Chapter 19: Head and Neck Cancer

Head and neck cancers are a diverse group of diseases, each with distinct epidemiologic, anatomic, and pathologic features. The natural history and treatment considerations may vary widely. In this chapter, our focus is on the primary management of squamous cell carcinomas (SCCs) of the head and neck (HNSCC). In recent years, we have observed significant advances in diagnosis and treatment and recognition of the human papillomavirus (HPV) as a significant causative agent and prognostic factor for cancers of the oropharynx. Tumor imaging is increasingly precise. Primary therapy eradicates disease in a majority of patients with early-stage HNSCC, and the long-term management of these patients currently involves an emphasis on general medical care, avoiding known carcinogens such as alcohol and tobacco, and participation in chemoprevention strategies to reduce the risk of second primary tumors. Therapy for patients with locally advanced disease is multimodal, and success has been achieved in improving local tumor control, disease remission, organ preservation, and overall survival. The integration of chemotherapy and novel “targeting” systemic treatment approaches with surgery and/or radiotherapy is under study and discussed.

In the United States, HNSCC is estimated to represent approximately 3% (46,000) of new cancer cases and 2% (9,000) of cancer deaths in 2015 (¹). However, the disease is more common in many developing countries, with a worldwide annual incidence of more than 500,000 (²).

The risk of developing head and neck cancer increases with age; most patients are older than age 50. There has been a clearly demonstrated association with tobacco and alcohol use. Molecular studies provide evidence that carcinogens found in these substances have a causal role. The prevalence and spectrum of p53 mutations are prominent in cancers of patients with a history of tobacco and alcohol use (³). Cancers of the oral cavity, larynx, and hypopharynx are uncommon in persons with no smoking history.

Human papillomavirus infection is now widely accepted as another etiologic factor for HNSCC. In the United States, more than 50% of cancers arising in the oropharynx, particularly in the palatine tonsils and tongue base, harbor oncogenic HPV (⁴). The incidence of oropharyngeal cancer in the United States is increasing, primarily due to HPV-associated cases in men (⁵). It appears that the HPV-positive oropharyngeal malignancy represents a distinct clinical and pathologic subgroup of HNSCC, with poorly differentiated basaloid histopathology (⁶) and marked tumor responsiveness to radiation and chemotherapy. Moreover, HNSCCs with transcriptionally active HPV-16 DNA are characterized by occasional chromosomal loss, whereas those lacking HPV DNA typically have gross deletions, involving chromosomal arms known to be abnormal in HNSCC (⁷). Thus, HPV-16 infection may be an early carcinogenic event. Patients with HPV DNA–positive tumors, particularly those associated with E6 and E7 proteins, have improved survival after chemoradiotherapy when compared to patients with HPV-negative tumors (^8,9). A recent case-control study reported that HPV-16–positive HNSCCs were independently associated with several measures of sexual behavior and exposure to marijuana but not with cumulative measures of tobacco smoking, alcohol use, or poor oral hygiene (¹⁰). These findings indicate that HPV-16–positive HNSCC and HPV-16–negative HNSCC have different risk factor profiles. In addition to detection of oncogenic HPV DNA by in situ hybridization, tumor overexpression of p16 by immunohistochemistry serves as a surrogate for HPV association in oropharyngeal cancer with prognostic implications, particularly when coupled with patient smoking status (⁸). In addition, HNSCC patients with heavy tobacco and alcohol exposures are at high risk of developing multiple cancers, with “field cancerization” throughout the upper aerodigestive tract and bladder. The observation that treated head and neck cancer patients may have a high risk (estimated to be 3%-4% per year) of metachronous tumors has driven chemoprevention trials designed to reduce the risk of second primary tumors.

The progression of HNSCC is thought to involve multiple stepwise alterations of molecular pathways in the squamous epithelium (¹¹). Aberrations in the p53 and Rb tumor suppressor pathways are the most common molecular events, resulting in uncontrolled cell proliferation. Approximately 50% to 80% of HPV-negative tumor samples harbor a p53 mutation (^12,13). The Rb pathway can be disrupted either through inactivation of the CDKN2A gene, which encodes p16, an inhibitor of cyclin-dependent kinase, or amplification or the CCND1 gene encoding cyclin D1 (^14,15,16).

The epidermal growth factor receptor (EGFR) is overexpressed in invasive HNSCC in a majority of sample tumors (¹⁷). Binding to EGFR by its natural ligands, mainly epidermal growth factor or transforming growth factor α (TGF-α), prompts a conformational change in the receptor through dimerization, which results in subsequent autoactivation of the tyrosine kinase from the intracellular domain of the receptor. This process activates an intracellular signaling pathway, leading to the inhibition of apoptosis, activation of cell proliferation and angiogenesis, and an increase in metastatic spread potential (¹⁸).

In 2011, Stransky et al performed whole exome sequencing on 74 HNSCC tumor specimens (¹⁹). Not surprisingly, the mutational landscape of the disease is quite complex, and smoking-related HNSCC specimens had a mutation rate approximately twice that of HPV-positive HNSCC. Mutations in genes previously implicated in HNSCC were detected at a high rate, including TP53, CDKN2A, PTEN, PIK3CA, and HRAS, validating this approach. However, this unbiased approach also identified a high rate of mutations in numerous genes not previously linked to HNSCC. In approximately 30% of cases, mutations were identified in genes that regulate squamous differentiation, such as NOTCH1, IRF6, and TP63. Mutations in genes regulating apoptosis were also frequent events. These novel findings have provided the rationale for testing novel therapeutic targets such as NOTCH inhibitors.

There are numerous challenges to the development of molecularly targeted therapies in HNSCC, including frequent tumor suppressor loss and difficulty in identification of true driver mutations within a complex mutational landscape. However, an improved understanding of the molecular biology of this disease should facilitate discovery of new prognostic markers and therapeutic targets.

Optimal therapy and treatment outcomes depend on the precise identification of the primary tumor (Fig. 19-1) as well as the local, regional, and distant extent of disease.

Patients with early-stage disease may present with vague symptoms and minimal physical findings, which is why a high index of suspicion for early diagnosis is needed, especially for tobacco users. A majority of patients will present with signs and symptoms of locally advanced disease, which vary according to the subsite in the head and neck. Sinusitis, unilateral nasal airway obstruction, and epistaxis may be early signs of cancers of the nasal cavity or paranasal sinuses. Otitis media that is recurrent or is refractory to antibiotics is an indication for a complete ear, nose, and throat evaluation to rule out a nasopharyngeal neoplasm. Chronic otalgia, dysphagia, odynophagia, and throat soreness lasting weeks may be the presenting symptoms of oropharyngeal or hypopharyngeal cancer. However, many patients with HPV-associated oropharyngeal cancer present with an otherwise asymptomatic neck mass and may have limited or no smoking history. Persistent hoarseness demands visualization of the larynx. Supraglottic laryngeal neoplasms do not usually present early, and a neck mass may be the presenting sign. Careful examination of lymph nodes in the facial, cervical, and supraclavicular regions is important because the anatomic patterns of lymphatic drainage may reflect the specific subsite of a head and neck primary tumor (Fig. 19-2) (²⁰). Level 2/3 adenopathy, for example, suggests a primary cancer of the oral tongue or oropharynx, and posterior cervical adenopathy is frequently a result of regional spread of a nasopharyngeal tumor.

