42

Print Preview
Chapter 42
Oropharynx
Peter C. Levendag
David N. Teguh
Ben J. Heijmen

Epidemiology

The oropharynx is the posterior continuation of the oral cavity and connects with the nasopharynx (above) and laryngopharynx (below). It is located between the soft palate superiorly, and the hyoid bone inferiorly. The main sites of the oropharynx consist of the posterior and lateral pharyngeal wall, faucial arches, tonsillar fossa (TF), soft palate (SP), and the base of tongue (BOT). These structures play a crucial role in swallowing and speech. By obstructing the “air space” or by infiltrating muscles or nerves, locally advanced oropharyngeal tumors can significantly impede these functions. The same holds for intensive treatment regimen: it can cause deformities and/or impairment of particular functional (sub) units, resulting eventually in severe (late) side effects. It has long been known that patients with a history of smoking or excessive consumption of alcohol are believed to be at increased risk for developing cancer in the oropharynx (31,37). Overall these cancers comprise less than 0.5% of all cancers in men in the United States, which amounts to approximately 5000 new cases each year (319). According to the Surveillance, Epidemiology, and End Results report of the National Cancer Institute, in 2001 the age-adjusted incidence was 1.5 per 100,000 white men and 3.2 per 100,000 black men (271). These cancers more often afflict men (4:1); they are diagnosed most frequently in the sixth and seventh decades of life. Oropharyngeal cancers are readily accessible to clinical examination and staging. Historically, in the early stage and in the moderately advanced tumors, radiation therapy (RT) has been the preferred therapy mode because of its organ function-preservation properties (300,321). Most (±95%) oropharyngeal cancers are squamous cell carcinomas (SCC). Although reports can be found of other histologic subtypes (1,14,72,93,158), such as minor salivary gland tumors, lymphoepitheliomas, malignant lymphomas, mesenchymal tumors, or metastases from other extracranial tumor sites, these will not be discussed in great detail as they are considered beyond the scope of the present chapter.

Anatomy

The SP, anterior faucial pillar, and the retromolar trigone are embryologically connected to the oral cavity. However, because of their clinical behavior, tumors of these structures are preferably classified with oropharyngeal malignancies. The inferior part of the TF is referred to as the glossopalatine sulcus (Fig. 42.1).

The lateral border of the retromolar trigone extends upward into the buccal mucosa, medially it blends with the anterior tonsillar pillar. Its base is formed by the last lower molar and the adjacent gingivolingual surface. The lateral walls of the oropharynx are limited posteriorly by the TF proper and the posterior tonsillar pillar. The anterior and posterior tonsillar pillars are the folds of mucous membrane that cover the underlying glossopalatine and pharyngopalatine muscles, respectively. Deep to the lateral wall of the TF are major vessels (Figs. 42.2 and 42.3) and muscular components such as the superior constrictor muscle, the upper fibers of the middle constrictor muscle, the pharyngeus and stylopharyngeus muscles, and the glossopalatine and pharyngopalatine muscles.

Stratified squamous epithelium covers all of these structures. The tonsil has a heavy lymphoid network. The pharyngeal wall is related to the second and third cervical vertebrae. Nerve supply is from the cranial nerves IX and X. The BOT lies posterior and inferior to the palatoglossal arch. It is bounded anteriorly by the circumvallate papillae, laterally by the glossopharyngeal sulci and oropharyngeal walls, and inferiorly by the valleculae and the pharyngoepiglottic fold. Embryologically, its epithelium is derived from the entoderm, unlike that from the oral tongue (ectoderm). The body of the BOT is formed by thick muscles, the genioglossus, styloglossus, palatoglossus, and hypoglossus muscles. The muscles originate from the margins of the mandible and are attached to the hyoid bone. The blood supply and the innervation are by the lingual arteries and hypoglossal nerve, respectively.

Natural History

In general, tumors of the anterior tonsillar pillar and soft palate are better differentiated and biologically less aggressive than those of the TF. For example, 50% to 60% of patients with primary tumors in the anterior tonsillar pillar, retromolar trigone, and SP had necks with clinically negative findings, in contrast to only 24% of those with TF primaries (149). Lesions of the TF (231), retromolar trigone (39), and BOT tend to grow more extensively. Perez et al. (234) observed that the primary tumor was confined to the TF in only 5.4%. Byers et al. (171) described 14% mandibular invasion in carcinomas of the retromolar trigone. At diagnosis, 75% of BOT cancers have invaded adjacent structures, including the glossopharyngeal sulcus, pharyngeal wall, larynx, and/or faucial arches. The most common complaint at presentation of tumors in the oropharynx is pain; this pain is either attributed to (severe) mucositis, deep infiltration of the tumor or is referred (Fig. 42.4).

However, patients with primary tumors of the oropharynx can also be asymptomatic or have only vague discomfort at presentation. BOT tumors, for example, typically grow insidiously. Because the BOT is devoid of pain fibers, they are mostly asymptomatic until they have progressed significantly. With local advancement and/or with infiltration of the pterygoid muscles, patients can experience trismus and, ultimately, bleeding or swallowing problems, or can have difficulty with speech.

Diagnosis is typically established by clinical examination in the outpatient clinic and/or examination under general anesthesia, including morphologic confirmation (biopsy) of the lesion and tattooing of the clinical target volume (CTV). In the Erasmus Medical Center—Daniel den Hoed, Rotterdam (Erasmus MC), at the time of diagnosis/staging, with the patient still under general anesthesia, the lesion is frequently marked with marker seeds. This enables the extensions of the lesion to be visualized on x-ray films. From a series of patients implanted with platinum markers, we found, for example, that the TF significantly moves during swallowing and even in rest (because of respiration). Maximum excursions in rest were found to be 3.6 mm.

This type of information contributes to a more





accurate determination of the planning target volume (PTV) margin (Figs. 42.5 and 42.6).

Conventionally, platinum (Pt) or gold (Au) marker seeds were used, particularly for those patients to be boosted by brachytherapy (BT) (175). Because of significant scattering properties on computed tomography (CT), nonmetallic seeds are being tested. Panendoscopy can reveal synchronous second primaries. Ultrasound fine-needle aspiration biopsy has become an indispensable tool for pro diagnosis and for staging, especially where it concerns the lymph nodes. Multislice CT and magnetic resonance imaging (MRI) scans are now obligatory imaging tools. CT scanning with contrast enhancement using 2-mm slices is better for detecting lymph nodes and for bone detail. MRI is preferred for the evaluation of the parapharyngeal space. Axial slices are usually sufficient; sagittal MRI is helpful for detecting early pre-epiglottic space infiltration. CT combined with positron emission tomography scanning seems an extremely promising, powerful tool for diagnostic and simulation purposes, but is not yet available in every institution. Several textbooks contain helpful overviews (8,59,123,171,208,232).

Tumors are staged according to the American Joint Committee on Cancer classification system (Table 42.1) (6). Dentulous patients are at increased risk for caries and osteoradionecrosis from the reduction and qualitative change of salivary flow, change in pH, and proliferation of bacteria believed to be responsible for caries. Panorex x-ray films, identification of nonrestorable teeth for pretreatment extraction, dental trays for fluoride rinse, protection against scatter radiation, as well as education about long-term oral hygiene, should be engaged before RT and/or chemotherapy (CHT) is applied. In fact, the quite common development of osteoradionecrosis in the past (17) should be prevented by adequate measures. Finally, given the complexity of head and neck tumors, all patients should be


formally discussed in a head and neck tumor board, with or without the patient being present, before the initiation of any treatment.

Lymphatics of the Oropharynx

The lymphatic drainage of the oropharynx and the neck was first described by Rouviere (262) in 1938 and has been refined since by others (113,149). The nodal groups in the neck were originally described along the lines of lymph node chains, located in particular anatomic regions and draining specific (sub) sites. Instead of using the term jugular chain nodes, Robbins et al.
(255,256,254) proposed the “level system” for classifying the location of lymph nodes in the neck relative to surgical-anatomic landmarks (Table 42.2). The level classification (levels I to VI) was recently refined with the addition of sublevels (Ia/Ib, IIa/IIb, and Va/Vb), also using some of the radiologically defined landmarks as proposed by Som et al. (283). The probability of lymphatic (regional) metastasis is related to size and location of the primary site within the oropharynx. The order of progression of metastatic cells is systematic, that is, it usually proceeds from the upper jugular chain nodes superiorly (level I/II; first echelon), to midcervical (level III) and to lower cervical nodes (level IV), inferiorly. Candela et al. (44), for that purpose, evaluated 333 untreated primary SCCs of the oropharynx and hypopharynx and ascertained the prevalence of neck node metastasis by neck level. Isolated skip metastases occurred in only one patient (0.3%); level I or level V involvement was always associated with metastases at other levels. Another important finding is that tumors in the midline or of the posterior pharyngeal wall exhibit a higher propensity for bilateral lymphadenopathy.

Delineating and Validation of Neck Node Levels

Node-Negative Neck

From a comprehensive review of the literature regarding the incidence of regional metastasis (16,40,44,45,52,178,276,277,279), a few observations can be made. First, for most of the tumor sites in the oropharynx, when using RT one needs to irradiate the neck either electively (46 to 50 Gy node-negative [N0] neck) or boost the lymph nodes to high cumulative doses (60 to 70 Gy neck nodes). Second, in organ function-preservation RT, image-based technology has become paramount. Third, in order to adequately spare the critical healthy tissues, the clinician needs to be as selective as possible with the levels incorporated in the irradiated volumes (standardization of treatment volumes; e.g., ipsilateral vs. bilateral neck). Finally, the lymph nodal drainage per tumor site follows a predictable pattern, such that the concept of selective treatment has a legitimate rationale for surgery as well as for RT (Tables 42.3 and 42.4). Nowak et al. (222) reported on an inventory in The Netherlands, showing a lack of standardization even when experienced physicians were asked to delineate the portals designed to cover the primary and neck (Fig. 42.7). They argued that a more precise three-dimensional definition on CT of the lymph node levels in the neck allows for a better standardization of the treatment portals and, in addition, for the development and application of more conformal (selective) RT techniques. Conceptually, with the radical and modified neck dissection, all lymph nodes are routinely removed (255,256). In contrast, selective neck dissection refers to preservation of one or more of the lymph node groups (for example, see Table 42.2 for selective neck dissection levels I, II, and III).






In order to be as selective as with surgery, first the clinical (research) groups of Rotterdam and Brussels have translated the surgical-anatomy boundaries (Table 42.2) to corresponding borders on CT (108,223). The Rotterdam guidelines have further evolved into a “simplified version” for routine clinical practice (Fig. 42.8) (318). In fact, the proposed simplified delineation guidelines were based on easy-to-identify anatomic landmarks on CT, leading to a simple-to-execute delineation procedure (“learning curve”). Besides the Rotterdam system being more generous, the comparison with the Brussels guidelines revealed small but essential discrepancies (Fig. 42.9). After adjustments, a common set of recommendations for the delineation of neck node levels for the N0 neck was proposed (110). This proposal was discussed with major cooperative groups in Europe (DAHANCA, EORTC, GORTEC, NCIC, RTOG) and after some minor modifications, it was fully endorsed (Table 42.5, Fig. 42.10).

