Endometriosis risk is associated with shorter anogenital distance by meta-analysis

23

Anogenital distance is a well-validated marker of prenatal testosterone, with 24 shorter distances indicating lower levels during early gestation. A suite of studies has linked 25 anogenital distance with risk of endometriosis, but the findings are variable, leading to 26 uncertainty in interpretation. The relationship of anogenital distance with endometriosis is 27 especially important because lower testosterone has been associated with endometriosis in 28 recent Mendelian Randomization studies, which implies causality in the association, with direct 29 implications for future research and treatment. 30

A systematic review and meta-analysis was conducted on the association of 31 endometriosis with anogenital distance. Three databases were queried in the identification 32 phase, and a random-effects meta-analysis was applied to the data in studies that met the 33 inclusion criteria. 34

Shorter anogenital distance AF, measured from the anus to the posterior fourchette, 35 was significantly associated with higher risk of endometriosis in the meta-analysis. By contrast, 36 there was no such association for anogenital distance AC, measured from the anus to the 37 clitoral surface. Both analyses demonstrated significant heterogeneity across studies. Too few 38 studies were available for robust investigation of publication bias. 39

The association of short anogenital distance with endometriosis risk provides 40 support for the hypothesis that endometriosis represents, in part, a disorder mediated by 41 relatively low testosterone levels in early prenatal development. This conclusions has notable 42 implications for understanding the causes and treatment of endometriosis. 43

endometriosis, prenatal testosterone, anogenital distance, meta-analysis 44 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.05.24300901doi: medRxiv preprint 3

45 Endometriosis, which has been reported in about 5-10% of reproductive-aged women, is 46 characterized by endometrial tissue establishing outside of the uterus, most often in the 47 ovaries, peritoneum, or rectovaginal region [1]. This 'ectopic' endometrial tissue undergoes 48 menstrual cycle changes similar to those of normally situated, 'eutopic' endometrium, 49 proliferating under the influence of estradiol. Endometriosis often causes severe pelvic pain, as 50 well as reduced fertility in many patients. Management typically involves various means to 51 reduce the effects of estrogenic stimuli that promote endometrial tissue proliferation, and 52 surgery to remove endometriotic tissue [2,3]. 53 54 Endometriosis risk is mediated by many genes each of small effect [4], exhibiting a heritability 55 of approximately 0.50 [5,6]. GWA (genome wide association) studies have identified a set of 56 single nucleotide polymorphisms (SNPs) that account for a small proportion of this heritability, 57 and that includes SNPs in genes with demonstrated roles in early sexual development (WNT4, 58 HOXC6), steroid hormone signalling (ESR1, GREB1, KDR), overall growth (IGF1), HPO axis 59 function (FSHB), among other phenotypes [4]. Endometriosis risk is also mediated by 60 environmental steroid effects, as exemplified by links of the disorder with the synthetic 61 estrogen diethylstilbestrol, via influences during prenatal development [7,8]. 62 63 A series of studies has tested for associations of endometriosis with anogenital distance (AGD), 64 which represents an indicator of levels of prenatal testosterone in early fetal development 65 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.05.24300901doi: medRxiv preprint 4 (reviews in [9,10]). Anogenital distance, which is measured from the anus to landmarks on the 66 genitalia, is substantially longer on average in males than in females among humans and other 67 mammals, due to variation between the sexes in levels of prenatal testosterone that control 68 growth of the perineal region. AGD also varies substantially within each sex, as indicated by 69 observational studies and extensive experimentation with non-human animals, as well as 70 studies of humans naturally subject to altered levels of testosterone [11–14]. Female mammals 71 that develop under relatively low levels of prenatal testosterone thus tend to develop relatively 72 short AGDs, and high levels result in longer AGDs [15]. 73 74 Anogenital distance in women with endometriosis was initially studied in the context of 75 developing new biomarkers for this disorder [16]. More recently, AGD has been analyzed in the 76 context of the developmental basis of endometriosis, whereby relatively low prenatal 77 testosterone differentially programs the developing hypothalmic-pituitary-ovarian (HPO) axis in 78 such a way as to increase risk in later life [17,18]. This paradigm provides a novel conceptual 79 and empirical framework for understanding and studying endometriosis, in a comparable way 80 to analyses of the well-established role of relatively high prenatal testosterone in risk for 81 development of polycystic ovary syndrome [17]. 82 83 A suite of studies of anogenital distance in women with endometriosis, compared to controls, 84 has been published to date. These studies all include the two main measures of AGD: AGD-AF 85 (from the anus to the posterior fourchette), and AGD-AC (from the anus to the clitoral surface), 86 and the results are heterogeneous both across studies and between these two measures of 87 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.05.24300901doi: medRxiv preprint 5 AGD. The purpose of this analysis is to conduct the first meta-analysis of AGD in relation to risk 88 of endometriosis, to ascertain the strength of evidence overall, and to help guide future work. 89 90

