Efficacy and safety of oral gonadotropin-releasing hormone antagonists in moderate-to-severe endometriosis-associated pain: a systematic review and network meta-analysis

Purpose The aim of this NMA is to comprehensively analyze evidence of oral GnRH antagonist in the treatment of moderate-to-severe endometriosis-associated pain. Methods Literature searching was performed to select eligible studies published prior to April 2022 in PubMed, Cochrane, Embase and Web of Science. Randomized controlled trials involving patients who suffered from moderate-to-severe endometriosis-associated pain and treated with oral nonpeptide GnRH antagonists or placebo were included. Results Elagolix 400 mg and ASP1707 15 mg were most efficient in reducing pelvic pain, dysmenorrhea and dyspareunia. Relugolix 40 mg was best in reducing the analgesics use. The rates of any TEAEs and TEAEs-related discontinuation were highest in relugolix 40 mg and elagolix 250 mg, respectively, while rates of hot flush and headache were highest in relugolix 40 mg and elagolix 150 mg. Significantly decreased spinal BMD was observed in elagolix 250 mg. Conclusion Oral GnRH antagonists were effective in endometriosis-associated pain in 12w, and most of the efficiency and safety outcomes were expressed in a dose-dependent manner, but linzagolix 75 mg was an exception.

and progestogens. However, NSAIDs are often ineffective and cause treatment-associated adverse effects [5], while COCs and progestogens are prone to cause bothersome side effects, such as weight gain, mood swings, and irregular uterine bleeding, which lead to drug interruption. Moreover, 25-33% patients are primarily resistant to COCs and progestogens [6,7]. For the second-line drug therapy, injectable depot formulations of gonadotropin-releasing hormone (GnRH) agonists are effective in managing endometriosis-associated pain. Nevertheless, flareup effects and hypoestrogenic adverse events relating to complete estrogen suppression limit the long-term use of injectable GnRH agonists [8].
Oral GnRH antagonists are oral short-acting treatments for endometriosis-associated pain. They inhibit the secretion of estrogen in a dose-dependent manner without flareup effects, and rapid reverse of estrogen-suppression effects can be achieved shortly after drug withdrawn. Thus, it is convenient to tailor dosage to balance efficacy and safety [9]. Currently, oral GnRH antagonists, including elagolix and relugolix, have been approved for endometriosis by FDA [10] (https:// www. conte mpora ryobg yn. net/ view/ fda-appro ves-myfem bree-forendom etrio sis-pain), while treating EAP with Linzagolix and ASP1707 has being evaluated in ongoing clinical trials [11,12]. However, there is lack of comprehensive comparison on efficiency and safety of different oral GnRH antagonists.
In the present study, we conducted systematic review and network meta-analysis (NMA) to evaluate the efficacy and safety of oral GnRH antagonists in treating endometriosisassociated pain.

Methods
This study was conducted according to guidelines of Preferred Reporting Items for Systematic Reviews Incorporating Network Meta-analyses of Health Care Interventions with minor modification [13].

Literature searching
Literature searching was performed to select eligible studies published prior to April 2022 in the following electronic databases: PubMed, Cochrane, Embase and Web of Science. The following combined relevant Medical Subject Heading terms and keywords were used: "elagolix" [All Fields] and "endometriosis" [All Fields], "relugolix" [All Fields] and "endometriosis" [All Fields], "linzagolix" [All Fields] and "endometriosis" [All Fields], "ASP1707" [All Fields] and "endometriosis" [All Fields], and "opigolix" [All Fields] and "endometriosis" [All Fields]. The searching was confined to English language and human studies. Following the searching, duplicate studies were removed by Endnote X7 for windows. The remaining studies were manually screened to identify additional potential studies by two independent authors (Q.X.H. and L.L.X.).

Inclusion and exclusion criteria
The candidate studies should satisfy the following inclusion criteria: 1. randomized control studies (RCT); 2. studies involving patients suffering from moderate-to-severe endometriosis-associated pain; 3. studies comparing placebo with oral non-peptide GnRH antagonists without addback; 4. human studies published in English; 5. studies reporting any of the following outcomes for 12 weeks: numeric rating score (NRS) of pelvic pain, modified Biberoglu and Behrman (M-B&B) score of dysmenorrhea, M-B&B score of dyspareunia, percentage of days using analgesics, rate of any grade treatment-emergent adverse effects (TEAEs), rate of TEAEs leading to treatment discontinuation, rate of hot flush, rate of headache, spinal and femoral bone mineral density (BMD); and 6. available full text. Studies met following criteria were excluded: 1. review studies, comments, letters, meta-analysis; 2. studies involving peptide GnRH antagonists; 3. treatment with oral non-peptide GnRH antagonists and other pharmaceuticals.