Finally, a series of clinical experiments was performed to validate the “international consensus guidelines” on the delineation of the CT-based lymph node levels (174,173). In the first experiment, the superior border of the N0 neck (cranial border level II) was determined. In 10 consecutive patients, clips were placed at the most cranial border of the neck at the time of a neck dissection. Anteroposterior and lateral x-ray films were obtained intraoperatively. The clips lined up at the caudal part of the transverse process of C1 (Fig, 42.11). Thus, the cranial border is situated much lower than the base of skull. From the position of the parotid gland, one can appreciate that this is of great importance when sparing the major salivary gland. In a second series of experiments, the neck levels I-VI in three patients first were contoured on preoperative contrast-enhanced CT scans according to the international consensus guidelines. Of each of these three patients, after placing clips at relevant surgical-anatomic–based level boundaries, an intraoperative CT scan was also obtained. The preoperative (CT-based delineated boundaries) and intraoperative (surgical-anatomic defined level boundaries) CT scans were then fused. The caudal border of level IV can be identified by the clips positioned next to traverse cervical artery (Fig. 42.12). The posterior border of level IV and Vb is determined by a virtual line drawn between the heads of the trapezius muscle (Fig. 42.12). The posterior border of surgical level IIa (spinal accessory nerve) did not fully match with posterior border of CT-based level IIa (internal jugular vein); the other surgical boundaries and CT-based contours were in perfect agreement (Fig. 42.13).

This second series of experiments was also designed to examine whether the subdivision into sublevels IIa and IIb, as suggested by Robbins (256), is of any benefit in selective RT. It can be argued that because of the heavy infestation of occult metastatic cells in the lymph channels around the IJV and carotid artery, the division of level II into radiologic sublevels IIa and IIb may not be relevant and may even be risky (174) (Figs. 42.14 and 42.15). In contrast to CT-based contoured sublevels in cases of selective RT, the division into surgical sublevels IIa/IIb makes sense as this could reduce serious morbidity, that is, could prevent potential damage to the spinal accessory nerve (174).

Node-Positive Neck

Retrospective intensity-modulated radiation therapy (IMRT) series have reported marginal recurrences in the neck of node-positive (N+) patients treated primarily or postoperatively with radiotherapy. Eisbruch et al. (74) reported a series of 135 patients treated bilaterally from 1994 to 2002 with three-dimensional conformal radiation therapy or IMRT for primary tumors located mainly in the oropharynx. With a median


follow-up of 32 months, 21 patients had a locoregional recurrence, 4 of which were marginal. Some of the marginal recurrences were observed near the base of skull above the upper limit of the delineated level II. Thus, it seems reasonable, as a first modification in case of infiltration of the upper part of level II, to include in the CTV the fatty space around the internal jugular vein and internal carotid artery up to the jugular foramen (base of skull) in the N+ neck (109). From an anatomic point of view, this space belongs to the upper most part of the retrostyloid space (Fig. 42.16). Also included are the tumor-infiltrated nonlymphoid structures. As a third modification with respect to the N+ neck, the guidelines for the caudal limits were modified. For the N+ neck, the caudal border was modified basically by lowering it down to the sternoclavicular joint. For a boost volume, the involved level and the directly surrounding uninvolved neck node levels are used (109). The optimal management of cervical metastases is still subject to considerable debate. For example, Roy et al. (263) reported on patients with TF/SP and BOT tumors and N2 neck disease. After a full course of RT, 65% of these patients still had pathologically confirmed disease in cases of clinical and/or CT-based evidence of persistent disease. For those patients with no evidence of residual disease, 33% still had remaining disease in the neck dissection specimen. Their data lend support to a planned neck dissection. In contrast, Su et al. (291) argued that for patients with oropharyngeal tumors, a neck dissection only improves regional control for those patients with a complete clinical response. In fact, a close follow-up is advocated for this category.

Postoperative Neck

In cases of postoperative RT (PORT), one relies on the long-established institutional guidelines (24,109). As an example, the routine clinical criterion for PORT as used in the Erasmus MC, Rotterdam, and those established by the University of Texas, M.D. Anderson Cancer Center, Houston, are summarized in Table 42.6. Basically, with regard to the tumor bed, if feasible, in PORT one tries to mirror the contralateral noninvolved neck. Finally, the previously discussed consensus guidelines established for the N0 neck will remain the foundation for PORT (and N+ neck).

Clinical Presentation

Patients with carcinoma of the oropharynx present most frequently with a sore throat. They also may complain of difficulty swallowing or pain in the ear, which is related to the anastomotic–tympanic nerve of Jacobson. Carcinomas of the tonsil usually are ulcerated and sometimes exophytic. They infiltrate the glossopharyngeal sulcus and BOT, many times with little or no mucosal involvement. Trismus may be a late manifestation of the disease if the masseter or pterygoid muscle is involved.




Cancers of the BOT, unlike those of the oral tongue, rarely are visualized by the patient and may grow to a large size before detection. The patient usually can point to the site of pain and the location of the tumor. Difficulty in swallowing because of pain is common, but dysphagia and impaired deglutition caused by massive infiltration of the tongue by tumor are less so. In advanced tumors that fix the root of the tongue, poor articulation is caused by impaired tongue mobility.

It is common for patients with oropharyngeal cancer to notice a mass in the cervical region, usually subdigastric (jugulodigastric), as the first manifestation of disease. Initially, distant metastases are extremely rare.

Diagnostic Work-Up

The evaluation of patients with carcinoma of the oropharynx always begins with a complete history and physical examination. The next step is a comprehensive examination of the head and neck, including oral cavity, oropharynx, nasopharynx, hypopharynx, and larynx. Mirror or fiberoptic examination of the nasopharynx, hypopharynx, and larynx always should be performed to detect any tumor extension or associated pathology. In many centers, panendoscopy has become a routine procedure because of the risk of second primaries in the upper digestive tract.

After indirect laryngoscopy, careful digital examination with a gloved finger should evaluate submucosal involvement of the glossopalatine sulcus, base of the tongue, buccal mucosa, or lateral pharyngeal wall. Direct laryngoscopy under anesthesia seldom is required for carcinoma of the faucial arch, but it is very useful to evaluate patients with larger tonsillar or BOT lesions.

Physical examination should include a thorough evaluation of the neck for detection of metastatic lymph nodes as well as a search for distant metastases. Examination of the neck should be done with the physician standing behind the seated patient. Anatomic position, size, consistency, tenderness, and mobility of palpable cervical lymph nodes should be recorded.

Histologic confirmation of a clinically suspicious malignant lesion always must be obtained, and biopsies should be performed, preferably at the margins of the tumor. Incisional or punch forceps biopsies can be performed with local anesthesia on an outpatient basis. When a lymphoma is suspected, the lesion may be submucosal; a large amount of tissue may be required for electron microscopy and immunologic typing of the tumor. Fine-needle aspiration biopsy of suspicious palpable neck nodes may be used to make the initial diagnosis.


Complete blood counts, chemistry profiles, and urinalysis should be obtained.

Plain films of the soft tissues of the neck or mandible may show involvement of soft tissues or bony structures. CT with contrast has become standard in delineating the extent of tumor and evaluating involvement of the mandible or extension into the base of the skull. MRI scans and positron emission tomography with their superior soft tissue contrast can be quite sensitive in detecting tumor extension and lymph node distributions (232). Criteria used for tumor involvement are abnormal contrast enhancement, soft tissue thickening, presence of a bulky mass, infiltration of lymphatic tissues (even without distortion of surrounding tissues), or a combination of these. Neck lymph nodes are considered pathologic when the smallest diameter is greater than 1 cm. In other studies, the presence of hypodense areas in more than 33% of the lymph nodes has been defined as nodal necrosis. Extracapsular extension is thought to be present when the nodal margin appears irregular, without clear distinction with the surrounding fat or when there is thickening of surrounding fibroadipose tissue or muscle.

Chest x-ray films should be routine. Bone scans should be requested only if bone involvement is suspected. X-ray films of the skeleton may be required in patients with positive findings on bone scans or clinical suspicion of bony lesions.

Follow-Up

Patients with oropharyngeal cancer usually are followed up with careful physical and indirect laryngoscopic examination as well as thorough cervical lymph node evaluation on a monthly basis for the first 6 months after therapy, every 3 months in years 2 and 3, every 6 months from year 3 to 5, and yearly thereafter. A recent report on 46 patients treated with irradiation showed that postirradiation CT scans may not add incremental information to the clinical examination for predicting local tumor control. Diffuse and symmetric changes of the soft tissue or asymmetry without detectable mass or less than 10 mm was associated with primary tumor control (174).

Staging

In staging oropharyngeal tumors, it is extremely important to include both the ulcerated and infiltrating components of the tumor and all of its submucosal extensions. Because of a tendency to overestimate the size of oropharyngeal tumors, a ruler or caliper must be used to measure the diameter of the lesion (160). Visual, palpatory, and radiographic findings are critical in accurate staging. The usual staging classification for carcinoma of the oropharynx, including lymph node involvement, is that of the International Union Against Cancer or the American Joint Committee on Cancer (Table 42.1) (6). A criticism of this staging system is that it is largely two-dimensional and does not take into account the third dimension, which determines tumor bulk and morphology (e.g., endophytic or exophytic lesions of similar size, which respond differently to similar treatment) (232).

Pathologic Classification

These tumors have characteristic features, including keratin in many cases, although some are nonkeratinizing; the tumors are


graded I to IV, depending on the degree of differentiation. Carcinomas that arise in the faucial arch, usually of the squamous cell type, tend to be keratinizing and more differentiated than tumors that arise from the TF.

SCCs, often poorly differentiated, account for more than 90% of cancers of the BOT. Although it has been suggested that keratinizing, more differentiated tumors have a somewhat better prognosis than others, no definite correlation between histologic type and pattern of behavior or response to therapy has been reported (232).

Lymphoepithelioma is much rarer in the tonsil or BOT (less than 1.5%) than in the nasopharynx. Most pathologists agree that lymphoepithelioma represents a poorly differentiated, nonkeratinizing SCC with a profuse lymphoid infiltration (232).

Other cell types include mucoepidermoid, adenocarcinoma, and adenoid cystic, which appear to behave more like salivary gland tumors of similar histology in other sites rather than like SCCs of similar size. Tumors of the salivary gland type are uncommon in the tonsil or faucial arch. Malignant melanomas of the TF are also rare (232).

Primary small cell carcinoma of the tonsil, with neurosecretory granules demonstrated by electron microscopy, have been infrequently described. As in other locations, this tumor has a high propensity for regional, nodal, and distant metastatic spread with a very poor prognosis.

Malignant lymphomas, usually non-Hodgkin's type, constitute 10% to 15% of malignant tumors of the tonsil and 1% to 2% of the BOT. They tend to grow submucosally and may reach large size without significant mucosal ulceration. The surface of the tumor is covered by the same mucous membrane that covers the soft palate. Primary Hodgkin's disease in the tonsil is extremely rare.

Metastatic carcinoma to the tonsil is rare, with only 92 cases reported in the world literature (232).

Prognostic Factors

Host and tumor factors have been correlated with survival but not consistently with primary, nodal, and distant relapses. Age and gender are host characteristics that may have prognostic significance (232).
In some reports, women have shown a better prognosis than men, possibly because of earlier detection of tumors in women although other authors observed no significant survival difference between the genders. Significant differences in survival based on age (e.g., patients younger or older than 40 years) have not been established in sufficiently large and uniform patient cohorts and need to be stratified by disease stage and performance status (232).

Tumor characteristics that have prognostic significance include tumor size and extension (stage); presence of palpable lymph nodes; and location, number, and size of involved lymph nodes. Tumor regression during radiation therapy and


histologic differentiation are additional prognostic factors reported. P53 and epidermal growth factor receptor overexpression has been associated with increased survival (232).

