Kisspeptin as a predictor of miscarriage: a systematic review

Abstract Objective A highly accurate serum marker for predicting viable pregnancy needs to be developed. Recent studies have demonstrated that kisspeptin is a potential biomarker for this purpose. Methods This systematic review evaluated the available data in the literature on the role of kisspeptin as a miscarriage biomarker. A literature search was conducted in the PubMed/Medline, Embase, Web of Science, and Scopus databases using the following keywords: (kisspeptin) AND (miscarriage OR pregnancy loss OR spontaneous abortion OR reproductive failure). Results Seven case–control studies were selected for the systematic review. The included papers described the potential role of kisspeptin as a putative biomarker of pregnancy loss. Furthermore, two studies reported that changes in kisspeptin levels may be associated with unexplained infertility and low rates of embryo implantation in women undergoing assisted reproductive technology. Conclusion Kisspeptin might be used as a potential biomarker of pregnancy viability in the near future. However, studies with better evidence are needed to establish the applicability of kisspeptin as a diagnostic and prognostic tool.


Introduction
Kisspeptin is a hormone belonging to the family of peptides encoded by the KISS1/KISS1 gene that binds to G-protein-coupled receptors [1]. In humans, its synthesis occurs in two regions in the hypothalamus, the anteroventral periventricular nucleus and the infundibular nucleus. Its main function is to stimulate the production of gonadotropin-releasing hormone (GnRH), from positive and negative feedback mechanisms, controlled by steroid hormones [2].
The KISS1/KISS1 gene encodes 145-amino-acids prepropeptides, with the first 19 amino acids contributing to signaling peptides. Preprohormones are sent to the endoplasmic reticulum and then cleaved into four biologically active peptides that are distinguished by the number of their amino acids: kisspeptin-10, kisspeptin-13, kisspeptin-14, and kisspeptin-54 (52 in rodent animals). All these peptides have a C-terminal region containing an Arg-Phe-NH2 radical, which allows these peptides to bind to and fully activate Kisspeptin receptors (KISS1R) [2,3].
Outside the brain, KISS1/KISS1 gene was discovered in human nonmetastatic melanoma in 1996, implying that kisspeptin expression conferred a nonmalignant phenotype to melanoma cells. The KISS1/KISS1 gene was first isolated from human placenta cells, specifically syncytiotrophoblasts, in 2001. The KISS1/KISS1 gene can also be found in the ovaries (follicles, oocytes, and corpora lutea) [4].
The hypothalamus, pituitary, pancreas, placenta, peripheral blood leukocytes, smooth muscles of some blood vessels, testes, spleen, thymus, adrenals, and lymph nodes all express KISS1R. KISS1R is also expressed in the upper frontal gyrus, caudate nucleus, cingulate gyrus, hippocampus, medullary bridge, and hypothalamus of adult humans [4]. Kisspeptin is a neuromodulator that plays an important role in the onset of puberty, maintaining the normal reproductive function, coordinating sex steroid feedback on the reproductive axis, and gender differentiation in the brain [3].
Puberty is initiated from excitatory signals on GnRH neurons, culminating in pulsatile secretion of this hormone from the hypothalamus. This secretory pattern triggers an increase in the release of gonadotropins and sex hormones, promoting the development of secondary sexual characters, gametogenesis, and rapid growth of height [2,5]. This developmental milestone depends on environmental and genetic factors. However, it has been demonstrated that puberty does not occur without adequate interaction between kisspeptins and their receptors. Kisspeptin acts on the arcuate nucleus of the hypothalamus, modulating GnRH secretion in frequency and amplitude. Deletions or mutations in KISS1R are associated with abnormal pubertal development, such as precocious puberty [2,5,6].
In women, kisspeptin regulates several mechanisms, including follicular development, oocyte formation, ovulation, ovarian steroidogenesis, and embryonic and placental implantation [7]. In this last mechanism, kisspeptin plays a role in controlling cell invasion, altering its motility and adhesion [2]. The process of implantation depends on the degree of trophoblast infiltration into the uterine extracellular matrix. In this process, kisspeptin, along with other proinflammatory cytokines, controls the trophoblastic invasion rate by stimulating apoptosis, ensuring that placentation occurs in a controlled and sequential manner [2,3].
The KISS1/KISS1 gene is primarily expressed in villous cytotrophoblast cells during early pregnancy in humans, but not in extravillous trophoblast cells. KISS1/KISS1 expression is detected in multinucleated villous syncyotrophoblast cells as well as in the placental bed as the placenta matures [3].
