Evidence for different aetiologies of low estradiol response to FSH: age-related accelerated luteinization of follicles or presence of ovarian autoantibodies

J.L. Luborsky1,4, P. Thiruppathi1, B. Rivnay2, R. Roussev3, C. Coulam3 and E. Radwanska1

1 Department of Obstetrics and Gynecology, Rush Medical College, Chicago, IL, 2 Repromedix Corporation, Woburn, MA and 3 Sher Institute and Millinova Laboratories, Chicago, IL, USA


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
BACKGROUND: We found granulosa cells of low responders (LR) expressed more LH receptors, suggesting that follicles were more luteinized than normal responders (NR). The objectives were to test the hypothesis that follicles of LR were more luteinized than follicles of NR, and to determine if LR with (LR+) and without (LR–) ovarian antibodies differed. METHODS: Hormone levels and ovarian autoantibodies (OVAB) were measured in follicular fluid from mature follicles (>17 mm), and in serum obtained on the day of oocyte retrieval during controlled ovarian stimulation. The gonadotrophin response was defined as a ratio of peak estradiol/number of FSH ampoules. RESULTS: NR (32.5 ± 4.6 years; n = 11) were similar in age to LR+ (33.4 ± 4.2 years; n = 9) and were younger than LR– (37.1 ± 3.8 years; n = 12) (P = 0.03). Likewise, dehydroepiandrosterone sulphate was lower in LR– compared with LR+ or NR (P < 0.01). FSH, progesterone, inhibin-A and vascular endothelial growth factor levels were higher in follicular fluid of LR than NR. LR– and LR+ differed. For example, the follicular fluid progesterone/estradiol ratio was similar in NR (11.1 ± 8.9) and LR+ (9.8 ± 6.6) but was lower than LR– (22.9 ± 19.6) (P = 0.05). Serum hormone levels did not reflect follicular fluid hormone profiles. CONCLUSIONS: In the absence of ovarian antibodies, low responses are associated with higher age and accelerated luteinization of mature follicles, rather than diminished responsiveness. Ovarian antibody may be an additional tool to predict and individualize treatment regimens in poor responders.

Key words: age-related ovarian response/low responders/luteinization/ovarian antibodies/ovarian stimulation


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Low estradiol responses to FSH occur sporadically or consistently in 20–30% of women undergoing controlled ovarian stimulation (Ben-Rafael et al., 1994Go; Scott 1996Go; Ferraretti et al., 2000Go). Since low responses are associated with poorer treatment outcomes, they present a significant impediment to successful infertility treatment. The basis for low responses to FSH is not well defined (Ferraretti et al., 2000Go; Surrey and Schoolcraft, 2000Go).

Low gonadotrophin responses are usually differentiated from normal responses by the level of serum estradiol attained in response to 7–14 days of FSH stimulation during controlled ovarian stimulation. It has become customary to designate low responses as those <1000 mIU/ml estradiol (Scott, 1996Go; Ferraretti et al., 2000Go). The number of ampoules of FSH required to attain an adequate response is also an indicator of low responses, although a specific cut-off value has not been defined. A reduced number of oocytes retrieved after hCG stimulation may also be associated with low responses (Dor et al., 1992Go). However, poor responses associated with a low number of follicles recruited during controlled ovarian stimulation may represent a different clinical entity than low responses to gonadotrophin (Ferraretti et al., 2000Go; Surrey and Schoolcraft, 2000Go).

Different aetiologies have been described for low responses. For example, low responses are associated with ageing and low ovarian reserve (Kligman and Rosenwaks, 2001Go), reduced blood flow (Pellicer et al., 1994Go), reduced aromatase activity (Hurst et al., 1992Go), hormone resistance associated with specific FSH receptor polymorphism (Perez Mayorga et al., 2000Go) and ovarian autoimmunity (Meyer and Lavy et al., 1990; Hovav et al., 1994Go). Although ageing appears to explain a proportion of low responses, many low responders are young (Jenkins et al., 1991Go; Kim 1995Go; Hanoch et al., 1998Go). Thus, it is likely that low responses represent multiple aetiologies.

In a previous study, we examined the possibility that differences in hormone response were due to differences in gonadotrophin receptor expression. We showed that FSH receptor expression was similar on granulosa cells of low and normal responders but LH receptor expression was slightly higher on cells of low responders (Thiruppathi et al., 2001Go). In addition, more of the granulosa cells from low responders had the large rounded, granular morphology of luteal cells (unpublished observation). Similarly, Whitman et al. observed that follicles of oocytes that failed to fertilize contained more luteinized cells (Whitman et al., 1988Go, 1989Go). These observations suggested that follicles of low responders might be more luteinized.

More advanced luteinization has been reported in follicles of older women during natural ovarian cycles (Klein et al., 1996aGo,bGo,cGo; Klein and Soules, 1998Go). Based on serum measures, premature luteinization was observed in women with infertility (Ubaldi et al., 1996Go; Lindheim et al., 1999Go; Muttukrishna et al., 2000Go). However, markers of luteinization have not been examined at the follicular level as a basis for altered responsiveness to gonadotrophin.