A significant correlation has been found between the stage of the primary tumor, the presence of involved cervical lymph nodes, and 5-year survival (234). Tonsil tumor extension into the base of the tongue is associated with decreased survival (232).

There is no definite correlation between histologic type or degree of tumor differentiation and patient survival. In patients treated surgically after irradiation, more than 90% with negative histologic specimens survived for 5 years compared with only 30% of patients with persistent tumor (232).

Overall, BOT cancers have a worse prognosis than do their oral tongue or tonsil counterparts because of greater size at diagnosis, more frequent spread to adjacent structures, and higher rates of lymphatic spread. However, stage for stage, they may have similar prognoses as oral tongue cancers. Small exophytic tumors (i.e., superficial surface lesions) have higher rates of local tumor control by surgery or irradiation and a better prognosis than infiltrating or large tumors. Patients with tumors confined to the BOT survive longer than those with tumors that extend to the faucial arch, oral cavity, or larynx and hypopharynx (232). The prognosis is better for patients without palpable lymph nodes (N0) and for those with small, ipsilateral, mobile lymph nodes rather than those with large, fixed, contralateral, or bilateral nodes (234).

Management Strategies, Results, and Outcomes

Lymphoepithelioma, representing a poorly differentiated, nonkeratinizing SCC with a profuse lymphoid infiltration, is rare (less than 1.5%) in the TF/SP and BOT. Mucoepidermoid, adenocarcinoma, and adenocystic carcinomas behave more or less like “salivary gland type of tumors” of similar histology in other sites of the head and neck, rather than like SCC. Moreover, these salivary glandlike tumors are particularly uncommon in the faucial arch and TF (287). Koss et al. (159) and Abedi and Sismanis (1) found only a few cases of small cell carcinomas of the tonsil. Malignant lymphomas, mostly of the non-Hodgkin's type, constitute 10% to 15% of the TF and 1% to 2% of the BOT tumors. In contrast to non-Hodgkin's lymphoma, primary Hodgkin's disease in the oropharynx is extremely rare (14). Metastatic lesions to the TF/SP and BOT are exceptionally rare as well. Oreggia et al. (224) found women to have a better prognosis than men (at 5-year overall survival [OS], 40% vs. 9%). In contrast, Vallis et al. (301) observed no significant difference between genders. Similarly, young age has not proven to be of prognostic significance (150).




Tumor characteristics like tumor size, stage, lymph nodal status, and tumor regression during therapy have been found to be of some prognostic significance. No definite correlation has been observed between histologic type, tumor differentiation, and patient survival (250,301). In contrast, p53 overexpression and targeting the epidermal growth factor receptor family do seem to be of prognostic value (267). For the more advanced tumors, there is a need for improvement, and the jury is still out on which treatment approach offers better tumor control and fewer side effects at the same time. One approach addressing the problem of improving the locoregional control in advanced tumors is reported by Hoogsteen et al. (136). They tried to modify the response of malignant cells to radiation by overcoming tumor hypoxia. The Head and Neck Oncology group of the department of Radiotherapy of the St. Radbond University Hospital in Nijmegen argued that this can be done in several ways, including, for example, hyperbaric oxygen, carbogen breathing combined with nicotinamide, hypoxic cell sensitizers, and erythropoietin. There is now compelling evidence that shows that low hemoglobin levels before and during treatment are associated with reduced tumor control and decreased survival. The authors investigated the impact of low hemoglobin levels on locoregional control in patients who have been treated with accelerated radiotherapy with carbogen and nicotinamide. This is another example of good local tumor control and survival by modulating tumor cell biology (e.g., hypoxia) (Tables 42.7 and 42.8). For BOT tumors treated with accelerated radiotherapy with carbogen and nicotinamide, Kaanders et al. (153) showed an actuarial local control (LC) rate of 84% and actuarial OS rate of 50%. Similar findings were observed for TF/SP tumors at 5 years: LC rate of 86% and OS rate of 33%. Hyperbaric oxygen as adjunctive therapy, although never proven in a randomized setting, is another effective way of modulating the oxygen status of normal tissues beneficially (92). In Erasmus MC, such a trial is currently ongoing. To measure the oxygen status of the tissues of these patients, a novel optical spectroscopic technique is used; this so called differential path-length spectroscopy allows for the in vivo measurement of hypoxia-related parameters such as blood oxygenation, blood content, and microvessel size in the most superficial layer of tissue (5). For the


advanced oropharyngeal tumors, there are many more innovative approaches feasible. For example, Suntharalingam (292) reviewed the beneficial effect of systemic and/or intra-arterial CHT as organ-preservation therapy. The author argued that the real focus in the treatment approach of the tumor, however, should be on newer biologic agents targeting cellular protein receptors. For more details one is referred to XII.

Tumors of the Tonsillar Fossa and/or Soft Palate

For general reading on TF and/or SP tumors treated by external beam RT (EBRT) and/or surgery, one is referred to the literature search as referenced in this section (3,7,14,15,18,51,66,70,87,96,97,101,102,138,145,164,170,175,189,190,193,196,203,209,225,226,233,234,246,250,268,273,301,313,316,322,320,323). The TF and/or SP tumors are also typically tumor sites very suitable for BT. For an overview on the techniques and results obtained with BT, see the following references for example (71,80,81,176,178,199,200,201,202,230,236,237,246,327). Perez et al. (234) addressed the important issue of ipsilateral and/or contralateral neck failure in a series of 384 patients treated in a single institution (Washington University, St. Louis, MO) (Fig. 42.17). A similar type of analysis was done for the 254 patients with TF and/or SP tumors treated at Erasmus MC, Rotterdam (Fig. 42.18). Interestingly, in the Erasmus MC series, 37 patients received no treatment to the contralateral clinically N0 neck; only 1 patient (3%) had a recurrence in the neck. Based on these findings, in recent years one has become more selective in treating the contralateral neck. That is, according to treatment protocol in Erasmus MC, the contralateral neck is only treated if the CTV is crossing the midline, or when dealing with N2c,3 disease, or if the TF tumor is infiltrating the glossopalatine sulcus and/or the BOT (175).





Treatment Results

T1 lesions less than 1 cm can be treated with surgical resection or irradiation to the primary only to a dose of 65 to 70 Gy in 7 weeks. The majority of T1 or T2 tumors of the TF and/or SP are treated by irradiation, the ipsilateral neck inclusive, be it electively or because of N+ disease (see previous section). For T3 and T4 tumors, surgery of the primary is often advocated; it can require removal of the primary tumor, partial removal of the mandible in combination with an ipsilateral neck dissection (combined resection). Because of the high incidence of a recurrence with surgery as a single modality (86), surgery is to be followed by PORT (23). Recent insights have demonstrated the potential beneficial effect of PORT combined with CHT (56). Because of the particular location of some of these tumors (e.g., tumor growth in the midline of the SP), surgical resection of advanced tumors can lead to a permanent functional defect (e.g., in the SP). This then needs to be repaired by reconstructive surgery; otherwise the patient is left with open nasal speech. For reasons of organ preservation, T1-T3 TF/SP tumors are therefore frequently treated by RT, albeit by EBRT alone (70 to 75 Gy) to the neck and primary or EBRT (40 to 50 Gy) to the primary and neck, followed by a boost to the primary tumor by means of low-dose-rate (LDR; total dose, 20 to 30 Gy) or high-dose-rate (fractionated HDR; total dose, 20 to 25 Gy) BT (for details on BT, see Clinical Section on Tonsillar Fossa and/or Soft Palate and Tumors of the Base of Tongue). For the advanced cases, EBRT is combined with concomitant CHT (22). Data from the literature on the surgical, EBRT only, and EBRT plus IRT results with regard to LC and survival are summarized in Tables 42.9, 42.10, and 42.11, respectively.

Tumors of the Base of Tongue

SCCs, often poorly differentiated, account for more than 90% of cancers of the BOT. It is often difficult to estimate exact tumor extension by clinical examination. Fullness in the soft tissue around the hyoid bone may be a sign of inferior penetration through the valleculae (the transition zone between the BOT and the epiglottis). Tumors in the valleculae tend to be exophytic; they frequently encroach on the lingual aspect of the epiglottis. Rarely do these tumors infiltrate the palatine tonsils. Bilateral and contralateral lymphatic spread is common; retrograde spread to retropharyngeal lymph nodes has been reported in advanced cases. Overall, patients with BOT cancers present with lymphatic metastasis in 50% to 80%, with the jugulodigastric and parapharyngeal nodes most commonly involved. Bilateral spread is observed in 37% to 55% (179,227).


The incidence of pathologic lymph nodes (pN+) in the ipsilateral clinical N0 neck is estimated to be 22% to 33%. Contralateral lymphatic metastasis at presentation is observed in 37%, albeit by RT or surgery. These data testify to the fact that in BOT cancer, the neck should be treated electively bilaterally (N0; levels II-IV), or therapeutically (N+; levels I-V). An overview of the pertinent literature on BOT cancer can be obtained from references 15,44,66,87,101,119,123,124,127,128,129,135,136,141,143,148,155,163,168,169,184,189,192,193,205,241,245,249,274,281,284,292,298,311.

Treatment Results

With regard to the management of BOT tumors, in short, the primary tumor in early-stage oropharyngeal cancers can be treated by either EBRT or IRT or surgery, whereas more advanced lesions often are treated by surgery plus PORT. Also EBRT, followed by a boost by IRT or intraoral cone and/or concomitant CHT for the more advanced cases, is frequently used. In the majority of institutions, however, RT is the preferred definitive treatment mode for T1, T2, and some of the exophytic T3, N0, N1 cancers. In general, a neck dissection is warranted only in these early cancer stages in patients treated by RT and experiencing a residual regional mass 6 weeks after completion of the therapy. In this respect, Doweck et al. (71) discussed the controversial role of selective neck dissection after definitive RT. For N1–3 disease, some protocols have successfully used routine neck dissections after preoperative RT (46 Gy), with excellent control rates. For the more advanced (endophytic) T3 lesions, as well as for the T4 tumors with significant extension into normal surrounding tissues, organ/normal tissue deformities are frequently the cause of clinical problems, for example, resulting in swallowing disability and trismus. For this category of patients, the treatment is frequently “tailor made” and surgery followed by PORT might sometimes be a more sensible option (299).

Reviewing the literature, however, the implementation of concomitant CHT has also been shown to be a highly effective treatment combination. BOT tumors may be resected transorally or via mandibulotomy; the last approach is frequently combined with reconstruction by tissue grafting (272). Patients with advanced tumors may require a glossectomy. In these cases, a tracheotomy (to avoid aspiration) with placement of a speech button and a percutaneous endoscopic gastrostomy to circumvent swallowing dysfunction (thus to secure adequate food intake during treatment and immediate follow-up) is often performed at the time of surgery. The relevant data taken from the literature with regard to locoregional tumor control and survival has been summarized in Table 42.12 (for surgery), Table 42.13 (for EBRT), and Table 42.14 (for EBRT combined with a BT boost).

Finally, two typical protocols, exemplified by Figures 42.19 and 42.20, illustrate different treatment approaches, but similar (good) LC and survival for oropharyngeal tumors. Figure 42.19 represents the oropharynx protocol of Erasmus MC, with emphasis on organ function-preserving properties by using accelerated fractionation during the first series of IMRT (6 fractions per week) and HDR-BT, or Cyber Knife as a boost technique. Figure 42.20 illustrates the protocol of the M.D. Anderson Cancer Center, with the main focus on the concomitant boost (altered fractionation) technique.

