Four-protein signature accurately predicts lymph node metastasis and survival in oral squamous cell carcinoma

Summary

The presence of lymph node (LN) metastasis significantly affects the survival of patients with oral squamous cell carcinoma (OSCC). Successful detection and removal of positive LNs are crucial in the treatment of this disease. Current evaluation methods still have their limitations in detecting the presence of tumor cells in the LNs, where up to a third of clinically diagnosed metastasis-negative (N0) patients actually have metastasis-positive LNs in the neck. We developed a molecular signature in the primary tumor that could predict LN metastasis in OSCC. A total of 211 cores from 55 individuals were included in the study. Eleven proteins were evaluated using immunohistochemical analysis in a tissue microarray. Of the 11 biomarkers evaluated using receiver operating curve analysis, epidermal growth factor receptor (EGFR), v-erb-b2 erythroblastic leukemia viral oncogene homolog 2 (HER-2/neu), laminin, gamma 2 (LAMC2), and ras homolog family member C (RHOC) were found to be significantly associated with the presence of LN metastasis. Unsupervised hierarchical clustering–demonstrated expression patterns of these 4 proteins could be used to differentiate specimens that have positive LN metastasis from those that are negative for LN metastasis. Collectively, EGFR, HER-2/neu, LAMC2, and RHOC have a specificity of 87.5% and a sensitivity of 70%, with a prognostic accuracy of 83.4% for LN metastasis. We also demonstrated that the LN signature could independently predict disease-specific survival (P = .036). The 4-protein LN signature validated in an independent set of samples strongly suggests that it could reliably distinguish patients with LN metastasis from those who were metastasis-free and therefore could be a prognostic tool for the management of patients with OSCC.

Introduction

Metastasis involves the migration and colonization of cancer cells beyond the primary tumor and is governed by molecular mechanisms that facilitate the multiple steps required for this process. Despite a better understanding of some of the factors influencing metastasis through in vitro and in vivo studies, metastatic disease still remains the principal event leading to death in patients with cancer. In oral squamous cell carcinoma (OSCC), metastasis spreads predominantly via a lymphatic route, where the cervical lymph nodes (LN) are the first location [1], whereas metastasis to distant sites is relatively uncommon. Cervical LN metastasis is an important determinant of the outcome of OSCC [2] and the status is correlated with patient survival [3]. An accurate assessment of the cervical LN metastasis status in OSCC helps to predict the prognosis of patients and also helps clinicians perform the appropriate treatment. Currently, the clinical assessment of cervical LNs is conducted by clinical examination of the neck region or by ultrasound, computed tomography, and magnetic resonance imaging. Notably, postoperative histology examination has demonstrated that approximately 30% of clinically diagnosed metastasis-negative (N0) patients have metastasis-positive LNs in the neck. Furthermore, 10% to 20% of clinically diagnosed metastasis-positive (N+) patients were actually metastasis-free, where the highest overall accuracy for current preoperative neck assessment is 76%, strongly indicating that the sensitivity of these methods is still limited [4], [5], [6]. Clearly, the development of other parameters that could help with the assessment of LN metastasis is urgently needed.

Recent evidence has demonstrated that primary tumors are preconfigured to metastasize, and this propensity is detectable at the time of initial diagnosis [7], [8], [9]. Despite the large amount of information from these studies, these markers have not yet been validated as clinical tools to predict LN status in patients. In this study, we used high-throughput tissue microarrays (TMAs) to validate biomarkers that could accurately predict LN metastasis in a cohort of patients with OSCC. We identified a 4-protein signature that could reliably distinguish patients with LN metastasis from those who were metastasis-free. The sensitivity and specificity of this gene signature suggest that it could be a useful tool to complement current diagnostic procedures for managing patients with OSCC.