This Article Abstract Full Text (PDF) Submit a related Letter to the Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Tong, S. Articles by Wallace, E. M. Search for Related Content PubMed PubMed Citation Articles by Tong, S. Articles by Wallace, E. M.

Increased Day 15–17 Serum Pro-C Inhibin Levels Specific to Successful Pregnancy

Centre for Women’s Health Research (S.T., L.R., B.M., E.M.W.), Department of Obstetrics and Gynaecology, Monash University, Monash Medical Centre, and Monash IVF (L.R., A.M.), Clayton 3168, Victoria, Australia; and Springfield Hospital (J.L.O.), Chelmsford CM1 7GU, United Kingdom

Address all correspondence and requests for reprints to: Dr. Stephen Tong, Department of Obstetrics and Gynaecology, Monash University, Monash Medical Centre, 246 Clayton Road, Clayton, Victoria 3168, Australia. E-mail: Stephen.Tong{at}med.monash.edu.au.

	Abstract

Top Abstract Introduction Subjects and Methods Results Discussion References

 
In early pregnancy, serum levels of the luteal-derived hormone pro-C inhibin peak by the second week after conception. Whether this early rise is biologically important and a consistent feature of only successful pregnancy is unknown. We undertook a prospective cross-sectional study to determine whether serum pro-C inhibin levels at d 15–17 are predictive of a successful clinical in vitro fertilization pregnancy and compared levels between fresh embryo transfer (ET) and frozen-thawed ET. Median (95% confidence interval) pro-C inhibin levels were 68 (57–76) pg/ml in 204 women who did not become clinically pregnant after ET, significantly lower than in either 90 women who became clinically pregnant after fresh ET and who had 3139 (1684–4220) pg/ml or in 39 women with a successful frozen ET who had 877 (678–1111) pg/ml. Pro-C was highly sensitive and specific in predicting clinical pregnancy success but did not improve on the performance of human chorionic gonadotropin. Pro-C inhibin levels were not correlated with progesterone or human chorionic gonadotropin. Levels were no higher in singleton compared with multiple pregnancies and did not increase across gestation, suggestive of a luteal source. The increase in circulating pro-C inhibin in very early pregnancy is highly specific to clinical pregnancy, suggesting a possible biological role in early gestation.

	Introduction

Top Abstract Introduction Subjects and Methods Results Discussion References

 
THE INHIBINS ARE glycoprotein hormones belonging to the transforming growth factor-ß superfamily and thought to have diverse roles in human reproduction (1, 2). Inhibin comprises an -subunit linked by a disulfide bond to either a ß_A (inhibin A) or ß_B (inhibin B) subunit. Of these two dimers, only inhibin A is detectable in the peripheral circulation in pregnancy (3, 4). However, both the - and ß- subunits are produced as pre-pro-proteins giving rise to a large number of different molecular weight forms relating to different stages of processing (5). Indeed, it is now recognized that many of these different molecular weight forms of inhibin A are present in the maternal circulation (4), including inhibin with a partially processed -subunit, termed pro-C inhibin.

Most pregnancy-related hormones that are detectable in the maternal circulation rise progressively across the first few weeks of pregnancy. These include the luteal-derived steroid progesterone and fetoplacentally derived hormones such as human chorionic gonadotropin (hCG), estradiol, and other inhibin forms, such as activin A and inhibin A (3, 6). In contrast, maternal serum levels of pro-C inhibin appear to peak at around d 16 after conception (6, 7), then gradually decline. Although a secondary rise in pro-C inhibin occurs in early second trimester and continues until the end of pregnancy, absolute levels do not surpass those seen in very early pregnancy (6). Given these apparent differences in the ontogeny of pro-C inhibin compared with most other pregnancy hormones, it is conceivable that the early rise in pro-C inhibin reflects an, as yet uncharacterized, biological role for these inhibin forms in the establishment of early pregnancy. Accordingly, we hypothesized that if pro-C inhibin did have such a role, then detectable levels should rise consistently with the establishment of pregnancy and that pro-C inhibin measured by around d 16 after conception in women undergoing in vitro fertilization (IVF) may be a useful predictor of whether a clinical pregnancy will result. We therefore undertook a prospective study to evaluate pro-C inhibin levels at d 13–14 after embryo transfer (ET), in both fresh and frozen ET cycles, and to assess the clinical utility of pro-C inhibin as a predictor of subsequent pregnancy success.

