Disorder of primordial follicle endowment
X chromosome dysfunction
Monosomy (Haploinsufficiency)
Trisomy or polysomy X, or mosaic variants
Autosomal/X translocations
Macrodeletion Xp or Xq
Xq (long arm)
Basic helix–loop–helix protein (BHLHB9)
Fragile X mental retardation (FMR1) syndrome
X-linked mental retardation
XIST and X-inactivation
DIAPH2
DACH2
POF1B
Xp (short arm)
X-propyl aminopeptidase 2 (XPNPEP2)
USP9X (Ubiquitin-Specific Protease 9)
Zf (Zink Finger X)
Bone Morphogenic Protein (BMP 15)
Transforming Growth Factor-Beta (TGFb)
Autosomal mutations
Autoimmune regulator (AIRE)
Deleted in azoospermia-like (DAZL)
Homologue of yeast disrupted meiotic cDB + NA 1(DMC 1)
Eukaryotic translation initiation factor 5B (EIF5B)
Estrogen receptor 1 (ESR1)
Homologue of murine factor in germline α (FIGLA)
Forkhead transcription factor (FOXL2)
β chain of follicle-stimulating hormone (FSHB)
Follicle-stimulating hormone receptor (FSHR)
Galactose 1-phosphate uridyltransferase (GALT)
Growth differentiation factor 9 (GDF9)
G protein-coupled receptor 3 (GPR3)
Type II 3-β-hydroxysteroid dehydrogenase deficiency (HSD3B2)
Inhibin alpha (INHA)
Luteinizing hormone, β polypeptide (LHB)
LIM homeobox gene 8 (LHX8)
Homologue of Escherichia coli MutS, 5 (MSH5)
Homologue of murine newborn ovary homeobox (NOBOX)
Homologue of murine noggin (NOG)
Nuclear receptor subfamily 5, group A member 1 (NR5A1)
Progesterone receptor membrane component 1 (PGRMC1)
DNA polymerase γ (POLG)
Transforming growth factor-β receptor, type 3 (TGFBR3)
γ box-binding protein 2 (γBX2)
Breast Cancer BRCA1, BRCA 2
NANOS
Enzyme deficiencies
CYP17 A1
Aromatase CYP19
Carbohydrate-deficient glycoprotein (PMM1 mutation)
Galactosemia (GALT)
Depletion/destruction of ovarian follicles
Spontaneous follicle loss
Ovarian autoimmunity
Infectious
Toxins environment lifestyle
Treatment-associated follicle loss
Surgery
Chemotherapy
Radiation
Dysfunction of ovarian follicles
Signaling defect
FSHb, FSHR, LHb, LHR mutations
GNAS (Pseudohypoparathyroidism type 1a)
1.
Disorders of primordial follicle endowment/genetic causes
2.
Depletion/destruction of ovarian follicles
3.
Dysfunction of ovarian follicles
Disorders of Primordial Follicle Endowment
Approximately, 15 % of POI is familial, an observation that first suggested the role of genetics in the pathophysiology of POI [1, 2]. The majority of described genetic causes involve the X chromosome, though there is increasing evidence also implicating autosomal genes. Although the list of discovered genetic mutations is extensive, the precise pathophysiologic mechanisms upon which these mutations exert their negative effect on folliculogenesis and oogenesis are still poorly understood. In general, most causative single gene mutations are extremely rare with few published descriptions.
The following represents a selection of known genetic causes of POI, including chromosomal and single gene mutations that potentially play a role in POI via impaired folliculogenesis or oogenesis.
The X Chromosome and POI
Structural or numerical abnormalities of the X chromosome constitute the largest proportion of identified genetic causes of POI, including deletions, trisomy X, X chromosome mosaicism, monosomy X (Turner syndrome), and balanced translocations between the X and autosomal chromosomes.