The objective of this study was to test the hypothesis that follicles of low responders were more luteinized than follicles of normal responders. Endocrine markers of luteinization such as progesterone, inhibin-A and vascular endothelial growth factor (VEGF) were assessed in follicular fluid and serum of low and normal responders at oocyte retrieval. A secondary objective was to determine if hormone profiles in the presence or absence of ovarian antibodies differed.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Study group
Women undergoing IVF for infertility were entered into the study. Women with polycystic ovarian syndrome (PCOS) or recurrent spontaneous fetal loss were excluded. Women with PCOS were excluded based on a history of oligomenorrhoea (menses interval >42 days) and ultrasound evidence of cystic ovary (Franks, 1989Go). Women with recurrent fetal loss were excluded based on a history of more than three unexplained spontaneous abortions. The primary infertility diagnosis included male factor (n = 12), pelvic factor (n = 5), uterine factor (n = 2), tubal factor (n = 4) and unexplained (n = 9). The average age of the study group was 34.5 + 4.6 years (range 27–42). Early follicular phase FSH levels were not routinely measured, particularly in younger women with menstrual cycles since the information has limited predictive value (Schipper et al., 1998Go; Bancsi et al., 2000Go). The average years of infertility was 3.3 + 2.0 (range 1–8). The average number of prior IVF cycles was 1 ± 1.1 (range 0–4). A blood sample was obtained by routine venipuncture on the day of oocyte retrieval. Follicular fluid was collected during oocyte retrieval (n = 32). All procedures were reviewed and approved by the Institutional Review Board.

Ovarian stimulation was initiated by pituitary desensitization with GnRH during the midluteal phase of the preceding cycle. In the subsequent cycle, hMG, urofollitropin, or urofollitropin HP) was injected daily starting with an initial dose of 300 IU. Highly purified urinary FSH (Fertinex®; Serono, Norwell, MA, USA) was used for 70% of the patients, with the remaining FSH preparations distributed similarly among the study groups. Doses were adjusted based on follicle growth by ultrasound and serum estradiol monitored every other day. Ovulation was induced with 10 000 IU of hCG (Profasi®; Serono) after follicles were >=17 mm in diameter and estradiol had increased over 8 days. The average estradiol response to gonadotrophin in the study cycle was 1341 ± 605 (range 564–2714) and the average number of ampoules of FSH administered was 52 ± 28 (range 16–120). Oocytes were obtained 34 h after hCG administration under transvaginal ultrasound guidance.

Measurement of hormones and antibodies in follicular fluid and serum
For each individual, the follicular fluid from three follicles, 18–20 mm in diameter, was pooled in order to obtain adequate volume for multiple measures. The similarity of results from individual follicles and the pooled follicular fluid of six individuals was verified (Thiruppathi et al., 2001Go). The follicular fluid was centrifuged to remove cells and debris and frozen at –20°C until assay. Serum was separated from whole blood and stored at –20°C. Hormones were measured by immunoassay with commercial kits. Progesterone and estradiol were measured in serum and follicular fluid (Cayman Chemical, Ann Arbor, MI, USA) with detection limits of 10 and 9 pg/ml respectively. Androstenedione was measured in follicular fluid and dehydroepiandrosterone sulphate (DHEAS) were measured in serum (Diagnostic Systems Laboratories, Houston, TX, USA) with detection limits of 0.03 and 15 ng/ml respectively. VEGF was measured in follicular fluid by enzyme immunoassay (R&D Systems, Minneapolis, MN, USA). FSH was measured in follicular fluid (Genzyme, CA, USA). Inhibin-A and inhibin-B were measured with Serotec kits (Harlan, Raleigh, NC, USA). The sensitivity of the inhibin-A immunoassay was 10 pg/ml and of inhibin-B was 15 pg/ml. Hormone levels in follicular fluid were significantly higher than the range of the immunoassay standard curves. Therefore, samples were diluted with sample dilution buffer supplied by the manufacturer, until the levels were in the mid-range of the standard curve. The inter-assay and intra-assay coefficients of variation of all commercial assays were <10%.

Ovarian and oocyte antibodies in serum were determined as described previously (Luborsky et al., 1990Go; Luborsky and Pong, 2000Go). In summary, anti-ovarian antibodies were detected by enzyme-linked immunosorbent assay (ELISA). Ovaries were obtained from rats superovulated with pregnant mare’s serum gonadotrophin and HCG. The 10000g pellet of ovarian homogenates was used as antigen (0.1 µg/ 0.1 ml Tris buffer/well). Results with rat or human ovary were highly correlated (r = 0.95, P < 0.001) (Luborsky and Pong, 2000Go). Anti-oocyte antibodies were detected in ELISA using homogenates of oocytes from mature follicles of rat ovaries. Oocytes were aspirated from follicles at the ovary surface, incubated in 0.5% collagenase (Sigma, St Louis, MO, USA) and 0.01% DNAase (Sigma) in PBS (30 min, 37°C) and separated from cells by gradient centrifugation in 3 ml of Hypaque (Sigma) (400 g, 30 min). After washing, oocytes were homogenized in batches of 100 in a 1 ml microfuge tube with a hand-held micro-pestle. Oocytes were added to ELISA plates at a concentration of ~1 oocyte/0.1 ml Tris buffer/well. Results with rat or human oocytes were correlated (r = 0.85, P < 0.001).

Sera were tested at a dilution of 1:100 and follicular fluid at a dilution of 1:10. Negative control sera from normally cycling women (n = 10) and a positive control were included in each assay. For follicular fluid, four control samples from women with male factor infertility were used as negative controls. The captured autoantibodies were identified by a goat antihuman IgG–alkaline phosphatase conjugate (FAB specific; Sigma). Bound alkaline phosphatase was reacted with substrate (AP substrate; Sigma) and the product read at 405 nm (Thermomax; Molecular Devices, Sunnyvale, CA, USA).

Data and statistical analysis
Statistical differences between normal and low responders and among the response groups were assessed with the Mann–Whitney U-test. Pearson’s correlation coefficient (r) was calculated to evaluate associations between variables. P < 0.05 was considered significant.

