Unraveling the Clinical FSH Conundrum: Insights From the Small Ovarian Reserve Heifer Model.

FSH is indispensable for fertility in mammals (Wang et al.  2021 ) and used by clinicians worldwide to stimulate the growth of multiple (e.g., a dozen) ovulatory follicles in an effort to obtain multiple high quality oocytes for IVF, which enhances the probability of achieving a live birth (Broekmans  2019 ; Macklon et al.  2006 ). During a typical ART cycle (SART  2021 ), the patient is injected beginning on days 1–2 of menses with human recombinant FSH (hrFSH; Lunenfeld et al.  2019 ) daily for 1–2 weeks to stimulate maximal growth of multiple ovulatory‐size follicles (Mahony et al.  2021 ). The hrFSH injections are often preceded by and sometimes coincide with injections of gonadotropin‐releasing hormone agonists or antagonists to prevent premature ovulatory LH surges, treatment with birth control pills to better manage the menstrual cycle and ovulation to optimize the timing of IVF, or treatment with selective estrogen receptor modulators (SERMs) to improve responsiveness to FSH (SART  2021 ). Before IVF, the patient is injected with human chorionic gonadotropin (hCG) to stimulate oocyte maturation within each ovulatory‐size follicle developing during ovarian stimulation. The hCG‐matured oocytes are recovered mechanically from each ovulatory‐size follicle 24–36 h after the hCG injection (before ovulation). Within several hours after recovery, the mature oocytes are subjected to either IVF or intracytoplasmic sperm injection (ICSI). Approximately 2–6 days after successful IVF or ICSI of the oocytes, one or more of the highest quality embryo(s) (Klitzman  2016 ) are transferred back into the uterus of the donor or a recipient hopefully resulting in a live birth. Any excess high‐quality embryos not transferred into the patient may be cryopreserved and used for frozen embryo transfer (FET) (SART  2021 ). ART also includes gamete intrafallopian transfer (GIFT) and zygote intrafallopian transfer (ZIFT), although these laparoscopic procedures are infrequently used (Van Voorhis  2006 ). The most common approach employs blastocyst transfer, as this allows in vitro assessment of morphological and other criteria of embryo quality, as well as genetic screening (Latham  2024 ) increasing the importance of obtaining high‐quality oocytes. Single embryo transfer is becoming increasingly desirable, as it minimizes the risk of multiple‐gestation pregnancies and may be necessary in some cases to comply with legislative restrictions. Single embryo transfer, however, places a premium on obtaining the highest quality oocytes possible. The increasing demand for obtaining the highest quality oocytes has enhanced the desire to minimize the risk of negative impacts of ART procedures, including FSH dosing, on oocytes.

ART results in live births for many women. Nevertheless, significant improvement of ART outcomes is needed to mitigate the high economic and emotional costs associated with failed ART cycles (Brezina and Zhao  2012 ; Carson et al.  2021 ; Chambers, Adamson, and Eijkemans  2013 ). Such advancements would clearly broaden the use of this technology by the millions of childless couples (“Family planning/contraception methods.  2020 ; “Infertility.  2023 ; SingleCare  2023 ). Development of new, more efficient ART procedures, however, remains a daunting challenge. Barriers to enhancing ART outcomes include the inherently high variability in the ovarian reserve and responsiveness to ovarian stimulation with FSH, high FSH‐induced oocyte wastage amongst patients seeking ART, and the ethical and experimental design limitations associated with improvement of ART procedures using human subjects. Consequently, we developed an experimentally adaptable, and relevant biomedical model to better understand the complex interrelationship between the ovarian reserve, FSH secretion, and ovarian stimulation with high FSH dose, and how they impact ovarian function, oocyte quality, and ART outcomes.

FSH is a pituitary glycoprotein hormone required for ovarian follicular growth and development (Wang et al.  2021 ). During the follicular phase of the menstrual cycle, women exhibit multiple “waves” of follicle growth (Baerwald and Pierson  2020 ; Baerwald, Adams, and Pierson Adams, and Pierson  2003 ,  2012 ; Kirillova et al.  2021 ). A single dominant ovulatory follicle develops during one of these waves and ovulates at mid‐cycle in response to a preovulatory LH surge. The remaining follicular waves sometimes also produce large non‐ovulatory follicles but these regress (Baerwald and Pierson  2020 ; Baerwald, Adams, and Pierson Adams, and Pierson  2003 ,  2012 ; Kirillova et al.  2021 ). At the beginning of the ovulatory follicular wave, a transient increase in the circulating FSH concentration (Baerwald and Pierson  2020 ; Baerwald, Adams, and Pierson  2012 ) is predicted to upregulate FSH receptors in the granulosa and cumulus cells (Leung and Steele  1992 ; Themmen and Huhtaniemi  2000 ) of dozens of preantral follicles as they transition into antral follicles (Baerwald and Pierson  2020 ; Baerwald, Adams, and Pierson  2012 ; Kumar et al.  1997 ). This activates diverse signaling systems (Babu et al.  2000 ; Touyz, Jiang, and Ram Sairam  2000 ) that both inhibit expression of proapoptotic genes (Casarini and Crépieux  2019 ; Regan et al.  2018 ; Shen et al.  2014 ; Shi et al.  2020 ; Zhou et al.  2013 ) and enhance the expression of many other genes that are important for the formation of antral follicles, as well as follicular growth and differentiation, estradiol and progesterone production (Glister et al.  2001 ), and the regulation of oxygen and glucose in antral follicles (Foreman, Kolettis, and Garris  1993 ). Some well‐established FSH‐induced genes include aromatase (Marsters, Kendall, and Campbell  2003 ; Silva and Price  2000 ), estradiol α, ß receptors (Richards  2001 ), FSH receptor (Richards  2001 ), LH receptor (Erickson, Wang, and Hsueh  1979 ; Marsters, Kendall, and Campbell  2003 ), inhibins (Boudjemaa et al.  2000 ; Glister et al.  2001 ), activins (Glister et al.  2001 ), follistatins (Glister et al.  2001 ), StAR (Balasubramanian et al.  1997 ), cyclin D2 (Sicinski et al.  1996 ), epidermal growth factor (EGF)‐like ligands, BMP‐15, and the fibroblastic growth factor (FGF) proteins and receptors (Caixeta et al.  2013 ). The decline from peak FSH concentrations during the follicular wave triggers selection of a single antral follicle. This follicle becomes dominant and continues its growth to ovulatory size while all other growing follicles undergo atresia (Ginther et al.  2001 ; Ireland et al.  2000 ; Mihm et al.  1997 ). FSH also promotes oocyte growth and competence (Demeestere et al.  2012 ), cumulus cell expansion, and production of critical factors for oocyte maturation in an ovulatory follicle (Turathum, Gao, and Chian  2021 ). Despite progress (Ginther et al.  2001 ; Mihm et al.  1997 ), however, the precise mechanisms explaining the FSH‐regulated selection of a single dominant ovulatory follicle during a follicular wave remain obscure.

Choosing the optimal FSH dose during ovarian stimulation of a patient to maximize the number of ovulatory‐size follicles and recovery of high‐quality oocytes for IVF is the clinician's chief conundrum during ART. Consequently, clinicians use a wide range of hrFSH doses (up to 20‐fold) during ovarian stimulation (Baker et al.  2015 ) which further contributes to the high variability in the responsiveness of individuals to ovarian stimulation during ART. The high circulating FSH concentrations during aging are postulated to be toxic to ovulatory follicle function and oocyte quality (Bernstein et al.  2024 ; Bernstein et al.  2023 ; Buratini et al.  2022 ). Nevertheless, it is unknown if the pharmacological amounts of hrFSH used during ovarian stimulation explain why a high proportion (˃ 90%) of oocytes recovered from ovulatory‐size follicles for IVF during ART are immature, degenerate, fail to fertilize, or yield embryos that develop abnormally after IVF (Bianchi et al.  2007 ; Chang et al.  2009 ; French et al.  2010 ; Griesinger, Felberbaum, and Diedrich  2005 ; Hardy  1999 ; Kovalevsky and Patrizio  2005 ; Levi‐Setti et al.  2016 ; Macklon.  2002 ; Meniru and Craft  1997 ; Patrizio and Sakkas  2009 ; Silber et al.  2017 ; Steward et al.  2014 ). This question is critical to resolve especially in DOR patients who have inherently higher circulating FSH concentrations and lower AMH concentrations compared with their counterparts with a larger ovarian reserve. Ovarian stimulation with high hrFSH doses during ART is linked to impaired embryo survival and low live birth rates (Baker et al.  2015 ). The pregnancy rate in women after a single natural insemination is usually high, ranging from 60% to 80% (Jarvis  2016 ). However, live birth rates for a single ART cycle are significantly lower, typically between 20% and 36%, despite transferring high‐quality IVF‐generated embryos (Baker et al.  2015 ; Clark et al.  2021 ; Gleicher, Kushnir, and Barad  2019 ; Gnoth et al.  2011 ; Hazekamp et al.  2000 ; Kovalevsky and Patrizio  2005 ; Luke et al.  2012 ; McLernon et al.  2016 ; Poulain et al.  2021 ; Prevention, C. f. D. C. a.  2020 ; Saket et al.  2021 ; Sanders et al.  2021 ; Smith et al.  2015 ; Sullivan et al.  2013 ; Sunkara et al.  2011 ; Toftager et al.  2017 ). This large discrepancy between natural and IVF pregnancy rates explains why many patients require multiple ART cycles (ranging from 1 to 9; Smith et al.  2015 ) before achieving a successful pregnancy, contributing to the high cost of treatment. Many studies report that the high doses of FSH during ovarian stimulation are detrimental to oocyte quality (Baker et al.  2015 ; Clark et al.  2021 ; Kovalevsky and Patrizio  2005 ; Kuokkanen and Pal  2023 ; Luo et al.  2022 ; Munch et al.  2017 ; Sanders et al.  2021 ; Shaia et al.  2020 ). The rampant FSH‐induced oocyte wastage indicates that the high FSH doses during ovarian stimulation may also be harmful to survival of the IVF‐generated embryos transferred into patients during ART. The impact of FSH dosage on ART outcomes in women, however, has long been debated (Hohmann, Macklon, and Fauser  2003 ; Katz‐Jaffe, Trounson, and Cram  2005 ; Klinkert et al.  2005 ; Lekamge et al.  2008 ; Myers et al.  2008 ; Olivennes et al.  2011 ; Out et al.  2000 ; Out.  2001 ; Pal et al.  2008 ; Sterrenburg et al.  2011 ; Tan et al.  2005 ). Several weaknesses in experimental designs have contributed to this controversy: (1) live birth rates were rarely reported, (2) studies often involved small groups with no consistent differences in pregnancy rates, and (3) the relationship between FSH dose, AFC, and ART outcomes was disregarded. To address these limitations, we (Baker et al.  2015 ) analyzed a large dataset of 658,519 women who underwent fresh autologous ART cycles between 2004 and 2012. Our analysis revealed a highly significant ( p  < 0.0001) decrease in both oocyte retrieval rates (Clark et al.  2021 ) and live birth rates (Baker et al.  2015 ) with increasing total FSH dose used during ovarian stimulation (Figure  1 ). Notably, this finding aligns with a separate European analysis of 400,135 IVF cycles, which also demonstrated the lowest live birth rates were associated with the lowest oocyte retrieval rates (Baker et al.  2015 ; Clark et al.  2021 ; Sunkara et al.  2011 ). Furthermore, the inverse relationship between total FSH dose and live birth rate (Figure  1 ) has been reported in other studies involving diverse patient groups and locations (Baker et al.  2015 ; Edwards, Lobo, and Bouchard  1996 ; Klinkert et al.  2005 ; Kuokkanen and Pal  2023 ; Luo et al.  2022 ; Munch et al.  2017 ; Pal et al.  2008 ; Santos, Kuijk, and Macklon  2010 ; Shaia et al.  2020 ; Tan et al.  2005 ; Xu et al.  2024 ) including a recent analysis of another large dataset from 2014 to 2016 (Shaia et al.  2020 ). The inverse relationship between total FSH dosage during ovarian stimulation and live birth rate during ART is best exemplified in our study where the highest FSH dosage compared to either of the two lowest resulted in a ~50% reduction in live birth rate (Figure  1 )! Taken together, these large correlative studies positively linked the use of high hrFSH doses during ovarian stimulation not only to a high degree of oocyte wastage but also diminished survivability of the IVF‐generated embryos during ART. Relationship of total FSH doses administered during ovarian simulation with oocyte recovery and live birth rate during 658,519 fresh autologous ART cycles in women (as adapted from Baker et al.  2015 ; Clark et al.  2021 ). Bars depict mean ± SEM (SEM not visible in figure). Number of ART cycles at each different total FSH dose range follows: ˂ 1000 IU = 10,916; 1001–2000 = 132,976; 2001–3000 = 184,182; 3001–4000 = 116,419; 4001–5000 = 105,635; ˃ 5001 = 108,391. Portions of the figure are reproduced from Baker et al.  2015 ; Clark et al.  2021 with permission from [Fertility and Sterility, https://www.fertstert.org/ and Journal of Assisted Reproductive Genetics, https://link.springer.com/journal/10815 ].