Physical examination should include careful inspection of the skin and oral/oropharyngeal mucosal surfaces; palpation of the tongue, floor of the mouth, and oropharynx; and systematic palpation of the neck. A complete examination also requires an indirect mirror examination of the oropharynx, hypopharynx, and larynx, complemented by fiberoptic endoscopy (²¹). Leukoplakia (white mucosal patches that cannot be removed by scraping) and higher risk erythroplakia (red or mixed red-white patches) are the most common premalignant lesions in the head and neck. Any suspicious surface in the oral mucosa should undergo biopsy.

Three-dimensional imaging with computed tomography (CT), magnetic resonance imaging (MRI), and/or ultrasonography is also needed to evaluate the extent of disease and to complete staging. Magnetic resonance imaging is the preferred local imaging modality for nasopharyngeal cancer. Because the lungs are the most common sites of distant metastases, a chest x-ray should be performed as well. Computed tomography scanning of the chest should be performed for symptomatic or high-risk patients. This would include patients with nasopharyngeal cancer or those with primary tumors of other sites presenting with N2b or greater nodal disease and low neck or supraclavicular metastases. Circulating tumor markers that would be reliable in early detection of HNSCC have not yet been identified.

Patients who present with a suspicious neck mass and no obvious primary mucosal lesion should undergo a systematic examination of the head and neck. Head and neck imaging may be helpful. If no obvious primary site is found, fine-needle aspiration of the mass may establish a diagnosis of cancer. Detection of HPV or Epstein-Barr virus DNA in a lymph node suggests a tumor of oropharyngeal or nasopharyngeal origin, respectively. If metastatic SCC is demonstrated, examination under anesthesia is often performed. Suspicious lesions are biopsied, and consideration is given to tonsillectomy and blind biopsies of the nasopharynx, base of the tongue, and hypopharynx, depending on the pattern of lymphadenopathy. Open biopsy of the neck mass may be performed if fine-needle aspiration biopsy and panendoscopy have failed to yield a diagnosis or in patients suspected of having an alternative process (eg, lymphoma). An experienced head and neck surgeon may be prepared to proceed with selective neck dissection if SCC of unknown head and neck primary origin is determined.

Staging criteria for head and neck cancers are based on the American Joint Committee on Cancer TNM staging system, which classifies tumors according to anatomic site and extent of disease (²²). Head and neck primary tumor (T) staging is complex, varying with the primary subsite in the head and neck region. Classifications for lymph node (N) and distant metastases (M) are uniform for sites (Table 19-1) other than nasopharynx (^22a).

Two-thirds of patients with HNSCC will present with stage III or IV disease. For patients with T1/2 disease (stage I/II), surgery or radiotherapy as a single modality is most often applicable and effective. Depending on the precise primary site and stage, a curative outcome will be achieved in 70% to 95% of cases. In patients with intermediate and locally advanced disease at diagnosis, combined treatment strategies have become the standard of care. Multimodal treatment plans are designed to balance competing goals of tumor eradication and organ preservation (^11,23,24). Despite optimal local therapy, 30% to 50% of patients may develop local or regional recurrence, and nearly 20% to 30% are at risk for distant metastases depending on the primary site and staging.

Nasopharynx

Over 95% of endemic nasopharyngeal carcinomas (NPCs) are associated with EBV. Nasopharyngeal carcinoma tends to occur in younger persons and is not associated with tobacco use. Nasopharyngeal carcinoma is an aggressive neoplasm with cervical lymph node metastases present in 60% to 90% of patients at diagnosis. Because of unique anatomic, biologic, and clinical characteristics, therapy for NPC is distinctive. Radiotherapy is the mainstay of local treatment. The anatomy of the nasopharynx and tumor sensitivity to radiotherapy limit the role of surgery to obtaining the initial biopsy and, for selected patients, resection of residual lymphadenopathy after radiotherapy. Nasopharyngeal carcinomas are highly sensitive to chemoradiotherapy. Because of the proximity of the nasopharynx to normal critical structures of the central nervous system and given the propensity of NPC for skull base invasion, intensity-modulated radiation therapy (IMRT) is the usual radiation therapy technique for NPC, which improves tumor coverage, reduces xerostomia, and improves patient quality of life compared to traditional techniques (^25,26). Locoregional tumor control following chemoradiotherapy approaches 90%.

For stages III and IV, concomitant chemotherapy and radiotherapy followed by adjuvant chemotherapy is the accepted standard of care based on the Head and Neck Intergroup NPC 0099 trial (²⁷). Compared to radiation alone, chemoradiation with cisplatin (100 mg/m² on days 1, 22, and 43), followed by adjuvant chemotherapy with cisplatin and 5-fluorouracil (5-FU) demonstrated significant improvement in 3-year survival (78% vs 47%). Chan et al (²⁸) have also demonstrated the efficacy of concomitant radiation and weekly cisplatin 40 mg/m². Phase III studies investigating the value of induction chemotherapy are under way. The NRG-HN001 study is an ongoing phase II/III study investigating the value of measuring plasma EBV DNA as a marker of the efficacy of concomitant chemoradiotherapy. With undetectable DNA, patients are randomized to observation or adjuvant chemotherapy. Patients with detectable DNA after chemoradiotherapy receive additional treatment, testing an alternative regimen consisting of paclitaxel and gemcitabine versus cisplatin and 5-FU.

Oral Cavity

The majority of oral cavity neoplasms occur in the anterior two-thirds of the tongue (oral tongue) and the floor of the mouth. Surgical resection, often with postoperative radiotherapy, is the most common and effective local treatment approach (^11,24). Depending on site and tumor volume, early cancers should be resected but may be treated with radiotherapy. Local tumor control rates of patients with stage I and II tumors are 80% to 90% and 50% to 80%, respectively (²⁹). For deeply invasive T1/2 disease, we favor surgical resection and neck dissection with postoperative concomitant chemoradiotherapy for selected patients with narrow margins or nodal metastases, particularly if there is extracapsular spread. Perineural invasion is also a significant negative prognostic sign. Forty percent of patients present with clinically evident lymph nodes, and bilateral nodal involvement is not uncommon. Although primary surgical approaches are preferred at our center, interstitial radiotherapy (brachytherapy) has been used in combination with external-beam therapy for selected cases to achieve higher control rates than external radiation alone.

For patients with locally more advanced disease (Fig. 19-3 shows an example of a patient with a retromolar trigone primary tumor invading bone, T4), surgery followed by radiation therapy (or chemoradiotherapy) is the most widely accepted approach. At the University of Texas MD Anderson Cancer Center (MDACC), selective neck dissections are routinely performed for patients with stages II to IVa disease.