For detection of ovarian and oocyte auto antibodies, samples were assessed in duplicate against four negative control samples and a positive control. Inter-assay and intra-assay coefficients of variation for specific optical density signals were <15% and <10% respectively. In this study, individuals positive for one or both antibodies were considered positive and results are referred to in the text as ‘ovarian antibody’ positive for simplicity. Although ovarian antibodies were correlated in follicular fluid and serum (r = 0.4, P = 0.035), the serum test results were used to differentiate positive and negative low responders. Results were compared by {chi}2-analysis.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Characterization of low and normal responders
In order to integrate the FSH dose and the estradiol response, the ratio of the peak estradiol/number of ampoules of FSH (gonadotrophin response) was used to define low and normal responses. This method of defining low responses was more consistent with resistance to FSH and was used to define response groups as described previously (Thiruppathi et al., 2001Go). The gonadotrophin response ratio ranged from 7 to 89; low responders were those <42. Ovarian antibodies were detected in low responders but not normal responders. There were 9/32 (28.1%) low responders with ovarian antibodies, 12/32 (37.5%) low responders without ovarian antibodies and 11/32 (34.4%) normal responders. The mean gonadotrophin response ratio was higher for normal responders than low responders, but did not differ for low responders with and without ovarian antibodies (Table IGo).


View this table:
[in this window]
[in a new window]
 
Table I. General patient information: descriptive statistics (mean ± SD)
 
Normal responders and low responders with ovarian antibodies were significantly younger than low responders and also significantly younger than low responders without antibodies (Table IGo and Figure 1Go). The average number of FSH ampoules administered and the peak estradiol were significantly different between normal and low responders but did not differ between low responders with and without ovarian antibodies (Table IGo). The mean number of oocytes retrieved and the mean percentage of fertilized oocytes were not different between the groups (Table IIGo). However, the lowest number of oocytes retrieved and the lowest percentage of oocytes fertilized occurred in low responders with ovarian antibodies, as reflected in the standard deviations.



View larger version (21K):
[in this window]
[in a new window]
 
Figure 1. Age of normal and low responders shown as box plots. Responses to FSH were defined as a ratio of the peak estradiol/number of FSH ampoules. In addition, the presence (POS) or absence (NEG) of ovarian antibodies (OVAB) was used to define groups. The median, range (whiskers), 25th–75th percentile (box) and outliers (circles) are shown. The mean values and significance of differences are summarized in Table IGo.

 

View this table:
[in this window]
[in a new window]
 
Table II. Oocyte parameters: descriptive statistics (mean ± SD)
 
Hormones in follicular fluid
Hormone levels in follicular fluid of mature follicles >17 mm were measured in order to determine if the local environment of maturing follicles differed for normal and low responders. Estradiol and androstenedione levels were similar between normal and low responders (Table IIIGo). Progesterone levels were significantly higher in follicles of low responders compared with normal responders (Table IIIGo).


View this table:
[in this window]
[in a new window]
 
Table III. Follicular fluid hormone levels: descriptive statistics (mean ± SD)
 
The mean progesterone level did not differ significantly between low responders with or without ovarian antibodies, although it tended to be lower and more variable in low responders with ovarian antibodies (Figure 2Go). Progesterone was significantly higher in low responders without ovarian antibodies compared with normal responders (Table IIIGo). The progesterone/estradiol ratio was significantly higher for low responders without ovarian antibodies compared with low responders with ovarian antibodies or normal responders (Table IIIGo).



View larger version (31K):
[in this window]
[in a new window]
 
Figure 2. Comparison of steroid hormone and inhibin levels in follicular fluid (FF) of mature follicles. Progesterone was elevated in low responders. Estradiol levels did not differ among normal and low responders. Compared with normal responders, the progesterone/estradiol ratio and inhibin-A was significantly higher only for low responders without ovarian antibodies. Specific mean values and the significance of differences are summarized in Table IIIGo. OVAB = ovarian antibodies; POS = present; NEG = absent.

 
Inhibin-A, inhibin-B and VEGF levels were measured as additional indicators of luteinization. Inhibin-B levels in follicular fluid of normal and low responders were similar (Table IIIGo). Inhibin-A and the inhibin-A/inhibin-B ratio were higher in follicles of low responders although the difference was not statistically significant compared with normal responders (Table IIIGo). Only low responders without ovarian antibodies had significantly higher inhibin-A levels compared with normal responders (Figure 2Go). Similarly, only low responders without ovarian antibodies had significantly higher VEGF levels in follicular fluid compared with normal responders (Table IIIGo).

The correlation of gonadotrophin response and indicators of luteinization were also assessed without separating the study group into low and normal responses. Response to gonadotrophin was negatively correlated with age (r = –0.4, P = 0.04), follicular fluid FSH levels (r = –0.5, P = 0.003) and the inhibin-A/inhibin-B ratio (r = –0.5, P = 0.013), but was not correlated with follicular fluid estradiol, progesterone or the progesterone/estradiol ratio. The lack of correlation of the progesterone/estradiol ratio with gonadotrophin response reflects a difference in the distribution of progesterone levels for normal and low responders. As seen in Figure 3Go, progesterone was correlated with gonadotrophin response in normal responders (r = 0.72, P = 0.006) but was not correlated among low responders (r = 0.05, P = 0.4, not significant). When ovarian antibody positive individuals were excluded, the correlation of response with progesterone (r = 0.4, P = 0.055) and inhibin-B (r = 0.4, P = 0.067) approached significance.



View larger version (18K):
[in this window]
[in a new window]
 
Figure 3. Relationship of gonadotrophin response to progesterone in follicular fluid (FF) (left panel) and serum (right panel). Response was defined by the ratio of peak estradiol/number of FSH ampoules. Progesterone levels were correlated with gonadotrophin response in normal responders. In contrast, levels of progesterone varied widely in follicles of low responders, and were not correlated with gonadotrophin response. Correlations are summarized in the text. OVAB = ovarian antibodies; POS = present; NEG = absent.

