EndoART: A proposed randomized controlled trial on endometriomas in assisted reproductive technologies, comparing the effect of no intervention, surgery, and prolonged GnRH downregulation on pregnancy rates

Background: Clinical management of endometriomas, prior to starting treatment with assisted reproductive technologies (ART), has since long been a matter of debate. Whereas cystectomy has been advocated in the past, recently more evidence has emerged on the potential negative effect of surgery on ovarian reserve. Parallel to this, prolonged downregulation with GnRH-a (gonadotropin-releasing hormone agonists) has been described to improve ART pregnancy rates in women with endometriosis. However, none of these strategies have been assessed in a large randomized controlled trial. The aim of the EndoART study is to assess whether ovarian surgery or prolonged GnRH-a downregulation result in higher pregnancy rates after ART compared to no intervention in women with endometrioma(s). Methods/design: A parallel randomized multi-center trial has been designed to compare ART pregnancy rates in three different treatment groups: no intervention, ovarian surgery, and prolonged hormonal suppression by GnRH-a prior to ART. The primary outcome measure studied is the clinical pregnancy rate with fetal heart-beat within 6 months after initiation of a fresh ART cycle. Secondary outcome measures studied include live birth rate after one initiated fresh ART cycle, cumulative clinical pregnancy rate with fetal heart-beat and live birth rates (after one fully completed ART cycle: initiated fresh + eventual associated frozen embryo transfer cycles), ART–specific data (e.g. number of oocytes, number of good quality embryos), complications, pelvic pain, and quality of life.

This trial may answer the most frequently asked questions by both women with endometriosis and physicians: how do you treat endometrioma in women prior to treatment with ART? 1 Department of Obstetrics and Gynaecology, Leuven University Fertility Center, University Hospitals Leuven, Leuven, Belgium 2 Department of Development and Regeneration, Organ Systems, KU Leuven, Leuven, Belgium 3 Palo Alto Medical Foundation Fertility Physicians of Northern California, East Palo Alto, CA, USA 4 Division of Reproductive Endocrinology and Infertility, Department of Obstetrics, Gynaecology and Reproductive Sciences, Yale University School of Medicine, New Haven, CT, USA 5 Department of Gynaecological Surgery, CHU Estaing, Clermont- Ferrand, France 6 CICE (Centre International de Chirurgie Endoscopique), Faculty of Medicine, Clermont-Ferrand, France 7 Vrije Universiteit Medical Center, Center for Reproductive Medicine, Amsterdam, The Netherlands 8 World Endometriosis Research Foundation, London, UK 9 The Robinson Research Institute, School of Medicine, University of Adelaide, The South Australian Health and Medical Research Institute, Adelaide, SA, Australia 10 Reproductive Health Group, Manchester, UK 11 Monash University, Melbourne, VIC, Australia Corresponding author: Carla Tomassetti, Department of Obstetrics and Gynaecology, Leuven University Fertility Center, University Hospitals Leuven, Herestraat 49, 3000 Leuven, Belgium. Email: [email protected] 784236 PEV0010.1177/2284026518784236Journal of Endometriosis and Pelvic Pain DisordersT omassetti et al. research-article2018 Clinical Trial Protocol Tomassetti et al. 159

Endometriosis, ovarian cyst, ART, operative surgical procedures, gonadotropin-releasing hormone agonist, randomized controlled trial Date received: 24 January 2018; accepted: 30 May 2018

literature background Endometriosis is a disease characterized by the presence of endometrium-like epithelium and stroma outside the endome- trium and myometrium. Intrapelvic endometriosis can be located superficially on the peritoneum (peritoneal endome- triosis), can extend 5 mm or more beneath the peritoneum (deep endometriosis) or can be present as an ovarian endome- triotic cyst (endometrioma). 1 Ovarian endometriotic cysts are rarely solitary, and are mostly found in association with adnexal endometriosis adhesions, deep and/or peritoneal endometriosis. Such disease is classified as moderate or severe endometriosis. 2 Endometriosis (including endometrioma) may cause infertility due to a combination of factors, including pelvic adhesion formation compromising ovum pick up mech- anism, impaired ovarian function, impaired oocyte quality, endometrial dysfunction, peritoneal inflammation, immuno- logical dysfunction, and potential other factors. 3 There is now molecular, histological, and morphologi- cal evidence to suggest that endometriosis is detrimental to the ovaries: toxic content from an endometrioma may lead to unfavorable events such as increased oxidative stress, increase fibrosis, loss of cortex-specific stroma, smooth muscle cell metaplasia, vascularization defect, and, later, reduced follicular maturation. 4 Whether this vicious cycle of damage can be ameliorated by surgical treatment before in vitro fertilization/intracytoplasmic sperm injection (IVF/ICSI) is still controversial. 5 Effect of endometriosis on assisted reproductive technologies pregnancy rates The influence of endometriosis on the pregnancy rates of assisted reproductive technologies (ART) is equivocal. In an older meta-analysis study, the pregnancy rates after ART were reported to be up to 56% lower in women with stage III–IV endometriosis as compared with those with tubal factor, due to a lower number of retrieved oocytes, a lower fertilization rate, and a lower implantation rate. 6 A more recent meta-analysis confirmed that women with stage III–IV endometriosis have inferior fertility out- comes, but that there is insufficient evidence to recom- mend surgery routinely before undergoing ART. 7 Of note, other studies and some large databases show that a diagno- sis of endometriosis does not adversely affect pregnancy rates (e.g. the Society for Assisted Reproductive Technology (SART) and the Human Fertilization and Embryology Authority (HFEA)). 8 When focusing on women with intact endometrioma undergoing ART com- pared to women without endometriosis, a recent system- atic review concluded that they had similar reproductive outcomes (live birth, clinical pregnancy, and miscarriage rate) compared with those without the disease, although their cycle cancelation rate was nearly three times higher and they had a lower number of oocytes retrieved; 5 this was only based on the information from maximally five studies (of which none were randomized for obvious rea- sons). This review further concluded that of all evidence found regarding endometrioma and ART, the majority of the 33 studies included were not randomized (except three). Furthermore, the authors highlighted the paucity of data available for analysis and their high heterogeneity and recommend that future studies should be carried out to examine adverse events, including cancelation rates. Effect of endometrioma surgery on ovarian reserve Laparoscopic excision of endometriotic cysts is superior compared to drainage and coagulation or laser vaporiza- tion of the cyst wall with respect to both spontaneous preg- nancy rate and cyst recurrence rate. 8,9 However, it is not clear if excision of endometriotic cysts prior to ART is beneficial in terms of fertility outcome, pain symptoms, recurrence rates, and complication rates when compared to expectant management. More recent systematic reviews, concerning the effect of endometrioma surgery on ovarian damage and reserve, also come to conflicting conclusions, depending on outcome measures that were studied. A sys- tematic review and meta-analysis by Somigliana et al. 10 showed—although heterogeneity was high—a statistically significant decline in postoperative anti-Müllerian hor - mone (AMH) serum levels after endometrioma cystec- tomy, attributed to the surgery itself, suggesting that this technique should be used with caution. A systematic review by Muzii et al., 11 however, concluded that ovarian reserve evaluated with antral follicle count (AFC) is not reduced after surgical treatment of an endometrioma, and a lower AFC is present for the affected ovary both before and after surgery. Less invasive techniques, such as the combined technique (combination of partial cystectomy and subsequent CO 2-laser ablation near the hilus12 or abla- tion of the cyst wall with low-energy sources such as plasma energy 13 appear to cause less damage to the ovarian cortex, although there are no large randomized series that draw firm conclusions). On these two surgical techniques 160 Journal of Endometriosis and Pelvic Pain Disorders 10(3) per se (cystectomy versus (bipolar) ablation), only limited non-randomized comparative evidence exists regarding their effect on the ovarian reserve. Three studies point toward an advantage of (bipolar) ablation over classical cystectomy on ovarian reserve testing: one study uses AFC as an ovarian reserve marker, 14 and the other two studies assess ovarian reserve by AMH. 15–16 Although both abla- tion and cystectomy had a negative effect on ovarian reserve, none of these studies were performed in a pre- ART setting. A small study on ART patients where women with endometriosis who underwent plasma energy abla- tion for the endometrioma(s) prior to ART were compared to women without endometriosis undergoing ART 17 con- cluded that ovarian endometrioma ablation using plasma energy is followed by good IVF/ICSI outcomes and sug- gested that this surgical procedure spares underlying ovar- ian parenchyma. However, this does not directly relate to the subject of the study as outlined in this article, since this is not a true comparison between treated and untreated endometrioma(s). The effect of endometrioma treatment on ovarian response in ART The effect of endometriotic ovarian cysts and of their surgi- cal or medical treatment on ovarian response during ART is equally controversial. Regarding surgical treatment, a ret- rospective analysis has shown that the number of oocytes and embryos obtained was not significantly decreased by laparoscopic cystectomy of cysts larger than 3 cm, suggest- ing that in experienced hands, this procedure might be a valuable surgical tool for the treatment of large ovarian endometriomas. 