Monosomy X (45,X) Turner Syndrome
Turner syndrome is due to partial or complete loss of the second sex chromosome and affects 1 in 2500 live-born female infants [3]. Approximately, 50 % of Turner syndrome females have a classic 45,X karyotype, while the remaining affected females demonstrate mosaic karyotypes or structural abnormalities of the second X chromosome [4]. The two most prominent phenotypic findings are short stature and primordial follicle depletion [5]. Normal ovarian development occurs in the fetus, but accelerated follicular atresia leads to the development of gonadal dysgenesis and infertility [6]. One possible explanation for the increased rate of oocyte attrition is haploinsufficiency for multiple loci on the X chromosome [7]. Alternatively, oocyte depletion may be secondary to dysfunctional meiosis caused by errors in homologous pairing, as the solitary X is lacking an appropriate homologue [8, 9]. Despite this, spontaneous puberty and even pregnancy are reported in approximately 15 % and 3 % of Turner syndrome patients, respectively, with higher rates found in mosaic females [10].
Trisomy or Polysomy X
Trisomy X is caused by nondisjunction of the X chromosome during meiosis. The incidence in the newborn population is estimated to be 1 in 1000 live-born females [11]. Although the association of trisomy X and POI has been reported in the literature, the prevalence of POI in this group is unknown. Several case reports have documented the association of trisomy X with POI in patients presenting with primary amenorrhea, secondary amenorrhea, or oligomenorrhea and hypergonadotrophic hypogonadism [12, 13].
X Chromosome Long Arm (Xq)
Two critical regions for ovarian function have been identified on the long arm of the X chromosome, termed premature ovarian failure 1 (POF1) (Xq21.3-qter) and premature ovarian failure 2 (POF2) (Xq13.3-q21.1) [14, 15]. Deletions or rearrangements involving these two regions have been described with primary and secondary amenorrhea possibly due to haploinsufficiency or disrupted regulation of transcription of critical genes, or failure of meiotic pairing in pachytene, leading to apoptosis [16, 17]. Balanced X-autosome translocations have been shown in some instances to result in the downregulation of autosomal genes important in oogenesis due to epigenetic effects via chromatin modifications [18].
Investigation into candidate genes within the POF1 and POF2 regions has been mixed, with a number of genes interrupted by translocations in affected families; however, the majority of balanced translocations occur in gene-poor regions [19, 20]. Bione et al. presented a POI family with an X; 12 balanced translocation that caused interruption of DIAPH2. The DIAPH2 protein is a member of a protein family important to cytokinesis and other actin-mediated processes critical to early development [21]. Mutations in the Drosophila homologue dia are known to cause sterility in male and female fruit flies [22]. The authors proposed that DIAPH2 plays an important role in oogenesis, however, corroborating evidence in humans is lacking [21].
Fragile X mental retardation 1 (FMR1) premutation is the most common single gene cause of POI and is found in 11–14 % of familial POI and 2–3 % of sporadic POI [23, 24] The FMR1 gene , located at Xq27.3, is associated with three distinct phenotypes, Fragile X syndrome, Fragile-X associated tremor/ataxia syndrome (FXTAS), and FMR1 related POI. All three conditions are X-linked dominant with incomplete penetrance and are related to a CGG trinucleotide repeat in the 5′ untranslated region of FMR1. Repeat numbers greater than 200 lead to silencing of FMR1 gene expression and Fragile X syndrome in males and approximately half of females. Females with 55–200 CGG repeats are considered premutation carriers and are at risk of having a child with Fragile X syndrome due to anticipation. Additionally, 15–20 % of FMR1 premutation carrier females develop POI [25]. Risk of ovarian failure does not appear to be completely linear with regard to repeat length, where the risk increases to lengths of around 99, but then plateaus [24] (Fig. 5.1). While the absence of gene expression clearly explains the Fragile X syndrome phenotype, the underlying mechanism(s) of FMR1 related POI are less certain. It is hypothesized that RNA-mediated toxicity may affect ovarian follicle dynamics, as mRNA is overproduced in the premutation carrier [26].

Fig. 5.1
Expression of FMR1 in normal women, premutation carriers, and full mutation carriers. Figure adapted from Hagerman and Hagerman [24]. Wittenberger. FMR1 premutation. Fertil Steril 2007
X Chromosome Short Arm (Xp)
The proximal region of the short arm of the X chromosome (Xp11.1–Xp21) is essential for ovarian germ cell function; deletions in this region result in primary amenorrhea or secondary amenorrhea in approximately equal proportions [16, 27]. Bone Morphogenic Protein-15 (BMP15) is located at Xp11.2 and is an oocyte-specific growth/differentiation factor known to play an important role in the regulation of folliculogenesis and granulosa cell proliferation, although the precise underlying mechanism remains incompletely characterized [28]. Additionally, BMP15 effects are likely significantly influenced by the actions of growth and differentiation factor 9 (GDF9) , due to the formation of biologically active BMP15: GDF9 heterodimers [28]. The first BMP15 mutation associated with POI was detected in two sisters with gonadal dysgenesis with a heterozygous mutation in BMP15 inherited from their hemizygous unaffected father. This mutated BMP15 likely exerted a dominant negative effect on the wild-type BMP15, possibly by producing an antagonistic effect on granulosa cells [29]. Many other BMP15 mutations have since been identified in women with POI [30].
