Management of the N3 Neck

Surgery followed by radiation has long been the criterion standard treatment option for advanced stage squamous cell carcinoma (SCCA) of the head and neck. Treatment has consisted of resection of the primary site, radical neck dissection, reconstruction as needed, and postoperative radiotherapy. To minimize the morbidity of radical neck dissection, Bocca (1967) introduced the concept of modified radical neck dissection, preserving the sternocleidomastoid muscle, internal jugular vein, and spinal accessory nerve, when possible.[4]

However, a modified radical neck dissection is often not possible in the N3 neck. Given the potential morbidity associated with radical surgery, surgeons and oncologists alike began looking for nonsurgical options for treating advanced stage head and neck cancer. Many chemoradiation trials have been conducted, with the Department of Veterans Affairs Laryngeal Cancer Study Group (1991) trial beginning the push for nonsurgical treatment of advanced stage head and neck cancer. To this end, many centers consider organ preservation chemoradiation as the standard treatment for advanced stage head and neck cancer.

A patient with N3 disease of the neck is automatically categorized as having stage IVb disease, and the average 2-year survival rate, considering all head and neck sites, is approximately 20%.[1] Patients with N3 disease also have a high rate of distant metastasis, up to 30%.[5] In addition, both radical surgery and concurrent chemoradiation bring morbidity to the patient. The goal of treatment in most patients with this degree of disease is palliation, but cure may still be possible.

Frequency

N3 neck disease is uncommon. The rate of patients presenting with the N3 neck ranges from 1.7-9.5% in large series.[1, 5, 6]

Squamous cell carcinoma (SCC) arising from the aerodigestive tract is the most common cause of cervical metastases. Primary tumors of the major and minor salivary glands can also present with cervical lymphadenopathy. The histopathology of malignant salivary gland tumors that have frequent neck disease include high-grade mucoepidermoid carcinoma, squamous cell carcinoma, carcinoma ex-pleomorphic, and high-grade adenocarcinoma. Thyroid carcinomas also manifest with cervical metastases, most commonly from papillary or medullary thyroid carcinomas. In addition, lymphoma can present with bulky lymphadenopathy.

As mentioned previously, N3 neck disease usually results from squamous cell carcinoma arising from the aerodigestive tract. Although the disease usually involves only structures in the neck, depending on the location of the involved node, it may extend into surrounding structures such as the skull base, clavicles, or mediastinum. In addition, patients presenting with N3 neck disease are at high risk for distant metastases. The lungs (83.4%), bone (31.3%), and liver (6%) are the most common site of distant metastasis, although the brain and other sites have been reported in the surgical literature.[7]

Aside from evaluation of the primary tumor, patients with an N3 neck disease present in the following clinical scenarios:

• N3 neck disease that has not been treated

• N3 neck disease after chemoradiation or radiation with no response or partial response

• N3 neck disease after chemoradiation with complete response

• N3 neck disease with unknown primary tumor

A complete head and neck examination is warranted to check the following:

• Identify primary site

• Fixation of mass around bone or paraspinal muscles

• Skin involvement

• Mediastinal spread

• Contralateral neck

The following may be presenting symptoms:

• Neck mass

• Otalgia

• Nerve deficit

• Horner syndrome

• Dysphagia

• Airway obstruction

The goals of treatment include palliation, improved quality of life, and cure, if possible. If the tumor is histologically proven to be squamous cell carcinoma based on findings of either fine-needle aspiration (FNA) of lymph node or biopsy of the primary tumor, the factors that influence treatment include operability and resectability of the both the primary tumor and the neck, the presence of distant metastases, as well as the presence of synchronous second primary tumors of either the head and neck or the lung. The exception to these considerations is nasopharyngeal carcinoma, which is treated with primary chemoradiation, regardless of the status of the neck. This should be the first step in assessing these patients.

The term operable refers to a reasonable degree of safety and chance of success. The term resectable refers to the ability to completely resect the tumor with adequate or clear surgical margins. Tumors that are inoperable or unresectable may still be treatable or even curable with nonsurgical therapy.

