Head and Neck Tumors

Treatment Approaches

In general, early-stage (stage I or II) head and neck tumors may be treated using a single modality (surgery or radiotherapy), whereas advanced disease (stage III or IV) frequently benefits from multimodality therapy.

The best therapeutic approach for the primary tumor depends on the anatomic site. Most neck disease can be treated equally well with surgery or radiation, thus the modality chosen to treat the neck is based on which modality is selected for the primary. When the primary tumor is treated with irradiation, the "at-risk" regional lymphatics are incorporated into the treatment fields. Neck dissections should remain standardized (ie, complete anatomic dissections, as opposed to "berry picking" or random biopsy) in these settings to avoid incomplete surgery.

Preoperative Assessment

Before surgical resection, preoperative assessment of the extent of disease is essential. Complete physical examination and appropriate radiologic evaluation are necessary. Direct laryngoscopy and esophagoscopy are frequently performed to determine tumor extent and rule out the presence of a second primary tumor. A chest x-ray or CT scan may be obtained to screen for distant metastases or a primary lung cancer.

Surgical Principles

Classic principles of surgical oncology apply to head and neck cancer. Complete resection is necessary. Securing sufficient margins may be challenging because of the many structures in this area. Reconstruction is complex after resection of head and neck tumors, as the surgery may have an impact on appearance, speech, and swallowing. Decisions regarding the extent of resection should be made by experienced surgeons.

Surgery Plus Radiation Therapy

The combination of radical surgery and radiation therapy has been used for several decades to treat patients with advanced head and neck cancers.

Postoperative versus preoperative radiation therapy

Postoperative radiation therapy (60 to 70 Gy in 6 to 7 weeks) reduces the rate of locoregional recurrence from approximately 50% to 15% for tumors with pathologic features predictive of locoregional recurrence. The indications for postoperative radiation therapy are well established and include a deeply infiltrative primary tumor (T4); close or positive margins; an involved lymph node > 3 cm or multiple involved lymph nodes; extracapsular extension; an open lymph node biopsy not followed by an immediate neck dissection; perineural invasion; invasion of the lymphovascular space, cartilage, bone, or deep soft tissue; and surgeon unease (an experienced surgeon's recommendation to deliver postoperative radiation should be credited by the radiation oncologist because the surgeon will often appreciate worrisome features not documented in a pathology report). The addition of postoperative radiation therapy reduces the risk of locoregional failure but does not decrease the risk of developing distant metastases.

Preoperative radiotherapy (45 to 50 Gy in 4 to 5 weeks) has been used for patients with advanced primary tumors, but rates of locoregional recurrence appear to be lower and complications fewer compared with postoperative radiation therapy; however, this observation was based on a study performed in the 1970s. Preoperative radiotherapy with or without chemotherapy is indicated for marginally resectable tumors, such as those with fixed cervical lymph nodes. In this setting, preoperative irradiation often permits resection of an otherwise unresectable tumor. However, this approach is used rarely because of the improvements in outcome with chemoradiation and the poor functional outcomes of patients requiring trimodality therapy.

Postoperative chemotherapy/radiation therapy

Two randomized clinical trials were launched to determine whether the addition of chemotherapy to radiation therapy enhanced locoregional tumor control and survival in high-risk patients with head and neck cancer following definitive surgical resection. The results of these trials were published in 2004. In the RTOG 95-01/Intergroup trial, patients with high-risk features including two or more involved lymph nodes, extracapsular extension, or positive margins following definitive surgical resection were randomized to receive 6,000 to 6,600 cGy of postoperative irradiation with or without cisplatin(Drug information on cisplatin) (100 mg/m² given on days 1, 22, and 43). With a median follow-up of 45.9 months, the estimated 2-year rate of local and regional tumor control was 82% in the combined-therapy group vs 72% in the radiotherapy group (P = .01). The disease-free (P = .04), but not overall (P = .19), survival was significantly longer in the combined-treatment group, although the primary endpoint was locoregional control and was not powered to demonstrate a survival advantage. The incidence of acute adverse effects (grade 3 or greater) was higher in the combined-treatment group (77% vs 34%; P < .001).

In the European Organisation for Research on the Treatment of Cancer (EORTC) 22931 trial, similar conclusions were reported. This study included patients with T3 or T4 disease, perineural invasion, and involvement of the lymphovascular space in addition to positive margins and extracapsular extension. The overall survival rate was significantly improved in the combined-therapy group compared with the radiotherapy group. After a mean of 60 months, the overall survival of 334 randomized patients was 53% in the combined-treatment group vs 40% in the radiation-alone group (P = .02). The discrepancy in effect on overall survival between the two studies was thought to be related to different entry criteria used in the RTOG and EORTC trials.

In a comparative analysis of the two trials, the presence of extracapsular extension and/or microscopically involved surgical margins were the only risk factors for which the impact of chemoradiation therapy significantly improved survival.