Oropharynx

The most common cancers of the oropharynx are of the base of tongue and tonsils, and an increasing percentage of these are HPV associated, with an improved prognosis. In an unplanned, post hoc analysis of RTOG 0129, Ang et al classified oropharyngeal cancer patients treated with concurrent chemoradiotherapy in to three risk-of-death groups using recursive partition analysis. The “low-risk” group were those with p16 positive tumors and minimal smoking history (94% 3-year overall survival). Conversely, the “high-risk” group were those characterized mostly by p16 negative tumors and greater smoking intensity (42% 3-year overall survival) (⁸). Radiation therapy serves as the principal treatment modality for the majority of oropharyngeal malignancies and is used as a single modality for T1 and many T2 tumors. Local control is obtained in over 90% of patients (³⁰). Regional lymph nodes are treated in all cases, and unilateral neck radiation is considered for well-lateralized earlier-stage tonsillar primaries, which reduces greatly the radiation dose to the contralateral parotid gland and key swallowing structures. Given recent technical advances and the popularity of minimally invasive transoral approaches, surgical resection for oropharyngeal cancer is being performed more frequently at many centers, and this approach is currently under study (Eastern Cooperative Oncology Group [ECOG] 3311). However, adjuvant radiation or chemoradiation may also be required, depending on the surgical pathology findings.

Concomitant chemoradiotherapy using IMRT is the current standard of care for patients with locally advanced disease. Under study, protons have unique physical properties compared with x-rays or photons due to the Bragg peak, where most of the proton dose is delivered at a finite depth, thus reducing dose to certain nontarget structures. Investigators at MDACC are currently conducting a phase II/III randomized trial of concomitant chemotherapy with intensity-modulated proton therapy vs IMRT for stage III/IV oropharyngeal cancer. Objectives are to compare tumor control and long-term toxicity.

Hypopharynx

With 75% of lesions occurring in the pyriform sinus, carcinoma of the hypopharynx is relatively uncommon but virulent (Fig. 19-4). Small-volume disease may be treated with surgery or radiation, but later-stage disease requires multimodal therapy. At presentation, more than 75% of patients have advanced disease (T3 or T4). The overall 5-year survival rate is lower than 30%. For many patients, surgical treatment also requires removal of the larynx. The European Organization for Research and Treatment of Cancer (EORTC), in a phase III trial, demonstrated that laryngeal preservation with sequential chemoradiotherapy is a feasible alternative to radical surgery for many patients with locally advanced disease (³¹). In a more recent trial of patients with resectable advanced SCC of the larynx or hypopharynx, Lefebvre et al (³²) compared sequential treatment with two cycles of cisplatin and 5-FU followed by radiotherapy with an arm of four cycles of cisplatin and 5-FU administered during weeks 1, 4, 7, and 10, alternating with radiotherapy. Survival with a functional larynx was similar in both arms, as was overall survival (median, 4.4 vs 5.1 years, respectively). Please see the “Organ Preservation” section for further discussion.

Larynx

Given the critical role of the larynx in communication, swallowing, respiration, and airway protection, organ preservation to maintain functional status and quality of life has been the focus of laryngeal cancer treatment since the 1970s. The most widely used treatment of T1 and T2 cancers of the larynx is radiotherapy, which has demonstrated control rates greater than 90% for T1 disease and approximately 70% to 80% for T2 tumors (³³). For carefully selected patients with intermediate-stage disease, sequential chemotherapy followed by radiation and surgical salvage, if needed, showed equivalent survival outcomes compared with surgery in the Veterans Affairs (VA) laryngeal study (³⁴). Although the rate of local failure is higher with organ preservation approaches, salvage surgery is effective, and this approach allows 60% of patients to preserve organ function (³⁵). Larynx cancer treatment strategies are discussed in detail in the “Organ Preservation” section.

Salivary Gland Cancers

Tumors of the salivary glands are uncommon, with approximately 5,000 cases per year in the United States. Histologies are diverse, and risk factors are poorly defined, although radiotherapy may be causative. The age range of patients affected is broad. Many salivary neoplasms are benign, often involving the parotid gland, accounting for approximately 80% of parotid tumors, 50% of tumors arising in submandibular glands, and 25% of tumors arising in minor salivary glands.

Table 19-2 lists primary salivary malignancies.

Primary treatment depends on tumor extent and histology. Notably, parotid lymphadenopathy may reflect metastatic involvement by squamous cancers of the scalp or melanomas, and this must be borne in mind when evaluating these patients. Following a complete head and neck evaluation, consideration may be given to CT imaging of chest and a bone scan because these are common metastatic sites.

Surgical resection is the fundamental primary treatment for most patients, and the approach will be influenced by the primary histology (³⁶). Adenoid cystic carcinoma (ACC) tends to track along nerves and may involve structures of the skull base, an important consideration in surgical and radiation therapy planning. C-kit is overexpressed in ACC (³⁷). Lymph node metastases are uncommon. Low-grade mucoepidermoid carcinomas tend to be localized and are most often treated by surgery alone. High-grade mucoepidermoid carcinomas carry a much higher risk of lymph node and distant metastases. Salivary ductal carcinomas may be high grade and share biomarker characteristics, such as estrogen or progesterone receptor and HER2/neu overexpression, with breast cancer.

As a generalization, large tumors or those with close surgical margins will require postoperative radiotherapy. Postoperative concomitant chemoradiation is now under study (Radiation Therapy Oncology Group [RTOG] 1008) in a randomized trial of high-risk patients and is a consideration for patients with good performance status with locally advanced resectable disease.

For the palliative treatment of patients with recurrent disease not amenable to further local treatment or those with distant metastases, treatment with systemic chemotherapy, most often with a platinum-based combination, is an option (³⁸). Cisplatin, 5-FU, cyclophosphamide, and doxorubicin are active compounds. The taxanes also have activity, although not demonstrated in patients with ACC. Combinations may be more effective, with response rates ranging from 20% to 30%. Salivary ductal cancers are much more sensitive to chemotherapy than ACC.

Treatment goals in the setting of distant metastatic disease are palliative because there has not been an overall survival advantage with chemotherapy. See Table 19-3 for a listing of tumor markers in salivary cancer that have prompted clinical trials. EGFR, KIT, HER2, and AR are prospective targets for systematic study. Fibroblast growth factor receptor (FGFR) expression (³⁹) and MYB-NFIB fusion oncogene (⁴⁰) have been identified in subsets of ACC. In a phase II trial, lapatinib, a dual inhibitor of EGFR and erbB2 tyrosine kinase activity, showed biologic activity in ACC (⁴¹).

Some patients with advanced salivary tumors will have a protracted and indolent clinical course. This has been frequently observed in patients with metastatic ACC involving lung, so it is important to assess the pace of the disease before committing a patient to systemic therapy.

For patients with locally or regionally advanced SCC, much effort has been directed toward improvements in primary management with the addition of chemotherapy to surgery, radiotherapy, or both. Toward this end, three general strategies have been undertaken: (1) induction, also known as neoadjuvant therapy, with chemotherapy given before surgery or radiation; (2) concomitant chemoradiation, with chemotherapy given simultaneously with radiation to enhance its effect; and (3) adjuvant therapy, where chemotherapy is given after surgery or radiation in an effort to decrease microscopic metastatic disease burden.

