Targeted Molecular Therapy in Head and Neck Squamous Cell Carcinoma Overview of Targeted Molecular Therapy in HNSCC

In 2001, 500,000 cases of head and neck squamous cell carcinoma (HNSCC) were reported worldwide. In the United States, approximately 40,000 new cases of HNSCC were reported, with nearly 12,000 deaths related to HNSCC. Early-stage HNSCC disease is treated relatively well with single-modality therapy (either surgery or radiation alone). However, nearly 66% of patients present with advanced disease (stage III and IV), and fewer than 30% of these patients are cured.

The management of advanced HNSCC consists of multiple-modality therapy with surgery, radiation, and chemotherapy. Despite significant improvements in these modalities, long-term survival rates in patients with advanced-stage HNSCC have not increased significantly in the past 30 years.

Selective versus nonselective therapies

The current conventional modalities (surgery, radiation, chemotherapy) are nonselective and can cause damage to normal tissue. In particular, chemoradiotherapy is associated with systemic toxicities that often reduce compliance and prevent timely completion of therapy.^[1] In an attempt to improve outcomes in HNSCC, most research in the field is now focusing on the molecular biology of HNSCC in an attempt to target select pathways involved in carcinogenesis.

With the increased understanding of molecular mechanisms and basic pathways in the pathogenesis of squamous cell cancer of the head and neck, these pathways may be modified, and rational approaches in cancer therapy at the molecular level may be created. Some of the new approaches depend on tumor biology and aim specifically to inhibit tumor growth and metastasis by targeting the tumor microenvironment or vasculature (leaving normal cells unaffected) or focusing on specific protein or signal transduction pathways.^[2]

The goal of specific molecular targets in cancer therapy is to create a “magic bullet” that selectively kills cancer cells. As our understanding of the molecular biology of HNSCC continues to develop, we can target the specific components of cancer cells that are not found in normal cells. Ideal targets should be both specific to cancer cells and commonly found in cancer cells.

Targeted molecular therapy, like therapy with monoclonal antibodies, gene therapy, and other therapies, has limited or nonexistent side effects on normal cells of the body, unlike present modalities such as surgery, chemotherapy, and radiotherapy. Targeted molecular therapy can also act as a complement to other existing cancer therapies.

HER and EGFR

The HER (erbB) family of transmembrane receptor tyrosine kinases is one of the cytostatic targets in tumor cell growth and survival. This family, which includes epidermal growth factor receptor (EGFR), plays a pivotal role in normal cell growth, lineage determination, repair, and functional differentiation. Overexpression of EGFR is recognized in more than 80% of squamous cell cancers, and this overexpression is associated with a poor prognosis.

Targeted molecular therapy against EGFR has shown promise as an adjuvant therapy in preliminary studies in several solid tumors, including head and neck cancer. Selective compounds have been developed that target either the extracellular ligand-binding region of the EGFR (including a number of monoclonal antibodies [MAbs], immunotoxins, and ligand-binding cytotoxic agents) or the intracellular tyrosine kinase region (including various small-molecule inhibitors).

Go to Imaging of Nasopharyngeal and Laryngeal Squamous Cell Carcinoma and Head and Neck Squamous Cell Carcinoma for complete information on these topics.

Various techniques have been developed for targeting cancer cells: gene therapy, monoclonal antibodies (MAbs), antibody toxin conjugates, small-molecule inhibitors, antisense molecules, and tumor vaccines.

The goal of gene therapy is to introduce new genetic material into cancer cells that selectively kills them without causing toxicity to the surrounding cells. This task can be accomplished by replacing tumor suppressor genes that have been lost or mutated, selectively inserting genes that produce cytotoxic substances, or modifying the immune system to destroy the tumor cells. The major barrier in successful gene therapy is producing a vector that selectively infects all tumor cells within a tumor.

MAbs and antibody toxin conjugates can be targeted to specific receptors or proteins found on cancer cells. MAbs can block receptors and prevent potential growth signals. Antibodies can also be conjugated to toxins and specifically kill the tumor cells they bind.

Antisense molecules are a small, complementary, single-stranded type of DNA that binds targeted messenger RNA (mRNA) within the cell and prevents specific protein translation. Antisense molecules can be targeted toward specific proteins that are crucial in tumorigenesis.

Small-molecule inhibitors can bind and inhibit specific receptors or enzymes in cancer cells. These small-molecule inhibitors can be targeted toward crucial steps in tumorigenesis.

Tumor vaccines act to stimulate the patient’s immune system to attack cancer cells. These tumor vaccines can be developed from a patient’s tumor cells. In this process, mRNA is isolated from a tumor biopsy sample, amplified, and incorporated into human antigen-presenting cells (APCs). The APCs are then intravenously given to the patient to stimulate the patient’s immune system to activate antitumor T cells.

Steps leading to carcinogenesis

Much research has been devoted to the events leading to carcinogenesis. A multistep model of carcinogenesis identifies discrete genetic events that occur during the progression from normal colonic mucosa to precancerous polyps to invasive tumors. The precise number and order of steps in tumorigenesis remains unknown.

