Application of anatomy unit resection surgery for lateral basicranial surgical approach in oral squamous carcinoma

Lip and oral cavity cancers are relatively common cancers worldwide, with approximately 350,000 new cases and 170,000 deaths being reported [1]. Histologically, squamous cell carcinoma accounts for more than 90% of malignant lesions in the oral cavity, with a 5-year overall survival rate of approximately 50% [2]. Comprehensive treatment (including surgery, chemotherapy, radiotherapy and other treatments) is universally accepted as the most appropriate management for oral squamous cell carcinoma (OSCC) patients, especially for advanced OSCC patients. Despite continuous therapeutic advancements in the past several years, the overall prognosis has remained poor [3, 4]. Undoubtedly, surgical treatment remains the mainstay of comprehensive treatment [5], and it exerts a considerable effect on the prognosis of OSCC patients [6, 7].

OSCC can be subdivided according to primary tumor sites, such as the tongue, buccal mucosa, maxillary/mandibular gingiva, floor of the mouth and palate. Due to the specific anatomical locations of OSCC, the overall survival rate of different OSCC subtypes can significantly vary [8, 9]. Previous studies have obtained similar conclusions that buccal squamous cell carcinoma (BSCC) has a worse prognosis than other OSCCs due to the high incidence of locoregional recurrence [8, 10]. In addition, differences in TNM stage may account for why BSCC appears to have a worse prognosis than other OSCCs [8]. BSCC has a high incidence of local recurrence, even with pathologically negative margins [9].

For posterior BSCC, we clinically observed that almost all recurrent tumors were located in the basicranial region and involved surrounding anatomical units (Fig. 1). The characteristic feature of the basicranial region is the absence of definite boundaries and the inclusion of various anatomical units. We developed a novel concept of the posterior oral anatomical complex (POAC) to identify these anatomical units, which consists of the posterior mandibular body, mandibular ramus, infratemporal surface of the maxillary body, pterygoid process, temporalis, medial pterygoid and lateral pterygoid (Fig. 2). Due to the indefinite anatomical boundaries and surrounding space of POAC, tumor cells can spread along muscle fibers and their surrounding spaces to the basicranial region [11]. OSCC with POAC involvement is known as posterior oral squamous cell carcinoma (POSCC). In our previous study, we found that anatomy unit resection surgery (AUSR) involves the removal of the entire anatomical unit (or subunit) with tumor involvement and significantly improves the overall survival rate in BSCC patients [11, 12]. Therefore, the principal aim of this study was to evaluate the effect of AUSR on patients with POSCC.

Among the patients, 114 patients were male, and 6 patients were female. The mean age of the patients was 49.8 years (range: 28–72 years). The clinical characteristics are presented in Table 1. The primary tumors of all of the patients in the AURS were classified as T4 stage. The T stage (p < 0.0001) and N stage (p = 0.0168) in the AURS group were more advanced than those in the conventional group. The size of the free flap (p < 0.0001) and the duration of operation (p < 0.0001) in the AURS group were significantly increased compared with those in the conventional group. In addition, there were no significant differences in other clinical characteristics between the control and experimental groups. The duration of follow-up ranged from 1 to 88 months, with a 100% follow-up rate. The overall survival rate (OS) was 62.5% and 37.5% in the AURS group and conventional group (hazard ratio: 0.59; p < 0.0001), respectively (Fig. 4A, Table 1). The disease-free survival rate (DFS) was 62.5% and 34.3% in the AURS group and conventional group (hazard ratio: 0.43; p = 0.0008), respectively (Fig. 4B, Table 1). Moreover, the local disease control rate in the AURS group was significantly higher than that in the conventional group (p < 0.0001) (Table 1). Upon further analysis, the OS and DFS of the patients (n = 41) with T4 stage in the conventional group was 26.8% (p = 0.0048) and 26.8% (p < 0.0001), respectively, which was dramatically decreased compared to that in the AURS group (Fig. 4C and D). Interestingly, the DFS of patients with metastatic lymph nodes or without metastatic lymph nodes was 50% (20 of 40 patients) or 93.7% (15 of 16 patients) in the AUSR group, respectively.

In the AURS group, the face-lateral approach I, face-lateral approach II, face-median approach or face-median and face-lateral combined approach were 18, 15, 21, or 2, respectively. A total of 20 patients (face-lateral approach I and face-lateral combined approach) had the temporofacial branch of the facial nerve disconnected. Fifteen patients presented with slight paresis, and five patients presented with moderate or severe paresis. Furthermore, the survival rate of the disconnected zygomatic arch was 94.6% (54 of 56 patients).

The oral cavity boundary comprises the lips anteriorly, the mouth floor inferiorly, the cheeks laterally, the oropharynx posteriorly and the palate superiorly [15, 16]. The bone tissue bases of the surrounding muscle tissue are attached to the mandible and maxillary areas. Due to the complexities of oral cavity anatomy, it is not easy to obtain adequate and safe resection margins for OSCC [17]. Previous studies have indicated that OSCC involving POAC can lead to a high locoregional recurrence rate and poor prognosis [18, 19]. To address this challenge, the present study aimed to develop a novel AURS for POSCC to improve treatment options for these patients.