Induction Chemotherapy

Induction chemotherapy has been investigated as an approach to improve outcomes in terms of overall survival and tumor control in patients with stage III/IV disease undergoing definitive local therapy. Theoretical advantages to this approach include reducing the risk of distant disease recurrence, enhancing organ preservation, improving response to definitive radiotherapy by reducing tumor bulk, and modification of subsequent local therapy to response.

This approach was first investigated in the 1970s after the cisplatin and 5-FU regimen proved to be highly active in metastatic disease. Trials over the next three decades investigated the role of chemotherapy added to local therapy in locally advanced disease. In the 2009 meta-analysis that established concomitant chemoradiation as a standard of care for nonsurgical management of stage III/IV SCC, direct comparisons of concomitant chemoradiation and induction chemotherapy indicated that although concomitant chemoradiation was superior to induction chemotherapy followed by radiation for local control and survival, induction chemotherapy was more effective at decreasing distant failure (⁴²). This conclusion lent credence to contemporaneous trials investigating induction chemotherapy followed by chemoradiation.

Following demonstration of the activity of taxanes in head and neck cancer, a series of trials investigated the combination of three-drug regimens with a platinum, taxane, and 5-FU. In 2007, two multicenter phase III trials, the European TAX 323 (⁴³) and the North American TAX 324 (⁴⁴), demonstrated the superiority of induction TPF (docetaxel, cisplatin, 5-FU) over PF (cisplatin, 5-FU). In TAX 323, a total of 358 patients with untreated, unresectable, and locally advanced tumors were randomized to receive either docetaxel 75 mg/m², cisplatin 75 mg/m², and 5-FU 750 mg/m²/d for 5 days (TPF) or cisplatin 100 mg/m² and 5-FU 1,000 mg/m²/d for 5 days (PF), followed by radiotherapy alone. The primary end point, median progression-free survival, was 11.0 months in the TPF group and 8.2 months in the PF group (hazard ratio [HR], 0.72; P = .007), and median overall survival was 18.8 months versus 14.5 months. The TAX 324 trial also compared TPF to PF, but in that study, the doses in the TPF arm were docetaxel 75 mg/m², cisplatin 100 mg/m², and 5-FU 1,000 mg/m²/d for 4 days. Both induction regimens were followed by concomitant chemoradiotherapy with weekly carboplatin area under the curve (AUC) of 1.5. The median overall survival was 71 months in the TPF group and 30 months in the PF group (P = .006). Notably, there was better locoregional control in the TPF group than in the PF group (P = .04). Rates of neutropenia and febrile neutropenia were significantly higher in the TPF group in both studies.

These two phase III trials led to US Food and Drug Administration (FDA) approval of induction chemotherapy TPF for patients with locally advanced HNSCC in 2007. Despite the impressive results of TAX 324, the value of adding induction chemotherapy to chemoradiation remained an unanswered question given the lack of definitive randomized trials comparing this approach to upfront chemoradiation. A randomized phase II trial by Paccagnella et al of induction TPF followed by chemoradiation versus chemoradiation alone reported a higher radiologic clinical response and a trend toward improved progression-free survival and overall survival (⁴⁵).

Two recent phase III trials, PARADIGM (⁴⁶) and DeCIDE (⁴⁷), were designed to test the hypothesis that induction chemotherapy followed by chemoradiation would confer a survival benefit over chemoradiation alone. These studies failed to meet their accrual targets and were therefore underpowered. The PARADIGM trial enrolled previously untreated patients with SCC of the oral cavity, oropharynx, hypopharynx, or larynx, patients with tumors deemed to be either unresectable or of low surgical curability on the basis of T stage (T3 or T4) and/or nodal status (N2 or N3), or patients who were candidates for organ preservation. The concomitant chemoradiation alone control group received cisplatin 100 mg/m² on days 1 and 22, and radiotherapy was given as an accelerated concomitant boost over 6 weeks for a total of 72 Gy in 1.8-/1.5-Gy fractions. Induction TPF was given as per TAX 324, and subsequent chemoradiation was adapted to response. Partial and complete responders received weekly carboplatin at AUC 1.5 for 7 weeks and 70 Gy of radiotherapy over 7 weeks in 2-Gy fractions. Patients who responded poorly were treated with an intensified regimen of weekly docetaxel 20 mg/m² for 4 weeks and 6 weeks of radiation to 72 Gy. The trial was powered to detect an improvement in 3-year survival from 55% in the control group (based on historical controls) to 70% in the induction group. However, due to slow accrual, the trial closed with 145 patients enrolled, less than half of the planned enrollment. Overall 3-year survival was 73% (95% confidence interval [CI], 60%-82%) in the induction TPF followed by chemoradiation group versus 78% in the chemoradiation alone group (⁴⁶).

The DeCIDE trial (⁴⁷) randomized patients with N2/3 M0 disease to either two cycles of induction TPF followed by chemoradiation or chemoradiation alone. Chemoradiation was given as DFHX (docetaxel, 5-FU, and hydroxyurea) with concurrent twice-daily radiotherapy or IMRT. Radiation doses were adaptive, with 74 to 75 Gy given to gross disease, 54 Gy to high-risk microscopic disease, and 39 Gy to low-risk microscopic disease. This study was similarly powered to detect an improvement in 3-year survival from 50% in the control group to 65% in the induction group. However, this trial also failed to accrue well, enrolling 280 patients versus the planned 400 patients. There were no statistically significant differences in overall survival (HR, 0.91; 95% CI, 0.59-1.41) or relapse-free survival. In competing risk analysis, there was a statistically significant reduction in risk of distant relapse without locoregional recurrence in the induction arm (P = .043).

Given that both studies failed to show survival benefit, but were ultimately underpowered to do so, the role of induction chemotherapy remains controversial. Importantly, in both studies, the 3-year overall survival rate of 70% to 78% in the chemoradiation alone arm was significantly higher than the historical control of 50% to 55% used in the pretreatment power calculations. This improvement in survival is likely multifactorial—a result of improved supportive care, technical advances in radiotherapy, and a shift in the biology of the disease the incidence as smoking-induced cancers declines and HPV-associated HNSCCs increases. Although the PARADIGM trial did not test for HPV, in DeCIDE, the HPV-positive rate of the 31% of patients tested was over 80%, and HPV-associated disease is known to have higher survival rates.

The Gruppo di Studio Tumori della Testa e del Collo (GSTCC) trial presented by Ghi and colleagues at the 2014 American Society of Clinical Oncology (ASCO) annual meeting, but not yet published, is a 2 × 2 factorial design of induction TPF versus no induction followed by chemoradiation with either cetuximab or PF in 415 patients. This trial showed a survival benefit of 53.7 months in the induction arm versus 30.3 months in the upfront chemoradiation arm (HR, 0.72; 95% CI, 0.55-0.96; P = .025) and additionally showed a reduction in distant metastases. The difference in outcomes observed in this trial versus the DeCIDE and PARADIGM trials is likely due to different patient populations and inclusion criteria.

