Chemoprevention of Head and Neck Cancer with Green Tea Polyphenols

Cancer is the leading cause of death for people under the age of 85 years (1). About 47,000 cases of squamous cell carcinoma of the head and neck (SCCHN) were estimated to occur in the year 2009, with an expected 11,000 deaths from the disease (1, 2). SCCHN causes significant morbidity and mortality with a 5-year survival rate of <50% (2–5). A common cause of mortality in SCCHN survivors is second primary tumor, which occurs at an annual rate of 3% to 5% (6–8). SCCHN occurs as a consequence of accumulating genetic instabilities from exposure to various carcinogens, such as cigarettes, alcohol, marijuana, and betel chewing (3, 9–11). Figure 1 illustrates the molecular progression of SCCHN.

Recently, SCCHN chemoprevention research has made major advances, including novel clinical trial designs suited for chemoprevention (12), use of biomarkers to identify high-risk patients (12), use of molecularly targeted agents and natural compounds in chemoprevention trials, and the development of an oral-specific carcinogenesis animal model (13). An ideal chemopreventive agent should be nontoxic, potent, inexpensive, and easily available. Research over the last several decades has identified numerous natural compounds, many of which are present in the diet and have the potential to suppress the development of multiple cancers (14). Among many natural compounds, green tea polyphenols (GTP) including (−)-epigallocatechin-3-gallate (EGCG) exhibit high promise for chemoprevention in epidemiologic, preclinical, and early clinical studies. EGCG also shows strong synergistic or additive antitumor activities with many natural or synthetic compounds. This review will provide an update of the potential of GTPs, particularly EGCG, for the chemoprevention of SCCHN.

Green tea is produced from the nonfermented leaves of the plant Camellia sinensis. Figure 2 shows the four major polyphenols present in green tea extract (GTE). EGCG is the most potent and abundant antioxidant present in green tea and has been extensively investigated for its chemopreventive and therapeutic potential (15). Several formulations of GTE have been used in clinical trials, in which the EGCG content varies from 13% to 70% (Table 1).

One of the first pieces of evidence of the chemopreventive effect of EGCG was reported in 1987, when the inhibitory effects of EGCG on teleocidin-induced tumor promotion in mouse skin were shown (16). Its antitumor effects were also shown by the regression of experimentally induced skin papilloma in mice by orally administered green tea, i.p.-administered GTP fraction, or i.p. EGCG (17). Topical application of EGCG was shown to induce apoptosis in UVB-induced skin tumors in mice (18). Numerous preclinical studies have followed since, showing the mechanisms of antitumor effects of GTP and EGCG in various cancer models.

EGCG exerts its chemopreventive actions by modulating multiple signaling pathways at various cellular levels, the ultimate outcomes of which are apoptosis, cell cycle arrest, growth inhibition, antiangiogenesis, and inhibition of metastasis. Figure 3 illustrates the molecular targets modulated by EGCG. At the cell membrane, EGCG inhibits the activation of receptor tyrosine kinases, such as epidermal growth factor receptor (EGFR), human EGFR 2 and 3, insulin-like growth factor-I receptor and vascular endothelial growth factor (VEGF) receptor (VEGFR), and their downstream effectors such as pAkt and pERK (19–29). Of these, EGFR seems to be the most active target of EGCG in in vitro and in vivo SCCHN models (19, 25, 28). A correlation between pEGFR inhibition, Akt phosphorylation, and tumor growth inhibition was observed in SCCHN xenografted tumor tissues (19). EGCG in combination with curcumin inhibited VEGFR-1 activation in a breast cancer xenograft model (30). EGCG also inhibited VEGFR-2 activation in a hepatocellular carcinoma xenograft model (31). Inhibition of VEGF by green tea preparations was also observed in animal models of breast (32, 33) and prostate cancers (34). Downregulation of angiogenic stromal VEGF was observed in a clinical trial with GTE (35). GTP treatment also decreased serum VEGF levels in prostate cancer patients (36). Laminin receptor was identified as a potential receptor for EGCG to modulate several important intracellular signaling pathways (37–39).

