Preimplantation genetic screening 2 0 An evolving and promising technique

Authors:
David Roy Meldrum, M.D., H. Irene Su, M.D., M.S.C.E., Mandy G. Katz-Jaffe, Ph.D., William B. Schoolcraft, M.D.

Abstract:

Reflections on “Effectiveness of in vitro fertilization with preimplantation genetic screening: a reanalysis of United States assisted reproductive technology data 2011–2012” by Kushnir et al.

  • Vitaly Kushnir

    N. Gleicher, D. Barad, V.A. Kushnir

    PGS: OVER 10 YEARS OF UNFULFILLED PROMISES

    We were somewhat surprised by the delayed commentary of Meldrum et al (1) regarding our manuscript (2) since Fertility & Sterility had not offered an editorial regarding an earlier manuscript from the Centers for Disease Control and Prevention (CDC) (3). That paper actually motivated our study because it, mistakenly, reported outcome benefits from preimplantation genetic screening (PGS) for in vitro fertilization (IVF).

    The editorial by Meldrum et al, interestingly, to considerable degree mirrors comments made in reference to our manuscript by other members of the PGS “establishment” (Grifo et al.), which we already answered (Kushnir et al., http://fertstertforum.com/kushnirv-ivf-pgs-effectiveness/). We, therefore, here do not want to be repetitive, and will direct the discussion to where the discourse on PGS in Fertility & Sterility really should be directed. Our Cornell colleagues recently asked an important question in that regard: PGS: who benefits? (4) We here hope to answer this question.

    Every scientist and/or physician will agree that before a diagnostic test is introduced to clinical practice, it has to be validated. This means we should know its diagnostic specificity and sensitivity, positive and negative predictive values, effects on desired outcomes and, potentially, also its cost effectiveness. We assume that even proponents of PGS, including Meldrum et al, and Grifo et al, now concur that the lack of any validation of PGS 1.0 prior to its clinical utilization represented a large misstep of IVF in its over 35-year history. Professional bodies, the American Society for Reproductive Medicine (ASRM) included (5), indeed, much too late declared PGS 1.0 ineffective in improving IVF outcomes and reducing miscarriage rates, therefore sharing part of the responsibility that PGS 1.0 went on for much too long.

    After such an embarrassment, and a for patients highly consequential failure in IVF practice worldwide, one would expect the scientific community to have shown some remorse, and attempt to determine why PGS 1.0 failed, when widely held opinions so strongly suggested that pre-transfer elimination of aneuploid embryos should lead to significant improvements in IVF outcomes? The PGS “establishment,” however neither demonstrated remorse nor did it even attempt to determine likely causes for PGS 1.0’s failure. With the same arrogance and self-righteousness that had led to the premature introduction of PGS 1.0 into clinical IVF, the PGS “establishment” now, instead, claimed, without even minimal research data in support, that the reasons for the failure of PGS 1.0 were only too obvious: Unquestionably, they were inadequate techniques (cleavage-stage embryo biopsy) and technologies (fluorescent in-situ hybridization of limited chromosome numbers) (6).

    Such absolute certainty, of course, also leads to indisputable conclusions. With PGS 1.0 finally disclosed as the failure it was (5), astonishingly, the PGS “establishment” demonstrated absolutely no hesitance in declaring PGS 2.0 effective where PGS 1.0 had failed, simply because it allegedly offered better techniques (trophectoderm biopsies) and technologies (genetic diagnostic platforms allowing full chromosomal complement analyses with clearly improved accuracy). The new PGS 2.0, consequently, was reintroduced to the market place even more aggressively, though incredulously, once again, without prior validation.

    As a consequence, in 2015 it has been estimated that as many as 20% of IVF cycles underwent PGS in the U.S. (7). Considering lack of validation, absence of clinical evidence that PGS 2.0 really offers outcome improvements with IVF and, most importantly, increasing evidence that the procedure in poor prognosis patients may actually negatively affect IVF outcomes, PGS should have attracted the attention of the Food and Drug Administration a long time ago.

    It, therefore, is rather astonishing that an official editorial of the official organ of the ASRM despite these facts, and despite a recently updated ASRM statement pointing out lack of supportive data (8), as the title of the editorial (“PGS 2.0: an evolving and promising technique”) well demonstrates, fully supports, indeed encourages the clinical utilization of PGS.2.0. What, however, really demonstrates the confusion in logic of authors and, likely, editors of Fertility & Sterility, is that the editorial, itself, concludes with the statement that: “…. data were not adequate to allow meaningful conclusions regarding the contemporary practice of PGS.”

    If data in support of PGS are not “adequate,” then why was PGS performed in such a high number of IVF cycles in the U.S. last year? More importantly, however, why does the commentary by Meldrum et al not conclude that, considering inadequate available data in support of PGS 2.0, the technique should be considered experimental and not routinely applied to clinical IVF?

    It is widely accepted in science that claimants of treatment effects face the burden of proof, and not their critics. In other words, it is up to the PGS “establishment,” which includes Meldrum et al, to establish the efficacy of PGS 2.0 before its clinical utilization in routine IVF. It is not our responsibility as critics of PGS 2.0 to prove the ineffectiveness of the procedure.

    Yet, there is increasing data for the ineffectiveness, indeed, harm caused by PGS 2.0 in certain patient populations. The PGS “establishment” controlling the editorial board and peer review process at Fertility & Sterility, however, has demonstrate little interest in presenting these data. The journal rejected reports from Israeli colleagues [since published elsewhere (9)] and from our laboratory, which have raised serious questions about the biological hypothesis behind PGS, offering convincing evidence that a single trophectoderm biopsy cannot reliably determine whether an embryo is euploid or aneuploid. Coincidentally, Bolton et al reported in an elegant mouse model of chromosomal mosaicism lineage-specific depletion of aneuploid cells in the cell lineage creating the inner cell mass and relative accumulation of aneuploidy in the placental lineage (i.e. trophectoderm), where biopsies are taken in PGS 2.0. The study further confirmed normal developmental potential for such mosaic embryos and concluded that these findings have clinical significance for PGS (10).