Induction chemotherapy may also have value as a component of a sequential approach in which chemotherapy is followed by radiotherapy as a single modality in select locally advanced patients, thus sparing some of the toxicity of concomitant chemoradiation. In our center, a phase II trial with 47 patients investigated the efficacy of combining cetuximab with paclitaxel and carboplatin in a 6-week induction regimen followed by risk-based local therapy (radiation, concomitant chemoradiotherapy, or surgery) based on tumor stage and site at diagnosis. Inclusion criteria were stage IV M0 with nodal staging of N2b/c/N3. Of note, local therapy was determined at diagnosis and was not adapted to response. After induction chemotherapy, 9 patients (19%) achieved a clinically complete response, and 36 patients (77%) achieved a partial response. Local therapy consisted of concomitant chemoradiotherapy in 23 patients, radiotherapy alone in 23 patients, and surgery in 1 patient. The 3-year progression-free survival and overall survival rates were 87% and 91%, respectively (⁴⁸). A recent update to this trial reported a 5-year overall survival rate of 89% and very favorable long-term speech and swallow functions (⁴⁹). This strategy is undergoing further testing. At the ASCO 2014 annual meeting, Cmelak et al (⁵⁰) presented preliminary results of E2399. Patients with locally advanced HPV-positive disease responding to induction chemotherapy with paclitaxel, cisplatin, and cetuximab were effectively treated with a reduced-dose cetuximab-IMRT regimen, 54 Gy, if they achieved a clinical complete response to the induction chemotherapy. An early outcomes analysis showed 84% progression-free-survival and 95% overall survival at 2 years.

Concomitant Radiotherapy and Chemotherapy

In patients with locally advanced but M0 disease, the strategy of concomitant radiotherapy and chemotherapy has led to improved local and regional tumor control compared to radiotherapy alone (⁴²). Synergy between chemotherapy and radiation is based on several mechanisms, including (1) inhibition of DNA repair; (2) redistribution of cells to sensitive phases of the cell cycle; and (3) promoting oxygenation of anoxic tissues. The net effect is to improve cellular cytotoxicity (^51,52,53). However, combined therapy also enhances acute mucocutaneous toxicity, which may prompt subsequent dose reductions and treatment interruptions in radiotherapy. Thus, in combining these two treatment modalities, it is essential that toxicity not preclude the delivery of therapy in an effective schedule to avoid compromise of efficacy.

In a landmark phase II trial in 1987, the RTOG administered cisplatin (100 mg/m²) every 3 weeks to 124 patients with locally advanced unresectable head and neck cancer (⁵⁴). Sixty percent of patients completed the combined treatment per protocol, and 69% of all patients achieved a complete response. A comparison to RTOG patients treated with radiotherapy alone suggested improvement in survival time for the combined treatment.

The use of concomitant combination chemotherapy and radiation has long been under intense study (⁵⁵). Meta-analysis (⁴²) of prospective clinical trials demonstrates an enhancement of local tumor control and improvement of survival with combined therapy over radiation treatment alone, and chemoradiation is the standard of care in locally advanced non-surgical disease.

Combining several drugs with radiation will enhance acute toxicity, which may be severe. Therefore, investigators have piloted trials designed with split-course radiation to allow for healthy tissue recovery. Most of these studies have been limited to patients with stage III or IV locally advanced SCC, with local control and improved survival time as the primary objectives. These regimens alternate chemotherapy and radiotherapy or use split-course radiotherapy to maximize tumor cell kill and minimize tissue toxicity. However, protracted radiation treatment times may result in decreased local control rates because of accelerated repopulation of cancer stem cells (^56,57). The strategy of alternating non–cross-resistant agents may potentially eliminate not only tumor cell repopulation but also primary drug resistance.

Brizel et al (⁵⁸) compared a hyperfractionated radiotherapy arm to total dose of 75 Gy versus concomitant PF and hyperfractionated radiation to 70 Gy followed by two cycles of adjuvant chemotherapy. There was a statistically significant improvement in local disease control and a strong trend toward improved overall survival for the combined-modality arm. In this trial, neck dissection was recommended in patients with N2/3 disease. Clayman et al (⁵⁹) have reviewed the MDACC experience, examining the indication for neck dissection in this patient population. Their report suggests that neck dissections are required only when there is radiographic evidence of residual disease 6 to 8 weeks following the completion of definitive chemoradiation. Wendt et al (⁶⁰) reported a statistically significant 3-year survival advantage after the concomitant use of cisplatin, 5-FU, and leucovorin with split-course radiotherapy versus radiotherapy given as a single therapeutic modality. Calais et al (⁶¹) compared a more standard once-daily fractionation radiation schedule with the same radiotherapy and concomitant carboplatin and 5-FU, demonstrating a statistically significant advantage in locoregional tumor control and overall survival at 3 years. Jeremic et al (⁶²) also investigated the value of adding cisplatin given daily to a hyperfractionated radiation therapy program versus the same radiation schedule given alone in patients with locally advanced HNSCC. In this report, locoregional and distant disease control and overall survival were improved at 5 years. Adelstein et al (⁶³) compared standard daily radiotherapy with two schedules of concomitant chemoradiotherapy in a large intergroup study. The addition of high-dose cisplatin to conventional single daily dose radiotherapy improved survival from 23% to 37% at 3 years. The clearest benefit in these studies was an improvement in locoregional control, which translated into a survival advantage. Acute toxicity was increased, especially mucositis and hematologic effects, but there was no obvious escalation of long-term sequelae. However, this may need further investigation. In aggregate, overall 3-year survival exceeded 50% in these experimental programs, underscoring the potential therapeutic efficacy of concomitant chemotherapy and radiation in patients with advanced head and neck cancers.

Cetuximab, a monoclonal antibody, is approved for use in combination with radiation in previously untreated patients. In a landmark study, patients with locoregionally advanced head and neck cancer were randomly assigned to receive either high-dose radiotherapy alone (213 patients) or high-dose radiotherapy plus weekly cetuximab (211 patients) at an initial dose of 400 mg/m² of body surface area, followed by 250 mg/m² weekly for the duration of radiotherapy (⁶⁴). The primary end point, median duration of locoregional control, was 24.4 months among patients treated with cetuximab plus radiotherapy and 14.9 months among patients given radiotherapy alone (HR, 0.68; P = .005). The median duration of overall survival was 49.0 months among patients treated with combined therapy and 29.3 months among patients treated with radiotherapy alone (HR for death, 0.74; P = .03). However, the rates of distant metastases at 1 and 2 years were similar in both groups. With the exception of acneiform rash and infusion reactions, the incidence of grade 3 or greater toxic effects, including mucositis, did not differ significantly between the two groups.

Cetuximab plus radiotherapy is directly compared to chemoradiation for patients with HPV-associated oropharyngeal cancer in a phase III randomized trial (RTOG 1016), but results from this trial are not yet available. This trial represents the recent trend in investigational treatment strategies undertaken by clinical trial cooperative groups that have focused on treatment “de-intensification” for selected patients (namely those with HPV-associated oropharyngeal cancer), with the goals of maintaining or improving established cure rates but reducing treatment-related toxicity.

