Advertisement

The influence of ethnicity on outcomes of ovulation induction with clomiphene citrate in women with PCOS

  • C. Meun
    Correspondence
    Corresponding Author: Cindy Meun, Department of Reproductive Endocrinology and Infertility – Floor Na16, Erasmus University Medical Centre, PO Box 2040, 3000 CA Rotterdam, the Netherlands
    Affiliations
    Division of Reproductive Endocrinology and Infertility, department of Obstetrics and Gynaecology, Erasmus University Medical Center, Rotterdam, The Netherlands
    Search for articles by this author
  • E. Oostingh
    Affiliations
    Division of Reproductive Endocrinology and Infertility, department of Obstetrics and Gynaecology, Erasmus University Medical Center, Rotterdam, The Netherlands
    Search for articles by this author
  • Y.V. Louwers
    Affiliations
    Division of Reproductive Endocrinology and Infertility, department of Obstetrics and Gynaecology, Erasmus University Medical Center, Rotterdam, The Netherlands
    Search for articles by this author
  • J.S.E. Laven
    Affiliations
    Division of Reproductive Endocrinology and Infertility, department of Obstetrics and Gynaecology, Erasmus University Medical Center, Rotterdam, The Netherlands
    Search for articles by this author
Open AccessPublished:December 26, 2021DOI:https://doi.org/10.1016/j.rbmo.2021.12.017

      Structured abstract

      Research question

      To study the influence of ethnicity on the outcome of ovulation induction with clomiphene citrate (CC) in women with polycystic ovary syndrome (PCOS).

      Design

      A retrospective cohort study. In total, 420 women diagnosed with PCOS from Northern European, Mediterranean, African, South-East Asian and South-American descent, who started ovulation induction treatment with CC, were included. All women were treated with CC according to standardized treatment regime. Minimal effective dose of CC, and the prevalence of clomiphene resistance (CRA) were assessed, and we predicted the chance to become ovulatory.

      Results

      We observed differences in BMI (p=0.002), waist circumference (p=0.036), LH, insulin, and androgen serum levels (all p<0.001) in PCOS women of different ethnicity. Compared to women of Northern European descent, the minimal effective dose of CC in women of other ethnic groups was not different (p>0.2). The prevalence of CRA (p=0.574) was similar in all ethnic groups. We predicted a similar chance of ovulation (p=0.5) in the different ethnic groups.

      Conclusions

      This is the first study aiming to link ethnicity to ovulation induction outcome in PCOS. Although PCOS women of different ethnicity exhibit variation in phenotype expression, it does not appear that there are differences in the prevalence of clomiphene resistant anovulation as well as the minimal effective dose of CC. Furthermore, a prediction model revealed no significant differences in the predicted chance to ovulate. A larger cohort is needed to validate these findings.