Tumors of the Lateral and Posterior Pharyngeal Walls

These tumors are less frequently reported in the literature and generally do not do so well with either RT or surgery





as opposed to the TF and/or SP tumors or the tumors of the BOT. For a short bibliography see references 53,59,63,83,144,152,191,206,208,242,285,293,295,326.

Treatment Results

Guillamondequi et al. (112) found 28% recurrences after surgery, with salvage in less than one third of the patients. Fein et al. (81) at the University of Florida compared retrospectively once-daily versus twice-daily fractionation. The observed LC rates were 100% versus 100% for the T1 category, 67% versus 92% for the T2 tumors, 43% versus 80% for T3 tumors, and 17% versus 50% for T4 tumors. Meoz-Mendez et al. (207) reported on a mixed group of patients with hypopharyngeal and pharyngeal wall carcinomas treated in the M.D. Anderson Cancer Center: the LC for T1 was 91%; for T2, 73%; for T3, 61%; and for T4, 37%. Those treated with surgery and PORT or preoperative RT fared better (LC, 75%) as opposed to RT alone (LC, 51%). A study by Marks et al. (194,195) compared preoperative RT with definitive RT. There was no difference in LC, but significantly more grade III/IV complications in the surgery group. Spiro et al. (286), from Memorial Sloan Kettering Cancer Center in New York, reported on 78 patients with posterior wall carcinomas. The cumulative 5-year survival was poor: 32%. Good results, albeit in a small selected series of patients, were obtained with an I125 or Ir192 implant; there was an LC rate of 82% at 5 years (284).


In general, the locoregional outcome and survival is significantly better for the early T1, T2, and T3N0,1 carcinomas as opposed to the (endophytic) T3,T4 and N2,3 tumors. Although RT alone most likely confers less functional impairment than is the case with surgery, surgery followed by PORT remains a valuable treatment option for advanced tumors.

Different surgical approaches have been proposed for the primary tumor (see Chapter 17 in Harrison et al. [126]). A (bilateral) modified neck dissection is also included in the treatment approach of these difficult-to-manage malignant tumors. Posterior pharyngeal wall tumors in particular pose a technical problem when one needs to deliver high doses of definitive radiation to the primary tumor because of the proximity of the spinal cord. Grimard et al. (111) described an elegant technique for radiating these tumors without compromising the spinal cord tolerance by using two posterior arcs with closure of one jaw beyond the central axis. The initial target volume encompasses the primary tumor and the bilateral neck levels II-V, with the parapharyngeal and retropharyngeal lymphatics inclusive. Finally, results in terms of tumor control, survival, and (severe) complications are summarized in Table 42.15.

Recurrent Disease and Salvage

The management of a locoregional failure in the head and neck remains a formidable challenge. Most recurrences (80%) manifest in the first 2 years following primary treatment. At least 50%





of patients who die from uncontrolled disease have local and/or regional disease as their sole site of failure. Moreover, the majority (80%) of those who develop distant metastases also have local and/or regional failure. Another, related clinical entity is the management of second primary tumors (about 3% per year [161,220]) occurring in previously irradiated regions. Selected patients with locoregional recurrences can be successfully salvaged with surgery and/or RT. Treatment options are more limited if initial treatment consists of surgery combined with RT or high-dose RT. The average cure rate of these patients has been reported to vary between 30% and 40%, and most failures are due to locoregional relapses. The use of surgery alone as a salvage procedure in case of recurrent BOT cancer was reported by Pradhan et al. (242). In approximately one third of the patients, LC was achieved for the duration of 1 year. Thirty-five patients required a total glossectomy. The role of RT is not widely appreciated as yet, mainly because of concerns about the tolerance of local tissues to reirradiation. In this regard, BT plays a crucial role (high-dose, small volume, rapid dose fall-off) (140,165,197). An equivalent EBRT dose of 60/2 Gy by five fractions per week is being applied mostly as the reirradiation dose schedule. An important prognostic factor favoring long-term LC is an interval of more than 1 year between the radiation courses. Langlois et al. (165), for example, report on 123 patients treated for recurrent cancer or a new cancer of the tongue or oropharynx, arising in previously irradiated volumes. The actuarial LC rate was 67% at 2 years and 59% at 5 years. Levendag et al. (176) analyzed a 13-year experience with reirradiation. An improvement in LC was observed (50% vs. 29%) for the EBRT plus IRT as opposed to the EBRT-alone series. The improvement in LC was typically not reflected in a survival benefit; that is, an actuarial OS of 20% at 5 years was observed in both series. Mazeron et al. (200) had similar results: actuarial LC was 72% at 2 years and 69% at 4 years. Although LC of the tumor was achieved in the majority of these patients, only 14% remained alive at 5 years. Best results were achieved in lesions of the faucial arch and posterior pharyngeal wall (LC, 100%).

Other ways of applying reirradiation is in an intraoperative setting for residual microscopic disease by means of a silicone flexible intraoperative template (Fig. 42.21). After the dose of 10 Gy (prescribed mostly at 1 cm from the afterloading catheter


in the flexible intraoperative template) has been delivered, the surgical defect can be closed by a reconstructive procedure using “fresh,” that is, donor tissue (e.g., deltopectoral flap) that has not been previously irradiated. The dose is mostly prescribed to a distance of 1 cm (157); subsequently, a course of fractionated EBRT is applied as an outpatient procedure (e.g., 26 × 1.8 Gy). As with any retreatment situation, the complication rate is substantially higher than with primary therapy.

Radiation Therapy Techniques

Conventional Radiation Therapy

Irradiation portals for oropharyngeal cancers should encompass the primary tumor and its local and regional “extensions,” with a margin for the CTV (approximately 0.7 cm) and for the PTV (approximately 0.5 cm). The concept of regional coverage has been eluded to before extensively. Patients are generally treated in the supine position with bite-block and thermoplastic mask immobilization, with daily treatment of all fields. Neck portals should extend superiorly until C1 for N0, and the base of skull (retrostyloid space) in case of N+ disease. Patient examples (T2N2b BOT tumor and T2N2b TF/SP tumor) with regard to the geometry of portals, treatment techniques used, and dose distributions are shown in Figures 42.22, 42.23, 42.24, 42.25, 42.26, 42.27, 42.28, 42.29, 42.30, 42.31, 42.32 and 42.33. In the examples presented in Figures 42.22, 42.23, 42.24, 42.25 and 42.26, the primary tumor and both sides of the upper neck are irradiated using a conventional lateral parallel-opposed technique for the upper neck in case of a T2N2b BOT tumor. Both sides of the lower neck are generally irradiated through a single anteroposterior field, sometimes with a midline block. In order to prevent overdose


at the junction line, a junction zone of 1 cm between the lateral fields and the anteroposterior portals is treated daily in Erasmus MC with maximum or minimum field sizes (the so called slip zone). If appropriate, the midline block shields the larynx and spinal cord. The spinal cord is shielded after administration of 46/2 to 50/2 Gy and if indicated, the posterior cervical triangles are boosted with 10-MeV electrons, therewith sparing the spinal cord. Tissue compensators (wedges) are used to ensure dose homogeneity in the lateral portals and to prevent excessive dose to the supraglottic larynx. After 46/2 Gy with 4- to 6-MV photon beams has been applied, the remaining dose can be delivered with high-energy photons (15 to 18 MV) in order to reduce the dose to the parotids, the mandible, and/or the temporomandibular joints (buildup). The middle ear and inner ear should be carefully shielded posteriorly. Small tumors of the TF, anterior tonsillar pillar, and retromolar trigone can be treated by ipsilateral wedged anterior and posterior fields or with BT. With the wedge technique, the dose to the mandible is high, and a greater incidence of complications (e.g., soft tissue necrosis and osteonecrosis of the mandible) can be anticipated. Limiting irradiation to the ipsilateral neck reduces the probability of xerostomia (18). This approach was confirmed by O'Sullivan and Grice (225), who reported a 3-year tumor control rate of 77% and cause-specific survival rate of 76% in 228 patients with carcinoma of the tonsillar region treated with ipsilateral-only RT techniques (oblique wedge pair arrangement). Contralateral neck failure was observed in only eight patients (3.5%). Levendag et al. had similar observations (Fig. 42.18). Contralateral failure in the untreated neck was only 3% (175).


The intraoral cone technique, using orthovoltage or electrons, has been used selectively in the treatment of patients with small lesions. Adequate tumor coverage is aided by CT-based treatment planning (with or without MRI fusion); moreover,


at the present time it is reasonable to consider CT-based treatment planning for head and neck cancer more or less obligatory. Many of the previously described techniques have therefore substantially changed since the introduction of new and innovative technology, such as IMRT, three-dimensional conformal radiation therapy, and cone beam CT (see also dedicated section XIII on IMRT). Figures 42.27, 42.28, 42.29, 42.30, 42.31, 42.32 and 42.33 represent a TF and SP tumor treated by IMRT techniques. The figures depict adequate target coverage and maximum effort to spare major salivary glands. We have compared, for bilateral irradiation, bilateral sparing of parotid glands (Fig. 42.27), as opposed to maximum sparing of contralateral parotid glands (Fig. 42.28), or reducing CTV contralateral side (Fig. 42.29). The corresponding dose-volume histograms are depicted in Figures 42.30, 42.31 and 42.32. Also, using this technology, new concepts can be incorporated in future treatment protocols. For example, Thorstad et al. (298) from M.D. Anderson Cancer Center, report favorable results for SCC of the oropharynx treated with IMRT. Multivariate analysis showed that the GTV (primary tumor ± nodes) became an independent risk factor determining locoregional control (GTV <50 mL LC ± 90% vs. GTV >50 mL LC ± 20%; p <.0001). From this type of data, the authors concluded that selecting “high-volume” patients for aggressive treatment protocols might be warranted. For more details, see IMRT section.

Dose and Fractionation Primary Sites in Oropharynx

In general for T1–2 lesions, doses of 66 to 70 Gy in 6.5 to 7 weeks with conventional fractionation (1.8 to 2.0 Gy per fraction daily, six fractions per week) are recommended for definitive radiotherapy. However, for T3–4 oropharyngeal cancers, several studies have demonstrated better locoregional control when either accelerated or hyperfractionated regimens are used with concurrent CHT (41,88,91,155). (For details on altered fractionation, see section Chemotherapy, Targeted Therapy, and Altered Fractionation Regimens). For locally advanced lesions in the head and neck, in recent years the addition of concurrent CHT to either conventional or altered fractionated radiation has shown to be beneficial compared to radiation therapy alone.

Side Effects of Conventional Treatment Techniques

Normal Tissue Toxicity Profile—Acute Effects

The major sequelae of RT can be divided into acute and chronic side effects. They are multifactorial. The potential acute effects on the oral cavity and pharynx after approximately 1 to 3 weeks of RT include mucositis (ulcer), sore throat, loss of taste, and xerostomia (if any of the major salivary glands are in the treatment portal). Approximately 5% of patients develop sialadenitis within 24 hours of the first irradiation treatment, but this usually resolves within 24 to 48 hours. The skin experiences erythema, peeling, and pigmentation. If the capacity of the basal cell layer to repopulate the epidermis is overwhelmed, the result is moist desquamation. Likewise, epilation of hair-bearing areas with accompanying loss of sweat and sebaceous gland function occurs.

