From January 2004 to January 2010, 78 consecutive OMM patients, including 64 primary OMMs and 14 recurrent OMMs, were retrospectively reviewed with their clinical data (Additional file 2). These patients were treated at the Department of Oral and Maxillofacial Surgery, Ninth People's Hospital, School of Medicine, Shanghai Jiao Tong University. From the clinical data, routine clinical examination was performed in all patients. In addition, computerized tomography (CT) of the craniomaxillofacial region, neck, and lung was also performed. 99Tc emission computerized tomography of entire body was performed in patients in whom bony distant metastasis was suspected. PET-CT was initially recommended in such cases, but the majority of these patients refused this due to economic reasons. After signing the informed consent forms, all patients with primary or recurrent OMMs received pre-operative incisional or excisional biopsy of the intraoral lesion at room temperature or pre-operative incisional biopsy during cryotherapy or intra-operative frozen section for pathological diagnosis.
For intraoral lesions, the pre-operative incisional or excisional biopsy was performed under local anesthesia with 1% lidocaine (Lidocaine Hydrochloride Injection, Shanghai Fuda Pharmaceutical Ltd, Shanghai, China). Routine pathological examination was performed using hematoxylin and eosin (HE) staining; if the diagnosis was unclear, immunohistochemical examination was performed using the S-100, Vimentin, HMB45 and MelanA antibodies, which would show positivity for melanoma. The incisional biopsy during cryotherapy or intra-operative frozen section was performed under local anesthesia with 1% lidocaine; frozen pathological examination was performed using HE staining. If the diagnosis was unclear, further specimens were taken until a definite pathological diagnosis was obtained. As part of the neck examination, if CT scan showed CLNM, fine needle aspiration cytology (FNAC) examination was not suggested; if in clinically suspicious CLNM cases whereby CLNM could not be verified by CT scan, FNAC examination was performed on the suspicious lymph node.
For patients with clinical stage III disease, cryotherapy was initially suggested for the treatment of intraoral lesions. However, when the lesion was large or in cases whereby bony exposure occurred after cryotherapy, surgical resection was performed. Surgery was also indicated if patients had refused cryotherapy. For patients with clinical stage IVA disease, cryotherapy and surgical resection were both performed. In such cases, the mucosal or submucosal lesion was treated by cryotherapy while the invaded deep structures were removed by surgery. Similarly, surgery was performed if patients refused cryotherapy. For patients with clinical stage IVB and IVC disease, chemotherapy with/without radiotherapy was recommended. Post-operative chemotherapy was recommended in patients with positive surgical margins.
Cryotherapy was performed under local anesthesia with 1% lidocaine. A special cryospray unit (HX17-YDQ-500, Beijing Western Vision Technology Co., Ltd., Beijing, China) was used to spray liquid nitrogen on to the oral mucosa to destroy the involved lesion of oral mucosa as well as 0.5 cm marginal mucosa. If the intraoral lesion could not be confidently controlled, surgical operation was performed. In cases whereby safety margins could not be achieved due to functional and aesthetic limitations, intra-operative cryotherapy was performed at these resection margins.
Surgical resection was performed in the majority of OMMs patients. For intraoral lesions, an extended resection with a 2 cm safety margin was performed. For lesions of the hard palate and maxillary gingiva, extended resection with subtotal maxillectomy for T3 lesions and total maxillectomy with or without preservation of the orbital floor for T4a lesions was preformed. For lesions of the mandibular gingiva, extended resection with marginal mandibulectomy for T3 lesions and segmental mandibulectomy for T4a lesions was performed. Radical neck dissection was planned in patients with clinically positive CLNM while functional neck dissection was planned in patients with clinically negative CLNM.
The following chemotherapy protocol of DTIC (Dacarbazine Injection, Nanjing Pharmaceutical Factory Co. Ltd., Jiangsu, China) and CDDP (cisplatin Injection, Qilu Pharmaceutical Co. Ltd., Shandong, China) was used: 80 mg/m2 of CDDP on the first day, 250 mg/m2 of DTIC from the first day to the fifth day. After two weeks, patients received a second cycle of chemotherapy (21 days of each cycle). Two to four cycles were recommended; if response was stable disease, more cycles were used; if patients could not tolerate the complications of chemotherapy, chemotherapy was stopped.
For patients with recurrent OMMs, extended resection with post-operative chemotherapy was performed. The chemotherapy protocol was the same as previously described. Post-operative radiotherapy was recommended for patients whose disease could not be controlled by surgery and post-operative chemotherapy.
Follow-up reviews were carried out for all patients after the initial treatment: every 2-4 months during the first year, every 4-6 moths during the second year, and then twice a year. Besides physical examinations, CT scans of the craniomaxillofacial region, neck, and lung and 99Tc emission computerized tomography of entire body were performed every half a year. PET-CT was performed if there was suspicion of distant metastasis to lung, brain or bones. If there was a suspicion of a recurrent intraoral lesion, incisional or excisional biopsy was performed. All data was analyzed using SPSS 13.0 for Windows (SPSS Inc., USA), and the survival analysis was performed using the Kaplan-Meier method. Statistically significance was set at P < 0.05.