According to the incessant ovulation hypothesis, early age at menarche and late age at menopause could increase the risk for OC through an increased number of ovulatory cycles. Conversely, according to the gonadotropin hypothesis, a late age at menopause delays the surge of postmenopausal gonadotropin hormones, possibly reducing OC risk. Numerous epidemiologic studies have examined the relation between lifetime menstrual history and OC risk. Results of studies that have examined the age at onset of menses are not terribly consistent [60–70]. For example, in a collaborative analysis of 12 US case-control studies conducted between 1956 and 1986, data from 2,197 White OC cases and 8,893 White controls detected no elevation in risk among women with onset of menses before 12 years of age [66]. Similarly, no statistically significant association was detected in the prospective Nurses’ Health Study cohort of 121,700 female registered nurses aged 30–55 years when the study began [69]. One Chinese study identified a significant protection with late age at menarche (after age 18) [71], while another study observed a slight increased risk with late age at menarche [72]. Additional research has failed to clarify the literature [53, 61, 73–78]. Data on age at natural menopause and OC risk are also inconsistent. Numerous case-control studies have identified an association between late age at menopause and the risk of OC, with odds ratios ranging from 1.4 to 4.6 in the highest category of age at menopause [60, 61, 63, 67, 71, 72, 76]. In the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort, age at menopause (>52 vs ≤45 years) was associated with an increased OC risk (HR = 1.57, 95 % CI: 1.16–2.13); however, after women diagnosed with OC within the first 2 years of follow-up were excluded, the risk was slightly attenuated and marginally statistically significant (HR = 1.40, 95 % CI: 0.98–2.00) [77]. The authors speculated that older women in the subclinical stage of OC may mistake bleeding for menses, which is why risk was attenuated when recently diagnosed cases were removed from the analysis. Other case–control studies [66, 68, 74, 75, 79–81] and several cohort studies [69, 73] found no such association. The collaborative analysis by Whittemore et al., for example, calculated an OR of 1.1 (95 % CI: 0.71–1.3) for menopause occurring after the age of 55 [66].

A recent study report from the Nurses’ Health Study and Nurses’ Health Study II found that age at natural menopause was associated with an increased risk of endometrioid tumors (RR = 1.13, 95 % CI: 1.04–1.22), but not serous invasive or mucinous tumors [17]. There are various explanations for the conflicting results regarding the relationship between ages at menarche and menopause and OC risk [82]. Besides the role of chance, it has been proposed that these differences may be explained through real differences between populations. Additionally, it is possible that the definition of menarche and menopause can be subject to recall and misclassification bias. It has also been pointed out that various populations have different age distributions and that some studies may have failed to adjust for age or other covariates in the analysis. In summary, it can be inferred from the available evidence that if early age at menarche and late age at menopause increase the risk of OC, the magnitude is likely small.

The association between pregnancy and OC risk has been studied extensively. Pregnancy causes anovulation and suppresses secretion of pituitary gonadotropins. Both the “incessant ovulation” and the “gonadotropin” hypotheses would predict that pregnancy reduces the risk of OC. Indeed, one of the most consistent findings is that parous women have a 30–60 % lower risk for OC than nulliparous women [53, 60, 67, 71–75, 80, 82–86]. Furthermore, each additional full-term pregnancy is estimated to lower risk by approximately 15 % [66, 73, 87]. While many case-control studies with hospital controls have shown positive associations with late age at first birth (≥30 years of age) [60, 65, 66, 74, 76, 83, 88–91], a reduced risk with late age at first birth has been identified in some case-control studies with population controls [64, 66, 92]. Recent data also suggests that OC risk does not vary by the time interval between the first and last birth [93].

It is unclear whether spontaneous or induced abortions impact OC risk. Although many investigations have found that an increased number of incomplete pregnancies may slightly decrease risk [53, 60, 65, 66, 72, 73, 94–96], others have reported risk to be increased among women with one or more incomplete pregnancies [75, 86], and a sizeable number of studies have yielded null results [64, 67, 68, 70, 74, 80, 83, 85, 97]. Induced abortions have been associated with lower risk in several studies [73, 95, 96], but not others [64, 76, 94]. With regard to spontaneous abortions and OC risk, positive [68, 83, 94], inverse [70], and null associations [71, 85, 95] have been reported. Interpretation of this literature is difficult because of the recognized potential for recall bias of spontaneous or induced pregnancies [59].