Normal Tissue Toxicity Profile— Late Effects

The late effects after definitive RT can include xerostomia, dental caries, altered sense of taste, swallowing problems, dysphagia, altered quality of voice, lymphedema, hypothyroidism,

epilation, trismus, cervical fibrosis, atrophy of the mucosa and skin, as well as soft tissue and bone necrosis. In a Rotterdam series on oropharyngeal tumors 25% grade III/IV mucositis (“pinpoint ulcer,” 47/190) and 10% trismus (19/190) were reported. In the process of osteoradionecrosis, radiation is believed to exert an avascular effect on tissues and epithelia that are thinner and more susceptible to injury. The process usually starts with ulceration of soft tissues, which can progress to bone exposure. For refractory cases, hyperbaric oxygen treatment has been advocated. Factors that can influence osteoradionecrosis include elective dental extraction after RT and treatment of tumors near bone. In the modern era, osteonecrosis should be an uncommon event (<5%) (17). Technique could also play a role, the BT nonlooping technique being associated with a higher reported injury rate than the looping technique (124,191). Of these late side effects, xerostomia is the most prominent (317), and will be discussed in more detail in the next paragraph. Dysphagia and trismus, although clinically important to prevent, are still somewhat underscored. These are given a prominent place in the discussion involving quality of life (QOL) in the section Performance Status Scale, Socioeconomic Outcomes, and Quality of Life. Finally, clinical reports on late side effects are summarized data in Table 42.16 (9,99,121,145,175,184,203,226,227,234,285,320).

Xerostomia

Given the way most patients were treated in the (recent) past, frequently using nonsparing parallel-opposed techniques, xerostomia seems to be the overriding side effect. Roesink, et al. from the Utrecht Medical Center reported important observations on the dry mouth syndrome, in particular related to the dose-effect relationship of the major salivary glands (Fig. 42.34). Irradiation of the salivary glands is obviously associated with loss of function, quantitatively and qualitatively, thus among other things resulting in a reduction in salivary flow and consequently dryness of the mouth. Moderate-to-severe xerostomia occurs in more than 75% of patients treated with conventional lateral beam arrangements. The best definition for objective parotid gland toxicity appeared to be reduction of stimulated output to <25% of the preradiotherapy output (260). Two dose-response curves for stimulated parotid saliva flow rates obtained from relatively large patient groups are available (76,259) (Fig. 42.34). Both studies conclude that the mean dose to the parotid gland best predicts its function after radiotherapy. The steepness of the dose-response curve and the TD50 value at 1 year after irradiation differ. However, we can conclude from these studies that it is rather safe, in terms of preservation of stimulated parotid gland function, to have a mean parotid gland dose of less than 25 Gy. When a mean dose is reached above 50 Gy, nearly all patients will have a severe decrease in parotid flow rate.

It is generally accepted that IMRT is a valuable tool for reducing the dose to the parotid gland. Several studies report on salivary flow after IMRT for oropharyngeal tumors. However, clinical studies that objectively demonstrate and quantify the advantages of IMRT compared with conventional beam arrangements are scarce. Chao et al. (49) found a correlation of the parotid flow ratio with the mean parotid dose, and a lower mean parotid dose in 27 IMRT patients compared with 14 patients treated conventionally. IMRT versus conventional treatment, however, did not independently influence the functional outcome of the salivary glands in this study. Roesink (259,260) and Terhaard et al. (297) prospectively evaluated a total of 56 patients with oropharyngeal cancer. Of these, 26 received conventional radiotherapy and 30 patients were treated with IMRT. The mean dose to the parotid glands was 48.1 Gy for

the conventional treatment and 33.7 Gy for IMRT. As a result, 6 weeks after treatment the number of parotid complications was significantly lower after IMRT (55%) than after conventional radiotherapy (87%).

There are several studies using toxicity scoring systems instead of saliva measurements. Eisbruch et al. report on parotid function after conformal radiotherapy and IMRT (76). In a matched case control study with a low number of patients treated with standard radiotherapy, the QOL scores of patients treated with IMRT improved over time after irradiation, and no improvement was seen in patients treated with standard radiotherapy (144). In the studies of Chao et al. (50), the dosimetric advantage of IMRT compared with conventional techniques did translate into a significant reduction of late salivary toxicity in patients with oropharyngeal carcinoma. One has to keep in mind that the submandibular glands also play a major role in producing saliva: in resting state, 70% of the saliva production is believed to be generated by the submandibular glands. The submandibular glands as well as the minor salivary glands are given more attention in clinical research at the present time, but much research on role of these structures still has to be performed. However, sparing of the submandibular gland in oropharyngeal cancer is extremely difficult in case of bilateral treatment of the neck. One possible future avenue is to routinely transfer the submandibular gland ventrally in the contralateral node-negative neck.

Performance Scale Status, Socioeconomic Outcomes, and Quality of Life

Performance Status Scale and Socioeconomic Outcome

Health-Related Quality of Life

Several questionnaires have been developed to assess health-related quality of life (HRQOL) for head and neck cancer patients (27,30,46,314). Each of the side effects can have a different impact on the HRQOL (28,29,105,106,115,116), varying from changes of speech and voice quality to impact on well-being and HRQOL in a broader sense. Interestingly, Nordgren et al. (220,221) evaluated the HRQOL of patients with pharyngeal carcinoma at diagnosis and after 1 and 5 years in a prospective multicenter study. Again, the HRQOL at diagnosis seems to be an important factor for the prognosis of both HRQOL over time and survival.

Trismus and Quality of Life

Trismus, severely restricted mouth opening, is a common problem in head and neck oncology. According to Dijkstra et al. (67,68), it is present at the time of diagnosis in approximately 2% of patients due to tumor growth; in tumors originating from or in the parapharyngeal space it is even more frequent (55%). Additionally, another 8% increase in trismus is due to treatment per se, be it surgery or RT (312). One of the reasons for this variation in reporting is the lack of uniform criteria. Dijkstra et al. proposed to use as a cut-off point for mouth opening 35 mm, irrespective of dental status, but they acknowledge that differences per subgroup may exist. The paired mastication apparatus facilitates opening of the mouth; it consists of the processes coronoideus and the condyle of the mandible, as well as the muscles responsible for jaw movement. The functionality of this muscular compartment (95) can be summarized as follows: depression (lateral pterygoid, gravity), elevation (temporalis, masseter and medial pterygoid), protrusion (lateral pterygoid, masseter, temporalis), retraction (posterior fibers temporalis, deep fibers masseter), and lateral movement (contralateral lateral pterygoid, bilateral temporalis). Surgery and RT may induce trismus by causing fibrosis of one of the aforementioned muscles. Fibrosis might significantly impact QOL of the patient

as it can affect the phonation, nutritional status, and dental hygiene of the patient (248). The development of some of these late effects typically depend on factors like previous treatment, total dose, fractionation, irradiated volume, and treatment techniques. Dijkstra reported mandibular function impairments in 18% of 89 patients with cancers in the oral cavity and oropharynx. In the Rotterdam series, the incidence was only 1% for the treatment group EBRT plus BT; for the surgery plus PORT series it amounted to 21% (67).

Ways to counteract this often long-lasting problem of trismus are mechanical appliances to reduce the severity of fibrosis (161), hyperbaric oxygen (156), pentoxifylline (54,312), surgical corrective measures (48), and IMRT. Figure 42.35 depicts an axial CT slice on which the relevant muscles for jaw movement are delineated, the processes coronoideus and condyle of the mandible inclusive. One may decrease the dose to the masseter muscle significantly with IMRT by putting a constraint on the masseter muscle (Fig. 42.36).

Speech and Quality of Life

For technical treatment planning reasons, voice, and speech can be affected at the time of the actual treatment of cancers in the oropharynx and during the follow-up period. In a recent article by van Gogh et al. (104), the authors also concluded that deviant voice quality can also lead to limitations in social life (302,303). A robust, short, five-item questionnaire was suggested to be able to detect voice deterioration and differentiate this in a busy outward clinic from a cancer in the larynx (104,152).

Dysphagia and Quality of Life

Swallowing function may be affected adversely by surgical and nonsurgical treatment of advanced oropharyngeal cancer (281). Gastrotomy tube (G-tube) dependence 6 to12 months after surgical management varies in the literature between 6% and 39% (75,280). Rates of swallowing dysfunction after chemoradiation are less well defined; G-tube dependence varies between 13% and 64% at short-term follow-up and between 13% and 33% at long-term follow-up (94,107,185,215,216,221,265). In general, after long-term follow-up (>1 year), one third of patients were reported to be G-tube–dependent (BOT 67% vs. TF/SP 25%; p = .049). The swallowing apparatus, being the wall of the pharynx (Fig. 42.37), is composed of two layers of muscles: the external three constrictor muscles (superior, middle, and inferior constrictor [with its cricopharyngeal and thyropharyngeal part]), the circular fibers of esophagus inlet, and the internal longitudinal levator muscles (stylopharyngeus and palatopharyngeus muscles). Deglutition or swallowing is a complex act of these seven muscular structures.
A study was recently initiated in Erasmus MC to get more insight in the problem of dysphagia. First, the components of the swallowing apparatus were determined and delineated on CT. After delineation, dose-volume histograms were constructed and mean doses calculated for every muscular structure. Fifty-five patients with cancer in the oropharynx who were treated


between 2000 and 2005 in Erasmus MC were used to study the problem of dysphagia in more detail. All patients were asked to respond to validated questionnaires PSS, EORTC H&N35, and the M.D. Anderson Dysphagia Inventory (Fig. 42.38). Using a univariate ordered logistic regression analysis technique, it was found that the probability for having serious complaints with swallowing increases significantly with dose (Fig. 42.39), but interestingly, this was significant for the superior and middle constrictor muscles. A multivariate analysis showed that the only significant factor was BT (dose). However, given the tight enveloping nature of the deglutition musculature, it needs very sophisticated three-dimensional treatment planning to spare the constrictor muscles without compromising on the dose to the primary tumor.

Brachytherapy

The history of BT dates back to the beginning of the 20th century, when the first BT procedures were performed using Radium-226 needles. Brachytherapy (“brachy” = Greek for “short”) is a treatment modality in which the tumor is irradiated by positioning the radioactive sources very close to (mould or endocavitary techniques) or even inside the tumor volume (interstitial implant), either by permanent (seed) implant or by temporarily inserted applicators or afterloading catheters. In principle, BT is a conformal type of radiation therapy technique. In recent years, artificial radionuclides such as Cs137, Co60, I125, and Ir192 have become available. Manual afterloading of the sources into applicators or afterloading tubes replaced direct loading of sources into the patient. The French developed the so-called “Paris system” for low-dose-rate dosimetry purposes; that is, for parallel-equidistant sources, the system suggests specifying the dose of the implant as being 85% of the average dose in the basal dose points (local minima). A similar type of dose prescription is used for HDR BT (Figs. 42.40 and 42.41). Also, computer-controlled afterloading devices, supported by sophisticated treatment planning software with optimization capabilities, became available. The BOT implant consists of afterloading catheters after the percutaneous introduction of trocars in a submental or submandibular approach (Fig. 42.42) (103). For patients with disease extension toward the pharyngoepiglottic fold, lateral loops are added. The spacing between each end of the “looping” catheters running over the dorsum of the tongue is ±1 cm. As a safety precaution, when removing the implant, a temporary tracheostomy is sometimes performed in patients immediately before the implantation. A typical case for fractionated HDR TF and SP implant is depicted in Figure 42.41. In the majority of cases, two to three catheters are implanted in TF and faucial arches (Fig. 42.41, inset). A temporary nasogastric feeding tube is placed at the completion of most of our BT procedures. The development


of radiobiologic models enables one to predict to a certain extent the tumor control probability and normal tissue complication probability after the application of BT, much depending on factors such as fraction size, dose rate, the tumor, and the normal tissues one is dealing with. Temporary BT has been used with several dose-rate categories.