Illingworth et al. (7) showed that the administration of hCG at around d 7 after ovulation in nonpregnant women increases pro-C inhibin levels. In addition, a recent study has shown that pro-C inhibin starts to fall very quickly upon administration of the antiprogestagen mifepristone in women undergoing medical termination of pregnancy (8). In light of these data, we also explored whether any correlation existed between pro-C inhibin and hCG or progesterone.

	Subjects and Methods

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Subjects and samples

We undertook a prospective, cross-sectional observational study of 334 women undergoing IVF. Blood was collected at d 13–14 after ET, equivalent to the 15th to 17th d after conception, or by convention, 4 wk of gestation. Blood samples were also collected from some women at a time equivalent to 22–24 d (n = 73) and 32–35 d (n = 39) gestation.

Baseline patient information was collected, including age, current IVF treatment cycle number, and whether the ET cycle was fresh or frozen-thawed. Women were grouped according to whether they became clinically pregnant or not and, if so, whether they had a singleton or twin pregnancy. Clinical pregnancy was defined by a positive pregnancy test (hCG) and an early ultrasound scan at approximately 6 wk of gestation confirming the presence of one or two gestational sacs with a fetal pole and visible cardiac activity. The nonpregnant group included both women with a negative pregnancy test and those who had had an initially positive pregnancy test but ultimately no demonstrable viable gestation on ultrasound (i.e. biochemical pregnancies).

Approval for the study was granted by the Monash IVF Ethics Committee before it was commenced. All women gave written informed consent before the collection of samples.

Treatment protocols

Women undergoing fresh IVF cycles were administered the GnRH agonist nafarelin (Synarel, Searle, High Wycombe, UK) for pituitary desensitization followed by sc recombinant human (rh)FSH (Gonal-F, Ares-Serono, Geneva, Switzerland). When the dominant follicle or follicles were greater than 17 mm in diameter, both nafareline and rhFSH were discontinued, and before oocyte retrieval, 5000–10,000 IU of hCG (Profasi, Serono Laboratories, Geneva, Switzerland) was administered. The oocytes were fertilized in vitro either with or without sperm microinjection. In fresh cycles, embryos were transferred after 3 d. After ET, luteal support was provided by administration of 400 mg daily progesterone vaginal suppositories.

In frozen-thawed cycles, embryos were maintained in culture for 2–3 d before freezing. Unlike fresh IVF cycles, a number of different transfer regimens were used, as briefly described in Table 1.

Serum hCG (immunometric assay) and progesterone (competitive immunoassay) were assayed at the time of blood collection using an automated system (VitrosECi, OrthoDiagnostics, Rochester, NY). Excess serum was frozen at –20 C for a period of less than 8 months for subsequent measurement of pro-C inhibin (Oxford Bio Innovation, Upper Heyford, Oxfordshire, UK). The assay for pro-C inhibin was performed in a single batch after a single thaw cycle. All three assays were done by scientists blinded to the clinical outcome.

Assays were performed according to manufacturer’s instructions. The intra- and interassay coefficient of variation for all assays was less than 10%. The sensitivities of the assays were as follows: 6 pg/ml for pro-C inhibin, 0.5 mIU/ml for hCG, and 0.08 ng/ml for progesterone.

Statistical analyses

Levels of all analytes were not normally distributed. Therefore, data were expressed as median, with a 95% confidence interval (CI), and comparisons between groups undertaken using either Kruskal-Wallis or Mann-Whitney U tests. Correlations between hormone levels were performed using Spearman’s correlation.