The treatment for the primary tumor dictates the management of neck disease. Therefore, if the primary tumor is treated surgically, then the neck disease will also be treated initially with neck dissection. If chemoradiation is the initial choice of therapy for the primary site, then the neck will also be treated with chemoradiation followed by completion neck dissection. However, if the goal is palliation, multimodality management (surgery, radiation, chemotherapy) is used to alleviate suffering and to minimize morbidity.

Operable primary tumors and N3 neck disease may be managed with surgical resection followed by radiation, with or without chemotherapy, to achieve the best chance of locoregional control.

Based on the head and neck surgeon's realistic and mature surgical judgment, and after the risk, benefits, alternative treatments, and potential complications are weighed, the ultimate decision concerning tumor resectability can sometimes be determined only at the time of surgery. During surgery, the surgeon should bear in mind that the execution of extended neck dissection, in which vital structures have to be sacrificed, has no benefit to the patient if the tumor cannot be completely resected.[8]

The indications for concurrent chemoradiation are as follows:

• Inoperable and unresectable primary tumor and/or nodal disease

• Organ preservation

• Patient refuses surgical treatment

Chemotherapy is typically indicated with adjuvant radiation therapy after surgical treatment for pathologic findings of extracapsular spread and close (within 5 mm) or positive margins.[9]

The cervical lymphatics have been divided into levels based on the patterns of spread of the primary tumor from different sites of the aerodigestive digestive site.[9] . Classification of the neck levels have helped to standardize surgical treatment of the neck as well as serve as a prognostic indicator of outcomes. The neck levels are as follows:

• Level Ia - Submental triangle

• Level Ib - Submandibular triangle

• Level IIa - Jugulodigastric nodes from the skull base to the hyoid bone, anterior to CN IX

• Level IIb - Jugulodigastric nodes from the skull base to the hyoid bone, posterior to CN IX

• Level III - Jugulodigastric nodes from the hyoid bone to the cricoid cartilage

• Level IV - Jugulodigastric nodes from the cricoid cartilage to the clavicle

• Level Va - Posterior triangle nodes superior to the horizontal plane of the cricoid cartilage

• Level Vb - Posterior triangle nodes inferior to the horizontal plane of the cricoid cartilage (supraclavicular nodes)

The following are contraindications to surgical management of N3 neck disease:

• Carotid artery encasement

• Paraspinous muscle invasion

• Vertebral column invasion

• Skull base invasion and/or extension

• Horner syndrome

• Phrenic nerve palsy

• Brachial plexus palsy

• Uncontrolled primary tumor site

Laboratory studies should include the following:

• Basic metabolic panel

• Complete blood count (CBC)

• Liver function tests

• Prealbumin and albumin

• Coagulation panel

• Blood type and cross-match

• P16 testing on biopsy specimen of oropharyngeal primary (surrogate test for HPV)

Liver function tests, as a part of the metabolic panel, may be used to identify liver metastasis. However, they are nonspecific and are not sensitive. Elevated results of liver function tests may reflect associated alcoholic liver disorders. Electrolyte abnormalities may reflect tumor-induced syndrome of inappropriate antidiuretic hormone (SIADH).

Evaluation of N3 neck disease is as follows:

• CT scan with contrast is useful in determining resectability and the extent of the primary tumor and nodal disease.

• MRI with gadolinium can demonstrate soft tissue changes and reveal perineural spread.

• Positron emission tomography (PET)–CT fusion study may be helpful in patients who present with N3 nodes with an unknown primary tumor. PET-CT scans were able to reveal an unknown primary tumor in 57% of cases compared with CT scan alone, which identified the unknown primary tumor in 23% of cases. Despite the increased ability of PET-CT scanning to reveal unknown primary sites, 43% of the primary sites were not identified.[10]

Evaluation for lung metastasis and second lung primary is as follows:

• Patients with N3 neck disease are at increased risk for developing distant metastasis. Garavello et al (2006) found patients with N3 disease developed distant metastases 29.55% of the time and had a 10.7 times increased risk of developing distant metastases.