 
The number of ampoules of FSH and levels of FSH in follicular fluid were correlated regardless of the inclusion (correlation = 0.71, P = 0.001) or exclusion (correlation = 0.76, P = 0.001) of individuals with ovarian antibodies. FSH in follicular fluid was positively correlated with age (r = 0.5, P = 0.003), progesterone (r = 0.6, P = 0.001), VEGF (r = 0.6, P < 0.001) and the ratio of inhibin-A/inhibin-B (r = 0.6, P = 0.001) in follicular fluid, and negatively correlated with inhibin-B (r = –0.5, P = 0.013). The correlation of FSH with the progesterone/estradiol ratio (r = 0.4, P = 0.067) approached significance in the absence of individuals with ovarian antibodies while other correlations remained similar.

Hormones in serum
Serum levels of hormones were measured as an indicator of the overall activity of follicles in the ovary. Serum inhibin-A and inhibin-B (Table IVGo) and peak estradiol (day of hCG) (Table IGo) were significantly lower in low responders than normal responders. Estradiol (day of oocyte retrieval), the inhibin-A/inhibin-B ratio, progesterone and the progesterone/estradiol ratio were not significantly different between low and normal responders. DHEAS was measured as an indicator of somatic ageing and of adrenal steroid biosynthesis and differed significantly between normal and low responders (Table IVGo). However, DHEAS in low responders without ovarian antibodies was significantly higher than DHEAS in low responders with ovarian antibodies and in normal responders (Figure 4Go).


View this table:
[in this window]
[in a new window]
 
Table IV. Serum hormone levels: descriptive statistics (mean ± SD)
 


View larger version (27K):
[in this window]
[in a new window]
 
Figure 4. Comparison of dehydroepiandrosterone sulphate (DHEAS) in serum (day of oocyte retrieval). DHEAS was significantly lower in low responders without ovarian antibodies compared with normal and low responders with ovarian antibodies. As an indicator of somatic age, DHEAS levels are consistent with mean chronological ages of each group shown in Figure 1Go. Specific mean values and the significance of differences are summarized in Table IVGo. OVAB = ovarian antibodies; POS = present; NEG = absent.

 
Responses to gonadotrophin were positively correlated with serum estradiol (day of oocyte retrieval) (r = 0.4, P = 0.03), inhibin-A (r = 0.6, P < 0.001) and inhibin-B (r = 0.6, P = 0.001) in serum, but were not correlated with progesterone, the progesterone/estradiol ratio, or the inhibin-A/inhibin-B ratio. Unlike the divergent distribution of progesterone levels in follicular fluid in relation to gonadotrophin response, progesterone levels in serum were more similarly distributed with a single trend from low to high in relation to gonadotrophin response (Figure 3Go). DHEAS (as an indicator of somatic age) was correlated with gonadotrophin response (r = 0.7, P = 0.001) only when ovarian antibody positive individuals were excluded from analysis.

The number of FSH ampoules was negatively correlated with serum inhibin-B (r = –0.5, P = 0.01) regardless of the inclusion of ovarian antibody positive individuals and was not correlated with serum levels of other hormones.

Relationship between serum and follicular fluid hormone levels
With a few exceptions, serum hormone levels did not reflect patterns of follicular fluid hormones regardless of the inclusion or exclusion of ovarian antibody positive individuals in the analysis. For example, progesterone in serum and follicular fluid were not correlated (P = 0.08, not significant). When ovarian antibody positive individuals were excluded, serum estradiol (day of oocyte retrieval) (r = –0.53, P = 0.02) and serum inhibin-A (r = –0.46, P = 0.04) were negatively correlated with the follicular fluid inhibin-A/inhibin-B ratio. In other words, in the absence of ovarian antibodies, the lower the serum estradiol and the higher the inhibin-A/inhibin-B ratio the more likely that individual maturing follicles would exhibit accelerated luteinization.


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
In this study, progesterone was the predominant sex steroid in mature follicles. This was expected since hCG initiates luteinization and progesterone production. However, progesterone was strikingly higher in follicles of low responders compared with normal responders. Likewise VEGF, inhibin-A and the inhibin-A/inhibin-B ratio were higher in mature follicles of low responders, consistent with more advanced luteinization. In contrast, hormone levels in serum did not reflect the accelerated luteinization of maturing follicles in low responders. Instead, serum hormone levels tended to be lower in low responders, consistent with reduced ovarian reserve. This is the first direct demonstration that low responses to gonadotrophin and accelerated luteinization of maturing follicles are associated.

Only low responders without ovarian antibodies showed a clear trend towards accelerated luteinization and higher age. Low responders with ovarian antibodies tended to be similar to normal responders with respect to age and hormone profiles, rather than to low responders without ovarian antibodies. Also, hormone levels and oocyte parameters varied more in low responders with ovarian antibodies.

The higher VEGF observed in this study is consistent with previous reports that higher levels of VEGF were associated with lower peak estradiol levels, lower number of oocytes retrieved, higher age and higher number of FSH ampoules required (Friedman et al., 1997Go, 1998Go). In addition, Neulen et al. showed that the ratio of VEGF/soluble VEGF receptor (sVEGFR-1) was inversely related to the number of oocytes retrieved, suggesting that decreased bioavailability of VEGF is associated with poor responses defined by oocyte yields (Neulen et al., 2001Go). VEGF is required for corpus luteum formation and increases with progressive luteinization and vascular growth (Ferrara and Davis-Smyth, 1997Go; Ferrara et al., 1998Go). VEGF production in luteinizing granulosa cells is stimulated by FSH and LH (Christenson and Stouffer 1997Go; Laitinen et al., 1997Go; Geva and Jaffe, 2000Go; Stouffer et al., 2001Go). Likewise, VEGF was correlated with FSH in this study. However, since doses of FSH are not equivalent, it is not clear if the higher levels of VEGF are a consequence of higher doses of FSH, if luteinization is enhanced by a slightly higher content of LH in the FSH preparation (Filicori et al., 2001Go) or if low responders have reduced control of the rate of follicle growth and differentiation (Eppig, 2001Go).