18 Several other studies, however, suggest that cystectomy for endometrioma(s) is followed by a reduced ovarian response during ovarian stimulation for ART, resulting in higher gonadotropin use when compared to patients without ovarian surgery for endometriomas before ART. 19 Unilateral excision or drainage and coagula- tion or laser vaporization of endometriomas before ART has also been reported to result in reduced responsiveness of the operated ovary during ART-stimulation. 20 In view of existing evidence that reduced ovarian responsiveness is present in women with non-operated endometriomas 21,22 especially when larger than 30 mm,23 even though the oocyte developmental competence per se is not affected,24 it is not clear to what extent surgical treatment of ovarian endometriotic cysts reduced ovarian response is caused by the presence of the cysts and/or by the damage caused by the surgical procedure. In the only randomized controlled trial (RCT) addressing this clinical question, 25 women with ovarian cystectomy for endometriomas followed by intra- cytoplasmatic sperm injection (ICSI) required a higher dose of gonadotropins and a longer duration of ovarian stimulation and still had fewer oocytes than women with endometriomas, who were treated directly with ICSI without ovarian cystectomy. This was also confirmed in a systematic review by Hamdan et al.: 5 compared to women without the disease, women with endometrioma require higher follicle-stimulating hormone (FSH) dosage for ovar- ian stimulation and have a lower AFC, suggesting that their ovarian reserve is diminished prior to ART. The authors further stated that surgical treatment of endometrioma prior to ART could exert a detrimental effect on ovarian reserve and that therefore a dogmatic recommendation on whether or not to perform surgery cannot be made. The use of medical treatment with prolonged pituitary downregulation before ART in patients with endometriosis in an older study is reported to cause no difference in total dose of gonadotropins used or days of stimulation when compared to classical pituitary downregulation, 26 but these data are hampered by small sample sizes and do not spe- cifically focus on patients with endometriomas. 8,27 A recent RCT evaluating GnRH-a (gonadotropin-releasing hor - mone agonists) versus no intervention prior to ART in women with surgically treated peritoneal endometriosis only showed no difference in terms of the number of meta- phase II (MII) oocytes obtained per cycle, but did find a significantly higher dose of FSH and a longer stimulation period to reach adequate follicle maturation in the GnRH-a group. 28 The effect of endometrioma with or without treatment on ART pregnancy rates The effect of endometriotic ovarian cysts and of their sur - gical or medical treatment on pregnancy rates after ART is again controversial. A Cochrane review evaluating sur - gery, medical treatment, combination therapy, or no treat- ment for improving reproductive outcomes among women with endometriomas prior to undergoing ART could not detect any evidence of an effect in the included trials (but only four trials were included). 29 According to a review paper including six retrospective controlled studies, the pregnancy rate after ART was either similar or reduced in women who previously underwent endometrioma surgery prior to ART, when compared to non-operated controls. 20 In the only RCT addressing this clinical question, 25 preg- nancy rates after ART were similar after cystectomy and after expectant management. A systematic review and meta-analysis 30 concluded that surgical management of endometriomas has no significant effect on pregnancy rates compared to no treatment and that there is insuffi- cient evidence to suggest superiority of one treatment strategy over another. This was confirmed in the review by Hamdan et al. 5 The effect of prolonged pituitary downregulation with GnRH-agonists (3–6 months) prior to ART has been sum- marized by a Cochrane review including only three small RCTs. 26,27,31,32 Although it was found to increase live birth rate by nearly four-fold compared to no pretreatment, the Tomassetti et al. 161 strength of this evidence is hampered by the small sample size and by the lack of data analysis in the subgroup of women with endometriomas. Potential adverse effects of the intervention (miscarriage, multiple pregnancy, and ectopic pregnancy) were not addressed in the included studies. International guidelines on management of endometrioma prior to ART Based on the lack of clear evidence in the literature on what to do with an endometriotic cyst prior to starting an ART treatment, clinicians mainly rely on guideline statements to make clinical decisions in these patients. However, a lack of evidence (or knowledge gap) does not imply a lack of potential benefit, and therefore, surgery is often chosen as a possible strategy. This is reflected by a survey on clinical practices in endometriomas detected prior to ART, which concludes that they are often operated on 33 as there is the intuitive feeling that the presence of an endometrioma reduces the success rate and increases the risks of treatment with ART, as many clinicians have experienced problems with an endometrioma complicating oocyte aspiration. On surgical treatment of endometriomas prior to ART, the World Endometriosis Society (WES) consensus on cur- rent management of endometriosis 34 states that—taking into account the lack of evidence that surgical treatment of endometriosis improves pregnancy rates through ART, the possible damage of the endometrioma and/or surgery itself to the ovary, and the possible complications of endometrio- mas during ART—laparoscopic cystectomy may some- times be recommended for women with endometriomas larger than 3 cm diameter. According to the European Society for Human Reproduction and Embryology (ESHRE) Guideline on the management of women with endometriosis, 8 it is recommended to clinicians that in women with endometrioma larger than 3 cm, cystectomy should only be considered prior to ART to improve endo- metriosis-associated pain or the accessibility of follicles (in contrast to the recommendations in the previous version of the guideline that recommended cystectomy in cysts larger than 4 cm). 35 They further recommend that clinicians coun- sel women with endometrioma regarding the risks of reduced ovarian function after surgery, the possible loss of the ovary, and consider that the decision to proceed with surgery should be taken carefully if the woman has had pre- vious ovarian surgery. Furthermore, the ASRM Practice Committee on Endometriosis and Infertility 36 concluded in 2012 that in each case, the benefits (prevention of possible cyst rupture in large cysts, facilitation of oocyte retrieval, detection of occult malignancy, avoidance of contamina- tion of follicular fluid with endometrioma content, and pre- vention of progression of endometriosis) and the risks (surgical trauma and complications, economic costs, poten- tial decreased ovarian response, and lack of evidence for improved ART results) should be balanced by clinicians and that therefore surgery should be considered if the endo- metrioma is larger than 4 cm. On medical treatment prior to ART, no specific state- ments exist on endometriosis cysts per se. With respect to guidelines on medical pretreatment prior to ART for endo- metriosis in general, both the WES consensus 34 and the ESHRE guidelines 8 state that clinicians can prescribe GnRH-a for a period of 3–6 months prior to treatment with ART (level B evidence). The opinion of the ASRM Practice Committee on Endometriosis and Infertility (2012) is less positive toward prolonged downregulation with GnRH-a prior to ART, as they acknowledge that results described in the literature are difficult to extrapolate to other popula- tions due to very high reported pregnancy rates and that it is unclear whether this therapy is equally beneficial for mild and severe stages of endometriosis and what the mechanisms might be. 36 Study objectives As described above, current statements on how to manage women with endometrioma(s) prior to ART cannot be made based on the available literature; clearly, more stud- ies are needed to address this important clinical question. The aim of this study is to provide an answer by conduct- ing a large international multi-center randomized clinical trial comparing the effect of three different treatment strat- egies on the rate of clinical pregnancy with fetal heartbeat (CPHB) after ART (no intervention, prolonged GnRH-a downregulation, and surgical treatment) prior to starting ART in women with at least one endometrioma with a diameter of 3 cm or more. Apart from CPHB, we want to evaluate further the effect of these three strategies on live birth rate, cumulative pregnancy rate, complications, qual- ity of life (QOL), ovarian reserve and specific ART param- eters (such as number of oocytes and embryo quality).