• The lung is the most common site of distant metastasis in head and neck cancer, as supported by both clinical and autopsy studies.

• Chest radiography has poor sensitivity in detecting lung metastasis (sensitivity of 50%, specificity is 94%.[11] )

• The incidence of pulmonary malignancy in head and neck cancer is 4.5-14%. Secondary lung malignancy is a high risk if the primary tumor originated from the larynx or pharynx.

• The lifetime risk in developing secondary malignancy in patients with head and neck cancer can be as high as 20%.

• CT scan is the single most sensitive imaging study used to reveal lung metastasis and, therefore, should be the modality of choice, at least in high-risk patients (stage 4 disease, T4 tumor, N2 or N3 nodal disease, tumors that arise from the oropharynx, larynx, hypopharynx, or supraglottis).[12]

• Yearly chest radiography is suggested to evaluate for second primary lung tumors or distant metastasis.

See the list below:

• Evaluation for liver metastasis

  • Liver metastasis usually presents in association with other metastases, especially lung metastasis. Liver metastasis alone is rare. Screening tests used to identify liver metastasis are nonspecific and are not sensitive. Elevated results of liver function tests may reflect associated alcoholic liver disorders.

  • Although ultrasonography, CT scanning, and MRI are sensitive imaging modalities for liver metastasis, if the index of suspicion is high, especially in the presence of lung metastasis, a liver ultrasound or CT scan can be performed to confirm clinical suspicion.[13]

• Evaluation for bone metastasis

  • Bone metastasis is invariably associated with lung metastasis; 50% of cases of bone metastasis involve multiple bony sites. The spine is the most common site of metastasis (12.7%), followed by the skull (4.2%), the rib (3.1%), and axial bone (femur, humerus [2.1%]).[7]

  • Alkaline phosphatase is sensitive for osteoblastic bone metastasis. It is not sensitive for revealing bone metastasis in head and neck squamous cell cancer head because this type of bone metastasis is osteolytic.

  • Bone scan is sensitive in the diagnosis of bone metastasis; however, it is useful only if the patient is clinically symptomatic. Routine bone scan is not necessary.

• Carotid artery evaluation

  • MR studies have shown that, if the carotid artery is surrounded by tumor in 270° or less, no carotid artery invasion has occurred.[14] Yoo et al compared preoperative CT findings with histologic findings in patients who underwent carotid artery resection.[15] If the carotid artery was encircled more than 180° by the tumor on CT scan, then the tumor invaded the elastic lamina of the carotid artery and patients had poorer outcomes.

  • In patients who have undergone chemotherapy or radiation treatment, surgical planes between the carotid artery and the tumor are obscured. In these cases, CT scanning or MRI are less accurate in predicting carotid artery invasion.[14]

  • Planned carotid artery resection or embolization requires preoperative testing to assure adequate collateral blood flow through the contralateral carotid artery and/or vertebral artery system. In addition, the presence or absence of carotid stenosis must be assessed.

Discussing the different indications and contraindications, complications, and advantages of each test is beyond the scope of this chapter. Invasive tests used to evaluate collaterals of the carotid artery can cause neurologic complications post-procedure. Several studies have conclusively shown that severe hemodynamic impairment is a strong predictor of stroke in patients who undergo carotid artery occlusion. In the St Louis Carotid Occlusion Study, a prospective study that evaluated cerebral hemodynamic and stroke risk, more than half of the symptomatic and asymptomatic patients had a normal oxygen extraction fraction.[16]

See the list below:

• Biopsy is necessary to confirm the diagnosis. Fine-needle aspiration biopsy (FNAB) of the neck mass is the only required test for the diagnosis. If readily available, a biopsy procedure can be performed on the primary tumor. The use of core biopsy has also been reported.[17]

• The use of an open biopsy procedure is necessary only if the diagnosis cannot be attained with FNAB and a lymphoma is suspected.