    Yet, when we submitted a paper to Fertility & Sterility reporting 5 healthy pregnancies in 8 attempts after transfer of allegedly aneuploid embryos (after PGS 2.0) in women who had no euploid embryos left for transfer, the manuscript was rejected. What stronger evidence is there that doubts about PGS 2.0 are in place than births of healthy children after transfer of allegedly aneuploid embryos, as reported by us (11) and Italian colleagues (12)?

    Under current editors and editorial board, Fertility & Sterility has persistently supported the viewpoint of the PGS “establishment,” and only rarely offered the opportunity of publication to opposing viewpoints. The self-invited editorial by Meldrum et al further makes our point (Meldrum is an editor of the journal). Another senior author of this editorial is also a strong proponent of PGS. Representing this view recently in a public panel discussion on the benefits of PGS 2.0 (The Best of ESHRE & ASRM, New York City, March 5-7, 2015), he is the principal owner of a prominent IVF center that performs PGS almost routinely. As pointed out by his opponent on the panel in politically irreverent fashion, such universal PGS utilization increases the annual revenue of the center to the tune of millions of dollars. If data on PGS are “not adequate,” as stated in the editorial, then why does this senior author of the editorial use PGS in so many of his center’s patients?

    It is not our intent to call into question any of our colleagues’ characters; but it does appear time to point out the contradictions in the positions taken by our colleagues in the PGS “establishment,” whatever their motivations may be. Moreover, we question why the leadership of ASRM permits even suspicions of conflicts of interest to permeate editorial decision making about acceptance and rejection of manuscripts for Fertility & Sterility.

    Such issues never before existed but have become commonplace in recent years not only in association with PGS but also in other areas of scientific discourse. Another area has been the utilization of closed incubation systems with time-lapse imaging. Most scientific journals have significantly tightened conflict of interest rules for journal editors in recent years. Especially self-publication of original scientific articles by associates of editors is increasingly frowned upon. Quantity and quality of accepted manuscripts from editors’ collaborators at Fertility & Sterility, have respectively reached unprecedented highs and lows in recent years, suggesting no such restrictions at the journal.

    In a recent commentary in Nature, Adam G. Dunn pointed out that conflicts of interest in research undertaken or funded by industry are rather easily measured if properly disclosed; yet ideology, religion, politics or personal relationships, all much more subtle potentially competing interests, blanket virtually every field of research and are clearly linked to biases. Such biases are then hidden in study designs, selective reporting of outcomes and conclusions that do not match results of studies (13). One would expect a professional society like ASRM to set standards for its official journal that avoid such compering interests.

    References

    1. Meldrum DR, Su HI, Katz-Jaffe MG, Schoolcraft WB, PGS 2.0: an evolving and promising technique, Fertil Steril 2016; in press;

    2. Kushnir VA, Darmon, S.K., Albertini, D.F., Barad, D.H., Gleicher N. Effectiveness of in vitro fertilization with preimplantation genetic screening: a reanalysis of United States assisted reproductive technology data 2011–2012. Fertil Steril. 2016 Mar 4. pii: S0015-0282(16)00140-0. doi: 10.1016/j.fertnstert.2016.02.026. [Epub ahead of print]

    3. Chang J, Boulet SL, Jeng G, Flowers L, Kissin DM. Outcomes of in vitro fertilization with preimplantation genetic diagnosis: an analysis of the United States Assisted Reproductive Technology Surveillance Data, 2011-2012. Fertil Steril 2016;105:394-400.

    4. Kang HJ, Melnick AP, Stewart JD, Xu K, Rosenwaks Z. Preimplantation genetic screening: who benefits? Fertil Steril. 2016 Apr 30. pii: S0015-0282(16)61124-X. doi: 10.1016/j.fertnstert.2016.04.027. [Epub ahead of print]

    5. Preimplantation genetic testing: a Practice Committee opinion. Fertil Steril. 2008 Nov;90(5 Suppl):S136-43.

    6. Munne, S., Wells, D., and Cohen, J. Technology requirements for preimplantation genetic diagnosis to improve assisted reproduction outcomes. Fertil. Steril. 2010; 94: 408–430

    7. Spath K, Wells D. Deep impact: sequencing embryo biopsy specimens at increasing depth. Reprod Biomed Online. 2015 Jul;31(1):1-3.

    8. ASRM Practice Committee Brief Communication on Pre-Implantation Genetic Screening for Aneuploidy: a Committee Opinion. Available for review by ASRM members at: http://www.asrm.org/proposed/ Accessed March 15, 2016

    9. Orvieto R, Shuly Y, Brengauz M, Feldman B. Should pre-implantation genetic screening be implemented to routine clinical practice? Gynecol Endocrinol. 2016 Feb 12:1-3.

    10. Bolton H, Graham SJ, Van der Aa N, Kumar P, Theunis K, Fernandez Gallardo E, Voet T, Zernicka-Goetz M. Mouse model of chromosome mosaicism reveals lineage-specific depletion of aneuploid cells and normal developmental potential. Nat Commun. 2016 Mar 29;7:11165.

    11. Gleicher N, Vidali A, Braverman J, Kushnir VA, Albertini DF, Barad DH. Further evidence against use of PGS in poor prognosis patients: report of normal births after transfer of embryos reported as aneuploid. Fertil Steril. 2015 ;104:e59.

    12. Greco E, Minasi MG, Fiorentino F. Healthy babies after intrauterine transfer of mosaic aneuploid blastocysts. N Engl J Med. 2015 Nov 19;373(21):2089-90.