There has been a series of trials investigating the use of EGFR antibodies with chemoradiation. The phase III RTOG 0522 trial randomized 940 patients to high-dose cisplatin-based chemoradiotherapy with or without cetuximab (⁶⁵). The combined biochemoradiotherapy failed to meet the primary end point of improving progression-free survival, with a 3-year rate of 61.2% versus 58.9% with cetuximab, and demonstrated a trend toward worse locoregional control. This trend was likely the result of significantly increased toxicities that led to radiation interruptions in 26.9% of patients. There was also a significant difference in treatment-related deaths (10 vs 3; P = .05). The CONCERT-1 and -2 trials have further explored bioradiotherapy with panitumumab (Table 19-4) (^66,67), showing no overall survival or local disease control advantage after matching chemoradiotherapy with the addition of the antibody.

The aggregate results of these trials indicate that improved disease-free and overall survival times have been obtained for patients with locally advanced HNSCC using concomitant chemotherapy and radiotherapy rather than radiotherapy as a single treatment modality. Combination chemotherapy with radiotherapy may increase response but causes increased toxicity. Well-designed clinical trials are still needed to determine optimal chemotherapy and radiotherapy schedules.

Adjuvant Chemoradiotherapy

Adjuvant chemotherapy is indicated in patients at high risk of recurrence after surgical resection, generally defined as having narrow or involved margins at the primary site, multiple nodal metastases, or extracapsular spread (Table 19-5) (^68,69,70).

Two large phase III studies, RTOG 9501 (⁶⁹) and EORTC 22931 (⁷⁰), tested cisplatin-based concomitant chemoradiotherapy in the adjuvant setting. Although with some variations between the studies, patients with high-risk features (positive margin, extracapsular spread, lymphovascular invasion, perineural invasion, and multiple positive lymph nodes) were randomly assigned to receive either radiotherapy alone or radiotherapy plus cisplatin at 100 mg/m² every 3 weeks for three cycles. In RTOG 9501, concomitant chemoradiotherapy significantly reduced the risk of locoregional recurrence compared with radiotherapy alone (HR for local or regional recurrence, 0.61; P = .01). However, no survival benefit was observed. In addition, the incidence of grade 3 or greater adverse effects was 34% in the radiotherapy group and 77% in the combined-therapy group (P < .001). In EORTC 22931, both the progression-free survival (HR, 0.75; P = .04) and overall survival (HR, 0.70; P = .02) rates were significantly higher in the combined-therapy group than in the radiotherapy group. Severe acute adverse effects were more frequent after combined therapy (41%) than in the radiotherapy group (21%).

More recently, based on the benefit of cetuximab bioradiotherapy in the definitive setting and the additive benefit of cetuximab to chemotherapy in the metastatic setting (⁷¹), RTOG 0234 explored the incorporation of cetuximab into adjuvant chemoradiation (⁷²). This phase II trial compared two biochemoradiotherapy regimens to historical high-dose cisplatin-based chemoradiotherapy in RTOG 9501 with the intent to select a regimen for further testing against standard high-dose cisplatin-based chemoradiotherapy in a phase III trial. Both docetaxel (15 mg/m²)/radiation/cetuximab and weekly cisplatin (30 mg/m²)/radiation/cetuximab outperformed the historical control with 2-year overall survival rates of 79% and 69% and 2-year disease-free survival rates of 66% and 57%, respectively (HR, 0.69 for the docetaxel arm vs control, P = .01; and HR, 0.76 for the cisplatin arm vs control, P = .05). Grade 3 or 4 myelosuppression was observed in 28% of patients in the cisplatin arm and 14% of patients in the docetaxel arm, and mucositis was observed in 56% and 54% of patients, respectively. Although these results are promising, as has been noted previously, comparison with historical controls is problematic given the shifting epidemiology from smoking-related cancer to better prognosis HPV-related cancers, which has likely contributed to the improvements in survival rates of the control arms seen in the recent induction trials (^46,47). RTOG 1216 is an ongoing phase II/III trial of surgery and postoperative radiation delivered with concurrent cisplatin versus docetaxel versus docetaxel and cetuximab for high-risk HNSCC.

Although adjuvant concomitant chemoradiotherapy has been demonstrated to be more effective than radiotherapy, there is significant associated toxicity. The two risk factors most associated with benefit from concomitant chemoradiotherapy are extracapsular extension and positive surgical margins (⁷³).

Many HNSCCs are diagnosed at a late stage. Stages III and IV tumors often necessitate extensive or radical surgery, which may alter organ function. Problems with radical surgery include loss of speech, loss of swallowing function, or disfigurement without a concomitant improvement in survival time. Therefore, preservation of function became one of the major challenges. This approach was first explored in laryngeal cancer given the high morbidity associated with laryngectomy.

The landmark VA study published in 1991 randomized 332 patients with stage III or IV SCC of the larynx to receive either induction chemotherapy consisting of PF followed by radiotherapy or surgery and postoperative radiotherapy (³⁴). Patients who experienced no tumor response to chemotherapy or those who had locally persistent or recurrent cancer underwent salvage laryngectomy. Two-year survival for both treatment groups was 68%, and 41% of patients randomly assigned to the experimental arm were alive with a functional larynx at 2 years. Thus, the efficacy of chemotherapy followed by radiotherapy (with surgical salvage) was similar to that of surgery followed by radiotherapy and established organ preservation as a realistic goal of nonsurgical treatment administered with curative intent. Patterns of failure differed, with patients in the chemotherapy arm more likely to have locoregional recurrence and distant recurrence more common in the surgical arm. Lefebvre et al (³¹) later reported that sequential chemotherapy and radiation could also be effective in selected patients with cancers of the hypopharynx.

The VA larynx study prompted further investigations of chemotherapy and radiotherapy for the treatment of larynx cancer using the sequential administration of induction chemotherapy, consisting of PF, followed by radiotherapy versus concomitant cisplatin-radiotherapy versus radiotherapy administered as a single treatment modality in RTOG 91-11 (⁷⁴). For all groups, totaling 547 patients, surgical salvage was reserved for patients with persistent or locally recurrent disease. Both chemotherapy groups demonstrated improved laryngectomy-free survival (the composite primary end point) compared to radiotherapy alone. The results indicated a significant advantage for concomitant cisplatin treatment, with preservation of the larynx in 88% of patients treated in the concomitant arm. In the recently published 10-year update (⁷⁵), locoregional control and larynx preservation were significantly improved with concomitant chemoradiotherapy compared with the induction arm RT (HR, 0.58; 95% CI, 0.37-0.89; P = .005), whereas the induction chemotherapy group showed a nonsignificant trend toward improved overall survival (HR, 1.25; 95% CI, 0.98-1.61; P = .08). This difference in survival was driven by non–cancer-related deaths, the cause of which remains unexplained.

Organ preservation has also been studied in hypopharyngeal cancer given that laryngectomy is often part of the surgical treatment of this disease. Similar to the VA study, the EORTC study compared induction chemotherapy with PF followed by radiation versus conventional surgery plus postoperative radiation. As in the VA study, survival between the two arms did not differ, and patients in the chemotherapy arm had a high rate of larynx preservation.