• Patients with head and neck carcinoma are at risk for synchronous primary tumors. Panendoscopy, including direct laryngoscopy, bronchoscopy, and esophagoscopy, has classically been performed to assess the primary tumor and to identify the presence of synchronous primary tumors.

As mentioned above, squamous cell carcinoma is the most common cause of carcinoma of the aerodigestive tract with cervical metastasis. For patients who undergo surgical treatment of their disease, the histologic features of the primary tumor and neck dissection are important for determining the need for adjuvant therapy.

At the primary site, the surgeon and pathologist alike are interested in the size of the tumor, the presence of positive surgical margins, perineural or lymphovascular invasion, and invasion of surrounding structures. In addition, the pathologist will comment on the grade of differentiation of the tumor, ranging from well differentiated, to moderately differentiated, and poorly differentiated.

However, in squamous cell carcinoma, the grade of the tumor in general does not impact prognosis. On the other hand, the grade of the tumor is important in salivary gland carcinomas. In the neck, features of interest include the size of the tumor, the number and location of involved nodes, and the presence of extracapsular extension of the tumor outside the lymph node.

High-risk features that have long been considered indications for postoperative radiotherapy include advanced T stage, perineural invasion, and multiple positive lymph nodes.

Patients who benefit from the addition of chemotherapy to postoperative radiotherapy are those with positive margins and those with extracapsular spread.[18, 9]

The reader is referred to the current AJCC (6th edition) and UICC guidelines for T staging of the primary tumor at the individual head and neck sites.

AJCC and UICC nodal categories (except thyroid and nasopharyngeal carcinoma) are as follows:

• Nx - Regional lymph nodes that cannot be assessed

• N0 - No regional node metastasis

• N1 - Metastasis in a single ipsilateral lymph node, 3 cm or smaller

• N2a - Metastasis in a single ipsilateral lymph node larger than 3 cm but not larger than 6 cm

• N2b - Metastasis in multiple ipsilateral lymph nodes, none larger than 6 cm

• N2c - Metastasis in bilateral or contralateral lymph nodes, none larger than 6 cm

• N3 - Metastasis in a lymph node larger than 6 cm

Distant metastasis categories are as follows:

• Mx - Distant metastasis cannot be assessed

• M0 - No distant metastasis

• M1 - Distant metastasis

The combination of the primary tumor, nodal status, and presence or absence of distant metastasis is used as a part of the overall staging of the patient’s disease according to AJCC and UICC guidelines, as follows:

Table 1. Staging (Open Table in a new window)

Local control with chemoradiation therapy has been reported to be as high as 90%. Clinical response, however, does not correlate with pathologic response (see Futures and Controversies). Although studies have shown that no clinical or pathologic parameters are able to predict a response to chemoradiation therapy, studies suggest that a tumor volume of less than 20 mL, tumor grade, lower epidermal growth factor receptor (EGFR) overexpression, and response to induction chemotherapy predict response to chemoradiation.

Concurrent chemoradiation therapy has survival rates similar to those of surgical therapy and preserves the function of important structures. Surgical therapy, however, carries better locoregional control rates.

Complete response to chemoradiation therapy requires a complete disappearance of all clinical, radiologic, and pathologic (if applicable) evidence of disease; however, not all clinically complete responses show histologic complete response after planned neck dissection. Anything less than a complete response requires surgery of the primary site and the neck nodes, if possible.

Treatment of the neck following chemoradiation is controversial. In the presence of a partial or incomplete response in the neck to chemoradiation, a completion neck dissection is mandatory (assuming the primary site is controlled or is resectable). In the presence of a complete response in the neck to chemoradiation, reports vary regarding the need for planned neck dissection (see Future and Controversies).

The following complications are associated with chemoradiation therapy:

• Feeding tube (required in 32% of patients)

• Mucositis

• Neutropenia

• Carotid stenosis (range, 30-50% of patients)[19]

• Renal insufficiency

• Ototoxicity

• Esophageal stricture

Surgery alone for the N3 neck carries a local failure rate of 21% and 5-year survival rate of 15%. Surgery for a stage N3 neck should always be in conjunction with chemoradiation or radiation (performed either preoperatively or postoperatively). The complication rates for a neck dissection after concurrent chemoradiation range from 8-35%.

