Abstract
Some evidence indicates endometriosis and migraine have a common genetic predisposition in sex-hormone genes, which could have important implications for the treatment of these two heterogenous conditions. To date, the genes responsibility remains unknown. Based on the biological hypothesis that polymorphisms of genes involved in sex-hormone pathways may influence estrogen levels and phenotypes of both disorders, we did a literature search for candidate sex-hormone genes and genes involved in the metabolism of estradiol. The aim was to review the evidence for shared sex-hormone-related polymorphisms between endometriosis and migraine and provide an exhaustive overview of the current literature. We included case-control studies investigating associations between candidate sex-hormone-related genes and the disorders endometriosis and migraine, respectively. Results showed three overlapping sex-hormone-associated polymorphisms in estrogen receptor genes that are associated with both conditions. To confirm possible associations with other sex-hormone genes, larger studies are needed.
Introduction
Endometriosis and migraine are two distinctive disorders associated with chronic pain, inflammation and a high grade of disability. Endometriosis is a common gynecological disease where endometrium-like tissue is manifested outside the uterine cavity. 1 Migraine, on the other hand, is a neurological disorder distinguished by recurrent headache attacks. 2 Based on the high prevalence in women during their reproductive life phase, it is assumed that estrogens or other female sex hormones might play a crucial role in the pathophysiology of both conditions. There is also some evidence that both conditions might share a common genetic background.3,4 The lifetime prevalence of endometriosis is at least 10% and the global prevalence of migraine in women is 18.9%, which corresponds to over a billion women suffering worldwide.5,6 Both conditions generate a huge social and economic burden and exert a significant negative impact on women’s quality of life.7–9 Migraine is the number one cause of disability in women during their reproductive years (aged 15 - 49 years).10,11 Large twin studies indicate that both disorders have a heritable trait, with heritability estimated to be between 30–60% for migraine and circa 50% for endometriosis, indicating that genes are of importance in the etiology of both conditions.12–16 Therefore, it has been suggested that endometriosis and migraine might be comorbidities, at least in a subset of women. 3 Menstruation is a major trigger for endometriosis-associated pain and menstrual-related migraine.17,18 In women, migraine typically starts during puberty and resolves during menopause. More than 50% of women report an association between migraine and their menstrual bleeding. 19 Both disorders are estrogen-dependent and may lead to exceptionally strong symptoms during menstruation.18,20
The role of female sex hormones in the pathogenesis of migraine is well established. 21 The drop of estrogen levels at the end of the menstrual cycle and at the beginning of the hormone-free interval in users of exogenous estrogens plays a crucial role in the pathophysiology of migraines in women.22–24 In addition, it has been shown that the use of combined oral contraceptives in the standard regimen may initiate and worsen migraine in predisposed women.25–28 Interestingly, a positive impact on the frequency of migraine episodes and the intensity of pain was observed in a continuous regimen with Desogestrel, a progestin-only contraceptive.29,30 Likewise, endometriosis symptoms can be successfully treated with hormones, in particular progestins.31,32
