DHEA and miscarriage?

When i carried out a (quick) review of the literature on DHEA and fertility the other day i kept the focus of looking for data that might be useful to my own situation (very low ovarian function). In the process some other interesting information came up though. Perhaps this post will be intriguing to anyone who stumbles across my blog, although it comes with the caveat that i haven’t done much analysis into this subject. I don’t have to worry about miscarriage seeing as i’m not releasing eggs.

The most interesting article that turned up in my search of recent papers  mentioning DHEA was not actually concerned with patients with diminished ovarian reserve. It was interesting because it is one of the few studies with DHEA and fertility that is from a double-blind, randomised, placebo-controlled trial*. Some findings of this study actually contradict earlier studies on DHEA and ovarian function, in that these women undergoing IVF did not produce more oocytes when given DHEA. Although, it must be remembered that this study involved women with normal ovarian reserve (high AMH level, high antral follicle count) and therefore is not directly comparable to studies of women with low ovarian reserve/function. Also, it is not known how DHEA (endogenous or otherwise) contributes to ovarian function. Without a mechanism of action to guide hypotheses, previous studies have investigated the use of DHEA to improve ovarian response in a quantitative manner, i.e. does DHEA treatment along with stimulation protocols result in more eggs to use in assisted reproduction?

This new study provides a suggestion that DHEA might not be useful to the issue of quantity, but rather quality. Women in the treatment group had a higher live birth rate (the one measure that the patients are really interested in!). This was not attributable to a higher pregnancy rate (similar in both groups) but seemed to be due to the miscarriage rate (none of the DHEA group pregnancies resulted in miscarriage, whereas 5 of the control group did). This is an interesting result due to the nature of the study. The patients enrolled in the study essentially had unexplained infertility, with age being the only indication of reduced ovarian function, and the randomisation of the patients appears to have been carried out without bias.

The authors (Tartagni et al.) speculate that their results are due to a difference in aneuploidy rates. They cite some earlier work from the Gleicher lab (at the Center for Human Reproduction, where the patent for using DHEA for DOR is held) to support this hypothesis. These earlier publications do not constitute strong evidence simply because they are case-control studies rather than randomised, prospective studies. Gleicher et al. 2010** for instance shows some difference in aneuploidy between women who had taken DHEA and controls who had not. However, I had some difficulty in understanding how the case reports were assigned to the two groups. The authors admit that the treatment group were those that were assessed as having lower ovarian reserve (surely this constitutes a confounding factor then?). Of note is a very odd error in that the authors at this point cite themselves (for AMH levels used to determine DOR), or at least appear to, in a citation that doesn’t exist:
“13. Barad DH, Weghofer A, Goyal A, Gleicher N: Age-specific anti-Müllerian hormone (AMH): Utility of AMH at various ages. Reprod Biomed Online .” This citation is clearly lacking some details (year?, journal volume?), additionally a search of the exact title as cited on google scholar or the journal website yields no such article.

A newer article from the Gleicher lab***, published after the Tartagni study confuses matters again. This analysis is a little different in that it is not concerned with DHEA supplementation really, but with endogenous androgens (DHEA and other hormones produced in the patients bodies themselves). Levels of androgens were expected to correlate with levels of aneuploidy in embryos from the women, but no such correlation was found. The authors claim there are previous reports of an association between aneuploidy and low androgen concentrations:

“Only two possible explanations come to mind: In the literature reported associations between LFOR and in-creased aneuploidy [4,5] and of increased aneuploidy with low androgen concentrations [7,8] are incorrect. While a possibility, we consider this a less likely explanation.”

7. Gleicher N, Weghofer A, Barad DH. Dehydroepiandrosterone (DHEA) reduces embryo aneuploidy: direct evidence from preimplantation genetic screening (PGS). Reprod Biol Endocrinol. 2010;8:140.

8. Gleicher N, Ryan E, Weghofer A, Blanco-Mejia S, Barad DH. Miscarriage rates after dehydroepiandrosterone (DHEA) supplementation in women with diminished ovarian reserve: a case control study. Reprod Biol Endocrinol. 2009;7:108.

Not only does this statement stand unjustified (no reason given for why they think it is ‘less likely’) but the references given in the case of the low androgen vs. aneuploidy are not adequate citations, even though they are their own work. That is, no association between androgen levels and aneuploidy is reported in these articles, no androgen levels are reported at all. The patients may well have been assumed to have low androgen levels, but this is hardly evidence of a ‘reported association’.

The alternative explanation given for no association is that there may have been bias in selecting the patients, in that they may form a strange group of ‘lean PCOS’ patients (with low androgens??). Then, in the fourth to last paragraph of the article, Gleicher et al. mention a possible threshold effect that doesn’t make sense to me at all. The idea seems to be that women with POI/DOR/LFOR (pick an acronym) have low androgen levels, and that this could lead to a higher miscarriage rate, although the women in this study had low androgen levels yet ovarian function was fine and miscarriage did not correlate with androgen level. Confusing? yes. And it’s my opinion that high quality data will be the only answer to ‘unfuddle’ this area. Tartagni et al. could very well be a start, and will hopefully be followed by more highly powered studies.

* Tartagni et al. 2015 http://www.biomedcentral.com/content/pdf/s12958-015-0014-3.pdf; DOI 10.1186/s12958-015-0014-3

**Gleicher et al. 2010 http://www.biomedcentral.com/content/pdf/1477-7827-8-140.pdf; 

***Gleicher et al. 2015 http://www.biomedcentral.com/content/pdf/s12958-015-0034-z.pdf;

DOI 10.1186/s12958-015-0034-z


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s