Table 2. Type of Neck Dissection Chosen as Dictated by What Structures are Involved (Open Table in a new window)

Classically, the N3 neck, if treated surgically, requires a radical neck dissection. Although a modified neck dissection may be possible in some cases, the classic radical neck dissection is necessary in most cases of the N3 neck.

Carotid artery resection is controversial. A 13-year retrospective study by Freeman et al (2004) reported on 41 patients whose carotid arteries were resected and reconstructed and 11 patients who underwent preoperative embolization or intraoperative ligation of the carotid artery.[20] The median disease-specific survival and the median disease-free interval were both 12 months. Distant metastasis developed in 24% of patients, and 20% of patients had recurrence within 6 months of the resection. Eight (20%) of 41 patients who underwent resection and reconstruction of the carotid artery developed stroke postoperatively. Three (27.7%) of the 11 patients who underwent embolization or ligation of the carotid artery developed stroke postoperatively.

Cancer seldom invades the lumen of the carotid artery; based on a study by Huvos et al, only 42% of patients had invasion of the adventitia and external elastic membrane.[21]

The treatment options in the management of cervical metastasis that involves the carotid artery are as follows:

• Permanent occlusion can be performed if collateral circulation is adequate.

• Resection with carotid shunt and reconstruction can be performed if collateral circulation is inadequate.

• Nonsurgical palliation is an option if collaterals are inadequate and reconstruction is not possible.

• Nussbaum et al (2000) developed a unique technique in the management of carotid involvement in head and neck cancer.[22] Endovascular stenting of the carotid artery is initially placed and followed by a staged neck dissection after 1 month of placement. Because the neoendothelial barrier has formed in the stent and prevents bleeding, dissection of the arterial wall with neck dissection can be performed. This is a technical case report that is currently under investigation.

Lore and Boulos reported that one third of their patients who underwent carotid artery resection lived longer than 2 years after the procedure, and a meta-analysis on carotid resection by Snyderman showed improved local control.[23, 24] The decision to resect the carotid artery depends on the risk-benefit ratio in relation to local control, survival, stroke risk, quality of life, mortality, and available expertise.

The external carotid artery should be resected as needed; however, the authors' view is that resection of the carotid bulb or internal carotid artery is rarely, if ever, indicated. This is because the chance for cure is exceedingly low. The hypoglossal and or vagus nerve can also be invaded aside from the carotid artery. Resection of both the hypoglossal and vagus nerves increases the morbidity and mortality. Stroke is a risk, even when results of balloon occlusion studies are favorable. Even with resection of the carotid, the median survival is only 12 months.

Consider the primary site during treatment planning. Generally, primary site management dictates the plan for the neck. Patients who will undergo surgery for the primary site must also have resectable neck disease. As discussed previously, incomplete resection of the neck disease is of no benefit to the patient (see Contraindications). Careful inspection of preoperative imaging is critical to making this assessment.

Certain patients may not have an operable primary tumor and would not normally undergo a neck dissection. However, they may have morbidity associated with advanced neck disease and could benefit from a palliative neck dissection. This situation includes patients with the following:

• Nonhealing ulcer

• Bleeding neck mass not attached to vascular structure

• Unresponsive to conservative wound care

• Unresponsive tumor embolization

• Neck mass is operable, resectable, and adequate margins can be achieve

This highly selected subset of patient may be a candidate for palliative neck dissection. However, one must carefully weigh the risks of creating a worse wound and having the patient hospitalized for his or her remaining life against potential quality-of-life improvements.

Intraoperatively, the surgeon must assess whether the N3 neck is completely resectable. Involvement of structures not normally included in the standard neck dissection must be assessed and resected as needed to achieve a complete resection. However, unless the tumor is completely resectable, resecting important neurovascular structures, causing unnecessary morbidity, is not advisable. (See Relevant Anatomy and Surgical Therapy.)