Study hypothesis and design The WERF EndoART study is designed as a parallel-group multi-center randomized controlled clinical trial to test the hypothesis that in women with at least one endometriotic ovarian cyst with a diameter of 3 cm or more, the CPHB rate after one initiated fresh ART cycle differs according to the treatment received before the start of the ART cycle. Women will be randomized into three different treatment groups: the Direct ART Group (no other treatment as preparation prior to the start of ART), the Surgery Group (surgical treatment of the ovarian endometrioma(s) prior to the start of ART), and the Ultra Long Group (prolonged (3–6 months) down- regulation with GnRH-a prior to the start of ART). This trial is an effectiveness trial: as opposed to an effi- cacy trial, it evaluates whether an intervention with proven 162 Journal of Endometriosis and Pelvic Pain Disorders 10(3) efficacy does more good than harm in the normally com- plex clinical setting.37 The goal is to measure the impact of real-time clinical practice in real patients in a realistic set- ting of clinical practice. In view of inherent heterogeneity, this trial cannot be an efficacy trial. In a surgical interven- tion, it is impossible to completely standardize all intraop- erative procedure, as every operation is different. Furthermore, due to international and multi-center recruit- ment, there will undoubtedly be variability in center-spe- cific practices of ART, surgery, and hormonal suppression with GnRH-a, which cannot be tolerated in an efficacy trial. However, in order to achieve as much standardization as possible, the methodology of an efficacy trial will be applied where possible. Furthermore, this study will take place as a superiority trial, as outlined further below. Sample size calculation Sample size is estimated based on the primary outcome: CPHB rate per randomized patient 6 months after initiation of one fresh ART cycle (= per-protocol analysis). Sample size is based on the baseline pregnancy rate expected in con- trols, the clinically important difference that might be expected with the surgical treatment and appropriate levels of alpha and beta error. With respect to the baseline preg- nancy rate, two retrospective studies and one prospective controlled but non-randomized trial have assessed surgical treatment versus no treatment of endometriomas prior to ART. 38–40 Although the treatments were different, the con- trol groups all received only ART treatment. The pregnancy rates were 20%, 23%, and 20%, respectively, in the three studies, resulting in an average pregnancy rate of about 21%. This trial is designed as a superiority trial, that is, to test the hypothesis that there is a significant difference in the primary outcome variable between three groups. Based on clinical grounds (patient burden, risks, and costs of surgery), it is reasonable to argue that unless the relative difference is as much as 33%, it may not be worthwhile to consider sur- gery. Given a baseline assumption of 21% of CPHB among patients with an endometrioma after one fresh ART cycle, it is reasonable to expect that the results should improve to 28% (i.e. 33% increase) if ART is preceded by surgery. The hypothesized effect size of prolonged downregula- tion in this study should be at least equal to the above hypothesized effect of surgery, taking into account addi- tional cost and patient burden (side effects and treatment delay) as well as literature data suggesting a possible four- fold increase of odds for pregnancy. 27 Given a baseline CPHB rate of 21%, a relative difference of 33% for the Surgery and the Ultra Long Groups, with the usual 5% alpha error (probability to wrongly reject the null hypoth- esis; type I error; ‘significance level’) and 20% beta error (probability to wrongly retain the null hypothesis; type II error; ‘power’) assumptions, the number of patients needed would be 563 per group using a chi-square test. To account for 5% losses to follow-up after randomization, it is pru- dent to aim for a total sample size of 1800 patients (600 per group). These assumptions are conservative and allow for the use of life table analysis and multivariate regression. No stratification by endometriosis stage will be used, as this decreases the sample per stratum and will make statis- tical analysis more complex. In addition, no stratification per center will be used, as including the trial center as a variable in the proportional hazards analysis should be suf- ficient to allow for the center effect (or clustering). It is even questionable whether within centers there is no vari- ation among surgeons and reproductive endocrinologists with respect to clinical techniques used (there probably is). Therefore, no sample size adjustments will be made to avoid clustering effects. Recruitment and study sites Patient recruitment will cover a period of three full years, and the follow-up of the last recruited patient is estimated to be completed 15 months later, with a possible extension of 6 months to be able to record the follow-up of all live birth rates. This study will be conducted worldwide, matching the global research goals of the World Endometriosis Research Foundation (WERF). Only infertility centers/networks with combined high-volume output and excellence in ART and in laparoscopic reproductive surgery for endometrio- sis (including experience and availability of techniques to perform ovarian sparing surgery) will be included. It is accepted that within the participating centers, women are treated surgically by a different (i.e. adequately experi- enced) physician than for ART, on the condition that at the start of the trial in each center, a list is provided of partici- pating physicians, stating their specific expertise (surgery, ART, or both). Eligible centers worldwide can apply for study partici- pation to the WERF EndoART consortium if they meet the following eligibility criteria: •• Quality assurance in ART; •• At least 250 fresh ART cycles per year; •• Existence of a quality monitoring system, prefera- bly with external validation; •• Quality assurance in surgery; •• At least 12 endometrioma surgeries per year; •• Demonstrated ability to surgically treat all stages of endometriosis at the same surgery at which the endometrioma is diagnosed and treated; •• Video of surgical technique used for treatment of endometrioma to be submitted to the WERF EndoART Consortium; •• Ability to recruit at least 20 randomized patients during a study period of 3 years; this is an essential condition to qualify for co-authorship. Tomassetti et al. 163 The WERF EndoART Coordinating Committee will review every application and reserves all rights to make a final decision regarding acceptance or not of the applying center, according to the formal list of conditions and quality assurance (see above). Only this committee can judge the expertise, quality and reputation of the apply- ing center. After approval of the WERF EndoART Coordinating Committee, the applying center becomes a member of the WERF EndoART Consortium. A partici- pating center will be excluded from further participation in this study if no patients have been randomized at 1 year after start of the study at the site. Randomization and blinding Patients who meet the study criteria and who have signed the informed consent form (ICF) are randomized into three groups with an equal distribution (1/3, 1/3, and 1/3): the Direct ART Group, the Surgery Group, or the Ultra Long Group. The study starts at the moment of randomization, including data recording on history, current symptoms, and QOL by the use of questionnaires (endometriosis patient questionnaire—minimum required (EPQ-M) World Endometriosis Research Foundation (WERF) Endometriosis Phenome and Biobanking Harmonization Project II (EPHect) Questionnaire, 41 30-item Endometriosis Health Profile (EHP-30) Oxford questionnaire,42,43 and other rele- vant data as summarized in Table 1). Randomization will be done by centralized computer assignment, ensuring that the allocation sequence is concealed from study personnel to minimize a possible allocation bias during recruitment. Block randomization by study center will be used to ensure allocation of equal numbers of subjects in each group per center. Blinding is neither applicable nor possible in this Table 1. Baseline data to be recorded. 1. WERF EPHect Questionnaire—Minimum (EPQ-M) 41 2. EHP-30 Oxford Questionnaire for QOL.42,43 3. Additional data History Age at randomization (male and female) Duration of infertility (months) Fertility work-up Male factor Moderate/severe; yes/no Date—classification of sperm sample (WHO 44) Ovulation factor Ovulatory Oligo/anovulation WHO types I–II Expected or proven poor ovarian response45 Implantation—date of ultrasound/hysteroscopy (if performed) Normal Congenital abnormality (definition ESHRE/ESGE 46) Large intramural fibroid (> 3 cm) Adenomyosis Tubal factor Normal Blocked left/right/both sides Karyotype male and female: date—results Ovarian reserve testing At randomization or maximum 6 months prior AMH (date—result—assay) FSH (date—result) Antral follicle count (date—result) Endometriosis N previous therapeutic surgeries for endometriosis N previous ovarian surgery For endometrioma left/right/both sides For non-endometriotic cyst left/right/both sides History of salpingectomy left/right/both sides Endometriotic cyst(s) Transvaginal ultrasound Date For each cyst: localization and mean diameter (mm) Current hydrosalpinx no/right/left/both sides Suspected peritoneal lesions (imaging, prior surgery): yes/no Suspected deep lesions (clinical exam, imaging): yes/no WERF: World Endometriosis Research Foundation; QOL: quality of life; ESGE: European Society of Gynaecological Endoscopy; ESHRE: European Society of Human Reproduction and Embryology; WHO: World Health Organization; AMH: anti-Müllerian hormone; FSH: follicle-stimulating hor- mone. 164 Journal of Endometriosis and Pelvic Pain Disorders 10(3) study, as neither surgery, GnRH-downregulation, nor ART can be blinded (no placebo surgery with mock incisions and no placebo medical pretreatment). However, biases to this are expected to be minimal because the primary outcome measured is clear and unambiguous. In theory, an influence on retention of participants in the trial may be present as a