The effects of EGCG on cytoplasmic signaling molecules in cell culture and animal models include inhibition of Akt, extracellular signal-related kinase (ERK)1/2, and mitogen-activated protein kinase or ERK kinase (MEK) phosphorylation (40–42), in addition to modulating phosphoinositide 3-kinase/Akt/mammalian target of rapamycin (41, 43), and signal transducer and activator of transcription 3 (25).

EGCG also modulates the function of certain transcription factors, namely NF-κB and activator protein (AP-1) in cell culture and animal studies (44–48). NF-κB facilitates the transcription of genes involved in inflammation, immunity, and carcinogenesis. In a normal human epidermal keratinocyte model, pretreatment with EGCG caused significant inhibition of UVB-induced NF-κB/p65 activation and its nuclear translocation (47). As a consequence of AP-1 inhibition, expression of its target molecules, such as cyclin D1 and cyclooxygenase-2 are reduced, inducing apoptosis and reducing inflammatory response (21, 25, 49). Furthermore, EGCG induces apoptosis and G₀-G₁ arrest in several cell lines through activation of p53 and its downstream targets p21, p57, and Bax (50–53). Evidence also suggests that EGCG induces the expression of p73, which is important for apoptosis and the expression of a subset of p53-target genes (54, 55).

EGCG has shown a dose-dependent inhibition of invasion and migration of human oral cancer, which is thought to be related to decreased production of matrix metalloproteinase-2/9 and urokinase plasminogen secretion (56). Topically administered GTP in UVB-induced tumors also inhibited the expression of matrix metalloproteinase-2 and matrix metalloproteinase-9 (57).

It seems that EGCG modulates multiple molecular targets, which largely depend on the experimental context. Moreover, the molecular targets affected by EGCG in cell cultures, animal models, and clinical samples are sometimes different. Furthermore, the doses of EGCG and administration routes used in these three situations are different. Although it is not currently clear which of these EGCG targets are most critical for its chemopreventive effects observed in clinical trials, this is an important issue that may become better elucidated once more clinical trial results are available.

Curcumin is another popular natural compound that has been studied extensively (58–62). EGCG showed synergistic effects with curcumin in SCCHN cells (63). The median effect analysis revealed that the combination of EGCG and curcumin exhibited synergistic growth inhibition of premalignant and malignant cells (63). Combination of topical curcumin and oral green tea also resulted in superior antitumor effects in 7,12-dimethylbenz[a]anthracene–induced carcinogenesis in Syrian hamsters (64). The combination significantly decreased the number and volume of visible oral tumors, purportedly mediated through the suppression of cell proliferation, induction of apoptosis, and inhibition of angiogenesis (64). This combination regimen has also shown enhanced or synergistic effects in other cancer models such as a hormone receptor–positive breast cancer model (30). The combination induced greater tumor volume reduction and inhibition of VEGFR protein compared with the single agents in nude mice (30).

It also seems that the sequence of administration of these agents affects the synergistic effect. In an in vitro study in chronic lymphocytic leukemia B cells, although each agent alone was active, simultaneous administration of the agents reduced apoptosis (65). However, treatment with EGCG followed by curcumin showed synergistic apoptotic effects (65).

EGFR overexpression seems to play a role in the early part of SCCHN carcinogenesis, correlates with progression of dysplasia (66), and associates with poor clinical outcome in invasive SCCHN (67, 68). Our laboratory has shown synergistic growth inhibition of SCCHN by EGCG and erlotinib, mediated through greater inhibition of pEGFR and pAkt (19). The combination of erlotinib and EGCG was associated with greater tumor growth inhibition compared with single agent treatments (19). The mechanism of synergy was thought to be mediated through more sustained inhibition of Akt phosphorylation compared with single agent treatment (19). A subsequent study suggested a critical role of activation of p53 and inhibition of NF-κB signaling pathways (69). The same combination also showed enhanced antiproliferative effects in several erlotinb-sensitive and erlotinb-resistant non–small cell lung cancer cell lines and in severe combined immunodeficient mice bearing erlotinib-resistant non–small cell lung cancer tumors (70).