When possible, patients benefit from adjuvant therapy following neck dissection for advanced neck disease. Chan et al (2003) showed that patients with N3 disease treated with surgery and postoperative radiotherapy had improved survival when compared with surgery and preoperative radiotherapy or surgery alone.[25] Patients with advanced neck disease derive additional benefit from postoperative concurrent chemoradiation.[18]

Following treatment for the N3 neck, patients require close follow-up, as they are at high risk for both regional recurrence and distant metastases. The risk of relapse is highest in the first 2 years following treatment. Many surgeons advocate monthly physical examinations, and imaging every 3 months in the first year following treatment.

The following are complications associated with surgical resection of the N3 neck:

• Perioperative stroke (range, 0.2-4.8%)

• Wound complications (range, 8-35%)

• Fistula

• Bleeding

• Infection

• Carotid artery rupture

• Chyle leak

Carotid artery rupture is a dreaded complication of advance neck disease. Rupture may be preceded by a sentinel bleed, wherein a patient has a short-lived episode of bleeding from the mouth, neck, or stoma. After bleeding subsides, all may appear to be well, but hospitalized patients may need to be placed on carotid artery precautions unless the patient refuses resuscitation. Precautions include a prepositioned stretcher to quickly take the patient to the operating room (OR) and rolled gauze bandages to obtain adequate pressure to the artery. Make no attempt to clamp the vessel. The following are the appropriate sequence of events in the management of carotid blowout:

1. Apply pressure and volume support.

2. Transfer the patient to the OR immediately.

3. Manage the condition either by endovascular stent graft or by surgically obtaining proximal and distal control followed by surgical repair or ligation, depending on hemodynamic status and condition of the tissue.

Outcomes are best when combined modality therapy is used. However, patients with the N3 neck have overall poor outcomes as a result of developing regional recurrence and distant metastases.[22, 24] Assessing the outcomes of the N3 neck is sometimes difficult because these patients frequently are grouped with patients with N2 neck disease.

Several recent studies have specifically addressed the outcomes of patients with N3 neck disease. Carvalho et al (2005) achieved the best regional control of 73% using surgery and postoperative radiotherapy, although overall 3-year survival was still poor at 17.9% in this group of patients.[26] Chan et al (2003) also found the best results in the N3 patient by treating with surgery followed by radiation.[25] Their 1-year, 3-year, and 5-year neck control rates in this group were 92.3%, 46.1%, and 46.1%, respectively. The overall disease free survival rates in their patients at 1, 3, and 5 years was 44.4%, 25%, and 22.2%, respectively.

Ballonoff et al (2008) reported their results in patients with N3 neck disease treated with primary chemoradiation, with or without planned neck dissection.[27] Their rates of locoregional control and distant control were 88% and 56%, respectively. Actuarial overall survival and disease-free survival at 2 years were 51% and 29%, respectively.

A study by Sellami et al indicated that although the overall survival (OS) rate is low in patients with locally advanced head and neck SCCA with an N3 neck, long-term locoregional control is more achievable in those with a body mass index (BMI) of 25 or greater at diagnosis. The 5-year OS rate in the report was 20%, but the 5-year locoregional control rate in patients with overweight or obesity was 55.0%, versus 15.3% for individuals with a lower BMI. Patients in the study were treated either with radiotherapy (with or without chemotherapy) or with nodal surgery (with or without primary tumor resection and, in most cases, with adjuvant radiotherapy with or without chemotherapy).[28]

A study by Alvarado et al found that in patients with clinical N3 (cN3) esophageal cancer, those who underwent surgery following neoadjuvant chemoradiation therapy had a better 5-year OS rate than did those who were treated with chemoradiation therapy alone (18.0% vs 11.8%, respectively).[29]

The following methods for local control of recurrent tumor in the neck are currently under investigation:

• Radiofrequency ablation[30]

• Microwave interstitial hyperthermia (915 MHz)[31]

• Intratumoral administration of cisplatin and epinephrine[32]

• Interstitial photodynamic therapy[33, 34]