      Keywords

      Introduction

      Normogonadotropic anovulation is the most common cause of subfertility in women of reproductive age. The majority of women with normogonadotropic anovulation are diagnosed with polycystic ovary syndrome (PCOS) (Balen et al. 2016). PCOS is most often diagnosed according the Rotterdam criteria, when either two or three of the key features (cycle irregularities, polycystic ovarian morphology, and elevated androgen levels) are present. This results in four different PCOS phenotypes (Rotterdam 2004). Using this definition, PCOS has a reported prevalence up to 13% (Bozdag et al. 2016). Multiple strategies are available for the treatment of anovulation, ranging from lifestyle modification to assisted reproductive techniques and laparoscopic ovarian drilling (Perales-Puchalt and Legro 2013, Teede et al. 2018a). The recently published international guideline on PCOS suggests letrozole as a first line pharmacological treatment for anovulation (Teede et al. 2018b). In many countries, such as India, Egypt, parts of the USA, and the Netherlands, Letrozole is not registered as an ovulation induction (OI) drug. Therefore, if used for OI, it is used off label which implies that many doctors will use clomiphene citrate (CC) (Abu Hashim 2016, Teede et al. 2018b). The chance to regain ovulatory cycles with CC has been estimated around 60-80%, and 30-50% will become pregnant (Imani et al. 2002, Lord et al. 2003, Ellakwa et al. 2016, Teede et al. 2018b). Currently, there are no ethnicity based treatment algorithms for the use of clomid in ovulation induction in women with PCOS.
      Response to drugs is difficult to predict and variability in the response might be caused by the pathophysiology and severity of the disease (Evans and Relling 1999). Genetic variants involved in the pathogenesis of PCOS have been associated with a higher chance of clomiphene resistant anovulation (CRA), and a lower chance to achieve an ongoing pregnancy (Valkenburg et al. 2015, Zilaitiene et al. 2018). Besides the genetic make-up of the patient, age, as well as PCOS phenotype characteristics influence the response to OI treatment with CC.
      Phenotype expression and the severity of PCOS are influenced by ethnicity. South-Asian women with PCOS tend to experience oligomenorrhea at a much younger age, and BMI and androgen levels are generally low in these women (Wijeyaratne et al. 2002, Zhao and Qiao 2013, Kim and Choi 2019). At the same time the prevalence of insulin resistance and metabolic syndrome is high in these women (Zhao and Qiao 2013, Engmann et al. 2017). Women with PCOS originating from sub-Saharan Africa as well as Mediterranean areas tend to have a higher BMI, together a higher prevalence of obesity and elevated androgen levels (Valkenburg et al. 2011, Zhao and Qiao 2013). In contrast, Hindustani women with PCOS generally present with the highest free androgen index (FAI) (Valkenburg et al. 2011, Engmann et al. 2017).
      Studies have shown that a high body mass index (BMI), as well as high androgen and serum insulin levels, the type of ovulatory disorder (i.e. oligomenorrhea versus amenorrhea), and a larger ovarian volume are associated with poor response to ovulation induction treatment with CC (Imani et al. 1998, Mulders et al. 2003, Rausch et al. 2009, Ellakwa et al. 2016). In addition, high AMH levels have been associated with a reduced response to clomiphene citrate (Ellakwa et al. 2016, Vagios et al. 2021).
      As stated ethnicity is related to differences in PCOS phenotype, which could also influence the response to CC. However, response to ovulation induction therapy has rarely been explored in relation to ethnicity. Further, available algorithms to predict to response to CC do not take ethnicity into consideration. The first aim of the current study, was to assess the influence of ethnicity on OI with CC, by assessing the minimal effective dose, and the chance to become ovulatory. The second aim of this study was to predict the chance to ovulate after OI treatment in women with PCOS of different ethnicity.
      To study these aims we have used an algorithm designed in our university hospital, which can be used to predict the chance to ovulate following CC therapy ((Imani et al. 1998). This model includes the free androgen index (FAI), BMI and the type of ovulatory dysfunction and was developed in a cohort of Northern-European women.

      Materials and Methods

       Study population

      Patients diagnosed with normogonadotropic anovulation and PCOS, who visited the outpatient clinic of a university in Rotterdam, the Netherlands, between January 1st 2005, and January 1st 2016, with a wish to conceive, and did not undergo previous fertility treatment, were eligible for inclusion. PCOS was diagnosed according to the Rotterdam criteria, when two or three of the key features; ovulatory dysfunction, hyperandrogenism and polycystic ovarian morphology, were present (Fauser et al. 2012). A standardized screening was performed by trained physicians. The screening protocol has been described in detail elsewhere (Valkenburg et al. 2008). Briefly, information on general, medical and obstetric history and previous fertility treatment was obtained through a questionnaire. Blood pressure, waist and hip circumference, height and weight were measured. An ultrasound probe of <8 MHz, was used to assess the presence of polycystic ovarian morphology, which was defined as ovarian volume >10ml or >12 follicles in one or both ovaries in the absence of a dominant follicle or cyst. Ovulatory dysfunction was defined as menstrual cycle length of <21 days or >35 days (oligomenorrhea) or the absence of a menstrual cycle ≥6 months (amenorrhea). Women with other reasons for infertility, including thyroid disease, were excluded. From 2009 onwards, in the Netherlands, the first line of treatment of anovulation in overweight (BMI >25) or obese (BMI >30) women with PCOS has become lifestyle modification. These women are enrolled in a year lifestyle modification program. Women enrolled in this program were not included in the current study. We excluded patients who did not start ovulation induction treatment with CC. Reasons for not starting OI treatment with CC were having received previous ovulation induction treatment elsewhere, the occurrence of a spontaneous pregnancy, switch to another form of infertility treatment than CC (intra-uterine insemination, IVF of ICSI), getting a second opinion and treatment elsewhere. Finally, some couples choose to not start with infertility treatment after all. Since this study neither implied that patients would receive a particular treatment, nor imposed on their behavior as described in the Medical Research Involving Human Subjects Act (WMO), the local institutional review board (IRB) of Erasmus Medical Center Rotterdam has officially stated that the WMO does not apply (MEC-2020-0534).

