Cancer treatment, such as chemotherapy, induces early ovarian follicular depletion and subsequent infertility. In order to protect gametes from the gonadotoxic effects of chemotherapy, several fertility preservation techniques—such as oocyte or embryo cryopreservation with or without ovarian stimulation, or cryopreservation of the ovarian cortex—should be considered. However, these methods may be difficult to perform, and the future use of cryopreserved germ cells remains uncertain. Therefore, improving the methods currently available and developing new strategies to preserve fertility represent major challenges in the area of oncofertility. Animal and ovarian culture models have been used to decipher the effects of different cytotoxic agents on ovarian function and several theories regarding chemotherapy gonadotoxicity have been raised. For example, cytotoxic agents might (i) have a direct detrimental effect on the DNA of primordial follicles constituting the ovarian reserve and induce apoptosis; (ii) induce a massive growth of dormant follicles, which are then destroyed; or (ii) induce vascular ovarian damage. Thanks to improvements in the understanding of the mechanisms involved, a large number of studies have been carried out to develop molecules limiting the negative impact of chemotherapy on the ovaries.

The evaluation of the number of mouse ovarian primordial follicles (PMF) can provide important information about ovarian function, regulation of folliculogenesis or the impact of chemotherapy on fertility. This counting, usually performed by specialized operators, is a tedious, time-consuming but indispensable procedure.The development and increasing use of deep machine learning algorithms promise to speed up and improve this process. Here, we present a new methodology of automatically detecting and counting PMF, using convolutional neural networks driven by labelled datasets and a sliding window algorithm to select test data. Trained from a database of 9 millions of images extracted from mouse ovaries, and tested over two ovaries (3 millions of images to classify and 2 000 follicles to detect), the algorithm processes the digitized histological slides of a completed ovary in less than one minute, dividing the usual processing time by a factor of about 30. It also outperforms the measurements made by a pathologist through optical detection. Its ability to correct label errors enables conducting an active learning process with the operator, improving the overall counting iteratively. These results could be suitable to adapt the methodology to the human ovarian follicles by transfer learning.

Is serum anti-Müllerian hormone (AMH) level predictive of cumulative live birth (CLB) rate after ART or in women trying to conceive naturally? Serum AMH level is linked to CLB after IVF/ICSI but data are lacking after IUI or in women trying to conceive without ART. Serum AMH level is a marker of ovarian reserve and a good predictor of ovarian response after controlled ovarian stimulation. It is unclear whether AMH measurement can predict CLB in spontaneous or assisted conception. A systematic review and meta-analysis was undertaken to assess whether serum AMH level may predict chances of CLB in infertile women undergoing IVF/ICSI or IUI and/or chances of live birth in women having conceived naturally. A systematic review and meta-analysis was performed using the following keywords: ‘AMH’, ‘anti-mullerian hormone’, ‘live-birth’, ‘cumulative live birth’. Searches were conducted from January 2004 to April 2021 on PubMed and Embase. Two independent reviewers carried out study selection, quality, and risk of bias assessment as well as data extraction. Odds ratios were estimated using a random-effect model. Pre-specified sensitivity analyses and subgroup analyses were performed. The primary outcome was CLB. A total of 32 studies were included in the meta-analysis. Overall, 27 articles were included in the meta-analysis of the relation between AMH and CLB or AMH and LB after IVF/ICSI. A non-linear positive relation was found in both cases. A polynomial fraction was the best model to describe it but no discriminant AMH threshold was shown, especially no serum AMH level threshold below which live birth could not be achieved after IVF/ICSI. After IVF–ICSI, only four studies reported CLB rate according to AMH level. No statistically significant differences in mean serum AMH levels were shown between patients with and without CLB, but with a high heterogeneity. After exclusion of two studies with high risks of bias, there was no more heterogeneity [I² = 0%] and the mean AMH level was statistically significantly higher in women with CLB. There were not enough articles/data to assess the ability of AMH to predict CLB rate or find an AMH threshold after IUI or in women without history of infertility trying to conceive without ART. The systematic review and meta-analysis had some limitations owing to the limits and bias of the studies included. In the present meta-analysis, heterogeneity may have been caused by different baseline characteristics in study participants, different stimulating protocols for ART, different serum AMH level thresholds used and the use of various assays for serum AMH. This could explain, in part, the absence of a discriminating AMH threshold found in this analysis. Serum AMH level is linked to CLB rate after IVF/ICSI but no discriminating threshold can be established, therefore low serum AMH level should not be used as the sole criterion for rejecting IVF treatment, especially in young patients. Data are lacking concerning its predictive value after IUI or in women trying to conceive without ART. Our findings may be helpful to counsel candidate couples to IVF-ICSI. No external funding was obtained for this study. There are no conflicts of interest. PROSPERO CRD42021269332.