Autosomal Mutations and POI
The advent of molecular technologies significantly advanced our knowledge of single gene mutations associated with POI. Despite this, the introduction of clinical genetic testing for most single gene causes has only recently been introduced. This at least in part stems from the incomplete understanding of the underlying mechanism of many mutations and the relative rarity of identifiable single gene causes of syndromic or isolated POI. The following section discusses some selected autosomal genes involved in POI, but is far from comprehensive. Additionally, other genetic causes with delineated mechanisms are described later under their appropriate categorizations (Fig. 5.2).

Fig. 5.2
Age-specific prevalence of premature ovarian failure by CGG repeat length. Based on a dataset of 429 FMR1 premutation carriers and 517 noncarriers. Wittenberger. FMR1 premutation. Fertil Steril 2007
Blepharophimosis–Ptosis–Epicanthus Inversus Syndrome
Blepharophimosis–Ptosis–Epicanthus Inversus Syndrome (BPES) is an autosomal dominant disorder characterized by distinct eyelid dysmorphology and, in BPES type I, POI. BPES is caused by mutations in FOXL2, which is highly expressed in the developing eyelid and ovaries, as well as in adult human ovaries [31] FOXL2 is one of over a hundred Forkhead Box transcription factors and plays a critical role in granulosa cell differentiation and ovary maintenance. Primordial cells consist of a single layer of squamous granulosa cells surrounding the oocyte which then develop into mature follicles during folliculogenesis. Mature follicles then produce activin-βA and AMH, inhibiting the remaining primordial follicle pool. In a murine model of Foxl2 homozygous mutants, the granulosa cell differentiation was blocked at the squamous follicle to mature cuboidal follicle transition. Consequentially, activin-βA and AMH are not produced and almost all primordial follicles are activated and undergo apoptosis. This results in premature depletion of the primordial follicle pool and may explain the underlying mechanism of POI in BPES [32].
Other isolated autosomal defects include transcription factor mutation such as BRCA, newborn ovary homeobox protein (NOBOX), growth factor mutations (GDF9), LIM homeobox (LHX8), and RNA binding protein (NANOS3) gene mutations.
BRCA Mutations
While overt POI does not appear to be associated with BRCA1 and BRCA2 carrier status, Oktay et al. in 2010 found that in a cohort of women undergoing ovarian stimulation for fertility preservation, low ovarian response was significantly more common in mutation carriers than noncarriers, which the authors termed occult POI. This relationship was strongest for BRCA1 carriers, who showed decreased oocyte yield when compared to noncarriers [33, 34] BRCA1 is one of many genes involved in the repair of DNA double-strand breaks. Titus et al. showed in a murine model that Brca1-deficient mice showed impaired reproductive function, with decreased total number of primordial follicles and increased DNA double-strand breaks in existing primordial follicles [35]. Therefore, oocytes may be more prone to DNA damage because of defective DNA repair mechanisms associated with BRCA mutations [34].
Newborn Ovary Homeobox
NOBOX is a homeobox transcription factor highly expressed in the oocyte and known to play a role in primordial to primary follicle development [36]. A mouse knockout model for NOBOX was consistent with a human POI phenotype [37]. Sequencing of NOBOX in POI cohorts has revealed mutation rates of 5–6 %, suggesting that NOBOX may be the most common autosomal gene implicated in POI [38, 39].
LHX8 and NANOS
LHX8 is a germ cell specific regulator of oogenesis through transcription factors NOBOX and FIGLA. These may then function as determinants of oocyte-specific gene expression throughout the remainder of folliculogenesis. NANOS is involved in the maintenance of primordial germ cells by supporting their proliferation and/or by suppression of cell death. Female knockout mice for either gene lack germ cells [40, 41].
