Although cancer remains a public health problem worldwide, survival has significantly improved over the past years thanks to major advances in both early detection and use of more effective anticancer treatments . Progress has been especially rapid for hematological malignancies but has also concerned the majority of solid tumors . Therefore, nowadays, addresing survivorship issues as early as possible is of crucial importance to avoid the potential serious long-term consequences of anticancer treatments . On this regard, chemotherapy-induced gonadotoxicity is of particular concern for newly diagnosed premenopausal patients being associated with negative side effects including menopause-related symptoms and other negative sequelae such as the possible risk of infertility . Therefore, at the time of diagnosis, all premenopausal women should be informed about the potential gonadotoxicity of chemotherapy as well as on the strategies available to counteract the risk of developing this side effect and its negative consequences [4–8].
Although cancer remains a public health problem worldwide, survival has significantly improved over the past years thanks to major advances in both early detection and use of more effective anticancer treatments . Progress has been especially rapid for hematological malignancies but has also concerned the majority of solid tumors . Therefore, nowadays, addressing survivorship issues as early as possible is of crucial importance to avoid the potential serious long-term consequences of anticancer treatments . On this regard, chemotherapy-induced gonadotoxicity is of particular concern for newly diagnosed premenopausal patients being associated with negative side effects including menopause-related symptoms and other negative sequelae such as the possible risk of infertility . Therefore, at the time of diagnosis, all premenopausal women should be informed about the potential gonadotoxicity of chemotherapy as well as on the strategies available to counteract the risk of developing this side effect and its negative consequences [4–8].
Temporary ovarian suppression by using a gonadotropin-releasing hormone agonist (GnRHa) during chemotherapy has been specifically developed as a medical intervention for trying to counteract chemotherapy-induced gonadotoxicity. So far, this remains the only option for protecting gonadal function during systemic cytotoxic treatment that has been tested in humans . Over the past 30 years, despite conducting several research efforts in both the preclinical and clinical settings, the role of this strategy has remained highly debated in the literature [10–13]. Nevertheless, recent large studies have better clarified the efficacy and safety of temporary ovarian suppression with GnRHa during chemotherapy. Therefore, while it should not replace well-established fertility preservation procedures, GnRHa administration is now recommended for clinical use as an option to reduce the risk of chemotherapy-induced gonadotoxicity particularly in breast cancer patients [6, 8, 14].
In this chapter, we review the existing evidence on the use of temporary ovarian suppression with GnRHa during chemotherapy in cancer patients from its biological rationale to the available preclinical and clinical available data. In addition, we highlight the still-existing gray zones requiring further research efforts in this field.
According to the two main hypotheses, the potential mechanism of action for the protective gonadal effect of temporary ovarian suppression with GnRHa during chemotherapy may be elicited through indirect and/or direct effects on the ovaries. However, both hypotheses have not been fully validated and their mechanism of action remains to be clarified.
The first hypothesized indirect effect is related to the GnRHa-induced prepubertal hormonal environment by suppressing the activity of the hypothalamic–pituitary–gonadal axis. In this scenario, it is expected that the follicles would be maintained at the quiescent stage thus being less vulnerable to chemotherapy-induced gonadotoxicity . Some studies in rodents have supported this hypothesis of a potential protective effect of GnRHa in quiescent follicles [16, 17]. However, it has been seriously challenged as the recruitment of quiescent follicles at the primordial stage is gonadotropin-independent and occurs irrespectively of follicle-stimulating hormone (FSH) levels . A recent study in mice even suggested a protective effect of gonadotrophins, specifically the luteinizing hormone (LH), against cisplatin-induced oocytes apoptosis .
On the other hand, FSH suppression obtained with GnRHa administration may protect against chemotherapy-induced gonadotoxicity by maintaining the quiescence of the early growing follicles through the inhibition of hypothalamic–pituitary–gonadal axis . Specifically, by maintaining FSH at a low level, the proliferation rate of follicular cells in growing follicles is reduced with subsequent indirect prevention of an accelerated recruitment of the quiescent follicular pool . Indeed, the anti-Müllerian hormone (AMH) secreted by the growing follicles can negatively regulate the recruitment of primordial follicles [21, 22]. By damaging the growing follicles, gonadotoxic chemotherapy is associated with a rapid and dramatic reduction in AMH levels leading to a massive recruitment of primordial follicles into the growing pool (i.e., “burnout effect” of chemotherapy) . Through the protection of the early growing follicles that secrete AMH, GnRHa administration may limit the burnout effect by reducing the recruitment of primordial follicles . Solid experimental evidence to confirm this hypothesis is still lacking.