91 Three databases were searched systematically on 23 November 2023. First, Web of Science was 92 searched using the search terms 'anogenital distance endometriosis' in the 'All fields' search 93 box, which returned a total of 33 entries. Second, PubMed was searched using the search terms 94 'anogenital distance endometriosis', which generated 21 returns. Third, Google Scholar, a more 95 broadly inclusive search engine, was also searched using the terms 'anogenital distance 96 endometriosis', for any date, and with sorting by relevance. For this latter database, the first 97 100 citations were included in the initial screening. The returned articles were searched to 98 determine if they met the search criteria (data on AGD from women with endometriosis, and 99 from controls), and the other criteria for inclusion (unique data set not used in published work 100 previously, and conduct of measurements for both AGD-AF and AGD-AC). A PRISMA diagram 101 depicts the search and inclusion processes (Figure 1). 102 103 A random-effects model meta-analysis was conducted using the program Meta MAR [19]. The 104 main goal of the meta-analysis was to increase statistical power to test for differences in AGD-105 AF, AGD-AC, or both, between women with endometriosis and controls. A random-effects 106 model was used so that the results can be applied beyond the included studies, and given that 107 women with different forms and severities of endometriosis may be subject to different 108 magnitudes of effects on AGD (see e.g., [20]). 109 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.05.24300901doi: medRxiv preprint 6 110 For the study [21] that included measurements of AGD-AF and AGD-AC from women with two 111 different forms of endometriosis (ovarian and deep-infiltrating) separately but not together, 112 the authors provided a copy of their data set, which allowed calculation and use of the data for 113 all women with endometriosis combined. 114 115

116 A total of six studies met the inclusion criteria (Figure 1), comprising a total of 883 subjects 117 overall, and were subject to meta-analyses for AGD-AF and AGD-AC. The random-effects meta-118 analysis demonstrated a significant difference between women with endometriosis and 119 controls for AGD-AF (SMD=-0.77, p = 0.026, Figure 2). By contrast, the meta-analysis indicated 120 no significant difference for AGD-AC (SMD=-0.29, p = 0.166, Figure 2). Both analyses 121 demonstrated significant heterogeneity between studies (AGD-AF, Cochran's Q = 47.4, P < 122 0.001; AGD-AC, Cochran's Q = 30.8, p < 0.001). There was an insufficient number of studies 123 available for robust interpretation of funnel plots in the context of testing for publication bias. 124 125