GDF9 Mutations
An important autosomal gene affecting folliculogenesis is growth differentiating factor 9 (GDF9). GDF9 is an oocyte-secreted growth factor that belongs to the TGF family and plays a pivotal role in differentiation of the oocyte, granulosa, and theca cells. GDF9 is located on chromosome 5 (5q31.1), expressed in oocytes, and believed to influence differentiation of the layers of the follicle compartments (oocyte, granulosa, and theca cells) [42]. Gdf9 null mouse models have provided new insight into the role of GDF9 in ovarian function. While a complete loss of Gdf9 in homozygous males showed that it is not required for male fertility, homozygous null females were infertile and had significantly smaller ovaries compared to the wild type. Furthermore, a loss of granulosa mitotic ability in the primary follicle stage appears to result in the absence of normal follicular development beyond the primary stage in homozygous null females. Aromatase and LH-receptor expression is also lacking in Gdf9 null ovaries and consequently theca cell recruitment is impaired [43]. Further evidence for the pivotal paracrine role of GDF9 in folliculogenesis comes from studies in ewes either homozygous for GDF9 mutations or immunized against GDF9, with all animals exhibiting a block in follicle growth at the primary stage [44, 45]. While not a common cause, variants in GDF9 thought to be likely causative have been reported in women with POI [46, 47].
Enzyme Deficiencies
A number of inherited enzymatic pathway disorders have been associated with ovarian follicle dysfunction leading to POI. Among the first described disorders is galactose 1-phosphate uridyltransferase (GALT) enzyme deficiency causing galactosemia, a rare autosomal recessive disorder. The GALT gene encoding this enzyme is located at 9p13. Almost all women with homozygous mutations in the GALT gene eventually present with POI, but fluctuating POI course and spontaneous pregnancies have been observed in girls with classic galactosemia [48]. FSH is often elevated even in early childhood. Neonatal ovaries have normal morphology and folliculogenesis. However, histological examination of adolescent women with galactosemia shows hypoplastic, streak-like ovaries containing severely decreased primordial follicles without any mature follicles [49, 50]. Although the mechanism underlying POI in women with galactosemia is still unknown, possibilities include the accumulation of toxic metabolites after birth leading to direct ovarian insult. Other possibilities involve FSH dysfunction through hypoglycosylation of glycoproteins or glycolipids, oxidative stress, and activation of oocyte apoptosis. Autoimmune mechanisms have also been discussed; however, no antibodies have been identified to date [51, 52]. Despite early diagnosis and treatment with newborn screening, development of POI is the most common long-term complication in affected females and appears to be treatment independent [50, 53, 54].
PMM2-CDG is an autosomal recessive congenital disorder of glycosylation characterized by abnormal glycosylation of N-linked oligosaccharides and multisystem involvement. Females with the syndrome have early-onset hypergonadotrophic hypogonadism, and the underlying mechanism is poorly characterized, but likely related to the global dysfunction of glycoproteins and glycolipids seen in the disorder [55].
Aromatase (CYP19A1) catalyzes the conversion of androgens to estrogens, thereby playing a key role in intact ovarian follicle function. CYP19A1 mutation s have been reported in two 46,XX women with cystic ovaries and primary amenorrhea. Compound heterozygosity existed for two different point mutations on exon 10 in both women, resulting in a mutant protein without enzymatic activity [56, 57].
Depletion/Destruction of Ovarian Follicles
Ovarian Autoimmunity
Approximately, 10–30 % of women with POI have a concurrent autoimmune disease, with autoimmune thyroid disease being most frequently associated. Other autoimmune associations described include systemic lupus erythematosus (SLE), Crohn’s disease, myasthenia gravis, type I diabetes mellitus, pernicious anemia, vitiligo, alopecia areata, ulcerative colitis, celiac disease, glomerulonephritis, rheumatoid arthritis, primary biliary cirrhosis, multiple sclerosis, autoimmune hemolytic anemia, idiopathic thrombocytopenic purpura, pernicious anemia, vitiligo, and hypoadrenalism [1].
