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

INTRODUCTION — Squamous cell carcinoma is the predominant malignancy that occurs in the oral cavity. Minor salivary gland cancers and sarcomas are less common.

The initial approach to squamous cell carcinoma of the oral cavity and the treatment of stage I and II (early) disease are discussed here. The treatment of locoregionally advanced oral cavity cancer, salivary gland tumors, and metastatic and recurrent head and neck cancers are discussed separately.

●(See "Treatment of locoregionally advanced (stage III and IV) head and neck cancer: The oral cavity".)

●(See "Salivary gland tumors: Treatment of locoregional disease".)

●(See "Malignant salivary gland tumors: Treatment of recurrent and metastatic disease".)

●(See "Treatment of metastatic and recurrent head and neck cancer".)

ANATOMY — The oral cavity extends from the skin-vermillion junction of the lips (a cutaneous site previously included as part of the oral cavity) to the junction of the hard and soft palate above, and to the line of circumvallate papilla of the tongue below. The anterior tonsillar pillars and glossotonsillar folds serve as the lateral boundaries between the oral cavity and oropharynx (figure 1).

Oral cavity cancer can involve the following anatomic sites (figure 1):

●Lip

●Lower alveolar ridge

●Upper alveolar ridge

●Retromolar trigone (retromolar gingiva)

●Hard palate

●Oral tongue (anterior two-thirds of the tongue)

●Floor of the mouth

●Buccal mucosa

Rarely, an oral cavity cancer can originate within the mandible (eg, osteosarcoma). Most odontogenic lesions, including the vast majority of ameloblastoma, do not metastasize or involve lymph nodes, although they may be locally destructive.

EPIDEMIOLOGY AND RISK FACTORS — There are approximately 378,000 cases of oral cavity cancer worldwide in 2020, which caused an estimated 178,000 deaths [1]. There is an increasing incidence of oral tongue cancer among White females less than 50 years old [2].

Some risk factors for the development of oral cavity cancers include:

●Environmental exposures – Tobacco smoking and alcohol use are the principal risk factors for developing oral cavity squamous cell carcinoma in the Western world. The chewing of betel nut quid is a significant factor throughout Asia and India. Oral tobacco use, periodontal disease, radiation, and conditions associated with immune deficiency or dysregulation have also been implicated [3-6]. For example, oral tobacco and betel nut quid chewing are important risk factors for buccal mucosa cancers [5,7]. (See "Epidemiology and risk factors for head and neck cancer", section on 'Risk factors'.)

●Is human papillomavirus a risk factor? – Infection with oncogenic human papillomavirus (HPV), especially HPV 16, is associated with the rising incidence of tonsil and base of tongue cancer. However, the percentage of oral cavity cancers associated with HPV infection is significantly less, and the majority of p16 positive tumors in this location will be negative for HPV unless they originate near the circumvallate papillae [8,9]. A possible etiologic relationship is unclear, as are the prognostic implications of HPV infection in patients with oral cavity squamous cell carcinoma [10]. Routine HPV testing of oral cavity cancer is not recommended. (See "Epidemiology, staging, and clinical presentation of human papillomavirus associated head and neck cancer", section on 'Nonoropharyngeal carcinoma'.)

CLINICAL PRESENTATION AND DIAGNOSIS — Oral cavity tumors often present with local invasion, tissue destruction, and lymph node metastases (table 1). (See 'Staging' below.)

Oral cavity cancers also tend to invade soft tissue early. Depth of invasion (DOI) is an important prognostic factor and is included in the tumor, node, metastases (TNM) classification system. Involvement of bone is usually limited to larger tumors, except for those that originate in gingival mucosa. Patients with oral cavity tumors rarely present distant metastases. Further details on the clinical presentation of oral cavity cancer are discussed separately. (See "Overview of the diagnosis and staging of head and neck cancer", section on 'Clinical presentation'.)

The diagnosis of oral cavity cancer requires a tissue diagnosis of a suspicious oral lesion or lymph nodes using biopsy or fine needle aspiration. A biopsy is not a definitive therapeutic procedure and should simply obtain tissue adequate for a diagnosis and not obliterate the site of the lesion. (See "Overview of the diagnosis and staging of head and neck cancer", section on 'Diagnosis and staging evaluation'.)

PRETREATMENT EVALUATION — Once the diagnosis of oral cavity cancer is made, a careful pretreatment evaluation is required to assess the extent of locoregional disease, as well as possible regional or distant metastasis. Clinical factors that are critical for treatment planning include tumor size, the extent or depth of invasion (DOI) which is not tumor thickness, and the presence or absence of regional lymph node metastases [11]. In particular, the DOI of early-stage squamous cell carcinoma involving the oral tongue is particularly difficult to assess preoperatively and is typically obtained on postoperative histology. Of note, DOI does not include the portion of the tumor superficial to the epithelium.

All patients should undergo a thorough inspection and palpation of the oral cavity. Additionally, the following pretreatment imaging is obtained:

●Computed tomography (CT) – CT with intravenous contrast is widely used to detect bone invasion; CT has largely replaced older techniques, such as plain radiography and panoramic radiography of the maxilla and mandible.

●Magnetic resonance imaging (MRI) – MRI may complement or replace CT scanning by providing better visualization of soft tissue involvement and gross perineural spread while also having good accuracy in detecting bone invasion. Whereas CT scanning may be limited by metallic dental restorations and cortical defects, MRI is limited by motion artifact and inflammatory reactions.

●Positron emission tomography (PET)/CT – Combined PET/CT scans may add accuracy in evaluating the extent of the primary tumor and aid in target delineation in patients receiving definitive radiation therapy (RT). PET scanning may help to identify pathologically involved lymph nodes or to clarify the status of suspicious nodes on CT or MRI that do not meet traditional size-based criteria for classification as being abnormal [12,13].

●Sentinel lymph node biopsy (SLNB) – SLNB is an emerging technique that may provide an additional option for the assessment of regional lymph node nodes. (See 'Sentinel lymph node biopsy' below.)