       Ethnicity

      Self-reported ethnicity and country of birth of the patient and both parents were registered via questionnaires. Based on self-reported ethnicity, eligible patients were classified as either Northern European, Mediterranean, African, South-East Asian or South-American. Due to the small sample size, patients with a mixed ethnicity were excluded from the analyses.

       Endocrine assessment

      Blood was drawn on a random cycle day after an overnight fast. Luteinizing hormone (LH) and follicle stimulating hormone (FSH), sex-hormone binding globulin (SHBG) and insulin levels were measured with immunoluminometric assay (Immulite® platform, Siemens DPC Los Angeles, CA, USA.). Testosterone, androstenedione (Adione) and dehydroepiandrosterone (DHEA) were measured with (LC-MS/MS). Glucose levels were measured with Cobas 8000 (Roche Diagnostics Almere, Netherlands). Estradiol was measured with Ria assay (Siemens DPC Los Angeles, CA, USA.). Gen II Beckman Coulter, (Beckman Coulter, Inc., Webster, TX) was used to determine serum anti-Mullerian hormone (AMH) serum levels. The free androgen index (FAI) was calculated as 100 x Testosterone/SHBG.

       Treatment protocol

      According to the standard treatment regimen, patients received an initial CC dose of 50 mg/d on cycle day 3-7 after spontaneous or induced withdrawal bleeding. In case of the absence of ovulation, the dosage was increased to 100mg/d or 150 mg/d. The minimal effective dose is the dosage of CC, after which a regular menstrual cycle was restored. Clomiphene resistant anovulation (CRA) was defined as the failure to ovulate during treatment with CC. The failure to conceive after ovulation has been restored for a minimum of 6 cycles is referred to as clomiphene citrate failure (CCF).

       Statistical analyses

      Statistical analyses were performed using IBM SPSS version 24 (IBM corp., Armonk, NY, USA). Baseline characteristics were displayed as medians and interquartile ranges (IQR) or numbers and percentages. Mann-Whitney-U was used to compare continuous variables between groups. Proportions were tested with the Chi-square test. An ordinal regression was performed with the polytomous universal model (PLUM) procedure, to assess the minimal effective dose of CC across different ethnicities. A p-value of <0.05 denoted a statistical significant difference. The chance to ovulate after the admission of CC was calculated with a previously designed prediction model (Imani et al. 2002). This model predicts the chance to ovulate following CC therapy, based on the height of the FAI, BMI and the presence of oligomenorrhea or amenorrhea. First the linear predictor (LP) is calculated with the formula LP = -4.1768+0.0626*BMI+0.1397*FAI+ 1.2047*(1 (in case of amenorrhea)). Next the probability for CRA is calculated with; Probability (PR) CRA = 1/(1+exp(-LP)). Finally, this results in the chance to ovulate (1-Pr(CRA)) (Imani et al. 2002). We calculated the chance of ovulation for each individual and then compared the group medians.

