Possible Role of Autoimmunity in Patients with Premature Ovarian Insufficiency

Premature ovarian insufficiency (POI) is characterized by hypergonadotropic amenorrhea due to cessation of ovarian function before the age of 40 years. The diagnosis is based on amenorrhea before the age of 40 associated with follicle stimulating hormone (FSH) levels >40 IU/l, detected on two occasions at least one month apart (1). POI causes female infertility, while is a significant psychosocial burden and a risk to women’s health. It occurs in 1% of women, of whom 10-28% have primary and 4-18% secondary amenorrhea (2, 3).

Although there are multiple etiologies of POI (genetic, chromosomal, infectious, and iatrogenic causes), the etiology cannot be identified in most patients and this is referred to as idiopathic POI; up to 30% of idiopathic cases may have an autoimmune cause (4). The most convincing evidence coming from the commonly observed association of POI with other autoimmune disorders (5, 6) are demonstration of anti-ovarian antibodies (AOA, 7, 8) and histological findings of ovarian tissue from affected women. Roughly, one third of POI patients have AOA and/or antithyroid antibodies in their serum (1, 9). Various organ-specific and systemic autoimmune diseases cause autoimmune ovarian insufficiency in up to 30% of women with POI (4). According to the literature, 2-10% of POI cases are known to be associated with adrenal autoimmunity (10). One of the first signs that autoimmunity may be responsible for ovarian function failure came from the observation that ovarian failure may precede the onset of Addison’s disease by 8-14 years (11). Autoimmune Addison’s disease seldom develops in isolation, whereas several other endocrine glands and organs are generally affected, leading to an autoimmune polyglandular syndrome (APGS, 12). Two main forms of APGS can be clinically discerned, APGS types 1 and 2. APGS type 1 is characterized by an association of mucocutaneous candidiasis, hypoparathyroidism and Addison’s disease. In about 60% of cases, there is also an association with ovarian insufficiency. Blizzard et al. (13) and Irvine et al. (14) found that POI commonly presents with adrenal cytoplasmic antibodies, called steroid cell antibodies (SCA); they react with cytoplasmic antigens of other steroid-producing cells present in the ovary, testis and placenta. Alteration of lymphocytes and their specific subsets, as well as T-cell mediated injury are likely to play an important role in the pathogenesis of autoimmune oophoritis. Surface markers of peripheral blood mononuclear cells (PBMC) have been shown to be deranged in early autoimmune phases and to be persisted through the disease, even after targeted disruption (15).

The presence of pathogenic factors might accelerate the process of apoptosis and atresia of ovarian follicles during the fetal and post-natal period (16). This interpretation is based on the dogma that the number of ovarian follicles at birth is final and that there is no possibility of regeneration or renewal of reserve follicles in adulthood (17). Experimental work on animals suggests a possibility of renewal of the follicle reserve from proliferative germinal ovarian cells even after birth; verification of which is being sought in studies on human ovaries (18, 19). It has been shown that undifferentiated ovarian stem cells differentiate into structures similar to egg cells under certain laboratory conditions (20). We faced two problems: whether to accept the standard understanding of a final number of ovarian follicles or to focus on the hypothetic possibility of renewal of the follicular reserve. The aim of this study was to evaluate the involvement of immune abnormality in patients with idiopathic POI.

To present POI as a possible autoimmune abnormality, we focused on three potentially interconnected factors: personal and familial history of autoimmune disorders, peripheral blood T-lymphocytes levels and presence of AOA.

We performed a targeted history of personal and familial autoimmune disorders and found associated autoimmune disorders in our patients. Thyroid disorders were common in personal histories and diabetes type 1 in familial histories, confirming the findings of previous studies (5, 6, 15). One of the reasons for suspecting an autoimmune etiology of POI is its frequent association with nearly all organ-specific autoimmune diseases (1, 3). Autoimmune diseases are significantly more frequent in young women than in men. This phenomenon may be explained by the effect of sex steroids on the components of the cellular immune system, which might contribute to the development and progression of autoimmune POI (23).

