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

Squamous cell cancers arising in the retromolar trigone and lower alveolar ridge comprise approximately 10 percent of all oral cancers and exhibit the same 3 or 4:1 male predominance of other head and neck cancers. The presenting symptom is typically pain, which is exacerbated by chewing.

Treatment options include radiation therapy (RT) and surgery. The local recurrence rate is higher with these tumors than for other sites in the oral cavity due to microscopic extension to the mandible and maxilla (for retromolar trigone tumors). In addition, the probability of occult regional lymph node metastases is higher than with most other oral cavity tumors, with the exception of tongue cancer and floor of mouth cancer. Thus, elective neck dissection is usually recommended for patients with a clinically negative neck.

Surgery — Surgical therapy involves wide local excision. Marginal or horizontal "rim" mandibulectomy may be required in order to achieve tumor free margins. Due to the normally thin overlying mucosa and the close proximity to the mandible, alveolar ridge and retromolar sites have a propensity for early invasion of this bone, as well as the maxilla for retromolar trigone lesions. Consequently, lesions that are clinically staged T1/T2 and treated with rim mandibulectomy may become pathologic stage T4 after histologic confirmation of bony invasion, the hallmark of T4 lesions. Segmental or composite resection is reserved for those tumors that are deeply invasive or that wrap around the mandible. In addition, segmental mandibulectomy may be necessary for early stage lesions in the thin, edentulous mandible in order to achieve negative margins.

It is extremely important to determine the true invasive margin, which may extend grossly or microscopically beyond the tumor front. Determining this invasive margin is challenging. For oral cavity lesions in general, computed tomography (CT) scans may be helpful for identifying bone invasion. In one report of 164 patients with oral cavity or oropharyngeal carcinoma treated with composite (segmental) resections, CT scan findings suspicious for bone invasion and primary tumor location (alveolus, retromolar trigone, tonsil and sulcus) were the only independent variables that predicted for the presence of bony invasion. However, in one report, preoperative CT scan failed to identify bone invasion in one-half of retromolar trigone lesions that histologically invaded bone. Potential reasons for this low sensitivity include the thickness of CT sections, the lack of bone windows and coronal imaging, and the presence of distortion from dental artifact.

A resection margin of at least 1 cm in all directions is recommended. At least for tumors involving the retromolar trigone, the optimal extent of surgery is controversial. In addition to stage, outcomes are dependent on the presence of bone invasion, deep infiltration of the masticator space, nodal involvement and treatment modality.

In an early series, surgery provided excellent local control (98 percent) among 61 patients with T1 to T4 disease. Among the patients with stage I and II disease, survival was 77 percent at five years. In a later series of 99 patients treated with definitive RT or surgery followed by RT, local control rates were better in surgically treated patients (approximately 71 versus 48 percent). Among all patients treated for stage I to III disease (RT or surgery plus RT), five-year rates of cause-specific and overall survival were 70 and 58 percent, compared to 57 and 42 percent for those treated for stage IV disease. Notably, in multivariate analysis, both cause-specific and overall survival were significantly better in the group undergoing RT in addition to surgery; only six patients in the definitive RT group received chemotherapy.

Management of the neck — For early lesions of the lower alveolar ridge and retromolar trigone, we generally recommend selective node dissection in levels I to IIl

The use of the Martinez-Gimeno Scoring System [MGSS] to predict the likelihood of neck metastases for oral cavity cancers is discussed above). Based upon points given for T-stage, tumor thickness, microvascular and perineural invasion, the histologic grade of differentiation, and presence of an inflammatory infiltrate, patients are classified into one of four groups with different rates of cervical nodal metastases on the basis of their MGSS score (total number of points). One of the main benefits of this scoring system is that it takes tumor thickness into account.

Postoperative RT — Combined modality therapy provides better locoregional disease control than single modality therapy. Postoperative radiation (in some cases with concomitant chemotherapy) is indicated for patients with positive resection margins (if not reresected), bone erosion, or pathologically positive lymph nodes after elective neck dissection. Others recommend that postoperative RT also be considered if there is vascular or perineural invasion in the primary tumor. The clinical situations where chemoradiotherapy is preferred over adjuvant radiotherapy alone are discussed in detail elsewhere.

Radiation therapy — Small, superficial lesions can be treated with combined external beam RT (EBRT) and an intraoral cone. Elective treatment of the upper neck nodes (levels I and II,) is indicated in all patients due to the high incidence of cervical nodal metastases. For well-lateralized lesions, elective nodal RT can be confined to the ipsilateral neck using either mixed photon-electron beams or a wedged-pair photon technique with an intraoral stent to reduce dose to the contralateral side.

For T1 lesions, we recommend 50 Gy/25 fractions by EBRT plus 16 Gy/8 fractions boost by EBRT or 15 Gy/6 fractions boost by intraoral cone. For T2 lesions, we use 50 Gy/25 fractions by EBRT plus 20 Gy/10 fractions boost.

Tumor stage is not a good indicator of outcome at this site and the results from RT for T1 and T2 lesions are similar. Although local faiure rates are high, most of the local failures can be successfully salvaged surgically. In one series of 159 patients with SCC of the anterior faucial pillar or retromolar trigone, for example, the local failure rate was 29 percent for T1 lesions, 30 percent for T2 lesions, 24 percent for T3 lesions, and 40 percent for T4 lesions. Following salvage surgery, which consisted of intraoral operation in one-third of the patients and a composite resection in the others, the ultimate failure rate was 0 percent for T1 lesions and 6 percent for T2 lesions.

Adverse effects of RT at this site include soft tissue necrosis, bone exposure, and bone necrosis. In the above report, 30 percent of patients developed some degree of bone exposure, but only 6 percent required a segmental mandibular resection for toxicity management.