Hahn and colleagues have demonstrated 6 steps believed to be necessary for the development of cancer, as follows^[3] :

• Acquisition of autonomous proliferative signaling
• Inhibition of growth inhibitory signals
• Evasion of programmed cell death
• Immortalization
• Angiogenesis
• Tissue invasion and metastasis

These 6 steps are necessary in tumorigenesis, and each offers a potential target for molecular therapy.

Characteristics of ideal molecular targets

Molecular physicians and researchers have discovered differences between cancer cells and normal cells. These differences led to the discovery of the targets that are only found on cancer cells and not in normal tissue and those targets that are differentially overexpressed in tumor cells compared to normal tissue. Not all molecules are good targets for cancer therapy at the molecular level.

The ideal target should have properties such as the following:

• Commonly found on cancer cells
• Differentially expressed or differentially functional in tumor versus nontumor host tissues
• Specific to cancer cells
• Causally related to tumor cell viability, progression, or both
• Involved in several aspects of the carcinogenesis pathway
• Measurable in diagnostic tumor material

Measurement of the target should provide some predictive information about the potential clinical response to the molecular targeted drug. The target, at some level, should validate mechanisms of resistance. Drug-mediated modulation of the molecular target should be validated in vivo.

Targets already used for head and neck cancer in animal models include epidermal growth factor receptors (EGFRs), interleukin-13 receptor (IL-13R), protein kinase activator, and others that are in various phase I, II, and III studies.

The acquisition of autonomous proliferative signaling occurs through the activation of proto-oncogenes into oncogenes. Proto-oncogenes are important in normal intracellular signaling pathways, which regulate cell growth and differentiation. When oncogenes are activated, cancer cells acquire the ability to proliferate without the need for exogenous growth factors.

Role of HER/EGFR family in signaling

The HER family consists of 4 closely related transmembrane receptors: HER-1 (ERBB1)/epidermal growth factor receptor (EGFR), HER-2 (ERBB2), HER-3 (ERBB3), and HER-4 (ERBB4). These receptors are structurally similar but have unique characteristics that dictate their signaling specificity. Each receptor has an extracellular ligand-binding domain, a transmembrane region that anchors the receptor to the cell, and an intracellular cytoplasmic domain that contains a tyrosine kinase region and a C terminal tail (see the image below).^[4]

Binding of ligands to the extracellular domain of EGFR results in receptor oligomerization, activation of the receptor’s tyrosine kinase activity, and receptor autophosphorylation in several C terminal tyrosine residues.^[5] These phosphorylated tyrosines serve as binding sites for a number of cytoplasmic signal-transducing molecules. The activation of these pathways downstream of the EGFR leads to cell proliferation, differentiation, and migration or motility and adhesion, protection from apoptosis, enhanced survival, and gene transcription.^[6]

The coexpression of EGFR and ligands at tumor sites allows for EGFR activation via autocrine/paracrine mechanisms. In support of the operational nature of these signaling pathways in EGFR-expressing tumor cells, interruption of signaling with various EGFR inhibitors has been shown to inhibit tumor cell proliferation and/or viability both in vitro and in vivo.^[7]

These observations, in combination with (1) the ability to identify EGFR-expressing human tumors in diagnostic tissue from patients, (2) the association of EGFR overexpression with poor patient prognosis, and (3) the lack of a critical physiologic role of EGFR in healthy adults, have all suggested this network as an ideal target for novel therapeutic strategies.

The specificity and potency of the signaling output from activated EGFR is highly dependent on the type of activating ligand, as well as on the cellular levels of coreceptors such as HER-2/neu (ERBB2), HER-3/neu (ERBB3), and HER-4/neu (ERBB4). Several combinations of activation of receptors ultimately generate effective synergistic actions that lead to optimum signal transduction and tumor progression. Hence, blockade of this network yields maximum results.

EGFR function is essential for embryogenesis and organogenesis.^[4] Mice lacking the EGFR gene have severely impaired development of multiple organs (eg, skin, brain, lung, kidney, liver, intestinal epithelium, and eye)^[8] and survive for only a short time after birth. In adults, EGFR has an important role in the repair of some epithelia, as supported by the skin and gastrointestinal toxicity observed in trials with EGFR inhibitors.^{[8, 9]} Both normal cells and cancer cells rely on EGFR signals, but in normal cells, the signal is strictly regulated (see the image below).

Role of other factors in signaling

Nuclear factor kappaB (NF-kB) is a transcription factor that activates cell growth in response to inflammation. Mutations in NF-kB are involved in tumorigenesis through increasing cell proliferation by up-regulating the production of cyclin D1. Small molecular inhibitors have been developed that inhibit NF-kB activation and reduce tumorigenicity of HNSCC in animal models.