Hoek et al. (24) found that 60% of patients with secondary amenorrhea and Addison’s disease have a detectable SCA serum titre. These antibodies are shown by 60-80% of patients with APGS type I. In 25% of our patients, the Synacthen test revealed an abnormal cortisol response. We interpreted this as the possibility that patients could be positive for SCA, but had a normal cortisol response. Patients with this disorder have an insidious onset; to confirm subclinical autoimmune adrenal insufficiency, measurement of adrenal antibodies might be a more effective screening method (25, 26).

Thyroid disorders and the presence of antithyroid antibodies are often mentioned in association with POI. Thyroid disorders are the most common of the autoimmune diseases associated with POI, found in 12-39% of women with POI (27- 29). Thyroid disorders are often associated with endometriosis and polycystic ovary syndrome, two conditions often resulting in infertility (2). We found a greater involvement of aTG and aTPO in the study than in the control group, and the explanation being that ovarian failure may have been present in the latent period of thyroid disease. We found a strong correlation between autoimmune thyroid disease and autoimmune POI; 50% of patients tested positive for aTG. We concluded that not all patients with positive aTG necessarily have a clinically expressed thyroid disorder and the disease can have an insidious onset. Furthermore, greater involvement of other immune-mediated diseases, particularly auto immune disorders, such as allergy, psoriasis, atopic dermatitis and vitiligo in the study group suggests an autoimmune cause of POI, consistent also with the findings in the literature (9, 15, 30, 31).

Genetic predisposition is known to be one of the primary causes of autoimmunity and it is generally observed that patients with autoimmune diseases have several types of antibodies, as also confirmed in our study. We found AOA in 20% of patients; these results are consistent with other studies (9, 21, 32-35). The prevalence of AOA in women with POI, studied with IF on ovary tissue, varies greatly from 2 to 70%, which supports their role as a marker of an immune dysfunction process against ovaries. There could be several reasons for the differences among study results, including the diverse origin of tissue sections, which include human and non-human primate ovaries, as well as different study inclusion criteria. Many studies have shown that the presence of serum AOA does not correlate with the clinical manifestation of POI. Despite these antibodies being present, their pathogenetic role is highly questionable (36, 37). AOA may occur several years before the occurrence of clinical symptoms, as detected in 33-61% of women with unexplained infertility; a situation that may indicate early stages of autoimmune ovarian insufficiency (38, 39). For most autoimmune diseases, screening for specific antibodies is probably the best way of evaluating immunological involvement. Many ovarian structures are potential targets for autoimmune events with POI. In addition to AOA, SCA, aTG and aTPO, antibodies to several potential ovarian antigens have been proposed as markers of ovarian autoimmunity, which could potentially mediate autoimmune damage in POI. Betterle et al. (40) found antibodies against steroid genetic enzymes in some cases with anti-adrenal autoimmunity. Antibodies against gonadotropin receptors and gonadotropins have also been found in patients with POI, but they are still the subject of research. The general opinion is that more research is needed (41, 42). Although some studies have managed to identify autoantibodies against zona pellucida, corpus luteum and ovarian cells, these findings have no real correlation with the clinical picture (43-45). In our preliminary study, AOA were determined semi-quantitatively by the IIF method. IIF is a basic method and widely used for determining auto antibodies, while more specific tests enabling detection of AOA for specific antigen targets are not available in our laboratory.

Infertile women with AOA also have a decreased response to gonadotropin stimulation and reduced pregnancy rate after treatment (46, 47). In 2002, it was found that low-responders to gonadotropins with AOA are younger than low-responders without AOA (48). Detection of autoimmune processes that affect the ovarian response should thus be included in the diagnostic workup before any infertility treatment, particularly in women with low or no response to gonadotropins (34). Determination of AOA, as a specific test, is important in the diagnosis of diseases with an autoimmune etiology, but it should not be the only reliable diagnostic tool for optimal selection of patients who may benefit from immune modulatory therapy that could, at least temporarily, re-establish their ovarian function and fertility.