    13. Dunn AG. Set up a public registry of competing interests. Nature 2016;533:9

    • David Meldrum

      In our final paragraph we were specifically summarizing the focus of our paper: the “analyzed (1) and reanalyzed (2) CDC data, for the reasons outlined, were not adequate to allow meaningful conclusions regarding the contemporary practice of PGS”. We pointed out the difficulty in using registry data to answer such questions because defining a valid control group is extremely difficult. Therefore the best data we have are from RCTs and cohort studies. Six had ongoing/delivered results per cycle that can be gleaned from their papers showing higher outcomes consistent with elimination of aneuploid embryos from transfer and no apparent adverse effect of biopsy. As pointed out, cumulative results must eventually be published. Elimination of aneuploid embryos from transfer would not be expected to increase the final delivery rate after transfer of all embryos unless reduced stress enhances implantation and reduces drop-outs, but would avoid the emotional toll and costs of futile transfers and miscarriages due to aneuploidy. It has also been shown that a vanishing twin can increase perinatal complications, so an aneuploid embryo transferred along with a euploid embryo
      can introduce significant risk (3). Nothing in my reflections piece
      would justify labeling me as a “proponent” or part of the “PGS establishment”. The piece did provide a critique and perspective on the calculated and recalculated CDC data and on other more pertinent studies, and advantages of PGS were stated. A comprehensive review of all papers published on PGS was not
      the goal nor was it possible within the “reflections” format.

      It appears that this commenter did not carefully read our “reflections” piece. “Evolving” means not yet fully fleshed out. “Promising” means it shows promise. Neither word suggested, nor did the article suggest, that PGS 2.0 should be widely adopted. In fact we made a major point of the fact that published trials to
      date have been mainly on good prognosis patients and in well-established laboratories. However, good prognosis patients are at the highest risk of twins with multiple embryos transferred. Most couples will accept elective single embryo transfer when fully informed of the risks of multiple pregnancies and reassured that a single euploid embryo will give a high chance of pregnancy in an experienced program with a good track record. In all programs a substantial proportion of couples have a normal prognosis. In IVF, transferring multiple embryos has been used for decades and is still used today as a way to achieve reasonable pregnancy rates in spite of low implantation rates, but the increased perinatal mortality, increased complications and stresses of twins on the couple’s relationship should be avoided whenever possible (4). Between the lines we were clearly
      suggesting that programs should use their implantation rate as a rough guide as to how their entire IVF process is functioning, and for individual programs, as a stimulus for improvement. We also implied that programs should not be attempting PGS 2.0 until their IVF program is functioning at a high level, including assuring
      high quality extended culture and adopting optimal biopsy and vitrification techniques. We gave two prominent examples of requirements for optimizing the production of high quality blastocysts – use of reduced oxygen and minimizing interruptions
      of the culture environment (5).

      I agree that the CDC paper could have stimulated an editorial and your article appropriately pointed out that results must be reported per cycle. It was fortunate that your recalculations made it evident that an unacceptably large proportion of cycles
      were lost between retrieval and transfer. Both papers nicely illustrated the importance of considering selection bias and performing intention to treat analysis, because lost data can bias the outcome toward or away from the null hypothesis. With such a large number of lost cycles, the data are simply not adequate for any meaningful conclusions to be made. It is unfortunate that the reviewers of both papers appear to have missed these two crucial deficiencies in the data being analyzed. If as the commenter suggests, the editorial board promotes the “PGS establishment”, the Kushnir et al paper should have been
      rejected. The commenter does not seem to be familiar with the review process, which is simply dependent on the two reviewers assigned by associate editors to each paper. We have all had papers rejected by the journal for reasons we could rightly dispute but the review process depends entirely on the quality of those chosen reviewers and is imperfect for most if not all journals.

      No one would dispute that PGS 1.0 was introduced prematurely, and clearly comprehensive chromosome screening is a huge step forward compared to FISH, which analyzed a limited number of chromosomes and was highly subject to artifacts. I have also suggested that defining a large number of intermediate
      reads using NextGen may be premature, in part because an increasing number of mosaic embryos are being reported as resulting in normal births (4), as these authors have also
      indicated. There were many failures before the Wright brothers’ successful flight and famous aeronautical experts expressed certainty that human controlled flight was impossible (6). Failed techniques of the past fortunately do not close most minds to future success.

      PGS 2.0 has been validated (largely by Richard Scott’s group) to a higher level of evidence than probably any other technique widely used in the treatment of infertility (7) (8) (9) (10) (11), and in an elegant study, they showed day 3 biopsy to be harmful (12). An adverse effect of trophectoderm biopsy was not evident (12), but Scott’s program reported the very highest implantation rate in women under 35 for 2012 aside from outliers reporting extremely small numbers. All 6 programs whose data I collated were
      reported per cycle. Those, including Scott’s program, used low oxygen culture during the genesis of their data, in most cases confirmed to include pre-equilibration of the media under reduced oxygen prior to zygote culture (5). That does make it difficult to generalize these data to the wide range of settings where IVF is performed in the U.S. and around the world, which we emphasized. To my recollection, none of those papers proposed wide introduction to “the marketplace” but they were
      important contributions to study of a potential new treatment. We also emphasized that data assessing use of the technique in women with lower prognosis is still forthcoming.

      Other comments made are far beyond the “reflections” piece under discussion. As with any evolving change in practice, each center must evaluate the published literature, their individual capabilities, track record and accumulating results, so as to offer what they consider to be the best outcomes for their patients. Discussions such as these will contribute to progress and certainly should be devoid of personal attacks. It is worth pointing out that one of the authors impugned by this commentary was routinely using ultrasound guided transfer for a decade before it was eventually widely adopted, and nurtured David Gardner’s research at a time when few people were taking notice of his
      early work on extended embryo culture. Just as with the Wright brothers, it is fortunate we have individuals such as Richard and David and his mentor in our midst.

      1. Chang J, Boulet SL, Jeng G, Flowers L, Kissin DM. Outcomes of in vitro fertilization with preimplantation genetic diagnosis: an analysis of the United States Assisted Reproductive Technology
      Surveillance Data, 2011-2012. Fertility and Sterility 2016;105:394-400.

      2. Kushnir VA, Darmon, S.K., Albertini, D.F., Barad, D.H., Gleicher N. . Effectiveness of in vitro fertilization with preimplantation genetic screening: a reanalysis of United States assisted reproductive technology data 2011–2012. Fertility and Sterility 2016;106.

      3. Evron E, Sheiner E, Friger M, Sergienko R, Harlev A. Vanishing twin syndrome: is it associated with adverse perinatal outcome? Fertility and Sterility 2015;103:1209-14.

      4. Meldrum DR. Lightening the burden of care in assisted reproductive technology. Fertility and Sterility 2016;105:1144-5.