Data extraction
The data extraction was performed by two independent authors (Q.X.H. and L.L.X.). The following information was extracted: the name of first author, year of publication, country, study design, sample size, age of the patients, treatment arms, duration of follow-up, pain-related outcomes, and safety-related outcomes. For pain-related outcomes, results of change of NRS of pelvic pain, M-B&B score of dysmenorrhea, M-B&B score of dyspareunia, and use of analgesics were collected. For safety-related outcomes, we extracted data of rate of any TEAEs, treatment discontinuation led by TEAE, hot flush, headache, percentage change of spinal and femoral BMD from baseline. If there were more than one study from one cohort with identical outcomes, the more comprehensive study would be included. When the complete data for quantitative synthesis was unavailable, we turn to the correspondence author for full data by email.

Quality assessment
The quality of eligible studies was assessed by the Cochrane Risk of Bias tool. There are seven components included in the qualification, consisting of random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome, selective reporting and other biases. In each component, the judgment is categorized as low risk of bias, unclear risk of bias or high risk of bias [14].

Statistics analysis
NMA was implemented with R software version 4.1.0 for windows. Heterogeneity across studies was evaluated with P value and I 2 ; P value > 0.1 and I 2 < 50% indicated low heterogeneity and a fixed-effects model was applied; P value < 0.1 and I 2 > 50% indicated significant heterogeneity and a random-effects model was applied. Polled continuous variables were expressed as mean difference (MD) with 95% confidence interval (CI), and pooled dichotomous variables were expressed as relative ratio (RR) with 95% CI. When the 95% CI did not include 0 for MD and 1 for RR, it was considered significantly different. The efficacy of treatments was ranked according to P score. A larger P score indicated worse pain-related outcomes, higher incidence of adverse effects or higher BMD.

Quality assessment results
All the 6 included studies were double-blind RCT with randomization, and allocation concealment and blinding were well implemented. There were no incomplete outcome, selective reporting and other biases in the 6 studies. The risk of bias was assessed as low risk (Fig. 2).    (Fig. 3B).
For rate of treatment discontinuation led by TEAEs, the data was reported by 3 studies involving 688 women in 10 cohorts. The results showed that the rates of treatment discontinuation in patients treated with elagolix 150 mg and elagolix 250 mg were remarkably higher than those receiving other treatments (RR = 2.1e + 4, 95% CI  (Fig. 4C).
Rate of headache was analyzed in 5 studies involving 1452 women in 20 cohorts. The highest rate of headache was found in patients receiving elagolix 150 mg (P score = 0.93), followed by that in patients receiving elagolix 250 mg (P score = 0.81) (Fig. 4D).  with elagolix 250 mg (P score = 0.08) (Fig. 4E). For percentage change in femoral BMD, no significant difference was found in patients treated with elagolix 150 mg or 250 mg comparing with those treated with placebo (Fig. 4F).

Heterogeneity and inconsistency
There was no significant heterogeneity across studies in all quantitative analysis.

