OPINIONs 002: June 23, 2014
CHR is seriously concerned about the uncontrolled and, in our opinion, premature reintroduction of preimplantation genetic screening (PGS) into routine infertility care. We are, furthermore, concerned about what, to us, appears as a biased peer review process at major medical journals within the Ob/Gyn specialty, when the issue of PGS is presented. These biases are, likely, unintended consequences of selecting expert reviewers with either unbalanced intellectual investments in the PGS concept or with potential economic conflicts as providers of PGS laboratory services. We, therefore, call upon journal editors to not repeat the errors of the first introduction of PGS, which caused significant harm to tens of thousands of women worldwide.
We encourage editors to allow for a more vigorous and unbiased discussion of PGS in the pages of their journals, as we are convinced that this will rapidly lead to a more balanced presentation of advantages and disadvantages of PGS in in vitro fertilization (IVF) practice. Since we consider PGS, performed in attempts to improve IVF outcomes, to represent experimental procedures, these procedures should be only performed under study conditions and with appropriate informed consents.
Our specialty has now, for the second time in a decade, “fallen in love” with preimplantation genetic screening (PGS), the concept of selecting euploid embryos and avoiding aneuploid embryos for embryo transfer in attempts to improve in vitro fertilization (IVF) cycle outcomes. PGS represents a seemingly impeccably logical concept, since aneuploid embryos, in a large majority, either do not implant or are miscarried. But seemingly impeccably logical hypotheses are very frequently proven wrong in actual clinical medical practice. If the seemingly impeccable logic of an idea could guarantee clinical success, clinical trials would become superfluous.
In this OPINION, CHR follows up on a number of publications on this website [for the most recent overview seehere and here; for a complete list of journal publications, see here], in which we repeatedly warned about the premature reintroduction of PGS after an earlier version of PGS, approximately 10 years ago, was established as a colossal failure of practice, which not only failed to improve IVF outcomes but, actually, harmed IVF results in older women.
We at the time attempted to warn our colleagues about the premature utilization of PGS because our reanalysis of published clinical trials by Belgian colleagues suggested that PGS may not be successful in improving IVF outcomes, as was widely claimed, but may actually reduce pregnancy chances in older women. We, however, were unable to publish our concerns because medical journals in our specialty uniformly rejected the submission, in which we outlined statistical arguments in support of our concerns.
Only once colleagues from The Netherlands published a prospectively randomized study of PGS in the prestigious New England Journal of Medicine,1 demonstrating outcomes compatible with our statistical predictions, did Fertility & Sterility, the official organ of the American Society for Reproductive Medicine (ASRM), publish our paper after expedited re-review.2 The delay of a full year in transmitting the message, however, resulted in thousands of additional women worldwide undergoing PGS for no good reasons and, potentially, even harming their pregnancy chances, while in addition paying for these unnecessary services.
Now, only a few short years later, history appears to repeat itself. An allegedly improved version of PGS is once again being introduced into routine IVF, this time claiming that the initial failure of PGS was only due to inadequate technology in assessing chromosomal abnormalities in embryos. Proponents of the new PGS promise that, since the new PGS greatly improved the technologies to assess chromosomal abnormalities in embryos, this time it will fulfill the promise of improving IVF pregnancy rates and reducing spontaneous miscarriage rates after IVF.
We believe otherwise. CHR investigations of published data strongly suggest that it was unlikely that the previous failure of PGS was due primarily to technical incompetence. As a much more likely explanation, data suggest that in a majority of infertility patients, PGS simply does not make mathematical sense. Whatever small minority in which the procedure may make statistical sense has so far not been identified, but, counter-intuitively, may actually primarily include young women who already have the best IVF pregnancy chances.3
Like the last time around, PGS proponents still control the peer review process in most medical journals within the Ob/Gyn specialty. Fortunately, however, the publication scene has changed in recent years. Due to the emergence of electronic journals, leading journals like Fertility & Sterility and Human Reproduction, the official organ of the European Society for Reproductive Medicine and Embryology (ESHRE) no longer possess a publication monopoly to the same degree as they once did. With the rapid ascendance of high-quality electronic journals, the chance of getting dissenting opinions into (electronic) print has, therefore, greatly improved.4
Biases, however, persist as the following example well demonstrates: A group of colleagues recently published in Fertility & Sterility a series of papers strongly supportive of the new PGS. Two principal scientists in this group’s PGS research team are members of the journal’s editorial board. While editorial board membership should not preclude publication in the journal, it should require special scrutiny during the review process for submitted papers before acceptance and publication. This, however, does not appear to be the case at this journal, as published manuscripts of editorial board members in the pages have skyrocketed in recent years.
In this case, the journal, in addition, rejected a rebuttal article and was only willing to consider a short letter-to-the-editor response (the journal recently also eliminated this form of written response, allowing comments on published articles only via the journal’s blog).
Interestingly, using data from the same study, the authors then published an additional paper in the American Journal of Obstetrics and Gynecology (AJOG),5 which, to a significant degree, overlapped with the earlier papers published in Fertility & Sterility (which raises potential concerns about double publication). In presentation of data and data analyses, the manuscript breached the same statistical rules, already previously criticized by CHR investigators in the Fertility & Sterility publication.6,7
Here are some details: Per original trial registration, the primary goal of this so-called BEST study was a non-inferiority trial between allegedly (see below) randomized patients, with one arm undergoing trophectoderm biopsy and chromosomal evaluation, followed by elective single embryo transfer (eSET), and the other arm receiving a two-embryo transfer (2ET) of untested embryos. For the relatively small number of patients, non-inferiority was set at a remarkably high 20%. Though eSET resulted in -4.5% absolute and -7.2% relative lower pregnancy rates, the authors, nevertheless, concluded that non-inferiority existed.