There are multiple findings supporting an autoimmune-mediated destruction in patients with suspected autoimmune oophoritis and subsequent progressive decrease in ovarian function. These findings include autoantibodies to steroid producing cells such as the adrenal cortex, testis, placenta, and the ovary, which can be isolated in the sera of 60–80 % of these women. Increased lymphatic and plasma cell infiltration of the theca cells of the growing follicle and T-cell CD4+ and CD8+ activation have also been described [58, 59]. Interestingly, inflammatory processes appear to spare early primordial and primary follicles and affect mostly later stages of folliculogenesis (pre-ovulatory follicles and corpora lutea) [60].
There are three categories that have been used to describe autoimmune ovarian insufficiency: those associated with adrenal autoimmunity, non-adrenal autoimmunity, and isolated idiopathic POI [61]. In cases where POI is associated with either non-adrenal immunity or idiopathic, the prevalence of autoantibodies directed against several types of steroid-producing cells such as the adrenal cortex, testis, placenta, and ovary is significantly lower (approximately 10 %). This compares with up to 80 % of women with adrenal autoimmunity and POI. Autoimmunity of the ovary and the presence of anti-ovarian antibodies (AOA) have also been associated with the outcome of in vitro fertilization. It has been suggested that the presence of AOA could potentially reduce fertilization rates, generate a poor response to gonadotropin stimulation, decrease pregnancy rates, affect egg and embryo development, and could be responsible for implantation failures [62]. In implantation failure, anti-ooplasm antibodies or antizonal antibodies can be isolated in up to 25 % of women, with even higher prevalence rates with fertilization failures. It has been proposed in the past to test women for the presence of AOA prior to initiation of assisted reproductive therapy programs as a prognostic factor for reproductive outcomes, and to help counsel patients accordingly [63, 64], though these are not currently standard of care because of lack of effective treatments which can address these and the high success rates of IVF/ICSI regardless of antibody status.
There is still little known about the exact antigenic targets within the ovary; however, ongoing research has identified AOA against various components and compartments of the ovary which can be subdivided into the germinal component, including autoantigens directed to the oocyte and the surrounding zona pellucida, and the somatic component, encompassing autoantigens to steroid producing cells, such as the corpus luteum, granulosa cells, and theca cells.
Anti-oocyte antibodies were first described in the hallmark study by Vallotton and Forbes in 1966 [65]. Since then, a number of studies have successfully demonstrated the presence of anti-oocyte antibodies in the cytoplasm of IVF patients and women with POI [66, 67]. The literature investigating the identity of the corresponding antigens in human ovaries is sparse, though mouse models in development have provided some rich information. POI appears to develop in mice after neonatal thymectomy. Tong and Nelson demonstrated that these mice develop anti-oocyte cytoplasm antibodies against an oocyte-specific protein termed OP1 (ooplasm-specific protein 1) and, therefore, proposed that OP1 as an antigen may play a role in murine autoimmune POI [68]. Antigen-heat shock protein 90-beta (HSP90β) is a second antigen that was demonstrated predominantly in the oocyte cytoplasm, but could also be identified in theca and corpus luteum cells of women with infertility. It has been suggested to use antibodies against HSP90 for diagnostic markers of ovarian failure [69].
The zona pellucida (ZP) , which is also a part of the germinal component surrounding the oocyte, is another important ovarian antigen in autoimmune POI. It has been suggested that anti-ZP antibodies correlate with lower fertilization rates in IVF patients and have been successfully isolated from sera of POI patients [70, 71]. However, further research is needed to investigate the role of anti-ZP antibodies in the pathophysiology of ovarian autoimmunity, as current research consists of studies with small samples sizes and questionable specificity of the methodology. The clinical significance of these antibodies is questionable in part due to a broad range of anti-ZP antibody prevalence (between 2.4 and 68 %) as well as positive reactions in healthy females and male patients [72–75].
Among AOA, autoantibodies directed to steroid producing cells (SCA) are recognized as the best markers for autoimmune POI [76, 77]. These antibodies are predominantly found in patients with Addison’s disease and to a lesser degree in non-adrenal or idiopathic POI [16, 78]. SCA are believed to target cytoplasmatic steroid enzymes, specifically anti-3β-hydroxysteroid dehydrogenase, anti-17 hydroxylase, and anti-side chain cleavage enzyme [79, 80]. It has furthermore been suggested that gonadotropins themselves and their corresponding receptors could potentially function as autoantigens. The concept of anti-gonadotropin and anti-gonadotropin receptor antibodies has been investigated, with evidence suggesting that circulating anti-FSH and anti-LH antibodies in sera of women undergoing IVF were associated with poorer reproductive outcomes [81, 82].
