Oral squamous cell carcinoma (OSCC) is a serious and increasing problem in Western and Asian countries. The 5-year survival rate is 59.4%, and this entity is responsible for 8000 deaths per year in the USA and has become the fifth most common malignant tumour affecting the male population. Apart from lymph node involvement, the prognosis is dependent on tumour size, and overall survival is impacted when jaw bone infiltration occurs. The prevalence of mandibular bone involvement by OSCC is reported to range from 11% to 56% of cases. Intraoperative evaluation of the soft tissue margins of OSCC is performed routinely by quick section analysis of samples obtained by frozen section technique. However, only a few studies have focused on the analysis of frozen bone sections during segmental mandibulectomy and the results have varied widely. Thus no reliable technique for routine clinical use has been identified to date.

Simultaneous reconstruction after segmental mandibulectomy is desirable as it improves quality of life. However, reliable investigation of the osseous margins requires tissue decalcification, which can take up to 14 days. This time frame is critical if immediate free flap reconstruction has been performed and further bone resection is necessary due to unexpected tumour invasion.

There appear to be no existing data on the average amount of resected tumour-free bone, because routine clinical practice involves the pathologist evaluating cross-sections of the distal and proximal ends and the adjacent 5 mm of bone to classify the margins as clear. Different methods for sampling and processing bone specimens for quick section analysis are reported in the literature. Only very few of them are used regularly in the clinical routine, mainly due to a lack of technical expertise or equipment.

Perineural invasion (PNI) or perineural spread (PNS) in OSCC, as well as in early lesions, has been emphasized as a very important prognostic factor, especially in relation to the inferior alveolar nerve (IAN). Therefore its role in tumour spread has been the subject of intense debate. While several authors dispute its role as a preferred route of tumour extension and have indicated that this phenomenon is an uncommon finding, others have shown cases of preferential spread along the IAN. Many authors have examined the prognostic value of clinically evident positive PNS with neurological disturbances in the case of extended OSCC. The perineural space lies between the enclosing protective nerve sheath and axons and is thought to enable some invasive malignant neoplasms to spread to distant sites through a pathway of reduced tissue resistance. As there are reported cases of free bony margins but infiltration of the IAN on the contralateral side of the mandible, it is possible from the theoretical point of view that the nerve might be a sensitive marker for tumour invasion of the mandible, predominantly in cases with tumour spread along the IAN.

Therefore, the aim of this prospective feasibility study was to critically evaluate whether a marginal IAN biopsy is representative of the complete mandible margin in certain cases. Furthermore it was sought to answer the question of how often the IAN is the dominant route of tumour entry and spread within the mandible and could therefore be used in these high-risk cases as a reliable and feasible additional marker to evaluate bony margins during the segmental resection of OSCC.

Materials and methods

All patients included in this study had OSCC in direct contact with the mandible as determined clinically and radiologically (multi-slice computer tomography). None of the subjects had any signs of numbness of the lower lip. Apart from the segmental mandibulectomy due to mandibular bone infiltration, all patients with a T3/T4 status who were pN0 underwent an ipsilateral selective neck dissection with clearance of the level I–III nodes, in accordance with the German guidelines for oral cancer. In the case of metastatic nodal disease (pN1 or pN2) that was confirmed intraoperatively by frozen section, a modified neck dissection with extended nodal clearance down to level IV and V and a contralateral neck dissection at levels I–III was done. Postoperative radiotherapy or radiochemotherapy was administered if indicated according to the German guidelines for oral cancer. To obtain a homogeneous sample, subjects who had undergone preoperative radiotherapy or chemotherapy, showed any gap between the mandible and tumour, and those with a poor prognosis for any reason were excluded from the study.

Preoperative diagnosis included the histological confirmation of the clinical diagnosis (if not already performed by the transferring dentist or surgeon), panoramic X-ray, computed tomography (CT), magnetic resonance imaging (MRI), and single photon emission computed tomography (SPECT)/CT performed according to a previously published protocol. Written informed consent for clinical and imaging diagnostic modalities, as well as for tissue biopsies for diagnostic and research purposes, was obtained from all patients at the time of initial presentation. The study was approved by the ethics committee of the hospital. Study subjects were enrolled in a clinical protocol reviewed and approved by the institutional cancer board according to the German and European guidelines for oral cancer treatment. During the segmental bone resection, a margin of at least 10 mm from the macroscopically suspected bone area was included ( Fig. 1 A) .

(A) Partial mandibulectomy performed according to the clinical extension of the tumour seen intraoperatively and the preoperative planning using different imaging modalities. (B) and (C) Extraction of the end of the IAN at the margin of the resected mandible. (D) Extracted part of the IAN. The extracted part of the IAN was subjected to intraoperative quick section analysis for the confirmation of clear margins and to determine the necessity for a further second-look resection of the soft tissue by the pathologist. The resected parts of the mandible were reconstructed with a free fibula flap ( n = 7) or a combination of a reconstruction plate and anterolateral thigh flap ( n = 7) or deep inferior epigastric artery perforator flap ( n = 1).

Immediately after bone resection, a representative sample was taken from the distal and proximal (if present) part of the IAN, which was sent for quick section analysis ( Fig. 1 B–D). The IAN tissue was prepared in a routine way by frozen section and cutting with a microtome. The complete specimens were analyzed further using paraffin embedding and staining ( Fig. 2 A, B ). Quick section analysis findings were compared with the definite histopathological results as the ‘gold standard’ regarding invasion of the nerve tissue. Based on the histological results of the resected specimens, a 2 × 2 contingency table was constructed to calculate sensitivity and specificity.

IAN tissue was prepared in a routine way by frozen section and cutting with a microtome. The complete specimens were further analyzed after paraffin embedding and staining. Definite histology results were compared to the initial quick section results. Left: Coronal section of the IAN (white arrow) surrounded by tumour tissue. Right: Axial section of the mandibular bone showing destruction by the invading tumour tissue (dotted line).