Abnormalities of cellular immunity of T lymphocytes, macrophages and dendritic cells play an important role in autoimmune events. Some of these abnormalities have been seen in women with POI, confirming the potential existence of an autoimmune mechanism of the disease. Mignot et al. (49) found that the absolute number and percentage of peripheral blood T lymphocytes, especially CD4+ T cells, are increased in patients with POI. Moreover, Miyake et al. (50) found that patients with POI have low levels of CD8+/CD57+ Tcells (cytotoxic T lymphocytes) and an increased ratio of CD4+ to CD8+ cells, which may reflect cell cytotoxic cell migration from blood to inflamed tissue. Multiple animal studies have suggested that the basis of POI is a cell-mediated autoimmune reaction caused by an alteration in T cell regulation (18). CD4+ T cells with constantly expressed α chain (CD25) receptor for IL-2 of were the first detected mediators of inhibition of autoimmune diseases in mice alteration of suppressor T cell subsets and T cell abnormalities are likely to play an important role in the pathogenesis of autoimmune diseases (51). Regulatory CD4+25+ T cells show a potent immunosuppressive function in vitro and in vivo, and contribute to immunologic self-tolerance by suppressing potentially auto-reactive CD4+ T cells. There is known to be a number of immunological mechanisms associated with the failure of immune tolerance and the development of autoimmunity. Although studying regulatory T cells in human autoimmune diseases is difficult, and at times, findings have been contradictory, the data suggest that defects in CD4+ CD25+ regulatory T cells mediated suppression (52) are a major subset of immune cells responsible for peripheral immune self-tolerance.

In agreement with studies in patients with systemic lupus erythematosus, multiple sclerosis, rheumatoid arthritis and autoimmune vasculitis, we confirmed a reduced number of CD4+CD25^+high T cells in the peripheral blood of our patients. High expressions of CD25 and CD4 surface markers have classically been used for identification of regulatory T cells. This may be problematic since CD25 is also expressed on antigen-responding activated non-regulatory T cells. The additional measurement of cellular expression of Foxp3 protein allowed a more specific analysis of Treg cells (CD4+CD25+Foxp3+ cells). However, Foxp3 is also transiently expressed in activated human effector T cells, thus complicating a correct Treg analysis. The large majority of Foxp3-expressing regulatory T cells express high levels of the interleukin-2 receptor alpha chain (CD25). Since there are no cell surface markers that are uniquely and specifically expressed on all Foxp3-expressing regulatory T cells, the measurement of CD4+CD25^+high T cells is still in use in clinical studies of peripheral blood lymphocytes.

We interpreted the reduced number of CD4+CD25^+high T cells as a possible mechanism contributing to the formation of an autoimmune response in association with the presence of AOA and aTG. We also found an increased number of B cells in peripheral blood. A similar picture has been observed with other autoimmune endocrinopathies; therefore, we interpreted the elevated B cell count as activation of the humoral immune system, crucial for autoantibody production. Some authors, though, have tried estrogen substitution in women with POI without any effect on peripheral B cell count (53, 54).

The hormones inhibin B, FSH and AMH have been proposed as potential markers for determining the functional ovarian reserve (55, 56). In young ovulatory women, measurement of AMH at 3-year intervals has shown that the AMH serum level decreases significantly over time, whereas other markers associated with ovarian aging, such as FSH, inhibin B and antral follicle count (AFC), do not change during this time period. Since it decreases at a time when concentrations of FSH and inhibin B are still normal, AMH has been proposed as the best indicator of ovarian reserve and as a marker of ovarian aging (55-59). In contrast, AMH is almost undetectable in women with POI (60), which our study also confirmed. In our group of patients, there was a small AFC or these structures were no longer seen on ultrasound, which agrees with the data in the literature (61).

Clinical and biological characteristics of women without known causes of disease suggest a possibility of autoimmune pathogenesis. In some patients, a combination of various autoimmune processes has been found. The presence of AOA and anti-thyroid antibodies, together with abnormalities of cellular immunity, potentially represent an autoimmune mechanism of POI. There is thus an increasing need to find a reliable and simple diagnostic procedure to determine the true prevalence of autoimmune ovarian disease. In women with POI, more attention should be paid to evaluation of associated autoimmune disorders.