The study by Margaret showed that the local recurrence incidence of T4 stage BSCC was 75% [20]. In this study, the OS and DFS of the patients with T4 stage in the conventional group was 26.8% and 26.8%, respectively, which is similar to the rates of a previous study. For POSCC patients treated with a traditional surgical approach, the OS and DFS were 39.6% and 37.5%, respectively [11]. Although all of the POSCC cases were classified as T4 stage and 71.4% of patients had metastatic lymph nodes in this study, the OS and DFS were 62.5% and 62.5% in the AURS group, respectively, which were dramatically higher than those in the conventional group. These results indicated that AURS is an effective approach to address patients with POSCC.

In POSCC patients, the tumor cells can spread along the muscle fiber and its surrounding space to the basicranial region, which may barely be determined by resection margins [20]. Similarly, the local disease control rate in the AURS group was significantly better than that in the conventional group in the present study. The findings indicated that the application of AUSR could provide a new and effective strategy for POSCC surgical treatment.

The presence of cervical lymph node metastasis is universally accepted as an important prognostic factor in patients with OSCC [21, 22]. In this study, the DFS of patients with metastatic lymph nodes (50%) decreased dramatically compared to those without metastatic lymph nodes (93.7%) in patients treated with AURS. The aggressiveness of tumor cells may explain the difference in DFS between the two groups. Metastatic lymph nodes serve as a key factor in determining the prognosis of POSCC patients [23]. The only patient without metastatic lymph nodes died of contralateral metastatic lymph nodes. None of these patients without metastatic lymph nodes had recurrence of the primary site tumor in the AURS group. Moreover, these findings indicated that the application of AUSR is an ideal method to completely resect the primary tumor of POSCC, especially for patients without metastatic lymph nodes.

In this study, the AUSR for the lateral basicranial surgical approach was classified as face-lateral approach I, face-lateral approach II, face-median approach or face-median and face-lateral combined approach, which is dependent on the extent of tumor resection. For tumors extending beyond the level of the zygomatic arch, an incision was made in the temporoparietal region to completely remove the temporalis anterior. For nonthrough resection POSCC, the face-median approach was performed [24]. Compared with the face-median approach, the face-lateral approach obtained enough exposure in the operation field; however, it is not ideal for the protection of the facial nerve and temporal fat pad. The anterolateral thigh free flap is preferred for filling the absence of muscle or soft tissue, which is particularly significant in facial depression resulting from the loss of tissue inside of the zygomatic arch. The donor site of anterolateral thigh free flaps includes abundant soft tissue and various types of flaps, such as chimeric, adipofascial, musculocutaneous and folded flaps [25]. Free flaps are preferred to reconstruct complex defects and fill dead spaces in head and neck defects [24].

Due to the difficulty and complexity of AUSR in POSCC patients, the flap size of the anterolateral thigh flap and duration of operation in the AUSR group were significantly increased compared with those in the conventional group. Similarly, the postoperative length of stay in the AUSR group was also longer than that in the conventional group. To completely remove the temporalis anterior, it is necessary to extend the incision upward to the temporoparietal region and to temporarily disconnect the temporofacial branch of the facial nerve. Facial paralysis is an inevitable complication for patients who undergo complete resection of the temporalis. The study by Eaton showed that 79% of patients with facial nerve anastomosis had some function, and the mean time to recovery was 7 months [26]. For lateral basicranial surgery, most studies have demonstrated at least 50% grade III postoperative recovery, and the duration of nerve palsy was a prognostic factor for postoperative recovery [27]. Our findings suggested that 75% of patients presented with slight paresis, and 25% of patients presented with moderate or severe paresis. This may be explained by the fact that the repair of the facial nerves was immediately performed after tumor extirpation, with a short interval time, thus leading to a good prognosis. Biological and mechanical factors play an important role in bone graft incorporation [28]. There is a consensus that essential techniques for successful bone grafts include adequate graft fixation without movement, the absence of infection and extensive soft tissue coverage [28, 29]. A previous study showed that the success rate of nonvascularized bone grafts for mandibular reconstruction was 87.6%. Our results indicated that the survival rate of disconnected zygomatic arches was 94.6%. The high success rate of free zygomatic arch grafts could be explained by the following factors: 1. the relatively small size of the graft; 2. the absence of dead spaces filled with anterolateral thigh free flaps; and 3. the presence of sufficient tension-free soft tissue. Furthermore, the clinical results of this study showed that patients in the AUSR group had better nerve function and a good prognosis after surgery.

This clinical study had several limitations. The study was a retrospective study with a limited sample size and biases in sample information. Moreover, it lacks the support of a large number of samples and a multicenter prospective study, which is another shortcoming. Further research and evaluation will be conducted in the future.

This lateral basicranial surgical approach based on AUSR improves the survival rate and enhances the local control rate while preserving a good prognosis without damaging the nerve and zygomatic bone. This surgical approach based on AUSR provides a novel and effective surgical treatment to address POSCC with better prognosis, especially for patients without metastatic lymph nodes.