The performance of pro-C inhibin in predicting viable pregnancy was determined by a logistic regression model that included pregnancy outcome, treatment method (fresh vs. frozen ET), and the categorized pro-C levels at the cutoff level. Statistical tests were done using Stata 8.1 (Stata Corp., College Station, TX).

	Results

Top Abstract Introduction Subjects and Methods Results Discussion References

 
The mean (SD) age of all women in the study was 34.1 (4.9) yr, and the median (95% CI) current IVF treatment cycle number was 3 (3, 4). Furthermore, baseline data for the women, grouped according to gestation at blood sampling, are summarized in Table 2.

Pro-C inhibin

At d 15–17, the median (95% CI) pro-C inhibin levels in the 204 women who did not become clinically pregnant was 68 (57–76) pg/ml. Median levels were significantly higher in the 107 women with a singleton pregnancy at 1372 (1003–1963) pg/ml (P < 0.0001) and in the 23 women who had a twin pregnancy at 2000 (897–3735) pg/ml (P < 0.0001) (Fig. 1).

View larger version (23K): [in this window] [in a new window]   FIG. 1. Serum pro-C levels (median and 95% CI) in the entire cohort according to gestation at sample collection as follows: no clinical pregnancy (), singleton gestation (), or twins (). Numbers in each cohort are shown above bars. For pro-C levels among singleton vs. twin pregnancies at all gestations, P > 0.09 for all comparisons (Mann-Whitney U). For pro-C levels across gestation within single or twin cohorts, P > 0.44 (Kruskal-Wallis).

View larger version (33K): [in this window] [in a new window]   FIG. 2. Serum pro-C levels (median and 95% CI) grouped to whether there was a clinical pregnancy and whether the ET was either fresh () or frozen-thawed (). Numbers in each cohort are shown above bars. Of the no-clinical-pregnancy subgroup, levels at d 20–21 and d 30–32 were not included because of low numbers within each subgroup.

Pro-C inhibin levels in women with a clinical pregnancy after a fresh ET did not differ, at any gestation, whether a twin or singleton pregnancy resulted (Kruskal-Wallis for viable singletons vs. multiples in fresh embryo cohort from all three gestations P = 0.42; six groups). This analysis was not repeated for the frozen ET group because of insufficient numbers of multiple pregnancies.

The pro-C inhibin levels at d 15–17 in the frozen ET group who underwent different regimens are shown in Table 3. There were no differences in levels between the nonpregnant and clinically pregnant women who had the hormone replacement therapy (HRT) regimen, but there were clear differences in the pregnant vs. nonpregnant subgroups among the natural cycle regimen or the clomiphene citrate regimen.

View this table: [in this window] [in a new window]   TABLE 3. Levels of d 15–17 pro-C in the frozen-thawed ET group according to the type of regimen undertaken

Correlation of pro-C levels with progesterone and hCG

Table 4 shows the levels of hCG and progesterone levels at d 15–17. Correlations between analytes were limited to those cases where a clinical pregnancy eventuated. There were 13 cases where progesterone levels were not available. Neither serum hCG (r = 0.14; P = 0.11) nor progesterone (r = 0.078; P = 0.16) were correlated with pro-C levels.

Assessment of pro-C inhibin as a predictive marker of clinical pregnancy

At 95% specificity, a pro-C level of 308 pg/ml at 15–17 d after ET afforded an 83% sensitivity in the prediction of a successful clinical pregnancy. In comparison, for the same specificity, hCG afforded a sensitivity of 90%. Of the 204 women who did not become pregnant, there were 21 who would have had a positive hCG at a 10–50 mIU/ml cutoff (false positive). Of these, 16 would have had a pro-C inhibin level less than 308 pg/ml, which would have been defined as a negative test for the presence of a pregnancy. Of 15 who had a hCG level at 50–524 mIU/ml but did not become pregnant, seven had pro-C inhibin levels less than 308 pg/ml. However, logistic regression modeling found that a combination of pro-C inhibin and hCG was not better than hCG alone in predicting a successful clinical pregnancy (Table 5).