STAGING — The tumor, node, metastases (TNM) staging system of the American Joint Committee on Cancer (AJCC) and the Union for International Cancer Control (UICC) is used to classify lip (excluding external and vermillion) and oral cavity carcinoma (table 1) [14,15]. The 2017 updated staging system includes important changes for patients with early-stage oral cavity cancer. Whereas prior iterations included only the maximum tumor size in a single plane (usually the surface), the AJCC eighth edition staging system now also incorporates maximum depth of invasion (DOI). Patients with >5 mm DOI have a minimum tumor stage of T2; patients with >10 mm depth of invasion have a minimum tumor stage of T3. Thus, patients with small but more deeply invasive tumors effectively have higher-stage tumors. Patients with more extensive local disease, regional lymph node involvement, or distant metastasis have stage III or IV disease. (See "Overview of the diagnosis and staging of head and neck cancer".)

MANAGEMENT OF THE PRIMARY TUMOR

Treatment modalities — Primary surgery and definitive radiation therapy (RT) are both treatment options for patients with early-stage oral cavity cancer. For most patients with early-stage oral cavity tumors, we suggest treating the primary tumor with initial surgery (with or without adjuvant RT) rather than definitive RT. We alternatively offer RT to patients who are ineligible for or decline surgery. For example, surgical resection of the primary tumor may not be feasible or desirable if it results in significant morbidity or functional impairment. However, the use of surgery versus RT for the primary tumor has not been directly compared in randomized trials.

For patients with stage I or II oral cavity tumors, surgery with or without adjuvant RT is associated with less long-term morbidity than RT [16,17] and offers more detailed pathologic staging, including depth of invasion (DOI). Definitive RT is reserved for patients who are ineligible for surgery (eg, due to medical comorbidities) or for whom surgical resection would result in severe functional impairment [18,19]. The latter circumstance is uncommon in stage I and II oral tongue and floor of mouth primaries but is sometimes applicable to early tumors of the retromolar trigone and lip since external RT can achieve similar outcomes to surgery with acceptable morbidity.

Surgery — For patients who undergo resection of the primary tumor with close (<2 mm) or positive surgical margins, we suggest reresection rather than postoperative RT. For those who are ineligible for or decline reresection, we offer adjuvant RT to the primary site.

Every attempt should be made to ensure negative resection margins since positive margins are associated with a worse prognosis [20,21]. If feasible, reresection of any positive margin is preferred. Otherwise, postoperative RT, with or without chemotherapy, is indicated. Close margins, typically defined as less than 5 mm, may also portend a worse prognosis, although a more recent study suggests a narrower margin may be acceptable [22]. (See 'Adjuvant radiation therapy' below.)

Traditional surgical approaches are generally used for patients with early-stage oral cavity cancer. Although minimally invasive surgery, such as transoral laser resection or robotic surgery, has been used, the relative advantages of one surgical technique versus another are not well established [23,24]. The primary benefit of minimally invasive surgery is its ability to access structures such as the oropharynx and larynx, which are not easily approached with standard instruments. However, early-stage oral cavity tumors are usually accessible via traditional transoral approaches.

Most early-stage oral cavity cancers can be excised without major, long-term, functional or cosmetic deformity. Swallowing function is generally excellent with appropriate reconstruction and postoperative rehabilitation. Likewise, short-term perturbations in speech (eg, difficulty pronouncing the letter "t" after tongue surgery) can be corrected to normal or near-normal function. As an example, an early-stage oral cavity cancer patient employed in telephone sales would generally be expected to be able to return to full-time work within three months of surgery.

Acute surgical complications can include infection, bleeding, aspiration, wound breakdown, flap loss, and fistula [25]. Surgical procedures such as hemiglossectomy, maxillectomy, and mandibulotomy can cause functional defects in speech and swallowing, although these may be reduced by optimal reconstruction. (See "Mandibular and palatal reconstruction in patients with head and neck cancer" and "Management of acquired maxillary and hard palate defects" and "Speech and swallowing rehabilitation of the patient with head and neck cancer".)

Radiation therapy — Both external beam RT using contemporary conformal techniques and brachytherapy can have a role in the management of early-stage oral cavity cancer. The general principles of RT and the role of radiation dose and schedule are discussed elsewhere. (See "General principles of radiation therapy for head and neck cancer" and "Definitive radiation therapy for head and neck cancer: Dose and fractionation considerations".)

Small tumors in most oral cavity sites can be managed with intraoral cone or interstitial brachytherapy in order to minimize exposure of normal tissue. These techniques do not treat regional lymph node basins, however, and therefore are only appropriate to use as a single modality in selected stage I patients where the risk of occult nodal involvement is very low. Factors associated with high-risk disease or lymph node involvement include deep invasion and lymphovascular invasion.

External beam RT to the draining lymph node regions is used as the primary mode of irradiation when regional lymph nodes are at significant risk for subclinical involvement, and intraoral cone-beam radiation or interstitial brachytherapy may be added as a boost to the primary tumor. (See 'Adjuvant radiation therapy' below.)

Significant acute RT toxicities include mucositis, skin reaction, loss of taste, and dysphagia. Late complications may include skin and soft tissue atrophy and fibrosis, osteoradionecrosis (particularly with brachytherapy), lymphedema, and trismus. Xerostomia commonly occurs during the course of RT and persists after treatment. (See "Management and prevention of complications during initial treatment of head and neck cancer" and "Management of late complications of head and neck cancer and its treatment".)

Xerostomia may recover after treatment with modern RT techniques, such as three-dimensional conformal or intensity-modulated radiation therapy. However, recovery of xerostomia in patients with oral cavity tumors may be more limited than with other head and neck sites because of higher doses of radiation to the submandibular glands and larger volumes of oral mucosa and minor salivary glands in the treatment fields. (See "Management and prevention of complications during initial treatment of head and neck cancer", section on 'Highly conformal RT technique'.)

Site-specific management

Lip (squamous cell carcinoma) — Squamous cell carcinoma can arise on the inner surface of the lip and is included with other oral cavity cancers. Cancers of the vermillion and external surface of the lip are classified now with skin cancers, as they are often the result of sun exposure and share characteristics with many skin cancers. Surgical excision is generally preferred for management. (See "Treatment and prognosis of low-risk cutaneous squamous cell carcinoma (cSCC)".)