Serum hormone levels did not reflect the hormone profiles of mature follicles. Instead, serum hormone levels tended to be lower in low responders. Both low responders with and without ovarian antibodies had lower levels of ovarian hormones consistent with reduced ovarian reserve that is thought to precede ovarian failure. In serum, the hormone levels reflect the additive activity of follicles of different maturation stages. Increasing estradiol in response to high doses of FSH may be due to increases within smaller follicles or to recruitment of additional, less mature, follicles. The trend to lower serum levels of ovarian hormones in low responders and in older women is consistent with other studies (Scott, 1996Go; Hall et al., 1999Go; Seifer et al., 1999Go). For example, Seifer et al. reported that age was associated with declining ovarian reserve, decreased serum inhibin-B, increased requirement for FSH, lower success rate and fewer oocytes retrieved (Seifer et al., 1999Go).

We did not observe a difference in serum progesterone between normal and low responders. Other reports describing the relationship of serum progesterone and responses to gonadotrophin are inconsistent. Lindheim et al. found that low serum progesterone responses to hCG were associated with poor estradiol responders and suggested that this reflected inadequate steroidogenesis (Lindheim et al., 1999Go). Women with premature luteinization, defined by serum progesterone >1.1 ng/ml on the day of hCG administration, required a higher number of FSH ampoules to attain an adequate estradiol response (Ubaldi et al., 1996Go). Likewise, premature luteinization, defined by a serum progesterone/estradiol ratio >1 on the day of hCG administration, was associated with low estradiol responses and ovarian reserve (Younis et al., 2001Go). Follicular fluid hormone levels were not measured in these studies, and thus there is no information on the activity of maturing follicles. Differences in conclusions among studies may reflect different types of information obtained by assessing serum or follicular fluid, differences in definition of low and normal responses, and differences in study design and the day of serum sampling.

Thus, it appears that rather than reduced responsiveness, follicles of older ‘low responders’ are maturing and differentiating at a faster rate than those of normal responders. The results are consistent with previous reports of accelerated luteinization of follicles in non-stimulated ovarian cycles of older women (Klein et al., 1996aGo,bGo; Klein and Soules, 1998Go). Greater luteinization could be due to higher levels of FSH, higher levels of LH associated with higher doses of FSH, enhanced responses to low levels of LH as cells acquire LH receptors (Filicori et al., 1999Go, 2001Go), or to a reduced ability of oocytes to inhibit luteinization (Vanderhyden and Macdonald, 1998Go) prior to oocyte retrieval.

The results are consistent with heterogeneous aetiologies of low responses. Tanbo et al. reported two types of low responses based on clomiphene citrate responses (Tanbo et al., 1990Go). Hanoch et al. also differentiated low responses by age and noted that young low responders had better pregancy rates (Hanoch et al., 1998Go). Low estradiol responses were also differentiated by day 3 LH (Noci et al., 1998Go). FSH receptor polymorphism at position 680 was associated with different requirements for FSH during stimulation, suggesting partial hormone resistance in some women (Perez Mayorga et al., 2000Go). Low responses have also been associated with decreased blood flow measured by Doppler ultrasonograpy (Pellicer et al., 1994Go). We showed that some low responders had ovarian antibodies (Meyer et al., 1990Go). In this study, we found that low responders differentiated by the presence or absence of ovarian antibodies differed in age and hormone profiles. Because of the heterogeneity of low gonadotrophin responses, it has been suggested that pre-cycle screening to differentiate potential responses may improve treatment outcomes (Surrey and Schoolcraft, 2000Go).

Although the primary objective of this research study was to obtain information on potential mechanisms of low responses to FSH, there are implications for clinical practice. The lack of correlation of FSH dose with peak serum estradiol, the correlation with follicular fluid markers of luteinization and the lack of difference in oocytes retrieved suggest that higher FSH doses are not required for older low responders. This is consistent with other studies (Tinkanen et al., 1999Go; Out et al., 2000Go). Premature luteinization associated with ageing may result in over-ripe or poorer quality oocytes (Eichenlaub-Ritter, 1998Go; Sauer, 1998Go). It has been reported that monitoring inhibin-B in addition to estradiol to determine the timing of hCG improves prediction of a successful outcome compared with estradiol monitoring alone (Seki et al., 1997Go). This would be supported by this study since serum inhibin-B was correlated with the follicular response to gonadotrophin. Since the rate of follicle growth is slower in response to recombinant FSH which lacks trace LH (Filicori et al., 1999Go, 2001Go), it has been suggested that outcomes for low responders may be improved by using recombinant FSH or a combination of recombinant FSH and controlled levels of trace LH, although treatments may need to be individualized (Pellicer et al., 1998Go; Howles, 2000Go; Surrey and Schoolcraft, 2000Go; Fawzy et al., 2001Go).

The significant differences among low responders with ovarian antibodies suggest that treatment responses to standard hormone therapy may be less predictable in low responders with ovarian antibodies. Thus, individuals with ovarian antibodies may benefit from closer pre-cycle screening and monitoring during treatment for markers of follicular growth. In some cases, hormone stimulation may overcome the autoimmune reaction and recruit some healthy follicles. However, there is evidence that treatment outcomes associated with ovarian autoimmunity are poor (Geva et al., 1999Go; Ivanova et al., 1999Go; Horejsi et al., 2000Go; Luborsky and Pong, 2000Go). Treatment with low dose immunosuppression has been successful (Luborsky et al., 1990Go; Geva et al., 1999Go), but treatment protocols have not been standardized in controlled trials.