126 The primary finding of this study is that meta-analysis provides support for the hypothesis that 127 AGD-AF is significantly shorter among women with endometriosis than among controls. By 128 contrast, the comparison for AGD-AC was non-significant. The results for AGD-AF indicate that 129 endometriosis shows a robustly replicated association with this measure of prenatal 130 testosterone, such that low levels of this hormone in early prenatal development represent a 131 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.05.24300901doi: medRxiv preprint 7 risk factor, even if the predictive value of AGD-AF as a biomarker for diagnosis of endometriosis 132 is limited [16,22]. These findings are of importance given that: (1) two studies have recently 133 demonstrated that genetically predicted lower serum testosterone is significantly associated 134 with higher risk of endometriosis [23,24], and (2) shorter anogenital distance (AGD-AF) has 135 been linked with lower serum teststerone in healthy women of reproductive age [25]. 136 137 The meta-analyses indicated significant heterogeneity among studies, for both AGD-AF and 138 AGD-AC. Some of this variation is likely due to the populations analyzed; for example, the 139 studies differed in the proportions of women with different forms of endometriosis (deep-140 infiltrating, ovarian, or peritoneal), in the body mass index (BMI) of participants, and in other 141 variables. For example, nine of the 43 women with endometriosis in one sample [26] exhibited 142 hyperandrogenism (and eight reported acne, and two showed hirsutism), which is characteristic 143 of polycystic overy syndrome (PCOS), and may be related to the lack of difference in AGD-AF 144 between women with endometriosis and controls in this study. 145 146 The differences in results reported here between AGD-AF and AGD-AC should be considered in 147 the context of the AGD-AF distance being a subset of the AGD-AC distance, such that AGD-AC is 148 equal to the sum of AGD-AF plus the distance between the posterior fourchette to the clitoral 149 surface. As such, AGD-AF and AGD-AC are related as part-whole, and are positively correlated 150 with one another, with correlation coefficients of between 0.33 and 0.60 (mean=0.48, N =7) 151 [16,26–31]. Given that the overall average distance of AGD-AF is about one-third that of the 152 overall average distance of AGD-AC, a substantial proportion of the mean difference between 153 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.05.24300901doi: medRxiv preprint 8 women with endometriosis and controls in AGD-AC can thus be attributed to the differences in 154 AGD-AF. As such, for the results reported here, AGD-AF can be considered as the metric that 155 differs significantly between women with endometriosis compared to controls, with any 156 differences in AGD-AC (in some individual studies) being attributable at least in part to its 157 variation. These inferences are demonstrated most clearly in the results from Mendiola et al. 158 [16] and Buggio et al. [21], where the distance AGD-AF is 50–100% of that for AGD-AC, such 159 that the former accounts for the bulk of the variation in the latter. 160 161 The biological significance of AGD-AF, compared to AGD-AC, is supported by the observations 162 that: (1) the approximately two-fold difference between females and males for distance of the 163 perineum (measured from the anus to the genitalia: posterior fourchette in females or base of 164 scrotum in males), and corresponding to AGD-AF, is proportionally much larger than the sex 165 difference in AGD-AC (measured in males from the anus to the anterior base of penis) (e.g., 166 [32]); (2) among rodents, the longer AGD in males than females, which is known from 167 experiments to be due to higher prenatal testosterone in males, corresponds to AGD-AF among 168 humans; (3) AGD-AF, but not AGD-AC, is positively correlated with levels of serum testosterone 169 in young women without endometriosis or PCOS [25,28]; and (4) AGD-AF is a better predictor of 170 endometriosis than is AGD-AC in studies that use AUC (Area Under receiver operating 171 Characteristic curve) analyses [27,33]. 172 173 Significantly shorter AGDs (for AGD-AF, AGD-AC, or both) have also been associated with four 174 correlates of endometriosis: premature ovarian insufficiency [34,35], dysmenorrhea [36,37], 175 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.05.24300901doi: medRxiv preprint 9 reduced responses to ovarian stimulation in IVF [29,38], and relatively regular menstrual cycles 176 [39]. These findings support the hypothesis of causal links between lower prenatal testosterone 177 and higher risk of endometriosis, as does evidence suggesting that women with endometriosis 178 exhibit a higher level of sexual difference from males than do healthy women for a suite of 179 sexually dimorphic traits [40]. Most broadly, these results provide evidence that endometriosis 180 represents, in part, a reproductive disorder mediated by early development whereby relatively 181 low testosterone affects the expression of many sexually dimorphic and sex limited traits [18]. 182 This hypothesis is convergently supported by recent Mendelian Randomzation studies showing 183 that lower genetically predicted testosterone is linked with higher risk of endometriosis, such 184 that the relationship between the two shows evidence of causality [23,24]. Future studies that 185 investigate effects of low prenatal testosterone in women, in the context of other correlates of 186 endometriosis (such as early menarche and higher pain sensitivity), should provide further 187 insights into the causes of this disorder. Studies testing for shorter AGDs in daughters of women 188 with endometriosis, compared to controls, would also represent useful tests for effects of low 189 prenatal testosterone in endometriosis risk. 190 191 The main limitations of this study include the relatively small number of studies conducted thus 192 far on AGD in relation to endometriosis, the low sample sizes of some of the studies, their 193 restriction thus far mainly to women of European descent, and the heterogeneity among 194 studies in the clinical characteristics of the particpants. These limitations should be addressed 195 with larger studies of more-diverse participants, and analyses of the genetic basis of AGD and 196 its relationship to endometriosis risk factors. 197 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.05.24300901doi: medRxiv preprint 10 198 In contrast to these studies of endometriosis and its correlates, analyses of PCOS provide 199 evidence for longer AGDs in women with this disorder. Thus, in three data sets, women with 200 PCOS showed significantly longer AGDs compared to controls [27,28] or to women with 201 endometriosis [26], and in a fourth study, the difference between women with PCOS and 202 controls was marginally non-significant (p=0.08 for AGD-AF, 0.17 for AGD-AC) [41]. Women with 203 PCOS also show evidence of bearing daughters with longer AGDs, compared to controls, from 204 three studies [42–44], whereas a fourth study showed no significant differences [45]. 205 206 Longer AGDs in women with PCOS, and in their daughters, may be related in part to their higher 207 body mass indices (BMIs) (women with PCOS had significantly higher BMIs than did controls in 208 most studies of AGD [26,27,41,46]), given that: (1) BMI is higher among women with PCOS than 209 in controls (e.g., [47]); (2) BMI is positively correlated with AGD among women [16,26,29,31, 210 33]; (3) maternal obesity is associated with longer AGD among daughters [48,49]; (4) AGD is 211 positively correlated with serum testosterone [25,28]; and (5) BMI is positively associated with 212 levels of testosterone, which represent a key correlate of PCOS [50]. Additional studies are 213 needed of AGD in women with PCOS, and in daughters of women with PCOS, in the context of 214 maternal BMI and testosterone, for sufficient data to be available for meta-analysis of AGD in 215 this disorder. 216 217 218 219 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.05.24300901doi: medRxiv preprint 11