The natural history and hence prognosis of squamous cell carcinoma of the lip varies depending upon whether the lower or upper lip is involved, and this has implications for management:

●Lower lip – Squamous cell carcinoma is more common than basal cell carcinoma in the lower lip. Lymph node metastasis usually is not seen, except in the presence of a large lesion or multiple recurrent primary tumors.

Maximum DOI is a predictor for metastatic spread to regional lymph nodes for squamous cell carcinoma of the lip and, therefore, may be important for treatment planning and assessment of prognosis [25,26]. In one report of 27 patients with squamous cell carcinoma of the lower lip, there was a significant difference in mean DOI between patients with or without cervical lymph node metastases (5.6 versus 3.8 mm, respectively) [26]. Whether DOI is independent of tumor stage as a prognostic factor is uncertain, but DOI (rather than thickness) is now a part of the staging system.

●Upper lip – Squamous cell carcinomas that arise in the upper lip and oral commissure have a worse prognosis than those originating in the lower lip. These tumors grow more rapidly, ulcerate sooner, and metastasize earlier (picture 1 and picture 2) [27-29].

Upper lip cancers also have the potential for invasion into the premaxilla and nasal cavity, which can add considerable complexity to the treatment planning and ultimate functional outcome. Therefore, a more aggressive treatment strategy including prophylactic treatment of the neck is generally advocated for even early-stage upper lip squamous cell carcinomas. (See 'Management of the neck' below.)

Cosmetic and functional outcomes are important in planning treatment because of the role of the lips in speech, facial expression, and eating. The extent of surgical excision depends entirely upon the size of the lesion. The optimal margin for resection has not been defined [30,31], but a margin ≥5 mm on final pathology is generally recommended.

There are numerous reconstructive options to restore function after surgery. Defects that involve less than two-thirds of the lip can be closed primarily, while defects involving up to three-quarters of the lip can be reconstructed with full thickness pedicled flaps from the uninvolved lip (Abbe-Estlander flaps) [32]. More complicated reconstructive options are available for larger defects, including the Karapandzic flap [33].

Mohs micrographic surgery, which is effective for treating early-stage skin cancer, has been used to treat early lip cancers but has not been directly compared with standard surgical excision [34,35]. Therefore, we advise caution when treating lip cancer with Mohs surgery, particularly for tumors >2 cm, since narrow margins may be less effective for these tumors. (See "Mohs surgery".)

In general, the prognosis for early-stage squamous cell carcinoma of the lip is very good, with 10-year recurrence-free survival rates of 94 and 78 percent for stage I and II disease, respectively [26].

Lower alveolar ridge and retromolar trigone — The retromolar trigone is the small mucosal space that begins at the third molar of the mandible and extends posteriorly and cranially to the maxillary tuberosity. It is continuous with the buccal mucosa, upper and lower gingiva, maxillary tuberosity, anterior tonsillar pillar, soft palate, and mouth floor.

Even early cancers of the retromolar trigone have a high local recurrence rate due to microscopic extension into the mandible and a high probability of occult regional lymph node metastases [36-38]. The five-year disease-free survival (DFS) rate for early-stage retromolar trigone cancer is approximately 70 percent [39-41].

Surgical therapy involves wide local excision, but resection can be technically challenging. Although small tumors may be managed with a transoral approach, many will require a lip-splitting cheek flap for exposure. Marginal mandibulectomy may be required in order to achieve tumor-free margins.

The degree of tumor invasion must be assessed before surgery as well as possible since this will determine the deep surgical margin [42]. A planned intraoperative mucosal resection margin of 1 cm is recommended to achieve a final pathologic margin >5 mm since there can be substantial shrinkage of the tumor specimen in formalin [43]. By contrast, a deep bone margin of 1 cm is unnecessarily morbid, and instead, a margin of one tissue layer deeper than the greatest DOI is suggested. Preoperative assessment of bone invasion is notoriously difficult.

Treatment of cancers of the lower alveolar ridge is similar to that for lesions arising in the retromolar trigone. These lesions often have mandibular involvement at presentation.

The status of the teeth plays a critical role in treatment of cancers involving the upper and lower alveolar ridge. Many patients with oral cavity cancer are edentulous or have very poor dentition. In these patients, porous cortical bone serves as a poor barrier to invasion, and the risk of mandibular invasion is increased, even for early-stage cancers. Therefore, edentulous patients with even minimal cortical bone invasion should be treated with a segmental resection of bone rather than a marginal resection. By contrast, healthy teeth are a better barrier to direct tumor invasion into the medullary space of the bone, and patients with healthy teeth with an intact dental ligament, and with early-stage cancers rarely have mandibular invasion.

Upper alveolar ridge and hard palate — Both surgery and RT offer excellent control for early-stage squamous cell cancers of the upper alveolar ridge and hard palate. However, surgery is generally preferred because of the late effects of radiation, including osteoradionecrosis of the maxilla. Except for the most superficial cancers, surgery usually involves resection of bone including alveolectomy, palatectomy, or infrastructure maxillectomy (ie, resection of the palatine process of the maxillary bone) [44]. The resulting surgical defects may be closed with local mucosal rotation flaps or free tissue transfer reconstruction for larger defects. However, some surgeons prefer prosthetic rehabilitation with a removable palatal obturator, which permits better inspection of the defect for disease recurrence. (See "Management of acquired maxillary and hard palate defects".)

Buccal mucosa — Buccal mucosa cancers are often neglected or misdiagnosed as an infection or consequence of trauma, and thus, rarely present as T1 lesions. For patients who are diagnosed with early-stage buccal mucosa tumors, three-year DFS rates of 75 to 85 percent for stage I and 65 percent for stage II cancers have been reported [45-48].