The ras oncogene stimulates tumorigenesis through a cascade of signals that stimulate cell growth. Farnesyl transferase inhibitors (FTIs) can specifically inhibit the ras oncogene. FTIs act by inhibiting farnesylation, which is a crucial posttranslation modification of the ras oncogene. FTIs have been shown to inhibit growth of tumor cell lines. Phase II clinical trials have not been able to show significant response rates, and further investigation is warranted.

Numerous other potential molecular targets are involved in the acquisition of autonomous proliferation signaling in neoplasms. Signal transducers and activators of transcription (STATs) are a family of proteins responsible for transmitting growth signals from the cell surface to the nucleus. Mutations in STATs, specifically signal transducer and activator of transcription 3 (STAT3), have been shown to play a role in the development of HNSCC.

Fibroblast growth factors are also up-regulated in HNSCC and, when inhibited in vitro, result in significant inhibition of cell growth. Hepatocyte growth factor has also been implicated in tumorigenesis.

Regulation of HER/EGFR activity

The epidermal growth factor axis is involved in the regulation of normal cell proliferation. This axis is up-regulated in many carcinomas and is detected in greater than 90% of head and neck squamous cell carcinomas (HNSCCs).^[10] Increased levels of epidermal growth factor receptor (EGFR) have been shown to be an independent predictor of a decreased chance of disease-free survival.

Dysregulation of HER-1/EGFR activity can occur because of several mechanisms, including receptor overexpression, ligand overproduction, the presence of constitutively active receptor mutants, and cross-talk with other amplified receptors and signaling systems, among others. Increased understanding of the structure and function of these receptors has led to the development of various molecular targeted agents, such as monoclonal antibodies (MAbs) and small-molecule tyrosine kinase inhibitors.

Overexpression of the EGFR tyrosine kinase has been documented across all stages of disease, including precancerous lesions, early cancers, and advanced cancers. Some ductal carcinomas in situ, precursor lesions to invasive breast carcinoma, express high levels of EGFR tyrosine kinase. Increased expression of EGFR and transforming growth factor–alpha (TGF-α) has been documented in early stage non–small-cell lung cancer (NSCLC) and in premalignant bronchial biopsy samples.

In light of a relationship between overexpression of EGFR and clinically aggressive malignant disease, EGFR has emerged as a promising target for treatment of patients with HNSCC. The identification of HER-1/EGFR as an important receptor in the pathogenesis of human tumors has prompted considerable research, focusing particularly on the HER family signaling network.

Therapeutic molecular targeting strategies

EGFR has many naturally occurring ligands, such as epidermal growth factor (EGF) and TGF-α. Multiple ligands have been developed to bind to the receptor. These ligands can be conjugated with toxin to produce antitumor responses. Azemar et al performed one of the first studies to show antitumor effects against HNSCC cell lines using bacterially derived toxins (eg, diphtheria, Pseudomonas); however, these therapies proved to be extremely hepatotoxic.^[11]

MAbs have also been developed to target EGFR and act by binding the receptor.^[12] Apart from blockade of EGFR signaling, EGFR antibodies may recruit Fc receptor–expressing immune effector cells; this leads to antibody-dependent cellular cytotoxicity and tumor lysis.^[13] However, high EGFR expression is not a predictor of tumor response to antitumor therapies.^[14] The most rigorously studied monoclonal antibody, cetuximab, has shown an enhanced ability to kill tumor cells in synergy with radiation and chemotherapy.

Small-molecule inhibitors have also been developed to inhibit the tyrosine kinase activity of EGFR. These molecules are typically adenosine triphosphate (ATP) analogs that compete with native ATP for binding. Because of the high intracellular concentration of ATP, higher concentrations of these inhibitors are required to block EGFR phosphorylation continuously in intact cells (in vivo) than to inhibit the purified EGFR tyrosine kinase in vitro.

Finally, nucleic acid–based molecules have been developed to interfere with translation of EGFR protein. These molecules include antisense oligodeoxynucleotides and small interfering mRNA. These molecules have shown promise increasing sensitivity to various chemotherapeutic agents in both in vitro and in vivo models.^[15] However, they are still in early stages of investigation.

Extracellular versus intracellular blockade

Members of the HER family are established therapeutic targets for the development of novel anticancer agents, and several approaches are being used to block these receptors. The blockade may be directed either to extracellular binding sites or to intracellular sites of EGFR. At present, predicting which of these 2 strategies will be more effective is difficult. A molecule that has dual action would be ideal.^[14]

The strong points of humanized EGFR MAbs (extracellular blockade) are as follows:

• Prolonged half-life
• Some cytolytic actions by immune mediated pathways
• Can induce receptor down-regulation
• No gastrointestinal toxicity

The strong points of low-molecular-weight EGFR tyrosine kinase inhibitors (intracellular blockade) are as follows:

• Long-term therapy with oral administration
• Can inhibit EGFR-homologous kinases such as HER-2
• Can directly inhibit HER-2
• Less potential for anaphylaxis or allergic reactions
• Can inhibit mutant EGFRvIII kinase found in some tumors

Preliminary results from early clinical trials of both anti-EGFR MAbs and EGFR small-molecule tyrosine kinase inhibitors are promising. The rapid evaluation of these target-specific noncytotoxics is limited by the lack of accurate information concerning the relevance of target expression and its modulation to this tumor type.