It is yet to be determined whether nulliparity and low parity per se, rather than difficulty becoming pregnant due to female infertility, is the relevant factor. Infertility is a term that is used to describe a heterogeneous group of biologically distinct conditions ranging from genital tract infections and tubal disturbances to medical conditions such as endometriosis and polycystic ovarian syndrome [98, 99]. Infertility appears to be associated with increased OC risk in most studies [60, 66, 70, 74, 80, 83, 85, 86, 91, 98], but not all [73, 100]. Infertility seems to pose the greatest risk among women who remain nulliparous, while periods of temporary infertility among parous women are of little concern [60, 66, 70, 74, 85]. For example, in a large Canadian case-control study in which most nulliparous women were so by choice, infertility was not associated with OC risk among parous women, but there was a trend towards elevated risk among a small group of infertile nulliparous women (OR = 2.5, 95 % CI: 0.6–4.1) [70].

Possible reasons for the inconsistent results may include the failure to examine the various types of infertility separately. Furthermore, it has been reported that some factors such as a personal history of endometriosis [101–103] or polycystic ovarian syndrome [104] may influence both infertility and OC risk. The definition of infertility used across studies is variable, including physician-diagnosed infertility, self-reported infertility, and periods of unprotected intercourse without becoming pregnant [59]. A particular challenge is trying to distinguish an influence of infertility from an adverse effect of fertility drug exposure. Although some studies report that women with a prior history of fertility drug use who remain nulliparous are at an elevated risk for ovarian tumors, particularly tumors of low malignant potential [66, 105], the results are not consistent [98–100, 106–108]. Early detection bias may explain the discrepant findings, as early stage cancers may be overdiagnosed in infertile women due to the close medical surveillance [109].

Lactation suppresses secretion of pituitary gonadotropins and leads to anovulation, particularly in the initial months after delivery [6]. If the incessant ovulation and gonadotropin hypotheses are true, lactation should reduce the risk of OC. Although the majority of studies have identified a slight decrease in OC risk with lactation, with odds ratios approximating 0.6–0.7 [66, 67, 70, 84–86, 110–113], some have not [64, 68, 80]. Despite the conflicting results, the overall impression is that lactation protects against epithelial OC, especially in the first few months following delivery.

Several gynecologic conditions have been examined as risk factors for OC, including polycystic ovarian syndrome (PCOS), endometriosis, and pelvic inflammatory disease (PID). PCOS is a heterogeneous disease often characterized by obesity, hirsutism, infertility, and menstrual abnormalities. The association between PCOS and OC risk was investigated using data from the Cancer and Steroid Hormone Study, a population-based case-control study [104]. Among 476 histologically confirmed epithelial OC cases and 4,081 controls, 7 cases (1.5 %) and 24 controls (0.06 %) reported a history of PCOS (OR = 2.5-fold, 95 % CI: 1.1–5.9) [104]. The association appeared to be stronger among women who never used oral contraceptives (OR = 10.5, 95 % CI: 2.5–44.2) and women in the first quartile of body mass index (13.3–18.5 kg/m²) at age 18 (OR = 15.6, 95 % CI: 3.4–71.0), but these estimates have wide confidence intervals. Larger studies that adjust for potential confounders of the PCOS-OC association are needed before conclusions can be drawn regarding these findings.

Endometriosis is one of the most common gynecologic disorders, affecting 10–15 % of women in reproductive years [114]. Even though endometriosis is considered a benign condition, it has been linked with OC in the medical literature since 1925. Sayasneh and colleagues [114] recently reported a systematic review of eight studies; seven found an increased risk of OC, with effect sizes ranging from 1.3 to 1.9. The strongest associations were evident among endometrioid and clear cell histologies, consistent with molecular data that supports the uterus as the origin of these subtypes [7]. However, the authors suggest that the association between endometriosis and endometrioid and clear cell ovarian carcinomas may represent sharing of similar risk factors rather than a causal association [114], a topic that merits further research.

PID causes inflammation of the endometrium, fallopian tubes, and ovaries. Previous studies from the 1990s that evaluated the association between PID and OC risk yielded inconsistent results [115, 116]. Recently, Lin and colleagues [117] evaluated this association in a large nationwide cohort from Taiwan that included 67,936 women with PID (42 of whom later developed OC) and 135,872 women without a history of PID (48 of whom developed OC). A history of PID was found to be a risk factor (adjusted HR = 1.92 (95 % CI: 1.27–2.92)), especially among subjects diagnosed with PID before the age of 35 and women who had at least five episodes of PID. Note, however, that the absolute rates of OC among women with PID are clearly low overall.