View this table: [in this window] [in a new window]   TABLE 5. Sensitivities of hCG and pro-C at d 15–17 in predicting the presence of a successful clinical pregnancy

In light of the possibility that the lack of any increase of pro-C in the HRT clinical pregnancy subgroup (Table 3) might have adversely affected the performance of the entire frozen ET group, we undertook a post hoc analysis to determine the performance of pro-C in predicting pregnancy, excluding this group. Because pro-C levels among women undergoing the clomiphene citrate regimen were no different from those who underwent the natural cycle (P = 0.58), these two groups were combined for this analysis. At 95% specificity, pro-C inhibin levels were 100% and hCG levels were 96.7% sensitive, respectively, in predicting pregnancy success.

	Discussion

Top Abstract Introduction Subjects and Methods Results Discussion References

 
In this study, we report the largest series to date on maternal serum pro-C inhibin levels measured during early IVF pregnancies, and we have been the first to evaluate its sensitivity as a predictive marker of achieving a successful clinical pregnancy at 6 wk gestation. We have found that serum levels of pro-C inhibin are consistently and considerably increased at 13–14 d after IVF ET in those women destined to achieve a clinical pregnancy. However, pro-C inhibin does not improve upon the predictiveness of outcome already afforded by hCG.

Currently, women undergoing IVF treatment are provided with their first indication of likely success from hCG levels measured at a time corresponding to approximately d 16 after conception. hCG is often very good at predicting whether a clinical pregnancy will occur, particularly if it is highly elevated (9, 10). However, when levels are around 50–100 mIU/liter, it may be difficult to accurately predict whether that IVF cycle will be successful (9). Other markers have been investigated, including inhibin A (11). Although inhibin A certainly increases in association with pregnancy success, there is too much overlap between successful and unsuccessful pregnancies for inhibin A to be clinically useful (11).

In contrast, levels of pro-C inhibin at d 15–17 were highly specific and sensitive in predicting pregnancy success. Nonetheless, pro-C inhibin did not add to the usefulness of hCG, the currently used marker. However, the 90% sensitivity at 95% specificity that we observed for hCG in our cohort may be better than that reported by others (9, 12). It is therefore possible that other centers might find pro-C inhibin to be more predictive than we have. Pro-aC inhibin may also prove to be a useful diagnostic tool in other disorders of early pregnancy, such as ectopic pregnancy, although this was not explored in this current study. Finally, we did not investigate whether pro-C at d 15–17 may be able to further discriminate those who have a clinically recognized pregnancy, but later miscarry.

We found that among those women undergoing frozen-thawed ET who underwent either a natural cycle regimen (expected to have only one corpus luteum) or clomiphene citrate (expected to have had one or more than one corpus luteum), both pro-C and hCG appear to offer extremely high sensitivities in predicting a subsequent clinical pregnancy. For this analysis, we excluded the HRT subgroup (expected to have no luteal tissue) after we found that there was no rise in pro-C in this group. Because this was a post hoc finding, it requires confirmation in a future study of larger numbers. Nonetheless, if this is verified, it may suggest that in these particular regimens, both these hormones might be extremely sensitive and specific at predicting clinical pregnancy outcome. The fact that the predictive ability may significantly increase in these specific regimens, especially when compared with fresh cycles, may be explained by the fact that with less luteal tissue, the distribution of serum levels of pro-C might be less in both the pregnant and nonpregnant groups.