As with TAX 323 and TAX 324, TPF was also explored in locally advanced but resectable SCC of the larynx or hypopharynx (⁷⁶). Compared to PF, TPF was shown to increase tumor responsiveness and lead to improved larynx preservation (70% vs 58% in the PF arm), with no compromise in overall survival. The efficacy of induction TPF versus concurrent chemoradiation has not been explored. The concurrent approach tends to be favored in the United States, whereas Europeans tend to prefer induction.

These trials indicate that for patients with intermediate-stage SCC of the larynx, a combined treatment program with the objectives of tumor eradication and laryngeal preservation is appropriate. It is also important to recognize that patients with locally advanced, destructive primary laryngeal cancers were not included in the more recent multigroup trial. These patients may require total laryngectomy for optimal tumor control and preservation of function.

It should be noted, however, that nonsurgical treatment also carries risk of morbidity and functional impact. Radiation produces tissue changes that can result in immediate and long-term alterations in speech and swallowing. The adverse impact of radiation may equal or exceed that associated with surgery, depending on the treatment dose and volume, and the sequelae of treatment may manifest or increase in severity years after the completion of treatment. Fibrosis may reduce the range of motion of the tongue and jaw and diminish pharyngeal wall motion. Historically, 20% to 40% of patients receiving chemoradiotherapy for SCC of the oropharynx and hypopharynx may require long-term gastrostomy tube feedings. However, the long-term gastrostomy rate for patients with oropharynx cancer treated with modern radiation therapy approaches (eg, IMRT) is less than 10% (⁷⁷). Radiation to the larynx often results in swallowing problems related to pharyngeal transport. To counteract the deleterious effects of radiation and chemoradiotherapy, there are rehabilitative options, which are best administered by a qualified speech pathologist.

Patients with tumor recurrence after primary treatment who are not candidates for surgical salvage may be offered palliative cytotoxic chemotherapy or investigational therapy. Methotrexate, cisplatin, carboplatin, bleomycin, 5-FU, and the taxanes are drugs with single-agent activity in the range of 15% to 25%. Previous studies have consistently demonstrated response rates of 30% to 40% for combination chemotherapy, usually cisplatin based, with a median survival of 6 to 9 months. There has been no clear demonstration of a survival advantage over single-agent treatment or even best supportive care. However, in the appropriate context with the goal of reducing symptoms, combination chemotherapy with PF or a platinum-taxane combination has become a frequently exercised practice in the care of patients with incurable HNSCC, and a fraction of patients treated with these combinations have extended survival.

Current investigations are under way in an attempt to develop effective targeted treatment approaches. Epidermal growth factor receptor is overexpressed in a majority of invasive HNSCCs. The small-molecule inhibitors gefitinib (500 mg/d) and erlotinib (150 mg/d) downregulate the phosphorylation of tyrosine kinase residues in the cytoplasmic domain of EGFR and have demonstrated single-agent activity in 11% and 5% of patients, respectively, with advanced disease (^78,79,80). Cetuximab is a chimeric murine-human monoclonal antibody directed against the extracellular domain of EGFR. Burtness et al (⁸¹) conducted a prospective randomized trial in patients with recurrent HNSCC and demonstrated responses in 26% of patients treated with cetuximab and cisplatin versus 10% of patients treated with cisplatin alone. However, the primary end point, progression-free survival, was not significantly different. Cetuximab was tested in a phase II trial as monotherapy in 103 patients with recurrent or metastatic HNSCC refractory to platinum-based therapy (⁸²). The response rate was 13%, disease control rate (complete response/partial response/stable disease) was 46%, and median time to progression was 70 days. There appeared to be no benefit in adding cisplatin to these patients. Afatinib is a small-molecule tyrosine kinase inhibitor that targets EGFR and HER2. This compound is under study in ECOG 1311 as an adjuvant systemic therapy after definitive chemoradiation and neck dissection for high-risk patients.

In a major phase III trial, cetuximab in combination with chemotherapy was investigated in patients with untreated recurrent or metastatic HNSCC (⁷¹). In this trial, 442 patients were randomized to receive either cisplatin or carboplatin plus 5-FU, with or without cetuximab. The cetuximab group had longer overall survival (10.1 vs 7.4 months) and median progression-free survival (5.6 vs 3.3 months). Thus, cetuximab plus PF chemotherapy improved overall survival when given as first-line treatment in patients with recurrent or metastatic HNSCC.

Further study of molecular biomarkers and selection of targeted therapies for trials both in definitive and palliative treatment settings is receiving much emphasis. Activation of the PI3k/Akt signaling pathway is under study in HNSCC, and phase I/II trials with PI3K inhibition are ongoing (⁸³). One of the more exciting agents currently under investigation is MK-3475, a PD-1 antibody that acts as an immune checkpoint inhibitor. This type of immunotherapy has been approved in melanoma and lung cancer and is being investigated in multiple tumor types with promising results. A phase Ib study presented by Seiwert et al (⁸⁴) at ASCO 2014 enrolled 61 patients with metastatic HNSCC, 36 HPV-negative patients and 23 HPV-positive patients. The response rate was 19.6%, and an additional eight patients experienced stable disease for over 6 months. Response was correlated with programmed death ligand 1 (PD-L1) expression. There was no difference in response between HPV-negative and HPV-positive disease. A phase III study is planned.

The decades-long history of clinical and translational study of retinoids in oral premalignant lesions, or intraepithelial neoplasia (IEN), has advanced our understanding of the biology of carcinogenesis and molecular-targeted drug development, even though definitive clinical testing has not shown that retinoids can prevent oral cancer (⁸⁵). One early trial in 1986 tested a high dose of the retinoid 13-cis-retinoic acid (13cRA) against placebo in 44 evaluable oral IEN patients for only 3 months (⁸⁶). The complete plus partial clinical response rate in the retinoid arm was 67% (vs 10% in the placebo arm) (P = .0002). Histopathologic responses also favored the 13cRA arm. Over half of the responders in the 13cRA arm, however, recurred or developed new lesions within 3 months of stopping the intervention. This high-dose, short-term trial led to another early trial in oral IEN patients, which was designed to reduce the toxicity of and prolong the response to 13cRA. A short-term (3-month) course of high-dose 13cRA (1.5 mg/kg/d) was followed by a 9-month maintenance course with low-dose 13cRA (0.5 mg/kg/d) or β-carotene (30 mg/d) in IEN patients who responded to or were stable after the induction phase (⁸⁷). The maintenance-phase progression rates were 8% in the 13cRA group and 55% in the β-carotene group (P < .001). Nonetheless, on long-term follow-up (median of 66 months), the incidence of in situ or invasive cancer was not different between the two arms (23% for low-dose 13cRA vs 27% for β-carotene) (⁸⁸).