Surgery is typically preferred for buccal mucosa cancers, despite high local recurrence rates and technical challenges [49]. Early-stage buccal mucosa cancers may appear easy to treat (picture 3). However, exposure of the cancer can be difficult via a transoral approach, which makes it difficult to obtain clear radial margins in an en bloc fashion. Furthermore, the short distance between the buccal mucosa and the buccal space permits early invasion to deep structures or to the skin of the anterior cheek. The surgeon must choose between taking a thin deep margin and risking recurrence versus removing skin and reconstructing both the inner and outer cheek surfaces. Although more aggressive surgery including exenteration of the buccal space and parotidectomy may improve oncologic results, the disfigurement and morbidity associated with these procedures are considerable.

The buccal mucosal surface must be reconstructed regardless of the depth of resection; inadequate soft tissue coverage will result in severe, irreversible trismus. Thus, free tissue transfer reconstruction (eg, radial forearm flap) is recommended for all but the smallest buccal cancers. In these cases, skin graft reconstruction will suffice. Even with adequate reconstruction, trismus is a significant clinical problem, and jaw exercises are required for the best functional outcome.

Postoperative radiation or chemoradiation is often necessary to maximize the likelihood of locoregional control. If RT is used, the initial treatment volume includes the primary tumor, with at least 2 cm margins. An intraoral stent may be used to displace the tongue medially to reduce the dose delivered to this organ. (See 'Radiation therapy' above.)

Oral tongue — Surgery is generally recommended for oral tongue cancer if good functional rehabilitation can be achieved with reconstruction (picture 4 and picture 5) [50,51].

An en bloc partial glossectomy with negative margins can preserve speech and swallowing for most stage I and II lesions of the oral tongue. The choice of reconstruction and intensity of rehabilitation determine the ultimate functional outcome (picture 6) [51]. (See "Speech and swallowing rehabilitation of the patient with head and neck cancer".)

Assessing surgical resection margins can be difficult. Deep tongue muscle margins are not found in a single plane, in contrast with the radial mucosal margins. In addition, striated tongue muscle fibers shred or fragment with tissue handling during and after surgery, leaving a less reliable surface to assess margin status. Therefore, close deep surgical margins should be interpreted with caution, and more aggressive treatment may be indicated compared with close radial mucosal margins or close margins in other disease sites.

Definitive RT of oral tongue cancer typically utilizes external beam RT. Small (≤10 mm) and superficial lesions can be treated with either an intraoral cone or interstitial brachytherapy alone. (See "General principles of radiation therapy for head and neck cancer" and "Definitive radiation therapy for head and neck cancer: Dose and fractionation considerations".)

Although the five-year overall survival (OS) for early-stage oral tongue cancer reported by a single institution was quite good (89 and 95 percent for stage I and II cancers), the five-year survival calculated from a population-based database was more modest (67 and 51 percent, respectively) [50,51]. Tongue cancer is often believed to have a more aggressive course in younger patients (ie, <40 years), but most reports have not found a difference in survival between younger and older patients [44]. However, patients are at significant risk for the development of a second primary head and neck cancer. (See "Second primary malignancies in patients with head and neck cancers".)

Floor of mouth — Primary tumor staging for floor of mouth cancers can be challenging due to difficulties in tumor inspection and measurement. These tumors tend to be locally invasive. There is a high risk of neck lymph node metastases, with occult node metastases reported in 21 and 62 percent of patients with clinical stage I and II disease [52,53]. (See 'Management of the neck' below.)

Surgery has been the preferred approach due to the concern of risk of radiation-induced bone necrosis and other long-term complications in patients treated with definitive RT. (See "Management of late complications of head and neck cancer and its treatment".)

Early-stage floor of mouth cancer can generally be managed via transoral excision. Care should be taken to avoid the lingual nerve, if oncologically feasible, as this nerve is relatively superficial in the floor of mouth and easily injured.

Small to medium defects can be closed primarily, left to close by secondary intention, or reconstructed with a skin graft. Large defects are best repaired with a vascularized graft, such as a radial forearm free tissue transfer or submental island skin flap, to avoid contracture of the tongue and difficulty with speech.

Results with definitive RT are similar to those attained with surgery [54,55]. Small (<1 cm) or superficial lesions (<4 mm thick) can be treated with either an intraoral cone or interstitial brachytherapy alone. An intraoral cone is more suitable for lesions located centrally or in the anterior part of the mouth. For lesions extending close to or into the tongue, an intraoral stent or bite block helps position the tongue. When a boost is given with an intraoral cone, it is delivered before external beam RT so that mucositis does not impede visualization of the lesion. When a boost is given with interstitial brachytherapy, it is delivered after external beam RT to allow for tumor shrinkage. (See 'Radiation therapy' above.)

Five-year OS rates for stage I and II cancers of 95 and 85 percent have been reported [52,54,55].

MANAGEMENT OF THE NECK — For patients with early-stage oral cavity cancers without clinical lymph node involvement, management of the neck is an essential component of treatment, as cervical lymph node metastases in these patients are associated with inferior survival rates.

Patients with lymph node involvement or invasion of deeper structures detected on initial pretreatment evaluation have locoregionally advanced (American Joint Committee on Cancer [AJCC] stage III or IV (table 1)) disease. For such patients, treatment is directed against both the primary tumor and neck, and management is discussed separately. (See "Treatment of locoregionally advanced (stage III and IV) head and neck cancer: The oral cavity".)

Treatment modalities — Available treatment modalities for the neck include elective neck dissection or elective radiation therapy (RT). While sentinel lymph node biopsy (SLNB) is a diagnostic and not a therapeutic technique, it can be used to guide subsequent management.

Elective neck dissection — Elective neck dissection is preferred over observation in most patients with resectable oral cavity tumors, as this approach improved both overall survival (OS) and disease-free survival (DFS) in randomized trials [56-58] and provides pathologic staging.

The optimal extent and nodal yield of neck dissection is relatively well established. For clinical N0 tumors, elective lymph node dissection should include at least 18 lymph nodes. Of course, limiting the extent of dissection reduces the potential for morbidity. A supraomohyoid neck dissection (levels I to III plus IV; for subsites other than oral tongue, levels I to III) is typically sufficient for clinically N0 oral cavity cancer since level IV and V lymph nodes are rarely involved without clinical disease at other levels (figure 2) [59-62]. This approach includes removal of the submandibular gland but preserves the spinal accessory nerve, the internal jugular vein, and the sternocleidomastoid muscle. Sublevel IIB may also be preserved in patients with early-stage oral cavity cancer, thus minimizing the risk of shoulder dysfunction from damage to the spinal accessory nerve, which runs through sublevel IIB.