Although the relative contributions of the corpus luteum and the fetoplacental unit to circulating serum levels of inhibin A in early pregnancy remain uncertain (6, 7, 13), it is generally accepted that the predominant source of pro-C inhibin is the corpus luteum (13, 14). Our data would support this. We reported that pro-C inhibin levels are no higher with advancing gestation, are increased in proportion to the expected amount of luteal tissue present (i.e. fresh vs. frozen ET), and are no higher if the pregnancy is multiple rather than single. These observations are more in keeping with a luteal than a placental origin. Furthermore, that we observed that the levels of pro-C inhibin in women with a frozen-thawed embryo pregnancy are significantly higher than levels previously reported (13, 14) reflects that the women in this current study were treated with protocols where there is often luteal function present, whereas in previous reports (13, 14), women receiving frozen-thawed ETs underwent a protocol of complete ovarian suppression and therefore had a total absence of luteal tissue.

Because there is an obvious discrimination in pro-C inhibin levels between successful and unsuccessful cycles and because it has been previously shown that administration of exogenous hCG rescues luteal production of pro-C inhibin (7), it is likely that hCG is the physiological stimulus for the higher luteal pro-C inhibin production observed in the successful pregnancies. However, we were unable to confirm an association between circulating pro-C inhibin levels and either progesterone (8) or hCG (7) at d 15–17 in the maternal serum. We did find a weak negative correlation between pro-C inhibin levels and progesterone in the viable frozen ET pregnancies, but because all other correlations were found to be strongly nonsignificant, this may be a type I error and interpretation should be cautious. In addition, we cannot exclude the possibility that a true relationship exists between endogenous progesterone and serum levels of hCG or pro-C inhibin but was affected by the administration of exogenous progesterone. The more convincing experimental design to explore relationships between progesterone and pro-C inhibin would be to undertake interventional studies, such as withholding luteal support and the administration of mifepristone.

Clearly, pro-C inhibin is not essential for successful pregnancy because protocols employing complete ovarian suppression causing the absence of luteal tissue are compatible with successful pregnancy (14). However, the fact that an early rise in pro-C levels appears to be a consistent feature specific to clinical pregnancy suggests that it may have a biological role. In this regard, implantation has been previously shown to be optimal at d 8–10 after fertilization (15). It has also been reported recently that in the week immediately after implantation, an early hCG rise and an abrupt increase in progesterone is associated with fewer pregnancy losses (16). It is therefore possible that the progressive rise of pro-C inhibin across this same period (7) may reflect the fact that these inhibin forms might also play important roles such as the facilitation of implantation and/or decidualization. In support of this possibility are immunohistochemical studies showing that the -subunit staining in the nonpregnant endometrium is weak but becomes strong after decidualization (17). Such speculation should be tempered with some caution because it is presently believed that only the dimeric inhibin forms are biologically active, and we have investigated an IVF cohort, not spontaneous pregnancies.

In summary, we have demonstrated a massive and specific early rise of serum pro-C inhibin levels in association with clinical pregnancy. The biological relevance of this, if any, remains uncertain, but the current data suggest further study of pro-C inhibins in early pregnancy would be merited.

	Footnotes

 
Abbreviations: CI, Confidence interval; ET, embryo transfer; hCG, human chorionic gonadotropin; HRT, hormone replacement therapy; IVF, in vitro fertilization; rh, recombinant human.

Received January 7, 2004.

Accepted June 16, 2004.

	References

Top Abstract Introduction Subjects and Methods Results Discussion References

 