The precise reasons clinicians used the highest FSH doses were not reported in our study (Figure  1 ; Baker et al.  2015 ). Nevertheless, the FSH dosages used during ART are typically tailored to patient characteristics such as age, body weight, biomarker concentrations of AMH, FSH concentration, AFC, and prior ART cycle response (Broekmans  2019 ; Fatemi et al.  2021 ; Phillips, Olanrewaju, and Omole  2023 ). This raises the caveat that the highest FSH doses were given to patients with the inherently poorest response and lowest fertility thus explaining the inverse relationship between FSH dosage and birth rate (Figure  1 ) (Baker et al.  2015 ). This concern, however, seems unlikely because the inverse relationship between FSH doses and live birth rate during ART (Figure  1 ) remained significant ( p  < 0.0001) regardless of donor age (Figure  1 ), oocyte recovery number, length of treatment, BMI, health, and reason for IVF (data not shown; Baker et al.  2015 ). This finding aligns with several other clinical trials (Klinkert et al.  2005 ; Pal et al.  2008 ; Tan et al.  2005 ) that confirmed that milder FSH stimulation (Edwards, Lobo, and Bouchard  1996 ; Santos, Kuijk, and Macklon  2010 ) improves embryo survival. In these studies, the higher FSH doses did not improve the response of typically poor‐responding patients (e.g., those with low AFC, low AMH or increased age which are all DOR symptoms) as measured by oocyte retrieval (Klinkert et al.  2005 ; Lekamge et al.  2008 ; Out et al.  2000 ; Pal et al.  2008 ). The benefits of low FSH dosages during ovarian stimulation were also observed in our analyses (Baker et al.  2015 ) as the lowest total FSH doses (≤ 2000 IU, Figure  1 ) were linked to the highest live birth rates regardless of oocyte recovery rate, age of donor, length of treatment, body mass index or reason for ART. These combined observations further amplified the long‐standing warning by Nobel Prize Winner Robert Edwards (Edwards, Lobo, and Bouchard  1996 ) that ovarian stimulation with excessive doses of FSH is detrimental to ART outcomes (Baker et al.  2015 ; Edwards, Lobo, and Bouchard  1996 ; Kuokkanen and Pal  2023 ; Luo et al.  2022 ; Munch et al.  2017 ; Santos, Kuijk, and Macklon  2010 ; Shaia et al.  2020 ).

The ovarian reserve is remarkably variable throughout life in cattle and women (Block  1953 ; Erickson  1966a ,  1966b ; Erickson, Reynolds, and Murphree  1976 ; Ireland et al.  2008 ), ranging from 1000 to 400,000 oocytes at birth (Erickson, Reynolds, and Murphree  1976 ) and ~14,000–250,000 in adults (Erickson  1966b ; Ireland et al.  2008 ). FSH secretion is heightened as the ovarian reserve decreases during aging but before ovulation and cycles cease in cattle and women (Block  1953 ; Faddy  2000 ; Klein et al.  1996 ; Malhi et al.  2006 ; Malhi, Adams, and Singh  2005 ; McTavish et al.  2007 ; Reame et al.  1996a ; Welt et al.  1999 ). Moreover, like women, cattle are a single‐ovulating species with multiple waves of growth of dozens of small antral follicles during a long reproductive cycle (Ireland et al.  2000 ) like women (Baerwald, Adams, and Pierson  2012 ). However, the pool of follicles available that respond to ovarian stimulation protocols is highly variable in cattle as it is in women. This is best exemplified in our studies (Burns et al.  2005 ; Ireland et al.  2007 ; Ireland et al.  2008 ) wherein serial ovarian ultrasound analysis in 11‐ to 12‐month‐old (young adult) beef or dairy heifers established that despite the uniformity of age, breed and weight, and length of the reproductive cycle, the AFC, or number of small antral follicles (≥ 3 mm in diameter, ovulatory size = ~15–20 mm) growing during follicular waves ranged from 2 to 54. Because the small antral follicles develop into ovulatory‐size follicles during ovarian stimulation with FSH, the high variability in AFC during follicular waves in cattle (Burns et al.  2005 ; Ireland et al.  2007 ; Ireland et al.  2008 ; Singh et al.  2004 ) and women (Baerwald, Adams, and Pierson  2012 ) is very likely a major contributing factor to the unpredictably erratic numbers of follicles developing in response to ovarian stimulation with FSH (e.g., 0–101 ova collected from presumably ovulatory‐size follicles in cattle; Hasler  1992 ) during ART in both species (Al‐Shawaf et al.  2001 ; Hasler  1992 ; Liu et al.  2023 ; Sunkara et al.  2011 ; Yildiz et al.  2020 ). The first calf was born from embryo transfer in 1951 (Willett et al.  1951 ) and the first calf born from IVF was in 1982 (Brackett et al.  1982 ). There are hundreds of commercial entities (Demetrio et al.  2020 ) using well‐established techniques for ovarian stimulation, IVF (~1,500,000 IVF‐produced embryos transferred in 2022; Viana  2023 ), in vivo derived embryos and transfer (~370,000 transfers in 2022; Viana  2023 ), and cryopreservation (e.g., ~30% of all IVEP‐generated embryos are cryopreserved; Do and Taylor‐Robinson  2020 ) in cattle, as during ART in women (SART  2021 ). Although recombinant bovine FSH products have been developed (Carvalho et al.  2014 ; Gutiérrez‐Reinoso et al.  2022 ; Hesser, Morris, and Gibbons  2011 ; Looney et al.  1988 ), they are not commercially available for use during ART in cattle. Pituitary extracts enriched for FSH (e.g., Folltropin‐V, Pluset, Stimufol; Galli and Lazzari  2024 ) are used worldwide to superovulate cattle. Folltropin‐V, an FSH‐enriched product produced commercially by Vetoquinol, Inc. (Princeville, Quebec), is the only FDA approved drug widely and successfully used for ovarian stimulation in the bovine embryo transfer industry. Folltropin‐V is an extract of porcine pituitary glands that contains 700 IU of FSH per 20 mL vial (equivalent to 400 mg NIH‐FSH‐P1), with a low 0.25% LH contamination, or < 1 mg NIH‐LH‐S19 (Freedom of Information Summary NADA 141‐431 Folltropin  2014 ). Hereafter, this commercial porcine pituitary extract of FSH is referred to as cpFSH. Whether cpFSH doses used during ovarian stimulation in cattle are potentially inversely linked to live birth rates as observed for women (Figure  1 ; Baker et al.  2015 ; Edwards, Lobo, and Bouchard  1996 ; Klinkert et al.  2005 ; Kuokkanen and Pal  2023 ; Luo et al.  2022 ; Munch et al.  2017 ; Pal et al.  2008 ; Santos, Kuijk, and Macklon  2010 ; Shaia et al.  2020 ; Tan et al.  2005 ) is unknown. Nevertheless, the detrimental effects of FSH doses on ovarian function and oocyte/embryo wastage in cattle is controversial as it is in women (Hohmann, Macklon, and Fauser  2003 ; Katz‐Jaffe, Trounson, and Cram  2005 ; Klinkert et al.  2005 ; Lekamge et al.  2008 ; Myers et al.  2008 ; Olivennes et al.  2011 ; Out et al.  2000 ; Out.  2001 ; Pal et al.  2008 ; Sterrenburg et al.  2011 ; Tan et al.  2005 ). For example, despite the differences in hormone action, pharmacokinetics, and dosages, the highest doses of cpFSH or FSH‐like factors (e.g., pregnant mare serum gonadotropin, human menopausal gonadotropin) used to induce superovulation in dose‐response studies decreased fertilization rate (McGowan et al.  1985 ; Pawlyshyn et al.  1986 ; Sugano and Watanabe  1997 ), embryo yield (Donaldson  1984 ; Sugano and Watanabe  1997 ) or quality (McGowan et al.  1985 ; Sugano and Watanabe  1997 ) and number of transferable embryos (Donaldson  1984 ; Greve, Lehn‐Jensen, and Rasbech  1979 ; Pawlyshyn et al.  1986 ; Sugano and Watanabe  1997 ). In addition, high cpFSH doses are linked to high oocyte and embryo wastage (Donaldson  1984 ; Donaldson and Perry  1983 ; Greve, Lehn‐Jensen, and Rasbech  1979 ; McGowan et al.  1985 ; Pawlyshyn et al.  1986 ; Saumande and Chupin  1986 ; Sugano and Watanabe  1997 ) and have negative effects on the numbers of corpora lutea (CL) observed (Kanitz et al.  2002 ; Saumande and Chupin  1986 ), circulating progesterone (Saumande and Chupin  1986 ) and estradiol (Kanitz et al.  2002 ) concentrations, and ovulation rate (Saumande and Chupin  1986 ). Alternatively, other studies show that the highest cpFSH doses increase the number of antral follicles (Gonzalez et al.  1990 ; Souza et al.  2007 ) and estradiol (Saumande and Chupin  1986 ) and progesterone concentrations (Kanitz et al.  2002 ), while other studies have reported no effects on CL number (Barati et al.  2006 ; Gonzalez et al.  1990 ; McGowan et al.  1985 ; Pawlyshyn et al.  1986 ; Souza et al.  2007 ; Sugano and Watanabe  1997 ), circulating progesterone (Gonzalez et al.  1990 ) or estradiol (McGowan et al.  1985 ) concentrations, or ovulation rate (McGowan et al.  1985 ; Souza et al.  2007 ). Many reviews (Armstrong  1993 ; Bó and Mapletoft  2014 ; Hasler  2014 ; Kafi and McGowan  1997 ; Mikkola, Hasler, and Taponen  2020 ; Seidel  1981 ) have identified a variety of factors that cause or contribute to the high variability in response of cattle to ovarian stimulation. These include breed (Hasler  2014 ; Kafi and McGowan  1997 ; Mikkola, Hasler, and Taponen  2020 ; Sartori et al.  2010 ), age (Hasler  2014 ; Kafi and McGowan  1997 ; Mikkola, Hasler, and Taponen  2020 ), parity (Hasler  2014 ; Mikkola, Hasler, and Taponen  2020 ), nutrition (Hasler  2014 ; Kafi and McGowan  1997 ; Mikkola, Hasler, and Taponen  2020 ), AFC and the ovarian reserve (Ireland et al.  2007 ), presence of a dominant follicle (Armstrong  1993 ; Bó and Mapletoft  2014 ; Kafi and McGowan  1997 ), timing of FSH treatments relative to the follicular wave (Adams et al.  1994 ; Armstrong  1993 ; Bó and Mapletoft  2014 ; Guilbault et al.  1991 ; Hasler  2014 ; Mikkola, Hasler, and Taponen  2020 ; Rajamahendran et al.  1987 ), progesterone treatments during ovarian stimulation (Bó et al.  2006 ; Goulding et al.  1994 ; Mikkola, Hasler, and Taponen  2020 ), season (Hasler  2014 ; Kafi and McGowan  1997 ; Mikkola, Hasler, and Taponen  2020 ), source of hormone (Armstrong  1993 ; Mikkola, Hasler, and Taponen  2020 ) and inadequate sample size (Donaldson  1984 ; Kanitz et al.  2002 ; Saumande and Chupin  1986 ). Variability in these experimental factors has confounded deriving a consensus interpretation of the effects of high FSH doses on ovarian function and oocyte/embryo wastage across studies and contributed to controversial views of FSH dose effects during ovarian stimulation of cattle. We discovered that, despite the high variability in AFC amongst individuals, peak AFC during each follicular wave across multiple reproductive cycles within individuals was remarkably highly repeatable (0.84–0.95, 1 = perfect; Burns et al.  2005 ; Ireland et al.  2007 ; Ireland et al.  2008 ; Singh et al.  2004 ; Figure  2 ). Thus, cattle could be reliably phenotyped into three AFC groups for our studies: low (≤ 15 follicles/wave), intermediate (16–24 follicles/wave) or high (≥ 25 follicles/wave) (Burns et al.  2005 ; Ireland et al.  2007 ; Ireland et al.  2008 ). Approximately 15%–20%, 60%–70%, and 15%–20% of the heifers in a herd were distributed into these AFC groups, respectively (Burns et al.  2005 ; Ireland et al.  2007 ; Ireland et al.  2008 ). The young adult heifers that comprised the SORH model, and the older cows (calved at least once) also with a low AFC and small ovarian reserve, exhibited the following phenotypic differences compared with their age‐matched counterparts with a high AFC and larger ovarian reserve: ~80% fewer morphologically healthy follicles and oocytes in both ovaries (Ireland et al.  2008 ), ~80% lower circulating AMH concentrations throughout the reproductive cycle (Figure  3 ; Ireland et al.  2011 ; Ireland et al.  2008 ), up to ~50% higher circulating FSH concentrations during follicular waves (Burns et al.  2005 ; Ireland et al.  2007 ) (Figure  3 ) and basal and episodic LH