Preclinical study of the antitumor effect of melatonin in different head and neck cancer models: evaluation of the role of mitochondria and oxidative stress

Abstract

Head and neck squamous cell carcinoma (HNSCC) is an aggressive malignancy, representing the sixth most common cancer worldwide and accounting for nearly 500,000 deaths every year. Current mainstay HNSCC treatment is the combination of surgery, cisplatin (CDDP)-based chemotherapy and concurrent locoregional radiotherapy. However, these treatments are associated with important related-toxicities and the development of resistance, leading to treatment failure and to the death of patients. The poor outcomes for HNSCC demonstrate the need for novel therapies with less toxicity and more efficacy. Because of its oncostatic influence and lack of association with adverse effects, melatonin is of relevance to the development of innovative cancer treatments. Several studies have demonstrated that melatonin not only shows synergistic anticancer activity with other treatments, but also reduce the general toxicity arising from other treatments. This could be due to the “dual effect” of melatonin. On one hand, melatonin protects normal cells from a variety of insults due to its antioxidant effect in these cells, and on the other, melatonin suppresses tumor cell growth due to its pro-oxidant effect in these cells. Its oncostatic properties include pro-oxidant, anti-proliferative, pro-apoptotic, anti-angiogenic and antimetastatic actions in tumor cells. Nevertheless, although a great deal of clinical evidence has confirmed the anticancer effects of melatonin, some conflicting results also exist when it is applied in vivo. One possible explanation is that high concentrations of melatonin within the tumor are necessary to exert its oncostatic effect, being therefore necessary to search for an alternative administration route to improve bioavailability and establish the optimal dosage for cancer treatment. On the other hand, an important factor for failure in clinical trials is the inadequate biology of preclinical models. Traditionally, established human cancer cell lines have been the fundamental tool in cancer research. However, the genetic composition and behavior of established cancer cell lines has been altered after thousands of generations in culture, not faithfully representing real patient´s tumors. To solve this problem, the concept of patient-derived models emerged and gained extensive acceptance in cancer research. Since they proceed directly and recently from a patient´s tumor, these models might better retain the heterogeneity and molecular characteristics of patient tumors and be a better alternative to study cancer biology and drug sensitivity. Therefore, the utilization of patient-derived models significantly increased in the recent 5 years. However, literature about the oncostatic effect of melatonin in patient-derived tumor models is scant. Considering all above, our hypothesis is that melatonin, due to its oncostatic properties, is a promising agent in anticancer therapy, as new therapeutic approach and as a coadjutant therapy in HNSCC treatment. However, to establish a melatonin treatment with proper application in clinical practice, it is necessary to search for an alternative formulation or an alternative route of administration, which ensures a high bioavailability of melatonin in the tumor. These studies should be conducted not only in established HNSCC cell line models, but also in patient-derived models, which further resemble real tumors and ensure clinical translation. Concerning this hypothesis, the general objective of this study was to analyze the effect of different formulations of melatonin in vivo in different HNSCC mice models. We analyzed the effect of melatonin in mice bearing established HNSCC Cal-27 and SCC-9 cell line xenografts or patient-derived tumor xenografts. Mice were treated with different formulations of melatonin, administered intratumorally each 24 hours for 63 days. Melatonin was administered alone or in combination with CDDP, one of the most common HNSCC treatment. After mice sacrifice, tumors were histologically analyzed to assess the oncostatic effects of melatonin. Mitochondria morphology and distribution, apoptosis and ROS levels were also measured in tumors. Finally, melatonin effects in migration process were also studied, both in established and patient-derived cells. Results showed that intratumoral treatment with melatonin 3% reduced tumor growth in Cal-27 and SCC-9 xenografts, which correlated with a significant increase of melatonin levels in the tumors. The histology of tumors treated with intratumoral melatonin showed a decrease in tumor active areas, as well as an increase in collagen-rich capsule surrounding the tumors and a reduction of adenosquamous differentiation areas, which has been associated with a better cancer prognosis. Intratumoral injection of melatonin also potentiated the effects of CDDP reducing tumor growth. In line with these results, melatonin increased LPO levels and oxidized protein expression in Cal-27 tumors, alone or in combination with CDDP, and noticeably triggered ROS-dependent apoptosis, which resulted in an increase in the Bax/Bcl-2 ratio and TUNEL-positive nuclei. Moreover, melatonin 3% reduced CD98 levels compared to the control and CDDP groups, suggesting an increase in tumor cell differentiation. On the other hand, considering that one of the main targets of melatonin is the mitochondria, we evaluated mitochondrial morphology and distribution by electron microscopy. According to previous in vitro results, in melatonin-treated tumors, mitochondria were larger and localized in the periphery of the cells and not in the perinucleus, which has been associated with a better cancer outcome. Next, it was attempted to generate patient-derived cells (PDC) and patient-derived xenografts (PDX) from HNSCC tumor biopsies. Results demonstrated that melatonin reduced cell proliferation and spheroid formation in the PDC model. In addition, the intratumoral administration of melatonin 3% reduced tumor growth in the three PDXs analyzed. Finally, it was demonstrated that melatonin impaired migration and invasive capacities of Cal-27 and SCC-9 cell lines, as well as altered epithelial to mesenchymal transition markers in the PDC model. As a conclusion, our study elucidated the roles of intratumoral melatonin reducing tumor growth and synergizing CDDP, proposing melatonin as a possible therapeutic option for cancer treatment. Specifically, the combined treatment with melatonin and CDDP may be a promising clinical approach to treat patients with HNSCC, encouraging a future clinical trial in cancer patients to establish a new melatonin treatment with proper clinical application.