The formation of the placenta begins on the 7th day after fertilization. The growth of trophoblasts and their incorporation into the decidual envelope of the uterus are critical in the development of the placenta and fetus. The invasion of trophoblasts into the uterine wall involves increased cell proliferation, decreased cell death, increased blood supply, and invasion of the surrounding tissues. This process is strictly regulated. Paracrine and autocrine factors synthesized by the trophoblast and decidual envelope are among the regulatory mechanisms that limit trophoblast invasion [8].
Kisspeptin is also involved in the induction of trophoblast invasion inhibition, along with GnRH and Tumor Necrosis Factor a (TNF-a) produced by decidual macrophages. Kisspeptin participation in placental invasion control is confirmed by decreased expression of KISS1/KISS1 and KISS1R in choriocarcinoma cells but is most pronounced in trophoblast cells during the first trimester of pregnancy [8].
Kisspeptins regulate trophoblast invasion by decreasing the activities of some matrix metalloproteinases and, as a result, the ability of trophoblast cells to migrate. Kisspeptin is also a factor that, in conjunction with other proinflammatory cytokines such as TNF-a, inhibits trophoblast growth by inducing apoptosis. Kisspeptin also inhibits neovascularization while restraining the invasive growth of the hemochorial placenta. Kisspeptin has been shown to inhibit Vascular Endothelial Growth Factor (VEGF) expression and thus control the formation of embryonic vessels as well as the growth of new vessels in the already existing vascular system [8].
There is a possibility to detect kisspeptin concentrations in the blood. Both plasma and serum are clinically used; however, assays using serum have exhibited greater sensitivity for detection as a biomarker [9]. Serum kisspeptin is present in nonpregnant patients at significantly lower levels compared with pregnant patients. Furthermore, the serum kisspeptin assay was shown to be stable, in contrast to previous studies that demonstrated its rapid degradation in serum [10].
These concentrations increase throughout pregnancy, which is probably a reflection of the increase in the number of trophoblastic cells [13]. However, this phenomenon is not observed in relation to the KISS1R expression in the placenta. During pregnancy, a higher expression of these receptors is observed in the first trimester and a decrease in the third trimester [3]. Therefore, the serum levels of kisspeptin can serve as markers of placental function and, consequently, gestational viability [13].
SAB is a gestational complication that affects about 20% of pregnancies with a relevant physical and emotional impact on the woman [3]. It is defined as spontaneous pregnancy loss before 20-24 weeks of gestation. Before the 6 th week, the main causes of miscarriage are cytogenetic abnormalities, such as trisomy. Thrombophilia, placental insufficiency, and intrauterine infection are the commonly observed etiologies later in pregnancy [14].
Gestational viability is routinely evaluated by serially measuring the beta subunit of human chorionic gonadotropin (bhCG) serum levels. However, this approach is limited by the fact that 20% of SAB are associated with high bhCG levels [14]. Thus, the development of highly accurate serum markers for predicting and/or diagnosing clinically unviable pregnancies is a possible approach [3,15]. The studies by Jayasena et al. [14] and Sullyvan-Pyke et al. [15] investigated the serum and plasma levels of kisspeptin, with adequate statistical analysis to control confounding variables, such as gestational and maternal age, indicating that the levels of this substance were significantly lower in the groups of patients who underwent SAB. A linear regression analysis conducted using clinical samples of kisspeptin and bhCG revealed that the serum levels of these two markers are positively associated with abortion in women between 6 and 10 weeks of gestation, whereas the same was not observed in women with a viable pregnancy [14].
These data indicate that kisspeptin can be considered a potential biomarker for the early diagnosis of gestational viability with possible clinical applicability [3]. However, a recent editorial argued that further studies with greater evidence are needed to establish the usefulness of kisspeptin as a diagnostic and prognostic biomarker of SAB [16].
Thus, this systematic review aimed to evaluate the current state of evidence available in the literature on the use of kisspeptin as a serum and/or plasma biomarker for predicting SAB in pregnant women using the following research question: "Is kisspeptin a biomarker predictor for SAB?"

