Treatment of early (stage I and II) head and neck cancer: The oropharynx

INTRODUCTION Cancers of the oropharynx are subgrouped according to their site of origin including the soft palate, tonsil, and base of the tongue Cancers involving the vallecula (a fold located between the base of tongue and the epiglottis) are also classified as oropharyngeal cancers Early stage cancers of the oropharynx are generally treated with radiation therapy. Although local control is similar with either surgery or radiation as primary modalities, surgical access may be difficult in these areas. In addition, functional status is often better after primary radiation than primary surgery. Only one small randomized trial directly compared radiation alone to surgery plus radiation for patients with oropharyngeal cancer, and no treatment differences were noted between the two groups By definition, patients with stage I and II disease have no evidence of nodal involvement. The management of the neck in patients with head and neck cancer and a clinically negative neck, including the role for neck dissection, is discussed separately. Those patients who fail to achieve local control are frequently eligible for salvage therapy, which is often similar for the different sites in the head and neck.

STAGING Staging for tumors of the oropharynx is accomplished according to the TNM staging system of the AJCC Differences in five year survival rates for stages I through IV disease are illustrated by the following data derived from the Surveillance, Epidemiology and End Results (SEER) database of the National Cancer Institute between 1985 and 1991 ,

Stage I 50 percent (cause-specific survival [CSS], 57 percent)
Stage II 48 percent (CSS, 54 percent)
Stage III 38 percent (CSS, 43 percent)
Stage IV 26 percent (CSS, 30 percent)

Since these data reflect unselected patients treated in general practice, they tend to be inferior to results reported in selected patients at experienced institutions.

SOFT PALATE CANCER Early stage soft palate cancers may behave in a relatively indolent manner and remain in the early stages. Most patients do not have palpable cervical lymph nodes, but there is a significant correlation between tumor thickness and nodal disease. In one study of 39 patients, for example, none of the 24 lesions <2.86 mm were associated with cervical adenopathy, while all of the 15 lesions >3.12 mm had palpable adenopathy Patients with thicker lesions also had a worse prognosis.

Patients with early soft palate cancer are usually treated with radiotherapy delivered as external beam radiotherapy (EBRT), interstitial brachytherapy, or both. The use of prophylactic neck irradiation or observation remains controversial. The neck should be electively treated if the risk of occult metastatic disease exceeds 15 percent, given the poor prognosis with observation in this setting. If surgery is chosen for the primary tumor, then we recommend elective surgical treatment of the neck since even patients with early lesions and a negative neck are at risk for lymph node metastases. If EBRT is chosen for the primary tumor, the neck should be electively irradiated.

Radiotherapy The initial volume includes the entire soft palate and adjacent pillars and the upper neck; EBRT is given using opposed lateral fields. Some institutions have suggested that, for well-lateralized lesions, elective nodal radiotherapy can be confined to the ipsilateral neck. The boost volume can be treated by EBRT, intraoral cone (performed prior to EBRT) or, in selected cases, interstitial brachytherapy. Our approach varies with tumor size (For T1 lesions, we administer 50 Gy in 25 fractions by EBRT plus 16 Gy in eight fractions boost by EBRT or 15 Gy in six fractions boost by intraoral cone.For T2 lesions, we administer 50 Gy in 25 fractions by EBRT plus 20 Gy in 10 fractions boost by EBRT. Some centers advocate the use of interstitial brachytherapy with iridium 192 in conjunction with EBRT for small soft palate lesions since this spares the major salivary glands and reduces the risk of xerostomia The EBRT dose is usually 50 Gy in 25 fractions, followed by a 15 to 30 Gy single plane implant.

Results External beam radiotherapy produces local control in approximately 87 to 92 of T1 lesions and 70 to 75 percent of T2 lesions The five-year survival in one study of 75 patients was 83, 78, and 38 percent, respectively, with stage I, II, and III disease, respectively Adverse sequelae including soft tissue necrosis are uncommon, occurring in less than 10 percent of patients. Severe complications, principally osteonecrosis of the mandible requiring surgical resection, occur in less than 5 percent Similar outcomes have been described in patients with soft palate cancer treated with EBRT and brachytherapy Estimated five year values for disease-free and overall survival have been 43 and 57 percent, respectively.

The duration of therapy may be important with combination therapy. In one series of 370 patients with soft palate and tonsillar carcinoma (two-thirds of whom had T1 or T2 disease), the total duration of irradiation and the delay between external irradiation and brachytherapy were significant prognostic factors for local control and overall survival It was suggested that the overall duration should not be more than seven weeks: five weeks of external irradiation plus a delay of 12 to 15 days, but not more than 20 days, between external therapy and brachytherapy. At five years, patients treated with the shorter duration had higher rates of both local control (85 versus 73 percent, p = 0.01) and overall survival (59 versus 38 percent, p<0.001).

