The hormonal composition of follicular fluid and its implications for ovarian cancer pathogenesis

Despite our growing understanding of the cancer genome and the evolution of targeted therapies, ovarian cancer remains the most lethal gynecological malignancy in the Western world [1]. The clinical outcomes for this disease have not changed significantly over the past four decades, in large part due the lack of early detection tools and the almost inevitable emergence of chemo-resistant disease [1]. Adding to the complexity of this cancer is its heterogenous nature. Although ovarian tumors can arise from three different cell types: epithelial cells, germ cells, and sex cord stromal cells, the vast majority of ovarian cancers are epithelial in nature. Even within epithelial ovarian cancers there exist various histologic subtypes and molecular subgroups. This has prompted the classification of epithelial ovarian cancers into two groups. Type I tumors, clearly linked to ovarian precursor lesions, encompass all histologic subtypes including low grade serous, endometriod, mucinous, and clear cell carcinomas. They are defined by their slow growth and multiple genetic mutations. In contrast, Type II tumors are highly aggressive, confer a much poorer prognosis, and many have been linked to precursors arising from the fallopian tube epithelium. High Grade Serous Ovarian Cancer (HGSOC) is the most common of the Type II tumors. Type I and Type II tumors are also genomically distinct. Type I tumors are frequently associated with specific mutations in oncogenes such as k-RAS and ARID1A [2]. In Type II tumors, TP53 is mutated in the vast majority of tumors (96-100%) and appears to be the earliest genetic event in HGSOC. BRCA1 and BRCA2 mutation carriers are particularly susceptible to Type II tumors [2, 3].

While much work has been done to characterize the pathology and genetics of ovarian cancer, we still lack a basic understanding of the early events and causes of this disease. Our current understanding of the risks factors of HGSOC derives primarily from epidemiological data. Lifetime ovulation is positively correlated with HGSOC, and factors such as parity and birth control, which decrease lifetime ovulation, have a protective effect against HGSOC [4, 5]. Establishing a molecular mechanism linking ovulation and HGSOC pathogenesis is critical to developing screening techniques and treatments for this disease.

The origins of ovarian cancer are complex and still under debate. New theories suggest that different ovarian tumor subtypes have different origins, with the ovarian surface epithelium implicated in Type 1 tumors and fallopian tube secretory epithelial cells implicated in high grade serous ovarian cancer [2, 3, 6]. The adherent mesothelium of the ovary attracts both shed fallopian tube cells as well as endometriosis-derived Müllerian tissue, further blurring the lines of the ovary vs. Müllerian tissue debate [7].

Understanding the various cells of origin in ovarian cancer similarly informs our understanding of ovarian cancer pathogenesis. Several models have been posited to explain how epidemiological factors such as menstruation and ovulation may lead to ovarian cancer. A long standing hypothesis, often called the Incessant Ovulation Hypothesis, suggests that the repetitive wounding and healing of the ovarian surface epithelium and adjacent tubal epithelium that is induced by monthly ovulation increases cell proliferation and thus the likelihood of genomic instability which could lead to oncogenesis (Figure 1) [4]. Another hypothesis, known as the Gonadotropin hypothesis, implicates excessive direct and indirect stimulation of the ovarian surface epithelium by gonadotropins, leading to differentiation, proliferation, and ultimately malignant transformation (Figure 1) [5]. More recently, the Incessant Menstruation hypothesis suggests that repeated exposure to retrograde menstruation exposes the ovary and fallopian tube to reactive oxygen species and oxidative iron from the blood [8]. Lastly, several recent papers have focused on damage induced by inflammation-mediated factors found in the follicular fluid [9, 10]. The recurring theme in all these hypotheses is the incessant ovulatory damage. This reinforces the importance of ovulation, but also makes it difficult to separate the impact of each hypothesis as they are physiologically interconnected. Women with altered steroid hormone levels, such as those with Polycystic Ovarian Syndrome (PCOS), tend to ovulate sporadically, while women who take oral contraceptives receive the protective benefits of both lowered gonadotropin levels and inhibited ovulation [5, 11]. The average age of onset in ovarian cancer is postmenopausal, at age 63, when hormonal levels have shifted and ovulation has stopped [12]. Thus, an unanswered question that also remains is why menopause is so temporally important to the onset of the disease.