Materials and methods
This systematic review and meta-analysis aimed to investigate the use of kisspeptin as a predictor of miscarriage. This review was conducted in accordance with the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) guidelines [17].
The studies included in this review were selected based on the population, index test, reference test, outcome, and study type (PIROS) criteria: (P) pregnant women under 20 weeks (natural conception or assisted reproduction technique), (I) plasma or serum kisspeptin levels, (R) plasma or serum bhCG levels, (O) SAB rate, and (S) peer-reviewed articles with original data (observational studies, case-control studies, and cohort studies) (Table 1) [18].

Search strategy
The search for studies was performed by two authors independently (PHAS and LGMR) in September 2021 in the following electronic databases: PubMed/Medline, Web of Science (previously known as Web of Knowledge), Scopus, and Embase. Boolean operators and the following keywords were used: (kisspeptin) AND (miscarriage OR pregnancy loss OR spontaneous abortion OR reproductive failure). The results obtained from the databases were entered in Mendeley Desktop V R reference manager version 1.19.4.

Selection (inclusion and exclusion) criteria
After the initial search of the databases, duplicate studies were excluded using Mendeley Desktop V R version 1.19.4. Subsequently, the resulting articles were exported to the Rayyan QCRI V R software for screening after reading the title and abstract. Then, the included studies were submitted to complete textual reading to determine whether the articles met the inclusion criterion: observational studies that used kisspeptin as a pregnancy loss biomarker.
These steps were carried out by two reviewers (PHAS and LGMR) independently. To ensure the quality of the process, differences were resolved by reaching a consensus. In cases where disagreements persisted, a third reviewer was requested to resolve the impasse (MBC). The bibliographic references listed in the included studies were also analyzed for the identification of articles that met the inclusion criteria and for additional citations.

Definition of outcomes
SAB was defined based on the following ultrasound criteria: when the ultrasound scan revealed a) an empty gestational sac with a mean sac diameter of >25 mm or a fetal pole with no heartbeat and a crown-rump length (CRL) >7 mm or a lack of fetal cardiac activity after previously recorded embryonic cardiac activity; b) complete SAB when the follow-up ultrasound revealed a thin endometrium; c) incomplete SAB when the follow-up ultrasound revealed retained products of conception; or d) a downstream bhCG after admission due to vaginal bleeding.
Pregnancy was confirmed by a bhCG serum level > 25 mIU/mL and embryonic cardiac activity was detected on transvaginal ultrasound.

Data extraction and quality Assessment
Data were extracted from the studies, including author names, year of study, country, sample size, characteristics of the participating groups, main results, hormonal measures, association measures, adjustment variables, and conclusions. A third reviewer (MBC) evaluated these data for accuracy and uniformity, and changes were made as needed.
All manuscripts that met the selection criteria were evaluated for their methodological quality using Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) [17]. According to QUADAS-2, four factors were considered to assess the quality of the included studies: (1) patient selection, which describes patient selection methods; (2) index text, which describes the index test and how it was conducted and interpreted; (3) reference test, which describes the reference standard and how it was conducted and interpreted; and (4) flow and timing, which describe any patients who did not receive the index tests or reference standard or who were excluded from the 2 Â 2 table and the interval and any interventions between the index tests and the reference standard. We rated the quality of the studies (low, high, and unclear) in each domain following the QUADAS-2 guidelines. The first three domains were also assessed with regard to applicability [19].
A "low"-quality score required at least one answer of "no" to any signaling question in each domain, whereas a "high"-quality score required all answers of "yes" to each domain. The "unclear" category should be used only when insufficient data are reported to permit a judgment. If a study is judged as "low" on all domains relating to bias or applicability, then it is appropriate to have an overall judgment of "low risk of bias" or "low concern regarding applicability" for that study. If a study is judged as "high" or "unclear" in one or more domains, then it may be judged as "at risk of bias" or as having "concerns regarding applicability" [19].