secretion during the reproductive cycle (Figure  4 ; Jimenez‐Krassel et al.  2009 ), similar or lower circulating estradiol concentrations (Burns et al.  2005 ; Ireland et al.  2007 ; Ireland et al.  2009 ) and ~50% lower circulating concentrations of progesterone (Jimenez‐Krassel et al.  2009 ) and testosterone (Mossa et al.  2010 ) throughout the reproductive cycle, suboptimal fertility (Jimenez‐Krassel et al.  2015 ; Mossa et al.  2012 ); although controversial (Alward, Cockrum, and Ealy  2023 ), and 50% and 30% fewer recovered oocytes/embryos and number of transferable embryos, respectively, following ovarian stimulation with the same doses of cpFSH for embryo transfer (Ireland et al.  2007 ). Alterations in antral follicle count (AFC) and diameter of the dominant and largest subordinate follicle during two follicular waves of an estrous cycle for four representative Holstein dairy cows. Each bar represents the mean for two ultrasound measurements (12 h apart) of the total number of antral follicles ≥ 3 mm in diameter. Each animal was arbitrarily assigned to Low, Intermediate, or High categories based on peak AFC per wave. O = changes in diameter of the dominant non‐ovulatory follicle (DNF) and largest subordinate follicle (SF) during the first follicular wave. ● = changes in diameter of the dominant ovulatory follicle (DOR) and largest SF during the ovulatory follicular wave. ↓ = day of ovulation. *  = peak number of follicles during the wave. Note that peak AFC during each wave differed markedly among cows but was highly repeatable within individuals (Burns et al.  2005 ). Portions of the figure are reproduced from Burns et al.  2005 with permission from [Biology of Reproduction, Oxford University Press; https://academic.oup.com/biolreprod ]. (A) Size of the ovarian reserve in the SORH model (Low) compared with age‐matched 11‐ to 12‐month‐old high‐AFC (High) heifers (Ireland et al.  2008 ). Ovaries were removed surgically 1–2 days after ovulation. A single ovary contralateral to the recent ovulation was subjected to histological analysis to determine the total number of morphologically healthy follicles in both ovaries. n = number of animals per AFC group. Bars depict means ± SEM. Portions of the figure are reproduced from Ireland et al.  2008 with permission from [Biology of Reproduction, Oxford University Press; https://academic.oup.com/biolreprod ]. (B) Alterations in circulating AMH concentrations before ovulation in young adult beef heifers with a consistently low, intermediate or high AFC (as modified from Ireland et al.  2011 ). Each point depicts the mean (± SEM) for AMH concentrations. n = number of cattle. Portions of the figure are reproduced from Ireland et al.  2011 with permission from [Reproduction, Fertility and Development, CSIRO; https://www.publish.csiro.au/rd ]. FSH secretion in age‐matched cattle with a low or high AFC. FSH measurements were made at different times before and after the peak FSH concentrations or emergence (first day ultrasound detects a follicle ≥ 4 mm in diameter) for the first follicular wave or before and after the peak LH concentrations during the ovulatory follicular wave. Top left panel: n  = 3–4 cows per time; remaining panels, n  = 8–11 heifers per time. Data were modified from Burns et al.  2005 ; Ireland et al.  2007 . Portions of the figure are reproduced from Burns et al.  2005 ; Ireland et al.  2007 with permission from [Biology of Reproduction, Oxford University Press; https://academic.oup.com/biolreprod , and Human Reproduction, Oxford University Press; https://academic.oup.com/humrep ]. Our studies (Burns et al.  2005 ; Ireland et al.  2011 ; Ireland et al.  2007 ; Ireland et al.  2009 ; Ireland et al.  2008 ; Jimenez‐Krassel et al.  2009 ; Jimenez‐Krassel et al.  2015 ; Mossa et al.  2010 ; Mossa et al.  2012 ) confirmed that the ovarian biomarkers, AFC and AMH concentration, which are moderately heritable genetic traits in cattle (Nawaz et al.  2018 ; Walsh et al.  2014 ), can be used reliably to identify the inherent differences in the ovarian reserve and ovarian function amongst age‐matched cattle. Most of the phenotypic features for the SORH model and low‐AFC cows with a small ovarian reserve have also been observed in women with DOR (Gleicher, Weghofer, and Barad  2011 ; Gleicher et al.  2013 ; Holte et al.  2011 ; Kawakita et al.  2023 ; Nelson, Yates, and Fleming  2007 ; Patrizio et al.  2015 ; Reame et al.  1996b ; Soares et al.  2020 ; Ulrich and Marsh  2019 ), which is the major reason for infertility and seeking ART in women (“Assisted Reproductive Technology National Summary Report for 2018,”  2021 ; Devine et al.  2015 ; Oudendijk et al.  2011 ; Patrizio et al.  2015 ). However, the link between DOR, a reduced ovarian response and poor ART outcomes in women is controversial and often confounded by age (Baker et al.  2015 ; Devine et al.  2015 ; Edwards, Lobo, and Bouchard  1996 ; Klinkert et al.  2005 ; Kuokkanen and Pal  2023 ; Luo et al.  2022 ; Munch et al.  2017 ; Pal et al.  2008 ; Santos, Kuijk, and Macklon  2010 ; Shaia et al.  2020 ; Tan et al.  2005 ; Xu et al.  2024 ). Consequently, the SORH model is uniquely relevant as a tool for better understanding the complex interrelationship between the ovarian reserve, FSH secretion and excessive FSH action on ovarian function and ART outcomes independent of age. Our studies using the SORH model and low‐AFC cows provided direct evidence that ovarian function and responsiveness to gonadotropin stimulation are suboptimal in individuals with small ovarian reserves. For example, intrafollicular estradiol and androstenedione concentrations in dominant follicles (Mossa et al.  2010 ) and the capacity of granulosa (Scheetz et al.  2012 ), thecal (Mossa et al.  2010 ) and luteal cells (Jimenez‐Krassel et al.  2009 ) to respond to FSH or LH stimulation and produce estradiol, testosterone and progesterone, respectively, was up to 50% lower in the SORH model and the older cows with a low AFC and small ovarian reserve compared with their age‐matched counterparts with a high AFC and larger ovarian reserve. A mechanism to explain the suboptimal ovarian function and ovarian responsiveness to gonadotropin stimuli in the SORH model and low‐AFC cows warrants consideration. Gonadotropin‐induced receptor downregulation and desensitization have a critical role in impeding gonadal responsiveness to an additional gonadotropin stimulus (Amsterdam et al.  2002 ; Menon and Menon  2012 ). Consequently, the chronically higher endogenous secretion of gonadotropins characteristic of the SORH model and low AFC cows (Figures  5 and  6 ; Burns et al.  2005 ; Ireland et al.  2007 ; Jimenez‐Krassel et al.  2009 ) very likely causes the ovary to become refractory to an additional gonadotropin stimulus. Thus, this may explain why the responsiveness of granulosa, thecal and luteal cells to gonadotropin treatments (Jimenez‐Krassel et al.  2009 ; Mossa et al.  2010 ; Scheetz et al.  2012 ) and the results of ovarian stimulation with Follitropin‐V (Ireland et al.  2007 ) are suboptimal in the SORH model. LH secretion on Day 4 or 11 of the estrous cycle in the low‐ or age‐matched high‐AFC heifers. Data represent the SAS adjusted means for LH concentrations determined at 10‐min intervals for 7–11 h for 3–4 animals in each AFC group (adapted from Jimenez‐Krassel et al.  2009 ). Portions of the figure are reproduced from Jimenez‐Krassel et al.  2009 with permission from [Biology of Reproduction, Oxford University Press; https://academic.oup.com/biolreprod ]. Schematic diagram depicting the schedule for all treatments and the rest period during the ovarian stimulation protocols for the SORH model. PG was injected three times (once on Day 1 and twice at 12‐h intervals on Day 11) to induce luteolysis and synchronize estrous cycles. The first injection of cpFSH was given 36 h after the last PG injection (± 1 day of ovulation and emergence of the first follicular wave) followed by seven additional cpFSH injections at 12‐h intervals. To regress the newly formed CL after ovulation and maintain low progesterone concentrations throughout the treatment period, a further three PG injections were given at 12‐h intervals starting at the time of the 7th cpFSH injection (Days 4–5 of estrous cycle). In some studies, a single 2500 IU injection of hCG was given coincident with the third PG injection (12 h after the last (or 8th) cpFSH injection) to induce ovulation. Nine days after hCG‐induced ovulation when development of CL was monitored, two PG injections were administered at 12‐h intervals to regress the CLs and allow a spontaneous, non‐stimulated estrous cycle to occur to allow circulating hormone concentrations and AFC to return to a basal level before the subsequent ovarian stimulation cycle would begin. This timeframe was referred to as the “rest period” and was a minimum of 21 days, with a half‐way check‐in (11 days) ultrasonography scan to monitor that ovary size and AFC were returning to normal, that CL were regressing, and no cystic follicles were present. This cpFSH and PG injection protocol was used in all studies to evaluate FSH action during ovarian stimulation on ovarian function. Portions of the figure are reproduced from Karl et al.  2021 with permission from [Biology of Reproduction, Oxford University Press; https://academic.oup.com/biolreprod ]. High FSH doses during ovarian stimulation are linked with high rates of oocyte wastage (Bianchi et al.  2007 ; Chang et al.  2009 ; French et al.  2010 ; Griesinger, Felberbaum, and Diedrich  2005 ; Hardy  1999 ; Kovalevsky and Patrizio  2005 ; Levi‐Setti et al.  2016 ; Macklon.  2002 ; Meniru and Craft  1997 ; Patrizio and Sakkas  2009 ; Silber et al.  2017 ; Steward et al.  2014 ) and low live birth rates during ART in women (Baker et al.  2015 ; Edwards, Lobo, and Bouchard  1996 ; Klinkert et al.  2005 ; Kuokkanen and Pal  2023 ; Luo et al.  2022 ; Munch et al.  2017 ; Pal et al.  2008 ; Santos, Kuijk, and Macklon  2010 ; Shaia et al.  2020 ; Tan et al.  2005 ; Xu et al.  2024 ). We have established that the SORH model has many phenotypic characteristics also observed for women with DOR (Gleicher, Weghofer, and Barad  2011 ; Gleicher et al.  2013 ; Holte et al.  2011 ; Kawakita et al.  2023 ; Nelson, Yates, and Fleming  2007 ; Patrizio et al.  2015 ; Reame et al.  1996b ; Soares et al.  2020 ; Ulrich and Marsh  2019 ). Because DOR is the major reason for infertility and seeking ART in women (“Assisted Reproductive Technology National Summary Report for 2018,”  2021 ; Devine et al.  2015 ; Oudendijk et al.  2011 ; Patrizio et al.  2015 ), the SORH model was used to test the hypothesis that high FSH doses during ovarian stimulation of individuals with a small ovarian reserve are detrimental to ovulatory follicle function, oocyte quality, and ART outcomes. To test the hypothesis that excessive FSH dosing can adversely affect ART outcomes, numerous precautions were taken in our studies to enhance interpretation of the effects of FSH dose during ovarian stimulation of the SORH model on ovulatory follicle function and oocyte quality (Table  1 ). For example, all heifers were purchased from a single commercial source, fed the same balanced diet, and housed in the same location during treatments. In addition, treatments with cpFSH were not preceded in our studies by nor did they coincide with other hormonal treatments that are sometimes employed in cattle and during ART in women (Jain and Singh  2023 ; SART  2021 ). It is also well‐established that progesterone impacts follicular development and oocyte maturation (Fair and Lonergan  2012 ) and response to FSH treatments (DiZerega