Early clinical trials are being designed to address these concerns. Current research goals include (1) defining the optimal dose and schedule in combinations with conventional chemotherapeutic agents and radiation therapy and (2) determining predictive factors that identify the best patient population in which to study and administer these agents. However, the clinical impact of EGFR inhibitors in patients with HNSCC must await the completion of randomized evaluations in combination with standard radiation and chemotherapeutic regimens.

Optimization of clinical response

Recent results highlight the notion that preclinical studies of targeted agents, particularly in combination with other agents, may not be good predictors of clinical response. Therefore, to optimize the use of these agents, alternative approaches are being explored.

One favored avenue of exploration is selecting responsive patients before therapy based on a predictive marker of response. A model for this approach involves trastuzumab; patients are selected for therapy based on the level of tumor HER-2 overexpression. This approach was obvious for trastuzumab because preclinical studies consistently showed that HER-2–overexpressing tumor cells, but not those with low HER-2 levels, were sensitive to trastuzumab-induced growth inhibition.

In contrast, preclinical and clinical studies have not found a strong correlation between HER-1/EGFR level and response to HER-1/EGFR-targeted therapies.^[16] This suggests that patient selection for trials with these agents should not be based solely on HER/EGFR expression. Numerous studies are in progress to identify markers that may predict for response to HER-1/EGFR inhibitors.

The extracellular strategy for blockade of the HER/epidermal growth factor receptor (EGFR) family uses antibodies to block the extracellular ligand-binding region of the receptor. Two anti–HER-2 monoclonal antibodies (MAbs), trastuzumab and pertuzumab 2C4, with different epitopes, have been developed.

Anti–HER-2: Trastuzumab

Trastuzumab (Herceptin; Genentech, Inc) which binds to and blocks HER-2, was one of the first targeted agents to be approved for cancer therapy. HER-2 is amplified in approximately 25-30% of human breast cancers, and this overexpression is associated with a poor prognosis.^[17] Trastuzumab acts against HER-2–overexpressing tumors, in part by inducing receptor endocytosis.^[18]

Clinical trials have shown that trastuzumab provides clinical benefits as significant as those monotherapy provides and improves the chance of survival when used in combination with chemotherapy as compared to chemotherapy alone in women with metastatic breast tumors that overexpress HER-2.^{[19, 20]} At present, trastuzumab is approved, alone or in combination with paclitaxel, for the treatment of metastatic breast cancer.

Anti–HER-2: Pertuzumab 2C4

Pertuzumab 2C4 is a recombinant humanized MAb that binds to extracellular domain II of the HER-2 receptor and inhibits its ability to dimerize with other HER receptors. Pertuzumab 2C4 represents a novel class known as HER dimerization inhibitors.

In a 2005 clinical study aimed at investigating the safety and pharmacokinetics of pertuzumab 2C4 and assessing HER dimerization inhibition as a therapeutic strategy, Agus and coworkers demonstrated that pertuzumab is well tolerated, has a pharmacokinetic profile which supports 3-week dosing, and is clinically active.^[21] Therefore, this study suggested that inhibition of dimerization may be an effective anticancer strategy.

Another study indicated that this unique mechanism of action could result in efficacy against tumors with low HER-2 expression that would not be targets for trastuzumab treatment.^[14]

Anti–HER-1/EGFR: Cetuximab C225

Several anti–HER-1/EGFR MAbs directed against the extracellular ligand-binding domain of the receptor are also in clinical development. Of these, cetuximab C225 (Erbitux, Imclone Systems Inc, New York) has been the most rigorously studied for targeted therapy of head and neck, lung, and colorectal cancer.

Phase I studies showed that cetuximab binds to HER-1/EGFR with an affinity comparable with that of the natural HER-1/EGFR ligands, epidermal growth factor (EGF) and transforming growth factor–alpha (TGF-α). This high-affinity binding of cetuximab to HER-1/EGFR prevents ligand binding and subsequent receptor activation. In preclinical studies, cetuximab has been shown to have significant inhibitory effects on the EGFR and the downstream pathways that it activates.

Several phase II clinical trials of cetuximab in combination with chemotherapy, radiotherapy, or both have been completed for a range of indications, including non–small cell lung carcinoma (NSCLC), head and neck squamous cell carcinoma (HNSCC), and colorectal cancer.^{[22, 23, 24, 25, 26, 27]} The US Food and Drug Administration (FDA) has recently approved this drug for clinical use.