220 This meta-analysis provides evidence that risk of endometriosis is mediated by lower anogenital 221 distance (specifically, AGD-AF), which suggests that lower testosterone in prenatal 222 development increases risk of this disorder. Further studies are needed on AGD-AF in relation 223 to the correlates and potential causes of endometriosis. 224 225 Abbreviations: 226 AGD: anogenital distance 227 AGD-AC: anogenital distance, anus to clitoral surfac 228 AGD-AF: anogenital distance, anus to posterior fourchette 229 BMI: body mass index 230 GWA: genome-wide-association 231 PCOS: polycystic ovary syndrome 232 SNP: single nucleotide polymorphism 233 234 235 Declarations: 236 Ethics approval and consent to participate: Not applicable 237 Consent for publication: Not applicable 238 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.05.24300901doi: medRxiv preprint 12 Availability of data and materials: The data (means, standard deviations, and sample sizes) 239 used in the meta-analysis were taken from the respective studies that were included in the 240 systematic review, and are shown in Figure 1. 241 Competing interests: The author declares that they have no competing interests. 242 Approvals: The author has seen and approved the manuscript. 243 Funding: This work was funded by Discovery Grant 2018-04208 from the Canadian Natural 244 Sciences and Engineering Research Council. 245 Authors' contributions: BC conducted all of the research and manuscript preparation. 246

I am grategul to L. Buggio for provision of information allowing calculation 247 of AGD across all participants in their article [21]. 248 249