and Hodgen  1982 ; Menchaca et al.  2018 ; Peluso and Pru  2014 ; Rivera et al.  2011 ), and progesterone concentrations are low during the follicular phase of the menstrual cycle in women (Ginther et al.  2005 ; Stricker et al.  2006 ). Thus, PG was used in our studies to not only synchronize estrous cycles but to maintain low progesterone concentrations throughout the FSH treatment period (Figure  6 ). Precautions used to improve interpretation of the effects of high FSH doses during ovarian stimulation of the SORH model on ovulatory follicle function and oocyte quality. Healthy, reproductively mature 11‐ to 12‐month‐old registered Holstein heifers 800–900 lbs, consistent low AFC per wave (≤ 15 follicles, range = 4–15) 1,2 and small ovarian reserve 3 . Single commercial source of cpFSH, same diet, and housing location during treatments. Healthy, reproductively mature 11‐ to 12‐month‐old registered Holstein heifers 800–900 lbs, consistent low AFC per wave (≤ 15 follicles, range = 4–15) 1,2 and small ovarian reserve 3 . Single commercial source of cpFSH, same diet, and housing location during treatments. All injections started ±1 day from ovulation, near the time of initiation of the first follicular wave and several days before selection of the dominant follicle (Figure  6 ). Single batch of cpFSH within study. PG used to synchronize estrous cycles before cpFSH stimulation to maintain low (< 0.5 ng/mL) progesterone concentration (Figures  6 and  7 ) 2 . Single 2500 IU injection of hCG, sufficient to ovulate 40–60 follicles in heifers, given 12 h after the last cpFSH injection to assess whether the ovulatory‐size follicle responded to an LH‐like/hCG ovulatory stimulus (Figure  6 ). Heifers treated only with cpFSH, PG or hCG (Figure  6 ). All injections started ±1 day from ovulation, near the time of initiation of the first follicular wave and several days before selection of the dominant follicle (Figure  6 ). Single batch of cpFSH within study. PG used to synchronize estrous cycles before cpFSH stimulation to maintain low (< 0.5 ng/mL) progesterone concentration (Figures  6 and  7 ) 2 . Single 2500 IU injection of hCG, sufficient to ovulate 40–60 follicles in heifers, given 12 h after the last cpFSH injection to assess whether the ovulatory‐size follicle responded to an LH‐like/hCG ovulatory stimulus (Figure  6 ). Heifers treated only with cpFSH, PG or hCG (Figure  6 ). A cross‐over sequence 4,5 of cpFSH doses during each ovarian stimulation protocol was unique for every heifer and each cpFSH dosage followed every other cpFSH dosage an even number of times, which provides control for possible carryover effects, and potential impact of other nuisance variables. A rest period (washout period) used between different cpFSH dosages to minimize carryover effect and ensure return to basal untreated AFC before next series of injections (Figure  6 ). A cross‐over sequence 4,5 of cpFSH doses during each ovarian stimulation protocol was unique for every heifer and each cpFSH dosage followed every other cpFSH dosage an even number of times, which provides control for possible carryover effects, and potential impact of other nuisance variables. A rest period (washout period) used between different cpFSH dosages to minimize carryover effect and ensure return to basal untreated AFC before next series of injections (Figure  6 ). Note: References: 1. (Burns et al.  2005 ). 2. (Karl et al.  2021 ). 3. (Ireland et al.  2008 ). 4. (Williams, 1949). 5. (Wang, 2009 #93). Our studies (Clark et al.  2022a ; Clark et al.  2022b ; Karl et al.  2022 ; Karl et al.  2021 ; Karl et al.  2023 ) determined whether excessive cpFSH doses during ovarian stimulation of the SORH model altered: (i) ovarian function (Karl et al.  2021 ), (ii) ovulatory follicle function (Clark et al.  2022a ), (iii) expression of genes critical for ovulatory follicle function and oocyte quality (Clark et al.  2022b ), (iv) cumulus function and oocyte quality (Karl et al.  2023 ), and (v) reliability of the ovarian reserve biomarkers, AFC and AMH, as predictors of responsiveness to ovarian stimulation with cpFSH (Karl et al.  2022 ). A summary of outcomes of excessive FSH action during ovarian stimulation of the SORH model is provided in Table  2 . Ex‐cpFSH‐induced phenotypic changes during ovarian stimulation of the SORH model. No further increase in the number or size of ovulatory‐size follicles (Figure  7A ) Decreased estradiol production and circulating estradiol concentrations (Figure  7A ) Decreased ovulation rate (Figure  7B ) No further increase in the number or size of ovulatory‐size follicles (Figure  7A ) Decreased estradiol production and circulating estradiol concentrations (Figure  7A ) Decreased ovulation rate (Figure  7B ) Follicle: High proportion of estrogen‐inactive vs. ‐active follicles that had higher intrafollicular concentrations of progesterone than estradiol and high intrafollicular concentration of oxytocin (Figure  8 ) Characteristics of premature luteinization (e.g., elevated expression of a subset of luteinization marker genes, low capacity to produce estradiol, high intrafollicular progesterone to estradiol ratio, and high intrafollicular oxytocin concentration) (Figure  8 ) Follicular Hyperstimulation Dysgenesis (Figures  8 and  9 ) Follicle: High proportion of estrogen‐inactive vs. ‐active follicles that had higher intrafollicular concentrations of progesterone than estradiol and high intrafollicular concentration of oxytocin (Figure  8 ) Characteristics of premature luteinization (e.g., elevated expression of a subset of luteinization marker genes, low capacity to produce estradiol, high intrafollicular progesterone to estradiol ratio, and high intrafollicular oxytocin concentration) (Figure  8 ) Follicular Hyperstimulation Dysgenesis (Figures  8 and  9 ) Disruption of multiple cell‐signaling pathways in granulosa and cumulus cells and oocytes critical for folliculogenesis, steroidogenesis, luteinization, cell survival, ovulation, and oocyte maturation and quality resulting in Follicular Hyperstimulation Dysgenesis in all ovulatory‐size follicles (Figures  8 and  9 ) Cumulus cell‐expressed genes elevated in all follicles assayed (Figure  9 ) Disruption of multiple cell‐signaling pathways in granulosa and cumulus cells and oocytes critical for folliculogenesis, steroidogenesis, luteinization, cell survival, ovulation, and oocyte maturation and quality resulting in Follicular Hyperstimulation Dysgenesis in all ovulatory‐size follicles (Figures  8 and  9 ) Cumulus cell‐expressed genes elevated in all follicles assayed (Figure  9 ) Premature cumulus cell expansion in vivo (Figure  10 ) Premature resumption of meiosis in vivo (Figure  10 ) Overexpression of cumulus cell genes critical for cumulus function and oocyte maturation in all ovulatory‐size follicles (Figure  9 ) Premature cumulus cell expansion in vivo (Figure  10 ) Premature resumption of meiosis in vivo (Figure  10 ) Overexpression of cumulus cell genes critical for cumulus function and oocyte maturation in all ovulatory‐size follicles (Figure  9 ) Antral follicle count and anti‐Müllerian Hormone concentrations prior to ovarian stimulation were correlated with the number, but not quality of ovulatory‐size follicles (Figure  11 ) Antral follicle count and anti‐Müllerian Hormone concentrations prior to ovarian stimulation were correlated with the number, but not quality of ovulatory‐size follicles (Figure  11 ) Reduced capacity to undergo COC expansion in response to hCG (Figure  12 ) Reduced capacity to resume meiosis during in vitro maturation in response to hCG 1 Reduced capacity of oocytes to complete progression to MII stage in vivo and in vitro (Figure  10 ) Reduced capacity of COCs to undergo in vitro fertilization (Figure  10 ) Reduced capacity to undergo COC expansion in response to hCG (Figure  12 ) Reduced capacity to resume meiosis during in vitro maturation in response to hCG 1 Reduced capacity of oocytes to complete progression to MII stage in vivo and in vitro (Figure  10 ) Reduced capacity of COCs to undergo in vitro fertilization (Figure  10 ) Note: References: 1. (Karl et al.  2023 ). The effects of different cpFSH doses on ovarian function were examined to determine if the relatively poor ovarian response of the SORH model to ovarian stimulation with cpFSH observed in our previous study (Ireland et al.  2007 ) was improved with higher or perhaps even lower FSH doses. Resolving this question had significant clinical relevance because a high proportion of women seeking ART not only also have a small ovarian reserve (“Assisted Reproductive Technology National Summary Report for 2018,”  2021 ; Devine et al.  2015 ; Oudendijk et al.  2011 ; Patrizio et al.  2015 ), but are usually treated with very high FSH doses during ovarian stimulation protocols to overcome their poor responsiveness (Figure  1 ; Baker et al.  2015 ). In this study (Karl et al.  2021 ), the SORH model was subjected to four different cpFSH doses (35 IU, 70 IU, 140 IU, 210 IU) reflecting a sixfold dose range during ovarian stimulation followed by an ovulatory dose of hCG (Figure  6 ). The 70 IU dose is near the recommended industry standard dose of 87.5 IU dose (Freedom of Information Summary NADA 141‐431 Folltropin  2014 ) and is hereafter referred to as IS‐cpFSH. The cpFSH dose range used in our studies was much narrower than the 20‐fold range in doses of rhFSH injected into women during ART (Clark et al.  2021 ). While the two highest cpFSH doses did not further increase number of ovulatory‐size follicles during ovarian stimulation (Figure  7 , top panel), these doses resulted in 40% to 60% decreases in circulating estradiol concentrations (Figure  7 , lower panel) and ~50% decline in ovulation rate in response to hCG (Figure  7 ) compared with the two lowest cpFSH doses (Karl et al.  2021 ). The different cpFSH doses did not alter circulating concentrations of AMH or progesterone, which remained low and unaltered throughout the study (data not shown, (Karl et al.  2021 ). These results indicated that the highest cpFSH doses did not disproportionately increase circulating concentrations of AMH or progesterone during ovarian stimulation, which in turn could have impaired FSH‐induced follicular growth and function (DiZerega and Hodgen  1982 ; Durlinger et al.  2002 ; Pellatt et al.  2011 ; Peluso and Pru  2014 ) and perhaps explained the reduction in estradiol production observed in this study (Figure  7 ). Effect of different cpFSH doses on number of ovulatory‐size follicles, circulating estradiol concentrations, and ovulation rate in the SORH model (modified from Karl et al.  2021 ). (A) Heifers were injected (depicted by arrows) beginning on Day 1 of the estrous cycle with four different doses (35 IU ●, 70 IU □, 140 IU ○, 210 IU ♦) of cpFSH and treated with PG and hCG (depicted by arrows) as explained in the legend for Figure  5 . Symbols depict means (± SEM) for the same eight heifers. The number of ovulatory‐size follicles (≥ 10 mm) were determined by daily ultrasonography measurements. Estradiol concentrations were determined at 24‐h intervals. (B) Ovulation rate was determined by dividing the number of corpora lutea on Day 7 post hCG by number of ovulatory‐size follicles at time of hCG. Asterisks in indicate means (± SEM) for number of follicles or estradiol concentrations differed (* =  p  < 0.05, ** =  p  < 0.01) when compared with the 210 IU dose for Panel A, or pooled means (± SEM) for ovulation rates differed (*** =  p  < 0.001) for Panel B. Portions of the figure are reproduced from Karl et al.  2021 with permission from [Biology of Reproduction, Oxford University