Ⅰ. INTRODUCTION
Oral squamous cell carcinoma (OSCC) is well-known as the most common malignancy in oral cavity. The five-year survival rate of OSCC patients has been reported to be 65% on average, but those of patients with high clinical stage remain still 43-57%.1,2) Although much understanding has been gained about the molecular mechanisms of OSCC over the past two decades, the majority of OSCC patients still die from local recurrence and metastasis of the tumor. Therefore, many studies have been conducted to find prognostic factors predicting tumor aggressiveness and poor prognosis of patients.
Cervical lymph node (LN) metastasis is a well-known prognostic factor in OSCC.4) Moreover, interest in perineural invasion (PNI) has recently increased as a new form of metastasis that has been underestimated. The neurotropism of tumor cells has been reported in many malignancies, including those of prostate, pancreas, and head and neck.4-7) PNI incidence has varied from 6.1% to 81% in head and neck squamous cell carcinoma.5,6) These variations in the assessment may be caused by the detection of PNI on a limited number of HE stained slides or by the subjective interpretation of PNI. Therefore, it is needed to develop the auxiliary diagnostic marker for PNI in the tumor tissues. PNI has been significantly associated with tumor aggressiveness, nodal involvement at the time of diagnosis, locoregional recurrence, and/or poor survival rate of patients in many kinds of cancer including OSCC.6,7) According to those previous studies, adjuvant radiotherapy and/or elective neck dissection may be needed for improving treatment outcome in patient with PNI.8) Recently, studies in the pancreatic cancer which shows PNI in more than 80% of patients demonstrated that paracrine interactions between Schwann cells and cancer cells promote PNI via neural cell adhesion molecule, L1CAM.9) It was also reported that presence of PNI correlated with L1CAM expression in pancreas cancer tissue.10)
L1 cell adhesion molecule (L1CAM) is a transmembrane glycoprotein of the immunoglobulin superfamily that plays a role in neural development increasing neural cell migration.11) Besides its expression in neural cells, L1CAM is also detected in other normal tissue such as vascular endothelial cells, renal collecting duct cells, and myelomonocytic cells.12) However, a new role of L1CAM outside the nervous system has been revealed to date. Overexpression of L1CAM was related to the tumor progression in many human cancers such as endometrial, gastric, colorectal, pulmonary, and salivary gland cancers.13) In particular, recent studies using large cohort of endometrial carcinoma emphasized the value of L1CAM as a specific prognosticator and/or a preoperative surrogate of lymph vascular space invasion.14,15) However, there was no report evaluating the clinical significance of L1CAM expression in OSCC. Therefore, the aims of our study were to examine PNI and L1CAM expression in OSCC tissue samples, to define the value of L1CAM as a candidate molecule supporting the diagnosis of PNI, and to evaluate the prognostic significance of PNI and L1CAM in OSCC.
Ⅱ. MATERIALS AND METHODS
1. Patient and tissue samples
61 OSCC samples from various location of oral cavity including tongue, gingiva, buccal mucosa, and floor of mouth were investigated by immunohistochemistry. All tumors were surgically removed at the Department of Oral and Maxillofacial Surgery, SNU Dental Hospital, between 2006 and 2007. All procedures followed in this study were in accordance with the guidelines of the Institutional Review Board of the Seoul National University Dental Hospital.(#CRI 20003) Table 1 shows the clinicopathological characteristics of the patients including age, gender, differentiation, tumor size (pT), LN metastasis (pN), TNM stage, and PNI. All procedures followed in this study were in accordance with the guidelines of the Institutional Review Board of the Seoul National University Dental Hospital (#CRI 20003). Detection of PNI in OSCC tissue samples was performed based on the current widely accepted definition: 1) tumor in close proximity to nerve and involving at least 33% of its circumference or 2) tumor cells within any of the 3 layers of the nerve sheath.16)
2. Immunohistochemistry
Immunohistochemical staining was performed according to the routine procedures. Tissue sections were incubated in a 1:100 dilution of mouse monoclonal anti-human L1CAM antibody (clone 14.10; BioLegend) at room temperature. Nerve tissues in each slide were used as an internal positive control. L1CAM expression was considered positive if 10% or more of the tumor cells showed moderated to strong membranous staining as mentioned previously.17)
3. Statistical analysis
All statistical analyses were carried out using IBM SPSS Statistics 25 (IBM, Armonk, NY, USA). Correlations between L1CAM expression and clinicopathological parameters were assessed with Pearson’s chi-square test. For all analyses, p values less than 0.05 were considered statistically significant.
Ⅲ. RESULTS
1. Perineural invasion (PNI) in OSCC and its clinicopathological significance
PNI was detected in 31.1% of OSCC cases (Fig. 1). PNI significantly correlated with presence of LN metastasis at the time of diagnosis (p＜0.001) and advanced clinical stage (p=0.039), but there were no significant correlations between PNI and age, gender, differentiation, pT stage, or recurrence (Table 2).
2. L1CAM expression in OSCC tissues
L1CAM expression was not detected in the normal oral mucosa, but strongly expressed in the peripheral nerve tissue within the tissues (Fig. 2A). Vascular endothelial cells also expressed L1CAM.