Curative radiation therapy

Radiation therapy with curative intent usually involves daily treatment for 6 to 7 weeks (total dose: 60 to 70 Gy). Although there is no tissue loss with radiation therapy, as there is with surgery, potential complications include dry mouth, tissue fibrosis, trismus, bone necrosis, hypothyroidism, and dysphasia. Some problems are common and sufficiently debilitating that they require significant care during treatment planning. Surgery often produces less morbidity in the oral cavity, whereas radiotherapy produces less morbidity in the oropharyngeal, laryngeal, and nasopharyngeal regions.

Radiation fractionation

The RTOG 90-03 trial was conducted to determine the efficacy of various fractionation schemes in the treament of locally advanced head and neck cancer. Four schedules were tested: (1) standard fractionation at 2 Gy/fraction/d, 5 d/wk, to 70 Gy/35 fractions/7 wk; (2) hyperfractionation at 1.2 Gy/fraction, twice daily, 5 d/wk to 81.6 Gy/68 fractions/7 wk; (3) accelerated fractionation with split at 1.6 Gy/fraction, twice daily, 5 d/wk to 67.2 Gy/42 fractions/6 wk, including a 2-week rest after 38.4 Gy; or (4) accelerated fractionation with a concomitant boost at 1.8 Gy/fraction/d, 5 d/wk, and 1.5 Gy/fraction/d to a boost field as a second daily treatment for the last 12 treatment days to 72 Gy/42 fractions/6 wk. A total of 1,113 patients were entered in the study, with a median follow-up of 23 months.

Patients treated with both hyperfractionation and accelerated fractionation with a concomitant boost had significantly increased locoregional tumor control rates (the primary endpoint) on RTOG 90-03 compared with patients on the other two arms. All three groups treated with the altered fractionation schemes had more acute, but not late, side effects. The study concluded that hyperfractionation and accelerated fractionation with a concomitant boost were both superior to conventionally fractionated radiotherapy.

An update of this trial further demonstrated that the hyperfractionated arm showed a trend toward an overall survival benefit (P = .06); the accelerated fractionation arm using the concomitant boost showed a trend toward increased late side effects (P = .06).

A recent, large meta-analysis that analyzed different fractionation schedules demonstrated an 8% improvement in overall survival with hyperfractionation similar to the benefit of added concurrent chemotherapy seen in the 1995 meta-analysis.

Intensity-modulated radiation therapy

Intensity-modulated radiation therapy (IMRT) is a more sophisticated approach for obtaining highly conformal radiation dose distributions needed to irradiate complex targets positioned near sensitive normal structures. Treatment planning for IMRT (also known as inverse planning) is extremely complex and different from that of conventional or three-dimensional radiation therapy planning. The starting point with IMRT is a description of the desired dose distribution rather than the application of traditional fields and beam modifiers to generate an acceptable plan. Conventional radiation treatment utilizes relatively uniform beams of radiation (typically between 2 and 4 beams), whereas IMRT, instead of using 4 beams of 50 cGy each, could use 50 beams of 4 cGy each. Each beam direction is divided into multiple segments to modulate the radiation dose.

The role of IMRT continues to evolve, although its use should not be considered standard for all head and neck tumors. One randomized trial in patients with early nasopharyngeal cancer showed a small improvement in salivary function. IMRT in head and neck cancer is ideal in the setting of a tumor adjacent to a critical structure (eyes, optic nerve, spinal cord, brainstem, optic chiasm, or spinal cord), which would not otherwise be treated adequately with conventional planning.

Various groups have examined dosimetric and quality-of-life differences between IMRT and conventional radiation techniques. The group at Memorial Sloan-Kettering Cancer Center compared its IMRT planned treatment for nasopharyngeal cancer against conventional treatment with a conformal boost. Locoregional tumor control was 97% vs 78%, with IMRT vs conventional radiotherapy, respectively, at 2 years. The University of Michigan and Washington University have reported reductions in late xerostomia with IMRT.

Proper patient selection is imperative when using IMRT. Marginal failures have been reported in the postoperative setting, in which no consensus guidelines exist regarding optimal target volume, and in the setting of bulky adenopathy. The postoperative setting is difficult, because disruption of normal draining lymphatics makes the appropriate target volume much more difficult to define. Bulky adenopathy may change the typical drainage pattern and lead to metastases in the base of the skull or parotid glands.

In addition, the RTOG-0234 group reported that IMRT in the postoperative setting increased acute side effects in high-risk patients when compared with three-dimensional conformal radiotherapy. This increase in side effects may be related to dosimetric parameters such as "hot spots" or a greater volume of normal structures receiving low doses.