Press; https://academic.oup.com/biolreprod ]. Estradiol production and responsiveness to an LH/hCG stimulus and ovulation are well‐established hallmarks of ovulatory follicle function. Thus, the significant reduction in both parameters by the highest cpFSH doses in our study (Karl et al.  2021 ) demonstrated these doses were not only excessive and wasteful from an economic viewpoint, but detrimental to ovulatory follicle function in the SORH model. From a clinical perspective, determining whether a dose of FSH is excessive and detrimental during ovarian stimulation poses practical challenges. Not only does it necessitate an FSH dose‐response protocol and use of a cross‐over experimental design (as employed in our study; Karl et al.  2021 ), but the excessive dose threshold is also likely to vary between individuals. Consequently, identifying reliable clinical markers for excessive FSH action would be valuable. Such markers could optimize ovarian stimulation protocols, reduce costs of ART, and ultimately improve ART outcomes. A series of additional studies (Clark et al.  2022a ; Clark et al.  2022b ; Karl et al.  2023 ) were conducted to develop a better understanding of the precise reason the excessive cpFSH doses impaired ovarian function. Healthy, growing dominant follicles during the follicular waves of unstimulated reproductive cycles in cattle have higher estradiol than progesterone concentrations and are classified as estrogen‐active (EA; Ireland and Roche  1982 ,  1983a ,  1983b ; Sunderland et al.  1994 ). In contrast, the dominant follicles that have ceased to grow during non‐ovulatory follicular waves and are destined for atresia have higher progesterone than estradiol concentrations and are classified as estrogen‐inactive (EI; Ireland and Roche  1982 ,  1983a ,  1983b ; Sunderland et al.  1994 ). The small growing antral follicles that comprise the first follicular wave differ in age, size and stage of differentiation (McNatty et al.  2010 ) as does the capillary network within the thecal layers of cells of each growing follicle (Figure  8 ; Acosta et al.  2005 ; Fraser  2006 ). Because of these differences in vasculature, the amounts of cpFSH interacting with the FSH receptors on the granulosa and cumulus cells must also be unequally distributed among the follicles in the wave. The unequal intrafollicular cpFSH distribution is hypothesized to induce development of the different phenotypes for the ovulatory‐size follicles observed in our study (Clark et al.  2022a ) as explained in Figure  8 . Schematic diagram illustrating that exposure of small antral follicles of the first follicular wave to varying amounts of cpFSH during ovarian stimulation resulted in the development of heterogenous ovulatory‐size follicle phenotypes (as modified from Clark et al.  2022a ). The SORH model ( n  = 6 or 7/dose) was injected (arrows) with the industry‐standard (IS‐cpFSH) or excessive (Ex‐cpFSH, 3X IS‐cpFSH doses) cpFSH doses and subjected to ovariectomy (Ovx) 12 h after the last cpFSH injection (before hCG) as explained in Figure  5 's legend. Green balls depict an FSH molecule. All ovulatory‐size follicles (≥ 10 mm) were excised, and follicular fluid (FF) removed to determine concentrations of estradiol ( E ), progesterone ( P ) and oxytocin ( O ). Size of hormone symbol reflects relative intrafollicular hormone concentration (Clark et al.  2022a ). The COCs recovered from FF were classified as compact or expanded as depicted by size of the cumulus layer around oocyte within each ovulatory‐size follicle (Clark et al.  2022a ). Each animal had ~1.5 dozen ovulatory‐ size follicles averaged ~12 mm in diameter at ovariectomy (Clark et al.  2022a ). The phenotype for each ovulatory‐size follicle was based E:P and O concentrations in FF and cumulus morphology. Percent reflects proportion of each ovulatory‐size follicle phenotype per animal . The different phenotypes are hypothesized to result from the unequal distribution of the total amount of cpFSH (green balls) entering the ovarian arteries and thecal capillary system of each different sized small antral follicle in the first follicular wave as they develop to ovulatory‐size during ovarian stimulation. Number adjacent to each ovulatory‐size follicle phenotype (Types 1, 2, 3, and 4) indicate that the granulosa and cumulus cells and oocytes for this phenotype were subjected to RNAseq and bioinformatic analyses. Portions of the figure are reproduced from Clark et al.  2022a with permission from [Biology of Reproduction, Oxford University Press; https://academic.oup.com/biolreprod ]. To directly examine the impact of excessive cpFSH doses on ovarian function in the SORH model, the phenotype for each individual ovulatory‐size follicle was established after the last cpFSH injection as illustrated in Figure  8 . This was accomplished (Clark et al.  2022a ) by measuring well‐established intrafollicular markers of follicular function and oocyte maturation for the ovulatory follicles in cattle, including estradiol (Ireland and Roche  1982 ,  1983a ,  1983b ; Sunderland et al.  1994 ), progesterone (Ireland and Roche  1982 ,  1983a ,  1983b ; Sunderland et al.  1994 ), oxytocin (Murphy  2000 ), and cumulus cell expansion (Assidi, Richard, and Sirard  2013 ). The intrafollicular ratio of estradiol:progesterone was also determined for each excised ovulatory‐size follicle to evaluate its health and function following ovarian stimulation with cpFSH (Clark et al.  2022a ). The intrafollicular concentrations of cpFSH for heifers treated with IS‐cpFSH during ovarian stimulation, as depicted in Figure  8 , are hypothesized to be within a physiological range for FSH concentrations thereby stimulating development of multiple predominantly healthy, EA ovulatory‐size follicles with low oxytocin levels and compact cumulus‐oocyte complexes (COCs) like ovulatory follicles in cattle (Ireland and Roche  1982 ,  1983a ,  1983b ; Sunderland et al.  1994 ), as observed in our study (Clark et al.  2022a ). However, when the amounts of intrafollicular cpFSH fall below the physiological threshold for FSH concentrations, as shown in Figure  8 , they cannot sustain the estradiol‐producing capacity of the ovulatory‐size follicles resulting in development of the EI phenotype for the IS‐cpFSH treated animals (Clark et al.  2022a ). The EI phenotype is typical for dominant follicles destined to undergo atresia as circulating FSH concentrations decline during non‐ovulatory follicular waves of estrous cycles in cattle (Ireland and Roche  1982 ,  1983a ,  1983b ; Sunderland et al.  1994 ). In contrast, the threefold higher doses of cpFSH (i.e., excessive, Ex‐cpFSH) during ovarian stimulation are above the threshold for the normal physiological range of FSH concentrations resulting in growth of predominantly dysfunctional EI ovulatory‐size follicles with expanded COCs as depicted in Figure  8 . Note that the most extreme ovulatory‐size follicle phenotype with the highest intrafollicular oxytocin concentrations is predicted to be linked to the highest intrafollicular cpFSH concentration (Figure  8 ). The morphological analysis of cumulus‐oocyte complexes in individual ovulatory‐size follicles revealed that, in the absence of an hCG‐ovulatory stimulus, the use of excessive cpFSH doses during ovarian stimulation resulted in development of predominantly dysfunctional ovulatory‐size follicles. The majority of the Ex‐cpFSH‐induced ovulatory‐size follicle phenotypes contained prematurely expanded COCs not observed in the IS‐cpFSH treated animals (Figure  8 ; Clark et al.  2022a ). This discovery suggested that the Ex‐cpFSH doses in the absence of an hCG‐ovulatory stimulus may have prematurely initiated oocyte maturation and thus impaired oocyte quality. We also discovered that nearly half of all the Ex‐cpFSH‐induced ovulatory‐size follicles were dysfunctional and exhibited a low capacity to produce estradiol, high intrafollicular progesterone:estrogen ratio, and high intrafollicular oxytocin concentration, which are symptomatic of premature luteinization. The high proportion of Ex‐cpFSH‐induced dysfunctional follicles observed in this study (Clark et al.  2022a ) also explains why circulating concentrations of estradiol were lower during ovarian stimulation of the SORH model with the Ex‐cpFSH as compared to IS‐cpFSH doses in our previous study (Figure  7 ; Karl et al.  2021 ). To directly evaluate if excessive FSH action induces granulosa cell luteinization in cattle, granulosa cells were isolated from small antral follicles of low‐ or high‐AFC cows and treated with a wide range of cpFSH doses (Scheetz et al.  2012 ). The small antral follicles used as the source of granulosa cells in this study (Scheetz et al.  2012 ) were similar in size to the small follicles at the beginning of a follicular wave in cattle, which develop into ovulatory‐size follicles during ovarian stimulation with FSH (Figure  2 ). The capacity of granulosal cells from these follicles to produce estradiol in response to FSH treatments was consistently lower for the low‐ versus high‐AFC cows (Figure  13 ). In addition, there was a distinct biphasic responsiveness of granulosa cells to the different FSH doses independent of AFC. For example, the relatively low FSH doses increased estradiol production in a linear fashion (Figure  13 ) while higher doses did the opposite. Other well‐established targets of FSH action also followed a similar biphasic response pattern like estradiol, including AMH production and expression of mRNAs for CYP19A1 , AMH , or FSHR (data not shown) (Scheetz et al.  2012 ). The peak responsiveness of granulosa cells to FSH, however, occurred at a lower FSH dose in the low‐ vs high‐AFC cows (e.g., at 0.1 vs. 0.5 ng/mL FSH; Figure  13 ). This finding demonstrated that the “window” of positive responsiveness of granulosa cells to FSH, as measured by estradiol production and other well‐established markers of FSH action, was restricted to relatively low FSH doses especially for the low‐AFC cows. The biphasic response of granulosa cells to FSH also showed that the highest FSH doses decreased the capacity of granulosa cells to produce estradiol (Figure  13 ) while increasing production of progesterone and the expression of mRNA for oxytocin (Figure  14 ). These alterations in estradiol, progesterone and oxytocin production are well‐established as markers for luteinization of granulosa cells in cattle and other species (Murphy  2000 ). Consequently, the relatively high FSH doses exceeding the threshold of positive FSH responsiveness mimicked LH action and induced granulosa cell luteinization as shown by others using a variety of models (Galway et al.  1990 ; Inaba et al.  1998 ; Tapanainen et al.  1993 ; Ubaldi et al.  1996 ; Zhan et al.  2024 ). These in vitro results demonstrated directly that excessive FSH doses not further enhancing estradiol production abrogated FSH action resulting, for example, in loss of the estradiol‐producing capacity and the luteinization of granulosa cells. The pronounced detrimental effects of excessive FSH action on granulosa cells in vitro provided compelling evidence that the Ex‐cpFSH doses used during ovarian stimulation of the SORH model induced premature luteinization of the granulosa cells in a high proportion of ovulatory‐size follicles per heifer, resulting in the decline in circulating (Figure  7 ) and intrafollicular estradiol concentrations, increased intrafollicular progesterone and oxytocin production (Figure  13 ), and decreased ovulation rates (Figure  8 ; Clark et al.  2022a ; Karl et al.  2021 ). The Ex‐cpFSH doses during ovarian stimulation of the SORH model reduced responsiveness