Anti-Müllerian hormone (AMH) is a member of the transforming growth factor (TGF)-beta family and a key regulator of sexual differentiation and folliculogenesis. While the serum AMH level has been used in reproductive medicine as a biomarker of quantitative ovarian reserve for more than 20 years, new potential therapeutic applications of recombinant AMH are emerging, notably in the field of oncofertility. Indeed, it is well known that chemotherapy, used to treat cancer, induces ovarian follicular depletion and subsequent infertility. Animal models have been used widely to understand the effects of different cytotoxic agents on ovarian function, and several hypotheses regarding chemotherapy gonadotoxicity have been proposed, that is, it might have a direct detrimental effect on the primordial follicles constituting the ovarian reserve and/or on the pool of growing follicles secreting AMH. Recently, a new mechanism of chemotherapy-induced follicular depletion, called the “burn-out effect,” has been proposed. According to this theory, chemotherapeutic agents may lead to a massive growth of dormant follicles which are then destroyed. As AMH is one of the factors regulating the recruitment of primordial follicles from the ovarian reserve, recombinant AMH administration concomitant with chemotherapy might limit follicular depletion, therefore representing a promising option for preserving fertility in women suffering from cancer. This review reports on the potential usefulness of AMH measurement as well as AMH’s role as a therapeutic agent in the field of female fertility preservation.

Oocyte and/or embryo vitrification after controlled ovarian stimulation (COS) represents the most established method of fertility preservation (FP) before cancer treatment. However, traditional COS regimens are associated with supraphysiologic serum estradiol and are therefore not recommended in estrogen-sensitive diseases such as breast cancer (BC). To protect the patients from the potential deleterious effects of elevated estrogen levels during COS for FP, protocols using aromatase inhibitors (letrozole) were developed. The present study aims at investigating whether COS with letrozole supplementation (COSTLES) modifies ovarian response in BC patients. One hundred and seventy-seven BC patients candidates for FP using oocyte and/or embryo vitrification following COS referred to our center between July 2013 and December 2016 were included in this retrospective case-control study. 94 patients underwent COSTLES while 83 had standard GnRH antagonist protocol. The number of oocytes retrieved, oocyte maturation rates, number of oocytes vitrified and follicle responsiveness to FSH assessed by the Follicular Output Rate (FORT) were assessed. Women in both groups were comparable in terms of age and ovarian reserve tests leading to a similar number of oocyte recovered (13.1 ± 10.0 vs. 12.2 ± 8.0 oocytes, respectively, NS). However, oocyte maturation rates were significantly lower in COSTLES compared to standard protocol (64.9 ± 22.8 vs. 77.4 ± 19.3%, p < 0.001). As a result, the number of mature oocyte vitrified was lower in COSTLES group (7.8 ± 5.3 vs. 10.3 ± 8.5 oocytes, p < 0.001 respectively). Despite similar response to exogenous FSH, BC patients having undergone COSTLES show reduced oocyte maturation rates in comparison with those having received standard stimulation regimen.