In OSCC tissue, the extent of L1CAM expression in tumor cells varied within the same tissues. L1CAM expression was considered positive when 10% or more of the tumor cells stained moderate or strong for L1CAM. OSCC cells showed the membranous pattern of expression of L1CAM (Fig. 2B and 2C). 44.3% of cases were positive for L1CAM. Most of tumor cells showing moderate to strong expression were found at the invasive border of tumors or in the less- differentiated and sometimes spindle-shaped tumor cells (Fig .2D and 2E). In the cases showing PNI, L1CAM expression was observed both in the invading tumor cells and in invaded nerves (Fig. 1B). On the other hand, L1CAM was rarely expressed in more differentiated, keratinizing tumors (Fig. 2F).
3. Correlation between L1CAM expression and clinicopathological parameters in OSCC
Correlations between L1CAM expression and clinicopathological parameters are shown in Table 3. There were significant correlations between L1CAM expression and differentiation, pN, PNI, and TNM stage in OSCC. L1CAM expression correlated with moderate differentiation (p=0.046), positive LN metastasis (p=0.022), PNI (p=0.046), and advanced stage (p=0.023). However, there were no significant correlations between L1CAM expression and age, gender, pT, or recurrence. These results indicate that L1CAM could play a role in PNI and tumor progression in OSCC.
Ⅳ. DISCUSSION
Perineural invasion (PNI) is a unique and underestimated metastatic route and has become a key pathologic feature in various human cancers such as pancreatic, prostate, gastric, and head and neck cancer.16) Head and neck squamous cell carcinoma is a representative cancer with high incidence of PNI. PNI incidence in OSCC has varied up to 81%,6) which may depend on the location of the tumor (ex. relatively higher prevalence in tongue). In our study, PNI was detected in 31.1% of 61 cases. Our cases included several different locations of the oral cavity together with tongue, so the incidence of PNI could be lower than that from the study of tongue. Several studies support that PNI significantly correlate with the nodal status at the time of diagnosis and the postoperative regional LN recurrence.6) According to these results, more aggressive treatments such as adjuvant radiation therapy or elective neck dissection have been recommended for either the early stage or cNO tumor showing PNI.6-9) In order to determine whether or not to provide adjuvant therapy, more objective and reproducible molecular markers predicting PNI beyond the histological assessment on the HE slides should be developed in the nearest future. Our results showed the significant association of L1CAM expression with PNI, suggesting the possibility of L1CAM as a predictive marker for the presence of PNI. However, more larger cohort study composed of many cases of tongue lesion is necessary to confirm the reliability of its expression as tongue squamous cell carcinomas frequently accompany PNI.
PNI is believed to occur through the active, specific and very complex interaction between cancer cells and nerve cells.18) These actions are known to be regulated by various neurotrophins, nerve growth factor receptors, chemokines and their receptors, and neural cell adhesion molecules such as NCAM and L1CAM.5,6) L1 cell adhesion molecule (L1CAM) is one of key members of the immunoglobulin-like CAM (Ig- CAM) family in addition to NCAM. It was first recognized to play critical roles in surface interactions of neurons during neural development. The important function of L1CAM is to promote cell motility that drives cell migration during neural development and promotes tumor progression and metastasis of cancers.9,13,19) As expected, overexpression of L1CAM has been shown to correlate with the aggressive pathological characters such as LN metastasis, and poor survival in many human cancers such as endometrial, vulvar, and pancreatic cancers.13,20,21) Our study also demonstrated that there were significant positive correlations between L1CAM expression and differentiation, pN, PNI, and TNM stage in OSCC. Regarding the correlation between PNI and L1CAM, however, there has been few studies except for one in pancreatic duct adenocarcinoma.10) Little is known about the molecular mechanism by which L1CAM plays a role in PNI. However, recent report by Na’ara et al. demonstrated that paracrine interactions between Schwann cells and cancer cells could promote PNI via L1CAM secretion.9) As this mechanism may be also applied to OSCC, further studies are needed.
It has been reported that L1CAM might be a promising new target for treatment in a variety of human malignancies as it is well-known to be involved in tumor progression and metastasis. In addition, L1CAM is overexpressed in several cancers such as ovarian or pancreatic carcinoma that have poor prognosis with conventional chemotherapy.13) Therefore, a new antibody therapy targeting L1CAM can overcome some limitations of traditional therapy in those tumors.22) To date, various studies using mice xenograft models have demonstrated the anticancer effects of L1CAM mAbs. Moreover, suppression of prostate cancer growth in mice bone using anti-L1CAM siRNA was also reported.23) Therefore, additional studies are needed to evaluate the possibility of L1CAM as a candidate target molecule for the treatment of OSCC.
In conclusion, we investigated L1CAM expression in OSCC and evaluated the diagnostic and prognostic value of L1CAM for the first time. We found that PNI significantly correlated with positive LN metastasis and poorer clinical outcome in OSCC. Moreover, L1CAM expression was shown to be a prognostic factor as well as a predictive marker for PNI in OSCC. These findings suggest that L1CAM likely play a role in tumor progression and could be a candidate marker for predicting PNI and clinical outcome in OSCC. However, more larger cohort study is needed to confirm the reliability and reproducibility as a predictive marker for PNI.