      Results

      In total, 1259 diagnosed with women were eligible for inclusion in the current study. We excluded 762 women (60.5%) who did not start with ovulation induction. Reasons for not starting treatment were the occurrence of a spontaneous pregnancy (N=130 (16.9%)), being overweight or obese (N=252 (32.8%)), previous treatment (N=63 (8.2%)) or because they had started with ovulation induction with recombinant FSH or IVF treatment as a first line treatment (N=185 (24.0%). Women who decided not to start their ART treatment (N=132 (17.2%) were also excluded. This resulted in 497 women who began with treatment with CC. Due to the small sample size (N=8) women with a mixed ethnicity were also excluded. Finally, N = 69 (14.1%) women who started treatment with CC were lost to follow up and therefore excluded, resulting in 420 women used in the analyses.
      We stratified women according to self-reported descent. We assessed N=278 Northern-European, N=60 Mediterranean, N=20 African, N=44 South-East Asian, and N=18 South-American women with PCOS. Women classified as South-American mainly originate from the Netherlands Antilles (Aruba and Curacao). The baseline characteristics of women with PCOS of different ethnicity are presented in table 1.
      Table 1Characteristics of the study population stratified by ethnicity
      EthnicityNorthern-EuropeanMediterraneanAfricanSouth-East AsianSouth-AmericanP-value
      No. of cases278 (66.2%)60 (14.3%)20 (4.8%)44 (10.5%)18 (4.3%)-
      Age29.6 (27.0-32.3)26.4 (23.6-28.9)26.1 (22.8-29.1)29.4 (26.6-32.4)29.5 (26.5-33.3)<0.001
      BMI (kg/m2)23.2 (21.0-27.6)25.6 (23.6-29.6)23.7 (21.1-31.4)24.2 (21.9-28.3)25.1 (23.5-34.3)0.008
      Waist (cm)78.0 (71.0-87.0)84.0 (74.0-90.0)75.5 (69.3-87.5)78.5 (71.8-87.3)86.5 (75.5-100.00)0.036
      BP systolic (mmHg)112.0 (110.0-120.0)110.0 (105.0-120.0)115.0 (105.0-120.0)110.0 (110.0-120.0)115.0 (107.5-125.5)0.725
      BP diastolic (mmHg)72.0 (70.0-80.0)70.0 (68.0-80.0)75.0 (65.0-82.0)70.0 (70.0-80.0)80.0 (72.5-83.5)0.348
      LH (IU/L)8.2 (5.2-12.5)12.1 (7.5-16.7)12.2 (8.1-18.1)9.2 (5.9-14.0)8.5 (4.6-12.2)0.002
      FSH (IU/L)5.9 (4.5-7.3)6.0 (5.0-7.2)5.9 (3.9-6.8)6.4 (4.6-7.0)5.5 (4.4-6.8)0.728
      Testosterone (nmol/L)1.6 (1.1-2.2)1.7 (1.2-2.4)2.8 (1.8-3.6)1.7 (1.3-2.1)1.6 (1.3-2.2)0.002
      SHBG (nmol/L)52.4 (35.9-69.5)34.1 (21.5-48.9)34.5 (24.6-52.1)30.4 (20.8-44.0)42.0 (28.8-59.3)<0.001
      FAI3.1 (1.8-5.1)4.7 (3.1-9.5)6.6 (4.6-11.4)5.9 (3.4-8.8)4.7 (2.6-6.5)<0.001
      FG score1.0 (0.0-3.0)4.0 (2.0-7.0)4.0 (0.0-5.5)3.0 (1.5-8.5)3.0 (1.5-4.5)<0.001
      Estradiol (pmol/L)224.0 (162.5-359.0)253.0 (154.30-361.8)292.5 (229.8-395.5)187.5 (133.3-282.3)252.0 (196.5-371.0)0.046
      Adione (nmol/L)9.2 (6.6-12.3)10.9 (8.1-14.6)15.1 (10.9-24.2)10.5 (7.9-13.9)8.9 (6.3-12.2)<0.001
      DHEA (nmol/L)31.3 (21.4-49.3)44.5 (28.2-66.0)51.4 (38.9-77.4)33.1 (22.9-50.5)32.0 (23.9-48.3)0.002
      Glucose (mmol/L)4.6 (4.3-4.9)4.7 (4.2-4.9)4.9 (4.3-5.2)4.9 (4.3-5.2)4.7 (4.2-4.8)0.172
      Insulin (pmol/L)40.0 (23.8-68.3)54.0 (31.0-83.0)46.0 (23.0-91.8)82.0 (27.5-112.7)53.0 (26.5-99.3)0.002
      AMH ((µg/L)8.1 (5.1-13.6)8.4 (6.3-11.9)11.3 (6.2-16.5)8.3 (5.1-14.5)8.7 (5.0-16.4)0.859
      Oligomenorrhea202 (72.7%)48 (80.0%)16 (80.0%)35 (79.5%)18 (100.0%)0.079
      Amenorrhea76 (17.3%)12 (20.0%)4 (20.0%)9 (20.5%)0 (0%)0.073
      PCOM265 (95.3%)55 (91.7%)20 (100%)41 (93.2%)16 (88.9%)0.596
      CRA43 (15.5%)10 (16.7%)5 (25%)8 (18.2%)1 (5.6%)0.574