In the definitive setting, all series have demonstrated improved outcomes. RTOG H-0022 was a multi-institutional study using IMRT for T1 or T2, N0 or N1 squamous cell carcinoma of the oropharynx to 66 Gy in 30 fractions. Investigators found that moderately accelerated hypofractionated IMRT without chemotherapy for early oropharyngeal cancer is feasible, achieving high tumor control rates (95% local control) and reduced salivary toxicity when compared with treatment of similar patients in previous RTOG studies.

Chemotherapy

Induction (neoadjuvant) chemotherapy vs concomitant chemotherapy and radiation therapy. A rationale for using induction chemotherapy for treating advanced-stage laryngeal cancer involves using chemotherapy as a marker of radiation sensitivity to select potentially radiocurable patients. In the Veterans Administration (VA) Laryngeal Cancer Cooperative Study, the major benefit for two-thirds of patients in the experimental arm was laryngeal preservation. In this study, the lack of a substantial tumor response to induction chemotherapy was not associated with reduced survival. Patients who failed to respond to induction chemotherapy underwent surgery, which had the advantage of not being performed in an irradiated field. One problem with this approach is that radiation sensitivity may not be related to chemotherapy. For example, in the RTOG 91-11 trial, 11 of 12 patients who refused laryngectomy because of lack of a response to chemotherapy were nevertheless salvaged with radiotherapy, suggesting that response to induction chemotherapy may not predict radioresponsiveness.

The use of concurrent treatment (as opposed to induction) using chemotherapy and irradiation is associated with increased locoregional tumor control and survival. When used concurrently with radiotherapy, the function of chemotherapy is thought to "radiosensitize" the tissue in the radiation field. The RTOG 91-11 trial demonstrated improved locoregional tumor control as well as an increased rate of laryngeal preservation in patients with stage III and IV resectable laryngeal cancer who underwent concurrent chemoradiation therapy (for details, see discussion later in chapter). However, the rate of mucosal toxicity in those receiving concomitant chemotherapy was double that of either of the other two arms. Patients with extensive T4 primaries, such as those with cartilage invasion or those with involvement of the tongue base, were not included in the study because of poor outcomes in the VA larynx study.

Investigators from the University of Michigan also studied the role of induction chemotherapy in advanced laryngeal cancer and conducted a phase II trial in which patients who had a response of greater than 50% to a single cycle of induction cisplatin (100 mg/m²) and 5-fluorouracil (5-FU) (1,000 mg/m²/d, days 1 to 5) went on to receive definitive chemoradiation. Of the 97 patients, 33% had T4 primary tumors. Seventy-three patients (75%) were able to undergo chemoradiation, and the overall laryngeal preservation rate was 70%. After a median follow-up of 41.9 months, the overall 3-year survival rate was 85%.

Biologic agents. The toxicity of chemotherapeutic agents has prompted the development of molecular agents with potentially greater tumor specificity and diminished toxicity. Targeting the epidermal growth factor receptor has thus been a subject of study in squamous cell head and neck cancers. In one study, 424 patients with locoregionally advanced head and neck cancer were randomly assigned to receive treatment with high-dose radiotherapy alone or with weekly cetuximab(Drug information on cetuximab) (Erbitux). The primary endpoint was locoregional control; secondary endpoints included overall and progression-free survival. The median duration of locoregional tumor control was 24.4 months among patients treated with cetuximab plus radiotherapy and 14.9 months among those given radiation therapy alone (HR for locoregional progression or death, 0.68; P = .005). With a median follow-up of 54 months, the median duration of overall survival was 49 months among patients treated with combined therapy and 29.3 months among those treated with radiation therapy alone (HR for death, 0.74; P = .03). Radiotherapy plus cetuximab significantly prolonged progression-free survival (HR for disease progression or death, 0.70; P = .006) In March 2006, cetuximab received US Food and Drug Administration (FDA) approval for use in combination with radiotherapy for patients with squamous cell cancer of the head and neck as well as for monotherapy metastatic disease. Combined-modality treatment was not associated with any increased mucosal toxicity or long-term differences in measured quality of life. Other agents are being investigated.

The cooperative groups are evaluating the benefit of cetuximab and conventional chemotherapy with radiation in the locally advanced and high-risk postoperative setting, since the added effects from cetuximab are low and not overlapping with cytotoxic therapy. The initial findings from RTOG 05-22 presented at the 2011 Annual Meeting of the American Society of Clinical Oncology (ASCO) did not demonstrate an improvement (median follow-up, 2.4 years) with the addition of cetuximab to cisplatin-based concurrent chemoradiotherapy, although longer follow-up and further analysis based on HPV status are still needed to confirm these findings.