Evidence supporting elective neck dissection includes a meta-analysis of five randomized trials that included patients with tongue and floor of mouth tumors and found that elective neck dissection of clinically negative neck nodes improved overall survival (HR 0.64, 95% CI 0.50-0.83) and reduced locoregional recurrence (HR 0.58, 95% CI 0.43-0.78) in comparison with therapeutic neck dissection [58]. Elective neck dissection was associated with more adverse events (risk ratio 1.31, 95% CI 1.11-1.54).

Similarly, one trial reported outcomes of elective neck dissection in 500 patients with early-stage (T1-T2N0) oral cavity tumors whose nodal status was assessed by ultrasound [56]. Most patients had primary tumors of the oral tongue (85 percent); and a minority had tumors of the buccal mucosa (14 percent) or tumors involving the floor of the mouth (1 percent). Following excision of the primary tumor, patients were randomly assigned to either elective node dissection or to observation followed by therapeutic node dissection for recurrence. Patients assigned to an elective lymph node dissection underwent an ipsilateral selective neck dissection that included clearance of level I, II, and III lymph nodes. If disease was detected during the elective node dissection, surgery was extended to include level IV and V nodes. Patients assigned to a therapeutic neck dissection underwent a modified neck dissection (levels I to V) if a nodal relapse was detected. Patients were also secondarily randomized to be followed by physical examination only or by ultrasound (but not CT or positron emission tomography [PET]).

At a median follow-up of 39 months, compared with therapeutic neck dissection, elective neck dissection improved both OS (three-year OS 80 versus 67.5 percent; hazard ratio [HR] 0.64, 95% CI 0.45-0.92) and DFS (three-year DFS 69.5 versus 45.9 percent; HR 0.44, 95% CI 0.34-0.58). On subset analysis, survival benefit was limited to patients with tumor depth >3 mm, although the number of patients with a smaller DOI was limited. It is unclear whether results might differ if eligibility was limited to patients with a negative neck on baseline CT or PET, or if patients treated without neck dissection had more intensive follow up imaging using CT or PET, rather than ultrasound.

Investigations are ongoing looking at the role, if any, that imaging may play in allowing some patients to avoid elective neck dissection. For example, in one study of patients with T2 to T4 squamous cell carcinoma of the head and neck who were clinically N0, a PET/CT at the time of diagnosis had a negative predictive value of 87 percent [63]. However, additional studies are needed to determine if elective neck dissection can be omitted outside of clinical trials.

Elective radiation therapy — Elective therapy to the neck can also be performed using RT in patients whose primary tumor is being treated using RT. While we prefer elective neck dissection which offers pathologic staging, RT is a reasonable alternative for patients who are ineligible for or decline surgery. Elective RT to the neck follows the same indications as for elective neck dissection, with attention to the same nodal levels.

Sentinel lymph node biopsy — Select early-stage oral tongue cancers may be appropriate for staging via SLNB. This technique utilizes preoperative lymphoscintigraphy, intraoperative blue dye, and a handheld gamma probe, similar to the approach used for cutaneous melanoma. (See "Evaluation and management of regional nodes in primary cutaneous melanoma" and "Evaluation and management of regional nodes in primary cutaneous melanoma", section on 'SLNB timing and technique'.)

SLNB is usually performed at the same time as surgical resection of tumor. This procedure can detect disease in sentinel nodes ipsilateral or contralateral to the primary tumor, depending upon the lateralization of the primary tumor and the pattern of lymphatic drainage identified on lymphoscintigraphy [64].

SLNB can be used for tumors with depth of invasion (DOI) <3 mm or tumors that are midline or approaching midline (regardless of DOI) (algorithm 1). If the SLNB is negative for disease, it can often replace a planned elective neck dissection, and the neck can be observed. If the SLNB is positive for disease, nodal dissection can subsequently be performed either unilaterally or bilaterally, depending upon the results of the SLNB. If the SLNB is positive for disease on the ipsilateral side only, we perform a unilateral neck dissection. If the SLNB is positive for disease on the contralateral side or if bilateral drainage is observed by lymphoscintigraphy, we perform a bilateral neck dissection.

Clinical judgment should be used in deciding between elective node dissection and SLNB. SLNB should be performed at centers with expertise in this technique, as specialized training is required for the procedure. SLNB has been more widely adopted in Europe than in the United States for early-stage oral cavity cancer. In observational studies, the negative predictive value of this procedure is estimated at 95 percent [65]. We also encourage participation in an ongoing randomized trial (NRG-HN006) evaluating the use of SLNB versus elective neck dissection in various early-stage oral cavity tumor sites.

SLNB is a safe and feasible technique [65-68] with similar survival and better functional outcomes compared with elective neck dissection in randomized trials [69-71]. In a phase III trial (Senti-MERORL), 307 patients with surgically resectable cT1-T2N0 oral cavity or oropharyngeal cancers were randomly assigned to elective nodal dissection or SLNB [69]. Patients with a negative SLNB received nodal biopsy alone, and those with a positive SLNB received nodal dissection either during primary tumor surgery (if detected on intraoperative histologic analysis) or as a second surgical procedure (if detected postoperatively on immunohistochemistry). Bilateral nodal dissection was performed if contralateral positive sentinel lymph nodes were detected or if bilateral drainage was observed by lymphoscintigraphy. Among the 279 evaluable patients, five-year neck node recurrence-free survival (89.6 versus 89.4 percent), disease-specific survival (88.6 versus 87.1 percent), and OS (81.8 versus 82.2 percent) were similar between the node dissection and the SLNB groups. Compared with SLNB, neck dissection resulted in worsened postoperative functional outcomes until six months after surgery, based on evaluation of neck-shoulder impairment, arm abduction, and physiotherapy prescription rates.