      Values are displayed as medians and interquartile ranges or numbers and percentages (%). Comparison of continuous variables between groups was done with the Kruskall-Wallis test, a P-value of <0.05 was considered statistically significant. Chi-square test was used to test proportions. Boldface indicates a significant association at P <0.05. Abbreviations: body mass index (BMI), blood pressure (BP), luteinizing hormone (LH), follicle stimulating hormone (FSH), sex-hormone binding globulin (SHBG), free androgen index (FAI), Ferrimann Galwey (FG) score, androstenedione (Adione), dehydroepiandrosterone (DHEA),anti-mullerian hormone (AMH), clomiphene resistant anovulation (CRA)
      We observed differences in age (p <0.001), BMI (p=0.002), and waist circumference (p=0.036), with the highest BMI and waist circumference found in women of Mediterranean (median BMI 25.6 kg/m2, waist 84.0cm) and South-American descent (BMI median BMI 25.1 kg/m2, waist 86.5). Lowest median BMI was seen in women of Northern-European descent (23.2 kg/m2) Highest median LH levels were found in women of Mediterranean (12 IU/L) and African (12.2 IU/L) descent, and lowest in Northern-European women (8.2IU/L), p=0.002. Furthermore, we observed differences in androgen levels (p<0.001), with highest testosterone (2.8 nmol/L) and FAI (median 6.6) in African women with PCOS, and the lowest in women of Northern-European . Insulin levels differed between women of different ethnicity (P<0.001), with the highest levels measured in South-East Asian women (82.0 pmol/L), and lowest in Northern-European women). Systolic and diastolic blood pressure, as well as the prevalence of oligomenorrhea and amenorrhea, PCOM, and FSH serum levels were similar in women with PCOS across all ethnic groups. Finally, we observed no significant differences in AMH serum levels in women of Northern-European, Mediterranean, African, South-East Asian and South-American descent.
      The prevalence of CRA in women of self-reported Northern-European, Mediterranean, African, South-East Asian, or South-American descent was not significantly different (p=0.574). We performed an ordinal regression analysis to assess the minimal effective dose of CC in women with PCOS of different ethnicity. For this analysis, we used the largest group, women of self-reported Northern-European descent as the reference group, and tested whether women of other ethnicities needed a higher dose of CC to become ovulatory (Figure 1). We found no significant differences in the minimal effective dose of CC in Mediterranean (0.35 (-0.12:0.82), β (95% CI), African (0.33 (-0.43:1.09)), South-East Asian (0.19 (-0.35:0.73)), or South-American (0.08 (-0.91:0.74)) women.
      Figure 1
      Figure 1Minimal effective dose of CC in women with PCOS of different ethnicity.
      We used ordinal regression to assess differences in the minimal effective dose CC in women with PCOS of different ethnicity. Results are expressed as beta's and corresponding 95% confidence intervals, with Northern-European women used as the reference group.
      Finally, we used a previously designed algorithm predict the chance to ovulate after treatment with CC (Imani et al. 1998). We observed no significant differences in the predicted chance of ovulation in the different ethnic groups (p=0.504). Median predicted chance to ovulate ranged from 0.84 to 0.88 women with PCOS across all ethnic groups (Figure 2).
      Figure 2
      Figure 2The predicted chance to ovulate after the administration of CC in women with PCOS of different ethnicity.
      Values are expressed as medians and interquartile ranges. Comparison of groups was done with the Kruskall-Wallis test, a p-value of <0.05 was considered statistically significant.
      We performed a sensitivity analysis to compare women of self-reported Northern-European and South-east Asian ethnicity. We compared the prevalence of CRA, the predicted chance to ovulate after CC and the minimal effective dose of CC. In these analyses we observed no significant differences (data not shown).