Chemotherapy and survival. Numerous clinical trials have shown an improvement in locoregional tumor control using concurrent chemoradiation therapy, but a survival benefit has been observed less consistently. Recently published information from an updated meta-analysis of chemotherapy in head and neck cancer reports individual patient outcomes for 50 trials of concomitant chemoradiation therapy compared with irradiation alone. The pooled HR was 0.81, with a P < .001 and an absolute survival benefit of 8% at 5 years for concomitant treatment. In contrast, neoadjuvant chemotherapy with cisplatin and 5-FU prior to radiation did not improve survival, demonstrating the importance of timing with respect to the integration of chemotherapy in primary management.

In the RTOG 91-11 randomized trial, chemotherapy given concurrently with irradiation, or sequentially prior to irradiation, suppressed the incidence of distant metastases and improved disease-specific survival relative to irradiation alone. However, overall survival was not different among all three arms.

Concurrent chemoradiation therapy has the greatest impact on survival in the setting of unresectable squamous cell cancer, based on a head and neck Intergroup study. In this study, 295 patients with stages III and IV head and neck cancer were randomized to participate in one of three arms: (A) radiotherapy alone to 70 Gy in 35 fractions; (B) 70 Gy in 35 fractions plus concurrent cisplatin on days 1, 22, and 43; and (C) split-course radiotherapy and three cycles or concurrent cisplatin/5-FU chemotherapy, with 30 Gy given with cycle 1 and 30 to 40 Gy given with cycle 3. Grade 3 or worse toxicity occurred in 53% of arm A patients, 86% of arm B patients, and 77% of arm C patients. The 2- and 3-year Kaplan-Meier projected survivals for arm A are 30% and 20%, compared with 43% and 37% for arm B (P = .016) and 40% and 27% for arm C (P = .13). Median survival was 12.6 months for arm A, 19.1 months for arm B, and 14 months for arm C. The addition of concurrent, high-dose, single-agent cisplatin to conventional radiotherapy significantly improved survival with acceptable toxicity. In addition, concurrent multiagent chemotherapy did not offset the loss of efficacy resulting from split-course irradiation.

The impact of adjuvant concomitant chemotherapy and radiotherapy on survival was assessed in two similarly designed trials in high-risk, postoperative head and neck cancer patients that tested adjuvant chemoradiation therapy against standard postoperative radiotherapy. In the RTOG 95-01 trial, disease-free survival was significantly increased in patients treated with adjuvant chemoradiation therapy relative to those treated with radiotherapy alone, but overall survival was not.

Multiple investigators have presented data on the ability of the excision repair cross-complementation group 1 (ERCC-1) enzyme to predict response in patients treated for head and neck cancer. ERCC-1 appears to be related to resistance to platinum-based chemotherapy among other cancer sites. In one retrospective study, patients who had low levels of ERCC-1 expression had four times greater odds of benefiting from an objective response to chemotherapy (P = .01), which corresponded to a lower risk of cancer-related death. A prospective study to evaluate the role of ERCC-1 as a predictive biomarker in head and neck cancer has yet to be conducted.

Induction chemotherapy in combination with concomitant chemoradiation therapy. Concurrent chemoradiation therapy is the current standard of care for patients with locally advanced squamous cell carcinoma of the head and neck. Many think this paradigm shift has influenced patterns of failure, and distant metastasis has become more important. Docetaxel(Drug information on docetaxel) (Taxotere) in combination with cisplatin and 5-FU (TPF) was approved by the FDA in 2007 for use as induction therapy for patients with locally advanced squamous cell cancer of the head and neck.

The trial that led to this indication was a multicenter, open-label, randomized, phase III evaluation of TPF compared with cisplatin and 5-FU (PF) for three cycles prior to chemoradiation with weekly carboplatin(Drug information on carboplatin). A total of 501 patients with stage III or IV disease, deemed either unresectable or candidates for organ preservation, were randomized. Locoregional control was improved in the TPF arm (70% vs 62%). This translated into an improvement in overall survival, with median survival rates of 71 vs 30 months and an HR for survival in the TPF arm of 0.70 (P = .006). Distant failure was similar in both arms and was uncommon, observed in less than 10% of patients. Neutropenia and neutropenic fever were more common in the TPF group, but with the addition of prophylactic antibiotics, the regimen was feasible. Patient selection for this treatment approach is critical, given the toxicities and increased length of total therapy. In 2011, researchers reported long-term follow-up of patients receiving TPF vs PF, the historic standard induction therapy, showing that the trial results endured with time. With a median follow-up of 72.2 months, the median survival was 70.6 months in the TPF arm vs 34.8 months in the PF arm.