of communication of personal ideas of health-care providers involved after patients have been randomized. To monitor this, both the number of screened/eligible versus randomized patients as well as the number of post-randomi- zation drop-outs versus randomized patients will be recorded per center. Study population The study population includes women with at least one ovar- ian endometriotic cyst with a diameter of at least 30 mm and who have an indication for treatment with ART. Inclusion criteria Indication for ART: based on existing guidelines 8,34,36 and can be summarized as follows: if tubal function is compromised, if there is male factor infertility, and/or other treatments have failed and/or prolonged infertility of more than 4 years. Age: from 18 until 40 years included at randomization (as from 40 years onwards, baseline success rates of ART are considerably lower and will therefore make it more difficult to observe a difference between both strategies). Ultrasound diagnosis of at least one ovarian endometri- otic cyst with a diameter of 30 mm or more, performed by an experienced practitioner using advanced ultrasound equipment. All patients in the WERF EndoART study must fit the International Ovarian Tumor Analysis (IOTA)-criteria for reliable diagnosis of endometriomas in premenopausal women 47 (positive predictive value (PPV) of 88.6%, sensitivity of 67.9%, and specificity of 97.8%): ground glass echogenicity of the cyst fluid, one to four locules, and no papillations with detectable blood flow (alternatively when no Doppler used: no solid parts). If there is doubt about the differential diagnosis with a functional ovarian (corpus luteum) cyst, a new ultrasound is performed to confirm the diagnosis of an ovarian endo- metriotic cyst 6–8 weeks after the initial ultrasound. The cyst diameter is the mean (in millimeter) of diameters measured in the three dimensions of the cyst. Diagnosis and ultrasound assessment of the cyst should be per - formed maximally 3 months prior to randomization. Exclusion criteria Recent therapeutic surgery for ovarian endometriotic cysts defined as excisional or ablative surgery of ovar - ian endometriotic cyst(s) with or without excision of all visible endometriosis and endometriosis-related adhe- sions in a period of 6 months before randomization. Women with significant pelvic pain requiring elective surgery, who are therefore unable to undergo direct ART or ART with ultralong GnRH-a downregulation. ART with sperm derived from testicular biopsy for non- obstructive azoospermia because of lower observed ART pregnancy rates compared to obstructive azoo- spermia. 48 The use of donor sperm is allowed but should be recorded. The use of donor oocytes. ART with preimplantation genetic testing (formerly preimplantation genetic screening (PGS) or preimplan- tation genetic diagnosis (PGD), viz. screening or diag- nosis), 1 as the number of embryos suitable for transfer or cryopreservation is significantly lower compared to normal ART. 49 History of unilateral oophorectomy or having an ovarian remnant, as in this case in clinical practice the threshold for surgery on the remaining ovary will be higher, mak- ing it unsuitable for randomization. Subject withdrawal. Withdrawal is defined as women leav- ing the study after they have been randomized and does not include mere screening failures. Women should not be randomized until they have been screened and found to meet the inclusion criteria. Following reasons will lead to subject withdrawal: Women allocated to the Surgery Group whose histologi- cal tissue diagnosis at surgery did not confirm the pres- ence of an endometrioma (e.g. corpus luteum cyst). Women allocated to the Direct ART Group or Ultra Long Group undergoing ovarian surgery between rand- omization and the completion of the study period. Women allocated to the Surgery Group who receive ART treatment before they undergo surgery. Participation in any other drug study during the dura- tion of the study. Withdrawal of consent. Other changes in medical condition that precludes con- tinuation in the study: must be clearly documented and medically justifiable. Study procedures The study starts at the moment of randomization, and the patient ends the study when CPHB is diagnosed within 6 months after initiation of the fresh ART cycle (i.e. fresh + associated frozen cycles) or when after one fresh Tomassetti et al. 165 cycle and its associated frozen embryo transfer (FET) cycles, no CPHB has been detected (until all frozen embryos have been used unless this follow-up time exceeds 12 months after the initiated fresh cycle). Per patient, only one fully completed ART cycle (fresh + associated frozen cycles within above-mentioned time frame) is performed within the study; in case an initiated fresh cycle is canceled, any subsequently initiated fresh cycles fall outside the scope of this study. An initiated fresh cycle is defined as follows: an ART cycle in which the woman received specific medication for ovarian stimulation (i.e. start of gonadotropin administra- tion) with the intention to treat, irrespective of whether or not follicular aspiration is attempted. 1 All women, who achieved a CPHB within the study period, will be con- tacted 8–12 months later for follow-up on further preg- nancy outcome including live birth. ART treatment (all groups). To avoid treatment bias, it is important that treatment with ART is standardized as described in this section. All women will be treated with ART according to routine practices in the participating centers. Table 2 shows which ART procedures and varia- bles need to be recorded during this study. Use of GnRH-agonists. The use of cyclic or continu- ous use of oral contraceptives or progestogens is allowed before the initiation of the fresh ART cycle and in between subsequent frozen cycles, but has to be recorded. In both the Direct ART Group and the Surgery Group, downregu- lation with GnRH-agonist should be used in a classical way and not last more than 28 days. More specifically, women randomized to the Direct ART Group or the Surgery Group are not allowed to use GnRH-a for treatment of endome- triosis in the 3 months prior to the start of the fresh ART cycle. Ultrasound-guided aspiration of persistent func- tional ovarian cysts before or during ovarian stimulation is only allowed if these cysts are clearly echolucent and unilocular and completely lack the typical endometrioma characteristics; aspiration of the endometrioma(s) is not allowed. Between the end of the fresh ART cycle and the end of the study period (in case frozen-thawed cycles are planned), the use of GnRH-agonists is not allowed except as part of a hormonal replacement preparation for endome- trium in a frozen-thawed cycle. Ovarian stimulation protocols. For both fresh and fro- zen-thawed cycles, different stimulation protocols are allowed; for the fresh ART cycle, natural cycle or intended mild stimulation (= aiming at maximum 5 oocytes) is not allowed. The type of gonadotropin (highly purified, uri- nary, or recombinant) can be freely chosen by each center, as these products are equally effective with respect to clin- ical pregnancy rate. 50,51 As block randomization per center will be used, it is expected that there will be no signifi- cant differences in ART protocols used in the three study groups. Oocyte retrieval. Oocyte retrievals (all study groups) will be performed under intravenous antibiotic prophy- laxis (cefazolin 2 g IV single dose or equal type and dose of other antibiotic) to prevent infection. 