of the ovulatory‐size follicles to an hCG‐ovulatory stimulus (Karl et al.  2021 ). Although it was not determined which ovulatory‐size follicle phenotypes (Figure  8 ) failed to ovulate, it is highly likely that independent of the cpFSH doses used during ovarian stimulation, the dysfunctional EI ovulatory‐size follicles undergoing premature luteinization, which comprised 21% and 44% of the ovulatory‐size follicles in the IS‐cpFSH‐ and Ex‐cpFSH‐treated cattle, respectively (Figure  8 ; Clark et al.  2022a ), were refractory to hCG stimulation (12 h after the last cpFSH injection) and incapable of ovulation as observed in our previous study (Figure  7 ; Karl et al.  2021 ). If this assumption is correct, this would explain why hCG‐induced ovulation rate was lowest in the Ex‐cpFSH‐treated heifers (Figure  7 ; Karl et al.  2021 ), which had the highest proportion of EI ovulatory‐size follicles (Figure  8 ; Clark et al.  2022a ). The mechanisms explaining why excessive cpFSH action during ovarian stimulation of the SORH model was detrimental to ovulatory follicle function and oocyte quality were examined by determining if the excessive cpFSH doses during ovarian stimulation altered gene expression in signaling pathways critical for ovulatory follicle function and oocyte quality. This was accomplished by subjecting granulosa and cumulus cells and oocytes from a subset (4–6 ovulatory‐size follicles, or heifers, per cell type per dose) of the excised ovulatory‐size follicles (Type 1 vs. 2, 3, 4 depicted in Figure  8 ) obtained in our previous study (Clark et al.  2022a ) to RNAseq and bioinformatic analyses (Clark et al.  2022b ). Types 1 and 2 follicles were highly similar to each other and presumably represent the most healthy, normal follicles having cCOC and EA characteristics, with the only difference being the cpFSH dose used during ovarian stimulation, 70 and 210 IU, respectively. With progressively more severe abnormalities, Types 3 and 4 follicles contained eCOC, and Type 4 follicles additionally displayed an EI endocrine profile, making them phenotypically the most divergent from Type 1 follicles. The predominant phenotype (EA with low oxytocin and compact COCs) observed in the SORH model treated with the IS‐cpFSH doses was considered the control (Type 1, Figure  8 ) in this study because it resembled ovulatory follicles during estrous cycles in untreated cattle (Ireland and Roche  1982 ,  1983a ,  1983b ; Sunderland et al.  1994 ). Of the ~22,000 genes in the bovine genome (Burt  2009 ), mRNA expression for ~15,000 genes was detected during RNA sequencing (RNAseq). Of these genes, differentially expressed genes (DEGs) were identified in Type 4 ovulatory‐size follicles for cumulus cells ( n  = 3131), granulosa cells ( n  = 1368), and oocyte ( n  = 283) (Clark et al.  2022b ). The EI ovulatory‐size follicles, which include the extreme Type 4 follicles, comprised 44% of the ovulatory‐size follicles developing in response to Exp‐cpFSH during ovarian stimulation of the SORH model (Clark et al.  2022a ). The DEGs were subjected to Ingenuity Pathway Analysis (IPA) and IPA Causal Network Analysis. The results predicted that Ex‐cpFSH‐induced activation of FSH signaling in granulosa and cumulus cells, coupled with the decreased activation of estradiol signaling in oocytes, altered the activities of numerous signaling pathways critical for folliculogenesis, steroidogenesis, luteinization, cell survival, ovulation, and oocyte maturation and oocyte quality (Clark et al.  2022b ). Importantly, pathway analysis indicated drastic alterations in ovulatory follicle function that do not follow the expected sequence for normal follicular developmental progression, such as concurrent activation of ovulation and luteinization rather than these processes occurring sequentially in the most severely affected follicles. The analysis also indicated predicted deficiencies in oocyte survival, stress response, and fertilization and likely reductions in oocyte health. These changes in gene expression combined with other follicular abnormalities indicated that Ex‐cpFSH was inducing a state of FHD. FHD was concluded to have contributed to, or caused, the reduced intrafollicular estradiol but enhanced progesterone and oxytocin production, premature luteinization, premature cumulus expansion, reduced responsiveness to hCG and ovulation rate, and potentially impaired oocyte quality observed in most ovulatory‐size follicles developing in response to Ex‐cpFSH doses during ovarian stimulation of the SORH model (Clark et al.  2022a ; Clark et al.  2022b ; Karl et al.  2021 ; Karl et al.  2023 ). Such effects could also contribute to adverse effects of high FSH doses during ART procedures. Because excessive cpFSH doses during ovarian stimulation of the SORH model resulted in FHD and premature cumulus expansion in nearly all ovulatory‐size follicles, which could impact oocyte quality (Clark et al.  2022a ; Clark et al.  2022b ), the RNAseq data (Clark et al.  2022b ) were interrogated to determine if the Ex‐cpFSH impacted genes critical for cumulus function and oocyte maturation (Figure  9 ). Ex‐cpFSH resulted in significant overexpression of the 17 genes in cumulus but not granulosa cells in all ovulatory‐size follicle phenotypes in the absence of an hCG‐ovulatory stimulus (Karl et al.  2021 ). The degree of overexpression of these cumulus genes was greatest in the Type 4 phenotype and this highlights the potential mechanisms by which Ex‐cpFSH actions may impair ART outcomes. Effect of treatment of granulosa cells with FSH on estradiol production (adapted from Scheetz et al.  2012 ). Cells isolated from 3 to 5 mm antral follicles of cows with a low‐ or high‐AFC were cultured serum‐free for 6 days. Each bar represents the mean ± SEM for 3 pools of cells from 3 to 5 cows per pool. Portions of the figure are reproduced from Scheetz et al.  2012 with permission from [Reproduction, Fertility and Development, CSIRO; https://www.publish.csiro.au/rd ]. A literature review indicated that all the overexpressed cumulus genes identified in our analysis have well‐established roles in regulation of cumulus function and oocyte maturation, such as responsiveness to gonadotropins ( GPR50, RGS2 ), cell survival ( IGFBP5 ), extracellular matrix ( PLAT, MEPE ), cumulus expansion ( AREG , PTX3, MGAT5 ), calcium movement ( VGF, NCS1, RGS2 ), tight junctions and cell communication ( CLDN11, NGFR ), ovulation ( MDF1 ), clotting regulation ( FGG ), and resumption of meiosis ( IGFBP3 , TGFα, LIF ) (Karl et al.  2023 ). It is important to note that the cumulus gene overexpression depicted in Figure  9 was observed even in follicles that did not manifest premature cumulus expansion or disruption in hormone production, such as those seen in the Types 3 and 4 ovulatory‐size follicles (Karl et al.  2023 ). Such changes indicate that disruption in the regulation of these cumulus cell‐expressed genes may initiate oocyte wastage. Our studies using the SORH model (Clark et al.  2022a ; Karl et al.  2023 ) revealed that excessive FSH doses during ovarian stimulation induced premature cumulus expansion and overexpression of cumulus cell expressed genes critical for cumulus function and oocyte maturation in the absence of an ovulatory‐hCG dose in nearly all ovulatory‐size follicles. Moreover, the excessive cpFSH doses impaired responsiveness of ovulatory‐size follicles (likely the EI dysfunctional phenotype; Figure  8 ) to an ovulatory dose of hCG which significantly reduced ovulation rates (Karl et al.  2021 ). Whether these alterations in cumulus gene expression and morphology coupled with the insensitivity of ovulatory‐size follicles to hCG impacted oocyte maturation and resumption of meiosis was not examined (Clark et al.  2022a ). To address this question, ovarian stimulation with the IS‐cpFSH or Ex‐cpFSH doses was applied to the SORH model and COCs recovered from ovulatory‐size follicles of unknown phenotypes (Figure  12 ) by oocyte pick up (OPU). OPU was conducted 12 h after the last cpFSH injection (with no hCG injection) or 24 h after the ovulatory dose of hCG but before ovulation (Figure  6 ; Karl et al.  2023 ). Before the ovulatory hCG injection, expanded COCs were rarely observed in ovulatory‐size follicles in the IS‐cpFSH treated heifers (Karl et al.  2023 ) as previously observed using excised follicles (Clark et al.  2022a ). However, these heifers had a relatively high proportion of the ovulatory‐size follicles with expanded COCs after the hCG ovulatory dose (Karl et al.  2023 ), as depicted in Figure  12 as would be expected following a preovulatory gonadotropin surge during the estrous cycle in cattle. Surprisingly, we also discovered that after hCG ~50% of the ovulatory‐size follicles had compact COCs, failing to respond appropriately to hCG. This observation implies that even the IS‐cpFSH doses may be excessive and potentially detrimental to ovulatory follicle function although this has not been directly evaluated in our studies. Most notable was the observation that expanded COCs were observed in 22%–32% of ovulatory‐size follicles in the Ex‐cpFSH‐treated heifers compared with near zero proportion in the IS‐cpFSH‐treated heifers in the absence of an ovulatory stimulus (Figure  12 ; Karl et al.  2023 ). In addition, ~50% of the expanded COCs in the Ex‐cpFSH‐treated heifers had oocytes at the MI or MII stage of nuclear maturation (Karl et al.  2023 ). This observation implied that the excessive cpFSH doses during ovarian stimulation of the SORH model caused premature cumulus expansion and resumption of meiosis in the absence of an hCG/LH‐like stimulus. We also observed that the proportions of ovulatory‐size follicles with expanded COCs in the Ex‐cpFSH‐treated heifers were similar before and after hCG (Figure  12 ). These findings implied that the Ex‐cpFSH doses impaired the capacity of the compact COCs in the ovulatory‐size follicles to respond to the hCG‐ovulatory dose. Insensitivity to the hCG stimulus was also observed in the Ex‐cpFSH‐treated heifers, which had a reduced hCG‐induced ovulation rate compared with IS‐cpFSH‐treated heifers in our previous study (Figure  7 ; Karl et al.  2021 ). The proportions of ovulatory‐size follicles per heifer with expanded COCs observed following OPU (24%; Figure  12 ) were lower than that reported for excised follicles (~75%; Figure  8 ) in our previous study (Clark et al.  2022a ). We suspect this discrepancy between studies was because the OPU technique used to recover COCs from the ovulatory‐size follicles of non‐anaesthetized cattle is inefficient (as low as 55% recovery rate; (Clark et al.  2022a ) and harsh, resulting in more COC fragmentation compared with the use of a needle and syringe to recover COCs from excised follicles (Karl et al.  2023 ). Nevertheless, the OPU study confirmed that Ex‐cpFSH doses induced premature cumulus expansion in ovulatory‐size follicles in the absence of an hCG‐ovulatory stimulus. The impact of the Ex‐cpFSH doses during ovarian stimulation on oocyte quality was examined in the SORH model in two ways (Karl et al.  2023 ). First, we determined whether the Ex‐cpFSH‐induced prematurely expanded COCs were fertilizable during IVF without in vitro maturation (IVM). Second, we examined whether the Ex‐cpFSH doses altered capacity of the oocytes in the compact or expanded COCs to undergo nuclear maturation during IVM. The Ex‐cpFSH‐induced prematurely expanded COCs were not fertilizable during IVF (without IVM; Figure  10 ). In addition, the proportion of ovulatory‐size follicles with compact COCs and at the MII stage of nuclear maturation during IVM was significantly decreased by