Novel schedules of fractionated high-dose-rate and pulsed-dose-rate brachytherapy, which simulate classical continuous low-dose brachytherapy, have been developed and used with some success in patients with squamous cell carcinoma of the soft palate and tonsillar fossa. In one study of patients with predominantly T2 and T3 lesions at these sites, the local relapse-free survival was approximately 90 percent, exceeding the value of 61 percent in historical controls treated with EBRT

TONSILLAR CANCER Cancers of the tonsillar pillar tend to be more superficial than those in the tonsillar fossa. Tonsillar fossa cancers more often present in advanced disease stage than either cancers of the tonsillar pillar or soft palate. They metastasize more frequently to regional lymph nodes, often with contralateral metastases Early cancers can be treated either with surgery or radiotherapy. Radiation is usually preferred because it results in excellent cure rates, provides treatment of nodes and neck at the same time, and produces a potentially better functional outcome. An additional advantage of radiotherapy is treatment of retropharyngeal and spinal accessory nodes, which are at risk for involvement particularly with tumors of the posterior pillar.

Surgery There are several surgical options for the treatment of tonsillar cancer. Historically, the typical surgical approach has involved a combined lip splitting incision coupled with an anterior midline or lateral mandibulotomy. Elective bilateral neck dissection of levels II-IV should be strongly considered because of a high incidence of bilateral occult disease.

However, the transhyoid approach to the oropharynx is being used with increasing frequency for both tonsillar and tongue base cancer (see below). This approach permits wide exposure of the superior tonsil and tongue base and avoids the potential complications of mandibular osteotomy. One study compared 19 sequential transhyoid procedures with 19 staged-matched controls who underwent resection of tongue base or tonsillar tumors via mandibulotomy The transhyoid group had significantly fewer complications related to the mandible; both groups had similar rates of positive margins and postoperative long-term dysphagia (five in each group).

Radiotherapy The initial radiation volume includes the tonsillar fossa and pillars, retromolar trigone, adjacent soft palate, and tongue and upper neck, and the radiation is delivered by EBRT. For T1 lesions, we use 50 Gy in 25 fractions by EBRT plus 16 Gy in eight fractions boost by EBRT. For T2 lesions, we use 50 Gy in 25 fractions by EBRT plus 20 Gy in 10 fractions boost by EBRT. An intraoral stent can be used to displace the tongue and reduce the dose delivered to the contralateral side. An additional boost is delivered by interstitial brachytherapy for residual palpable disease in the base of tongue.

Ipsilateral irradiation techniques reduce the acute radiation reaction in the contralateral pharynx and late damage to the contralateral salivary tissue, thereby lowering the incidence of xerostomia. Survival results are at least as good as those reported with bilateral treatment, with fewer side effects and a very low risk of failure in the contralateral neck Bilateral upper neck radiotherapy is indicated for invasion of the tongue or of the soft palate near the midline, both of which are important risk factors for contralateral node involvement.

Results Several factors contribute to the results of treatment of early stage tonsillar cancer, including the frequent presentation with more advanced disease and the preference of most centers for either radiotherapy alone or surgery combined with preoperative or postoperative radiation (in some cases with concomitant chemotherapy). The available data suggest that the outcome with surgery alone for T1 and T2 disease is comparable to radiotherapy with local control rates of approximately 80 and 70 to 75 percent, respectively With radiotherapy, local control for T1 and T2 lesions ranges from 75 to 94 percent and 70 to 78 percent, respectively In one study for example, 137 patients with squamous cell carcinoma of the tonsillar fossa were treated with EBRT Elective bilateral neck irradiation was routinely performed. Absolute local control rates for T1 and T2 lesions at a minimum follow-up of two years were 94 and 79 percent, respectively; the values for T3 and T4 lesions were lower at 58 and 50 percent. Only four patients with primary disease control subsequently failed in the neck.

In a second single institutional series of 400 patients with stages I through IVb tonsillar cancer, of whom 140 underwent planned neck dissection (130 unilateral, 10 bilateral), the five year local control rates for T1-T4 disease were 83, 81, 74, and 60 percent The five-year cause-specific survival rates by stage were:

Stage I 100 percent
Stage II 86 percent
Stage III 82 percent
Stage IVa 63 percent
Stage IVb 22 percent

In multivariate analysis, the performance of a neck dissection was a significant predictor for locoregional control.

The choice of radiotherapy alone or in conjunction with surgery as preoperative radiation or postoperative radiation or chemoradiotherapy appears to have no impact on survival. This was illustrated in a series of 384 patients with carcinoma of the tonsillar fossa of whom 154 were treated with irradiation alone (55 to 79 Gy), 144 with preoperative radiation (20 to 40 Gy), and 86 with postoperative radiation (50 to 60 Gy) Actuarial 10-year disease-free survival rates for T1 and T2 tumors were 65 and 60 percent; these values were independent of treatment modality.