      Discussion

      To our knowledge, this is the first study assessing the influence of ethnicity on ovulation induction treatment with clomiphene citrate in women with PCOS. The main finding of the current study is that although we observed differences in the phenotype of women with PCOS of different ethnicity, we found no significant differences in the prevalence of clomiphene resistant anovulation and the minimal effective dose of CC. Furthermore, we used a previously designed algorithm to predict the chance to ovulate after treatment with CC. This model revealed no significant differences in the predicted chance to ovulate, suggesting that the differences in phenotype might balance each other out and result in a similar response to treatment and this model could be used in clinical practice.
      We assessed the phenotype of women with self-reported Northern-European, Mediterranean, African, South-East Asian, and South-American ethnicity. In line with existing literature, we report significant differences in BMI and waist circumference, and in insulin, LH, and androgen levels in women with PCOS of different ethnicity (Valkenburg et al. 2011, Zhao and Qiao 2013, Engmann et al. 2017). Similarly, no significant differences were observed in systolic and diastolic blood pressure, and the prevalence of oligomenorrhea or amenorrhea, PCOM, and in FSH serum levels. Despite these observed ethnicity based phenotypic differences in women with PCOS, no significant differences in the prevalence of CRA were observed, nor did we observe differences in the minimal effective dose of CC in women of different ethnicity. One potential explanation for that could be the fact that AMH levels were similar for all ethnic groups. High AMH levels have been associated with a reduced response to treatment (Ellakwa et al. 2016). The absence of differences in AMH serum levels in women of different ethnicity one the reasons why we did not observe differences in CC response. Another explanation might be that an increase in one of the predicting factors, could outbalance the impact of other predicting factors resulting in a similar outcome of OI treatment with CC for all ethnicities. Alternatively, the weight of the predictors might be different in women of different ethnicity. It could also be that the current study lacked the power to detect significant differences. We did observe some differences in the prevalence of CRA, but this did not reach statistical significance. This could be due to the relatively small sizes of some of the ethnic groups. We need to be cautious when interpreting these results, as it is one of the first addressing this topic. Validation of our findings in a larger cohort are necessary. However, based on the current study, we conclude that although ethnic differences exist in the phenotype of women with PCOS, this does not impact the response to CC. Hence, the treatment of women with PCOS of different ethnic descent according to the same standardized protocol seems appropriate.
      We used a previously designed algorithm, which uses BMI, the free androgen index and the type of cycle irregularity, to predict the chance to ovulate after treatment with CC in women with PCOS of different ethnicity (Imani et al. 2002). We predicted a similar chance to become ovulatory in the different ethnic groups. This is in line with the actual observed prevalence of CRA, which was similar amongst the different ethnic groups. Most algorithms to predict ovulation induction outcome are developed in women of Northern European descent. The weight of the predictors might be different in women of different ethnicity. Possibly the impact of for instance hyperandrogenism on fertility is bigger in some ethnic groups than in others. If is this the case for multiple phenotypic traits, it could be that the combination of traits with a bigger or lesser impact, result in a similar response to OI with CC. However, it is also possible that we did not detect differences either due to the design of the study or other physiologic influences. By not including overweight and obese women who were enrolled in the lifestyle program in our analyses, one could hypothesize that this is why we found no significant difference in the predicted chance to ovulate. However, because the multiplier in BMI is so small, compared to that of the FAI type of ovulatory dysfunction, we expect its influence to be minor.
      The current study is one of the largest studies, assessing ethnic differences in women with PCOS and the first study aiming to link ethnicity to ovulation induction treatment outcome. We were able to assess many phenotypical features and had detailed information regarding the follow up of treated women. A limitation of this study is the difference in size of our ethnic groups. We had to categorize our study population into five ethnic groups, to be able to use them for analysis. Some of the ethnic groups, are still relatively small, and we may not have detected small associations. We performed a sensitivity analysis to compare the outcome of OI with CC in women of Northern-European and South-East Asian ethnicity. This yielded no significant differences. When comparing additional ethnicities, it does not appear that there are significant differences. However a larger cohort is needed to validate these findings. Furthermore, we assessed a relatively lean group of women with PCOS, this is because in the Netherlands the first step in the treatment of ovulation is lifestyle modification and weight loss. Before starting pharmacological treatment, women with PCOS who are overweight or obese must attempt to achieve sustained weight loss on their own or by entering a lifestyle program, or sometimes are referred for bariatric surgery. With this, a proportion of these women will regain ovulatory cycles and become pregnant and further treatment might not be necessary. This may have introduced a selection bias. However, we feel that women enrolled in the lifestyle modification program, and their response to OI treatment after completing this program, need to be analyzed separately.
      The definition of ethnicity is complex including non-genetic elements such as diet, language, and other cultural elements as well as elements that are expected to have a genetic component such as biogeographic origin (i.e., genetic ancestry) or the genetic background in general, which may or may not affect the variability of the PCOS phenotype including its characteristics. In medical studies, including those on PCOS, ethnicity is usually assessed by self-reported data. We are aware that self-reported ethnicity can be biased and does not necessarily capture all existing population substructures. We partly address this bias by not only asking women for their ethnicity, but also include country of birth of the patient as well as her parents. We previously showed that the association between self-reported ethnicity and genetic ancestry was moderate ((Louwers et al. 2014). It would be very interesting to include genetic ancestry data in the prediction of ovulation induction outcome.
      To what extent these results are generable to overweight and obese women, needs to be examined. Validation of the current study in a larger multi-ethnic cohort are necessary to validate our findings and to provide more insight on the outcome of treatment with CC.