In summary, mature follicles of low responders appear to be more luteinized than follicles of normal responders. In the absence of ovarian antibodies, low responses are associated with higher chronological and somatic age and accelerated luteinization of mature follicles, rather than diminished responsiveness. However, the follicular environment of low responders with ovarian autoantibodies differs from low responders without autoantibodies, suggesting different mechanisms for low responses in the presence of ovarian autoimmunity. In addition, use of ovarian antibody testing provides an additional tool for predicting response to gonadotrophin and may contribute to more individualized regimens for controlled ovarian stimulation.


    Acknowledgements
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
This study was supported in part by an educational grant from Serono Laboratories, Inc. (Norwell, MA, USA). This work was presented in part at the Second International Conference on Experimental and Clinical Reproductive Immunobiology (2000) and the annual meeting of the Society Study of Reproduction (2000).


    Notes
 
4 To whom correspondence should be addressed at: Department of Obstetrics and Gynecology, 1653 W Congress Parkway, Rush Medical College, Chicago, IL 60612, USA. E-mail: jluborsk{at}rush.edu Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Bancsi, L.F., Huijs, A.M., den Ouden, C.T., Broekmans, F.J., Looman, C.W., Blankenstein, M.A. and te Velde, E.R. (2000) Basal follicle-stimulating hormone levels are of limited value in predicting ongoing pregnancy rates after in vitro fertilization. Fertil. Steril., 73, 552–557.[ISI][Medline]

Ben-Rafael, Z., Orvieto, R. and Feldberg, D. (1994) The poor responder patient in an in vitro fertilization embryo transfer (IVF–ET) program. Gynecol. Endocrinol., 8, 277–286.[ISI][Medline]

Christenson, L.K. and Stouffer, R.L. (1997) Follicle-stimulating hormone and luteinizing hormone/chorionic gonadotropin stimulation of vascular endothelial growth factor production by macaque granulosa cells from pre- and periovulatory follicles. J. Clin. Endocrinol. Metab., 82, 2135–2142.[Abstract/Free Full Text]

Dor, J., Seidman, D., Ben-Shlomo, I., Levran, D., Karasik, A. and Mashiach, S. (1992) The prognostic importance of the number of oocytes retrieved and estradiol levels in poor and normal responders in in vitro fertilization treatment. J. Assist. Reprod. Genet., 9, 228–232.[ISI][Medline]

Eichenlaub-Ritter, U. (1998) Genetics of oocyte ageing. Maturitas, 30, 143–169.[ISI][Medline]

Eppig, J.J. (2001) Oocyte control of ovarian follicular development and function in mammals. Reproduction, 122, 829–938.[Abstract/Free Full Text]

Fawzy, M., Harrison, R.F., Knight, P.G., Groome, N., Anderson, A., Robertson, W.R. and Lambert, A. (2001) The effect of gonadotrophins with differing LH/FSH ratios on the secretion of the various species of inhibin in women receiving IVF. Hum. Reprod., 16, 1092–1097.[Abstract/Free Full Text]

Ferrara, N. and Davis-Smyth, T. (1997) The biology of vascular endothelial growth factor. Endocr. Rev., 18, 4–25.[Abstract/Free Full Text]

Ferrara, N., Chen, H., Davis-Smyth, T., Gerber, H.P., Nguyen, T.N., Peers, D., Chisholm, V., Hillan, K.J. and Schwall, R.H. (1998) Vascular endothelial growth factor is essential for corpus luteum angiogenesis. Nat. Med., 4, 336–340.[ISI][Medline]

Ferraretti, A.P., Gianaroli, L., Magli, M.C., Bafaro, G. and Colacurci, N. (2000) Female poor responders. Mol. Cell. Endocrinol., 161, 59–66.[ISI][Medline]

Filicori, M., Cognigni, G.E., Taraborrelli, S., Spettoli, D., Ciampaglia, W., de Fatis, C.T. and Pocognoli, P. (1999) Luteinizing hormone activity supplementation enhances follicle-stimulating hormone efficacy and improves ovulation induction outcome. J. Clin. Endocrinol. Metab., 84, 2659–2663.[Abstract/Free Full Text]

Filicori, M., Cognigni, G.E., Taraborrelli, S., Spettoli, D., Ciampaglia, W., Tabarelli De Fatis, C., Pocognoli, P., Cantelli, B. and Boschi, S. (2001) Luteinizing hormone activity in menotropins optimizes folliculogenesis and treatment in controlled ovarian stimulation. J. Clin. Endocrinol. Metab., 86, 337–343.[Abstract/Free Full Text]

Franks, S. (1989) Polycystic ovary syndrome: a changing perspective. Clin. Endocrinol., 31, 87–120.[ISI][Medline]

Friedman, C.I., Danforth, D.R., Herbosa-Encarnacion, C., Arbogast, L., Alak, B.M. and Seifer, D.B. (1997) Follicular fluid vascular endothelial growth factor concentrations are elevated in women of advanced reproductive age undergoing ovulation induction. Fertil. Steril., 68, 607–612.[ISI][Medline]

Friedman, C.I., Seifer, D.B., Kennard, E.A., Arbogast, L., Alak, B. and Danforth, D.R. (1998) Elevated level of follicular fluid vascular endothelial growth factor is a marker of diminished pregnancy potential. Fertil. Steril., 70, 836–839.[ISI][Medline]

Geva, E. and Jaffe, R.B. (2000) Role of vascular endothelial growth factor in ovarian physiology and pathology. Fertil. Steril., 74, 429–438.[ISI][Medline]

Geva, E., Fait, G., Lerner-Geva, L., Lessing, J.B., Swartz, T., Wolman, I., Daniel, Y. and Amit, A. (1999) The possible role of antiovary antibodies in repeated in vitro fertilization failures. Am. J. Reprod. Immunol., 42, 292–296.[ISI][Medline]

