**5. Clinical Applications of Current Knowledge and Directions for Future Research** *5.1. Molecular Basis for a Specific Therapeutic Approach*

Collectively, studies performed over the last decade shed new light on the pathophysiology of endometriosis. Linkage and sequencing studies have identified genes and pathways important for endometriosis development and have highlighted potential causal links between endometriosis and endometriosis-associated cancer. The identification of women with endometriosis who are at risk of cancer development provides a basis for improved diagnosis and prognosis and is likely to aid in improved cancer surveillance of patients at risk.

As treatment of endometrial cancer was based on histological characteristics and staging [170,171], prognosis was not promising, especially if the stage is advanced [172–174]. Therefore, in the last couple of years, efforts have been directed to molecular aberrations within the specific tumor, as a novel biological targeted therapy with promising outcomes in clinical trials [175].

Various biomarkers [176], such as *mTOR* pathway disruptions, loss of estrogen and progesterone nuclear expression, TP53 mutation, changes in Wnt-signaling, or *L1CAM* expression, were identified as a link to endometrial cancer development [177]. All these mutations were associated with poor prognosis, but their clinical utilization is still questionable.

Preclinical investigations related to molecular-targeted therapies in ECs enabled deeper understanding of underlying mechanisms and highlighted different approaches to EC patients.

According to genomic characteristics of 373 endometrial carcinomas, The Cancer Genome Atlas (TCGA) classified EC into four molecular subtypes [178,179], which differ a lot from the molecular point, underlying risk factors, clinical and pathological features, treatment modalities, and prognosis [180–182]. These four distinct prognostic groups are *POLE* ultramutated, microsatellite instability/hypermutated, copy number-low microsatellite stable, and copy number-high/serous like.

Compared with other subtypes of tumors, prognosis for the copy number-high/serouslike group of patients is poor [183]. Poor prognosis is related to the loss of tumor suppressor *TP53* resulting in a high degree of genomic instability and rapid tumor progression and invasion [184,185]. As one-quarter of serous-like tumors have ERBB2 overexpression, there is a need to investigate the role of human epidermal growth factor receptor 2 (HER2) targeted inhibitors [186–188]. Considering molecular similarities between high-grade endometrioid and serous carcinomas, patients of this subtype may benefit from treatment

as if their tumor was serous. Moreover, specific mutations and overexpression of molecular targets in these tumors could tailor treatment in both the primary and recurrent setting.

From the clinical point of view, there is a need to create an integrated molecular risk profile for endometrial cancer. For that purpose, these four molecular subgroups were combined with additional molecular markers. Integrated molecular and clinico-pathologic risk assessment was based on a multivariate analysis of four molecular subgroups, clinical and histopathological characteristics of tumors, and various molecular classifiers. Molecular markers involved were *TP53* expression, *MSI*, *POLE* mutation, protein expression of *L1CAM*, *ARID1a*, *PTEN*, *ER/PR*, as well as analysis of 13 genes found to have variable expression in the TCGA classification groups (*BRAF*, *CDKNA2*, *CTNNB1*, *FBXW7*, *FGFR2*, *FGFR3*, *FOXL2*, *HRAS*, *KRAS*, *NRAS*, *PIK3CA*, *PPP2R1A*, and *PTEN*). This integrated model for prediction of endometrial cancer recurrence was confirmed to be more reliable than the traditional one relying on clinical and pathologic factors [189]. Moreover, this classification system enhances risk stratification of endometrial cancers. Importance for molecular subtyping was confirmed in clinical practice, as sorting of patients into molecular subgroups was confirmed to predict response rates to conventional, targeted systemic and radiotherapy [190,191]. For clinicians, molecular subtype stratification could be used in both preoperative evaluation (whether to prepare the whole set up for lymph node dissection or not) and postoperative treatment (the need for eventual adjuvant therapy). Moreover, molecular subtyping was confirmed to be very important for targeted therapy in patients with recurrent and metastatic diseases [175,177,192–194]. Thus, once the diagnosis of endometrial cancer has been established, there is a need to perform molecular subtyping, which will enable proper therapeutic approach.