The apoptosis-inducing effects of EGCG were drastically enhanced by sulindac and tamoxifen in lung cancer cell lines (71). Cotreatment with EGCG plus celecoxib induced synergistic apoptosis in lung cancer cell lines (72). Although neither EGCG nor celecoxib alone induced growth arrest or the expression of DNA damage–inducible-153 (GADD153) mRNA and protein, cotreatment with both compounds strongly induced the expression of both GADD153 mRNA and protein. Moreover, inhibition of ERK1/2 activation using chemical inhibitors inhibited the expression of GADD153 and apoptosis, suggesting that combination of EGCG and celecoxib induced synergistic apoptosis by upregulating GADD153 through ERK1/2 (72). Combined treatment with GTE and sulindac resulted in a significantly greater reduction in tumor number (from 72.3 ± 28.3 to 32.0 ± 18.7) than that achieved with either single agent alone (to 56.7 ± 3.5 and 49.0 ± 12.7 with GTE or sulindac alone, respectively) in multiple intestinal neoplasia mice that harbored a mutated Apc gene (73).

EGCG combined with the selective cyclooxygenase-2 inhibitor NS-398 resulted in synergistic cell growth inhibition, apoptosis induction, and inhibition of NF-κB in prostate cancer cell lines and enhanced tumor growth inhibition with reduced serum levels of prostate-specific antigen (PSA) and insulin-like growth factor-1 in nude mice bearing androgen-sensitive prostate cancer cells (74).

Several studies have also shown enhanced or synergistic antitumor effects in in vitro and in vivo models of breast cancers when EGCG was combined with tamoxifen or soy phytochemical concentrate (75–77). Furthermore, in prostate cancer models, dihydrotestosterone increased the sensitivity of prostate cancer cells to EGCG (78). Moreover, in a carcinogen-induced lung tumorigenesis mouse model, the combination of atorvastatin and EGCG at relatively low doses was able to induce synergistic tumor growth inhibition (79).

Despite mounting preclinical evidence to support the efficacy of GTPs, only a few clinical trials have been conducted (Table 1). The first clinical trial using green tea for oral premalignant lesions (OPL) was a double-blind, placebo-controlled randomized trial in patients with oral leukoplakia receiving either 760 mg of mixed tea (40% GTPs) capsules q.i.d. plus 10% mixed tea ointment topically, or placebo plus topical glycerin (80). After 6 months of treatment, treatment groups achieved a response rate of 37.9%, compared with 10% in the control group (80). This finding correlated with histopathologic results showing a reduced number of EGFR-positive cells (80). Pisters et al. (81) reported a phase I trial using a GTE in an attempt to find the maximum tolerated dose. The percentage of total catechins, EGCG, epigallocatechin (EGC), epicatechin-3-gallate (ECG), epicatchins (EC), and caffeine contents of the GTE preparation were 26.9, 13.2, 8.3, 3.3, 2.2, and 6.8, respectively. Patients were given GTE either once or thrice daily for 4 weeks, up to a maximum of 6 months (81). The dose-limiting toxicities were tremors, cough, constipation, and headache, which were thought to be related to caffeine components of the GTE (81). Although there was no clinical response observed, oral GTE at 1.0 g/m² thrice daily for at least 6 months was recommended (81).

Recently, Tsao et al. (35) reported a randomized phase II trial using the same GTE preparation as Pisters et al. (81) for patients with high-risk OPLs. Patients were randomized into one of the four arms—500, 750, or 1,000 mg/m² GTE, or placebo, t.i.d. for 12 weeks, and clinical response was measured by the Response Evaluation Criteria in Solid Tumors (35). At 12 weeks, there was 50% OPL clinical response rate in the treatment arms, versus 18.2% in the control arm. A higher response rate was observed in the two higher dose GTE arms, of 58.8% versus 36.4% in the lower dose arm, suggesting dose-response effects of GTE (35). GTE was very well tolerated with only three grade 3 toxicities, including insomnia, diarrhea, and oral/neck pain, and no grade 4 toxicity. Analysis of patients' demographics among clinical responders showed a higher response rate in never drinkers (P = 0.001). Other demographic characteristics did not affect the clinical response. At a median follow-up of 27.5 months, there was no difference in oral cancer–free survival between the GTE and placebo arms. Baseline biomarker characteristics, such as higher stromal VEGF expression, were associated with clinical but not histologic response. Interestingly, stromal VEGF and cyclin D1 expression were downregulated in clinically responsive GTE patients and upregulated in nonresponsive patients (35). These biomarker characteristics would be important predictive factors in profiling the patient population that would benefit most from GTE treatment.

