Head and neck squamous cell carcinoma (HNSCC) is an upper aerodigestive tract malignancy that arises from the squamous epithelium and by over 600,000 cases diagnosed annually, it comprises the 6th most common non-skin cancer worldwide(1) and the 9th most common cancer in men in the United States (2, 3). HNSCC causes high morbidity and mortality, particularly in advanced and recurrent cases. The primary known causes of HNSCC are alcohol and tobacco use, and infection with human papillomavirus (HPV)(4). For decades same treatment methods ( combination of surgery, radiation, and non-targeted cytotoxic chemotherapy) have been used for HNSCC patients independent of the underlying biology(5). Despite the effort to optimize treatment protocols (6, 7) HNSCC has a relatively poor prognosis, with a five-year survival between 25% to 60% based on cancer subsite(8). Moreover, these traditional therapies are often associated with high toxicity and morbidity(9).
There has been great interest in the optimization of therapies to “personalize” them and a paradigm shift toward a tailored and precision treatment for each patient is emerging(10). Treatment stratification designs have been developed for HNSCC to evaluate clinical response to cytotoxic chemotherapy induction (11, 12) and clinical and pathological staging are used to stratify the treatment intensity (13). There are great efforts and research to explore novel therapeutics for HNSCC. Despite decades of research, Cetuximab (Erbitux®), a chimeric monoclonal anti-EGFR antibody, is the only FDA-approved targeted therapy for HNSCC(5, 14) which has a response rate of only 7-13% (15).
To overcome these issues, a promising approach is to test and analyze the therapeutic response of the patient’s own cells in order to better predict the individual response to a specific therapy regimen. Viable cancer cells can be isolated from tumor tissue and exposed to therapeutic drugs under controlled conditions.(16-23) Small amount of patient’s tissue available for testing is one of the challenges as a few micrograms of tissue is achieved routinely from fine needle biopsies which 0.5 to 1 million of cells are typically extracted(24). Considering the heterogeneity and inevitable loss of cells during the enzymatic digestion procedure, the available number of cancer cells to perform reliable assays is very limited. There are different methods for proliferating primary cancer cells in vitro ranging from conventional, 2D cell monolayers to more advanced 3D culture systems(25). Another challenge in personalized medicine is the lack of appropriate in vitro models that can predict the chemotherapeutic response. In the past, the most common in vitro methods for tumor chemotherapeutic response testing were based on dissociating solid tumors and tumor cell were cultured two-dimensionally, but this method was proven inefficient as only limited and highly selected population of cells was able to explant and grow. Moreover, enough evidence shows that 2D monolayers lack many of characteristics found in tumor tissue that results in dramatically different therapeutic responses to clinical responses(26). So, they are starting to be replaced by 3D culture systems and spheroid models. Extensive evidence shows that 3D culture and spheroid (fragments of human tumor that form rounded multicellular structures in 3D culture) model of cancer cells better reflects in vivo tumor conditions than conventional 2D culture (27-29). 3D cultures more closely mimic in vivo behavior of cells in tumor tissue with respect to morphology and structure (27, 30) heterogeneity (28, 31) protein and gene expression patterns (28, 32, 33) distribution pattern of proliferating and apoptotic cells and growth (34) cell-cell and cell-matrix interaction(27) metabolic gradients, (35, 36) and mechanical and biochemical factors(37, 38)