Discussion
Oral non-peptide GnRH antagonists are novel treatment options for endometriosis-associated pain [20]. However, evidence of direct comparison among different oral nonpeptide GnRH antagonists was lacking. In this NMA, we assessed 6 studies with 2732 women that compared varying types and dosage of oral non-peptide GnRH antagonists with placebo in treating moderate-to-severe endometriosis-associated pain. For pain-related outcomes, almost all oral non-peptide GnRH antagonists were effective, except for linzagolix 50 mg and ASP1707 3 mg. For safety-related outcomes, most of the oral non-peptide GnRH antagonists brought about more adverse effects than placebo.
Elagolix is the first oral GnRH antagonists approved by FDA for the management of endometriosis-associated pain [21]. In our NMA, elagolix 400 mg was the most effective in managing pelvic pain and dyspareunia. In consistent with our results, elagolix 400 mg was recommended in patients with co-existing dyspareunia [10]. We also found that lower dose of elagolix (250 mg) could ameliorate pelvic pain significantly and remarkable reduced analgesics use. Moreover, elagolix 150 mg has been proposed to long-term use in treating endometriosis-associate pain [22]. Consistently, in our study, significant reductions in pelvic pain and dyspareunia were achieved in patients treated with elagolix 150 mg, though more analgesics were used compared with placebo. Thus, the effect of analgesics on pain controlling could not be ruled out, and more evidence is needed. For safety-outcomes, both dosages (250 mg and 150 mg) of elagolix increased the incidence of headache, one of the most common adverse effects reported by previous study leading to treatment discontinuation [22], which suggested a higher probability of TEAEs when receiving elagolix treatment.
Relugolix is an oral GnRH antagonists approved by FDA in uterine fibroids in 2019 [23], and its combination tablets were approved for endometriosis in 2022. In treating endometriosis-associated dysmenorrhea, we found that the efficacy of relugolix 40 mg ranked the second, and a dose-dependent reduction of analgesics use was detected. In the safety assessment, the rate of hot flush in patients treated with relugolix 40 mg was the highest, while treatment discontinuation due to TEAEs was similar with placebo. Additionally, previous study has reported that longterm application of relugolix 40 mg was well tolerable [24]. Unexpectedly, relugolix showed little effect in treating dyspareunia, regardless of the dosage. We speculated this might be due to the small sample size and less sexual intercourse due to pain, future study with larger sample size is needed.
ASP1707, which is developed for the treating endometriosis and rheumatoid arthritis by Astellas Pharma [25], was in the leading position in relieving dysmenorrhea and dyspareunia in our analysis, which supports the potential benefit of ASP1707. Our results also showed that the efficacy and adverse effects of ASP1707 were in a dose-dependent manner. Due to limited clinical trials, the optimal strategy of ASP1707 needs to be explored further.
Linzagolix is a novel type of oral GnRH antagonist and is currently in late experimental clinical trial [11]. In managing overall pelvic pain, linzagolix 75 mg was superior to linzagolix 200 mg in this NMA. For safety outcomes, the TEAEs of linzagolix were dose-dependent, which is that the rate of TEAEs was the highest in patients receiving linzagolix 200 mg, and the rate was at a medium level without significant difference comparing with placebo in patients treated with linzagolix 75 mg. Taken together, linzagolix 75 mg might be the optimal strategy in 12w.
In the treatment of ASP1707 and linzagolix, our results showed that a higher dose sometimes gains a worse effect in terms of pain in 12w. We speculated these might be due to following reasons: Firstly, the patient-reported outcomes used to assess pain were subjective; secondly, we could only obtain data at 12w, which was not long enough to obtain obvious pain relief; moreover, studies demonstrated that with the extension of the treatment duration; the effect of pain relief showed a dose-dependent manner [12,18]. Taken together, more objective pain assessment methods and longterm treatments need to be explored in the future.
In the present analysis, the spinal and femoral BMD of patients treated with elagolix were assessed. We found spinal BMD decreased more significantly in patients treated with elagolix 250 mg than those with elagolix 150 mg. Decreased BMD was considered to be a key factor constrained the longterm use of oral GnRH antagonists [26]. Nevertheless, a long-term study concluded that treatment with elagolix had minimal impact on BMD over a 24-week period [27]. Furthermore, our results showed that the femoral BMD in patients treated with elagolix was increased. The authors inferred that the different population enrolled might explain the increased femoral BMD partially [19]. Therefore, multicenter and multination RCTs covering different races should be implemented to determine the role of elagolix or oral GnRH antagonist on BMD.
Several limitations should be noted in our NMA. Firstly, limited number of eligible studies may constrain the confidence of our findings. Secondly, the sample size of some included studies is relatively small. Thirdly, the population was restrained to few nations, and data of long-term effects, headache before treatment and sexual activity could not be obtained. Last but not least, unified tools for outcome measurement such as pain and BMD should be adopted to obtain more objective evaluation.

Conclusion
In the present NMA, our findings indicated that oral GnRH antagonists were effective in treating endometriosis-associated pain in 12w, and the efficacy and safety of oral GnRH antagonists were dose-dependent. Except for linzagolix 75 mg, high dose of oral GnRH antagonists was favorable. Multicenter and multination RCTs with larger sample size and variety of races were urgently needed in the future.
Author contributions L-lX performed literature searching and manuscript writing; Y-hM performed data extraction and manuscript editing; MY carried out data extraction; LC and F-zL have done quality assessment; Q-xH contributed to project development and literature searching.
Funding The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.
Data availability Not applicable.

Conflict of interest
The authors have no relevant financial or non-financial interests to disclose.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.