Literary search
The initial search yielded 2152 potentially relevant articles in the databases (PubMed/Medline, n ¼ 1,499; Embase, n ¼ 133; Web of Science, n ¼ 85; and Scopus, n ¼ 435). A total of 1790 studies were obtained by the reviewers after downloading 2152 publications into the Rayyan QCRI system and Mendeley Desktop and deleting duplicate records (n ¼ 362 publications). After a preliminary review of the titles and abstracts, 1773 papers were excluded due to animal and review studies, non-English language, and conference abstracts. A total of 17 publications were selected for full-text review; finally, 7 studies were selected for systematic review, all of which were case-control studies ( Figure 1) [7,9,14,15,[20][21][22].
The ability of kisspeptin to predict SAB was measured at different thresholds. The limit of detection of each assay ranged from 0.5 to 24 pg/mL [23]. All the assays employed in the included studies were able to detect kisspeptin-54, which was described as the dominant circulating form in human pregnancies [9,10,14,15,[20][21][22].
In the seven studies evaluated, the number of pregnant women in the sample ranged from 60 to 981 (Table 2). Five studies compared the performances of kisspeptin and bhCG as biomarkers of miscarriage [10,14,15,20,22]. Samples were obtained from participants during their first trimester. Most of the studies included patients in early pregnancy with gestational ages ranging from 6 to 14 weeks. They all measured the kisspeptin levels every week or every 2 weeks after admission.
Only 3 studies [20][21][22] included nonpregnant women. Two of these studies [21,22] included women who conceived via in vitro fertilization (IVF) and measured the kisspeptin level at a certain point of time (by the 14 th and 21 st days after embryo transfer [22] and the 12 th day after embryo transfer [21]). Furthermore, Mumtaz et al. [20] included nonpregnant women with unexplained infertility who underwent intracytoplasmatic sperm injection (ICSI). Kisspeptin serum samples were collected on the 2 nd day of the menstrual cycle and at the 14 th day after the transfer of embryo.
The obstetric outcome evaluated in the included studies was new SAB in patients with infertility (n ¼ 3 studies) [20][21][22] with spontaneous intrauterine pregnancy in the first trimester (n ¼ 4 studies) [9,10,14,15]. Five studies compared the accuracies of kisspeptin and bhCG in predicting SAB [10,14,15,21,22]. In one study, the accuracy of kisspeptin was higher than that of bhCG [14]. The accuracies of both biomarkers were similar in two studies [10,15]. However, one study found that the accuracy of the combination of the two biomarkers was higher than those of kisspeptin and bhCG alone [10]. The accuracy of kisspeptin was lower than that of bhCG in two other studies (Table 2) ( Figure 2) [21,22]. Among the studies that only compared the kisspeptin levels in cases of SAB and normal pregnancies, one found decreased kisspeptin levels among patients who had SAB compared with those with normal pregnancies [20], and another study observed similar kisspeptin levels between the groups (Table 2) [9].

Quality Assessment
The results from QUADAS-2 revealed an overall low score for risk of bias and applicability concerns for five studies [10,14,15,21,22]. Only two studies [9,20] may be judged as "at risk of bias" or as having "concerns regarding applicability." The first study has a high risk of bias with regard to patient selection [20]. The second study has two judgments "unclear" both  regarding patient selection in each domain (risk of bias and applicability) ( Table 3 and Figure 3) [9].