However, interpretation of the study results is limited by the inclusion of patients with oropharyngeal cancers, which are often human papillomavirus (HPV) associated tumors that frequently present with nodal disease. Although a majority of patients received surgery alone, approximately one-fourth of patients received additional adjuvant therapy (RT, chemoradiation, or brachytherapy). Additionally, the use of neck node recurrence-free survival may be an inadequate surrogate endpoint for OS. More rigorous functional outcome evaluations than those utilized in this study are also available [72]. The study also did not perform an assessment based on DOI. Further randomized trial data are necessary to confirm which patient groups benefit from this approach.

A separate randomized noninferiority trial conducted in 271 patients with early-stage oral cavity tumors also confirmed similar survival outcomes between SLNB and elective neck dissection, but better neck functional outcomes with SLNB [70].

Management of the neck based on primary tumor site — The treatment approach to the neck is guided by the location of the primary tumor, tumor size and DOI, as well as available clinical expertise in the procedure of choice. Although there is disagreement regarding some details, our approach to management of the neck is generally consistent with guidelines from the American Society of Clinical Oncology (ASCO) and the National Comprehensive Cancer Network (NCCN) [73,74] and is as follows:

Lip — The management of the neck in patients with lip cancers is based primarily on size. Lip squamous cell cancers involving the vermillion are now classified as cutaneous malignancies rather than oral cavity tumors [14,15]. (See "Treatment and prognosis of low-risk cutaneous squamous cell carcinoma (cSCC)".)

●Lower lip – For patients with T1 lower lip cancers, we suggest observation rather than elective neck dissection, as the latter procedure is usually not performed in these tumors. For superficial lower lip cancers with DOI <3 mm, SLNB may be offered as an alternative to observation. SLNB may be a particularly attractive option for such tumors since lymphatics from the lip drain to both sides of the neck. (See 'Sentinel lymph node biopsy' above.)

For T2 or larger lower lip cancers, we suggest elective neck dissection rather than observation or SLNB. Elective neck dissection should at least include levels IA and IB (suprahyoid dissection) since lower lip cancers usually do not metastasize to lower cervical nodes without first invading facial, submental, and submandibular lymph nodes [75].

●Upper lip – For patients with early-stage upper lip squamous carcinomas, elective neck dissection is generally advocated, given the aggressive nature of these tumors. (See 'Elective neck dissection' above.)

Lower alveolar ridge and retromolar trigone — For all patients with lower alveolar ridge and retromolar trigone cancers, regardless of size, we suggest elective neck dissection rather than SLNB, including levels I to III (figure 2), given the particularly high incidence of occult nodal metastases [37]. (See 'Elective neck dissection' above.)

Upper alveolar ridge and hard palate — For patients with early-stage cancers of the upper alveolar ridge and hard palate, we suggest elective neck dissection rather than SLNB, particularly for those with T2 disease or greater. Although occult nodal metastases were previously thought to occur infrequently in these tumors, data suggest that the incidence is high enough to warrant treatment with elective neck dissection, particularly for those with ≥T2 disease [76,77]. The challenge in this location is that the oral mucosa is very thin over the bone, and bone invasion can occur relatively easily, upstaging the cancer to T4 disease. For these reasons, SLNB is particularly challenging for these tumors and is less preferred. (See 'Posttreatment surveillance' below.)

Buccal mucosa — Most patients with buccal mucosa tumors are treated with elective neck dissection. Additionally, the facial lymph nodes adjacent to the facial artery and vein at the mandibular ramus are at high risk of metastasis, and particular attention should be paid to this area during surgery.

Oral tongue and floor of mouth — For patients with oral cavity cancers involving the oral tongue or floor of mouth, the treatment approach is based on tumor size, DOI, and proximity to midline. In particular, DOI is one of the most important factors that determines both disease staging and management of the neck [14,15]. While other retrospective studies suggest a DOI threshold of 4 mm, we use 3 mm as an appropriate cutoff.

DOI can be grossly estimated clinically but requires histology for definitive determination. The DOI of the primary tumor may be unknown prior to surgery for early-stage oral cavity cancers. In that case, once the primary tumor has been surgically resected, the surgeon has the option to either immediately assess intraoperative frozen sections (using the assistance of a pathologist with expertise in head and neck tumors) to determine the need for elective neck dissection during the same operation, or wait for permanent histology specimens and perform elective neck dissection in a separate operation. Improved imaging techniques, including functional or molecular-based studies prior to surgery, may eventually prove useful in selecting N0 oral cavity cancer patients for either observation, SLNB, or elective neck dissection [63]. (See 'Sentinel lymph node biopsy' above.)

Our treatment approach for oral tongue tumors, based on DOI, tumor laterality, and available clinical expertise in the procedure of choice are discussed below (algorithm 1). The approach to most patients with floor of mouth tumors is typically similar to that of oral tongue tumors. However, SLNB is less reliable for floor of mouth tumors. Due to the close proximity of the radiotracer injection site (ie, the primary tumor-bearing area) to the lymph nodes at risk, radioactivity from the injection site may mask sentinel node activity in level I lymph nodes (figure 2) [64].

Depth of invasion >3 mm, tumor lateralized — For patients with early-stage, lateralized oral tongue tumors with DOI >3 mm (table 1), either unilateral elective neck dissection or SLNB (with further surgery in the event of sentinel node involvement) are appropriate options. Either approach is reasonable, as both demonstrated similar OS advantage when directly compared in one randomized trial [69]. Additionally, elective neck dissection improved OS over observation with therapeutic neck dissection at the time of recurrence in a separate randomized trial [56]. The choice of therapy typically is based upon the skill and expertise of the treating surgeon. These data are discussed above. (See 'Sentinel lymph node biopsy' above and 'Elective neck dissection' above.)

●Unilateral elective neck dissection – For patients who receive unilateral elective neck dissection ipsilateral to the primary tumor, treating levels I to III/IV reduces the risk of locoregional recurrences [78-81]. Patients may have "skip metastases" with involvement of level III or IV, without involvement of levels I and II (figure 2) [61]. Thus, elective neck dissection of levels I to IV may be more appropriate than a supraomohyoid dissection of levels I to III. Levels IIB and IV are dissected at the discretion of the surgeon. Level V dissection is usually unnecessary. (See 'Elective neck dissection' above.)