Moving forward, the detection of multiple antibodies in women with POI would likely prove to be most valuable in predicting reproductive outcomes. However, current ovarian antibody assays have a poor specificity and the association between the aforementioned antibodies with autoimmune POI has not been consistently confirmed. Therefore, testing for these antibodies is not recommended at this time. Furthermore, with regression of autoimmune disease, recovery of ovarian function may occur.
Infectious
The evidence whether or not viral infections such as mumps oophoritis, herpes zoster, varicella, HIV, or cytomegalovirus could trigger POI remains controversial. The true incidence of post-mumps oophoritis-related POI is unknown, but it has been reported in up to 7 % of cases, with return of ovarian function in the majority of patients following resolution of the disease. There have also been reports describing microbial infections such as shigella and malaria associated with POI, though a causative link has not been confirmed [83].
Recent studies have focused on POI in patients with HIV. In 2010 a prospective study evaluated 78 HIV-seropositive women for four markers of ovarian function and found that all women showed abnormalities in antral follicle counts, FSH, inhibin B, and anti-Müllerian hormone. These results imply that HIV infection may impair fertility, have a negative effect on ovarian function, and consequently lead to POI. A multicenter cohort study with 1139 HIV-seropositive women found that these women were approximately three times more likely to have prolonged amenorrhea, defined as no vaginal bleeding for at least 1 year, compared to seronegative women. However, elucidating whether the underlying cause of the prolonged amenorrhea was of ovarian origin proved to be problematic and remains to be investigated. At any rate, the data highlight the importance of monitoring this population early for signs and symptoms of POI [16, 84–88].
Toxins, Environmental, and Lifestyle Factors
Studies examining the association of chemical substances and POI in humans are rare; however, the plausibility of a causal relationship was illustrated in a group of Korean women exposed from 4 to 16 months to 2-promopropane (2BP or isopropylbromide), which is used as a solvent for cleaning electrical components. Sixteen out of 26 women were subsequently found to develop POI, with FSH values corresponding to menopausal levels. Regular menses ceased in the majority of women with only 13 % resuming menstruation later [89, 90]. A large, case–control study of 443 cases and 508 controls in 2009 demonstrated that hairdressers who are regularly exposed to chemicals may be at a higher risk for POI compared to other occupational groups [91]. With the assistance of animal models, further light has been shed on the effect of various other environmental and occupational factors, such as polycyclic aromatic hydrocarbons (PAH) , whose corresponding receptor can be isolated in oocytes as well as male spermatozoa, thereby potentially contributing to male infertility. The impact of these toxins has been shown to range from decreased ovarian function to induction of POI [92–94].
Smoking is a known cause of ovarian follicle damage, decreasing the age of menopause by up to 4 years in large epidemiologic studies, though the exact mechanisms and toxins are unclear [95–99]. PAH is one of the toxins found in tobacco and may affect the ovarian follicle pool by accelerating follicular atrophy and atresia via increased apoptosis in primordial germ cells [100]. Nicotine also inhibits aromatase, which results in decreased estradiol levels and downstream negative effects on fertility [101]. A causal relationship of smoking and POI is supported by a dose-dependent effect of cigarettes on the reproductive life span, starting at one half pack per day.
4-Vinylcyclohexene diepoxide (VCD) has also been reported to be an ovotoxic occupational chemical and is produced during the manufacture of rubber tires, flame retardants, insecticides, plasticizers, and antioxidants. Similarly to PAH, it targets ovarian primordial follicles and accelerates ovarian failure in rodents via cell death by apoptosis and altered expression of specific genes. Women exposed to VCD are, therefore, at a higher risk for POI and should be monitored carefully [102–107].