52 During oocyte retrieval, ultrasound-guided puncture of the endometrioma should be avoided in order to prevent infection and abscess formation; in case this accidentally happens, it must be recorded and the content of the fluid must be sent for cyto- logical examination. The cyst must not be drained if an accidental puncture happens. IVF and embryo transfer. Fertilization will be performed using IVF or ICSI based on the routine practice of each center. Embryo culture is done using commercially avail- able culture media. The main standardizing restriction in the ART treatment protocol is the embryo transfer policy (cleavage or blasto- cyst stage, as per center routine practice): •• A maximum of two embryos in women of 36 years old or younger; •• A maximum of three embryos in women of 37 until 39 years old inclusive; •• A maximum of four embryos in women of 40 years onwards. Luteal phase support. Luteal phase support regimen and duration are chosen freely by each center according to their standard practice, but should be documented. A new egg aspiration cycle (i.e. after the end of the study period) is only initiated when all cryopreserved embryos have been used. Direct ART group. Women in the Direct ART Group will initiate their fresh ART cycle, without any intervention for the endometrioma, within 6 months of randomization. Obviously, prolonged downregulation with GnRH-a (other than the short-term downregulation (maximum of 28 days) as in a classical agonist protocol) within 3 months before the start of ART is not allowed in this group, as described above. In case women are on GnRH-a treatment at the time of randomization, this means the use of GnRH needs to be discontinued and followed by a wash-out period of at least 3 months before they can start ART. Surgery group. To minimize post-randomization withdrawal of subjects, women in the Surgery Group need to receive surgery within 2 months after randomization and initiate their fresh ART cycle within 6 months after surgery. Pri- marily, a laparoscopic approach should be used (state-of- the-art, to enhance patient recovery), while recording any conversions to laparotomy. The technique used for 166 Journal of Endometriosis and Pelvic Pain Disorders 10(3) Table 2. ART procedures and variables. Pretreatment None (natural cycle prior to ART) Combined oral contraceptives (COC; generic name, start + end date) Progestogens only (POP; generic name, start + end date) GnRH-a (generic name, injection dates, and only in Ultra-Long Group) Ovarian stimulation Type protocol Long agonist (generic name, date start agonist) Short agonist (generic name, date start agonist) Antagonist (generic name, date start antagonist) Clomiphene citrate + gonadotropin + antagonist (generic name, date start antagonist) Gonadotropin: Type, generic name, start date Start dose (min 75–max 450 IU/day), highest daily dose (IU) Total dose used (IU)a OHSS (ovarian hyperstimulation syndrome) risk present: yes/no Coasting performed—yes/no, if yes: N days coasting Estradiol and progesterone level on day of decision: LH/HCG trigger (mg/dL) Endometrial thickness (mm) on day of decision: LH/HCG trigger N mature follicles (>13 mm) on day of decision: LH/HCG trigger LH/HCG trigger: generic name, dose, and date Oocyte aspiration Cancelation: yes/no If yes, state reason why (OHSS risk, insufficient ovarian response, premature LH/progesterone rise, other medical reasons, non-medical reasons) Age male and female at cancelation or aspiration Antibiotic prophylaxis: generic name, dose, and type of administration Date of aspiration, N hours between LH/HCG and aspiration Repeat US measurement of endometrioma(s) during aspiration Accidental puncture of endometrioma: yes/no If yes, result of cytology to be reported N mature follicles (>13 mm) not aspirated If >0: state reason why (endometrioma interpositioning, inaccessible ovary due to other reason, other) N follicles aspirated N oocytes obtained in total N mature oocytes a N immature oocytes—N postmature oocytes Fertilization N mature oocytes for IVF/N mature oocytes for ICSI N 2pn from IVF/N 2 pn from ICSI Fertilization rate = N 2pn/N mature oocytes for IVF/ICSIa N embryos from IVF/N embryos from ICSI Embryo transfer (fresh) Stage (cleavage, blastocyst), date of embryo transfer Total N embryos available N good quality embryos available per aspirationa/per N 2pna N embryos frozena, freezing method and date N embryos transferred N top quality embryos transferred Embryo utilization ratea: (N embryos transferred + N embryos frozen)/total N of embryos available Ultrasound guided: yes/no Luteal support (fresh and frozen-thawed) Progesterone intramuscular/oral/vaginal HCG subcutaneous/intramuscular End of luteal support (N days after LH/HCG trigger) Frozen-thawed embryo transfer Endometrial preparation Clomiphene citrate Natural cycle ± HCG trigger Gonadotropins ± HCG trigger Substitution protocol with GnRH-agonist Substitution protocol without GnRH-agonist Endometrial thickness on day of decision of thawing Date of thawing and date of embryo transfer Stage at embryo transfer, further cleaving after thawing: yes/no N embryos thawed—N embryos survived N embryos transferred Ultrasound guided: yes/no Total N embryo transfers performed (fresh + frozen) a (Continued) Tomassetti et al. 167 the surgical treatment of the endometrioma is aimed at simultaneously treating the endometrioma while minimiz- ing ovarian damage. Although direct comparison of the effect of different techniques on ovarian reserve is not pos- sible according to current literature, in this study, only those techniques associated with minimal ovarian damage are allowed. This strategy is defendable since this study focuses on ART success and not on long-term recurrence risk of the cyst(s). In case a complete cyst ablation is performed, it is mandatory to also biopsy the cyst wall to confirm the diag- nosis of endometriosis through histological examination. Any of the following surgical techniques may be used: the combined technique (stripping of the peripheral capsule and CO 2-laser ablation near the hilus 12 or complete cyst ablation with micro-bipolar, CO2-laser, or plasma energy, preferably using the lowest energy source available).53 This variation in surgical technique is allowed as there is a need for an intervention where small variability will not matter. If deemed necessary according to the size of the cyst (e.g. larger than 6 cm), a two-step laparoscopic procedure can be performed. During the first step, the cyst is drained. After this step, ovarian suppression with combined oral contra- ceptives, oral progestogens, or GnRH-agonists will be pre- scribed during 6–8 weeks before the second surgical step. During this second step, surgical treatment of the ovarian endometrioma is performed as mentioned above. Standard complete cystectomy and unipolar ablation are not allowed due to the larger anticipated effect on ovarian tissue dam- age. Not only the ovarian endometriotic cyst(s) must be treated, but also excision of all visible endometriosis and endometriosis-related adhesions must be performed. Since there is no evidence that one surgical treatment is superior to another regarding peritoneal endometriosis (resection or destruction/coagulation), deep lesions (in case of bowel endometriosis: partial or full thickness disk excision, seg- mental bowel resection), and related adhesions, all these techniques can be used in this study but need to be reported specifically. The details of surgery will be recorded in a standard surgical form, merging information from the WERF EPHect Standard Surgical Form 54 and the CORDES Deep Endometriosis Surgical Sheet 55 as summarized in Table 3. Obviously, postoperatively prolonged downregulation with GnRH-a (other than the short-term downregulation (maximum of 28 days) as in a classical agonist ART- stimulation protocol) is not allowed in this group, as described above. In case women are on GnRH-agonist treatment at the time of randomization, this means the use of GnRH needs to be discontinued and followed by a wash-out period of at least 3 months before they can start ART. Preoperative use of GnRH-a is allowed (in case of a two-step surgical approach also between the first and the second interventions), but again in that case patients will need a wash-out period of 3 months postoperatively before they can start ART treatment. Ultra long group. Within 2 months of randomization, GnRH-a pretreatment (triptorelin, goserelin, and leupro- lide) should be started and continued during a minimum of 3 months and a maximum of 6 months prior to commence- ment of gonadotropin stimulation for ART. Add-back with low-dose estrogens, with or without progesterone, or tibo- lone is allowed but needs to be documented. Patients will start the ART treatment within 6 months of randomization. Due to the treatment allocation, the stimulation protocol will invariably be an agonist protocol. Participants flow. The participants flow within the WERF EndoART study is summarized in Figure 1. Trial monitoring and patient visits. Follow-up visits will be planned (with local principal investigator (PI) or investiga- tor) in each center and are part of routine clinical care and therefore will not add a significant burden to either patients or clinicians: One visit at screening and initiation of the trial (study information, completion of ICF, and randomization). One visit after (each) surgery for patients in the Surgery Group. Severe adverse events No ovarian response in fresh cycle (no follicle development) Inability to puncture mature follicles Significant bleeding following oocyte aspiration Requiring in-patient hospitalization ± blood transfusion ± surgical intervention Significant pelvic infection within 14 days of oocyte aspiration Raised temperature (>38°C), lower abdominal pain, biochemical signs of infection (elevated white cell count, elevated C-reactive protein) Requiring in- or out-patient antibiotic treatment Ovarian abscess ± surgical intervention OHSS requiring in-patient hospitalization ART: assisted reproductive technologies; GnRH-a: gonadotropin-releasing hormone agonists; LH: luteinizing hormone; HCG: human chorionic gonadotropin; OHSS: ovarian hyperstimulation syndrome; IVF: in vitro fertilization; ICSI: intracytoplasmic sperm injection. aSecondary outcome variables. Table 2. Continued 168 Journal of Endometriosis and Pelvic Pain Disorders 10(3) One visit after the first initiated ART cycle, regardless of its outcome (i.e. with or without oocyte aspiration or embryo transfer) and (if applicable) its associated fro- zen-thawed embryo transfer cycle, in any case at the end of the study period if no CPHB is obtained. In case the primary outcome is reached (CPHB), a study visit is optional and may be replaced by a visit to the obstetrician who will follow the pregnancy; in the latter case, a formal written report with confirmation of CPHB by ultrasound is necessary. Outcome measures Primary outcome. The primary outcome studied is CPHB. This outcome parameter is a suitable and realistic surrogate outcome.37 More specifically, the CPHB rate per rand- omized patient within 6 months after initiation of the fresh ART cycle (= per-protocol analysis) is the main outcome studied, upon which also power and sample size calcula- tions are based. As explanatory information, it will be recorded how these CPHBs were achieved (pregnancy after ART in fresh or frozen-thawed cycle, escape intrauterine insemination (IUI), and spontaneous pregnancy after cancelation of the cycle or between cycles). The CPHB should be documented by a written report in the patient record (a copy of the ultrasound image is not compulsory). Secondary outcomes. The secondary outcome measures related to reproductive outcome include CPHB rate per fully completed (= fresh + all associated frozen-thawed Table 3. Surgical procedure: variables, work-up, staging, and adverse events. 