Hall, J.E., Welt, C.K. and Cramer, D.W. (1999) Inhibin A and inhibin B reflect ovarian function in assisted reproduction but are less useful at predicting outcome. Hum. Reprod., 14, 409–415.[Abstract/Free Full Text]

Hanoch, J., Lavy, Y., Holzer, H., Hurwitz, A., Simon, A., Revel, A. and Laufer, N. (1998) Young low responders protected from untoward effects of reduced ovarian response. Fertil. Steril., 69, 1001–1004.[ISI][Medline]

Horejsi, J., Martinek, J., Novakova, D., Madar, J. and Brandejska, M. (2000) Autoimmune antiovarian antibodies and their impact on the success of an IVF/ET program. Ann. NY Acad. Sci., 900, 351–356.[Abstract/Free Full Text]

Hovav, Y., Almagor, M., Benbenishti, D., Margalioth, E.J., Kafka, I. and Yaffe, H. (1994) Immunity to zona pellucida in women with low response to ovarian stimulation, in unexplained infertility and after multiple IVF attempts. Hum. Reprod., 9, 643–645.[Abstract]

Howles, C.M. (2000) Role of LH and FSH in ovarian function. Mol. Cell. Endocrinol., 161, 25–30.[ISI][Medline]

Hurst, B., Zacur, H., Schlaff, W. and Berkovitz, G. (1992) Use of granulosa–luteal cell culture to evaluate low and high clinical responses to menotropin stimulation. J. Endocrinol. Invest., 15, 567–572.[ISI][Medline]

Ivanova, M., Djarkova, T., Mollova, M., Petrov, M., Tikhomirova, T. and Dakhno, F. (1999) Zona pellucida autoantibodies in women undergoing ART. Folia Biol., 45, 59–62.

Jenkins, J., Davies, D., Devonport, H., Anthony, F., Gadd, S., Watson, R. and Masson, G. (1991) Comparison of poor responders with good responders using a standard buserelin/ human menopausal gonadotropin regime for in-vitro fertilization. Hum. Reprod., 6, 918–921.[Abstract]

Kim, S. (1995) Female aging and superovulation induction for IVF. J. Obstet. Gynaecol., 21, 75–82.[Medline]

Klein, N.A. and Soules, M. (1998) Endocrine changes of the perimenopause. Clin. Obstet. Gynecol., 41, 912–920.[ISI][Medline]

Klein, N.A., Battaglia, D., Miller, P., Branigan, E., Giudice, L. and Soules, M. (1996a) Ovarian follicular development and the follicular fluid hormones and growth factors in normal women of advanced reproductive age. J. Clin. Endocrinol. Metab., 81, 1946–1951.[Abstract]

Klein, N.A., Battaglia, D.E., Fujimoto, V.Y., Davis, G.S., Bremner, W.J. and Soules, M.R. (1996b) Reproductive aging: accelerated ovarian follicular development associated with a monotropic follicle-stimulating hormone rise in normal older women. J. Clin. Endocrinol. Metab., 81, 1038–1045.[Abstract]

Klein, N.A., Illingworth, P., Groome, N., McNeilly, A., Battaglia, D. and Soules, M. (1996c) Decreased inhibin B secretion is associated with the monotropic FSH rise in older, ovulatory women: a study of serum and follicular fluid levels of dimeric inhibin A and B in spontaneous menstrual cycles. J. Clin. Endocrinol. Metab., 81, 2742–2745.[Abstract]

Kligman, I. and Rosenwaks, Z. (2001) Differentiating clinical profiles: predicting good responders, poor responders, and hyperresponders. Fertil. Steril., 76, 1185–1190.[ISI][Medline]

Laitinen, M., Ristimaki, A., Honkasalo, M., Narko, K., Paavonen, K. and Ritvos, O. (1997) Differential hormonal regulation of vascular endothelial growth factors VEGF, VEGF-B, and VEGF-C messenger ribonucleic acid levels in cultured human granulosa–luteal cells. Endocrinology, 138, 4748–4756.[Abstract/Free Full Text]

Lindheim, S.R., Cohen, M.A., Chang, P.L. and Sauer, M.V. (1999) Serum progesterone before and after human chorionic gonadotropin injection depends on the estradiol response to ovarian hyperstimulation during in vitro fertilization–embryo transfer cycles. J. Assist. Reprod. Genet., 16, 242–246.[ISI][Medline]

Luborsky, J. and Pong, R. (2000) Pregnancy outcome and ovarian autoimmunity in infertility patients undergoing controlled ovarian hyperstimulation. Am. J. Reprod. Immunol., 44, 261–265.[ISI][Medline]

Luborsky, J.L., Visintin, I., Boyers, S., Asare, T., Caldwell, B. and DeCherney, A. (1990) Ovarian antibodies detected by immobilized antigen immunoassay in patients with premature ovarian failure. J. Clin. Endocrinol. Metab., 70, 69–75.[Abstract]

Meyer, W., Lavy, G., DeCherney, A., Visintin, I., Economy, K. and Luborsky, J. (1990) Evidence of gonadal and gonadotropin antibodies in women with a suboptimal ovarian response to exogenous gonadotropin. Obstet. Gynecol., 75, 795–799.[Abstract]

Muttukrishna, S., Child, T., Lockwood, G., Groome, N., Barlow, D. and Ledger, W. (2000) Serum concentrations of dimeric inhibins, activin A, gonadotrophins and ovarian steroids during the menstrual cycle in older women. Hum. Reprod., 15, 549–556.[Abstract/Free Full Text]

Neulen, J., Wenzel, D., Hornig, C., Wunsch, E., Weissenborn, U., Grunwald, K., Buttner, R. and Weich, H. (2001) Poor responder–high responder: the importance of soluble vascular endothelial growth factor receptor 1 in ovarian stimulation protocols. Hum. Reprod., 16, 621–626.[Abstract/Free Full Text]