Cetuximab has shown an enhanced ability to kill tumor cells in synergy with radiation. Studies have demonstrated increased locoregional control, improved progression-free survival, and overall survival when used in combination with radiotherapy.^[28] Experiments suggest that the enhanced antitumor activity observed when cetuximab is combined with radiation derives not only from inhibiting proliferation but also from inhibiting several important processes, including DNA repair after exposure to radiation and angiogenesis.^[29]

Cetuximab has also shown an enhanced ability to kill tumor cells in synergy with platinum-based chemotherapy. In a phase II trial, Herbst et al found better disease control rates and improved overall survival in both patients with stable disease and those patients with progressive disease following 2 cycles of cisplatinum-based chemotherapy.^[30]

In a study of 96 patients with refractory progressive head and neck cancer, Baselga et al found a response rate of 10% with a disease control rate of 53% and improved median time to progression and overall survival.^[31]

Data presented at the American Society of Clinical Oncologists in 2007 by Vermorken et al showed that patients with recurrent or metastatic head and neck cancer were randomized to receive cetuximab plus cisplatin (or carboplatin) and 5-fluorouracil (5-FU) or chemotherapy alone with promising initial results.^[32]

Kies et al conducted a phase II trial of induction chemotherapy using paclitaxel, carboplatin, and cetuximab in treatment-naive patients with head and neck cancer followed by local treatment with good response and acceptable toxicity.^[33]

The efficacy of cetuximab in combination with cisplatin has also been demonstrated in phase III clinical trials. Of interest, the degree of response to cetuximab was correlated with the development of acneiform rash.^[34]

The Erbitux in First-Line Treatment of Recurrent or Metastatic Head and Neck Cancer (EXTREME) study is a European multicenter phase III trial presented at the American Society of Clinical Oncologists in 2007. The results showed that adding cetuximab improves the impact of platinum-based chemotherapy. Patients treated with cetuximab survived a median of 10.1 months compared with 7.4 months for those patients who receive chemotherapy alone.

Clinical trials are currently under way to investigate the use of cetuximab as a single agent in patients refractory to platinum-based chemotherapy.

The most frequently reported cetuximab-related adverse events were asthenia, fever, and nausea (flulike symptoms), as well as elevated transaminases and allergic reactions. Acneiform rash and allergic reactions were clinically relevant. Eighty percent of patients developed rashes in the form of sterile folliculitis that usually affected the face, upper chest, and/or back. These rashes were usually mild to moderate in severity and resolved without treatment.

Significant infusion reactions were uncommon but can be severe. The allergic reactions and anaphylactic reactions were rare and usually appeared within minutes of starting the initial infusion; were responsive to standard treatment; and could be prevented with the prophylactic administration of antihistamines, a prolonged infusion duration, or both.^[35]

A second antireceptor approach is based on the observation of mutations in the ATP-bikinase function, which suggest that the receptor’s tyrosine kinase is critical for epidermal growth factor (EGFR)-mediated tumor progression. HER-targeted agents acting at an intracellular level are low-molecular-weight tyrosine kinase inhibitors.

Tyrosine kinase inhibitors

Inhibitors of the EGFR tyrosine kinase have the theoretical advantage of also blocking activating cytoplasmic signals when compared to agents that block activation at an extracellular level (ie, receptor monoclonal antibodies [MAbs]). Several small-molecule tyrosine kinase inhibitors of the EGFR tyrosine kinase are currently being studied. Of these, the US Food and Drug Administration (FDA) has already approved the oral quinazolines erlotinib (OSI-774, Tarceva; Genentech, South San Francisco, CA) and gefitinib (ZD 1839, Iressa; AstraZeneca, Wilmington, DE) for clinical use.

These agents inhibit the purified EGFR tyrosine kinase enzyme in vitro with an inhibitory concentration of 50% (IC₅₀) in the low nanomolar range.^[14] They competitively inhibit binding of adenosine triphosphate (ATP) to the intracellular domain of the EGFR, causing inhibition of ligand-induced cell growth by hindering autophosphorylation, leading to dose-dependent tumor stasis and even tumor regression. Erlotinib was initiated in the setting of recurrent and/or metastatic head and neck cancer.^{[35, 36]}

The FDA has recently added additional warnings to the prescribing information for erlotinib (Tarceva). The following have been reported: gastrointestinal perforation (including fatalities); bullous, blistering, and exfoliative skin conditions, including cases suggestive of Stevens-Johnson syndrome/toxic epidermal necrolysis, which have been fatal in some cases; and ocular disorders, including corneal perforation or ulceration.

Interestingly, limited preclinical data suggest that erlotinib, but not gefitinib, can inhibit EGFR.^{[37, 38]} However, these preliminary data require confirmation. Whether this difference between the agents, if true, is caused by differences in structure or potency or other, as yet unknown, cell type–dependent factors is still undetermined. In addition, these agents are administered orally, which makes them suitable for long-term therapy.

Extensive preclinical studies with erlotinib and gefitinib show that both agents effectively inhibit tumor cell growth when used alone and in combination with various chemotherapeutic agents, and both are well tolerated.^{[39, 40, 41, 42, 43, 44]} Adverse effects of both agents include acneiform rash (72%), diarrhea, and fatigue.