• Intratumoral administration of ONYX-015 adenovirus and chemotherapy[35]

• Reirradiation therapy and chemotherapy with amifostine[36]

• Electroporation with chemotherapy[37]

A controversial topic in managing advanced neck metastasis treated with definitive chemoradiation is whether to offer the patient a planned neck dissection after a complete response. Undoubtedly, a patient with resectable neck disease with no response or a partial response after chemoradiation should undergo salvage neck dissection; however, there may be a role for planned neck dissection even after clinical and radiological complete response. The debate stems from the fact that after a clinical and radiological complete response from chemoradiation, there is still 20%[38, 39] of patients with pathological nodes confirmed with histology after planned neck dissection.

Some authors advocate that patients with N2-N3 neck have improved regional control after planned neck dissection, which reduces the need for surgical salvage, which has a low likelihood to succeed. There is limited ability for physical examination and imaging to predict viable cancer cells in lymph nodes despite apparent complete response. A planned neck dissection performed 4-12 weeks after surgery has a 10% morbidity rate, with improved local control in specimens found to have viable tumor cells.[39] In addition, an observation approach to complete responders is complicated by the increased morbidity of surgery outside of the 4- to 12-week window following chemoradiation therapy, owing to increased fibrosis.[39]

Other authors routinely monitor patients after a complete response to chemoradiation therapy because of the lack of survival benefit from planned neck dissection and the low risk of neck recurrence. The strategy of planned neck dissection originated in the era of primary radiation therapy as a single-modality treatment, which showed poor response to advanced neck disease.[40]

With improved imaging and radiation therapy advancements, treatments have become more effective, with a better ability to detect residual disease.[41] In a series of 60 patients with N2-N3 disease with complete response to chemoradiation therapy, there was no isolated neck failure despite not undergoing planned neck dissection.[42] There is an 80% negative specimen rate for patients who undergo planned neck dissection,[38] and those patients who have viable tumor cells in the specimen have aggressive disease that may be more likely to develop distant metastasis or primary site failure.[40] In addition, there is a reported better response in HPV-positive tumors, so this could influence decision making about treatment of the neck.[40, 43]

There is new debate about the utility of positron emission tomography (PET) scanning to detect viable tumor cells in clinical complete responders. There have been disappointing data in the ability of PET to accurately predict histologic response to treatment when used prior to 12 weeks after completion of chemoradiation therapy.[44, 45] The optimal timing of PET scanning is 12 weeks following chemoradiation therapy, which is outside the window for ideal timing of planned neck dissection, but a negative PET scan may avoid an unnecessary surgery.[46]

In a study of 31 patients who underwent chemoradiation for N3 neck disease, 28 had a complete response at 12 weeks post therapy at the primary site and of these, 20 showed complete response in the neck as well on PET scan.[47] In this study, complete response was defined as having a fluorodeoxyglucose (FDG) PET complete nodal response, regardless of the size of any residual nodal abnormality. Of these 20 patients who were observed after a negative PET study, 1 had nodal recurrence that was inoperable, 3 had metastatic disease, and 16 were without nodal recurrence. Of these 20 patients, 12 had residual nodes seen on CT—median size 13 mm (range, 6–40 mm)—whereas another had diffuse thickening that was not able to be measured.[47] Using CT criteria for residual disease instead of PET criteria would have led to an increased number of neck dissections in patients who ultimately were without nodal relapse with observation alone.

Selection criteria for planned neck dissection versus observation among patients who have had a complete response to chemoradiation therapy need to be established. Yao et al (2004) correlated residual pathology in postradiation or postchemoradiation neck specimens to the standard uptake value (SUV) in post-treatment FDG PET-CT scans.[48] They found that an SUV of less than 3.0 had a negative predictive value of 100% and a positive predictive value of 80% for the residual tumor in the neck specimen.

Other authors have found that FDG PET-CT scanning adds little value over traditional CT scanning in determining who requires a postchemoradiation neck dissection for advanced neck disease.[49] The exact role of FDG PET-CT is still unclear and requires further investigation.