1. WERF EPHect Standard Surgical Form (EPHect SSF)54 2. If deep endometriosis is present: CORDES Deep Endometriosis Surgical Sheet55 3. Other information to be recorded: General information Surgeon Date of (second step) surgery Preoperative work-up Hormonal treatment between randomization and surgery None (natural cycle prior to ART) COC (generic name, start + end date) POP (generic name, start + end date) GnRH-a (generic name, injection dates) Presumed diagnosis apart from endometrioma(s): Peritoneal endometriosis: yes/no Deep endometriosis lesion: yes/no Surgical procedure summary Endometrioma surgery Two-step procedure: yes/no If yes: date of first intervention (marsupialization) Per cyst: describe side and diameter and technique used: Ablation only or combined resection/ablation near hilus Energy source Ultrasound Bipolar (unipolar ablation not allowed!) Laser CO 2 Other laser Plasma energy Ovarian reconstruction (suture used) Ovariectomy performed: yes/no Picture of each ovary before and after intervention Histology result: endometrioma confirmed: yes/no Salpingectomy: no/left/right/both sides Non-endometriotic cyst removed: yes/no (histology result) Other endometriotic lesion treatment as in CORDES Conversion to laparotomy and reason why (bleeding, bowel resection, ureter reimplantation, other) Severe adverse events Within 6 weeks after surgery Classification according to Clavien-Dindo56 Specification of the problem Bleeding Pelvic infection Rectovaginal fistula Thrombo-embolic complications Postoperative premature ovarian failure (amenorrhea and hypergonadotropic hypogonadic state) WERF: World Endometriosis Research Foundation; ART: assisted reproductive technologies; COC: combined oral contraceptives; POP: progesto- gens only; GnRH-a: gonadotropin-releasing hormone agonists. Tomassetti et al. 169 cycles) ART cycle, cumulative live birth delivery rate per initiated fresh ART cycle and per fully completed ART cycle, and CPHB rate and live birth delivery rate per patient during the study period, including also eventual expectant fertility management pregnancies or after escape-IUI (intention-to-treat analysis) with reporting of the mode of pregnancy. ART-specific outcome data for fresh and associ- ated frozen cycles, more detailed pregnancy outcomes (Table 4), and QOL and adverse events will be recorded and analyzed. Since, after randomization, women in the Surgery and Ultra Long Groups are expected to start later with their fresh ART cycle than women in the Direct ART Group, also the time to pregnancy of any CPHB detected during the study period will be reported. Next to this, the time interval between randomization and the start of the first ART cycle will also be recorded in all three groups, since the time factor (waiting period before start of ART) is a relevant variable for women. In order to quantify the effect of no intervention, sur - gery, and prolonged GnRH-agonist downregulation on ovarian function, ovarian reserve tests will be done at ran- domization (or not more than 6 months prior) and at base- line (maximum 1 week before the start of gonadotropin stimulation for the fresh ART cycle). For AMH serum level measurements, the assay used for each center will be recorded to enhance comparability. Longer term re-assess- ment (e.g. 1 year after start of ART) of ovarian reserve is currently not attempted due to a perceived higher risk of losses to follow-up in that particular case. Pelvic pain and QOL will be assessed at two time points: randomization and the end of the study. If CPHB occurs during the study period, this assessment will be done within 6 weeks of confirmation of CPHB. Visual analog scales (V AS) will be used to assess pain, with different assess- ments of dysmenorrhea (menstrual pain), dyspareunia (pain during sexual intercourse), and non-menstrual pelvic pain (NMPP). 57 The validated WERF EPHect Questionnaire— Minimum (EPQ-M)41 and the EHP-30 Oxford question- naire42,43 will be filled in at randomization and at the end of the study. For the EPQ-M, only relevant clinical parts regarding current situation will be filled in at the end of the study (questions C2–C11, C17–C24, C27–C35, D5–D7, E1–E2) to avoid duplication. Reporting of adverse events and complications will comply with the principles of Good Clinical Practice and the local regulations and according to the rules in the countries/centers where the study is con- ducted. Study-specific recording of adverse events and other secondary outcomes are in the tables describing the ART and surgical procedures. Statistical analysis Data will be analyzed using a per-protocol approach (pri- mary outcome). Similarly, an intention-to-treat analysis will be performed. Comparisons will be made using the chi-square tests for proportions and one-way analysis of variance (ANOV A) for continuous variables (if conditions are not satisfied, Kruskal–Wallis will be used). Figure 1. Summary of participants flow in the WERF EndoART. 170 Journal of Endometriosis and Pelvic Pain Disorders 10(3) Subgroup analysis per center will not be carried out as it is normal to have variability in success rates between and within centers just by chance in the number and pop- ulation (e.g. age) of subjects studied in this trial. Subgroup analysis per center can therefore lead to incorrect conclu- sions on surgical and/or ART performances of individual centers. A subgroup analysis is proposed to account for the possible effect of previous ovarian surgery for endo- metriomas 22 to compare women with and without previ- ous ovarian surgery for endometrioma. Further subgroup analysis will also be carried out according to the size and number of the ovarian endometrioma, the co-existence of deep lesions, and (in the surgery group) for the surgical technique used (CO 2-laser ablation, plasma energy abla- tion, micro-bipolar ablation, and combined CO 2-laser technique).53 An interim analysis will be performed for the primary endpoint only, to assess proven effectiveness or futility, by a designated Data and Safety Monitoring Committee (DSMC) when half of the randomized patients have been assessed on the primary outcome (the presence or absence of CPHB within 6 months after initiation of the fresh Table 4. Pregnancy outcome variables. Pregnancy diagnosis Date of HCG serum testing Positive/negative result If positive: value of serum HCG in IU/L If clinical pregnancy: date of first ultrasound confirmation If fetal heartbeat: date of first ultrasound confirmation First trimester pregnancy evolution Maximal outcome at 12 weeks Biochemical pregnancy Clinical pregnancy (CP) but ectopic (N ectopic sacs) Clinical pregnancy, no fetal heartbeat, miscarriage Clinical pregnancy, fetal heartbeat = PRIMARY OUTCOME: within 6 months of ART start Miscarriage Ongoing at 12 weeks In case of clinical pregnancy Maximal N of intrauterine and/or ectopic sacs Maximal N of intrauterine fetal heartbeats In case of multiple pregnancy N fetuses N chorion N amniotic sacs Vanishing multiple: yes/no (N vanished) Fetal reduction: yes/no (N of sacs removed) End of pregnancy Date Induced abortion: yes/no Delivery (>20 weeks): yes/no Mode of delivery: vaginal/caesarian section N of live born babies N of still born babies Birth weight and sex (of each baby) Admission to neonatal unit: yes/no (for each baby) Secondary outcomes studied CPHB rate at the end of the study period Time to CPHB: from moment of randomization, moment of surgery, and start of ovarian stimulation during the fresh ART cycle Live birth rate CP and CPHB rate Per initiated/aspirated/transferred fresh ART cycle Per frozen-thawed transferred cycle Per embryo transferred (fresh, frozen) Per completed cycle (fresh + frozen) Maternal/fetal pregnancy complications Pregnancy-induced hypertension Pre-eclampsia Eclampsia Abnormal placentation (previa/accreta/percreta) Gestational diabetes Major congenital anomaly 1 HCG: human chorionic gonadotropin; CPHB: clinical pregnancy rate with fetal heart-beat; ART: assisted reproductive technologies. Tomassetti et al. 171 cycle). The results of this analysis will not be disclosed to the investigators, unless it is necessary to stop the trial (e.g. proven effect). The study will only be stopped if signifi- cance for the primary outcome variable is reached before the anticipated number of patients have been recruited. Furthermore, the DSMC can terminate the study if in their opinion, for whatever reason, a significant number of adverse events have occurred as a result of performing the study, although unexpected severe adverse events are not anticipated since this trial compares three existing and accepted treatment strategies.