Finally, the indirect effect of temporary ovarian suppression with GnRHa during chemotherapy can be mediated by a decrease in utero-ovarian perfusion with subsequent possible reduced exposure of the follicles to chemotherapy . Estrogen-induced increase of ovarian perfusion and endothelial vessel area can be inhibited by GnRHa as shown in both rat  and human [26, 27] experiments, although other studies did not confirm these results [28, 29]. However, on the other hand, a decrease in ovarian perfusion may have a detrimental effect worsening chemotherapy-induced ovarian fibrosis that is one of the mechanisms responsible for treatment-associated ovarian injury [30, 31]. Both in vivo rat experiments  and in vitro culture models of human granulosa cells  showed that GnRHa administration can reduce mRNA expression of the vascular endothelial growth factor (VEGF) thus altering the neovascularization and with potential increased ovarian injury . Nevertheless, no definitive conclusions can be drawn on this regard considering the limited available data and the small sample size of the studies.
It has been hypothesized that GnRHa use during chemotherapy can also have a direct effect on the ovaries through the GnRH receptors (GnRHR) that are present in both interstitial cells and granulosa cells of ovarian follicles at different stages .
GnRHa use appears to influence folliculogenesis with an inhibitory action on immature follicles by reducing steroidogenesis and GnRHR expression as well as a stimulatory effect on mature follicles by promoting the maturation of the oocytes and follicular rupture . Moreover, GnRHa can protect the ovaries by decreasing apoptotic events and mitochondrial stress , for example, through the upregulation of anti-apoptotic molecules such as sphingosine-1-phosphate (S1P) . Nevertheless, this direct effect in the ovaries of GnRHa use during chemotherapy remains poorly understood.
Another potential direct mechanism is the possible protective action on ovarian primordial germ cells by interacting with crucial pathways implicated in cell growth/survival and primordial follicle activation after exposure to chemotherapy [37–39]. However, this remains also to be biologically proven in the ovaries.
Over the past 30 years, several in vivo experimental studies on this topic have been conducted in both female mice [17, 24] and rats [40–43]. On the contrary, more limited evidence exists on the protective effect of GnRHa use during chemotherapy in female primates or human models [44–46] (Table 15.1).
|Authors||Experimental model||Type of gonadotoxic treatment||Overall results|
|Ataya et al., 1995 ||In vivo study in rhesus monkey||Cyclophosphamide||Protection|
|Imai et al., 2007 ||In vitro study on human granulosa cells||Doxorubicin||Protection|
|Bildik et al., 2015 ||In vitro study on human granulosa cells and ovarian tissue fragments||No protection|
Overall, most of the available experimental studies support the protective gonadal effect of GnRHa use during chemotherapy. Interestingly, some studies have also suggested potential differences in its protective effect according to the type and dose of administered chemotherapy . Nevertheless, cross-study comparison remains difficult due to the significant discrepancies between the various studies that include differences in terms of experimental models, type, and doses of chemotherapy used as well as way and duration of GnRHa administration. A few important points should be considered. Notably, although both mice and rats are rodents, there are differences between these two species and compared to humans. Few experimental studies have documented the inhibitory effect of GnRHa administration on the hypothalamic–pituitary–gonadal axis . In animal experiments, ovarian reserve is usually evaluated based on follicular count at only one specific time point with very limited evidence on AMH that would be a more reliable indicator of chemotherapy-induced ovarian damage. In addition, there is lack of evidence on the long-term effects of GnRHa co-treatment. Inconclusive and limited data are available on the direct action of GnRHa in the ovaries through GnRHR; moreover, differences between species in GnRHRs localization and pathways involved in their activation should be taken into consideration in the interpretation of the different studies.
