F Horicks1, G Van Den Steen1, C Gervy2, H J Clarke3, I Demeestere1. 1. Research Laboratory on Human Reproduction, Université Libre de Bruxelles (ULB), 808 route de Lennik, 1070 Brussels, Belgium. 2. Laboratory of Chemistry, Erasme Hospital, 808 route de Lennik, 1070 Brussels, Belgium. 3. Departments of Obstetrics and Gynaecology and Biology, McGill University; Research Institute - McGill University Health Centre, Glen Research Building, 1001 Decarie Blvd, Montreal, QC H4A3J1, Canada.
Abstract
STUDY QUESTION: Does fertility preservation using gonadotrophin-releasing hormone (GnRH) analogues during chemotherapy act through a direct effect on the ovary or through inhibition of FSH secretion? SUMMARY ANSWER: The absence of FSH in vivo and the direct exposition of ovarian follicles to GnRH analogues in vitro did not prevent chemotherapy-induced ovarian damage. WHAT IS KNOWN ALREADY: The potential mechanisms of action of GnRH analogues in protecting ovaries against chemotherapy damage remain poorly understood. We previously showed that GnRH analogues have a limited inhibitory effect on gonadotropin secretion and follicular growth in mice. STUDY DESIGN SIZE, DURATION: Mouse models were developed to independently evaluate (i) the indirect effect of FSH depletion on chemotherapy-induced ovarian damage using Fshb-deficient (-/-) mice to mimic the profound inhibition of FSH secretion during GnRH analogues treatment and (ii) the direct in vitro effect of GnRH agonist and antagonist in follicles exposed to chemotherapy using a follicular culture system. PARTICIPANTS/MATERIALS, SETTING, METHODS: To assess the indirect effect of GnRH analogues through FSH inhibition, Fshb-/- mice were treated with 1 IU pregnant mare serum gonadotropin (control group) or saline (study group) for 7 days and with cyclophosphamide (200 mg/kg) on Day 5. Ovaries were collected 48 h post-cyclophosphamide to evaluate ovarian reserve, cellular apoptosis and proliferation. To evaluate the direct effects of GnRH analogues on growing follicles, isolated preantral follicles from prepubertal mice were cultured in vitro for 13 days with 1 μM GnRH analogues and 20 μM of 4-hydroperoxycyclophosphamide or not at Day 4. Oocytes were matured by adding epidermal growth factor (EGF)/hCG on Day 12. Follicular development, follicular survival, oocyte maturation rates, cAMP production, and steroidogenesis were evaluated. To assess the direct GnRH analogues effects on follicular reserve, whole neonatal ovaries were cultured in vitro under the same conditions for 2 days. Ovaries were processed 24 h post-chemotherapy for ovarian reserve, cellular apoptosis and proliferation analysis. MAIN RESULTS AND THE ROLE OF CHANCE: Cyclophosphamide induced a significant follicular loss of more than 50% in Fshb-/- mice regardless of previous treatment with gonadotropins and no difference was observed in cell proliferation or apoptosis. In vitro experiments on growing follicles showed that 4-hydroperoxycyclophosphamide significantly decreased preantral follicle survival and maturation rates (55% and 37%, respectively) and delayed follicular development, regardless of the presence of GnRH analogues. Chemotherapy reduced granulosa cell numbers in all groups, while no change in cAMP production/106 granulosa cells was observed. Similarly, 4-hydroperoxycyclophosphamide induced apoptosis and significant follicular loss in cultured neonatal ovaries irrespective of GnRH analogues exposure. LIMITATIONS REASONS FOR CAUTION: As ovarian GnRH receptors expression differs in humans and mice, further studies are needed to validate our results in human ovaries. WIDER IMPLICATIONS OF THE FINDINGS: Our findings demonstrate that ovarian damage occurred even in the absence of FSH, suggesting that inhibition of the pituitary-gonadal axis is not involved in ovarian protection during GnRH analogues treatment. Using in vitro models, no evidence for direct protective effect of GnRH analogues against cyclophosphamide metabolite damage was observed. At present, clinical efficiency of GnRH analogues to prevent chemotherapy-induced ovarian damage remains highly debated and these experimental results reinforced the question as they did not bring evidence of direct or indirect mechanisms of protection. LARGE SCALE DATA: N/A. STUDY FUNDING AND COMPETING INTEREST(S): This work was supported by the Belgian FNRS, 'Le Fonds Emile DEFAY', and 'La Fondation Rose et Jean Hoguet'. Authors have no conflict of interest to declare.
