Precision medicine of ovarian cancer for adults and children : molecular, hereditary and clinical aspects

Abstract

Ovarian cancer was 2018 ranked as the eighth most common cancer diagnosis and cause of death in women. Of these, the most common histological type is high-grade serous carcinoma. There is a diagnostic challenge in the early detection of ovarian tumors leading to more advanced stages at diagnosis and poor prognosis which especially concerns the high-grade subtypes. This increases the need for continuous development of treatment strategies and there is an ongoing focus on personalized medicine. For example, BRCA1/2 pathogenic variant carriers respond well to platinumbased chemotherapies as well as PARP inhibitors. Not only high-grade ovarian cancer but also other subtypes may be associated with hereditary variants representing different cancer predisposition syndromes (CPS) pointing to the need for genetic counseling and testing.

Paper I of this doctoral thesis regards high-grade epithelial ovarian tumors and their association with the most common CPS associated with ovarian tumors, namely the Hereditary Breast and Ovarian Cancer Syndrome (HBOC). Molecular characterization was performed to evaluate potential diagnostic and actionable variants. 274 high-grade tumors were DNA sequenced using a targeted panel. In addition to mutations in TP53, BRCA1, and BRCA2 there were also findings in other homologous recombination repair pathway genes as well as other genes such as PTEN and CDKN2A. Sixteen previously unknown BRCA1/2 carriers were identified. Twenty patients received PARP inhibitor treatment based on the sequencing results.

The development of molecular methods during recent years has entailed findings of diagnostic and prognostic markers as well as potential treatment targets. These have shown the value of molecular characterization in parallel to morphology and immunohistochemical analyses. One of the discovered molecular markers is a specific somatic FOXL2 mutation in certain types of sex cord-stromal tumors (SCSTs). Germline and somatic DICER1 mutations in Sertoli Leydig cell tumors and SMARCA4 mutations in Small Cell Carcinoma of the Ovary, Hypercalcemic Type (SCCOHT) are two other examples. The role of the FOXL2 C134W mutation in SCSTs is further investigated in paper II of this doctoral thesis in which morphologic features were evaluated in relation to FOXL2 mutation status in 44 cases of SCSTs. The dominating growth pattern (72.7%) was diffuse/solid with several tumors showing marked heterogenous architecture. In the initial histopathological review, 12 cases were given an altered diagnosis based on morphology alone. The study concluded that one cannot reliably identify FOXL2 mutation-positive tumors solely by morphologic features.

In the pediatric population, ovarian tumors are rare. Also, the type of ovarian tumor differs from the adult population. Another difference is that the care for pediatric patients demands special consideration regarding the choice of treatment due to the constant growth and development of the affected child. These are important factors to support research into pediatric ovarian tumors. Paper III of this doctoral thesis has gathered a population-based cohort of 345 ovarian tumors in children aged 0-19 registered in the Swedish Cancer Registry between 1970-2014. Descriptive statistics are reported and when the material was available (n=260), morphologic review was performed. The study concluded that pediatric ovarian tumors are rare and distinct from their adult counterparts regarding incidence and frequency. The review diagnosis and original diagnosis had a strong concordance. However, 1/5 of cases had a change in diagnosis after review pointing to the value of central review. The major diagnostic challenge was the subtyping of epithelial tumors as well as differentiating between entities within the SCST group.

There is a further clinical need to improve molecular diagnostic analyses of ovarian tumors in order to understand what molecular events drive tumor development. Also, to find new potential treatment targets with the goal to offer more personalized treatment. So far, precision medicine regarding ovarian tumors has been mainly focused on searching for different targets in epithelial tumors, being much more common in the adult population. Recently, the focus is more and more on whole genome sequencing (WGS) and several large projects involve pediatric tumors, e.g. “The Individualized Therapy for Relapsed Malignancies in Childhood (INFORM) precision medicine”-study in Heidelberg, Germany, and the Swedish national project Genomic Medicine Sweden (GMS). One section of GMS focuses on pediatric tumors in collaboration with the Swedish Childhood Tumor Biobank and SciLife lab. The project has implemented a national structure for prospective whole genome sequencing of all pediatric malignant tumors at the time of diagnosis. Evaluation of the first analyzed 117 patients in the GMS pediatric project is described in paper IV.