●Sentinel lymph node biopsy – The approach to SLNB in such patients is discussed above. (See 'Sentinel lymph node biopsy' above.)

Depth of invasion >3 mm, tumor midline or approaching midline — For patients with early-stage oral tongue tumors with DOI >3 mm (table 1) that approach or involve the midline, either bilateral elective neck dissection or SLNB (with further surgery in the case of sentinel node involvement) are appropriate options. The choice of therapy typically is based upon the skill and expertise of the treating surgeon. (See 'Sentinel lymph node biopsy' above and 'Elective neck dissection' above.)

Depth of invasion ≤3 mm, tumor lateralized — Only patients with the most superficial oral tongue cancers (eg, ≤1 mm DOI) may be offered observation. For patients with early-stage, lateralized oral tongue tumors with DOI >1 mm and ≤3 mm (table 1), we suggest SLNB rather than proceeding directly to a unilateral elective neck dissection, provided patients are being treated at a center with expertise in this technique. If such expertise is not available, we suggest a unilateral elective neck dissection rather than other surgeries.

Depth of invasion ≤3 mm, tumor midline or approaching midline — For patients with early-stage oral tongue tumors with DOI ≤3 mm that approach or involve the midline, we suggest SLNB rather than proceeding directly to a bilateral elective neck dissection, provided that patients are being treated at a center with expertise in this technique. If such expertise is not available, we suggest a bilateral elective neck dissection rather than other surgeries.

For those with primary tumors approaching the midline, SLNB can be used to assess the dominant lymph node drainage pattern. When performed properly, a negative SLNB may replace planned elective neck dissection. However, caution should be used for patients with floor of mouth tumors; the accuracy of SLNB in such tumors may be lower than that of oral tongue tumors [67,82], due to the close proximity of the radiotracer injection site (ie, the primary tumor-bearing area) to the lymph nodes at risk.

ADJUVANT RADIATION THERAPY — We suggest adjuvant radiation therapy (RT) to the primary site and unilateral or bilateral neck rather than observation for patients with high-risk features on postoperative pathology to reduce the risk of locoregional relapse. Such high-risk features include positive or close final resection margins without reresection, bone invasion, and pathologically positive lymph nodes. Adjuvant RT may also be indicated in patients with depth of invasion >4 mm, lymphovascular, or perineural invasion in the primary tumor [83], even in the setting of a negative unilateral neck dissection.

Adjuvant concurrent chemoradiation is reserved for patients with positive surgical margins or extranodal extension. Further details on the indications for adjuvant RT, with or without chemotherapy, in head and neck cancer are discussed separately. (See "Adjuvant radiation therapy or chemoradiation in the management of head and neck cancer".)

The combination of surgery and RT is associated with certain treatment-related morbidities, such as xerostomia and fibrosis. Contemporary conformal RT techniques should be used to minimize such toxicities, particularly late xerostomia [84]. (See "Adjuvant radiation therapy or chemoradiation in the management of head and neck cancer" and "General principles of radiation therapy for head and neck cancer" and "Overview of approach to long-term survivors of head and neck cancer", section on 'Xerostomia' and "Overview of approach to long-term survivors of head and neck cancer", section on 'Neck stiffness'.)

For patients with high-risk features who are ineligible for or unable to tolerate adjuvant RT, an alternative strategy is close monitoring of the neck for two to three years with ultrasound, CT, or positron emission tomography (PET) scanning.

There is controversy regarding the use of adjuvant RT in patients without pathologic lymph node involvement. While our approach to offer adjuvant RT in a subset of patients with high-risk features is consistent with the National Comprehensive Cancer Network (NCCN) guidelines, it differs from American Society of Clinical Oncology (ASCO) guidelines, which do not recommend RT in this setting.

In a retrospective observational study of 164 patients with early-stage oral tongue cancer treated with surgery alone and then observed, the regional recurrence rate was 24 percent for tumors ≥4 mm [85]. Of the patients with regional recurrence, 39 percent occurred in the contralateral neck, suggesting the potential need for bilateral neck irradiation, particularly for primary tumors that approach or involve the midline.

IS THERE A ROLE FOR SYSTEMIC THERAPY? — There is limited evidence to support the use of chemotherapy alone in patients with early-stage oral cavity cancer. The use of immunotherapy in both the preoperative (ie, neoadjuvant) and postoperative (ie, adjuvant) setting also remains investigational.

POSTTREATMENT SURVEILLANCE — Regular posttreatment follow-up is an essential part of the care of patients after potentially curative treatment of head and neck cancer. Patients should be educated about possible signs and symptoms of tumor recurrence, including hoarseness, pain, dysphagia, bleeding, and enlarged lymph nodes. Further details on posttreatment follow-up are discussed separately. (See "Posttreatment surveillance of squamous cell carcinoma of the head and neck".)

In general, the intensity of follow-up is greatest in the first two to four years, since approximately 80 to 90 percent of all recurrences after curative intent treatment will occur within this timeframe. Continued follow-up beyond five years is generally suggested because of the risk of late recurrence and for second primary malignancies. Patients with continued tobacco use may be offered more frequent surveillance visits for a longer duration (ie, beyond five years) due to the higher risk of recurrence and for second primary malignancy in such patients.

PROGNOSIS — For patients with stage I or II oral cavity cancer, the five-year overall survival rate was 70 percent [86]. Neck dissection is also associated with an improved prognosis, reflecting more accurate staging, and reclassification of patients with lymph node involvement or invasion of adjacent structures with American Joint Committee on Cancer (AJCC) stage III or IV disease (table 1).

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Head and neck cancer".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topic (see "Patient education: Tongue cancer (The Basics)" and "Patient education: Mouth sores (The Basics)")

SUMMARY AND RECOMMENDATIONS

●Sites of involvement for oral cavity tumors – Oral cavity squamous cell carcinomas include tumors that originate in the lip, lower alveolar ridge, upper alveolar ridge, retromolar trigone, hard palate, oral tongue, floor of the mouth, and buccal mucosa. (See 'Anatomy' above.)