To date, however, there have been no data suggesting that this combination is better than concurrent chemoradiation alone. Physicians who oppose this approach are concerned about the risk of tumor progression on chemotherapy (10%) and, ultimately, the risk of patients not being able to initiate and/or complete the definitive chemoradiation component of this treatment. Two randomized trials evaluating these two approaches were reported at ASCO 2012. Unfortunately, the PARADIGM trial closed early secondary to poor accrual, with only 145 of the 300 planned patients. The induction regimen consisted of three cycles of cisplatin (100 mg/m²), docetaxel (75 mg/m²), and 5-FU (1000 mg/m² for 4 days). This was followed by radiation with concurrent weekly carboplatin or weekly docetaxel among the nonresponders to induction. The control arm of chemoradiation used a 6-week course of accelerated boost radiation with two cycles of cisplatin (100 mg/m²). With a median follow-up of 49 months, there was no clear difference in progression-free or overall survival. The DeCIDE trial similarly accrued slowly and was amended to reduce the planned sample size, but the trial was eventually completed. This trial, which was led by the University of Chicago, used an accelerated hyperfractionated approach with planned split course (150 cGy twice daily for 5 days followed by a week break) to a total dose of 75 Gy as the standard radiotherapy arm given with concurrent docetaxel, hydroxyurea, and 5-FU. The induction arm included two cycles of docetaxel (75 mg/m²), cisplatin (75 mg/m²) and 5-FU (750 mg/m², days 1 to 5). The use of induction therapy increased the incidence of grade 3 neutropenia and leukopenia but did not seem to impare the ability for patients to tolerate the full course of radiotherapy. The primary endpoint was overall survival, which based on the 280 subjects accrued provided 80% power to detect a, HR of 0.5. The 3-year outcomes demonstrated a decrease in distant metastases (19% to 10% favoring the induction arm; HR 0.46; P = .025), but this did translate into an overall survival advantage (75% vs 73% favoring the induction arm; HR 0.92; P = .70). Neither of these trials had prospectively assayed for HPV status among the oropharyngeal cancers, and thus this was not a factor for stratification in either study.

Adjuvant chemotherapy following surgery or irradiation. Adjuvant chemotherapy has been given following initial surgery or radiation therapy to eliminate microscopic residual disease and distant metastases. Although this approach has resulted in a reduced rate of distant metastasis, it has not been associated with improved locoregional tumor control or survival, except in a large meta-analysis. As concomitant approaches evolve and locoregional tumor control for advanced disease becomes the rule rather than the exception, the value of additional chemotherapy (as induction or adjuvant therapy) will need to be re-explored to address the problem of distant metastasis. Currently, there is interest in studying the role of targeted therapy in the adjuvant setting after definitive radiation, but this remains a research question.

Locally advanced head and neck cancer. Data from prospective trials continue to support the use of altered fractionation and concurrent chemotherapy and irradiation as an alternative to surgery or conventionally fractionated irradiation alone for locally advanced cancers of the head and neck. The long-term results of RTOG 90-03, which randomized 1,113 patients with stages III and IV squamous cell carcinoma of the oral cavity, oropharynx, or supraglottic larynx and stages II to IV squamous cell carcinoma of the base of the tongue or hypopharynx to receive one of four different schedules of irradiation alone, revealed that locoregional tumor control and disease-free survival were superior for the hyperfractionated arms and accelerated fractionation with the concomitant boost, compared with the conventional fractionated radiotherapy arm. The hyperfractionated arm demonstrated a trend toward improved overall survival.

Updated results for RTOG 99-14, a phase II prospective study designed to integrate the altered fractionated radiation regimen using concomitant boosts from RTOG 90-03 with chemotherapy, were also reported in 2005. Cisplatin was given during weeks 1, 3, and 5. Longer-term results for 76 of 84 patients were reported, and the 3-year disease-free, overall, and cause-specific survival rates were 45.8%, 57.7%, and 61.2%, respectively. The 3-year local recurrence and distant metastasis rates were 38.7% and 23.3%, respectively. Of the 35 patients who were alive, 22 had a temporary feeding tube and 17% (6 patients) still required a feeding tube at the time of their last follow-up.

RTOG 97-03, a three-arm randomized phase II trial, evaluated the feasibility of concurrent multi-agent chemotherapy with radiation therapy. In this study, three different regimens of chemotherapy were used with irradiation (70 Gy/35 fractions) for oral cavity, oropharyngeal, and hypopharyngeal cancers. Arm 1 received cisplatin (10 mg/m²) and 5-FU (400 mg/m²) via continuous infusion daily for the final 10 days of radiotherapy. Treatment in arm 2 consisted of hydroxyurea (1 g/d) and 5-FU (800 mg/m²) via continuous infusion delivered concurrently with radiation therapy. Arm 3 received paclitaxel(Drug information on paclitaxel) (30 mg/m²) and cisplatin (20 mg/m²) weekly. The incidence of acute toxicity was high, with the lowest incidence of persistent feeding tube dependence in arm 3. Arm 2 was associated with a slightly higher rate of grade 4 toxicity and treatment interruption. All arms demonstrated improved survival relative to historic controls treated with radiation therapy alone.