To address the short-lived chemopreventive effects of 13cRA, Papadimitrakopoulou et al designed a follow-up study comparing an extended, 3-year treatment period with 13cRA at lower doses (0.5 mg/kg/d for 1 year followed by 0.25 mg/kg/d orally for 2 years; control arm) to β-carotene (50 mg/d) plus vitamin A (ie, retinyl palmitate 25,000 IU/d; experimental arm) in 162 patients with leukoplakia, using a noninferiority design (⁸⁹). During the study, β-carotene had to be dropped from the experimental arm due to emerging data demonstrating an increased risk of lung cancer incidence and mortality in other ongoing chemoprevention trials at that time. The study showed an inferior 3-month response rate in the vitamin A alone arm, lower tolerance to treatment with 13cRA, a lack of statistical significance in the test for noninferiority between the control and the experimental arm(s), and, more importantly, a similar oral cancer–free survival across all groups. This study, which is one of the longest term performed to date in patients with leukoplakia, demonstrated that 13cRA is still not well tolerated for long-term treatment, even at reduced doses and that less toxic regimens (ie, vitamin A alone) are ineffective. Furthermore, an impact on oral cancer incidence has yet to be demonstrated with any of these regimens. In addition to trials involving oral IEN, retinoids have also been studied for prevention of second primary head and neck cancers. A randomized, placebo-controlled study of high-dose 13cRA (50–100 mg/m²/d for 12 months) in definitively resected head and neck cancer patients demonstrated a lack of effect of the retinoid on distant, nodal, or local recurrence rates, but there was a statistically significant decrease in the incidence of second primary tumors (4% vs 24%, P = .005) that persisted on long-term follow-up (^90,91). Unfortunately, a follow-up trial of a tolerable low dose of 13cRA (30 mg/d for 3 years) in 1,190 early-stage patients did not prevent second primary tumors (⁹²). Randomized studies in this setting with the second-generation retinoid etretinate (n = 316 patients) (⁹³), vitamin A and/or N-acetylcysteine (n = 2,592) (⁹⁴), β-carotene (n = 264) (⁹⁵), and α-tocopherol plus β-carotene (n = 540) (⁹⁶) also did not demonstrate any clinical benefit in terms of prevention of second primary tumors.

Although the randomized retinoid trials failed to produce a chemoprevention strategy that could be considered standard of care, they were embedded with translational studies that helped to advance the overall understanding of the biology of intraepithelial carcinogenesis, molecular markers—for example, retinoic acid receptor (RAR) β, p53, p16, EGFR, and genetic instability—for developing drugs, monitoring interventions, and assessing cancer risk and pharmacogenomics.

In terms of cancer risk assessment, cyclin D1 genotype (^97,98) and loss of heterozygosity at certain chromosomal sites (^99,100) have emerged in multiple studies as prognostic factors that could be potentially useful in the clinic. Building on these data, investigators at MDACC led a clinical study evaluating the effects of erlotinib (150 mg/d for 1 year) on the incidence of invasive cancer in patients with oral premalignant lesions (with or without a prior history of oral cancer) selected for high risk based on loss of heterozygosity testing—the Erlotinib Prevention of Oral Cancer (EPOC) study (^101,102). This was the first large-scale study in oral IEN that used molecular risk assessment as part of the inclusion criteria, thus bringing the concept of precision medicine to the chemoprevention field. Loss of heterozygosity high-risk profiles were indeed associated with increased oral cancer incidence on long term follow-up and are now considered the most robust molecular markers of cancer risk in oral IEN. Erlotinib, however, did not improve oral cancer-free survival in this high-risk population.

As the role of HPV-16 in the pathogenesis of a subgroup of HNSCC becomes substantiated, preventive strategies targeting this infectious agent could be explored as well. Human papillomavirus vaccination is already being used to prevent cervical cancer. Human papillomavirus vaccine has been shown to reduce the prevalence of oral HPV infections, but its impact on incidence of HNSCC is yet to be determined (¹⁰³). One anticipates, though, that widespread HPV vaccination may contribute to reducing the burden of HPV-induced HNSCC in the coming decades.

Head and neck SCC is a major international health problem. General public health strategies such as reducing tobacco usage and increasing awareness of associated risks are of primary importance.

The demonstration of HPV as a causative agent for oropharyngeal cancers is of great importance because this will carry implications for prevention and also will influence decision making in treatment planning as well as the conduct of clinical trials.

The optimal care and treatment of head and neck cancer patients are multidisciplinary. Surgical resection and/or radiotherapy are powerful local modalities and the care of treatment for most patients. Emerging data support the administration of chemotherapy as a component of combined-modality treatments, especially in patients with advanced HNSCC. For patients with locally recurrent or metastatic disease, combination chemotherapy may produce response rates of 30% to 40%. However, responses tend to be brief, lasting a median of 3 to 6 months, and are associated with only a modest prolongation of survival. Thus, chemotherapy for these patients is palliative. An exception to this is for patients with NPC, with higher response rates and a small proportion of long-term disease-free survivors. For patients with metastatic HNSCC of any primary site, the addition of cetuximab to platinum–5-FU appears to improve tumor responses and overall survival. Targeted agents are currently under investigation. Prognostic and predictive markers are needed to improve selection of patients who are most likely to benefit from palliative treatment. Enrolling in investigational studies is strongly supported.

In the newly diagnosed patient with locally advanced disease, high response rates have been observed with induction chemotherapy. The addition of a taxane to the more traditional PF platform has increased overall activity, but the use of induction treatment for most patients remains an investigational endeavor. The potential for augmentation of local control with a substantial response to chemotherapy followed by definitive surgery or radiation is also under investigation. Three large multicenter randomized trials have been successfully conducted in larynx cancer with preservation of function in subsets of patients.

Chemotherapy administered concomitantly with radiotherapy has improved local control and survival in a sequence of studies and is recognized as the standard of care for nonsurgical therapy of patients with locally advanced squamous cancers of the pharynx and larynx. The increase in toxicity associated with these regimens should be carefully considered in selecting patients for combined treatment.

Patients with earlier-stage disease (ie, stage I or II) generally should receive therapy with either surgery or radiotherapy or both. Patients with locally advanced M0 (stage III/IVA/B) disease may be considered for nonsurgical therapy, most often with chemotherapy and radiation, or entered into a combined chemoradiation treatment protocol. Patients with “resectable” disease can be further divided by site. Patients with primary oral cavity tumors are best served with surgery followed by radiotherapy (or chemoradiotherapy if there are high-risk pathologic features), whereas those with oropharyngeal, hypopharyngeal, or laryngeal tumors are often treated with radiation, with or without chemotherapy, depending on precise site and stage.

Basic and translational chemoprevention research in head and neck carcinogenesis is advancing our understanding of the molecular characteristics of carcinogenesis and cancer risk. We have studied EGFR inhibition in a prospective, controlled trial in high-risk patients. This project illustrates the convergence of prevention and therapy, whereby a molecularly targeted agent known to have efficacy in the setting of invasive cancer is brought into the premalignant space.

The management of head and neck cancer is a multidisciplinary activity. The identification of effective chemotherapeutic agents and their integration into the initial therapy of head and neck cancer have the potential to improve survival time and preserve organ function. Moreover, studies are under way to reduce treatment intensity and thereby long-term toxicity. Through well-designed and executed clinical trials, coupled with basic research of the biology of upper aerodigestive tract tumors, further advances in the management and prevention of these cancers can be achieved.