Clinical management of endometriomas prior to starting treatment with Assisted Reproductive Technology (ART) has long been a matter of debate, but so far no large (rand- omized) trials have been performed to settle firmly the dis- cussion on the ideal patient management strategy. Whereas cystectomy prior to ART has been advocated in the past, recently more evidence is emerging on the potential nega- tive effect of surgery on ovarian reserve. Next to this, pro- longed downregulation with GnRH-a prior to starting ART has been described to improve ART pregnancy rates in women with endometriosis. The WERF EndoART trial is essential to answer one of the most frequently asked ques- tions by both patients and physicians regarding endome- triosis and ART: what to do with an endometrioma in the context of ART? Therefore, this study is designed to assess whether ovarian surgery (but not complete cystectomy near the hilum) or prolonged GnRH-a downregulation

in higher pregnancy rates after ART compared to no intervention in women with endometrioma(s), by means of a parallel randomized multi-center trial. The history of this trial protocol is long: back in 2007 the study question of this trial was recognized by WERF as one of the most important knowledge gaps in endometriosis and infertility treatment and selected as a top priority project eligible for funding. Between 2008 and 2015, this protocol has been rewritten and re-edited by the authors many times, with the aim to appeal to a large group of international endometrio- sis and ART specialists and to comply with the ever evolv- ing scientific literature. Unfortunately, this study did not commence due to various reasons. First (and most important), a lack of funding and finan- cial support, despite efforts made by WERF and the authors of this article to attract industry partners. Second, to reach the sufficient power to answer the study question, there is a need for a large international consortium with extensive logistic support to organize this trial, at present not possible with funding by academic resources only. The publication of this protocol is important for future investi- gators and sponsors, as it has been thoroughly developed according to the state of the art in trial design by a large group of international experts. Authors’ note Dr. Adamson is no longer associated with Palo Alto Medical Foundation Fertility Physicians of Northern California, East Palo Alto, CA, USA. He is Clinical Professor, ACF at Stanford University and Associate Clinical Professor at University of California San Francisco, San Francisco, CA, USA. Ben-Willem Mol is now affiliated to Monash University, Melbourne, VIC, Australia. Declaration of conflicting interests The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Dr D’Hooghe reports grants from Merck Serono and Ferring; personal consultancy fees paid to him via Leuven Research and Development for consultancy work for WHO, Actavis, Astellas, Cartagenia, Bayer, Roche, and Proteomika; and clinical trial financial support from Actavis, Roche, and Proteomika, In October 2015, Dr D’Hooghe became Vice-President and Head of Global Medical Affairs Infertility for Merck Serono. He continues on a part-time basis his academic appointment as Professor of Reproductive Medicine at the University of Leuven (KU Leuven) in Belgium and as Adjunct Professor at the Department of Obstetrics and Gynecology at Yale University, New Haven, CT, USA. The work described in this paper was carried out before Dr D’Hooghe joined Merck Serono in October 2015. Funding The author(s) received no financial support for the research, authorship, and/or publication of this article.

1. Zegers-Hochschild F, Adamson GD, Dyer S, et al. The international glossary on infertility and fertility care, 2017. Fertil Steril 2017; 108(3): 393–406. 2. American Society for Reproductive Medicine. Revised American Society for Reproductive Medicine classification of endometriosis: 1996. Fertil Steril 1997; 67(5): 817–821. 3. De Ziegler D, Borghese B and Chapron C. Endometriosis and infertility: pathophysiology and management. Lancet 2010; 376(9742): 730–738. 4. Sanchez AM, Viganò P, Somigliana E, et al. The distin- guishing cellular and molecular features of the endome- triotic ovarian cyst: from pathophysiology to the potential endometrioma-mediated damage to the ovary. Hum Reprod Update 2014; 20(2): 217–230. 5. Hamdan M, Dunselman G, Li TC, et al. The impact of endo- metrioma on IVF/ICSI outcomes: a systematic review and meta-analysis. Hum Reprod Update 2015; 21(6): 809–825. 6. Barnhart K, Dunsmoor-Su R and Coutifaris C. Effect of endometriosis on in vitro fertilization. Fertil Steril 2002; 77: 1148–1155. 7. Hamdan M, Omar SZ, Dunselman G, et al. Influence of endometriosis on assisted reproductive technology out- comes: a systematic review and meta-analysis. Obstet Gynecol 2015; 125(1): 79–88. 8. Dunselman GA, Vermeulen N, Becker C, et al. ESHRE guideline: management of women with endometriosis. Hum Reprod 2014; 29(3): 400–412. 172 Journal of Endometriosis and Pelvic Pain Disorders 10(3) 9. Hart R, Hickey M, Maouris P, et al. Excisional surgery versus ablative surgery for ovarian endometriomata: a Cochrane review. Hum Reprod 2005; 20: 3000–3007. 10. Somigliana E, Berlanda N, Benaglia L, et al. Surgical exci- sion of endometriomas and ovarian reserve: a systematic review on serum antimüllerian hormone level modifica- tions. Fertil Steril 2012; 98: 1531–1538. 11. Muzii L, Di Tucci C, Di Feliciantonio M, et al. The effect of surgery for endometrioma on ovarian reserve evaluated by antral follicle count: a systematic review and meta-analysis. Hum Reprod 2014; 29: 2190–2198. 12. Donnez J, Lousse JC, Jadoul P, et al. Laparoscopic man- agement of endometriomas using a combined technique of excisional (cystectomy) and ablative surgery. Fertil Steril 2010; 94(1): 28–32. 13. Roman H, Auber M, Mokdad C, et al. Ovarian endome- trioma ablation using plasma energy versus cystectomy: a step towards better preservation of the ovarian parenchyma in women wishing to conceive. Fertil Steril 2011; 96(6): 1396–1400. 14. Georgievska J, Sapunov S, Cekovska S, et al. Effect of two laparoscopic techniques for treatment of ovarian endome- trioma on ovarian reserve. Med Arch 2015; 69(2): 88–90. 15. Giampaolino P, Bifulco G, Di Spiezio Sardo A, et al. Endometrioma size is a relevant factor in selection of the most appropriate surgical technique: a prospective rand- omized preliminary study. Eur J Obstet Gynecol Reprod Biol 2015; 195: 88–93. 16. Saito N, Yamashita Y, Okuda K, et al. Comparison of the impact of laparoscopic endometriotic cystectomy and vaporization on postoperative serum anti-Mullerian hor- mone levels. Asian J Endosc Surg 2018; 11(1): 23–29. 17. Motte I, Roman H, Clavier B, et al. In vitro fertilization outcomes after ablation of endometriomas using plasma energy: a retrospective case-control study. Gynecol Obstet Fertil 2016; 44(10): 541–547. 18. Canis M, Pouly JL, Tamburro S, et al. Ovarian response during IVF-embryo transfer cycles after laparoscopic ovar- ian cystectomy for endometriotic cysts of >3 cm in diam- eter. Hum Reprod 2001; 16: 2583–2586. 19. Nargund G, Cheng WC and Parsons J. The impact of ovar- ian cystectomy on ovarian response to stimulation during in-vitro fertilization cycles. Hum Reprod 1995; 11: 81–83. 