Author:

Abstract:

Endometrial cancer is the most common gynecological malignancy in the western world, with yearly increasing incidence and mortality rates, stressing the need to (re-)evaluate and improve current therapies for this disease. Most endometrial cancers are carcinomas (ECs, 97%), originating from the epithelial lining of the uterus and are traditionally classified into type I and type II, based on the histopathology of the tumor. The generally more aggressive type II ECs account for the smallest group and mainly include tumors of serous (10-20%), and less commonly, clear cell (5%) and carcinosarcoma (<5%) histology. These tumors are classified as high-grade by the ESGO guidelines, and thus surgery as well as adjuvant therapy (chemo-and radiation therapy) are used as the standard treatments. Here, we focused on type II uterine serous carcinoma (USC), which has a high recurrence rate and a dismal five-year survival rate of less than 30%, due to its aggressive nature and general resistance towards the standard therapies. Over the last decade, the TCGA objectively classified endometrial tumors based on their molecular profiles, dividing EC into four major genomic groups. Additionally, these studies identified several molecular markers, which could lead to new targeted therapies and/or predictive biomarkers to improve current treatment options. Remarkably, somatic mutations in PPP2R1A, encoding the Aα subunit of the tumor suppressive Ser/Thr phosphatase PP2A, occur at high frequencies in type II ECs (up to 40%). In this thesis, we focused on the p.R183W Aα hotspot variant in two molecularly different USC cell lines. While ectopic expression of p.R183W Aα did not contribute to the oncogenic characteristics (growth, migration) of these USC cells, it did alter response to several commonly used and pre-clinical anti-cancer treatments. More precisely, we demonstrated that p.R183W Aα expression generally predicted increased resistance to kinase inhibitors targeting the TOR pathway. This was especially true for the USC cell line that has driver mutations in the PI3K pathway. Moreover, p.R183W Aα expression in this cell line rendered the cells more sensitive to the standard chemotherapies, cisplatin and paclitaxel. Furthermore, we showed that the small molecule activator of PP2A, ATUX-792, could be a potential new treatment option for patients with EC, especially, as part of a combination treatment with the tested kinase inhibitors. Importantly, the combination of PP2A activation with kinase inhibition was beneficial, regardless of the presence of PPP2R1A p.R183W mutation. Lastly, we investigated the effect of the p.R183W PPP2R1A hotspot variant on treatment response to the nucleoside analogue clofarabine. p.R183W Aα expressing USC cells showed increased resistance to clofarabine treatment in vitro and in vivo, corroborated by decreased clofarabine-induced apoptosis, G1 phase arrest, DNA-damage (γH2AX) and activation of ATM and Chk1/2 kinases. Phenotypic rescue by pharmacologic PP2A inhibition or dsiRNA-mediated B56δ subunit knockdown supported a gain-of-function mechanism of p.R183W Aα-B56δ complexes, promoting dephosphorylation and thereby inactivation of deoxycytidine kinase (dCK), the cellular enzyme responsible for the conversion of clofarabine into its bioactive form. Therapeutic assessment of related nucleoside analogues (gemcitabine, cladribine) revealed similar effects, but in a cell line-dependent manner. Expression of two other PPP2R1A USC mutants (p.P179R or p.S256F) did not affect clofarabine response in our cell models, arguing for mutant-specific effects on treatment outcome as well. In conclusion, our results indicate that p.R183W Aα holds potential as a predictive biomarker for USC, but especially in combination with other molecular alterations. Also, we indicate the potential of nucleoside analogues as a new treatment option for patients with USC, taking the PPP2R1A mutational status of the tumor as well as the tumor background into account. Additionally, we highlight the value of combination treatment with PP2A activators as well as inhibitors, depending on the type of anti-cancer compound. Importantly, we showed that such combination treatment can be valuable, regardless of PPP2R1A mutation, which could circumvent time and money consuming molecular profiling of the tumor for patient stratification.