In line with the biological hypothesis that polymorphisms of genes involved in sex-hormone pathways may influence estrogen levels, we aimed to study the main sex-hormone genes (ESR1, ESR2, PGR, FSHR, AR, SHBG) and selected the most regularly studied genes affecting the metabolism of estradiol (COMT, NRIP1, CYP1A1, CYP17A1, CYP19A1). 33 There are two types of estrogen receptor (ESR) genes; estrogen receptor 1 (ESR1 or ER-alpha) and estrogen receptor 2 (ESR2 or ER-beta). Estrogen levels rise in the early phase of the cycle as a result of increasing levels of follicle stimulating hormone (FSH). During the menstrual cycle, not only estrogen levels but also levels of FSH, progesterone, and androgens fluctuate. Therefore, FSH receptor (FSHR), progesterone receptor (PGR), and androgen receptor (AR) genes must also be investigated as a potential source for a genetic predisposition. The PGR gene has two protein isoforms that modulate the biological action of progesterone: isoform A (PRA), which is capable of inhibiting the activation of the estrogen receptors, and isoform B (PRB), which has the capacity to activate the estrogen receptors. 34 Similar to the ESR, the PGR can undergo ligand-independent activation and is involved in various intracellular signaling pathways. A specific polymorphism in the PGR gene called PROGINS seems to impact the ligand-binding and the entire signaling pathway. 35 Moreover, the sex hormone-binding globulin (SHBG) gene, which codes for a glycoprotein that binds to androgens and estrogens, is of interest. For the final step of estrogen biosynthesis, aromatase enzymes such as Cytochrome P450, family 19, subfamily A, polypeptide 1 (CYP19A1), which is involved in the conversion of androgen to estrogen, are essential. Another enzyme of this family, Cytochrome P450, Family 17, subfamily A, polypeptide 1 (CYP17A1), mediates both 17-alpha-hydroxylase and 17,20-lyase, which play a key role in androgen biosynthesis. 36 Finally, Cytochrome P450, family 1, subfamily A, polypeptide 1 (CYP1A1) participates in this process by catalyzing estrogen hydroxylation in extrahepatic tissues. 37 Other genes involved in estrogen metabolism, like catechol-estrogen or its products, including catechol-O-methyl-transferase (COMT) and nuclear receptor interacting protein 1 (NRIP1), could also have an impact. NRIP1 negatively regulates the transcription of estrogen receptors, particularly ESR1. 38 Catechol-O-methyltransferase (COMT) is an omnipresent enzyme of the estrogen metabolizing pathway that catalyzes O-methylation and subsequently inactivates estradiol metabolites. 39
The association between most of these sex hormone gene variants (ESR1, ESR2, PGR, FSHR, AR, SBHG) and estradiol-metabolizing enzymes (NRIP1, COMT, and CYP family members) has been studied in women with migraine or endometriosis, however mostly for each disorder separately.40–42 One large Australian family study with 815 monozygotic and 457 dizygotic twin-pair sisters with surgically confirmed endometriosis investigated whether the combination of the two conditions is the result of chance, selection bias, or common environmental and genetic factors. The findings suggest that endometriosis and migraine have common genetic predispositions with a bivariate heritability of 17%. 3
To our knowledge, no previous study has specifically assessed the shared polymorphism between endometriosis and migraine. A possible genetic link could lead to better therapeutic care of these patients who suffer from chronic pain.
The aim of this narrative review is to obtain evidence for shared sex-hormone related polymorphisms in migraine and endometriosis. We aspire to present a comprehensive review of the existing literature and rationale for new research. We included studies focusing on sex-hormone receptor polymorphisms as well as those investigating enzymes affecting the metabolism of estradiol.
Materials and methods
Search strategy
We searched the PubMed database for publications from between January 2000 and July 2021 on sex hormone polymorphisms in endometriosis and migraine, respectively. To search and include as many related studies as possible, we used the MeSH terms “endometriosis” and “migraine disorders” with different combinations of the keywords for each sex hormone receptor. For Estrogen Receptor 1, for example, we used: ESR1, ESR1 gene, ESR alpha, ESR alpha gene, ESR1 polymorphism, ESR alpha polymorphism, estrogen receptor alpha polymorphism, estrogen receptor 1 polymorphism. We did the same for ESR2, FSHR, PGR, AR, SHBG and the estrogen-metabolizing association genes NRIP1, CYP1A1, CYP17A1, CYP19A1 and COMT. For each sex hormone-related gene included in the present study, a literature search was performed to find case-control studies and potential previous reviews and/or meta-analyses in order to identify potential additional records through other sources.