20. Somigliana E, Vercellini P, Viganó P, et al. Should endo- metriomas be treated before IVF-ICSI cycles? Hum Reprod Update 2006; 12: 57–64. 21. Somigliana E, Infantino M, Benedetti F, et al. The pres- ence of ovarian endometriomas is associated with a reduced responsiveness to gonadotropins. Fertil Steril 2006; 86: 192–196. 22. Benaglia L, Bermejo A, Somigliana E, et al. In vitro ferti- lization outcome in women with unoperated bilateral endo- metriomas. Fertil Steril 2013; 99: 1714–1719. 23. Coccia ME, Rizzello F, Barone S, et al. Is there a criti- cal endometrioma size associated with reduced ovarian responsiveness in assisted reproduction techniques? Reprod Biomed Online 2014; 29: 259–266. 24. Filippi F, Benaglia L, Paffoni A, et al. Ovarian endometrio- mas and oocyte quality: insights from in vitro fertilization cycles. Fertil Steril 2014; 101: 988–993. 25. Demirol A, Guven S, Baykal C, et al. Effect of endome- trioma on IVF outcome: a prospective randomized study. Reprod Biomed Online 2006; 12: 639–643. 26. Surrey ES, Silverberg KM, Surrey MW, et al. Effect of prolonged gonadotropin-releasing hormone agonist therapy on the outcome of in vitro fertilization-embryo transfer in patients with endometriosis. Fertil Steril 2002; 78: 699–704. 27. Sallam HN, Garcia-Velasco JA, Dias S, et al. Long-term pituitary down-regulation before in vitro fertilization (IVF) for women with endometriosis. Cochrane Database Syst Rev 2006; Nov 25(1): CD004635. 28. Decleer W, Osmanagaoglu K, Verschueren K, et al. RCT to evaluate the influence of adjuvant medical treatment of peri- toneal endometriosis on the outcome of IVF. Hum Reprod 2016; 31(9): 2017–2023. 29. Benschop L, Farquhar C, van der Poel N, et al. Interventions for women with endometrioma prior to assisted reproduc- tive technology. Cochrane Database Syst Rev 2010; (11): CD008571. 30. Tsoumpou I, Kyrgiou M, Gelbaya T, et al. The effect of surgi- cal treatment for endometrioma on in vitro fertilization out- comes: a systematic review and meta-analysis. Fertil Steril 2009; 92: 75–87. 31. Rickes D, Nickel I, Kropf S, et al. Increased pregnancy rates after ultra long postoperative therapy with gonadotropin- releasing hormone analogy in patients with endometriosis. Fertil Steril 2002; 78: 757–762. 32. Dicker D, Goldman JA, Levy T, et al. The impact of long- term gonadotrophin-releasing hormone analogue treatment on preclinical abortion in patients with severe endometrio- sis undergoing in vitro fertilisation embryo transfer. Fertil Steril 1992; 57(3): 597–600. 33. Gelbaya TA, Gordts S, D’Hooghe TM, et al. Management of endometrioma prior to IVF: compliance with ESHRE guidelines. Reprod Biomed Online 2010; 21: 325–330. 34. Johnson NP and Hummelshoj L; World Endometriosis Society Montpellier Consortium. Consensus on current management of endometriosis. Hum Reprod 2013; 28: 1552–1568. 35. Kennedy S, Bergqvist A, Chapron C, et al; on behalf of the ESHRE Special Interest Group for Endometriosis and Endometrium Guideline Development Group. ESHRE guideline for the diagnosis and treatment of endometriosis. Hum Reprod 2005; 20(10): 2698–2704. 36. Practice Committee of the American Society for Reproductive Medicine. Endometriosis and infertility: a committee opinion. Fertil Steril 2002; 98(3): 591–598. 37. Arce JC. Methodology and clinical issues to be considered in the design of efficacy trials in assisted reproductive tech- nologies. PhD Thesis, University of Brussels, Brussels, Belgium, 3 June 2009. 38. Garcia-Velasco JA, Mahutte NG, Corona J, et al. Removal of endometriomas before in vitro fertilization does not improve fertility outcomes: a matched, case-control study. Fertil Steril 2004; 81: 1194–1197. 39. Tinkanen H and Kujansuu E. In vitro fertilization in patients with ovarian endometriomas. Acta Obstet Gynecol Scand 2000; 79: 119–122. 40. Pabuccu R, Onalan G, Goktolga U, et al. Aspiration of ovar- ian endometriomas before intracytoplasmic sperm injection. Fertil Steril 2004; 82: 705–711. Tomassetti et al. 173 41. Vitonis AF, Vincent K, Rahmioglu N, et al. World Endometriosis Research Foundation Endometriosis Phenome and Biobanking Harmonization Project: II. Clinical and covariate phenotype data collection in endome- triosis research. Fertil Steril 2014; 102: 1223–1232. 42. Jones G, Kennedy S, Barnard A, et al. Development of an endometriosis quality-of-life instrument: the endometriosis health Profile-30. Obstet Gynecol 2001; 98: 258–264. 43. Jones G, Jenkinson C and Kennedy S. Evaluating the respon- siveness of the Endometriosis Health Profile Questionnaire: the EHP-30. Qual Life Res 2004; 13(3): 705–713. 44. Cooper TG, Noonan E, von Eckardstein S, et al. World Health Organization reference values for human semen char- acteristics. Hum Reprod Update 2010; 16: 231–245. 45. Ferraretti AP, La Marca A, Fauser BC, et al. ESHRE con- sensus on the definition of ‘poor response’ to ovarian stimu- lation for in vitro fertilization: the Bologna criteria. Hum Reprod 2011; 26(7): 1616–1624. 46. Grimbizis GF, Di Spiezio Sardo A, Saravelos SH, et al. The Thessaloniki ESHRE/ESGE consensus on diagnosis of female genital anomalies. Hum Reprod 2016; 31(1): 2–7. 47. Van Holsbeke C, Van Calster B, Guerriero S, et al. Endometriomas: their ultrasound characteristics. Ultrasound Obstet Gynecol 2010; 35: 730–740. 48. Osmanagaoglu K, Vernaeve V, Kolibianakis E, et al. Cumulative delivery rates after ICSI treatment cycles with freshly retrieved testicular sperm: a 7-year follow-up study. Hum Reprod 2003; 18: 1836–1840. 49. Staessen C, Platteau P, Van Assche E, et al. Comparison of blastocyst transfer with or without preimplantation genetic diagnosis for aneuploidy screening in couples with advanced maternal age: a prospective randomized controlled trial. Hum Reprod 2004; 19: 2849–2858. 50. Van Wely M, Westergaard LW, Bossuyt PM, et al. Human menopausal gonadotropin versus recombinant follicle stimulation hormone for ovarian stimulation in assisted reproductive cycles. Cochrane Database Syst Rev 2003; (1): CD003973. 51. Ziebe S, Lundin K, Janssens R, et al; MERIT (Menotrophin vs Recombinant FSH in vitro Fertilization Trial) Group. Influence of ovarian stimulation with HP-hMG or recombi- nant FSH on embryo quality parameters in patients under- going IVF. Hum Reprod 2007; 22: 2404–2413. 52. Benaglia L, Somigliana E, Iemmello R, et al. Endometrioma and oocyte retrieval-induced pelvic abscess: a clinical concern or an exceptional complication? Fertil Steril 2008; 89(5): 1263–1266. 53. Saridogan E, Becker CM, Feki A, et al; Working group of ESGE, ESHRE, and WES. Recommendations for the sur- gical treatment of endometriosis: part 1: ovarian endome- trioma. Gynecol Surg 2017; 14: 27. 54. Becker CM, Laufer MR, Stratton P, et al. World Endometriosis Research Foundation Endometriosis Phenome and Biobanking Harmonisation Project: I. Surgical phenotype data collection in endometriosis research. Fertil Steril 2014; 102: 1213–1222. 55 Vanhie A, Meuleman C, Tomassetti C, et al. Consensus on recording deep endometriosis surgery: the CORDES state- ment. Hum Reprod 2016; 31: 1219–1223. 56. Clavien PA, Barkun J, de Oliveira ML, et al. The Clavien- Dindo classification of surgical complications: five year experience. Ann Surg 2009; 250: 187–196. 57. Vincent K, Kennedy S and Stratton P. Pain scoring in endo- metriosis: entry criteria and outcome measures for clinical trials. Report from the Art and Science of Endometriosis meeting. Fertil Steril 2010; 93: 62–67.