Outcomes that are as good or better can be achieved with combined EBRT and iridium 192 brachytherapyThe local control for T1 and T2 tonsillar carcinomas ranged from 95 to 98 percent and 79 to 90 percent, respectively. The overall survival at five years in one series was 66 percent The reduced dose of EBRT used with combination therapy appears to be associated with fewer complications and a lower incidence of xerostomia.

High dose rate brachytherapy has also been utilized as a boost following primary hyperfractionated EBRT, an approach that shortens the treatment time and limits the volume of normal tissue exposed to high doses of external beam radiation. In one series of patients with primary squamous cell cancer of the oral cavity and oropharynx, the use of high dose rate brachytherapy was associated with manageable toxicity and local control was achieved in 87 percent of T1/T2, and 47 percent of T3/T4 tumors

As described above with cancers of the soft palate, the duration of therapy may be important when EBRT is used in conjunction with brachytherapy. It has been suggested that the overall duration should not be more than seven weeks: five weeks of external irradiation plus a delay of 12 to 15 days, but not more than 20 days, between external therapy and brachytherapy Soft tissue necrosis, trismus, and bone necrosis are other complications of radiotherapy. In the above series of 127 patients with T1 or T2 oropharyngeal squamous cell cancer who were treated with EBRT (45 Gy) and iridium 192 brachytherapy (30 Gy), soft tissue ulceration occurred in 17 patients; all healed spontaneously The most severe complications with EBRT occur when the dose to the primary exceeds 67.5 Gy, which occurs in patients with T3 and T4 lesions Severe soft tissue and/or bone complications are exceptionally rare among patients with T1 to T2 lesions treated with EBRT

BASE OF TONGUE CANCER Cancers of the base of tongue are frequently advanced at presentation due both to the silent location and the aggressiveness of the tumor. These tumors have a high propensity for metastatic spread to lymph nodes: up to 70 of patients with T1 lesions have clinically palpable disease in the neck and the risk for occult metastases ranges from 21 to 45 percent Treatment should usually include both sides of the neck, regardless of the mode of therapy, because of a high incidence of bilateral occult disease, similar to the tonsil.

Surgery and radiotherapy have been used to treat early squamous cell carcinomas of the base of tongue. These modalities provide similar rates of local control and survival. However, functional status is usually better with radiation, which is therefore the treatment of choice for early lesions in most centers

Surgery Surgery for early unilateral base of tongue cancer consists of hemiglossectomy with frozen section marginal control. These tumors have been traditionally removed by resecting the mandible or using a translabial transmandibular approach. However, these procedures involve significant morbidity including lip and chin scars, malocclusion, compromised deglutition, chronic aspiration, and altered speech articulation. As a result, efforts have been made to make surgical approaches to the oropharynx technically easier and less morbid for the patient. Several investigators have reported success with transpharyngeal approaches which allow wide exposure of the tongue base and superior tonsil and avoid the potential complications of mandibular osteotomy One study compared 19 sequential transhyoid procedures with 19 staged-matched controls who underwent resection of tongue base or tonsillar tumors via mandibulotomy The transhyoid group had significantly fewer complications related to the mandible; both groups had similar rates of positive margins and postoperative long-term dysphagia (five in each group). A retrospective report analyzed the course of 13 patients with base of tongue tumors treated with a transpharyngeal approach compared with 18 patients who underwent a transmandibular resection Patients who underwent transpharyngeal resection had significantly better speech and swallowing and less aspiration, again without a difference in survival or tumor-free margins.

Radiotherapy The initial radiation volume includes the base of the tongue, suprahyoid epiglottis, upper preepiglottic space, and bilateral upper (including the retropharyngeal nodes) and lower neck. It is treated by EBRT with opposed lateral upper fields matched with a single anterior lower field. An intraoral stent is used to open the mouth and depress the tongue. In most cases, the match (inferior border of the upper field) should be placed at the level of the thyroid notch just above the arytenoids, and a midline block should be used on the lower field to exclude the larynx and significant portions of the spinal cord and pharyngeal wall. A "low" match should be used when the lesion extends into the vallecula.

For T1 lesions, we use 50 Gy in 25 fractions by EBRT plus 16 Gy in eight fractions boost by EBRT.
For T2 lesions, we use 50 Gy in 25 fractions by EBRT plus 20 Gy in 10 fractions boost by EBRT.