STUDY QUESTION: Does fertility preservation using gonadotrophin-releasing hormone (GnRH) analogues during chemotherapy act through a direct effect on the ovary or through inhibition of FSH secretion? SUMMARY ANSWER: The absence of FSH in vivo and the direct exposition of ovarian follicles to GnRH analogues in vitro did not prevent chemotherapy-induced ovarian damage. WHAT IS KNOWN ALREADY: The potential mechanisms of action of GnRH analogues in protecting ovaries against chemotherapy damage remain poorly understood. We previously showed that GnRH analogues have a limited inhibitory effect on gonadotropin secretion and follicular growth in mice. STUDY DESIGN SIZE, DURATION: Mouse models were developed to independently evaluate (i) the indirect effect of FSH depletion on chemotherapy-induced ovarian damage using Fshb-deficient (-/-) mice to mimic the profound inhibition of FSH secretion during GnRH analogues treatment and (ii) the direct in vitro effect of GnRH agonist and antagonist in follicles exposed to chemotherapy using a follicular culture system. PARTICIPANTS/MATERIALS, SETTING, METHODS: To assess the indirect effect of GnRH analogues through FSH inhibition, Fshb-/- mice were treated with 1 IU pregnant mare serum gonadotropin (control group) or saline (study group) for 7 days and with cyclophosphamide (200 mg/kg) on Day 5. Ovaries were collected 48 h post-cyclophosphamide to evaluate ovarian reserve, cellular apoptosis and proliferation. To evaluate the direct effects of GnRH analogues on growing follicles, isolated preantral follicles from prepubertal mice were cultured in vitro for 13 days with 1 μM GnRH analogues and 20 μM of 4-hydroperoxycyclophosphamide or not at Day 4. Oocytes were matured by adding epidermal growth factor (EGF)/hCG on Day 12. Follicular development, follicular survival, oocyte maturation rates, cAMP production, and steroidogenesis were evaluated. To assess the direct GnRH analogues effects on follicular reserve, whole neonatal ovaries were cultured in vitro under the same conditions for 2 days. Ovaries were processed 24 h post-chemotherapy for ovarian reserve, cellular apoptosis and proliferation analysis. MAIN RESULTS AND THE ROLE OF CHANCE: Cyclophosphamide induced a significant follicular loss of more than 50% in Fshb-/- mice regardless of previous treatment with gonadotropins and no difference was observed in cell proliferation or apoptosis. In vitro experiments on growing follicles showed that 4-hydroperoxycyclophosphamide significantly decreased preantral follicle survival and maturation rates (55% and 37%, respectively) and delayed follicular development, regardless of the presence of GnRH analogues. Chemotherapy reduced granulosa cell numbers in all groups, while no change in cAMP production/106 granulosa cells was observed. Similarly, 4-hydroperoxycyclophosphamide induced apoptosis and significant follicular loss in cultured neonatal ovaries irrespective of GnRH analogues exposure. LIMITATIONS REASONS FOR CAUTION: As ovarian GnRH receptors expression differs in humans and mice, further studies are needed to validate our results in humanovaries. WIDER IMPLICATIONS OF THE FINDINGS: Our findings demonstrate that ovarian damage occurred even in the absence of FSH, suggesting that inhibition of the pituitary-gonadal axis is not involved in ovarian protection during GnRH analogues treatment. Using in vitro models, no evidence for direct protective effect of GnRH analogues against cyclophosphamide metabolite damage was observed. At present, clinical efficiency of GnRH analogues to prevent chemotherapy-induced ovarian damage remains highly debated and these experimental results reinforced the question as they did not bring evidence of direct or indirect mechanisms of protection. LARGE SCALE DATA: N/A. STUDY FUNDING AND COMPETING INTEREST(S): This work was supported by the Belgian FNRS, 'Le Fonds Emile DEFAY', and 'La Fondation Rose et Jean Hoguet'. Authors have no conflict of interest to declare.
Authors: L M Del Castillo; A Buigues; V Rossi; M J Soriano; J Martinez; M De Felici; H K Lamsira; F Di Rella; F G Klinger; A Pellicer; S Herraiz Journal: Hum Reprod Date: 2021-08-18 Impact factor: 6.353