Patients with high-grade prostate intraepithelial neoplasia (HG-PIN) received either 200 mg of green tea catechin containing ∼103.6 mg of EGCG, orally, t.i.d., or placebo. The primary end point was prevalence of prostate cancer. After 1 year of follow-up, 3.3% patients were diagnosed with prostate cancer in the treatment group compared with 30% in the placebo group (82). Multivariate analysis including age, PSA, prostate volume, HG-PIN, and monofocal or plurifocal HG-PIN lesions showed no significant differences between the two arms. In a 2-year follow-up, only one prostate cancer was diagnosed among 13 green tea catechin–treated patients and 2 among 9 placebo-treated patients (83). A daily dose of 311 mg of EGCG in green tea catechin formulation was able to induce a clinical response in these patients. Although the results of this clinical study for prostate cancer prevention are dramatic and highly promising, the data need to be validated in larger randomized trials.

Oral administration of GTE, three 500 mg tablets, each containing 52.5 mg EGCG, 12.3 mg EC, 34.6 mg EGC, 11.1 mg ECG, and 15.7 mg caffeine, daily for 12 months, in addition to a tea drinking life-style, showed its efficacy in preventing incidence of metachronous adenoma in patients 1 year postpolypectomy (84). Twelve-month follow-up colonoscopy showed 31% incidence of colonic adenoma in the control arm versus 15% in the GTE arm (84). It was not reported whether the GTE treatment had any effect on the incidence of colon cancer. Furthermore, patients with human papillomavirus–infected cervical premalignant lesions were treated with various formulations of GTE: 200 mg polyphenon E orally t.i.d., 200 mg EGCG orally t.i.d., 200 mg polyphenon E orally t.i.d. plus topical polyphenon E twice weekly, or polyphenon E topical treatment only twice weekly for 8 to 12 weeks (85). Compared with 4 of 39 nontreated patients, 35 of 51 treated patients achieved a clinical response in reduction in human papillomavirus DNA titer or improvement in cytology or tissue biopsy after treatment (85).

Although these clinical trials used different GTE formulations with different percentages of EGCG content, none have encountered serious adverse effects. It seems that a GTE formulation containing EGCG levels as low as 158 mg daily for 12 months was able to induce a clinical response (84). GTE is also well tolerated at doses as high as 4,200 mg/m², containing 554.4 mg/m² of EGCG, daily for at least 6 months (81).

Since the first clinical chemoprevention study in 1986 (86), the field of SCCHN chemoprevention has made remarkable advances and entered into mainstream cancer research. Although none of the agents has yet been translated into clinical practice, there are several agents under clinical investigation that hold strong promise. Among the most promising compounds are GTPs. The recent phase II clinical trial by Tsao et al. (35), showing a dose-response relationship of GTE against OPLs that correlates with biomarker response, has generated a tangible momentum in SCCHN chemoprevention. Several approaches should be pursued to maximize yields from this promising agent in future investigations. First, GTE should be used in combination to generate synergy and to reduce toxicity because this agent has shown synergistic/additive antitumor effects in many cancer models (19, 63, 65, 72, 74, 76, 79).

Second, investigations to enhance the bioavailability and potency of GTE should be encouraged and validated more vigorously in preclinical studies. One approach is the use of a prodrug formulation of EGCG, many of which have shown increased activity and bioavailability in vitro and in vivo (87–90). Another promising approach is to formulate nanoparticles for effective delivery. Siddiqui et al. (91) have shown ∼10-fold higher potency of nanoparticle-encapsulated EGCG. Both of these approaches are novel and hold high promise for chemoprevention, thus warranting further validation in animal studies and clinical settings. However, efforts must be made not to compromise the safety or cost of this agent.

No potential conflicts of interest were disclosed.

We thank Dr. Anthea Hammond for her critical and editorial review of this article.

Grant Support: Grants from the NIH (P50 CA128613, U01 CA101244, and R01 CA112643). D.M. Shin is Distinguished Cancer Scholar of the Georgia Cancer Coalition. A.R.M. Ruhul Amin is a recipient of Specialized Programs of Research Excellence Career Development Award.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.