      5. Gardner DK, Lane M. Alleviation of the ‘2-cell block’ and development to the blastocyst of CF1 mouse embryos: role of amino acids, EDTA and physical parameters. Human
      reproduction 1996;11:2703-12.

      6. McCullough D. The Wright Brothers. Simon & Schuster, New York, NY 2015.

      7. Scott RT, Jr., Ferry K, Su J, Tao X, Scott K, Treff NR. Comprehensive chromosome screening is highly predictive of the reproductive potential of human embryos: a prospective, blinded, nonselection study. Fertility and Sterility 2012;97:870-5.

      8. Forman EJ, Hong KH, Ferry KM, Tao X, Taylor D, Levy B et
      al. In vitro fertilization with single euploid blastocyst transfer: a
      randomized controlled trial. Fertility and Sterility 2013;100:100-7.

      9. Forman EJ, Tao X, Ferry KM, Taylor D, Treff NR, Scott RT, Jr. Single embryo transfer with comprehensive chromosome screening results in improved ongoing pregnancy rates and decreased miscarriage rates. Human reproduction 2012;27:1217-22.

      10. Scott RT, Jr.,Upham KM, Forman EJ, Hong KH, Scott KL, Taylor D et al. Blastocyst biopsy with comprehensive chromosome screening and fresh embryo transfer significantly increases in vitro fertilization implantation and delivery rates: a randomized controlled trial. Fertility and Sterility 2013;100:697-703.

      11. Werner MD, Leondires MP, Schoolcraft WB, Miller BT, Copperman AB, Robins ED et al. Clinically recognizable error
      rate after the transfer of comprehensive chromosomal screened euploid embryos is low. Fertility and Sterility 2014;102:1613-8.

      12. Scott RT, Jr., Upham KM, Forman EJ, Zhao T, Treff NR.
      Cleavage-stage biopsy significantly impairs human embryonic implantation potential while blastocyst biopsy does not: a randomized and paired clinical trial. Fertility and Sterility 2013;100:624-30.

      • Norbert Gleicher, MD

        Response to David Meldrum, MD

        ___________
        Vitaly A. Kushnir,MD (1,2), David H. Barad, MD, MS (1,3)
        and Norbert Gleicher, MD (1,3,4,5)
        1) The Center for Human Reproduction, New York, NY;
        2) Department of Obstetrics and Gynecology, Wake Forest School of Medicine, Winston-Salem, NC;
        3) Foundation for Reproductive Medicine, New York,
        NY;
        4) The Rockefeller University, New York, NY;
        5) Department of Obstetrics and Gynecology, Medical University, Vienna, Austria
        ____________

        Though we greatly appreciate Dr. David Meldrum’s detailed effort to clarify his and his co-authors’ editorial, we have to respectfully disagree with many of his comments. All, of course, starts with the state of mind behind the editorial process. Dr. Meldrum suggests in his response that nothing in his “Reflection” piece would justify the label of a “proponent” or of being part of the “PGS establishment.” Yet, three of his recent publications in Fertility & Sterility strongly argue otherwise (1-3). More importantly, however, state of mind of a journal editor is also reflected by which articles are chosen for an editorial commentary in the journal (especially if written by the editor himself). In this instance, the choice was not made after colleagues from the Centers for Disease Control and Prevention (CDC) published a study supportive of favorable outcome effects of PGS (4), which after reanalysis of their data led to our publication, clearly refuting their conclusions (5).

        Dr. Meldrum suggests that publication of our manuscript was the best evidence against any editorial pro-PGS bias by Fertility & Sterility. We, indeed, were pleasantly surprised that our study was accepted for publication because a shorter uninvited commentary on the CDC publication had been outright rejected without review. Dr. Meldrum, however, does not explain why, if this subject was deserving of an editorial by him and associates, such an editorial was not published concomitantly with our manuscript as is usually the norm, but only after other proponents of PGS, criticized our publication in the pages of this blog [Grifo et al].

        We greatly appreciate that in his response (though not in the original editorial) Dr. Meldrum now states that “it was fortunate that our recalculations made it evident that an unacceptable large proportion of cycles were lost between retrieval and transfer.” The editorial, however, not only did not reflect such appreciation but also lacked self-criticism about the journal’s peer review process, having allowed an obviously flawed manuscript (4) to appear in print.

        The suggestion that we are not familiar with the peer review process at Fertility & Sterility is somewhat condescending, – but also misleading, considering our group’s decades long senior editorial journal experience and publication history in Fertility & Sterility and innumerable other medial and basic science journals. To claim that any review process “is simply dependent on the two reviewers assigned by associate editors to each paper” is also misleading in describing real editorial control at every journal. We highlighted a recently published editorial in Nature, which deftly describes common editorial biases in scientific publishing (6). Editorial boards are increasingly recognizing these weaknesses of current peer review, and are putting controls into place to minimize such biases as much as possible.

        That selection of associate editors and reviewers can severely bias the peer review process has been known for decades. Since experienced editors usually know the review patterns of their associate editors and reviewers well, they, through those selections alone, can determine a manuscript’s fate. The claim that this fate exclusively depends on two unbiased reviewers, therefore, is a deflection from the real responsible individuals within the editorial process, which are those who assign reviewers and/or authors of editorials. Of course self-assignements are even more subject to potential biases, as Dr. Meldrum’s publication history in regard to PGS demonstrates (1-3).

        Bias in regard to PGS is further demonstrated by the opinion that “PGS 2.0 has been validated (largely by Richard Scott’s group) to a higher level of evidence than probably any other technique widely used in the treatment of infertility.” One can really not imagine a more astonishing statement from a senior editor of Fertility & Sterility!

        We have previously in a number of publications, and in our response to Grifo et al in this blog (Kushnir et al., http://fertstertforum.com/kushnirv-ivf-pgs-effectiveness/), addressed the shortcomings of all attempted validation studies of PGS 2.0 in the literature, including those by Richard Scott’s group (8,9). Though claiming to be grateful that we did point out in our recent publication the effects of patient selection in the CDC study, Dr. Meldrum, unfortunately, for a second time overlooks what favorable patient selection does to outcome reporting in IVF (10) by claiming “very highest implantation rates” achieved by Scott’s group utilizing PGS 2.0,. It seems curious that he, at once, acknowledges to previously have overlooked this in the CDC study; yet, does it all over again when commenting on the publications of Scott’s group. These statements, very obviously, do not reflect the statistical realities of PGS 2.0.