Several gynecologic procedures appear to influence the risk for OC. It is well established that among high risk women, bilateral prophylactic oophorectomy decreases OC risk by at least 90 % [118]. Numerous studies have identified a reduced risk of OC associated with either a hysterectomy or tubal ligation (without oophorectomy), with the protective effect for each of these procedures ranging from 30 to 40 % [60, 70, 119–124]. For example, a recent meta-analysis estimated that tubal ligation reduced OC risk by 34 % [125]. Furthermore, the risk reduction from these procedures appears to last for at least 10–15 years, which argues against screening bias (due to selective removal of subclinical ovarian tumors) as the basis for the findings [81, 120, 126, 127]. Although it is uncertain how these procedures reduce the risk of OC, removal of the uterus and/or blockage of the tubes may prevent potential carcinogens from ascending the genital tract [62] and decreases blood flow to the ovaries [127]. In particular, Vercillini and colleagues [128] hypothesize that retrograde menstruation (i.e., menstrual fluid flows backwards into the fallopian tubes instead of leaving the body through the vagina) may promote iron-induced oxidative stress and subsequent cancer development in the fallopian tubes and ovaries.

The 30–40 % lower risk of ovarian cancer among women who ever used oral contraceptives is firmly established. The findings are consistent over the past several decades, even as the drug formulations evolved from high estrogen and progestin content popular in the 1960s to decreasing hormone content in the mid-1970s, and in the early 1980s when the sequential compounds (biphasic and triphasic) were introduced [129]. The risk reduction increases with duration of use [66, 70, 130–133] by at least 5 % per year, with about a 50 % reduction in risk for long-term use of 10 years or greater, [134] and persists long after use has ceased [80, 84, 132, 135–138]. Moreover, the risk reduction is not confined to any particular type of combined oral contraceptive formulation [139, 140] or to any histologic type of ovarian cancer, although the inverse relation is less consistent for mucinous cancers [11, 13, 16, 141]. There are few epidemiologic studies that have evaluated progestin-only contraceptives, mostly due to the rarity of this exposure, but the existing data suggest they too lower risk of ovarian cancer [84, 132, 142].

Oral contraceptive use corresponds to the avoidance of approximately 3,000–5,000 ovarian cancer cases and 2,000–3,000 deaths per year in both Europe [20] and in North America [143]. The use of OCs therefore has implications for individual risk assessment and on a public health scale.

Few recent studies have examined methods of contraception other than oral contraceptives and tubal ligation. In a population-based case-control study of 902 epithelial OC/tubal/peritoneal cases and 1,800 controls, Ness and colleagues [124] found that ever use of an intrauterine device (IUD) was associated with lower risk of OC (adjusted OR = 0.75, 95 % CI: 0.59–0.95). The benefit was evident with short duration of IUD use (≤4 years), but risk progressively increased with longer duration of IUD use (albeit nonsignificantly). The authors suggested that shorter use may reduce upper genital tract inflammation by killing sperm, while longer use may imply more insertions and greater risk of infection and inflammation. IUD use has previously been associated with an increased OC risk (RR = 1.76, 95 % CI: 1.08–2.85) among women in the Nurses’ Health Study [144]; however, most IUD use in their study occurred in the 1970s–1980s prior to the newer IUD formulations. Contrary to results from the Nurses’ Health Study [144] in which spousal vasectomy was not associated with OC risk (multivariate adjusted OR = 0.87, 95 % CI: 0.63–1.19), Ness and colleagues [124] observed vasectomy to be protective (adjusted OR = 0.77, 95 % CI: 0.61–0.99). The authors [124] speculated that vasectomy may confer a slight risk reduction from reduced exposure to sperm. Given that contraceptive methods are modifiable, these findings need to be replicated.

The benefit of oral contraceptives on OC risk is well established; however, the data on another exogenous hormone, HRT, is less clear. It has been postulated that HRT may reduce OC risk by decreasing the secretion of gonadotropins. However, the reduced levels are still above those of premenopausal women [145]. Conversely, postmenopausal HRT may increase OC risk due to increased estrogen-induced proliferation of ovarian cells [146].