      Conclusions

      In conclusion, although women with PCOS of different ethnicity exhibit variation in the phenotypic expression of PCOS, there seem to be no differences in the prevalence of clomiphene resistant anovulation or the minimal effective dose of CC. Furthermore, a prediction model revealed no significant differences in the predicted chance to ovulate. A larger cohort is needed to provide more insight on the outcome of OI treatment with CC in women with PCOS of different ethnicity.

      Acknowledgments

      We would like to thank Professor Eijkemans for providing us with the opportunity to use the prediction model for ovulation induction outcome in the current study.

      References

      Abu Hashim, H., 2016. Twenty years of ovulation induction with metformin for pcos; what is the best available evidence? Reprod Biomed Online. 32, 44-53
      Balen, A.H., Morley, L.C., Misso, M., Franks, S., Legro, R.S., Wijeyaratne, C.N., Stener-Victorin, E., Fauser, B.C., Norman, R.J., Teede, H., 2016. The management of anovulatory infertility in women with polycystic ovary syndrome: An analysis of the evidence to support the development of global who guidance. Hum Reprod Update. 22, 687-708
      Bozdag, G., Mumusoglu, S., Zengin, D., Karabulut, E., Yildiz, B.O., 2016. The prevalence and phenotypic features of polycystic ovary syndrome: A systematic review and meta-analysis. Hum Reprod. 31, 2841-2855
      Ellakwa, H.E., H.E., Sanad, Z.F., Hamza H.A., Emara, M.A., Elsayed, M.A., 2016. Predictors of patient responses to ovulation induction with clomiphene citrate in patients with polycystic ovary syndrome experiencing infertility. Int J Gynaecol Obstet. 133, 59-63
      Engmann, L., Jin, S, Sun, F., Legro, R. S., Polotsky, A. J., Hansen, K. R., Coutifaris, C.,Diamond, M. P., Eisenberg, E., Zhang, H., Santoro, N., Reproductive Medicine, Network, 2017. Racial and ethnic differences in the polycystic ovary syndrome metabolic phenotype. Am J Obstet Gynecol. 216, 493 e1-493 e13
      Evans, W.E., Relling, M.V., 1999. Pharmacogenomics: Translating functional genomics into rational therapeutics. Science. 286, 487-91
      Fauser, B.C., Tarlatzis, B. C., Rebar, R. W., Legro, R. S., Balen, A. H., Lobo, R., Carmina, E., Chang, J., Yildiz,
      B. O., Laven, J. S., Boivin, J., Petraglia, F., Wijeyeratne, C. N., Norman, R. J., Dunaif, A., Franks, S., Wild, R. A., Dumesic, D., Barnhart, K., 2012. Consensus on women's health aspects of polycystic ovary syndrome (pcos): The amsterdam eshre/asrm-sponsored 3rd pcos consensus workshop group. Fertil Steril. 97, 28-38 e25
      Imani, B., Eijkemans, M. J., te Velde, E. R., Habbema, J. D., Fauser, B. C., 1998. Predictors of patients remaining anovulatory during clomiphene citrate induction of ovulation in normogonadotropic oligoamenorrheic infertility. J Clin Endocrinol Metab. 83, 2361-5
      Imani, B., Eijkemans, M.J.C., te Velde, E.R., Habbema, J.D.F., Fauser, B.C.J.M., 2002. A nomogram to predict the probability of live birth after clomiphene citrate induction of ovulation in normogonadotropic oligoamenorrheic infertility. Fertil Steril. 77, 91-7
      Kim, J.J., Choi, Y.M., 2019. Phenotype and genotype of polycystic ovary syndrome in asia: Ethnic differences. J Obstet Gynaecol Res.
      Lord, J.M., Flight, I.H., Norman, R.J., 2003. Insulin-sensitising drugs (metformin, troglitazone, rosiglitazone, pioglitazone, d-chiro-inositol) for polycystic ovary syndrome. Cochrane Database Syst Rev. CD003053
      Louwers, Y.V., Lao, O., Fauser, B.C., Kayser, M., Laven, J.S., 2014. The impact of self-reported ethnicity versus genetic ancestry on phenotypic characteristics of polycystic ovary syndrome (pcos). J Clin Endocrinol Metab. 99, E2107-16