        How biased the analysis of PGS 2.0 is can also be deducted from the analogy of PGS to the pioneering flight by the Wright brothers. The Wright brothers risked their own lives; they did not sell tickets to passengers. In contrast, unvalidated PGS 2.0 is sold at considerable cost, under often incorrect informed consent (we have seen consents describing PGS 2.0 as a “routine” IVF procedure) to misinformed patients. A much more appropriate analogy would be that by another famous Wright (Frank Lloyd Wright, the architect): “The physician can bury his mistakes, but the architect can only advise his client to plant vines.”

        Dr. Meldrum suggests that all the editorial in principle wanted to say is that national SART registry data are not yet adequate to evaluate the efficacy of PGS in routine clinical practice. If that was, indeed, the principle purpose of the editorial, then its final conclusion would have been that PGS 2.0, until adequate outcome data exist for its validation, should only be performed under experimental conditions.

        Available IVF outcome data, while still far from perfect, increasingly appear to disprove the ability of PGS 2.0 to improve IVF outcomes. These include national IVF registry data in the publication pipeline, which not only demonstrate no benefits from PGS 2.0 on live birth and miscarriage rates after IVF but actually demonstrate unfavorable outcome effects.

        Once published, these studies can be perceived as what the National Institutes of Health (NIH) define as Phase IV, post-marketing studies, meant to gather information on treatment effects in various populations and side effects associated with … use (7). There are countless examples of treatments which, following clinical trials, gained FDA approval, only to be later removed from the market following unanticipated side effects and/or lack of efficacy in routine clinical practice documented in national registries.

        In summary, we greatly appreciate Dr. Meldrum’s contributions to IVF over many decades and his editorial dedication to Fertility & Sterility. We are convinced that his ultimate goal and motivation are continuous improvements in IVF outcomes. We all, however, at times need reminders that things are not always what they appear to be on first glance, and that what looks logical and promising in theory, does not always work out in clinical practice.

        We previously referred to a recent publication by Bolton et al, which in a mouse model elegantly demonstrated why single trophectoderm biopsies for biological reasons are incapable of reliably defining embryos as euploid or aneuploid (11). That this also appears to apply to human embryos is strongly suggested by recent healthy births after transfers of allegedly aneuploid embryos by our group (12) and Italian colleagues (13). Other publications, showing ineffectiveness of PGS 2.0 are in the publication pipeline.

        We are not claiming the absolute truth for our position on PGS 2.0. All we are striving for is fairness of opportunity at Fertility & Sterility for scientifically well supported differences of opinion, and an unbiased editorial process. We hope that both goals are shared by the editorial board of Fertility & Sterility as well as the leadership of ASRM.

        ___________
        References
        1. Meldrum DR, Su HI, Katz-Jaffe MG, Schoolcraft WB. Preimplantation genetic screening 2.0: an evolving and promising technique. Fertil Steril. 2016 May 24. pii: S0015-0282(16)61231-1
        2. Meldrum DR. Introduction: Preimplantation genetic screening is alive and very well. Fertil Steril. 2013 Sep;100(3):593-4
        3. Meldrum DR, Scott RT Jr, Schoolcraft WB. Reply of the authors. Fertil Steril. 2013 Nov;100(5):e37-8.

        4. Chang J, Boulet SL, Jeng G, Flowers L, Kissin DM. Outcomes of in vitro fertilization with preimplantation genetic diagnosis: an analysis of the United States Assisted Reproductive Technology Surveillance Data, 2011-2012. Fertil Steril 2016;105:394-400.
        5. Kushnir VA, Darmon, S.K., Albertini, D.F., Barad, D.H., Gleicher N. Effectiveness of in vitro fertilization with preimplantation genetic screening: a reanalysis of United States assisted reproductive technology data 2011–2012. Fertil Steril. 2016 Mar 4. pii: S0015-0282(16)00140-0. doi: 10.1016/j.fertnstert.2016.02.026. [Epub ahead of print]
        6. Dunn AG. Set up a public registry of competing interests. Nature 2016;533:9
        7. Available at: https://www.nlm.nih.gov/services/ctphases.html)
        8. Gleicher N, Kushnir VA, Barad DH. Preimplantation genetic screening (PGS) still in search of a clinical application: a systematic review. Reprod Biol Endocrinol. 2014 Mar 15;12:22.
        9. 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 Nov;29(11):1159-66.
        10. Gleicher N, Kushnir VA, Barad DH. The impact of patient preselection on reported IVF outcomes. J Assist Reprod Genet. 2016 Apr;33(4):455-9
        11. Bolton H, Graham SJL, Van der Aa N, Kumar P, Theunis K, Gallardo EF, Voet T, Zernicka-Goetz M. Mouse model of chromosome mosaicism reveals lineage-specific depletion of aneuploid cells and normal developmental potential. Nature Communications 2016;7:11165
        12. Gleicher N, Vidali A, Braverman J, Kushnir VA, Albertini DF, Barad DH. Further evidence against use of PGS in poor prognosis patients: report of normal births after transfer of embryos reported as aneuploid. Fertil Steril. 2015 ;104:e59.
        13. Greco E, Minasi MG, Fiorentino F. Healthy babies after intrauterine transfer of mosaic aneuploid blastocysts. N Engl J Med. 2015 Nov 19;373:2089-90.

        • David Meldrum

          Dr. Gleicher is obviously correct that it is difficult, if
          not impossible, to eliminate bias from the process of choosing reviewers and of the reviewers themselves. I recall one of my papers was rejected by this journal within the past few years due to an obvious and unsupported bias on the
          part of one of the reviewers. The paper was improved by emphasizing studies showing the reviewer’s bias was incorrect and it has been published in another quality journal. We must strive to limit these biases but it would be impossible for the editors to vet associate editors regarding such a wide field and likewise impossible for the associate editors to be familiar with every reviewer’s potential biases on papers to be assigned.