●Risk factors – Risk factors for oral cavity tumors include tobacco, alcohol, and betel nut quid use, among others. Human papillomavirus (HPV) infection is uncommon. (See 'Epidemiology and risk factors' above.)

●Pretreatment evaluation – Upon the diagnosis of an oral cavity tumor, pretreatment evaluation should assess the extent of locoregional disease (tumor size, depth of invasion [DOI], involvement of regional lymph nodes) and distant metastases. (See 'Pretreatment evaluation' above.)

●Management of primary tumor – For most patients with early-stage oral cavity tumors, we suggest treating the primary tumor with initial surgery (with or without adjuvant radiation therapy [RT]) rather than definitive RT (Grade 2C). For patients who are ineligible for or decline surgery, we alternatively offer definitive RT. Treatment of the primary tumor is also influenced by site. (See 'Management of the primary tumor' above and 'Site-specific management' above.)

•Close (<2 mm) or positive surgical margins – For patients who undergo resection of the primary tumor with close (<2 mm) or positive surgical margins, we suggest reresection rather than postoperative RT (Grade 2C). For those who are ineligible for or decline reresection, we offer adjuvant RT to the primary site. (See 'Surgery' above and 'Adjuvant radiation therapy' above.)

•Other high-risk features on postoperative pathology – For patients who undergo resection of the primary tumor and present with other high-risk features on postoperative pathology, we suggest postoperative RT to the primary site and unilateral or bilateral neck rather than observation (Grade 2C). Such high-risk features include positive or close (<2 mm) surgical margins, bone invasion, and pathologically positive lymph nodes. RT may also be indicated for primary tumors with DOI >4 mm, lymphovascular, or perineural invasion. (See 'Adjuvant radiation therapy' above.)

Adjuvant concurrent chemoradiation is reserved for those with positive surgical margins or extranodal extension. (See "Adjuvant radiation therapy or chemoradiation in the management of head and neck cancer".)

●Management of the neck for primary sites other than the oral tongue – For patients with resectable primary sites other than the oral tongue, the choice of elective therapy to the neck is guided by the location of the primary tumor, tumor size and DOI (table 1), as well as available clinical expertise in the procedure of choice. (See 'Management of the neck' above.)

Treatment modalities include elective neck dissection and sentinel lymph node biopsy (SLNB). For patients whose primary tumor is being treated with RT, treatment options also include elective RT to the neck. (See 'Treatment modalities' above.)

•For patients with T1 lower lip cancers, we suggest observation rather than elective neck dissection (Grade 2C). For tumors with DOI <3 mm, SLNB may be offered as an alternative to observation, since lymphatics from the lip drain to both sides of the neck. (See 'Lip' above.)

•For patients with T2 or larger lower lip cancers, upper lip, lower alveolar ridge/retromolar trigone, upper alveolar ridge/hard palate, or buccal mucosa tumors, we suggest elective neck dissection rather than observation or SLNB (Grade 2C). (See 'Lip' above and 'Lower alveolar ridge and retromolar trigone' above and 'Upper alveolar ridge and hard palate' above and 'Buccal mucosa' above.)

•For patients with floor of mouth tumors, the management to the neck is similar to that of oral tongue tumors (algorithm 1). However, SLNB is less reliable for floor of mouth tumors, due to the close proximity of the radiotracer injection site to the lymph nodes at risk. (See 'Oral tongue and floor of mouth' above.)

●Management of the neck for primary oral tongue tumors – For most patients with resectable primary tumors of the oral tongue, we proceed with elective treatment to the neck (either elective neck dissection or SLNB) depending on the location and DOI of the tumor as well as available clinical expertise in the procedure of choice. Our approach is as follows (algorithm 1) (see 'Oral tongue and floor of mouth' above):

•DOI >3 mm, tumor lateralized – For patients with lateralized tumors with DOI >3 mm, either unilateral elective neck dissection or SLNB (with further surgery in the case of sentinel node involvement) are appropriate options. The choice of therapy typically is based upon the skill and expertise of the treating surgeon. (See 'Depth of invasion >3 mm, tumor lateralized' above.)

•DOI >3 mm, tumor midline or approaching midline – For patients with tumors with DOI >3 mm that approach or involve the midline (table 1), either bilateral elective neck dissection or SLNB (with further surgery in the case of sentinel node involvement) are appropriate options. The choice of therapy typically is based upon the skill and expertise of the treating surgeon. (See 'Depth of invasion >3 mm, tumor midline or approaching midline' above.)

•DOI ≤3 mm, tumor lateralized – For patients with the most superficial oral tongue cancers (≤1 mm DOI) that are lateralized, we suggest observation rather than surgery (Grade 2C).

For patients with tumor lateralized and DOI >1 mm and ≤3 mm, we suggest SLNB rather than proceeding directly to an elective unilateral neck dissection, provided patients are being treated at a center with expertise in this technique. If such expertise is not available, we suggest a unilateral elective neck dissection rather than other surgeries (Grade 2C). (See 'Depth of invasion ≤3 mm, tumor lateralized' above.)

•DOI ≤3 mm, tumor midline or approaching midline – For patients with tumors with DOI ≤3 mm that approach or involve the midline, we suggest SLNB rather than proceeding directly to a bilateral elective neck dissection, provided that patients are being treated at a center with expertise in this technique. If such expertise is not available, we suggest a bilateral elective neck dissection rather than other surgeries (Grade 2C). (See 'Depth of invasion ≤3 mm, tumor midline or approaching midline' above.)

●Management of the sentinel node – In cases where an SLNB is performed and is negative for disease, we suggest observation rather than unilateral neck dissection (Grade 2C).

If the SLNB is positive for disease on the ipsilateral side only, we suggest a unilateral neck dissection (Grade 2C), whereas if it is positive on the contralateral side or if bilateral drainage is observed by lymphoscintigraphy, we suggest a bilateral neck dissection (Grade 2C), rather than observation or other surgeries. (See 'Sentinel lymph node biopsy' above.)