Ex‐cpFSH‐ as compared to IS‐cpFSH‐treated (Figure  10 ). Moreover, compared with compact COCs from either the IS‐cpFSH‐ or the Ex‐cpFSH‐treated heifers, the prematurely expanded COCs from the Ex‐cpFSH‐treated heifers had a significantly lower proportion of ovulatory‐size follicles with oocytes at the MII stage of nuclear maturation but a significantly higher proportion of oocytes classified as degenerated (Figure  10 ). It is noted that the precise ovulatory‐size follicle phenotype from which each COC was removed from during OPU could not be determined in our study (Karl et al.  2023 ). Nevertheless, these results indicate that the excessive cpFSH doses not only cause ovulatory follicle dysfunction but also impair cumulus function and oocyte maturation including responsiveness to an LH‐like/hCG ovulatory stimulus. Effect of FSH on granulosa cell progesterone production (top panel) and expression of oxytocin mRNA expression (lower panel, adapted from Scheetz et al.  2012 ). See Figure  8 's legend for details. Portions of the figure are reproduced from Scheetz et al.  2012 with permission from [Reproduction, Fertility and Development, CSIRO; https://www.publish.csiro.au/rd ]. Schematic diagram depicting Ex‐cpFSH‐induced overexpression of 17 cumulus cell genes in the Types 2, 3, and 4 versus Type 1 ovulatory‐size follicle phenotypes with well‐established roles in FSH action, cumulus expansion and oocyte maturation (adapted from Clark et al.  2022a ; Clark et al.  2022b ; Karl et al.  2023 ). This study (Karl et al.  2023 ) manually interrogated the RNAseq data generated in our previous study (Clark et al.  2022b ) to determine if the excessive FSH doses during ovarian stimulation of the SORH model significantly altered expression of cumulus genes important for regulation of cumulus function and oocyte maturation. The SORH model was treated with the different cpFSH doses and ovulatory‐size follicles excised and subjected to RNAseq and bioinformatic analyses as explained in Figure  7 's legend (Karl et al.  2023 ). RNAseq and bioinformatic results indicated that the same 17 cumulus genes were differentially expressed in Types 2, 3, and 4 compared with Type 1 (control) ovulatory‐size follicles (Karl et al.  2023 ). Portions of the figure are reproduced from (Karl et al.  2023 ) with permission from [Molecular Human Reproduction, Oxford University Press, https://academic.oup.com/molehr ]. Schematic diagram depicting morphology for the cumulus‐oocyte complexes recovered before or after an ovulatory dose of hCG by oocyte pick up (OPU) from the different ovulatory‐size follicle phenotypes developing in response to different cpFSH doses during ovarian stimulation (adapted from Clark et al.  2022a ; Karl et al.  2021 ; Karl et al.  2023 ). Percent beside each ovulatory‐size follicle phenotype reflects proportion per animal (Clark et al.  2021 ). The SORH model was subjected to ovarian stimulation with the 70 IU IS‐cpFSH or 210 IU Ex‐cpFSH cpFSH dose at 12‐h intervals and an ovulatory dose of hCG (Karl et al.  2021 ) was injected 12 h after the last cpFSH dose as explained in Figure  5 's legend. Cumulus‐oocyte complexes (COCs) were removed from each follicle by oocyte pick up (OPU) 12 h after the last cpFSH injection (before hCG) or 22 h after the hCG injection before ovulation. The recovery rate of COCs from ovulatory‐size follicles was ~55% (Karl et al.  2023 ). Percents below COCs reflect the proportion of ovulatory‐size follicles per heifer that had COCs classified as compact or expanded (Clark et al.  2022a ; Karl et al.  2023 ). Portions of the figure are reproduced from Clark et al.  2022a ; Karl et al.  2021 ; Karl et al.  2023 with permission from [Biology of Reproduction, Oxford University Press; https://academic.oup.com/biolreprod ]. Schematic diagram depicting different ovulatory‐size follicle phenotypes and capacity of compact or prematurely expanded cumulus‐oocyte complexes (COCs) developing in response to ovarian stimulation with different cpFSH doses to undergo in vitro fertilization (IVF) or resume meiosis during in vitro maturation (IVM) (adapted from Clark et al.  2022a ; Karl et al.  2021 ; Karl et al.  2023 ). Percent beside each ovulatory‐size follicle phenotype reflects proportion per animal (Clark et al.  2021 ). The SORH model was subjected to ovarian stimulation with the 70 IU IS‐cpFSH or 210 IU Ex‐cpFSH cpFSH dose at 12‐h intervals (Karl et al.  2021 ) and oocyte pick up (OPU) was conducted 12 h after the last cpFSH dose as explained in Figure  5 's legend. COCs were removed from each follicle by OPU 12 h after the last cpFSH injection (before hCG). The IVF arrow indicates that the prematurely expanded COCs were subjected to IVF without IVM resulting in 5% and 0% rate of IVF and blastocyst formation, respectively, compared with 68% and 28% rate of IVF and blastocyst formation, respectively, for controls (Karl et al.  2023 ). COCs were also subjected to IVM and stage of nuclear maturation determined (Karl et al.  2023 ). The IVM arrows reflect proportions of COCs at MII or degenerated (skull symbol) after IVM. Portions of the figure are reproduced from Clark et al.  2022a ; Karl et al.  2021 ; Karl et al.  2023 with permission from [Biology of Reproduction, Oxford University Press; https://academic.oup.com/biolreprod ]. Schematic diagram illustrating that circulating concentrations of anti‐Müllerian hormone (AMH) and antral follicle count (AFC) were highly positively correlated independent of the cpFSH dose used during ovarian stimulation of the SORH model with the number but not function (e.g., phenotype) of the ovulatory‐size follicles. AMH concentration and AFC were determined in the SORH model before ovarian stimulation and correlated with the peak number of ovulatory‐size follicles developing during ovarian stimulation (Karl et al.  2022 ). The phenotypes for the ovulatory‐size follicles and proportions (%) per heifer that developed in response to ovarian stimulation with the industry‐standard (IS‐cpFSH) or excessive (Ex‐cpFSH) dose was based on results of our previous study using the SORH model (Figure  7 ; Clark et al.  2022a ). Portions of the figure are reproduced from (Clark et al.  2022a ; Karl et al.  2022 ) with permission from [Theriogenology, Elsevier; https://www.sciencedirect.com/journal/theriogenology and Biology of Reproduction, Oxford University Press; https://academic.oup.com/biolreprod ]. In summary, Ex‐cpFSH leads to premature COC expansion in vivo without an ovulatory stimulus, premature resumption of meiosis in vivo, but reduced capacity of expanded COCs to complete progression to MII stage of nuclear maturation in vivo or in vitro. Based on these combined results of our studies (Clark et al.  2022a ; Clark et al.  2022b ; Karl et al.  2021 ; Karl et al.  2023 ), we concluded that the Ex‐cpFSH doses during ovarian stimulation of the SORH model cause FHD resulting in premature luteinization, overexpression of cumulus genes that normally regulate cumulus function and resumption of meiosis, and inhibition of the responsiveness of ovulatory‐size follicles to hCG. This syndrome of excessive FSH‐induced alterations in ovulatory follicle function raises the risk of recovery of a high proportion of poor‐quality prematurely expanded COCs (incapable of being fertilized) that are morphologically indistinguishable from healthy hCG‐matured expanded COCs for IVF during ART. Ovarian stimulation of the SORH model with excessive FSH doses results in development of many dysfunctional ovulatory‐size follicles (Clark et al.  2022a ; Clark et al.  2022b ; Karl et al.  2021 ; Karl et al.  2023 ). However, it is unknown whether the use of excessive cpFSH doses during ovarian stimulation impact reliability of the well‐established ovarian reserve biomarkers, AFC and AMH, to predict responsiveness of the SORH model to ovarian stimulation. To address this question, AFC and AMH concentration were determined before ovarian stimulation of the SORH model with four different cpFSH doses, which included the industry‐standard and excessive cpFSH doses depicted in Figure  6 (Karl et al.  2022 ), and correlated with the peak number of functional and dysfunctional ovulatory‐size follicles (determined 12 h after the last cpFSH injection) developing in response to ovarian stimulation with the same four cpFSH doses. Functional ovulatory‐size follicles were defined as those ovulating in response to an ovulatory dose of hCG whereas the number of dysfunctional follicles was determined by subtracting the total number of CL that formed following hCG from total number of ovulatory‐size follicles developing during ovarian stimulation (Karl et al.  2022 ). The number of follicles that failed to ovulate were considered dysfunctional. Numerous precautions were taken in this study (Karl et al.  2022 ) to minimize variability in the responsiveness of individual heifers that comprised the SORH model to ovarian stimulation (Table  1 ). The variability in responsiveness of each heifer to ovarian stimulation was high ranging from 3 to 70 ovulatory‐size follicles. The reason for this high variability in response of the SORH model which has an AFC ≤ 15 follicles ≥ 3 mm in diameter to ovarian stimulation is likely used by the large number of preantral and antral follicles ˂ 3 mm in diameter (Ireland et al.  2008 ) which may also have varying degrees of responsiveness to ovarian stimulation with cpFSH. Despite this variability, the dose of cpFSH, including the Ex‐cpFSH doses, did not impact the degree of positive correlation between AFC or AMH with the total number of ovulatory‐size follicles or the number of functional or dysfunctional ovulatory‐size follicles. Thus, all the AFC or AMH correlation data with number of ovulatory‐size follicles were pooled for final analyses. Results showed that measurements of AFC and circulating AMH concentrations before ovarian stimulation were highly positively correlated with the total number of ovulatory‐size follicles, which included both the functional and dysfunctional ovulatory‐size follicles, independent of cpFSH dose used during ovarian stimulation (Figure  8 ). We also observed that AFC and circulating AMH concentrations were positively correlated with the ratio of dysfunctional ovulatory‐size follicles developing in response to ovarian stimulation. This observation implied that an inverse relationship existed between size of the ovarian reserve (as measured by AFC or AMH) and the number of functional ovulatory follicles capable of ovulation in response to an hCG ovulatory stimulus during ovarian stimulation, independent of the cpFSH dose used. This finding indicated that ovarian stimulation per se may be detrimental to ovulatory follicle function and perhaps oocyte quality. We concluded that measurements of the ovarian reserve biomarkers, AFC and AMH, before ovarian stimulation are predictive of the total number of ovulatory‐size follicles (functional + dysfunctional) in the SORH model. Consequently, these biomarkers are unlikely to be useful to consistently improve IVF or embryo transfer outcomes in the SORH model. Once new protocols are developed to maximize the number of functional estrogen‐active ovulatory‐size follicles developing during ovarian stimulation, however, AFC and AMH measurements could be useful biomarkers to identify groups of relatively uniform responders to ovarian stimulation, albeit time‐consuming and expensive. This positive outcome could allow excessive FSH doses to be avoided, a better estimate of the number of oocytes potentially available for IVF and the number of transferable embryos, and to better inform patients/clients of expected ART outcomes.