In a phase I dose-escalation study of erlotinib combined with docetaxel and radiation in locally advanced head and neck cancer, Savvides et al found no significant pharmacokinetic interactions between erlotinib and docetaxel and concluded that this regimen is feasible and active.^[45] Kim et al studied the therapeutic effects of cisplatin, docetaxel, and erlotinib in advanced head and neck cancer and found an overall response rate of 66%.^[46]

A phase I/II trial by Doss that used induction chemotherapy plus gefitinib followed by concurrent chemotherapy/radiation/gefitinib for locally advanced head and neck cancer had an overall recurrence rate of 85%, with improved progression-free survival (68%) and improved overall survival at 1 year (86%).^[47]

In addition, combined therapy along with MAbs directed against EGFR to maximize inhibition of the EGFR pathway has demonstrated success in preclinical studies.^[48] The additive effects of tyrosine kinase inhibitors and cetuximab were studied by Jimeno et al, who concluded that cetuximab interferes with erlotinib-induced EGFR up-regulation, resulting in antitumor effects.^[49]

Another tyrosine kinase inhibitor, lapatinib (GW572016), is a selective and potent dual competitive inhibitor of EGFR and HER-2. Harrington et al conducted a phase I study of lapatinib (dose escalation) plus chemoradiation (using cisplatin) in patients with locally advanced head and neck cancer with preliminary evidence of clinical activity.^[50] Abidoye et al found no significant clinical response in their phase II study evaluating lapatinib in patients with recurrent/metastatic head and neck cancer.^[51]

Preclinical and early clinical data from trials with erlotinib and other targeted agents show that these inhibitors are well tolerated and could benefit patients with a variety of cancers. When data with other agents become available, their clinical application will become clearer, and new questions and challenges will emerge.

Further understanding of the HER family signaling pathways and their interactions with other networks within tumor cells is necessary to optimize the clinical development of these targeted agents. Finally, a predictive marker that may help select patients for treatment with HER-1/EGFR inhibitors is sorely needed to maximize the information derived from ongoing studies.

Tumor suppressor genes regulate the cell cycle and prevent cells from undergoing uncontrolled growth and division. When tumor suppressor genes are inactivated or mutated, the normal restraints on cellular proliferation are removed, and uncontrolled cell growth is allowed. This is one step in the process of carcinogenesis.

TP53 is the most commonly mutated tumor suppressor gene in human cancers. Mutations in TP53 are found in approximately 50% of head and neck squamous cell carcinomas (HNSCCs). The TP53 gene arrests the cell cycle by allowing for DNA repair or, in cases in which the DNA damage is too great for repair, by inducing programmed cell death in a process named apoptosis. By either repairing gene mutations or inducing apoptosis in heavily damaged cells, the TP53 gene acts as a guardian of the genome and prevents the accumulation of genetic damage.

Gene therapy techniques are being used to restore TP53 function lost in HNSCC. Using gene therapy, tumor regression has been observed with the increase of apoptosis of cancer cells in patients with HNSCCs. The adenovirus ONYX-015 specifically attacks TP53 -null cells by deleting the E1B gene. This adenovirus can only replicate in TP53 -null cells; thus, eventual lysis and selective killing of tumor cells occurs.

In 2002, Portella et al evaluated a therapeutic approach to anaplastic thyroid carcinoma that was based on the ONYX-015 virus.^[52] They reported that ONYX-015 induced cell death in 3 anaplastic thyroid carcinoma cell lines. The ONYX-015 virus worked synergistically with 2 antineoplastic drugs (doxorubicin and paclitaxel).

The INK4 family of genes (p15, p16, p18, p19) regulates the presynthetic gap (G₁) phase (phase of cells prior to DNA synthesis) of the cell cycle. p16 binds to and inhibits CDK4, which inhibits the cell in the G1 phase. p16 activity is lost in approximately 80% of HNSCCs. Gene therapy targeted at the restoration of p16 has been shown to inhibit HNSCC cell lines in animal models.

Cyclin D1 is an important tumor suppressor gene product involved in regulating the cell cycle. Overexpression of cyclin D1 has been demonstrated in 64% of HNSCCs and has been shown to be an independent prognostic indicator of recurrence. Transfection of antisense cyclin D1 DNA into tumor cells has been shown to decrease HNSCC tumor growth in animal models.

Several other potential molecular targets are involved in the inhibition of growth inhibitory signals in HNSCC. The retinoblastoma gene (RB) prevents progression through the cell cycle by inactivating transcription factor E2F. Although RB is a well-studied tumor suppressor gene, decreased Rb protein expression is only found in 20% of HNSCCs. The p21 gene family is also an important tumor suppressor gene family involved in regulating the cell cycle. Low expression of these genes has been correlated with poorer survival rates in patients with HNSCC.

Apoptosis

Programmed cell death, apoptosis, is an important regulatory step in cell differentiation and mutation regulation. In tumorigenesis, cancer cells must resist apoptosis and continue cell division. Extracellular and intracellular initiators can initiate apoptosis. Each pathway activates a cascade of caspases, which are enzymes that cause cell destruction.