      Mulders, A.G., A.G., Laven, J.S.E., Eijkemans, M.J.C., Hughes, E.G., Fauser, B.C.J.M., 2003. Patient predictors for outcome of gonadotrophin ovulation induction in women with normogonadotrophic anovulatory infertility: A meta-analysis. Hum Reprod Update. 9, 429-49
      Perales-Puchalt, A., Legro, R.S., 2013. Ovulation induction in women with polycystic ovary syndrome. Steroids. 78, 767-72
      Rausch, M.E., Legro, R. S., Barnhart, H. X., Schlaff, W. D., Carr, B. R., Diamond, M. P., Carson, S. A., Steinkampf, M. P., McGovern, P. G., Cataldo, N. A., Gosman, G. G., Nestler, J. E., Giudice, L. C., Leppert, P. C., Myers, E. R., Coutifaris, C., Reproductive Medicine, Network, 2009. Predictors of pregnancy in women with polycystic ovary syndrome. J Clin Endocrinol Metab. 94, 3458-66
      Rotterdam, Eshre-Asrm-Sponsored Pcos concensus workshop group., 2004. Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome (pcos). Hum Reprod. 19, 41-7
      Teede, H.J., Misso, M. L., Costello, M. F., Dokras, A., Laven, J., Moran, L., Piltonen, T., Norman, R. J., International, Pcos Network, 2018a. Recommendations from the international evidence-based guideline for the assessment and management of polycystic ovary syndrome. Hum Reprod. 33, 1602-1618
      Teede, H.J., Misso, M. L., Costello, M. F., Dokras, A., Laven, J., Moran, L., Piltonen, T., Norman, R. J., International, Pcos Network, 2018b. Recommendations from the international evidence-based guideline for the assessment and management of polycystic ovary syndrome. Fertil Steril. 110, 364-379
      Vagios, S., Sacha, C. R., Hammer, K. C., Dimitriadis, I., James, K. E., Bormann, C. L., Souter, I., 2021. Response to ovulation induction treatments in women with polycystic ovary syndrome as a function of serum anti-mullerian hormone levels. J Assist Reprod Genet. 38, 1827-1833
      Valkenburg, O., Lao, O., Schipper, I., Louwers, Y., Uitterlinden, A. G., Kayser, M., Laven, J. S., 2011. Genetic ancestry affects the phenotype of normogonadotropic anovulatory (whoii) subfertility. J Clin Endocrinol Metab. 96, E1181-7
      Valkenburg, O., Steegers-Theunissen, R. P., Smedts, H. P., Dallinga-Thie, G. M., Fauser, B. C., Westerveld, E. H., Laven, J. S., 2008. A more atherogenic serum lipoprotein profile is present in women with polycystic ovary syndrome: A case-control study. J Clin Endocrinol Metab. 93, 470-6
      Valkenburg, O., van Santbrink, E. J., Konig, T. E., Themmen, A. P., Uitterlinden, A. G., Fauser, B. C., Lambalk, C. B., Laven, J. S., 2015. Follicle-stimulating hormone receptor polymorphism affects the outcome of ovulation induction in normogonadotropic (world health organization class 2) anovulatory subfertility. Fertil Steril. 103, 1081-1088 e3
      Wijeyaratne, C.N., Balen, A.H., Barth, J.H., Belchetz, P.E., 2002. Clinical manifestations and insulin resistance (ir) in polycystic ovary syndrome (pcos) among south asians and caucasians: Is there a difference? Clin Endocrinol (Oxf). 57, 343-50
      Zhao, Y., Qiao, J., 2013. Ethnic differences in the phenotypic expression of polycystic ovary syndrome. Steroids. 78, 755-60
      Zilaitiene, B., Dirzauskas, M., Verkauskiene, R., Ostrauskas, R., Gromoll, J., Nieschlag, E., 2018. The impact of fsh receptor polymorphism on time-to-pregnancy: A cross-sectional single-centre study. BMC Pregnancy Childbirth. 18, 272

      Biography

      Cindy Meun completed her medical degree in 2013 and is a PhD candidate at Erasmus Medical Center in Rotterdam, the Netherlands. Her research focusses on genetics and cardiovascular disease risk in women with polycystic ovary syndrome.
      Key message
      Although women with PCOS of different ethnicity exhibit variation in the phenotypic expression of PCOS, there seem to be no differences in the prevalence of clomiphene resistant anovulation or the minimal effective dose of clomiphene citrate.