1. de Kretser DM, Hedger MP, Loveland KL, Phillips DJ 2002 Inhibins, activins and follistatin in reproduction. Hum Reprod Update 8:529–541[Abstract/Free Full Text]
2. Tong S, Wallace EM, Burger HG 2003 Inhibins and activins: clinical advances in reproductive medicine. Clin Endocrinol (Oxf) 58:115–127[CrossRef][Medline]
3. Wallace EM, Riley SC, Crossley JA, Ritoe SC, Horne A, Shade M, Ellis PM, Aitken DA, Groome NP 1997 Dimeric inhibins in amniotic fluid, maternal and fetal serum in human pregnancy. J Clin Endocrinol Metab 82:218–222[Abstract/Free Full Text]
4. Thirunavukarasu P, Stephenson T, Forray J, Stanton PG, Groome N, Wallace EM, Robertson DM 2001 Changes in molecular weight forms of inhibin A and pro-C in maternal serum during human pregnancy. J Clin Endocrinol Metab 86:5794–5804[Abstract/Free Full Text]
5. Robertson DM, Sullivan J, Cahir N 1995 Inhibin forms in human plasma. J Endocrinol 144:261–269[Abstract/Free Full Text]
6. Fowler PA, Evans LW, Groome NP, Templeton A, Knight PG 1998 A longitudinal study of maternal serum inhibin-A, inhibin-B, activin-A, activin-AB, pro-C and follistatin during pregnancy. Hum Reprod 13:3530–3536[Abstract/Free Full Text]
7. Illingworth PJ, Groome NP, Duncan WC, Grant V, Tovanabutra S, Baird DT, McNeilly AS 1996 Measurement of circulating inhibin forms during the establishment of pregnancy. J Clin Endocrinol Metab 81:1471–1475[Abstract]
8. Lahiri S, Anobile CJ, Stewart P, Ledger WL 2003 Changes in circulating concentrations of inhibin A and pro-C during first trimester medical termination of pregnancy. Hum Reprod 18:744–748[Abstract/Free Full Text]
9. Sugantha SE, Webster S, Sundar E, Lenton EA 2000 Predictive value of plasma human chorionic gonadotrophin following assisted conception treatment. Hum Reprod 15:469–473[Abstract/Free Full Text]
10. Homan G, Brown S, Moran J, Homan S, Kerin J 2000 Human chorionic gonadotropin as a predictor of outcome in assisted reproductive technology pregnancies. Fertil Steril 73:270–274[CrossRef][Medline]
11. Treetampinich C, O’Connor AE, MacLachlan V, Groome NP, de Kretser DM 2000 Maternal serum inhibin A concentrations in early pregnancy after IVF and embryo transfer reflect the corpus luteum contribution and pregnancy outcome. Hum Reprod 15:2028–2032[Abstract/Free Full Text]
12. Urbancsek J, Hauzman E, Fedorcsak P, Halmos A, Devenyi N, Papp Z 2002 Serum human chorionic gonadotropin measurements may predict pregnancy outcome and multiple gestation after in vitro fertilization. Fertil Steril 78:540–542[CrossRef][Medline]
13. Lockwood GM, Ledger WL, Barlow DH, Groome NP, Muttukrishna S 1997 Measurement of inhibin and activin in early human pregnancy: demonstration of fetoplacental origin and role in prediction of early-pregnancy outcome. Biol Reprod 57:1490–1494[Abstract]
14. Lockwood GM, Ledger WL, Barlow DH, Groome NP, Muttukrishna S 1998 Identification of the source of inhibins at the time of conception provides a diagnostic role for them in very early pregnancy. Am J Reprod Immunol 40:303–308
15. Wilcox AJ, Baird DD, Weinberg CR 1999 Time of implantation of the conceptus and loss of pregnancy. N Engl J Med 340:1796–1799[Abstract/Free Full Text]
16. Baird DD, Weinberg CR, McConnaughey DR, Wilcox AJ 2003 Rescue of the corpus luteum in human pregnancy. Biol Reprod 68:448–456[Abstract/Free Full Text]
17. Jones RL, Salamonsen LA, Findlay JK 2002 Potential roles for endometrial inhibins, activins and follistatin during human embryo implantation and early pregnancy. Trends Endocrinol Metab 13:144–150[CrossRef][Medline]

This article has been cited by other articles:

				S. Tong, L. Rombauts, J. Onwude, B. Marjono, and E.M. Wallace Highly specific and sensitive rise in Days 14 17 pro-{alpha}C inhibin with clinical pregnancy after frozen embryo transfer with ovulatory cycles Hum. Reprod., August 1, 2007; 22(8): 2249 - 2253. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Submit a related Letter to the Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Tong, S. Articles by Wallace, E. M. Search for Related Content PubMed PubMed Citation Articles by Tong, S. Articles by Wallace, E. M.