250 1. Vercellini P, Viganò P, Somigliana E, Fedele L. Endometriosis: pathogenesis and 251 treatment. Nat Rev Endocrinol. 2014;10:261-75. 252 2. Kalaitzopoulos DR, Samartzis N, Kolovos GN, Mareti E, Samartzis EP, Eberhard M, et al. 253 Treatment of endometriosis: a review with comparison of 8 guidelines. BMC Women's 254 Health. 2021;21:1-9. 255 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.05.24300901doi: medRxiv preprint 13 3. Vercellini P, Bandini V, Viganò P, Di Stefano G, Merli CE, Somigliana E. Proposal for 256 targeted, neo-evolutionary-oriented, secondary prevention of early-onset endometriosis 257 and adenomyosis. Part I: pathogenic aspects. Hum Reprod. 2023;dead229. 258 4. Rahmioglu N, Mortlock S, Ghiasi M, Møller PL, Stefansdottir L, Galarneau G, et al. The 259 genetic basis of endometriosis and comorbidity with other pain and inflammatory 260 conditions. Nat Genet. 2023;55:423-36. 261 5. Treloar SA, O’Connor DT, O’Connor VM, Martin NG. Genetic influences on endometriosis 262 in an Australian twin sample. Fertil Steril. 1999;71:701–10. 263 6. Saha R, Pettersson HJ, Svedberg P, Olovsson M, Bergqvist A, Marions L, et al. Heritability 264 of endometriosis. Fertil Steril. 2015;104:947-52. 265 7. Benagiano G, Brosens I. In utero exposure and endometriosis. J Maternal-Fetal Neonatal 266 Med. 2014;27:303-8. 267 8. Gaspari L, Soyer-Gobillard MO, Paris F, Kalfa N, Hamamah S, Sultan C. Multigenerational 268 endometriosis: consequence of fetal exposure to diethylstilbestrol? Environ Health. 269 2021;20:1-5. 270 9. Pan Z, Zhu F, Zhou K. A systematic review of anogenital distance and gynecological 271 disorders: endometriosis and polycystic ovary syndrome. Front Endocrinol. 272 2021;12:696879. 273 10. Zamani P, Hemati Z, Kelishadi R, Kolahdozan S, Dianatinasab M, Keikha M. Association 274 between anogenital distance as a noninvasive index in the diagnosis and prognosis of 275 reproductive disorder: a systematic review. Int J Reprod BioMed. 2023;21:599. 276 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.05.24300901doi: medRxiv preprint 14 11. Dean A, Sharpe RM. Anogenital distance or digit length ratio as measures of fetal 277 androgen exposure: relationship to male reproductive development and its disorders. J 278 Clin Endocrinol Metab. 2013;98:2230-8. 279 12. Liu C, Xu X, Huo X. Anogenital distance and its application in environmental health 280 research. Environ Sci Pollution Res. 2014;21:5457-64. 281 13. Thankamony A, Pasterski V, Ong KK, Acerini CL, Hughes IA. Anogenital distance as a 282 marker of androgen exposure in humans. Andrology. 2016;4:616-25. 283 14. Schwartz CL, Christiansen S, Vinggaard AM, Axelstad M, Hass U, Svingen T. Anogenital 284 distance as a toxicological or clinical marker for fetal androgen action and risk for 285 reproductive disorders. Arch Toxicol. 2019;93:253-72. 286 15. Svingen T, Hass U, Schwartz CL, Christiansen S. Anogenital distance: features, measures, 287 and uses as a biomarker for toxicity in utero. In: Gupta RC, editor. Biomarkers in 288 toxicology. Cham: Springer International Publishing; 2023. p. 609-21. 289 16. Mendiola J, Sánchez-Ferrer ML, Jimenez-Velazquez R, Canovas-Lopez L, Hernandez-290 Penalver AI, Corbalan-Biyang S, et al. Endometriomas and deep infiltrating 291 endometriosis in adulthood are strongly associated with anogenital distance, a 292 biomarker for prenatal hormonal environment. Hum Reprod. 2016;31:2377-83. 293 17. Dinsdale NL, Crespi BJ. Endometriosis and polycystic ovary syndrome are diametric 294 disorders. Evol Appl. 2021;14:1693-715. 295 18. Dinsdale N, Nepomnaschy P, Crespi B. The evolutionary biology of endometriosis. Evol 296 Med Public Health. 2021;9:174-91. 297 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.05.24300901doi: medRxiv preprint 15 19. Beheshti A, Chavanon ML, Christiansen H. Emotion dysregulation in adults with 298 attention deficit hyperactivity disorder: a meta-analysis. BMC Psychiatry. 