Initial studies on the topic focused on unopposed estrogen therapy. In the collaborative reanalysis of 12 US case-control studies, no association was identified with duration of HRT use in either hospital-based (OR = 0.90 for a 5-year increment of use, p = 0.37) or population-based (OR = 1.10 for a 5-year increment of use, p = 0.21) studies [66]. Several case-control studies [147, 148], cohort studies, [149] and meta-analyses [150, 151] found no association with duration of use, although two have observed either a significant increase or a suggestive trend towards increased risk [13, 152]. Data from recent studies, including four meta-analyses, now indicate an increased OC risk for ever users of HRT [153–156]. Furthermore, several prospective studies have found that longer durations of HRT use are associated with OC risk or death [157–160]. For example, in the Nurses’ Health Study cohort, both current and past HRT users of 5 or more years had a significantly higher risk for OC than never users current users (RR = 1.41, 95 % CI: 1.07–1.86) and past users (RR = 1.52, 95 % CI: 1.01–2.27) [161]. Based on their statistical modeling, the authors concluded that the elevated risk appeared to be driven largely by duration rather than by status of use. Additionally, in the UK Million Women Study [153], 2,273 incident ovarian cancers were observed among 948,576 postmenopausal women who did not have a prior cancer history or a bilateral oophorectomy. For current users of HRT, incidence of OC increased with increasing duration of use, but did not differ significantly by type of preparation used and its constituents or mode of administration.

Only recently have studies had sufficient statistical power to evaluate associations between combined estrogen and progestin use and OC risk. The effects of unopposed estrogen therapy (ET) are thought to be more detrimental to the ovaries than estrogen plus progestin (EPT) [162]. It is postulated that progestins promote apoptosis, while estrogen promotes proliferation of ovarian epithelial cells [162]. Most studies that investigated the association between EPT use and OC risk have found no association or a weak association [141, 153, 154, 156, 159, 161–164]; however, not all studies support a protective role for EPT. A few prospective studies [153, 158, 165] and meta-analysis [155] have reported a small increased risk for EPT users. In support of a weaker association for EPT, a recent meta-analysis of 14 population-based studies found that ET is associated with an increased risk of OC (RR = 1.22 for a 5-year increment of use, p < 0.0001); however, the risk among women who used EPT was attenuated (RR = 1.10 for a 5-year increment of use, p = 0.001) [154]. The authors suggest that the addition of progestin mitigates the effect of estrogen, because the increased risk of OC among EPT users was statistically significantly lower than the risk among ET users, p = 0.004 [154]. However, a large nationwide prospective cohort study of Danish women observed an increased risk both for ET users (RR = 1.31, 95 % CI: 1.11–1.54) and for EPT users (RR = 1.50, 95 % CI: 1.34–1.68) [165].

A recent cohort study investigated the association between HRT use and obesity on OC risk [166]. Among HRT nonusers, weight gain, waist circumference, and waist-to-hip ratio but not BMI increased the risk of OC [166]. HRT use of more than 5 years increased OC risk, but risk was not further increased for women who were overweight and used HRT. For example, while substantial weight gain (greater than 40 lbs) and HRT use of more than 5 years individually increased the risk of OC, the joint effect did not further increase the risk, which may imply a threshold effect [166]. Some studies have pointed to an increased risk only for certain histologic subtypes of OC. For example, the Nurses’ Health Study cohort observed that the association with ET was slightly stronger for endometrioid tumors, which is consistent with other studies [17, 148, 167]. A link between ET and the development of endometrioid ovarian tumors is biologically plausible because endometrioid tumors are histologically similar to endometrial tissue [168], and ET use increases the risk of endometrial cancer [146]. However, although risks associated with HRT use varied significantly according to tumor histology (p < 0.0001) in the UK Million Women Study [153], the relative risk for current versus never users of HRT was greater for serous than for mucinous, endometrioid, or clear cell tumors (1.53 (1.31–1.79), 0.72 (0.52–1.00), 1.05 (0.77–1.43), or 0.77 (0.48–1.23), respectively).

It can be concluded from the available evidence that if an association exists between HRT use and OC, the magnitude is probably moderate, but women should be counseled about the potential increase in risk with long-term use of unopposed estrogen. Evidence suggests that the OC risk with ET alone is higher than the risk associated with EPT. Since many women are exposed to HRT several years before the peak age-specific incidence of OC, even a small change in risk may have a significant impact on disease rates at the population level.