A DMBA-induced ovarian cancer model: Could it be clinically useful to test OncoTherad and erythropoietin as anticancer approaches?

Abstract

e17589 Background: The term ovarian carcinoma (OC) refers to a heterogeneous collection of five distinct diseases known as histotypes. While histotype-specific treatment is still a clinical challenge, well-characterized models are required for testing new therapies. OncoTherad is a nanoimmunotherapy developed by University of Campinas/Brazil, which exhibits antitumor properties. Erythropoietin (EPO) has cytoprotective effects including in the ovaries. We assessed whether a chemically-induced animal OC model was representative of human disease by evaluating which histotype it best represents. We evaluated both mutational and immunohistochemical (IHC) biomarkers routinely used for human OC and used the observed features to provide context in the evaluation of OncoTherad and EPO effects. Methods: Thirty-five Fischer rats were distributed into 5 groups: Control (Sham surgery); Cancer (7,12-dimethylbenzoanthracene – DMBA injection in the ovarian bursa, 1.25 mg/kg); OncoTherad (20mg/kg IP); EPO (8.4 µg/kg IP); and OncoTherad+EPO (same doses). Ovary specimens were formalin-fixed into paraffin-embedded donor blocks. After DNA extraction and tissue microarray construction, we assessed typical gene mutations directly by Sanger sequencing (Pik3ca, Ctnnb1, and Kras) or indirectly using IHC surrogates (Arid1a and p53). Finally, we examined biomarkers typical of different OC histotypes (Wt1, Pr, Hnf1β) as well as lymphocyte density (CD3) by IHC. Results: The results were consistent across the cancer-induced animals. The majority of abnormal epithelial cells were Wt1+, Pr-, and Hnf1b-. Therefore, our rat model of DMBA-induced ovarian cancer was most likely serous type ovarian carcinoma, in agreement with the histopathological analysis. The ovarian lesions were molecularly similar to low-grade serous ovarian carcinoma as abnormal pattern of p53 staining was rarely observed. Loss of immunoreactivity for Arid1a protein was seen in some abnormal epithelium. However, the interpretability of mutation surrogates in rat tissues may not always be consistent with IHC analysis systems applied for human tissues. Furthermore, DNA sequencing did not reveal driver mutations in any of the sampled specimens. The treatments, especially OncoTherad+EPO, increased the number of CD3 positive immune cells. After EPO treatment, the tumor and abnormal areas showed a higher number of CD3+ lymphocytes than the normal regions; following a previous work, this could be due to substantial presence of Foxp3 positive cells. Conclusions: The features analyzed favored a low-grade serous carcinoma model in which treatments with OncoTherad and EPO showed immunomodulatory properties related to the reduction of ovarian lesions seen in previous work. The overall data highlight the importance of further characterizations of animal models to provide a complete and specific panel for testing new drugs.