2020;20:1-1. 299 20. Tufanaru C, Munn Z, Stephenson M, Aromataris E. Fixed or random effects meta-300 analysis? Common methodological issues in systematic reviews of effectiveness. JBI 301 Evidence Implementation. 2015;13:196-207. 302 21. Buggio L, Somigliana E, Sergenti G, Ottolini F, Dridi D, Vercellini P. Anogenital distance 303 and endometriosis: results of a case-control study. Reprod Sci. 2022;29:3508-15. 304 22. Sánchez-Ferrer ML, Jiménez-Velázquez R, Mendiola J, Prieto-Sánchez MT, Cánovas-305 López L, Carmona-Barnosi A, et al. Accuracy of anogenital distance and anti-Müllerian 306 hormone in the diagnosis of endometriosis without surgery. Int J Gynecol Obstetr. 307 2019;144:90-6. 308 23. McGrath IM, International Endometriosis Genetics Consortium, Montgomery GW, 309 Mortlock S. Polygenic risk score phenome-wide association study reveals an association 310 between endometriosis and testosterone. BMC Med. 2023;21:482. 311 24. Zhao B, Wang Z, Liu D, Zhang S. Genetically predicted serum testosterone and risk of 312 gynecological disorders: a Mendelian randomization study. Front Endocrinol. 2023;14. 313 25. Mira-Escolano MP, Mendiola J, Mínguez-Alarcón L, Melgarejo M, Cutillas-Tolín A, Roca 314 M, et al. Longer anogenital distance is associated with higher testosterone levels in 315 women: a cross-sectional study. BJOG. 2014a;121:1359-64. 316 26. Peters HE, Laeven CH, Trimbos CJ, van de Ven PM, Verhoeven MO, Schats R, et al. 317 Anthropometric biomarkers for abnormal prenatal reproductive hormone exposure in 318 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.05.24300901doi: medRxiv preprint 16 women with Mayer-Rokitanksy-Küster-Hauser syndrome, polycystic ovary syndrome, 319 and endometriosis. Fertil Steril. 2020;114:1297-305. 320 27. Sánchez-Ferrer ML, Mendiola J, Hernández-Peñalver AI, Corbalán-Biyang S, Carmona-321 Barnosi A, Prieto-Sánchez MT, et al. Presence of polycystic ovary syndrome is associated 322 with longer anogenital distance in adult Mediterranean women. Hum Reprod. 323 2017;32:2315-23. 324 28. Wu Y, Zhong G, Chen S, Zheng C, Liao D, Xie M. Polycystic ovary syndrome is associated 325 with anogenital distance, a marker of prenatal androgen exposure. Hum Reprod. 326 2017;32:937-43. 327 29. Fabregues F, González-Foruria I, Peñarrubia J, Carmona F. Ovarian response is associated 328 with anogenital distance in patients undergoing controlled ovarian stimulation for IVF. 329 Hum Reprod. 2018;33:1696-704. 330 30. Li Z, Xu M. Clitoris to urethral meatus distance, an assumed marker of prenatal androgen 331 exposure is not correlated with the obesity compared to the anogenital distance. Clin 332 Exp Obsetr Gynecol. 2022;49:138. 333 31. Liu X, Ding D, Shen M, Yan D, Guo SW. Shorter anogenital distance in women with 334 ovarian endometriomas and adenomyosis, but not uterine leiomyomas. Biomedicines. 335 2023;11:2618. 336 32. Loreto-Gómez C, Farías P, Moreno-Macías H, Romano-Riquer SP, Riojas-Rodríguez H. 337 Anogenital distance: a longitudinal evaluation of its variants and indices in boys and girls 338 of Sonora, Mexico. Reprod Toxicol. 2017;73:167-74. 339 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.05.24300901doi: medRxiv preprint 17 33. Crestani A, Arfi A, Ploteau S, Breban M, Boudy AS, Bendifallah S, et al. Anogenital 340 distance in adult women is a strong marker of endometriosis: results of a prospective 341 study with laparoscopic and histological findings. Hum Reprod Open. 342 2020;2020:hoaa023. 343 34. Dural O, Kurbanova T, Yasa C, Ugurlucan FG, Selvi E, Ozdemir PK, et al. Idiopathic 344 primary ovarian insufficiency is associated with anogenital distance, a marker for 345 prenatal environment. Eur J Obstetr Gynecol Reprod Biol. 2021;258:304-8. 