Extracellular initiators (Fas, tumor necrosis factor [TNF] receptor, transgenic adenocarcinoma of the mouse prostate [TRAMP], TNF-related apoptosis-inducing ligand receptor 1 [TRAIL-R1], TNF-related apoptosis-inducing ligand receptor 2 [TRAIL-R2], death receptor 6 [DR6]) are activated by ligands that bind these receptors on the cell surface.

Mutations in these pathways can lead to the evasion of apoptosis and are potential targets for molecular therapy. The overexpression of transcription factor nuclear factor kappaB (NF-kB) has been shown to decrease TNF-mediated apoptosis in HNSCC. NF-kB inhibitors can reverse this effect.

Intracellular pathways initiate apoptosis when genetic mutations are detected within a cell. TP53 is instrumental in this pathway. Mutations in TP53 can lead to the evasion of apoptosis and tumorigenesis. The presence of mutated TP53 has an inverse correlation with the amount of apoptosis in head and neck squamous cell carcinoma (HNSCC). Gene therapy techniques are being used to restore TP53 function lost in HNSCC. Tumor regression has been observed by increasing apoptosis of cancer cells in HNSCC patients.

Immortalization

Normal cells can only replicate a finite number of times. Telomeric DNA, which is found at the end of chromosomes, regulates this process. In the normal cell, a small portion of telomeric DNA is lost with each replicative cycle. Once enough telomeric DNA is lost, the chromosomes become unstable, leading to cell death. An important step in tumorigenesis is maintaining telomere length. One mechanism for maintaining telomere length is the enzyme telomerase, which prevents the loss of the telomeric DNA. Telomerase activity is found in 90-100% of HNSCCs. Telomerase is also a potential target for molecular therapy in the future.

In order for a tumor to grow, invade, and metastasize, angiogenesis, which is the formation of new blood vessels, is critical for tumor cells to acquire the necessary nutrients. A number of factors regulate angiogenesis. Vascular endothelial cell growth factor (VEGF) has potent angiogenic effects. The presence of VEGF has been reported in approximately 40% of head and neck squamous cell carcinomas (HNSCCs), and its presence is associated with a poor prognosis.

In vitro studies using antisense VEGF mRNA have shown a down-regulation of VEGF and decreased endothelial migration. A tumor vaccine that targets VEGF has also been shown to suppress angiogenesis and tumor growth in animal studies. Recently released data from a phase III study of bevacizumab (Avastin), a humanized monoclonal antibody (MAb) that inhibits VEGF, showed prolongation of life in patients with metastatic colon cancer when bevacizumab was used in combination with chemotherapy as compared to chemotherapy alone.

Vokes et al, in a study of erlotinib with bevacizumab, showed stable disease in 31 of 44 patients with improved median progression-free survival and overall survival.^[53]

Seiwert et al showed no major synergistic toxic effect in a phase I study of bevacizumab and 5-fluorouracil (5-FU) and hydroxyurea with concomitant radiotherapy for poor-prognosis head and neck cancer.^[54]

Of interest, resistance to epidermal growth factor receptor (EGFR) inhibitors has been demonstrated to be secondary to increased VEGF levels.^[55] Investigation is under way to evaluate the efficacy of dual EGFR-VEGF inhibitors, with preclinical trials using head and neck xenografts demonstrating excellent responses.^[56]

Basic fibroblast growth factor (bFGF), platelet-derived endothelial cell growth factor (PD-ECGF), and interleukin-8 (IL-8) are also potent angiogenic factors. High levels of bFGF and the overexpression of PD-ECGF have been identified in HNSCC cell lines. IL-8 receptors have also been found in HNSCC tumors. The up-regulation of these angiogenic factors in HNSCC makes them ideal targets for molecular therapy.

As tumorigenesis occurs, cancer cells develop the ability for tumor invasion and metastasis. This process consists of 3 steps:

• Attachment of tumor cells to the basement membrane
• Proteolysis of the extracellular matrix
• Migration of tumor cells

Epithelial basement membranes are composed of collagen type IV, laminin, collagen type VII, and heparin sulfate proteoglycans. Integrins, E-cadherin, and catenins mediate the adherence of head and neck squamous cell carcinoma (HNSCC) tumor cells to the basement membrane. These adherence proteins may also provide potential molecular therapy targets in the future.

Proteolysis of the extracellular matrix is a critical step in tumor invasion. Matrix metalloproteinases (MMPs) are a diverse group of proteinases that work by degrading the extracellular matrix. MMPs are up-regulated in 50% of HNSCC cell lines. The development of MMP inhibitors that prevent tumor invasion is also a potential goal for molecular therapy.

Urokinase-type plasminogen activator (uPA) and its receptor (uPAR) are up-regulated in HNSCC and are believed to promote tumor invasion and metastasis. The use of anti-uPA antibodies has been shown to prevent tumor invasion in HNSCC cell lines. Likewise, blocking uPAR with molecular inhibitors has been shown to prevent tumor invasion in HNSCC cell lines.