The French have published extensively on interstitial radiation therapy of TF and/or SP tumors, as well as on cancers of the BOT (32,60,69,79,86,98,139,175,201,229,235,236,249,274,327); most of these data regard LDR implants. For example, Mazeron et al. (199) report on a subset of patients with early-stage (T1, T2) tumors of the TF and/or SP, with a LC rate of approximately 85%, that is, a regional control rate of 97% for N0, and 88% N1–3 disease. Patients were typically treated by 45 Gy EBRT and a 30 Gy LDR Ir192 boost. Soft tissue ulceration occurred in 17 patients. Similar locoregional control rates were reported by Pernot et al. (235,236) (LC LDR boost TF/SP T1, T2N0 tumors 90% vs. T1, T2N1–3 86%) and Levendag et al. (175,177) (TF and/or SP tumors LRC 87% at 5 years). The series of patients in Rotterdam were treated by fractionated HDR BT (daytime regimen) or PDR (24 hours regime) (Table 42.18). Esche et al. (79) described 43 patients with carcinoma of the SP and uvula with LC rate of 92%. Overall survival was 60% at 3 years and 37% at 5 years. The cause-specific survivals were 81% and 64%, respectively. The leading cause of death was other aerodigestive cancers (these cancers occur with an actuarial rate of 3% per year posttreatment). The “BT school” of Memorial Sloan Kettering Cancer Center in New York pioneered large-volume implants in particular for cancer of the BOT, a technique initially designed by Vikram and Hilaris (307) and Vikram et al. (308). Harrison et al. (120,121,125,127) elaborated on cancer of the BOT and also related outcomes to QOL. Some of the control rates with IRT can be taken from Table 42.11 (TF/SP) and Table 42.14 (BOT). In skillful, well-trained, hands, BT remains an extremely gratifying technique for applying high doses of radiation for small-volume disease located in the midline (e.g., SP tumors) with (in case of fractionated HDR) highly conformal and accelerated properties.

Finally, IRT can also be a very rewarding technique, given the high doses in small-volume disease and the rapid dose falloff in the treatment of recurrent cancers and/or in case of reirradiation (58). For the future, image-guided BT will become routine; summation of dose distributions of BT and EBRT will become mandatory (Fig. 42.43). Moreover, by the development of soft x-ray sources and afterloading machines that carry multiple sources and have multiple drives, the flexibility of intraoperative BT has increased. One of these sources that is currently being tested is yyterbium (169).

Chemotherapy Targeted Therapy, and Altered Fractionation Regimes

Concurrent CHT and altered fractionated irradiation have shown independently to improve the outcome for head and neck cancer patients. The combination maximizes the chance for preservation of organ function and has the potential to improve the results even more by integration with new biologic agents (10,36,187,206,261,264,288). Many of the hyperfractionated and/or accelerated schedules have resulted in improved locoregional control. Concomitant CHT appears to result in improved LRC and OS, in contrast to neoadjuvant CHT and maintenance CHT (22,42,217). Not infrequently, these treatment regimes increase toxicity as well. Finally, the role of intra-arterial CHT (206,213,257) as well as the benefit of induction (neoadjuvant) and adjuvant CHT remains to be determined. In their concise review on randomized trials concerning multimodality treatment approaches, Bernier and Bentzen (22) emphasized that, to maximize outcome, each of the components of a particular treatment regime needs to be optimized separately. Importantly, Benasso et al. (20) and Taylor et al. (295) conducted multivariate analyses of patients treated in chemoradiotherapy head and neck trials, and pointed out that the second most important prognostic factor is the experience of the Center.

Regarding the effects on OS and locoregional control by altered fractionation and/or concomitant CHT, in 2004 Rosenthal and Kian (261) made some recommendations for treatment selection: conventional fractionation (and dose) for T1 and favorable T2N0,1 tumors, altered fractionation for unfavorable T2 or exophytic T3N0,1 (with or without neck dissection in case of N2,3 disease), and concurrent CHT for the more advanced cancers. Meanwhile, toxicity amelioration and identification of predictive biomarkers and effective molecularly targeted therapy should be pursued (261,292). Salama et al. (264)


published on aggressive trimodality treatment for the subset of patients with recurrent and/or second primary cancers in the head and neck. They evaluated 115 patients treated with a median lifetime radiation dose of 131 Gy. The locoregional control, OS, and freedom from distant metastasis rate at 3 years were 22%, 51%, and 61%, respectively. However, of note is that 19 patients died of treatment-related toxicity, 5 of these because of carotid blow-out. Suntharalingam (292) reviewed the early trials in 2003.

Recognizing the mostly nonspecific nature of the toxicities of healthy tissues consequential to combined modality therapy, he argued that the real focus should be on researching newer biologic agents, targeting cellular protein receptors. Epidermal growth factor receptor is one of these receptors critical to cellular proliferation, differentiation, and survival. As it has been shown to be widely expressed in SCC cells of the head and neck, it was suggested that anti–epidermal growth factor receptor therapy could become a powerful agent in combined modality therapy in the future.

Nonrandomized Studies

Some of the studies reviewed in this section are designed to treat advanced cancers in the head and neck in general and not focused solely on tumors in the oropharynx. As has been shown by the meta-analyses, concomitant CHT and/or altered fractionation result in improved locoregional control and OS, but also a substantial amount of toxicity has been observed (35,238). In fact, with regard to concomitant CHT, Pignon and Bourhis (238) showed at 5 years an 8% increase in OS and a hazard ratio of 2.17 for overall toxicity. Harrison (128) published the results of a phase II trial treating 82 patients with unresectable head and neck cancer using the delayed concomitant boost technique with concurrent cisplatin. The 3-year LC for oropharynx cancers was 64%. Twenty-four percent of patients required a treatment break. Two deaths due to sepsis occurred during treatment. Severe chronic toxicity occurred in three patients: one osteoradionecrosis, one frontal lobe necrosis, and one case of lung toxicity secondary to adjuvant CHT. Bieri (25) reported on delayed concomitant boost radiation in which a planned total dose of 69.6 Gy was given in 5.5 weeks; one third of the patients received concurrent cisplatin-based CHT. Among the 55 patients with oropharynx carcinoma, LRC at 3 years was 69.5%. Eighty-two percent experienced grade 3 and 4 mucositis. Patients receiving CHT had more grade 3 dysphagia (68% vs. 25%; p = .003), hospitalization (37% vs. 14%; p = .08), and a need for nasogastric tube (68% vs. 22%; p = .001). Nathu (214) published the results of induction CHT followed by RT for patients with oropharyngeal carcinomas treated at the University of Florida. Neoadjuvant CHT consisted of cisplatinum (100 mg/m2) and 5-fluorouacil (1.0 mg/m2/day × 5 days) for three cycles, and was followed by definitive RT (83% received hyperfractionated RT from 74.4 to 81.75 Gy). Outcome was compared with oropharyngeal tumors treated with a similar radiation regimen, but without CHT. Multivariate analysis showed no difference in local failure or distant failure. However, disease-specific survival and OS were improved in those who received induction CHT (58% vs. 27% and 42% vs. 17%, respectively). Because of the nonrandomized nature of the study and the lack of statistically significant improvement in parameters of tumor control, the authors cautioned against any conclusions regarding the benefit of induction chemotherapy.

A phase II study on 61 patients with advanced oropharyngeal carcinoma using induction chemotherapy followed by concurrent chemoradiation was reported by Vokes (309). Neck dissections (n = 35) were performed for N2 to N3 disease. At a median follow-up of 39 months (68 months among survivors), LRC was 70%, distant metastasis-free survival was 89%, disease-free survival was 64%, and OS was 51%. Acute toxicity was substantial, with severe or life-threatening mucositis and leukopenia during the induction phase, whereas 81% had grade 3 or 4 mucositis during the concurrent chemoradiotherapy. The authors concluded that the treatment sequence of induction chemotherapy followed by concurrent chemoradiotherapy and optional organ-preservation surgery is promising but that less toxic regimens need to be identified. Bensadoun (21) reported on 54 patients with unresectable oropharynx and hypopharynx carcinoma treated with concomitant hyperfractionated radiation (75.6 to 80.4 Gy) and three cycles of 5-FU/cisplatin in weeks 1, 4, and 7. Four percent mortality was observed from treatment related septicemia, 86% grade 3/4 mucositis but no patient required a treatment break greater than 4 days because of


mucositis. Grade 2 xerostomia was observed in 70% of the patients and grade 2 cervical fibrosis in 45% of the patients. At a median follow-up of 16 months, disease-specific survival was 72%.

There are many other examples of chemoradiotherapy regimen for oropharyngeal carcinomas with encouraging LRC rates, but with short-term follow-up and/or too small patient numbers (10,100,187). A promising approach was presented by the Memorial Sloan Kettering Cancer Center. Arruda et al. (62) studied 50 patients treated by IMRT in conjunction with concurrent CHT (86%). At 2 years, local progression-free OS and distant metastases-free survival is 98%, 98%, and 84%, respectively. Six of 42 patients remained with their percutaneous endoscopic gastrostomy until the time of analysis.

Randomized Trials

A prime example of a multinational, randomized trial of molecularly targeted therapy is the study by Bonner et al. (33) that was recently published in the New England Journal of Medicine. It compares patients with advanced cancers in the head and neck treated with high-dose RT alone (n = 213) or with RT plus weekly cetuximab, a monoclonal antibody against epidermal growth factor. The outcome of the study showed a significant improvement of locoregional control (hazard ratio locoregional progression or death 0.68; p = .005) and OS (49 months for combined therapy vs. 29.3 months for RT alone [hazard ratio for death, 0.74; p = 0.03]). It reduced mortality without increasing the common side effects of radiation. Studies for future targeted therapies combining cetuximab with chemotherapeutic agents such as Taxotere, cisplatin, and 5FU are now underway. Concurrent CHT with hyperfractionated radiation was explored by Brizel (36) in a phase III randomized trial. One hundred sixteen patients with advanced head and neck cancer were randomized to hyperfractionated radiation alone treated with 1.25 Gy twice daily 5 days per week to 75 Gy during a 6-week period versus a concurrent CHT arm consisting of 5-FU/CDDP given on weeks 1 and 6 of split-course hyperfractionated radiation. Both groups received two adjuvant courses of 5-FU/CDDP after completion of radiation. At a median follow-up of 41 months, the concurrent CHT showed improved LRC (70% vs. 44%; p = .01) and a trend toward improved 3-year OS (55% vs. 34%; p = .07) and relapse-free survival (61% vs. 41%; p = .08). However, patients in the chemoradiotherapy arm developed more acute toxicity, including the requirement for more feeding tubes (44% vs. 29%) and worse hematologic suppression. Chronic toxicity was no different, with about a 10% incidence of necrosis of the skin or bone in both arms. The trial has been criticized, not only for the added toxicity, but also because of the imbalance in the proportion of advanced neck disease (44% vs. 63%) treated in the concurrent chemoradiotherapy, which may have accounted for the difference in LRC. Jeremic (148) reported a phase III randomized study testing whether daily low-dose cisplatin improved outcome for patients undergoing hyperfractionation radiation compared with those treated with the same hyperfractionated radiation alone in locally advanced head and neck cancers (37% were oropharynx). One hundred thirty patients with stage III or IV disease were randomized to 1.1 Gy twice daily to 77 Gy per 7 weeks with or without cisplatin (6 mg/m2/day). At a median follow-up of 79 months, the investigational arm showed improved LRC (50% vs. 36% at 5 years; p = .041), progression-free survival (46% vs. 25% at 5 years; p = .0068), and OS (46% vs. 25% at 5 years; p = .0075), and fewer distant metastases (14% vs. 43% at 5 years; p = .0013). Daily concurrent CHT was well tolerated, with no increase in acute grade 3 mucositis and esophagitis. There were no increases in late skin or severe effects to bone or salivary gland. A multicenter randomized trial reported by Staar (288) tested whether the combination of hyperfractionated accelerated radiation (69.9 Gy/5 × 5.5 weeks) with carboplatin (70 mg/m2) and 5-FU (600 mg/m2/day × 5 days) on weeks 1 and 5 of RT improved outcome compared with the same radiation regimen alone. At a median follow-up of 22 months, the 1- and 2-year respective rates of LRC were 69% and 52% after chemotherapy/RT compared with 58% and 45% after RT alone (p = .14). Patients with oropharyngeal carcinomas had a trend toward improved 2-year LRC with chemoradiotherapy compared with RT alone (51% vs. 42%; p = .07).