To diminish toxicity in locally advanced oropharyngeal and laryngeal cancers, a phase II trial using taxane-based induction followed by concurrent chemoradiation therapy was conducted through the Eastern Cooperative Oncology Group (ECOG) 2399 group. A total of 111 patients (T2, N+ or T3–4N0–3) underwent induction therapy with two cycles of paclitaxel (175 mg/m²) and carboplatin followed by concurrent weekly paclitaxel (30 mg/m²) and 70 Gy of radiotherapy. Patients with progressive or stable disease at the primary site after induction therapy underwent surgery. Patients with primary laryngeal cancers fared significantly worse using this regimen. Patients with laryngeal cancer had a lower rate of major response to induction chemotherapy, a lower 2-year progression-free survival rate, and a trend toward higher distant failure than did those with primary oropharyngeal cancer. The 2-year progression-free survival was 50% for patients with laryngeal cancer, compared with 75% for those with oropharyngeal cancer (P = .05). The improved survival among patients with oropharyngeal cancer may be due to the increased number of HPV-positive patients in this subset. The presence of genomic HPV-16, HPV-33, and HPV-35 DNA was assayed on tumor samples of patients in this study. HPV-positive tumors had a higher response rate to induction therapy (82% vs 55%) as well as to chemoradiation (84 % vs 57%), compared with HPV-negative tumors. In addition, patients with HPV-positive tumors had an improved 2-year overall survival of 95%.

Given the known differences in biology and prognosis between HPV-positive and HPV-negative tumors, current trials for locally advanced oropharyngeal cancers are studying these two groups separately. In particular, the cooperative groups are interested in whether it is possible to de-intensify treatment for HPV-positive disease, with the eventual goal of minimizing late toxicites. In ECOG 1308, patients with HPV-positive tumors underwent induction chemotherapy with cisplatin, paclitaxel, and cetuximab. Those who had a complete response in the primary site received a lower dose of definitive radiation with cetuximab. This trial has completed accrual, but results have not been reported to date. The RTOG 1016 trial, which is currently ongoing, randomizes patients to either cetuximab or cisplatin with concurrent radiation.

Determination of tumor response after chemoradiation. For patients with persistent disease after radiation, prompt diagnosis and treatment may improve survival. However, the sensitivity of the physical examination is poor as a result of the presence of submucosal disease and inflammatory changes in the upper aerodigestive tract from radiation. The best method for assessing tumor response to radiation remains under investigation. The sensitivity of CT imaging in the early postradiotherapy period (ie, within 10 weeks of the completion of radiation) is 80% for the primary site and 87% for the regional lymphatics. This modality, however, lacks specificity. FDG-PET/CT imaging has been shown to improve the positive predictive value of CT scans in determining response to radiation. A prospective study of 98 patients who underwent radiation for head and neck cancer did not support the use of FDG-PET/CT rather than CT imaging alone in evaluating treatment response early after the completion of radiotherapy. Overall, CT imaging is more sensitive when used during the early postradiation period. However, for high-risk patients (eg, current or former smokers, those with HPV-negative and non-oropharyngeal cancers), PET/CT scans enhanced the positive predictive value when compared with CT scan alone (100% in the primary site and 75% in the cervical lymph nodes compared with 71% and 37%, respectively, for CT imaging). This suggested that early PET/CT scanning may accurately identify patients at high risk for failure who may benefit from salvage surgery.

Chemotherapy for recurrent or metastatic disease. The combination of cisplatin/5-FU produces overall response rates of approximately 30% and survival rates of 6 months. In randomized trials comparing this combination with single-agent cisplatin, 5-FU, or methotrexate(Drug information on methotrexate), response rates with the single agents are lower, but survival is equivalent. However, because many practitioners believe response is a surrogate for palliation, cisplatin/5-FU has been used widely in this setting.

The taxanes are the most active cytotoxins yet identified in head and neck cancer, with overall response rates of approximately 35% in patients with recurrent or incurable disease. In one randomized trial, the combination of paclitaxel/cisplatin was compared with cisplatin/5-FU in patients with recurrent disease. A total of 194 patients who were therapy-naive for recurrent disease with an ECOG performance status of 0–1 were randomized. Efficacy outcomes were similar in the two arms, with median survival rates of 8 and 9 months for paclitaxel/cisplatin compared with cisplatin/5-FU, respectively. Paclitaxel/cisplatin was less toxic and more convenient, representing an effective and safe alternative to the traditional 5-FU–containing regimen.