James J Ireland: conceptualization, investigation, funding acquisition, writing–original draft, methodology, validation, visualization, writing–review & editing, project administration, supervision, resources, data curation, formal analysis. Kaitlin R Karl: investigation, writing–original draft, writing–review & editing, methodology, validation, visualization, software, formal analysis, data curation, conceptualization. Keith E Latham: conceptualization, investigation, funding acquisition, writing–original draft, methodology, validation, visualization, writing–review & editing, software, formal analysis, project administration, data curation, supervision, resources.

The authors have nothing to report.

The authors have nothing to report.

ART is defined here as a group of medical procedures involving recovery of oocytes from ovaries of women, combining the oocytes with sperm for IVF to produce embryos in the laboratory, returning the IVF‐produced embryos into the donor's or another recipient's uterus, and cryopreservation of excess embryos. ART has resulted in birth of over 9 million babies worldwide (Berntsen et al.  2019 ; Kuhnt and Passet‐Wittig  2022 ) since the first successful embryo transfer in women in 1978 (Steptoe and Edwards  1978 ). The World Health Organization reports that ~1 of every 6 individuals of reproductive age (currently 1.9 billion women age 15–49; “Family planning/contraception methods.  2020 ) worldwide experiences infertility (defined as a failure to achieve a pregnancy after 12 months of unprotected sexual intercourse) during their lifetime (Harris  2023 ; Infertility.  2023 ; SingleCare.  2023 ). Moreover, infertility is increasing 5%–10% per year globally (Harris  2023 ; “Infertility.  2023 ; SingleCare  2023 ). Approximately 30% of women seek ART because they have a DOR ( a reduction in the total number of morphologically healthy oocytes (“Assisted Reproductive Technology National Summary Report for 2018,”  2021 ; Devine et al.  2015 ; Oudendijk et al.  2011 ; Patrizio et al.  2015 ). Other common causes of female infertility potentially alleviated during ART include ovulatory dysfunction, blocked oviducts, endometriosis, uterine factor, polycystic ovarian disease, recurrent pregnancy loss, and unexplained reasons (“Assisted Reproductive Technology National Summary Report for 2018,”  2021 ; Devine et al.  2015 ; Oudendijk et al.  2011 ; Patrizio et al.  2015 ). The use of ART has been limited by the cost per ART cycle that ranges from $5000 to $73,000, with 85% of these costs being paid out of pocket (SingleCare  2023 ), as well as unequal patient access to such procedures (e.g., geographic disparities; Mikhael, Gaidis, and Gavrilova‐Jordan  2021 ). As a result, only 5% of all infertile couples can afford to use ART to have children. While ART alleviates many infertility problems, the procedure is not only costly but inefficient requiring multiple ART cycles for a live birth (Smith et al.  2015 ). Addressing these issues of cost, limited access and low efficiency are important goals for the field.

The above observations illustrate why FSH dosing should be valuable for the success of ART cycles. However, excessive FSH dosing is likely counterproductive. The treatment of ART patients with pharmacological amounts of rhFSH during ovarian stimulation overrides the natural FSH‐regulated selection process by promoting development of multiple ovulatory‐size follicles (Adams et al.  1994 ; Armstrong  1993 ; Bó and Mapletoft  2014 ; Broekmans  2019 ; García Guerra et al.  2015 ; Ireland et al.  2000 ; Xu et al.  1995 ). Consequently, the number of these rhFSH‐induced ovulatory‐size follicles with meiotically mature high‐quality oocytes following an ovulatory dose of hCG that can be recovered for IVF is crucial to the success of ART. High variability exists amongst women in their response to ovarian stimulation with hrFSH during ART. For example, 8–15 oocytes can be retrieved per patient during ART but some women produce as many as 52 ovulatory‐size follicles and 40 retrievable oocytes per ART cycle while others have none (Al‐Shawaf et al.  2001 ; Liu et al.  2023 ; Sunkara et al.  2011 ; Yildiz et al.  2020 ). The major cause of the erratic responsiveness to ovarian stimulation is very likely the inherently high variability amongst individuals in the total number of morphologically healthy follicles and oocytes in the ovarian reserve, which is remarkably variable at birth ranging from 350,000 to 1,100,000 (Block  1953 ; Forabosco et al.  1991 ; Gougeon, Ecochard, and Thalabard  1994 ). Although some laboratories have suggested otherwise (Akahori, Woods, and Tilly  2019 ; White et al.  2012 ), new oocytes are generally not generated after birth (Wagner et al.  2020 ; Woodruff  2008 ; Zhang et al.  2014 ). Thus, the number of primordial (non‐growing) follicles, which comprise the vast bulk of the different follicle types in the ovarian reserve, is finite and decreases rapidly and permanently during aging (Gougeon, Ecochard, and Thalabard  1994 ; Hansen et al.  2008 ; Wallace and Kelsey  2010 ). Coincident with the decline in the ovarian reserve and the pool of small, growing FSH‐responsive antral follicles in a follicular wave (Baerwald, Adams, and Pierson  2012 ; Block  1953 ; Forabosco et al.  1991 ; Gougeon, Ecochard, and Thalabard  1994 ) are the increase in circulating FSH concentrations (Burger et al.  2007 ; MacNaughton et al.  1992 ; Soares et al.  2020 ) and the decrease in circulating AMH concentrations (de Vet et al.  2002 ; Raeissi et al.  2015 ). Several studies report that AMH attenuates FSH actions (Chang, Klausen, and Leung  2011 ; Durlinger et al.  2001 ; Visser and Themmen  2014 ; Yu et al.  2022 ). Thus, the age‐related decline in the potential inhibitory effects of circulating AMH concentrations on FSH action very likely enhances responsiveness of the growing follicles in follicular waves to the heightened circulating FSH concentration during the menstrual cycle. Nevertheless, the impact of the enhanced FSH secretion and action on follicular function during aging is poorly understood. The age‐related decline in the ovarian reserve, coupled with the decrease in the pool of FSH‐responsive follicles and circulating AMH concentrations but enhanced FSH secretion, would also be expected to contribute to the variable responsiveness of individuals to ovarian stimulation with hrFSH during ART.

The results of our studies using the SORH model (summarized in Table  2 ) provided direct evident to support the longstanding hypothesis that high FSH doses during ovarian stimulation are detrimental to ovulatory follicle function and oocyte quality and support two conclusions. Firstly, ovarian stimulation with excessive doses of FSH in the absence of an hCG‐ovulatory stimulus alters gene expression and dysregulates numerous signaling pathways in granulosa and cumulus cells and oocytes critical for ovulatory follicle function, responsiveness to an ovulatory stimulus, and oocyte quality as measured by ability to undergo maturation and be fertilized. These detrimental effects of excessive FSH action increase oocyte wastage and the likelihood of decreased ART outcomes. Secondly, AMH concentration and AFC are reliable biomarkers to predict number but not functionality (health) of potential ovulatory follicles developing during ovarian stimulation of the SORH model. Thus, neither AMH nor AFC are predictive of the number of high‐quality oocytes available for ART. We have provided new insights using the SORH model into the potential impact of the FSH dose used during ovarian stimulation on ART outcomes. Firstly, direct evidence was presented using the SORH model that FSH doses above those needed to achieve the maximum number of ovulatory‐size follicles during ovarian stimulation are excessive and detrimental to ovulatory follicle function and oocyte quality. Secondly, direct rather than correlative evidence was presented using the SORH model that individuals with a small ovarian reserve respond best to relatively low FSH doses during ovarian stimulation. Thus, use of low rather than high FSH doses during ovarian stimulation should improve rather than reduce the recovery of high‐quality oocytes for IVF. Thirdly, excessive FSH action during ovarian stimulation in the SORH model disrupted expression of cumulus cell genes that are well‐known to regulate cumulus cell function and oocyte quality. These discoveries may warrant collaborations between clinical and animal scientists to overcome the clinical barriers of experimentation with humans. Such collaborations could take advantage of the bovine model to develop new low‐FSH‐dose procedures and other diagnostic methods that can reveal when FSH doses are excessive during ovarian stimulation and thereby improve ovarian stimulation protocols and ART outcomes. We propose that optimized ovarian stimulation protocols will mitigate the high costs associated with failed ART cycles by reducing the amount of costly FSH administered and the amount of oocyte wastage. Such improvements could enhance the recovery of high‐quality oocytes for IVF, and thereby improve IVF success and ART outcomes.

Nearly three decades ago, the Nobel laureate, Robert G. Edwards, who pioneered in vitro fertilization (IVF), raised a “red flag” that the chances of successful pregnancy in women undergoing assisted reproductive technology (ART) are reduced by the clinical practice of using high doses of follicle‐stimulating hormone (FSH) during ovarian stimulation (Edwards, Lobo, and Bouchard  1996 ). Although the challenge of choosing the optimal FSH dose for ART remains unmet, our understanding of FSH effects during ovarian stimulation have improved substantially through the development and application of relevant animal models as well as analysis of outcomes in women. Because of the challenges inherent in human studies to assess optimum FSH dosing, animal models that provide insight into this question are invaluable. The small ovarian reserve heifer (SORH) model was developed and found to have characteristics mimicking women with a diminished ovarian reserve (DOR), a key cause of infertility. The SORH model has provided new insights into the effects of excessive FSH dosing (defined as doses that do not further increase the number of ovulatory‐size follicles compared with a lower dose) on ovulatory function and toxic effects on oocytes. These observations may explain the inverse relationship between FSH doses and ART outcomes. The SORH model may also provide a valuable tool for developing new methods to optimize FSH doses to minimize ovulatory follicle dysfunction. Such new methods could improve the reliability of using antral follicle count (AFC) and circulating anti‐Müllerian hormone (AMH) concentrations as biomarkers to predict responsiveness to ovarian stimulation, reveal new biomarkers to facilitate individualized management of FSH dosing, improve the recovery of high‐quality oocytes for IVF, and improve ART outcomes. This review summarizes the clinical implications of excessive FSH dosing in women, the discoveries that have emerged using the SORH model, and potential paths for mitigating potential risks that may come with FSH dosing during ART.

The authors declare no conflicts of interest.