A number of additional ovotoxins have been described in the literature, but it has been challenging to illustrate the underlying mechanisms leading to POI and conflicting results regarding the effect of these toxins on the human ovary have been reported [108]. Different mechanisms involved in chemical-induced POI have been postulated. First, a follicular ovarian pool that is initially normal is destroyed by an ovotoxic external agent, thereby inducing POI. Multiple chemical agents associated with increased apoptosis or atresia of the primary or primordial follicles have been identified and at least four of these toxins [ethylene glycol methyl ether, 2,2-bis (bromomethyl)-1,3-propanediol, benzo[a] pyrene, and dimethylbenzanthracene] are potentially capable of inducing POI in the next generation after prenatal exposure in rodents [94, 109, 110].
Another theoretical pathway leading to POI by follicular destruction is ovarian hypoxia, as chronic hypoxia in men is already a known cause for oligoasthenospermia and in women ovarian torsion with subsequent ovarian hypoxia and necrosis may result in subfertility [111, 112]. Furthermore, it has been postulated that the follicular pool is prematurely depleted via increased follicular recruitment, either by agents acting on hormonal regulation of follicular recruitment or inhibiting mechanisms that maintain follicle dormancy. Two synthetic organic compounds have been associated with increased follicular recruitment: methoxychlor , which is an organochlorine insecticide, and bisphenol A [113, 114]. Yet another mechanism hypothesizes that chemical agents may not interfere with follicle pool quantity but influence follicle quality by impairing follicle maturation in the setting of normal follicle counts. Chemical agents that match this mechanism have been only hypothetically described in the literature.
Body composition and adiposity have also been evaluated in relation to reproductive life span. Interestingly, POI is not associated with weight gain and affected patients are usually less likely to be obese than unaffected controls [115]. Overall, environmental and lifestyle factors are thought to be the minor causes of POI at this time, but further research is needed to evaluate circumstances, dose, and frequency of exposure to ovotoxic agents in order to accurately assess occupational risks.
Treatment (Iatrogenic) Associated Accelerated Follicle Loss
Surgery
Surgical menopause and POI can result from bilateral oophorectomy for ovarian disease, prophylactic bilateral oophorectomy, such as in cases of BRCA-positive women, excessive laparoscopic drilling in women with polycystic ovary syndrome (PCOS) , or ovarian damage during the removal of large ovarian cysts or endometriomas. Hysterectomy, salpingectomy for ectopic pregnancies or hydrosalpinx, and almost any other pelvic surgery have been theorized to compromise blood supply to the ovaries and induce local inflammation, thereby placing women at a higher risk for POI [116]. This theory is supported by the observation that the average age of menopause in women who had previously undergone hysterectomies is significantly earlier (by approximately 4 years) than the average age at menopause for women who retained their uteri [117, 118]. Similarly, uterine artery embolization has the potential to induce POI by affecting the vascular supply to the ovaries [16]. While excision of endometriomas prior to assisted reproductive therapies is not recommended in order to improve reproductive outcomes, it can provide other benefits, such as decreasing pelvic pain. Laparoscopic removal of endometriomas carries a risk of POI of about 2.4 % [119]. The negative effect of surgery on ovarian reserve has been illustrated in a meta-analysis of 237 patients by Raffi et al. in 2012, in which a significant postsurgical decrease in AMH levels was observed after ovarian endometrioma excision [120].
Even though some studies suggest that ovarian reserve is diminished in women with PCOS who undergo laparoscopic ovarian drilling (reflected by postsurgical changes of hormonal levels), it has been alternatively postulated that the changes in ovarian reserve markers such as FSH, inhibin B, and AMH levels after ovarian drilling represent a normalization of ovarian function and do not reflect diminished ovarian reserve [121]. There is a general consensus that in women who desire to preserve their fertility, recurrent pelvic surgery for benign causes should be limited, care should be taken to avoid iatrogenic ischemia to the ovaries, and PCOS drilling needs to be appropriately and conservatively applied to restore normal ovary morphology and normalize endocrinologic properties.