Another German multicenter randomized trial compared hyperfractionated accelerated radiotherapy alone (77.6 Gy) with hyperfractionated accelerated radiochemotherapy (70.6 Gy) using mitomycin C and 5-FU (130). For patients treated inside the trial, no significant difference in survival was observed. A randomized phase II EORTC trial explored the feasibility of concomitant cisplatin and RT with conventional fractionation or multiple fractions per day (MFD). The MFD schedule was designed to achieve higher tumor concentrations of cisplatin at the time of irradiation by reducing the number of radiation treatment weeks from 7 to 3. No difference in acute and late side effects in both treatment arms while better tumor response was obtained with MFD. It is argued that the better tumor response in the MFD might be due to a (67%) higher daily dose of cisplatin concomitant with RT being given in a 3-week period (13). Hao et al. (119) updated the meta-analyses outcome of concomitant CHT trials to date in SCC of the head and neck. They confirmed an 8% benefit in 5-year absolute survival. Toxicity in general seems to be more pronounced with combined modality regimens using hyperfractionated RT or when the concomitant CHT regimen included carboplatin plus 5-FU. Several other randomized studies (e.g., Horiot et al. [139] or Fu et al. [93]) have demonstrated the beneficial effect of hyperfractionation and/or accelerated fractionation over standard fractionation. Also, Calais (42,43) demonstrated better locoregional control when altered fractionation is used with concurrent CHT. According to Hao et al. (119), the current state of the evidence supports strongly to offer platinum-based concurrent CHT with conventional fractionated RT as a treatment option for patients with advanced head and neck cancers treated outside a clinical trial.

Three-Dimensional Conformal RT/IMRT

The introduction of the multileaf collimator and three-dimensional treatment planning systems (TPS) in the 1990s has been instrumental for the development and application of three-dimensional conformal radiation therapy and IMRT. The major advantages of IMRT for irradiation of the complex head and neck anatomy are now generally recognized. The possibility of tightly shaping the higher isodose surfaces around the often concave target volumes allows for substantial sparing of critical structures. The use of electrons for irradiating the posterior neck, without exceeding the cord dose, has become almost obsolete. In this section, procedures are described for a safe and beneficial application of this powerful tool with focus on the IMRT techniques as used in the Erasmus MC.

IMRT

Treatment Planning

In the Erasmus MC, in case of radical radiation therapy of oropharyngeal cancer, IMRT is used to deliver a total dose of 46 Gy, 2 Gy per fraction, 6 fractions per week, to the primary tumor and neck, generally followed by a BT- or CyberKnife boost (Fig. 42.19). In line with the International Commission on Radiation Units and Measurements criteria for dose homogeneity in the PTV, it is generally required that 100% of the PTV must


obtain more than 95% of the prescribed dose (143), although small underdosages (e.g., around the salivary glands) are acceptable in specific cases. Tolerating minor PTV underdosages has also been described by Fogliata et al. (83) and Wu et al. (325). Recently, we have studied this trade-off between full PTV coverage and sparing of the parotid glands, using a model for calculation of the subclinical disease control probability (163). For the patients in the study, the mean parotid gland dose decreased by more than 10 Gy by allowing for a small underdosage in the PTV, corresponding with a reduction in the calculated subclinical disease control probability of typically 1% and a little higher.

The applied planning constraints for the critical structures for IMRT as used in Erasmus MC are presented in Table 42.19 and compared with the RTOG H-0022 protocol (247). For plan design, the constraints for the cord and the PTV are overriding, and the criteria for the parotids and oral cavity are planning objectives rather than hard constraints. To create a safety margin, the cord constraint is set for the spinal canal, rather than for the cord per se.

Depending on the patient geometry, different planning strategies are used. The most favorable strategy is to spare both parotid glands. This is done using a nonequiangular, five-field technique, with gantry angles of 0 degrees, ±60 degrees and ±140 degrees (optimized for each individual patient), using 6-MV beams. Especially when the boost is also delivered with IMRT, significant sparing of both parotids is frequently not feasible. It may then be decided to largely relax the constraint for the ipsilateral parotid gland and to focus on sparing of the contralateral gland. Generally, a nonsymmetrical four-field technique is then applied, with two parallel-opposed beams at gantry angles of around 350 degrees and 160 degrees (or 10 degrees and 200 degrees, depending on tumor position). With such an approach, that is, sparing of a single parotid gland structure Eisbruch et al. (73) observed a salivary flow increase after 2 years.

In the absence of positive nodes, the lower neck region is treated with two non-IMRT anterior fields, positioned on either side of the cord with sparing of the larynx (midline block) (Fig. 42.33). The International Commission on Radiation Units and Measurements dose homogeneity criterion is then less strictly enforced. Another technique to cover the lower neck region is to extend the upper IMRT fields.

Figures 42.27 and 42.30 show a five-field technique (0 to 46 Gy) for a patient with a TF tumor (T2N1) to be treated by RT to the primary and bilateral neck. For this bilateral parotid sparing treatment plan, mean doses to the parotids are 22 and 23 Gy, respectively. For comparison purposes, Figures 42.28 and 42.31 show a treatment plan with focus on maximum sparing of the contralateral parotid, yielding mean doses of 17 Gy (contralateral parotid gland) and 30 Gy (ipsilateral parotid gland).

Patient Setup Verification, Correction, and PTV-Margins

IMRT is most effective when used in combination with narrow PTV margins that have to be in line with the geometrical uncertainties for the patient involved. This implies a proper knowledge of the setup variations. Each patient has a setup error that occurs during all fractions (the systematic or mean error) and day-to-day variations around this mean setup error (the random errors) (26,278). By its nature, the systematic error of an individual patient can be obtained only from measurements during each fraction, and is therefore not known at the time of treatment planning. The setup uncertainties are generally quantified by three standard deviations Σx, Σy, and Σz, describing the distribution of systematic setup errors in the patient group, and the standard deviations σx, σy, and σz that represent the day-to-day variations. For head and neck cancer patients, these standard deviations are mostly derived from measurements with electronic portal imaging devices (EPIDs). Stroom et al. (289,290), from our institution, derived for each direction, i, the required PTV margin, Mi, given by:
Mi = 0.7·σI + 2·Σi.

In this approach, the PTV margin of each new patient is fully based on setup measurements performed for previously treated patients. The formula reflects the idea that systematic errors, potentially leading to an underdosage of a specific part of the tumor in all fractions, are more severe than random errors. The equation was confirmed by the work of van Herk et al. (132). Setup errors can be minimized using EPID measurements and a correction protocol. Deviations in the patient setup are then quantified by comparison of the EPID images with digitally reconstructed radiographs derived from the planning CT scan. It is essential that the demarcations on the patient's skin or mask, used for setup at the linac, are in exact agreement with the isocenter of the planning CT scan. These demarcations should not be adjusted in a session at a conventional simulator, neither should verification be based on acquired simulator images (19). In an online protocol, the patient setup error is assessed in each fraction using a few monitor units (MUs), followed by a subsequent correction and delivery of the remainder of the MUs. With such a protocol, both the systematic error and the random component can be substantially reduced. However, a disadvantage of online protocols is the involved workload at the treatment unit and the unavoidable increase of the fraction time. For this reason, so-called off-line protocols are more often applied than online protocols. In an off-line protocol, EPID images are only acquired in a limited number of fractions, and all image analyses are performed off-line, that is, not during the time of the delivery of the fractions. The latter excludes the possibility of reduction of random errors, which is


of lesser relevance for the determination of the required margin (equation). Instead, the aim of an off-line protocol is to reduce the more important systematic patient setup errors by estimating the optimal a priori setup correction for subsequent fractions.

In the Erasmus MC we have developed and implemented the no-action level (NAL) protocol for off-line corrections (63), which is now applied for most patient groups, including those with oropharyngeal cancer. For each patient, the protocol starts with acquisition of EPID images during the first Nm fractions (Erasmus MC Nm = 2 for head and neck sites), without applying any setup corrections. The involved systematic setup error for the complete fractionated treatment is then estimated by calculation of the mean setup error in these first Nm fractions. In the remainder of the fractions, the patient is first set up using the (original) marks on the patient mask. Then, prior to dose delivery, an a priori setup correction is performed as prescribed by minus the (estimated systematic error), followed by irradiation; no images are acquired. The first application of the NAL protocol for head and neck cancer was described by de Boer et al. (65). The patients in this study were treated with parallel-opposed laterals, and the NAL protocol was therefore only applied in two directions. Table 42.20 shows the setup errors for IMRT patients, as derived in a recent analysis (not published). As previously outlined, for each patient, the setup correction is based on an estimate of the systematic setup error, derived from measurements in only two fractions. As a consequence, application of the NAL protocol will diminish the systematic errors, but not cancel them out. In Table 42.20, both the distribution of the residual systematic errors, ΣNAL, and the distribution of (calculated) initial systematic errors, Σinit, that would have occurred without application of NAL, are presented. The presented margins are calculated with the equation provided. In clinical practice, margins of 5 mm are used for all directions, leaving some room for delineation uncertainty. Recently, the NAL protocol has been extended (eNAL) to systematically update setup corrections based on weekly follow-up measurements (64).

