeal carcinomas rather than early intervention or detection of disease. The high prevalence of HPV-related oral infection and the increase in HPV-related cancers call for chemopreventive measures to these malignancies.
HPV vaccines are effective in preventing HPV-related malignancies [128-131]. Despite this, the use of these vaccines is relatively limited at the present. The value of HPV vaccination in reducing the risk of oral HPV infection has not been established.
There has been an interest in exploring whether a "Pap test equivalent" would be useful for early detection of HPV-related oropharyngeal carcinoma. Cytologic abnormalities in addition to the presence of HPV 16 were strong predictors of the occurrence of oropharyngeal carcinoma [132]. There has also been no clear association between HPV infection and cytologic abnormalities in the oral mucosa, raising the question as to whether HPV infection leads to a precancerous lesion that would precede the occurrence of invasive oropharyngeal carcinoma. The problem may also reside in the location rather than the existence of these lesions in HPV positive disease, which if occurring in the lower part of the tonsillar crypts, would pose difficulty in both detection and monitoring [50].
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Head and neck cancer".)
SUMMARY AND RECOMMENDATIONS
●Screening for oral cavity cancer – We recommend oral cavity cancer screening for squamous cell carcinoma of the head and neck (SCCHN) in all patients, including high-risk groups of alcohol and tobacco users. Regular screening reduced mortality from oral cavity cancer in a randomized trial, likely due to detection of early-stage disease. (See 'Screening' above.)
●Is there an established approach to chemoprevention? – Despite the advances with surrogate markers in premalignant lesions, which can assess the risk of developing invasive carcinoma, there is no established approach for chemoprevention of SCCHN. (See 'Rationale for chemoprevention' above and 'Chemoprevention trials' above.)
●Potential targets for chemoprevention
•Podoplanin expression may be a powerful predictive factor for oral cancer development in patients with leukoplakia and may guide future therapeutic trials in chemoprevention. (See 'Surrogate markers and potential target agents' above.)
•Natural compounds, including green tea polyphenon E (PPE), luteolin, resveratrol, and others, are potentially attractive to be further explored as chemopreventive agents since they are components of daily food or beverage and are potentially nontoxic or less toxic compared with synthetic compounds. (See 'Natural compounds' above.)
•Chemopreventive strategies based on known or potential molecular targets are warranted to help prevent the development of both human papillomavirus (HPV)-related and non-HPV-related oropharyngeal carcinoma (ie, tobacco-induced carcinogenesis) in patients with premalignant disease. (See "Head and neck squamous cell carcinogenesis: Molecular and genetic alterations".)
•Data are limited on the efficacy of epidermal growth factor receptor (EGFR) inhibitors, such as cetuximab or erlotinib, in SCCHN chemoprevention. (See 'Epidermal growth factor receptor inhibition' above.)
ACKNOWLEDGMENTS — The UpToDate editorial staff acknowledges Charles Rudin, MD, PhD, and Shanthi Marur, MD, MBBS, who contributed to an earlier version of this topic review.