Based on nearly universal expression of EGFR in squamous cell cancer of the head and neck, EGFR-targeted agents have been evaluated in patients with recurrent disease. Although the single-agent response rates of the monoclonal antibody cetuximab and the tyrosine kinase inhibitors erlotinib (Tarceva) and gefitinib(Drug information on gefitinib) (Iressa) in this setting are only in the range of 5% to 15%, their use in combination with chemotherapy, particularly cetuximab, holds more promise. Results of the EXTREME (Erbitux in First-Line Treatment of Recurrent or Metastatic Head and Neck Cancer) trial, which accrued 442 patients with recurrent or metastatic squamous cell cancer of the head and neck, showed that when compared with platinum-based chemotherapy alone, the combination with cetuximab improved median (7.4 months vs 10.1 months) and 1-year (31% vs 39%) survival rates, with an HR of 0.8 (P = .036). Similar promising, albeit preliminary, results have been reported from phase II, single-arm studies of cetuximab/paclitaxel and erlotinib/docetaxel/cisplatin. Overall, these observations lend impetus to further investigation of EGFR-targeted therapy in combination with neoadjuvant chemotherapy and with chemoradiation therapy in patients with curable disease.

Other Therapies

Photodynamic therapy

Photodynamic therapy may have some promise in the treatment of mucosal dysplasia and small head and neck tumors.

Small studies of photodynamic therapy, performed at several institutions, suggest that widespread areas of carcinoma in situ or severe dysplasia, as well as cancer, are often extirpated after photodynamic therapy. Some patients have experienced durable remissions, but the long-term efficacy of this modality remains uncertain.

Pulsed dye laser therapy for early laryngeal lesions

The pulsed dye laser initially used to treat vascular lesions has recently been investigated in the treatment of respiratory papillomatosis, dysplastic cancers, and early invasive glottic carcinomas. The efficacy of the laser is mediated through its antiangiogenic properties. The nontoxic treatment preserves the laryngeal microarchitecture and preserves or restores near-normal vocal quality. In addition, this modality can be used in an outpatient clinic setting. Preliminary findings are favorable. At this time, however, pulsed dye laser therapy should be considered investigational.

Chemoprevention

The area of chemoprevention has received a great deal of attention in recent years, and the concept of field cancerization is important in this context. As mentioned previously, this concept refers to the diffuse epithelial injury incurred by upper aerodigestive tract mucosa due to chronic exposure to carcinogens (most commonly, alcohol(Drug information on alcohol) and tobacco). These mucosal changes increase the risk of developing premalignant lesions (leukoplakia and erythroplakia) as well as multiple primary lesions.

Retinoids (vitamin A analogues) have been investigated as chemopreventive agents in the aerodigestive tract based on their efficacy in tumor models, such as the hamster buccal pouch. Retinoids mediate changes in gene expression through interaction with nuclear retinoic acid receptors, which function as transcription factors. Nutritional epidemiology studies also have indicated that low serum levels of both carotenoids and retinoids contribute to the risk of cancer development in the epithelium of the upper aerodigestive tract.

The role of retinoids in the prevention of second primary tumors in patients treated with curative intent for an early-stage index squamous cell cancer of the head and neck has been investigated in three randomized trials. The initial study was a small, single-center, phase III trial of cis-retinoic acid (1 to 2 mg/kg/d) by Hong et al. Although this study demonstrated decreased second primary tumors in the experimental arm, compliance and morbidity with this high dose were problematic. A second study using a low dose of cis-retinoic acid (10 mg/d) was performed by ECOG, and results were negative. A third, large, multicenter trial with an intermediate dose of 30 mg/d has also reported negative results in regard to second primary prevention.

Rehabilitation

Rehabilitation also is very important in the care of patients with head and neck cancer. It includes physical and occupational therapy, speech and swallowing rehabilitation, and nutritional support. For example, resection of the spinal accessory nerve, which innervates the trapezius muscle, leads to scapular winging, an inability to abduct the arm fully and, eventually, severe pain around the shoulder. These symptoms may be ameliorated with appropriate physical therapy.

Adapting to the loss of the larynx also requires intensive rehabilitation and patient motivation. Voice rehabilitation options include esophageal speech, artificial larynges (portable, battery-operated devices), and tracheoesophageal shunts.

Patients who undergo nonsurgical treatment also benefit from early swallowing rehabilitation. Intensive chemoradiation protocols are associated with prolonged feeding tube dependence in 20% of patients. Early swallowing therapy improves treatment-related dysphagia.

Nutritional support is facilitated by temporary nasoduodenal tubes or gastrostomy tubes (which impose added morbidity but are more socially acceptable and ease the patient's transition to normal activities).

Management of Symptoms and Treatment Side Effects

Photodynamic therapy. At the time of diagnosis, many patients with head and neck cancer will have lost a significant amount of weight. Maintaining adequate nutrition is a major problem for these patients, as both the tumor and treatment side effects, such as mucositis from chemotherapy and radiation therapy, may be contributory. For patients who are unable to eat or who are being treated with aggressive concomitant chemotherapy and radiation therapy protocols, placement of a gastrostomy tube is often necessary to maintain caloric intake and adequate hydration.

Pain. Clinicians must also be aware of the significant pain associated with these lesions and use narcotic analgesics appropriately to relieve discomfort.