          I do readily admit that I find new technical advancements exciting. The concept of reducing miscarriage and futile transfers and their stress on our patients is alluring. Likewise the ability to help couples to accept single embryo transfer is an extremely important benefit of PGS. Virtually all IVF practitioners agree that the most important complication of IVF is multiple pregnancies. We hope to see the day when the very large majority of IVF couples accept transfer of a single embryo and the major morbidity, increased perinatal mortality, and stresses resulting from multiple pregnancies are reduced to an absolute minimum (1). If in some eyes those goals make me seem a proponent of PGS, I am happy to be considered as such.

          As I previously indicated, the CDC report could have instigated editorial comments, but the loss of cycles was not evident. Your paper did a service by making further calculations and pointing out the importance of presenting outcomes per cycle. Unfortunately, both papers’ conclusions should be given very little weight (2) because the loss of so many cycles could have biased results away from (or toward) favoring PGS. The best available data are the RCTs and cohort studies which allowed collating results on a per cycle basis. The increases in ongoing/birth rates were consistent with what would be expected by excluding aneuploid embryos from transfer and showed
          no evident harm of the biopsy in those predominantly good prognosis patients.

          My “reflections” (2) was motivated by seeing your
          Epub and was timed to be in the same issue as your paper. In spite of publication of papers in press, discussions of papers draw the most attention when published in the same issue of the journal. These four paragraphs and most of the comments regarding my “reflections” paper nicely illustrate how the discussion can be so easily diverted from the real issues needing to be discussed.

          The most important issue is the variable quality of
          blastocyst culture represented by the full range of programs submitting data to the registry and to journals for publication. There are many factors that can markedly stress an embryo. As just one example, David Gardner has amply shown the marked effects of 20% oxygen on blastocyst development using a murine model. The percentage of blast development and number of cells per blastocyst were doubled by using a physiologic concentration of oxygen and approached those observed in vivo (3). This year David has reviewed the extensive data from various animal species showing the damaging effects of exposing embryos to four-fold the concentrations normally surrounding the embryo in vivo (4). During the period when the data under discussion were being accumulated only approximately one quarter of IVF programs were using 5% O2 (5). As discussed in my previous
          comments, that undoubtedly contributed to the difference between registry data and the RCT’s and cohort studies (of which upon contacting the authors, all were confirmed to have been carried out using 5% O2). Reviewers should insist that manuscripts specifically state the concentration of O2 used during the gathering of PGS data and that incubator types are controlled for, which is
          unfortunately not the case in at least one study “in the pipeline”.

          The issue of mosaicism in the trophectoderm is extremely important. As the commenters have pointed out, the biopsy can only be highly representative if the incidence of mosaicism is low. The fascinating study referred to by Bolton (6) used a model that artificially created mosaic murine embryos. They elegantly demonstrated that with further growth the Inner Cell Mass (ICM) becomes progressively more normal through apoptosis of aneuploid cells, whereas the trophectoderm becomes progressively more normal through a severe retardation of replication of aneuploid cells. Their study provided no evidence of actual rates of mosaicism, but only referred to other publications. They did state their findings were “Potentially an explanation for why cleavage stage embryo biopsy was found to be detrimental to IVF success rates in couples with a poor prognosis for IVF” and that “it also supports the current move from biopsy at the cleavage stage towards blastocyst biopsy”.

          In a meta-analysis of published studies on blastocyst mosaicism (7), most were by analyzing embryos previously stressed by day 3 biopsy, most used spare embryos not chosen for fresh transfer or cryopreservation, and 88% of the programs used FISH for analysis. In one report the incidence of mosaicism in cleavage embryos with FISH was 100% compared to 31% using CCS by SNP microarray (8).

          Most of the publications on mosaicism have used a room air concentration of O2. In a model of non-disjunction, use of 20% oxygen increased
          chromosome segregation errors and use of 5% reduced the rate to in vivo levels (9). In one study using 5% O2, culturing 55 embryos found to be aneuploid on day 3 using FISH, the resulting
          ICM showed no mosaicism, in spite of using a liberal definition of what constituted a mosaic embryo (10). My co-authors in Denver use 5% O2 throughout the culture and have reported high implantation rates. They have recently analyzed donated blastocysts and found the rate of mosaicism was less than 4% (look for an ASRM presentation this fall). These data indicate that
          embryo stress may be a cause of mitotic errors and mosaicism and such an effect is certainly logical. Programs with a low implantation rate and those not using 5% O2 should focus their attention on removing embryo stress from their
          system. Such programs could have an embryo that would have been largely or entirely euploid (and therefore designated as euploid) but due to that stress and resulting mosaicism of the trophectoderm, the embryo might then be reported as mosaic or aneuploid (depending on the bioinformatics used). It is not surprising given the variable quality of embryo culture that examples of false positive embryo designation have been reported. In the quoted abstract by
          Gleicher there are insufficient details regarding the method of PGS, timing of the biopsy or whether reduced oxygen was used during the during the entire culture. In the study by Greco (whose lab has used 5% O2), their mosaic embryos resulting in 6 normal births are consistent with the discussion above and support the transfer of such embryos with careful informed consent and followup prenatal testing. The only sizeable randomized study of delivery rate using reduced oxygen exposure throughout the culture found a 30% higher delivery rate with physiologic oxygen levels (11).

          The commenters criticize the work of Scott and Treff, presumably because their conclusions markedly differ regarding the benefits of PGS. Hopefully the discussion above illustrates that there are cogent reasons why their views differ that have nothing to do with the quality of their
          studies. Having been on the editorial board for many years I can say that I have observed very few groups whose quality of publications I admire as much.

          Incidentally, although the Wright brothers did not sell tickets for accompanying them on flights, they did take many passengers along, obviously for the purpose of popularizing their invention and with the result of maximizing eventual profits from their work. Their success illustrated the importance of not simply accepting that flawed technology demonstrated that failure was inevitable. The most important reason they succeeded was that they recognized prior aerodynamic principles might not be correct. They consequently redefined how aerodynamics influence controlled flight and the rest is history.
          McCullough’s book (12) is a great read for anyone interested in technological advances.