Study selection
The eligibility of each publication was assessed independently by two reviewers (JFV and GM). After the titles and abstracts were screened, the full text of articles eligible for our review was examined. Studies had to meet the following inclusion criteria: 1) published between January 2000—July 2021, 2) original case-control study investigating associations between polymorphisms in the candidate sex hormone genes (ESR1, ESR2, FSHR, PGR, AR, SHBG) and the estradiol metabolizing association genes (NRIP1, CYP1A1, CYP17A1, CYP19A1, COMT) and 3) diagnosis of endometriosis confirmed by surgery and/or histology and diagnosis of migraine according to the criteria of the International Headache Society (IHS). 2 While reviewing publications, a novel gene, the FSH beta-subunit (FSHB) gene, consisting of a FSH beta-subunit (FSHβ), was added to our gene selection. 43 A new literature search for this FSHB gene was carried out. We excluded systematic reviews and/or meta-analyses, commentary articles, and duplicated studies. In the endometriosis studies, control groups with unrelated males or neonates were excluded, considering endometriosis is a disease occurring in women of reproductive age. However, we included one study with a very large sample size that included both parents of the women with endometriosis, due to their comparable hereditary genetic makeup. Studies with self-reported or self-diagnosed endometriosis or migraine were excluded. We excluded two case-control studies that investigated the combination of two sex hormone polymorphisms44,45 (Supplemental Table S1). Figure 1 summarizes the study selection process for identifying eligible studies.

Flow diagram of the literature search and study selection process.
Data extraction
For each eligible study, data were extracted by one investigator (JFV) using a standardized Microsoft Excel spreadsheet and checked by a second investigator (GM). Discrepancies were resolved through discussion. The following data were listed for each study: Author, year of publication, country in which the research was conducted, ethnicity of the women, sample size, sex hormone gene, dbSNP (RefSNP or rs), 46 associated alleles if available, and the type of genetic analysis. For the endometriosis studies, we also extracted the type of control population and the endometriosis stage according to the American Society for Reproductive Medicine (ASRM) classification or the American Fertility Society score (AFS) respectively, if available. 47
Results
In total, 92 studies fulfilled the inclusion criteria for this review (Figure 1). The characteristics of the included studies are listed in Table 1. For some of the candidate genes, a broad variety of SNPs were investigated. SNPs that were mentioned more than once in our literature search are grouped, see Table 1. Other SNPs that have been mentioned only once in our search are listed at the end. In the CYP17A1 gene, SHBG gene, and FSHB gene, the SNPs have only been assessed in one of both conditions. Since we are looking for overlapping SNPs between both conditions, these SNPs will not be described in our results but can be found in Table 1.
Summary table of the included studies according to sex-hormone genes.
ASRM: American Society for Reproductive Medicine; IHS: International Headache Society; ESR: estrogen receptor; PCR: polymerase chain reaction; PCR-SSCP: polymerase chain reaction-single-strand conformation polymorphism; PCR-RFLP: polymerase chain reaction-restriction fragment length polymorphism; HRM: high resolution melt; MALDI TOF MS: matrix-assisted laser desorption ionization-time-of-flight mass spectrometry; SNP: single nucleotide polymorphism; PGR: progesterone receptor; AR: androgen receptor; FSH: follicle stimulating hormone; FSHR: follicle stimulating hormone receptor; FSHB: FSH beta-subunit; SHBG: sex hormone-binding globulin; NRIP: nuclear receptor interacting protein; COMT: Catechol-O-methyl-transferase.
Not mentioned.
ESR1 gene
We found two overlapping polymorphisms in the ESR1 gene that could possibly play a role in the common genetic cause (Table 2).
Overview of overlapping polymorphisms associated with endometriosis and migraine.
SNP: single nucleotide polymorphism; ESR: estrogen receptor; PGR: progesterone receptor; AR: androgen receptor; FSHR: follicle stimulating hormone receptor; FSHB: FSH beta-subunit; NRIP: nuclear receptor interacting protein; COMT: Catechol-O-methyl-transferase.
Yes: all studies showed an association.
Probable: ⩾ 50% with association AND largest sum of sample sizes.
Possible: ⩽ 50% with association BUT largest sum of sample sizes.
No: ⩾ 75% without association AND/OR large sample size without association.
More studies needed: ⩾ 2 studies needed for each disorder to make a statement.