Follicular fluid is often studied in the context of livestock reproduction, and abattoirs provide a readily available source of whole ovaries and fallopian tubes. While valuable research has been obtained using non-human models, the advantages of increased follicle volume and sample availability must be considered against the limited relevance of working with non-human FF [24]. Humans are the only known mammals to develop ovarian cancer naturally, and the levels of major hormones in cow and camel follicles is known to be several orders of magnitude lower than their human counterparts [25, 26]. Domesticated laying hens, despite being oviparous, provide a unique model in that they are extremely susceptible to ovarian cancer and reflect several aspects of human disease [27]. Thus, a close examination of human FF, though harder to obtain and manipulate, is most relevant to studying the relationship between FF composition and ovarian cancer pathogenesis in humans.

Many recent studies have been conducted on the composition of human FF to assess conditions which are linked to ovarian cancer risk, as well as the efficacy and prognostic value of various in vitro fertilization (IVF) methods. This review gathers the most recent studies where human FF factors have been measured and comprehensively examines their fluctuation both under normal conditions and due to various conditions or treatments which are linked to HGSOC risk, and any role they may play in cancer pathogenesis (Table 1). Ultimately, developing a better understanding of FF and its hormonal composition will illuminate its potential role in ovarian cancer pathogenesis.

The hormonal profile of follicular fluid is complex and constantly in flux, changing dramatically over the course of follicle development, as well as throughout a woman’s lifetime depending on her age, health status and fertility. Understanding the hormonal microenvironment of ovulation is critical to establishing a molecular link between incessant ovulation and early ovarian cancer pathogenesis. While studying the role of hormones in ovulation is important, no less important is the role of hormones in the transition to menopause, when ovarian cancer is typically diagnosed. Why this disease so often manifests itself after risk factors such as ovulation have come to an end, and when supposedly protective progesterone levels have risen and likely damaging estrogen levels are low is a puzzle at the crux of the ovarian cancer problem [21, 35]. In the efforts to understand disease onset, cessation of ovulation and the hormonal milieu accompanying it will be as important as ovulation itself.

Currently, one of the most promising leads providing a molecular basis for the contribution of FF to ovarian cancer pathogenisis was reported by Bahar-Shany et al. [15]. Using human fallopian tube ex-vivo cultures and pooled human FF, they found that exposing fallopian tube epithelium to FF stimulated inflammatory and DNA repair pathways and resulted in the upregulation of pro-angiogenic and pro-inflammatory interleukin 8. They noticed that follicular fluid exposure led to DNA double stranded breaks and, consequently, the stabilization of the tumor suppressor TP53. Early precursors of high grade serous ovarian cancer are also defined by their high expression of TP53 and high levels of DNA damage, although in the vast majority of these cases TP53 is also mutated, often with a gain of function mutation [71]. Understanding the link between the temporary induction of TP53 in response to FF exposure and the aquisition of mutations in P53 in early precursor lesions will be key in the future of ovarian cancer research.

The role of hormones in the later stages of ovarian cancer is also a field ripe for study. The relationship between hormonal microenvironment and tumor is complicated, in no small part because the hormone receptors of many primary tumors and ovarian cancer cell lines have been inactivated either directly or indirectly. The lack of response to hormone signaling is evidenced by the largely unsuccessful use of hormonal therapy in ovarian cancer, especially compared to its sweeping successes in breast cancer [72, 73].

In the fight against ovarian cancer, many factors beyond the hormonal milieu play a role, and follicular fluid is by no means the sole initiator of tumorigenesis. Other theories, including telomere shortening [74, 75] have been proposed as well. Ultimately, providing the molecular link between epidemiologic risk factors and disease mechanisms will have broad implications not only for ovarian cancer, but for infertility and development as well.