Resistant ovary syndrome, previously known as Savage Syndrome, is a rare disease of unknown etiology. So far there are only a few reports described in the literature. The manifestations of ROS involve primary or secondary amenorrhea, normal secondary sexual characters, an age-compatible number of small antral follicles, normal chromosome, elevated gonadotropin levels of menopausal range, and unresponsiveness to gonadotropin stimulation. ROS might be misdiagnosed as primary ovarian insufficiency (POI). The main difference between ROS and POI is the presence of a normal amount of antral follicles in ROS, while few or no follicles are found in POI. In the past ovarian biopsy has been required to clarify a diagnosis of ROS, but nowadays many non-invasive approaches are helpful in proving the presence of antral follicles. It is well known that antral follicles as small as 2 mm in diameter can be visualized by high-resolution transvaginal ultrasonography [
7]. Furthermore, hormone biomarkers such as inhibin B and AMH have been suggested to have a diagnostic role in women with ROS [
4,
7], which are secreted by granulosa cells of growing follicles and closely related to ovarian reserve. In the report of Arici et al, Inhibin B has a diagnostic role in two women with ROS who manifested by secondary amenorrhea, high gonadotropin level but normal inhibin B level and normal primordial follicles after ovarian biopsy, while AMH level was not measured [
4]. However, in the report of Grynberg et al, the case with ROS showed very low inhibin B level but normal AMH level [
6]. The case in this report showed normal both inhibin B and AMH level. The conflicting results could be explained by the existence of different sizes of antral follicles in the ovary among different women with ROS. The intrafollicular concentrations of AMH become progressively lower with increasing follicle diameters but concentrations of inhibin B increased with increasing follicle diameter in human small antral follicles [
8]. Thereafter AMH could be a better biomarker than inhibin B for discriminating ROS and POI, due to it correlates with ovarian primordial follicle number even after adjustment for chronological age [
9]. According to the clinical manifestations and biomarkers, the patient in this report should be diagnosed as ROS.
ROS has long been an enigma. Mueller et al [
3] reported there was no indication for any autoimmune disease in a patient with ROS and tests for antinuclear antibodies, antiphospholipid antibodies, lupus anticoagulant and anticardiolipin antibodies were all negative. Arici et al [
4] also found no autoantibodies against thyroid, adrenals, or ovaries in two women with ROS. Grynberg et al [
6] conducted more detailed genetic and autoimmune explorations in a 29-year-old patient, including karyotype, sequencing of FSH and LH receptor genes, analysis of GDF9, FOXL2, BMP15 genes and FRAXA mutation, as well as adrenal cortex autoantibodies, steroid cell autoantibodies, serum 21-hydroxylase, 17-hydroxylase and P450 side-chain cleavage enzyme autoantibodies, yet revealed no abnormality. Similarly, our case showed no history of autoimmune diseases, normal karyotype, no deleterious mutations in FSHR gene, and some negative autoantibodies. Gonadotropin receptor or hormone mutations were rarely found in sporadic cases of gonadotropin resistance [
10,
11]. While ROS showed few links with genetics, since family history, chromosome and gonadotropin receptor as well as post-receptor defects analysis invariably proved to be normal [
6,
12,
13]. Thereafter, gonadotropin resistance in ROS women might be due to defects in interactions of FSH and its receptor, mostly secondary to autoimmune activities [
14,
15]. Some reports demonstrated that antigonadotropin antibodies were found in women with gonadotropin resistance [
13,
16,
17]. ROS could be another special condition of reproductive autoimmune failure that has been described elsewhere [
18,
19]. The patient in this report had a medical history of pelvic tuberculosis and oligomenorrhea or amenorrhea occurred after anti-tuberculosis therapy. While mycobacterial tuberculosis infections are known to induce the development of autoantibodies and even autoimmune diseases [
20,
21]. In a recent study of ROS case, it demonstrated that IgG/IgM-antibodies directed against some forms of gonadotropins were found in serum and respective antibodies in this serum only react with some but not all forms of FSH [
13]. And the expressed fashion of FSHR, LHR as well as estrogen receptor beta and progesterone receptor A induced by exogenous gonadotropins was not different in granulosa cells from ROS patient and from control patients [
13]. But the ROS case in this report showed normal response to ovarian stimulation with daily 75IU recombinant beta and 225IU hp-HMG for 14 days after hormone replacement and GnRH agonist treatment. 11 mature oocytes were obtained and a successful live birth after embryo transfer [
13]. Several common autoantibodies associated with infertility were not detected in our ROS case but it is limited that antigonadotropin antibodies have not been tested. It need be further investigated whether the possible mechanism of autoimmunity can explain gonadotropin resistance in all ROS patients, and the titers or types of antigonadotropin antibodies are related to that ovarian response has or not and which kind of exogenous gonadotropin should be effective.
Infertility is a difficult issue for ROS patients at reproductive age, and the chances of fertility with their own oocytes are unpredictable and most likely poor. Attempts to stimulate follicle development with high doses of gonadotropins to override the resistant state seem to be invalid [
3,
4,
6]. Indeed we tried to induce follicle growth with 300IU daily hp-HMG for 15 days, but no response. Taking into consideration the first pregnancy and live birth after IVM in ROS case was reported in 2013 [
6], we tried the same method. Three out of five oocytes were matured in vitro, and the transfer of two top-quality embryos resulted in a successful live birth. These two successful cases with ROS by IVM also confirm that granulosa cells can response to exogenous gonadotropin in vitro and IVM overcomes the state of gonadotropin resistance in ROS patients. Finally, no choice but IVF using donor oocytes was the only option for ROS in case of IVM and all other treatment failures [
4], undoubtedly the reproductive prognosis was acceptable in the donor oocyte program.