The largest amount of data from randomized trials exploring the efficacy and safety of temporary ovarian suppression with GnRHa during chemotherapy as a strategy to preserve ovarian function and potential fertility are available for premenopausal women with breast cancer (Table 15.2). A total of 14 different randomized trials have been conducted in this setting [49–63]. Globally, all but four trials showed that concurrent administration of GnRHa during chemotherapy was effective in reducing the risk of chemotherapy-induced POI; consistent results have been observed in the three largest trials (PROMISE-GIM6 [54, 61], POEMS/SWOG S0230 , Anglo Celtic Group OPTION ). More limited information is available on posttreatment pregnancies considering that this was a preplanned secondary endpoint only in the POEMS/SWOG S0230 trial  and that the majority of the other studies reported their primary endpoint (i.e., chemotherapy-induced POI) after a short follow-up.
|Authors||Type of disease||POI definition (timing)||No. patients||Overall results|
|Li M et al. 2008 ||Breast cancer||63||Protection|
|Badawy A et al. 2009 ||Breast cancer||78||Protection|
|Sverrisdottir A et al. 2009 ||Breast cancer||94||Protection|
|Gerber B et al. 2011 ||Breast cancer||60||No protection|
|Sun JB et al. 2011 ||Breast cancer||21||Protection|
|Del Mastro L et al. 2011 & Lambertini M et al. 2015 [54,61]||Breast cancer||281||Protection|
|Munster P et al. 2012 ||Breast cancer||49||No protection|
|Elgindy EA et al. 2013 ||Breast cancer||100||No protection|
|Song G et al. 2013 ||Breast cancer||183||Protection|
|Jiang FY et al. 2013 ||Breast cancer||21||Protection|
|Karimi-Zarchi M et al. 2014 ||Breast cancer||42||Protection|
|Moore HCF et al. 2015 ||Breast cancer||218||Protection|
|Leonard RCF et al. 2017 ||Breast cancer||221||Protection|
|Zhang Y et al. 2018 ||Breast cancer||216||No protection|
|Waxman JH et al. 1987 ||HL||18||No protection|
|Guiseppe et al. 2007* ||HL||29||No protection|
|Behringer K et al. 2010 ||HL||23||No protection|
|Demeestere I et al. 2013 & Demeestere I et al. 2016 [67,68]||HL and NHL||84||No protection|
|Gilani M et al. 2007 ||Ovarian cancer||30||Protection|
* The inconsistencies in methods and results pose strong doubts about the randomized nature of the study
Abbreviations: POI, premature ovarian insufficiency; FSH, follicle-stimulating hormone; E2, estradiol; HL, Hodgkin lymphoma; NHL, non-Hodgkin lymphoma; AMH, anti-Müllerian hormone.
In women with tumors other than breast cancer, more limited evidence exists on this regard (Table 15.2). The majority of the randomized trials have been conducted in women with hematological malignancies with only one study in ovarian cancer patients [64–69]. The protective effect of temporary ovarian suppression with GnRHa during chemotherapy could not be observed in any of the hematological trials with limited information reported on posttreatment pregnancies . On the contrary, the small study conducted in women with ovarian cancer showed a significant reduction in chemotherapy-induced POI with the use of GnRHa but no data on post-treatment pregnancies were reported.
To summarize the results from prior randomized trials, several meta-analyses have been performed [70–89] (Table 15.3). Globally, all but two meta-analyses showed that the concurrent use of GnRHa and chemotherapy significantly reduce the risk of chemotherapy-induced POI in premenopausal cancer patients with a larger and clearer benefit in those including only breast cancer trials. The potential protective effect of GnRHa use during chemotherapy as a strategy for fertility preservation was not observed in the oldest meta-analyses while a significantly higher pregnancy rate with the administration of GnRHa was observed in the most recent meta-analyses that included a larger number of studies and the largest trials.