In the subsequent AJOG paper,5 the authors further obfuscated their real results by comparing delivery outcomes by combining pregnancy rates for an initial fresh and a first frozen-thawed cycle. Doing this, cumulative pregnancy rates were 69% and 72%, respectively, in favor of 2ET. That eSET plus a first frozen-thawed cycle result in similar (though still slightly lower) pregnancy rates as one 2ET cycle has, even in absence of trophectoderm biopsy and aneuploidy testing (PGS), been widely reported in the literature for years. The authors’ claims that PGS in their study established outcome benefits are, therefore, incorrect. All they demonstrated is that PGS added cost to an already very expensive IVF procedure.
Claiming randomization of patients was also misleading since randomization occurred only among patients who had embryos that reached the blastocyst stage on days 5/6 of culture. Unfavorable patients whose embryos did not make it that far, therefore, never entered randomization, suggesting a very strong selection bias towards favorable patients, which was, of course, unreported by the authors. By assessing their outcomes per embryo transfer rather than by “intent to treat” (i.e. cycle start), they rendered their data uninterpretable. Moreover, this kind of biased data presentation, of course, very obviously artificially inflates pregnancy rates in comparison to a properly conducted intent to treat analysis, with reference point cycle start.
Despite having been made aware of the shortcomings of their statistical methods in their initial publication, the authors, nevertheless, repeated in the AJOG paper their earlier claim from the Fertility & Sterility manuscript that eSET resulted in “dramatically” lower multiple births (mostly twins) than 2ET. That eSET reduces twin pregnancies is, of course, widely known. Similar reductions in twin pregnancies have also been reported in earlier published studies without any aneuploidy testing if eSET was performed. Nothing in their manuscript, therefore, allows for the conclusion that testing embryos by PGS reduces twin pregnancies.
Aneuploidy testing, thus, just appears to add costs, and with proper statistical outcome evaluations, likely, actually reduces pregnancy chances by forcing even unfavorable patients into blastocyst stage cultures. This approach would frequently withhold from these unfavorable patients even the opportunity of embryo transfer.
This OPINION is not the space to further discuss whether twin pregnancies in a fertility paradigm really should be considered adverse outcomes, as these authors have done (and many colleagues concur with). We do not concur with such an opinion, as frequently repeatedly noted on this website [here, here and here].8 eSET will, therefore, in the near future be subject of another OPINION.
The AJOG study is discussed here in so much detail because the journal rejected a letter to the editor that criticized here described shortcomings. This is emblematic of what we, as noted above, consider to represent the peer review process biased in favor of PGS. Journal editors, of course, select PGS experts as reviewers, and those, more frequently than not, are intellectually invested in the concept of PGS or, consciously or subconsciously, may be biased because of economic interests in PGS laboratories.
We consider PGS literature published in the last two years to be especially biased in favor of PGS. Caution, therefore, should be exerted in interpreting PGS studies published in this time period.
What makes us issue an OPINION on this subject is our concern that history is in the process of repeating itself, with thousands of women worldwide, once again, going through PGS, without gaining any outcome benefits and, in many cases, actually being hurt in their IVF pregnancy chances. The patients, in addition, are also asked to pay for these services.
In our opinion, PGS, if applied to improve IVF cycle outcomes, should be considered an experimental procedure. The procedure should only be offered under study protocols and with appropriately obtained informed consents.
- Mastenbroek S, Twisk M, van Echten-Arends J, Sikkema-Raddatz B, Korevaar JC, Verheve HR, Vogel NE, Arts EG, deVries JW, Bossuyt PM, Buys CH, Heineman MJ. Repping S, van der Veen F. In vitro fertilization with preimplantation genetic screening. N Engl J Med 2007;357:917
- Gleicher N, Weghofer A, Barad D. Preimplantation genetic screening for aneuploidy screening: established and ready for prime time? Fertil Steril 2008;89:780-788
- Gleicher N, Barad DH. A review of, and commentary on the ongoing second clinical introduction of preimplantation genetic screening (PGS) to routine IVF practice. J Assist Reprod Genet 2012;29:1159-1166
- Gleicher N, Kushnir VA, Barad DH. Preimplantation genetic screening (PGS) still in search of a clinical application: a systematic review. Reprod Biol Endocrinol 214;12:22
- Forman EJ, Hong KH, Franasiak JM, Scott Jr. RT. Obstetrical and neonatal outcomes from the BEST Trial: single embryo transfer with aneuploidy screening improves outcomes after in vitro fertilization without compromising delivery rates. Am J Obstet Gynecol 2014;210:157:e1-6
- Gleicher N, Barad DH. A review of, and commentary on, the ongoing second clinical introduction of preimplantation genetic screening (PGS) to routine IVF practice. J Assist Reprod Genet 2012;29:1159-1166
- Gleicher N, Kushnir VA, Barad DH. Preimplantation genetic screening is alive and well: Really? Fertil Steril 100;e:36
- Gleicher N, Kushnir VA, Barad DH. What ‘misguided campaign’ against single embryo transfer? Hum Reprod 2014;29:380-381