Mucositis. The use of chemotherapy concomitantly with radiation therapy increases the occurrence of mucositis.

Nephrotoxicity and ototoxicity. For patients treated with cisplatin-containing regimens, renal insufficiency and ototoxicity are potential serious side effects.

Xerostomia. Following radiation therapy of a head and neck cancer, xerostomia may be a significant long-term side effect. In some patients, pilocarpine(Drug information on pilocarpine) has been useful in stimulating the production of saliva.

The use of organic thiophosphates, such as amifostine(Drug information on amifostine) (Ethyol), in patients undergoing radiotherapy for head and neck cancer reduces the severity of late xerostomia without compromising the antitumor activity of the irradiation.

Gastroesophageal reflux disease (GERD). Often asymptomatic or "silent," GERD is a common finding in patients treated for pharyngolaryngeal squamous cell carcinoma. In addition, cisplatin-containing chemotherapy may aggravate GERD.

Treatment of the Neck

Either irradiation alone or radical neck dissection will control metastatic squamous cell cancer to a single small neck node more than 90% of the time if there is no extracapsular tumor spread. Hence, radiation treatment may easily provide prophylactic treatment of the neck if control of the primary tumor is undertaken with irradiation. Traditionally, if the tumor in the neck was N2 or greater, or if there was tumor beyond the confines of a node, neck dissection and irradiation were combined for optimal control of the neck tumor. The need for neck dissection in patients with clinically staged N2–N3 tumors with a complete clinical and radiographic response following radiation therapy remains under investigation.

TABLE 2

Types of neck dissection and structures removed in the treatment of head and neck cancer

Types of dissection. There are several approaches to the surgical treatment of the neck nodes in patients with head and neck cancer (Table 2). This discussion will be limited to two types of neck dissection: comprehensive and selective.

• Comprehensive neck dissection—This entails complete removal of all lymphatic tissue from the neck (levels I–V). A radical neck dissection includes comprehensive node dissection with removal of the sternocleidomastoid muscle, jugular vein, and spinal accessory nerve. Modified radical neck dissection was developed to diminish the morbidity of the classic operation. The most important structure to preserve is the spinal accessory nerve.

• Selective neck dissection—This consists of the removal of lymph node groups at highest risk for containing metastases from a primary cancer. In such procedures, the lymph nodes removed correspond to the most significant drainage basins of specific head and neck tumor sites. These are staging operations usually performed in patients with clinically N0 neck cancer. If metastases are identified, further treatment to the neck will be required. A selective neck dissection should not be employed as the sole treatment of clinically palpable disease.

Sentinel lymph node biopsy for oral cavity lesions has been evaluated. Forty patients with clinically N0 neck cancer underwent sentinel lymph node biopsy followed by complete neck dissection. A sentinel node was identified in 90% of necks, with a 97% accuracy rate in predicting the nodal status of the remainder of the neck. This finding corresponded to a sensitivity of 94% and a specificity of 100%. Although these results are encouraging, they need to be validated in a larger trial. The completed American College of Surgeons Oncology Group (ACOSOG) study (Z0360) examined this technique in patients with T1 or T2, N0 oral cavity cancer. The final conclusions are pending. While sentinel lymph node biopsy may prove useful in small lesions without deep penetration, it remains investigational.

Follow-Up of Long-Term Survivors

As mentioned, head and neck cancers are aggressive tumors. The majority (80%) of recurrences will develop within 2 years. Since many recurrences are treatable with curative intent, patients should be followed closely during the months following their treatment. This period coincides with the time of greatest need from the standpoint of rehabilitation.

After 2 years, second primary tumors of the head and neck and lungs become important causes of death and morbidity. Late complications of treatment, such as radionecrosis, dental decay, radiation-induced fibrosis, hypothyroidism, and sequelae of spinal accessory nerve sacrifice or injury, may develop even after years. Complications and second primary cancers are more common in patients who continue to smoke.

Timing of Follow-Up Evaluations

Follow-up evaluations at regular intervals should be complete and should include a focused history and examination, as outlined previously. Physicians who are able to perform a head and neck examination (including laryngoscopy) should direct follow-up. After surgical treatment, this evaluation will usually require visits with the head and neck surgeon. Patients treated with irradiation should be followed by both their radiation oncologist and a head and neck surgeon or otolaryngologist.

Evaluations should be scheduled every 1 to 2 months during the first year after treatment, every 2 to 4 months during the second year, every 3 to 6 months during the third year, and every 6 months thereafter for several more years.

Imaging and Laboratory Studies

Any mucosal abnormality should be biopsied. There are no tumor markers or other useful laboratory studies to follow. Chest x-rays should be obtained yearly. There is little justification for performing CT or MRI in the follow-up of asymptomatic patients. Thyroid-stimulating hormone (TSH) should be measured yearly in patients.