Discussion
In the existing literature, several studies demonstrated that serum kisspeptin levels exhibited great accuracy in discriminating miscarriage from viable intrauterine pregnancy. The most abundant kisspeptin in human circulation is kisspeptin-54; this form has also been investigated as a biomarker of pregnancy viability [14]. In our review, all the included studies evaluated the levels of kisspeptin-54 as a marker of gestational viability. Three [10,14,20] of the seven studies demonstrated that the circulating level of kisspeptin (alone or in combination with bhCG) was a superior SAB biomarker to bhCG alone [10,14,15,20]. One of these included studies also demonstrated that the hormone levels of kisspeptin in patients who had a miscarriage were lower compared with those in controls [14]. Thus, based on this qualitative analysis of the studies, kisspeptin may be a promising biomarker of gestational viability in the early stages of spontaneous pregnancies.
Interestingly, the two studies that found kisspeptin to have lower accuracy than bhCG in predicting SAB evaluated pregnancies obtained via IVF/ICSI with fresh transfer or vitrified-warmed embryo transfer [21,22]. Furthermore, blood collection was performed between 14 and 21 days after embryo transfer, which was much earlier than in the other included articles, a period that may not correspond to the ideal plasma levels for its measurement [21,22].
Routinely, serial bhCG assessment is employed to establish the viability of early pregnancy. A serum bhCG level that doubles in 48 h after initial dosing is related to a lower risk of pregnancy loss [23]. The included studies did not have a standardized sample collection; some performed serial measurements of kisspeptin, whereas others only performed an isolated dose. It is important to know the rate at which kisspeptin levels increase in a normal pregnancy. Unlike the bhCG curve, in which there is a peak in its concentration during the first trimester, to subsequently decline, kisspeptin concentrations rise throughout pregnancy [24].
Another bias in the evaluation of the studies included in this systematic review is the lack of   standardization of laboratory tests for the detection of kisspeptin and its circulating isoforms. It is known that both plasma and serum can be used; however, the detection of kisspeptin as a biomarker is more sensitive in serum, and the assays employed for its detection are more stable [25][26][27]. The detection of kisspeptin-54 was described in the methodology of all the included studies, but there is no clear description of the detection of the other isoforms (À10, À13, and À14). Furthermore, kisspeptin has been studied as a predictor of success in fertility treatments, such as IVF/ICSI, and as a probable cause for cases of unexplained infertility, which corresponds to about 30% of couples with difficulty in conceiving [19]. For these couples, IVF and ICSI are proposed as the main treatments; however, they have a low success rate in those couples compared to those with an identified cause of infertility. This suggests that other factors may be involved, even in the absence of abnormalities [28].
The success of pregnancy is related to the development of immune tolerance mechanisms that prevent syncytiotrophoblast rejection during the implantation and invasion processes. The immune tolerance mechanisms of pregnancy involve the following: a) induction of the enzyme indolamine-2,3dioxygenase by the antigen-presenting cell; b) increase in the number of T-regulatory cells that suppress the immune response against the conceptus; c) decrease in the number of T-helper lymphocytes, which produce interleukin 17, and decrease in Th1 cells that stimulate cytotoxic reactions; and d) dominance of the Th2 response, protective to the conceptus. Kisspeptin has a proinflammatory activity and can regulate trophoblastic invasion, which leads to successful implantation [28][29][30].
KISS1R is expressed in most cells of the immune system; it is suggested that this hormone can jointly modulate these mechanisms of immune tolerance. Furthermore, kisspeptin is produced by trophoblastic cells and may exert a systemic effect on the cells of the immune system. The gestational levels of kisspeptin increase the production of regulatory T-cells and inhibit the differentiation of T-helper lymphocytes, thus decreasing the production of interleukin 17, which, in turn, acts by determining the suppressive potential of T-regulatory cells [30]. In general, the main objective of the immunological control of kisspeptin during pregnancy is to inhibit cytotoxic reactions, favoring the Th2-type response and increasing the population of T-regulatory cells [30].
The study conducted by Mumtaz et al. [20], included in this review, recruited 124 women with different causes of infertility and aimed to correlate the baseline kisspeptin levels with implantation failure, risk of miscarriage, and pregnancy after undergoing ICSI.  Patients with unexplained infertility who underwent ICSI had the lowest kisspeptin levels and the worst rate of embryo implantation. The strength of this systematic review is to debate the use of kisspeptin as a potential predictor of early pregnancy loss. In addition, to understand the changes in kisspeptin levels in infertility cases and patients undergoing IVF cycles, which may be associated with unexplained infertility and lower rates of embryo implantation in women undergoing IVF/ICSI. On the other hand, the main limitation is the quality and availability of data from studies already published on this topic. Unfortunately, the included studies proved to be very heterogeneous, making it difficult to conduct a meta-analysis and construct a summary receiver operating characteristic curve, not allowing a general analysis of the sensitivity and/or specificity of kisspeptin as a biomarker. This is partly due to the lack of standardization of studies with regard to the kit used to measure this hormone, the appropriate gestational age for its measurement, and the collection frequency of samples obtained from the included participants.

Conclusion
In conclusion, kisspeptin appears as a new biomarker candidate in naturally conceived pregnancies, with similar or superior accuracy to bhCG. The accuracy of kisspeptin as a predictor of SAB in pregnancies resulting from in vitro fertilization/intracytoplasmatic sperm injection was lower than that of bhCG. Thus, new studies with rigorous methodological designs are warranted, with a focus on the standardization of the collection period, the type of assay to be used, and the group of patients evaluated.

Disclosure statement
No potential conflict of interest was reported by the author(s).

Funding
The author(s) reported there is no funding associated with the work featured in this article.