          1. Meldrum DR. Lightening the burden of care in assisted reproductive technology. Fertility and Sterility 2016;105:1144-5.

          2. Meldrum DR. Preimplantation genetic screening 2.0: an evolving and promising technique. Fertility and Sterility 2016.

          3. Gardner DK, Lane M. Alleviation of the ‘2-cell block’ and development to the blastocyst of CF1 mouse embryos: role of amino acids, EDTA and physical parameters. Human reproduction 1996;11:2703-12.

          4. Wale PL, Gardner DK. The effects of chemical and physical factors on mammalian embryo culture and their importance for the practice of assisted human reproduction. Hum Reprod Update 2016;22:2-22.

          5. Christianson MS, Zhao Y, Shoham G, Granot I, Safran A, Khafagy A et al. Embryo catheter loading and embryo culture techniques: results of a worldwide Web-based survey. J Assist Reprod Genet 2014;31:1029-36.

          6. Bolton H, Graham SJ, Van der Aa N, Kumar P, Theunis K, Fernandez Gallardo E et al. Mouse
          model of chromosome mosaicism reveals lineage-specific depletion of aneuploid cells and normal developmental potential. Nat Commun 2016;7:11165.

          7. van Echten-Arends J, Mastenbroek S, Sikkema-Raddatz B, Korevaar JC, Heineman MJ,
          van der Veen F et al. Chromosomal mosaicism in human preimplantation embryos: a systematic review. Hum Reprod Update 2011;17:620-7.

          8. Treff NR, Levy B, Su J, Northrop LE, Tao X, Scott RT, Jr. SNP microarray-based 24 chromosome aneuploidy screening is significantly more consistent than FISH. Mol Hum Reprod 2010;16:583-9.

          9. Bean CJ, Hassold TJ, Judis L, Hunt PA. Fertilization in vitro increases non-disjunction
          during early cleavage divisions in a mouse model system. Human reproduction 2002;17:2362-7.

          10. Li M, DeUgarte CM, Surrey M, Danzer H, DeCherney A, Hill DL. Fluorescence in situ
          hybridization reanalysis of day-6 human blastocysts diagnosed with aneuploidy on day 3. Fertility and Sterility 2005;84:1395-400.

          11. Waldenstrom U, Engstrom AB, Hellberg D, Nilsson S. Low-oxygen compared with high-oxygen atmosphere in blastocyst culture, a prospective randomized study. Fertility and Sterility 2009;91:2461-5.

          12. McCullough D. The Wright Brothers. Simon & Schuster, New York, NY 2015.

          • Norbert Gleicher, MD

            We appreciate Dr. Meldrum’s detailed latest comment but fail to understand the relevance of oxygen concentrations in IVF incubation to the very basic question whether PGS 2.0 is a diagnostic procedure we, as scientists, should recommend to colleagues and, as physicians, to our patients. We, too, get exited by advances in our field, and we, too, admire Richard Scott and co-workers for much of their work; but advances need to be validated before their introduction into routine medical care, and PGS 2.0, like PGS 1.0 before, did not undergo proper validation. As much as we appreciate much of the work of Scott’s group, like the original CDC paper that has been subject to much discussion in these blog exchanges, Scott et al in our opinion are incorrect in the interpretation of their own data when it comes to PGS.

            It appears time to stop looking for technical excuses why PGS does not work. It is not the technique that does not work. As we in an abstract submitted to ASRM 2016 demonstrate, PGS 2.0 is technically, indeed, a
            great improvement over PGS 1.0; but it is not enough that PGS 2.0 reduces IVF live birth rates less than PGS 1.0 did. The original purpose of PGS was to improve IVF pregnancy and live birth rates, – not to diminish them, and as we quite convincingly demonstrate in a series of 3 abstracts submitted to ASRM 2016, national U.S. data even in best prognosis patients demonstrate significant declines in successful IVF outcomes following PGS when compared to cycles that did not have the misfortune of undergoing PGS.

            And the principal reason for these findings is obvious; PGS does not work as a biological concept because mosaicism in the trophectoderm is practically a physiological constant in blastocyst stage embryos, as aneuploid cell lineages disproportionally accumulate in trophectoderm. This is, however, where PG 2.0 biopsies are derived from. Biopsy of an aneuploid cell lineage, therefore, frequently will not represent the inner cell mass, resulting in false positive results, and the discarding of perfectly normal embryos.

            Since Greco et al performed only a single trophectoderm biopsy involving on the average 6 cells, the prevalence of such mosaicism is not, as Dr. Meldrum (based on the report by Greco et al) suggests ca. 3%. Assuming a 60-cell trophectoderm, the chance would be ca. 30%, with a more realistic 120-cell count, it would be 60% and with a trophectoderm encompassing 180 cells, Greco’s data actually suggest that mosaicism may be as high as 90%. Even assuming a trophectoderm mosaicism rate of only 50%, a coin toss would beat the predictability of a single trophectoderm biopsy, as currently the norm in PGS 2.0. Is this really the level of accuracy we want to rely on? We don’t think so and, therefore, feel obliged to speak up, and warn colleagues and patients alike.