The boost volume includes the primary tumor with 1 to 2 cm margins by EBRT, using either opposed lateral photons or appositional submental photons or electrons The "concomitant boost" technique may promote high rates of disease control with limited toxicity Interstitial brachytherapy can be used to boost the base of tongue. However, this approach does not appear to improve local control or reduce complications compared to EBRT alone

Results In a review of 16,188 cases of squamous cell base of tongue cancers extracted from the National Cancer Data Base between 1985 and 1991, five-year disease-specific survival rates were

Stage I 65 percent
Stage II 54 percent
Stage III 44 percent
Stage IV 30 percent

For patients with stage I and II cancers, survival rates appeared to be highest for patients treated with surgery alone, followed by surgery plus radiotherapy, and finally radiotherapy alone (75, 63, and 46 percent, respectively). However, these data must be interpreted cautiously. Because they reflect unselected patients treated in general practice, they tend to be inferior to results reported in selected patients at experienced institutions. Moreover, they are also reflective of selection bias in that patients who undergo surgery alone tend to be those with resectable disease and a more favorable prognosis.

There are no large single institution series describing the efficacy of surgery alone for early cancers of the base of tongue. Results reported in series dominated by patients with more advanced tumors suggest that local control with T1/T2 lesions is approximately 80 to 85 percent Similar findings have been reported with combined surgery and planned postoperative radiotherapyThe clinical situations where chemoradiotherapy is preferred over adjuvant radiotherapy alone are discussed in detail elsewhere. With radiotherapy, the reported rates of local control for T1 and T2 lesions are approximately 90 to 100 percent and 80 to 96 percent, respectively At five and ten years, the reported rates of overall survival for all stages of base of tongue cancer treated by either radiation, surgery or a combination have been 77 to 86 percent and 52 percent the respective values for disease-free survival have been 51 to 100 percent and 67 to 100 percent Complications of radiotherapy include insufficient ability to swallow necessitating a permanent gastrostomy, soft tissue necrosis (12 to 18 percent), bone exposure (2 to 6 percent), and rarely, hypoglossal nerve palsy (which causes dysarthria and problems with deglutition, otitis media, and trismus). In one study of 216 patients with cancer of the base of the tongue, 49 percent of whom had stage I or II disease, the incidence of severe radiation complications was 4 percent Functional outcomes and quality of life appear to be significantly better with radiotherapy compared to surgery.

One study assessed quality of life in 36 patients treated with primary radiotherapy for carcinoma of the base of the tongue, 25 of whom had T1 or T2 disease The overwhelming majority of patients achieved excellent functional status and quality of life and could maintain their prediagnosis earning potential and employment status after therapy.

The same group of investigators compared quality of life parameters (diet normalcy and speech intelligibility) in 30 patients (21 of whom had T1 or T2 disease) who underwent radiation and compared these parameters with 10 patients (five of whom had T1 or T2 disease) who underwent primary surgery for base of the tongue carcinomas The patients treated with radiation consistently had better performance status scores and quality of life than those treated with primary surgery. These differences applied to both early and advanced disease.

NEOADJUVANT CHEMOTHERAPY FOR OROPHARYNGEAL CANCER Patients with oropharyngeal squamous cell cancer may benefit from the addition of neoadjuvant chemotherapy prior to locoregional therapy. In one trial, 318 patients with all sites and stages of oropharyngeal cancer were randomly assigned to locoregional therapy (surgery or radiation) with or without neoadjuvant chemotherapy, which consisted of two or three cycles of cisplatin (100 mg/m2 on day 1) plus 5-fluorouracil (5-FU, 1000 mg/m2 over 24 hours per day, days 1 to 5) In each treatment group, 25 percent of the randomized patients had T2N0 disease, and 25 to 27 percent had stage IV disease. Despite premature closure of the trial after six years due to slowness of accrual (initial target sample size 700 patients), the use of neoadjuvant chemotherapy was associated with a significantly better median overall survival (5.1 versus 3.3 years).

Other trials that examine the role of induction chemotherapy and chemoradiotherapy in patients with locally advanced HNC, including oropharyngeal cancer, are discussed elsewhere.

POSTTREATMENT FOLLOW-UP The goal of posttreatment surveillance is improved survival through early detection of recurrent disease and identification of second primary cancers. Patients with HNC are more likely to develop second primary cancers than any other group of patients with malignancy. This probably reflects the wide distribution of the toxic effects of tobacco and alcohol, the major risk factors for HNC. The late development of second primary tumors is the most common cause of posttreatment "failure" after 36 months. The major sites are head and neck, lung, and esophagus section on Second and multiple primaries).

Despite the lack of defined survival benefit from any posttreatment surveillance strategy, surveillance protocols are in widespread clinical use after curative-intent therapy for HNC In general, the intensity of follow-up is greatest in the first two to three years, which is the period of greatest risk for disease recurrence.

The topic of surveillance after treatment for head and neck cancer is discussed in detail elsewhere.