Epithelial cell adhesion molecule (EpCAM) is a transmembrane glycoprotein that behaves unlike standard cell adhesion molecules. In fact, studies demonstrate that EpCAM negatively affects normal cadherin-mediated cell adhesion, loosening cell-cell connections and promoting cell migration, a requirement for metastasis.^{[57, 58]}

EpCAM has previously been demonstrated to be overexpressed in various non–head and neck carcinomas, including colorectal, prostate, liver, lung, breast, pancreas, and esophagus. More recently, it has been shown to be upregulated in HNSCC. It has also been implicated in modulation of proliferation and differentiation.

Various targeted therapies have been attempted to inhibit EpCAM. Monoclonal antibodies such as edrecolomab have had limited success.^[59] VB4-845 (Proxinium; Viventia Biotech, Inc) is currently being evaluated in 2 phase II clinical trials. This drug is a recombinant fusion protein produced by E coli, expressing humanized single-chain antibody fragment specific for EpCAM and linked to a truncated Pseudomonas exotoxin A.

Lastly, other strategies such as use of RNA interference techniques to inhibit expression of EpCAM show promise in pre-clinical trials.^[60]

Interleukin-13 receptor

In 2002 and 2003, Kawakami et al demonstrated that although several head and neck cancer cell lines overexpress interleukin-13 receptor (IL-13R), most cell lines express only low levels of IL-13R.^{[61, 62, 63]}

The investigators found that the primary IL-13–binding protein IL-13Rα2 chain plays an important role in ligand binding and internalization. They showed that the gene transfer of IL-13Ra2 chain into various solid-tumor cell lines that express few IL-13Rs could dramatically sensitize cells to the cytotoxic effect of a recombinant chimeric protein composed of interleukin-13 and a mutated form of Pseudomonas exotoxin A, IL13-PE38QQR.

By plasmid-mediated stable gene transfer, not only IL-13Rα2 chain–positive head and neck cancer cell lines, but also IL-13Rα2 chain–negative cell lines, can dramatically increase sensitivity to IL-13 toxin. The results reported by Kawakami et al demonstrated that by using a combination approach of gene transfer and systemic or locoregional cytotoxin therapy, the IL-13R represents a new potent target for head and neck cancer therapy.^{[62, 63]}

Cyclooxygenase-2

Cyclooxygenase-2 (COX-2) is overexpressed in several premalignant and malignant mucosal conditions of the head and neck. Increased levels of COX-2 may contribute to carcinogenesis by modulating xenobiotic metabolism, apoptosis, immune surveillance, and angiogenesis.

Newly developed, selective COX-2 inhibitors suppress the formation of tumors in experimental models. This may be a rationale for chemoprevention trials that are already under way.^[64] Limburg et al demonstrated the chemopreventive effect of celecoxib in oral precancers and cancers in animal models.^[65]

In animal studies, these agents also modulate the anticancer activity of radiotherapy and chemotherapy. Selective COX-2 inhibitors suppress the growth and metastases of established tumors.

Prellop et al conducted a phase IB/II trial to evaluate the toxicity and efficacy of celecoxib administered concurrently with chemoradiotherapy (cisplatin, paclitaxel) for locally advanced or recurrent head and neck cancer.^[66] The study was temporarily suspended in December 2004 because of the cardiotoxic effects of COX-2 inhibitors but was restarted with a modified schedule in May 2006 because of the initially promising results.

Carcinoembryonic antigen

Human carcinoembryonic antigen (CEA) is an oncofetal glycoprotein overexpressed in many gastrointestinal carcinomas. Expression of CEA in head and neck cancer is not widely recognized. Immunohistochemical analysis of tumor tissue from 69 cases of head and neck squamous cell carcinoma (HNSCC) using a CEA-specific monoclonal antibody (MAb) showed most to be positive for CEA. These results suggest that CEA may be considered as a possible target for specific vaccine-mediated immunotherapy against HNSCC.^[67]

tgDCC-E1A

Targeted genetics research has confirmed the results of preclinical studies that suggested that tgDCC-E1A has multiple cancer-fighting effects. One of these properties is the ability to down-regulate the oncogene for HER-2/neu. Importantly, overexpression of HER-2/neu occurs in a significant number of cancers. The level of HER-2/neu expression correlates with poor prognosis, increased metastasis, and resistance to chemotherapeutic agents.

By reducing or inhibiting the expression of oncogenes such as that for HER-2/neu, tgDCC-E1A may inhibit the growth of tumors and help prevent metastasis. Studies also show that tgDCC-E1A can make cancer cells more sensitive to chemotherapy and radiation, and that the gene may induce immune system cells to attack cancer cells.

Phase II studies of tgDCC-E1A in patients with head and neck cancer have been conducted. In the phase II head and neck cancer study, a 7% complete response rate was observed, and nearly 50% of the patients had stable disease. A phase II study of tgDCC-E1A in combination with radiation therapy in patients with head and neck cancer has been initiated.