|Authors||Type of disease||No. of included studies||No. of patients||Overall results|
|Clowse MEB et al. 2009 ||Autoimmune disease, HL, and NHL||9*||366||Protection|
|Ben-Aharon I et al. 2010 ||Autoimmune disease, breast cancer, HL, and NHL||16*||681|
|Kim SS et al. 2010 ||Autoimmune disease, breast cancer, HL, and NHL||11*||654||Protection|
|Bedaiwy et al. 2011 ||Breast cancer, ovarian cancer, and HL||6||340|
|Chen H et al. 2011 ||Breast cancer, ovarian cancer, and HL||4||157|
|Yang B et al. 2013 ||Breast cancer||5||528|
|Wang C et al. 2013 ||Breast cancer,||7||677||Protection|
|Zhang Y et al. 2013 ||HL, and NHL||7*||434|
|Sun X et al. 2014 ||Breast cancer, ovarian cancer, and HL||8||621|
|Del Mastro L et al. 2014 ||Breast cancer, ovarian cancer, HL, and NHL||9||765||Protection|
|Vitek WS et al. 2014 ||Breast cancer (hormone receptor-negative only)||4||252||No protection|
|Elgindy E et al. 2015 ||Breast cancer, ovarian cancer, HL, and NHL||10||907||No protection|
|Shen YW et al. 2015 ||Breast cancer||11||1,062|
|Lambertini M et al. 2015 ||Breast cancer||12||1,231|
|Munhoz RR et al. 2016 ||Breast cancer||7||856|
|Silva C et al. 2016 ||Breast cancer||7a||1,002a||Protection|
|Bai F et al. 2017 ||Breast cancer||15a||1,540a|
|Senra JC et al. 2018 ||Breast cancer, HL, and NHL||13||1,208|
|Hickman LC et al. 2018 ||Breast cancer, ovarian cancer, HL, and NHL||10||1,051||Protection|
|Lambertini M et al. 2018 ||Breast cancer||5b||873|
* Included also non-randomized studies.
a Data from the two publications from the PROMISE-GIM6 trial (Del Mastro et al. JAMA 2011 & Lambertini M et al. JAMA 2015) were considered twice instead of as from the same study.
b Meta-analysis based on individual patient-level data.
Abbreviations: RCTs, randomized controlled trials; HL, Hodgkin lymphoma; NHL, non-Hodgkin lymphoma.
Taken together, several considerations can be made considering the clinical data available on the role of temporary ovarian suppression with GnRHa during chemotherapy as a strategy to preserve ovarian function and potential fertility in premenopausal patients.
First, while the results on the efficacy of this strategy in reducing the risk of chemotherapy-induced POI are consistent for breast cancer patients, negative data were shown in the hematological studies. Nevertheless, it should be noted that a higher number of studies including more than 1,600 patients have been conducted in the breast cancer setting as compared to only four trials for approximately 150 randomized women with hematological malignancies. In addition, the two populations of patients are different in terms of both age at diagnosis and type of chemotherapy regimens to be used. Specifically, women with hematological malignancies are usually diagnosed around the age of 25 years and are candidates to receive chemotherapy regimens with low (e.g., ABVD) or high (e.g., conditioning regimens for hematopoietic stem cell transplantation) gonadotoxicity . On the contrary, premenopausal breast cancer patients are usually diagnosed at an older age close to 40 years and are candidates to chemotherapy regimens with an intermediate gonadotoxicity risk .
Second, various definitions and time points of its evaluation have been used to define the primary endpoints (i.e., chemotherapy-induced POI) of the different randomized trials. Although no standard definition of chemotherapy-induced POI is available to date, a composite definition including amenorrhea for ≥2 years and post-menopausal hormonal profile is the preferred definition according to experts [6, 91]. Only few randomized trials applied this definition while the majority considered amenorrhea only as marker of chemotherapy-induced POI. In addition, few trials reported AMH levels at baseline and during study follow-up in these patients.
Third, limited data are available on the fertility preservation potential of temporary ovarian suppression with GnRHa during chemotherapy. However, these trials were neither designed nor powered to detect differences in posttreatment pregnancies, and desire of a future pregnancy was not an inclusion criteria for any of them; in addition, the follow-up of the majority of the trials is too short to assess the fertility preservation potential of this strategy. Notably, the most recent meta-analyses showed a significantly higher number of posttreatment pregnancies in women who received GnRHa during chemotherapy when breast cancer trials were considered [83, 84, 86, 87, 89], but no benefit in women with hematological malignancies was observed .