No data.
rs2234693, Pvull
Four studies in endometriosis patients found an association of this polymorphism in Caucasian and Asian women,51,48–50 while three studies did not.52–54 This, in spite the sample sizes of the seven studies were comparable. Only two of four studies investigating the base-pair change T > C in different ethnicities found an association.48,49,52,53 Three out of four studies found an association with migraine, all conducted in Asian populations, while no association with migraine was found in the study with a Caucasian population.55–58 Only one study specified the affected base-pair change T > C. 57 Two large studies with Caucasian migraineurs did not find a significant association.58,67
rs9340799, Xbal
Two of four studies found an association with endometriosis, and one of two studies found a significant association with migraine (Table 1). The associations with endometriosis were found in trials with Asian women.48,54 The association was not confirmed in Caucasian women in studies with similar sample sizes.50,53 All four endometriosis studies investigated the same base-pair change, A > G. For migraine, the studies for this SNP showed contradictory results55,56 (Table 1).
rs2228480
In an Asian study, no association with endometriosis was found. 61 For migraine, an association was significant in two of eight studies with the same alleles G > A55,62 (Table 1).
rs1801132
No association with endometriosis was found and a significant association with migraine was reported in three studies including large sample sizes of Caucasian women, two of them investigating the base-pair change G > C. The largest study investigated a C > G change. 64 Seven studies did not find an association for this SNP with migraine, two also studying the C > G change.57,58 The majority of these seven studies included Asian populations.55–57,121
ESR2 gene
One overlapping polymorphism that could possibly play a role in the common genetic cause of the two diseases was found in the ESR2 gene (Table 2).
rs4986938
Endometriosis and migraine were both associated with this polymorphism in four of eight studies. Two studies with Caucasian endometriosis patients found a significant association with the base-pair change G > A,73,74 while such an association was not confirmed in women of Asian descent.61,75 Moreover, two studies in Caucasian migraineurs, but not those including Asian populations, reported an association with this SNP.55,67,68,121
rs1256049
One of two studies found an association with endometriosis. 76 A total of three Asian studies investigated this polymorphism, but none found an association with endometriosis or migraine.55,56,61
PGR gene
PROGINS
Four studies showed an association with endometriosis. All four studies included Caucasian participants.78,79,81,93 The sample sizes of these four studies were relatively small compared with the studies without an association.49,82,84,85One association was found in a migraine trial with Caucasian women. 86 Two other migraine trials with Caucasian women and slightly smaller sample sizes did not find an association.66,67
rs1042838
No associations were found, for neither endometriosis nor migraine.40,57,61,83,87,89
AR gene
CAG repeat
Three studies showed an association with endometriosis. All three studies consisted of participants of Asian ethnicity.90–92 Three other studies did not find an association, one of which was a study investigating an association with migraine.80,86,94
FSHR gene
rs6166
One of five studies found an association with endometriosis in Turkish women, and one of three studies found an association with Caucasian migraineurs.68,95
NRIP1 gene
rs2229741
No association was found with endometriosis in a small study with Caucasian women. 102 One of three studies showed an association in Caucasian migraineurs. 67
COMT gene
rs4680
In six studies investigating an association with endometriosis, none were found.39,40,72,98,104,105 Two of eight studies found an association with migraine. One of the trials included more than 1000 participants. 107 On the contrary, another very large trial did not find an association. 112 In both groups, the ethnicity was considered Caucasian. The other association was found in a small Turkish trial focused on the G > A pair change. 106 Six studies with female migraineurs did not find an association.107–109,111–113 Four of these six investigated the same base pair change, G > A.
CYP1A1 gene
rs4646903 (MspI)
Two of eight studies showed an association with endometriosis. The case-control study with the smaller sample size investigated the T > C change. 115 Four Asian studies also investigated this base pair change, but found no association.72,105,116–119 No association with migraine was found. 110
rs1048943
No associations were found in endometriosis patients or migraine patients.72,110
CYP19A1 gene
rs10046
Within five studies, no significant associations with endometriosis were found in either Asian or Caucasian study populations.49,61,74,124,126 Two of five studies, both of them in Asian populations, found an association with migraine. In contrast, a large Chinese case-control study did not find an association. 55
rs4646
No association with endometriosis was found in a Caucasian study population. 74 In one Indian study, migraine was associated with this polymorphism. 56 However, two other large studies found no association.55,67
rs700519
None of the three studies examined found an association with endometriosis, and neither did one study with migraineurs.37,72,98,110
Discussion
The objectives of the present study were to provide an extensive review of sex hormone-related polymorphisms studied in endometriosis and migraine. Improved understanding of this comorbidity might facilitate early diagnosis and specific therapy. We found many probable overlapping SNPs in the candidate genes (Table 2), with mostly contradictory results, presumably due to some limitations of the included studies.