346 35. Tan Z, Gong X, Li Y, Hung SW, Huang J, Wang CC, et al. Impacts of endometrioma on 347 ovarian aging from basic science to clinical management. Front Endocrinol. 348 2023;13:1073261. 349 36. Hewitt G. Dysmenorrhea and endometriosis: diagnosis and management in adolescents. 350 Clin Obstetr Gynecol. 2020;63:536-43. 351 37. Saputra NP, Noviardi, Sonia N,Sartika D, Nadisa S. Anogenital distance as a predictor of 352 dysmenorrhea. 4th Riau Medical Scientific and Expo 2022. NST Proceedings; 2022. p. 353 122-125. 354 38. Hamdan M, Omar SZ, Dunselman G, Cheong Y. Influence of endometriosis on assisted 355 reproductive technology outcomes: a systematic review and meta-analysis. Obstet 356 Gynecol. 2015;125:79-88. 357 39. Mira-Escolano MP, Mendiola J, Mínguez-Alarcón L, Roca M, Cutillas-Tolín A, López-Espín 358 JJ, et al. Anogenital distance of women in relation to their mother’s gynaecological 359 characteristics before or during pregnancy. Reprod BioMed Online. 2014b;28:209-15. 360 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.05.24300901doi: medRxiv preprint 18 40. Crespi B, Dinsdale NL. The sexual selection of endometriosis. Evol Behav Sci. 361 2023;17:123. 362 41. Simsir C, Pekcan MK, Aksoy RT, Ecemis T, Coskun B, Kilic SH, et al. The ratio of anterior 363 anogenital distance to posterior anogenital distance: a novel-biomarker for polycystic 364 ovary syndrome. J Chinese Med Assoc. 2019;82:782-6. 365 42. Barrett ES, Hoeger KM, Sathyanarayana S, Abbott DH, Redmon JB, Nguyen RH, et al. 366 Anogenital distance in newborn daughters of women with polycystic ovary syndrome 367 indicates fetal testosterone exposure. J Dev Origins Health Disease. 2018;9:307-14. 368 43. Perlman S, Toledano Y, Kivilevitch Z, Halevy N, Rubin E, Gilboa Y. Foetal sonographic 369 anogenital distance is longer in polycystic ovary syndrome mothers. J Clin Med. 370 2020;9:2863. 371 44. Bayraktar B, Taner CE. The effect of polycystic ovary syndrome history on neonatal 372 anogenital distance: a prospective study in Turkish population. J Health Sci Med. 373 2023;6:30-4. 374 45. Glintborg D, Jensen RC, Schmedes AV, Brandslund I, Kyhl HB, Jensen TK, et al. Anogenital 375 distance in children born of mothers with polycystic ovary syndrome: the Odense Child 376 Cohort. Hum Reprod. 2019;34:2061-70. 377 46. Prieto-Sánchez MT, Hernández-Peñalver AI, Sánchez-Ferrer ML, Mendiola J, Torres-378 Cantero AM. Anogenital distance and anti-Müllerian hormone combined improves the 379 diagnosis of polycystic ovary syndrome. Hum Fertil. 2022;25:274-82. 380 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.05.24300901doi: medRxiv preprint 19 47. Lim SS, Norman RJ, Davies MJ, Moran LJ. The effect of obesity on polycystic ovary 381 syndrome: a systematic review and meta-analysis. Obesity Rev. 2013;14:95-109. 382 48. Kloboves ME, Pacyga DC, Gardiner JC, Flaws JA, Schantz SL, Strakovsky RS. Associations 383 of maternal anthropometrics with newborn anogenital distance and the 2:4 digit ratio. 384 Hum Reprod. 2022;37:2154-66. 385 49. Halici-Ozturk F, Yetiskin FD, Gurlek B, Ocal FD, Yakut K, Engin-Ustun Y, et al. Longer 386 anogenital distance in female fetus of diabetic and obese pregnant women. Taiwanese J 387 Obstetr Gynecol. 2023;62:530-6. 388 50. Diamanti-Kandarakis E, Bergiele A. The influence of obesity on hyperandrogenism and 389 infertility in the female. Obesity Rev. 2001;2:231-8. 390 391 392 393 394 395 396 397 398 399 400 401 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.05.24300901doi: medRxiv preprint 20 402 Figures 403 404 Figure 1. PRISMA diagram showing identification, screening, and inclusion process for the meta-405 analysis. 406 407 408 409 410 411 412 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.05.24300901doi: medRxiv preprint 21 Figure 2. Main results of meta-analyses for (a) AGD-AF and (b) AGD-AF. 413 414 415 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.05.24300901doi: medRxiv preprint