Noci, I., Biagiotti, R., Maggi, M., Ricci, F., Cinotti, A. and Scarselli, G. (1998) Low day 3 luteinizing hormone values are predictive of reduced response to ovarian stimulation. Hum. Reprod., 13, 531–534.[Abstract]

Out, H., Braat, D., Lintsen, B., Gurgan, T., Bukulmez, O., Gokmen, O., Keles, G., Caballero, P., Gonzalez, J., Fabregues, F. et al. (2000) Increasing the daily dose of recombinant follicle stimulating hormone (Puregon) does not compensate for the age-related decline in retrievable oocytes after ovarian stimulation. Hum. Reprod., 15, 29–35.[Abstract/Free Full Text]

Pellicer, A., Ballester, M., Serrano, M., Mir, A., Serra-Serra, V., Remohi, J. and Bonilla-Musoles, F. (1994) Aetiological factors involved in the low response to gonadotrophins in infertile women with normal basal serum follicle stimulating hormone levels. Hum. Reprod., 9, 806–811.[Abstract]

Pellicer, A., Gaitan, P., Neuspiller, F., Ardiles, G., Albert, C., Remohi, J. and Simon, C. (1998) Ovarian follicular dynamics: from basic science to clinical practice. J. Reprod. Immunol., 39, 29–61.[ISI][Medline]

Perez Mayorga, M., Gromoll, J., Behre, H.M., Gassner, C., Nieschlag, E. and Simoni, M. (2000) Ovarian response to follicle-stimulating hormone (FSH) stimulation depends on the FSH receptor genotype. J. Clin. Endocrinol. Metab., 85, 3365–3369.[Abstract/Free Full Text]

Sauer, M. (1998) The impact of age on reproductive potential: lessons learned from oocyte donation. Maturitas, 30, 221–225.[ISI][Medline]

Schipper, I., de Jong, F. and Fauser, B. (1998) Lack of correlation between maximum early follicular phase serum follicle stimulating hormone concentrations and menstrual cycle characteristics in women under the age of 35 years. Hum. Reprod., 13, 1442–1448.[Abstract]

Scott, R.J. (1996) Evaluation and treatment of low responders. Semin. Reprod. Endocrinol., 14, 317–337.[ISI][Medline]

Seifer, D.B., Scott, R.T., Jr, Bergh, P.A., Abrogast, L.K., Friedman, C.I., Mack, C.K. and Danforth, D.R. (1999) Women with declining ovarian reserve may demonstrate a decrease in day 3 serum inhibin B before a rise in day 3 follicle-stimulating hormone. Fertil. Steril., 72, 63–65.[ISI][Medline]

Seki, M., Itoh, M., Takeuchi, T. and Ibuki, Y. (1997) Factors affecting the development of oocytes and embryos in IVF–ET. Hum. Cell, 10, 247–254.[Medline]

Stouffer, R.L., Martinez-Chequer, J.C., Molskness, T.A., Xu, F. and Hazzard, T.M. (2001) Regulation and action of angiogenic factors in the primate ovary. Arch. Med. Res., 32, 567–575.[ISI][Medline]

Surrey, E.S. and Schoolcraft, W.B. (2000) Evaluating strategies for improving ovarian response of the poor responder undergoing assisted reproductive techniques. Fertil. Steril., 73, 667–676.[ISI][Medline]

Tanbo, T., Abyholm, T., Bjoro, T. and Dale, P. (1990) Ovarian stimulation in previous failures from in-vitro fertilization: distinction of two groups of poor responders. Hum. Reprod., 5, 811–815.[Abstract]

Thiruppathi, P., Shatavi, S., Dias, J., Radwanska, E. and Luborsky, J. (2001) Gonadotropin receptor expression on granulosa cells of low and normal responders to FSH. Mol. Hum. Reprod., 7, 697–704.[Abstract/Free Full Text]

Tinkanen, U., Blauer, M., Laippala, P., Tuohimaa, P. and Kujansuu, E. (1999) Prognostic factors in controlled ovarian hyperstimulation. Fertil. Steril., 72, 932–936.[ISI][Medline]

Ubaldi, F., Camus, M., Smitz, J., Bennink, H., Van Steirteghem, A. and Devroey, P. (1996) Premature luteinization in in vitro fertilization cycles using gonadotropin releasing hormone agonist (GnRH-a) and a recombinant follicle stimulating hormone (FSH) and GnRH-a and urinary FSH. Fertil. Steril., 68, 275–280.

Vanderhyden, B. and Macdonald, E. (1998) Mouse oocytes regulate granulosa cell steroidogenesis throughout follicular development. Biol. Reprod., 59, 1296–1302.[Abstract/Free Full Text]

Whitman, G., Luciano, A., Maier, D. and Peluso, J. (1988) Influence of human chorionic gonadotropin (hCG) and hCG internalization by granulosa cells on the rate of in vitro fertilization and embryonic development of human oocytes. Fertil. Steril., 50, 607–611.[ISI][Medline]

Whitman, G., Luciano, A., Maier, D. and Peluso, J. (1989) Human chorionic gonadotropin localization and morphometric characterization of human granulosa–luteal cells obtained during in vitro fertilization cycles. Fertil. Steril., 51, 475–479.[ISI][Medline]

Younis, J.S., Matilsky, M., Radin, O. and Ben-Ami, M. (2001) Increased progesterone/estradiol ratio in the late follicular phase could be related to low ovarian reserve in in vitro fertilization–embryo transfer cycles with a long gonadotropin-releasing hormone agonist. Fertil. Steril., 76, 294–299.[ISI][Medline]

Submitted on March 4, 2002; accepted on May 23, 2002.