Both endometriosis and migraine are complex conditions with a variety of phenotypes. The heterogeneity of the control groups in the endometriosis studies constitutes a source of critique, as some women who have had a laparoscopy might not be representative of the normal population, and women in control groups without a laparoscopy might suffer from asymptomatic endometriosis with a probability of 2–11%. 5 Preferably, controls would be pain-free, fertile women in whom the absence of endometriosis is confirmed by surgery. 126 Such a control group will not only be difficult to establish in large study populations, but is also not acceptable for ethical reasons. A recent systematic review of polymorphisms and endometriosis confirmed the importance of being cautious regarding the criteria for selecting the control population. 33 To reduce heterogeneity, we excluded studies with self-reported endometriosis, recurrent endometriosis, female newborns, and men as controls. Arguably, including postmenopausal women in control groups is associated with the problem that these now asymptomatic women might have suffered from endometriosis in their reproductive years. Postmenopausal women could still have the genetic makeup predisposing them to endometriosis.
Another possible explanation is failed replication in subsequent studies due to the candidate-gene approach. Candidate-gene association studies have been widely used in the genetics of complex traits and diseases. This approach is based on the a priori selection of candidate genes with a hypothetical role in the pathogenesis of the disease and uses indirect genotyping methods. 127 Nowadays, these methods are mostly obsolete due to the rise of direct-sequencing technologies. 128 Indirect genotyping methods have a higher chance of yielding false positives compared to direct-sequencing technologies. In a simulation study, 968 of 1000 simulations (96.8%) produced at least one false positive. 129 Other drawbacks of candidate-gene studies are the small sample sizes and the lack of standardized genotyping methodologies. However, there is no universal genotyping method because choosing a suitable genotyping method for a certain variant depends on multiple factors, for example, the number of variants in the specified gene. Finally, problems with population stratification can occur in candidate-gene studies when cases and controls are poorly matched and, consequently, are responsible for significant associations. 130 We are aware that this approach has a limited ability to include all possible causative genes and polymorphisms. However, from a clinical point of view, we carefully decided to only include studies examining the candidate sex-hormone genes in question to better understand the role of these sex-hormones and to enable more targeted treatments for the comorbidity. Although subject to criticism, this approach still proves to be a robust tool for studying the genetic makeup, especially for diseases with complex traits. 131
We found three overlapping sex hormone-associated polymorphisms in the estrogen receptor genes (ESR1 and ESR2), in particular the SNPs rs2234693 (Pvull), rs9340799 (Xbal) and rs4986938 (AluI)48–51,54–57,67,68,73,74 (Table 2). Both Caucasian and Asian women with endometriosis showed an association between the disease and rs2234693 (Pvull), but for migraine the association was found only in Asian women.55–58 While rs9340799 (Xbal) was associated with endometriosis and migraine in Asian women, no association was found in Caucasian women.50,53 Both conditions were significantly associated with rs4986938 (AluI) in specifically Caucasian women. The differences between the findings in Asian and Caucasian populations demonstrate the importance of ethnicity when performing and interpreting genetic studies. The prevalence of endometriosis seems to differ among Asian and Caucasian populations. Asian women are significantly more likely to be diagnosed with endometriosis (OR 1.63, 95% CI 1.03–2.58). 132
Some studies did not specify the allelic change in the polymorphism. However, one SNP can have several different base-pair changes and allele frequencies can vary in different ethnicities. This information was reviewed in the current dbSNP database. 46
In the ESR1 gene, rs1801132 has been found to have an association with migraine in women, but no studies have investigated this polymorphism in women with endometriosis. Three repeat polymorphisms, TA repeat (ESR1 gene), CAG repeat (AR gene), and TTTA repeat (CYP19A1 gene) all have probable associations with endometriosis, but have not yet been investigated in migraineurs. We found overlapping polymorphisms for PROGINS (PGR gene) and rs6166 (FSHR gene), but more powered studies are required to understand if there is a significant association. Regarding the FSHB gene, rs11031006 is associated with endometriosis, but migraine studies are needed. General overlaps in the metabolizing enzymes have been assessed in our literature search for COMT, NRIP1, CYP1A1, CYP19A1, but no overlapping SNPs have been found and more data is needed (Table 2).