Dosimetric Quality Assurance

In our institution, IMRT is delivered with dynamic multileaf collimation (DMLC), using the sliding window technique. Because of the complexity, a dedicated quality assurance (QA) protocol is instituted, supplementing the QA procedures for non-IMRT treatments. All involved dosimetric measurements for IMRT are performed with EPIDs (Fig. 42.44) (89,90,131). For daily linac QA, the sliding-gap method as proposed by LoSasso et al. (181), measuring the leaf positioning accuracy with an ionization chamber for a single leaf pair, has been extended to two-dimensional, using the EPID (305). The measurements take 3 minutes, including the analyses. Errors in leaf positioning as small as 0.1 to 0.2 mm can be detected. Apart from the daily verification of the leaf motions, QA procedures are performed for each individual IMRT patient (228,304,306,328,329). These procedures aim at (a) verification of the final TPS dose calculation for the optimized treatment parameters such as the leaf


trajectories, and (b) verification of the correct execution of the plan at the linac. Currently, the TPS dose is only verified by an independent dose calculation for a single or few points in the center of the tumor. A fully three-dimensional procedure is being developed. For verification of the correct fluence delivery at the linac, EPID dose measurements are performed both prior to the first treatment fraction (pretreatment verification [229,306]), and during treatment (“in vivo” verification [305]). For pretreatment verification, portal dose images (a two-dimensional dose distribution in the plane of the fluorescent screen of the EPID) measured with the EPID are compared with predictions. Differences point at errors in leaf sequencing, data transfer from the TPS to the linac, or to dosimetric/mechanical linac performance problems. Presently, portal dose image comparison (Fig. 42.45) has been fully integrated in the applied EPID software (Theraview NT, Cablon Medical, Heusden, The Netherlands); a method for automated image analysis also has been implemented. Images are only reviewed by a physicist in case of a failure to pass the automated test. Because of the high spatial resolution, EPIDs are suited for detection of tongue-and-groove underdosage effects. For a group of 270 IMRT patients, the pretreatment procedure has revealed four serious errors prior to the start of treatment (329). Recently, methods have been developed for back-projection of fluence profiles, measured with the EPID, in the planning CT scan or in an in-room acquired cone beam CT scan, allowing full three-dimensional analyses (328).

Deviations in in vivo measured PDIs may be due to errors in fluence delivery, but may also be caused by changes in patient anatomy or variations in patient setup. To discriminate between the two, the split IMRT field technique (305) has been developed, which is now routinely applied for all head and neck cancer patients.

Alternative IMRT Approaches

Beam Orientations

Instead of dedicated orientations, often a relatively large number of equiangular beams are used. Whereas some articles report techniques with seven equiangular beams for oropharynx tumors (62,239), others advocate nine beams (324). Generally, an odd number is used to avoid opposing beams.

Simultaneous Integrated Boost

For oropharyngeal cancer patients treated in the Erasmus MC, the boost is generally delivered with BT or the CyberKnife after 46 Gy. When using IMRT for full-dose delivery, a simultaneous


integrated boost technique may be applied (166,247,325). The involved simultaneous optimization of the large field and the boost technique does generally result in superior plans, compared with sequential optimization (211). With the simultaneous integrated boost technique, an enhanced fraction dose may be selected for the primary tumor, yielding two simultaneous opportunities for biologic dose escalation: a shortening of the total treatment duration and an increased LC as a result of the higher daily tumor dose. However, the possibilities for application of escalated fraction doses are limited by the risk of increased toxicity (166,325). Alternatively, to minimize complications, the fraction dose in the elective regions may be reduced. The current RTOG study H-0022 applies a simultaneous integrated boost technique, prescribing a GTV total dose of 66 Gy at 2.2 Gy/fraction, and a dose for the subclinical disease region of 54 Gy (1.8 Gy/fraction) (247).

Plan Optimization and Evaluation Using Radiobiologic Models

Instead of using dose- and dose-volume–based objectives and constraints, plan optimization and evaluation can, in principle, also be done using radiobiologic criteria such as the tumor control probability, normal tissue control probabilities, and the equivalent uniform dose for the tumor and organs at risk. For the head and neck region, several parameter sets for biologic models exist, derived from observed tumor control and toxicity data (76,77,259,269) (also used in tumor control probability/normal tissue control probabilities calculating modules [315]). Unfortunately, the results vary considerably with the applied parameter set: for a group of oropharynx cancer patients, van Vulpen et al. (310) reported predicted normal tissue control probabilities differences for the parotid glands ranging from -3% to +35% when applying different parameter sets. To our knowledge, articles describing a decisive role in clinical decision-making for the treatment of oropharyngeal cancer patients have not yet been published.

Step-and-Shoot or Segmental IMRT (SMLC)

Apart from the DMLC technique, intensity-modulated profiles can also be generated by sequential delivery of static field segments, with a variable shape and number of monitor units (SMLC). In the transition period from end of delivery of one segment to shaping of the next segment, the beam is switched off. DMLC allows for more precise realization of the optimized fluence profiles. However, some studies have concluded that the differences are of minor clinical importance (4,55,83). It has also been reported that DMLC treatments require more MUs and SMLC treatments take more time.

A leaf-sequencing algorithm for DMLC has been developed that fully prevents the occurrence of tongue-and-groove underdosage reported by van Santvoort et al. (266). Currently, we use the Cadplan TPS (Varian Medical Systems, Espoo, Finland) for inverse planning and leaf sequencing. It was demonstrated that for extreme profiles, tongue-and-groove underdosage of up to 30% may occur with this TPS (80). However, the protocol for pretreatment verification of the fluence profiles of each individual IMRT patient has never revealed a clinically relevant tongue-and-groove error. Also for SMLC, leaf-sequencing algorithms have been developed that reduce or prevent the occurrence of tongue-and-groove underdosage (61,182).

Clinical Results

Some of the clinical results are presented in the section Xerostomia. Excellent reviews are presented by Puri et al. (243) and Lee et al. (168). Table 42.21 summarizes the preliminary clinical results of several studies of IMRT treatment for oropharyngeal carcinoma. The studies confirm the high rates for (loco-) regional control, distant metastases-free survival, disease-free survival, and OS in combination with reduced toxicity in comparison to conventional radiotherapy. Finally the multi-institutional RTOG study (H-0022) using IMRT for early-stage oropharyngeal cancer has completed accrual, and final results are to be expected shortly (247). IMRT allows dose to be concentrated in the tumor volume while sparing normal tissues. However, the downside to IMRT is the potential to increase the number of radiation-induced second cancers. The reasons for this potential are more MUs and, therefore, a larger total-body dose because of leakage radiation and, because IMRT involves more fields, a larger volume of normal tissue is exposed to lower radiation doses. In fact, Hall (114) calculated that IMRT may double the incidence of solid cancers in long-term survivals. In contrast to older patients, if balanced by an improvement of local tumor control, the use of IMRT might not be acceptable in children. An alternative might be to replace x-rays with protons in case of scanning pencil beams.

Alternative External Beam Approaches

Helical Tomotherapy

Apart from IMRT with linear accelerators, helical tomotherapy (186) (HiArt, TomoTherapy Inc., Madison, WI) can also be used for highly conformal dose delivery. Several articles report on dose distributions for head and neck cancer patients that might be superior to those obtained with linacs, regarding sparing of critical structures (82,230,310). Long-term clinical evaluations are not available as yet. Compared to linac-based IMRT, tomotherapy requires more MUs to deliver the same target dose, because of the applied fan-beam (212). This increases the whole-body dose equivalent, which may increase the risk for radiation-induced secondary malignancies (84). The clinical


implications of irradiating larger volumes to lower doses with tomotherapy, compared with smaller volumes with intermediate doses in linac IMRT, are unknown.

CyberKnife

The robotic CyberKnife system (Accuray Inc., Sunnyvale, CA) is another means of applying high dose of radiation with high accuracy (2). Some preliminary experience with the CyberKnife is available from the Erasmus MC for cancer in the oropharynx (Fig. 42.19 shows the protocol). This regards the delivery of a boost treatment of three fractions of 5.5 Gy on each consecutive day, prescribed at the 80% isodose. Patients are immobilized with the regular thermoplastic mask with a three-point fixation. Highly conformal plans with steep dose gradients are generated using 100 to 200 noncoplanar and nonisocentric coned beams. Figure 42.46 shows a typical dose distribution for a CyberKnife boost with the applied beam orientations. The CyberKnife image-guidance system and the patient skull are used for frequent measurement of the patient setup during treatment. Observed translations and rotations are used for immediate correction of the position and direction of the next beams. Because of these continuous adjustments, a PTV margin of only 2 mm was applied originally. Recently, the images obtained with the CyberKnife image-guidance system have been retrospectively analyzed, to quantify patient motion during delivery of a treatment fraction (135). For head and neck cancer and brain cancer patients, the maximum observed displacement in a 2-minute period was 2.8 mm in a single direction; a maximum rotation of 2.3 degrees was observed after 3 minutes. The overall systematic and random three-dimensional errors after 15 minutes are 1.3 and 1.2 mm (2 SD), respectively. With the CyberKnife image-guidance system, these in-fraction patient movements are automatically compensated using the robotic manipulator.

Cone-Beam CT

An important next step in image-guided RT for head and neck tumors may be the use of the recently introduced cone-beam CT scanners, integrated in linacs (Fig. 42.44) (146,172,202). In contrast to EPIDs, these systems allow for visualization of soft tissues. So far, the image quality is not as good as for modern diagnostic scanners. During the fractionated head and neck treatment, various processes may result in a gradual change of the patient anatomy, such as postoperative changes/edema, weight loss, and shrinking of the primary tumor and/or nodal masses (11,118). Large changes in the size of the GTV and the size and position of the parotid glands have been observed. These changes may result in suboptimal treatment as the dose delivery in all fractions is usually based on a treatment plan that is designed for the patient anatomy in the planning CT scan, which is acquired prior to the start of treatment. Studies have been performed to investigate the impact of replanning based on, or triggered by, anatomy changes observed in acquired cone beam CT scans (118,210). A major clinical question to be answered is the target definition in case of a shrunken gross target volume.

As part of the IMRT QA protocol, cone-beam CT scans may also be used to assess the “dose of the day” (328).

Future Technical Developments

Dose-Calculation Algorithms

It is well known that, especially in the presence of low-density inhomogeneities, significant dose-calculation errors may occur for single beams, even when using a modern commercial TPS. Such errors have also been observed for clinical, multibeam head and neck treatment plans (34,251,270). Improved accuracies can be obtained with Monte Carlo dose-calculation algorithms, and vendors of TPSs have started to offer this tool (129,251). However, to obtain clinically acceptable calculation times, approximations and simplifications are often used that could jeopardize the potential advantages of the full Monte Carlo technique. A comprehensive overview is provided by Reynaert et al. (252).



Paperless Electronic Records

In the previous section one is confronted with innovative, highly technological care, but also with clinical research regarding QOL issues. These processes will undoubtedly go on with virtually no limitations. From the organizational (data-retrieval) point of view, one could envisage that most departments of radiation oncology will eventually be structured as a “paperless office” (Fig. 42.47). Direct architecture changes the conventional workflow into a productive workspace environment; that is, with a click of a button it combines the ease of the use of Windows with access to all types of vendor applications, including record and verify, e-mail, IMRT-QA, and office applications. This server client architecture gives users the freedom to access their applications anywhere in the hospital or in the world for collaboration, consultation, or to access particular applications for personal use. Features like pen-enabled computing, centralized storage, and secure remote access of the applications via broadband are not new in the information technology arena, but definitely are not routine to radiation therapy. It is setting a stage for any type of new application to fit into the existing infrastructure without adding new workstations or PCs in the already fully taken workspace. The future generation of connectivity between radiotherapy applications is true flexibility at the physician's desk, a solution without constraints.

Acknowledgments

The authors of this chapter are greatly indebted to the scientific work of the many authors of referenced articles (well over 300 references), as well as to the many coworkers of the Department of Radiation Oncology of the Erasmus MC, Rotterdam, for their enthusiastic and skillful support. In particular, we would like to mention the laborious effort to help revise and to provide new information by Inge Noever, Peter Voet, Henrie van der Est, Erik Franken, Mischa Hoogeman, Eric van ‘t Hooft, Johan Pöll, Peter van Rooij, Eliana Vásquez Osorio, Jeanette Bruinsma-v.d Hill, Jeanette Schilperoord, Schandra Manusama, and Marlies de Gelder.