Chemotherapy
Due to advanced treatment regimens for malignant diseases, survival rates for many childhood cancers treated with chemotherapy have continued to improve over time. However, one of the long-term consequences for female survivors exposed to chemotherapy is POI, which significantly impacts quality of life and the ability to have biological offspring. The main predictive factors for the development of POI following chemotherapy include the age of the individual, the drug type, higher cumulative dosage, and the addition of radiation to the treatment [122]. Alkylating agents, such as cyclophosphamide, used for the treatment of some cancers and autoimmune disorders, are considered to be highly gonadotoxic. This is presumably because the drugs are not cell cycle specific and therefore both resting and growing primordial follicles can be damaged. Alkylating agent cause POI in 42 % of women and even up to 80 % of women if additional chemotherapeutics are concomitantly used [123, 124]. Cell cycle specific agents such as methotrexate, 5-fluouracil, bleomycin, and vinca alkaloids are associated with mild gonadotoxicity, whereas cisplatin and adriamycin are moderately toxic. Several studies have demonstrated a dose-dependent relationship between chemotherapy and the onset of POI, though the effect of duration of treatment is not well understood [125]. Rodents treated with cyclophosphamide had primordial follicle destruction in a dose-dependent relationship [126, 127].
Interestingly, prepubertal girls are relatively resistant to the gonadotoxic effect of chemotherapy, while older women have a much higher incidence of POI following treatment [128]. Women between ages 21–25 years have a 27-fold increased risk for POI after chemotherapy, compared to an approximately fourfold risk in the adolescent group. Women aged 35–40 years are at highest risk for POI following chemotherapy, although it is difficult to ascertain to what extent this is attributable to ovotoxic effects of the treatment or a naturally smaller primordial follicle pool in this age group at the time of therapy [16, 129]. The impact of chemotherapy on the ovary is similar to the physiologic changes during normal ovarian aging and includes follicular apoptosis, premature atresia, blood vasculature compromise, and cortical fibrosis leading to a reduced follicle pool and eventually to morphologic changes of the ovary [130–132]. Furthermore, administration of chemotherapy has been shown to have both transient and permanent effects on ovarian function. Transient effects result from the growing follicle pool being destroyed by the treatment, resulting in transient amenorrhea, though recruitment from an intact primordial follicle pool is subsequently initiated and the population of growing follicles can be restored. Permanent effects and POI result if the primordial follicle pool itself is sufficiently impacted and diminished. Whether POI occurs immediately after treatment or years later depends on whether the primordial pool has been completely or partially depleted. This hypothesis is supported by a prospective study in 2006 that aimed to investigate ovarian function as reflected by monthly menses in premenopausal women undergoing breast cancer treatment. Five hundred and ninety five women received predominantly cyclophosphamide, doxorubicin, paclitaxel, fluorouracil, methotrexate, and docetaxel resulting in rapid decrease of monthly menses throughout all age groups after the first dose of adjuvant therapy was given. Thereafter, menstrual cycling recovered rapidly in 85 % of women younger than 35 years, within 6 months after the end of treatment. Recovery was less pronounced in older women, with only 61 % of 35 to 40-year-old women resuming normal menses 6 months after treatment [133]. However, permanent ovarian dysfunction arises not only from direct cytotoxic effects of chemotherapy on primordial cells, but also by repetitive loss of activated growing follicles occurring with repeated chemotherapy. Ultimate exhaustion of the primordial cells from repetitive treatment has been referred to as “burn-out” or “second-hit” hypothesis, where the presence of less than a thousand primordial follicles is associated with menopause [134]. Therefore, both indirect and direct mechanisms are capable of reducing the size of the primordial pool, thereby leading to permanent ovarian follicle depletion and POI [131]. These findings lead to the question of whether certain stages and associated cell types involved in folliculogenesis are more susceptible to the ovotoxic effects of chemotherapy than others. Unfortunately, there is little data available regarding the relative sensitivities of the processes of follicle initiation, growth, selection, and maturation to chemotherapy and resultant effects on primordial, primary, secondary, pre-antral, early antral, and pre-ovulatory follicles. A decrease in number of primordial follicles is well established in both human and experimental rodent models following chemotherapy [135, 136]. Petrillo et al. exposed mice ovaries to phosphoramide mustard, a metabolite of cyclophosphamide, and demonstrated an induction of primordial and small primary follicle depletion in >90 % as early as 24 h following treatment [137]; however, whether this is a primary or secondary effect remains to be determined. Doxorubicin seems to affect secondary follicles, where changes in ovarian reserve markers during chemotherapy suggest declining numbers of pre-antral and early antral stages. The current data, however, does not allow definitive conclusions regarding the stage of folliculogenesis that is more susceptible to chemotherapy damage [138, 139].

Stay updated, free articles. Join our Telegram channel

Full access? Get Clinical Tree