The strengths of this study were the strict inclusion and exclusion criteria to reduce selection bias. We included only case-control studies. To ensure the diagnosis, we only included studies with endometriosis diagnosed by laparoscopy and/or histology, and the majority was ASRM-classified. Migraineurs were diagnosed based on the criteria specified by the IHS. 2 A limitation of our study was the stratification into two ethnic groups, resulting in recruitment bias. Arguably, this could have been more specific–for example, stratifying for European, Hispanic, and Turkish populations. It is important to note that ethnicity and race have different definitions, although within medical literature these are often used interchangeably. 132 In female lifetime, the clinical pattern of migraine is linked to reproductive events with an increase around puberty, a peak during fertile age, and a decrease after menopause. 133 Age was considered, but studies for women in menopause or before menarche were not available and therefore could not be included. Other mechanisms might be involved in childhood migraine.
Migraines appear to occur more commonly in patients with endometriosis than in the general population. 134 The prevalence of migraine is significantly higher in women with endometriosis as compared to women without endometriosis.135–137 We hypothesize that endometriosis and migraine are comorbidities in a subset of women. A recent study confirmed this comorbidity and suggested a non-causal relationship between the two traits. 4 Nevertheless, mechanistic insights for both conditions are still lacking. In addition to differences on the hormone receptor level or hormone metabolism level, this also could be a shared problem in the immune response in women with the comorbidities.
Table 2 shows an overview of overlapping genes in both conditions, which were only found in estrogen receptor genes. It seems estrogen plays a central role in the genetic link underlying the comorbidity of endometriosis and migraine. Biologically, estrogens exert their effects via the estrogen receptors localized in epithelial, stromal, and vascular cells. Progestins antagonize estrogen actions in the reproductive tissues, brain, and nerve cells by reducing estrogen receptor expression. 138 Women with overlapping estrogen receptor polymorphisms could experience a higher improvement in symptoms with estrogen-suppression by continuous treatment with progestins. The majority of women in the reproductive years use hormonal contraception. However, combined hormonal contraception frequently has a negative impact on migraine. 139 Whereas treatment with progestin-only has a positive impact on both conditions.30,140–143 Therefore, patients with the comorbidity of migraine and endometriosis would profit if their attending physician could identify the co-occurrence to optimize hormonal treatment.
Conclusion
This literature review gives an overview of the shared sex hormone polymorphisms in women with migraine and endometriosis. Furthermore, we have identified SNPs potentially related to these conditions, which are relevant for future research. To confirm possible associations with other sex-hormone genes, larger studies are needed, in which ethnicity needs to be taken into account. We hypothesize that ESR1 and ESR2 may play a role in the genetic cause of endometriosis and migraine. For optimal treatment and patient care, we recommend actively exploring the comorbidity of migraine and endometriosis.
Supplemental Material
sj-docx-1-whe-10.1177_17455057221111315 – Supplemental material for Sex hormone-related polymorphisms in endometriosis and migraine: A narrative review
Supplemental material, sj-docx-1-whe-10.1177_17455057221111315 for Sex hormone-related polymorphisms in endometriosis and migraine: A narrative review by Joy-Fleur van der Vaart and Gabriele Susanne Merki-Feld in Women’s Health
Footnotes
Author contribution(s)
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Ethical approval
According to the cantonal Swiss Ethic Committee of Zürich http://www.kek.zh.ch/ and Swiss Association of Research Ethics Committees
ethics approval is not applicable for literature reviews.
Supplemental material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