            • David Meldrum

              We can agree to disagree regarding Scott’s group’s validation of PGS 2.0

              I will try to more clearly explain the relevance of culture under physiologic levels of oxygen, which is that culture conditions may modify the efficacy of PGS. In my “reflections” paper I quoted data that anyone can verify that for 2012 (therefore
              representing the time during which the CDC PGS data were accumulated), showing that implantation rates (IRs) in patients under age 35 varied from 2.6 to 64% (mean 37.0) and the mean of the upper quartile (49%) was over two-fold that of the lower quartile (24) (1). While there are many factors that can influence IR, the quality of an embryo is obviously very prominent in determining its chance of implanting. A room air concentration of oxygen is highly stressful for the embryo. For mouse embryos the proportion of embryos reaching the blastocyst and the number of cells per blastocyst were reduced by half (2), and the damaging effects of 20% oxygen have been extensively documented in a wide range of animal models (3). In the only sizable RCT in the
              human comparing use of 5% to 20% O2 from insemination throughout culture to blastocyst transfer, the delivery rate was increased by 30% (4). Clearly having more and healthier blastocysts
              to biopsy and more cells per blastocyst would be expected to markedly influence outcomes with trophectoderm biopsy. Prior to 2014, only about one quarter of U.S. and European programs were using 5% O2 for all patients (5). Exposing the embryos to 4 times as much of this highly reactive compound than levels surrounding the embryo in
              vivo has been used to assess the effect of embryo stress on chromosome maldistribution; 20% O2 increased chromosome non-disjunction and 5% reduced it to the level of in vivo blastocysts (6). The earliest stages of zygote
              development are the most sensitive to 20% O2 (2), when mitotic errors in the initial cell divisions would have their greatest impact on mosaicism of the fully developed blastocyst. The variable quality of embryo culture reflected by the wide range of implantation rates and the high incidence of use of 20% O2 for PGS data submitted to CDC and SART and for publications could readily explain the variable enthusiasm for PGS 2.0. If your 2016 abstracts are based on national data those will also fail to show the performance of PGS 2.0 with optimal blastocyst culture techniques. Registry data also suffer from the almost impossible task of identifying a valid control group. As previously emphasized, the only valid data we have that are analyzed per cycle are from published RCT and cohort studies in good prognosis patients and using 5% oxygen. The higher ongoing/delivery rates reported are consistent with eliminating aneuploid embryos from transfer and show no evident harmful effect
              of biopsy (1).

              Again as previously emphasized, most studies of blastocyst mosaicism have been with culture following cleavage stage biopsy, followed by culture in 20 % O2, and analyzed by FISH, which
              vastly overestimates the incidence of mosaicism (7). In a recent publication, culture under 5% O2 resulted in blastocysts with only a 6.3% incidence of “inconclusive” results that could have been called mosaic but were not, because artifacts of
              whole genome amplification could have been partially responsible (8). “At risk for mosaicism” has been suggested as a preferable label for these intermediate reads (9). In another study using 5% O2 culture of 55 aneuploid embryos following cleavage stage biopsy there was no mosaicism noted of the inner cell mass (10). If you are aware of studies of mosaicism in donated blastocysts cultured entirely with 5% O2 from insemination through to blastocyst transfer and not subjected to any known stress such as cleavage stage biopsy, please provide those references. As mentioned, my co-authors have
              submitted an abstract to ASRM showing a 3.8% incidence of mosaicism examining donated blastocysts cultured using 5% O2 while minimizing embryo stress.

              I believe your calculations of mosaicism based on the report by Greco are incorrect. Only 181 of their 3802 blastocysts (4.8%) were mosaic, as stated in their paper (11). In your example it appears that you are somehow using the 3% of all blastocysts figure divided by 6 (the average number of cells biopsied) = 0.5 x 60 cells per blastocyst (a reasonable number) to arrive at 30% mosaicism. I, my co-authors and others in the field I have consulted fail to see the logic behind such a calculation. If 4.8% of their blasts were mosaic, that means that 95.2% had no detectable mosaicism; 3% would indicate that 97% had no detectable mosaicism. The level of mosaicism in that small percentage of biopsied embryos would depend on the method (and bioinformatics) of the CCS technique used. As pointed out previously, robust embryo self-correction appears to occur due to apoptosis of aneuploid cells in the ICM and marked retardation of replication of aneuploid cells in the trophectoderm (12). The less than 2% incidence of mosaicism in CVS samples of almost 400 IVF pregnancies (and not different from infertile and normal controls), the majority of whom (Huang, personal communication) were from a program that has reported excellent implantation rates and has used 5% O2 (13), is consistent with a low rate of mosaicism in the blastocyst and further self-correction. In that study, follow-up of the pregnancies showed mosaicism was rarely found in fetal tissues; that
              same program found no mosaicism in the ICM after extended culture of aneuploid embryos using 5% O2 (10).

              I hope these exchanges and references have been of value to those readers who have struggled to make PGS 2.0 a fully realized asset for their infertile couples. For those IVF programs having a low implantation rate and/or using 20% O2, efforts should first be focused on removing sources of embryo stress. Fortunately today, new small desktop incubators not only use 5% O2 but also achieve rapid recovery of culture conditions, which removes another major source of embryo stress (2). It is a significant investment of time and expense to switch one’s entire lab to a new incubator system, as we did about 15 years ago, but well worth it. Although series of lower prognosis patients with PGS 2.0 and excellent results are beginning to appear (8), readers should keep in mind that studies to date have been primarily on good prognosis patients. Those couples, who are at greatest risk for multiple pregnancies with more than a single embryo transferred, should be their primary focus when initially utilizing this “promising” technique.

              I would respectfully suggest that each side of this exchange has had more than ample opportunity to present their views and further responses will not significantly contribute to the readers’ understanding of these issues. For this forum to be successful, authors must have a reasonable expectation that their time will be respected.

              1. Meldrum DR. Preimplantation genetic screening 2.0: an evolving and promising technique. Fertil Steril 2016.

              2. Gardner DK, Lane M. Alleviation of the ‘2-cell block’ and development to the blastocyst of CF1
              mouse embryos: role of amino acids, EDTA and physical parameters. Human Reproduction 1996;11:2703-12.

              3. Wale PL, Gardner DK. The effects of chemical and physical factors on mammalian embryo culture and their importance for the practice of assisted human reproduction. Hum Reprod Update 2016;22:2-22.

              4. Waldenstrom U, Engstrom AB, Hellberg D, Nilsson S. Low-oxygen compared with high-oxygen atmosphere in blastocyst culture, a prospective randomized study. Fertil Steril 2009;91:2461-5.

              5. Christianson MS,Zhao Y, Shoham G, Granot I, Safran A, Khafagy A et al. Embryo catheter loading and embryo culture techniques: results of a worldwide Web-based survey. J Assist Reprod Genet 2014;31:1029-36.

              6. Bean CJ, Hassold TJ, Judis L, Hunt PA. Fertilization in vitro increases non-disjunction during early cleavage divisions in a mouse model system. Human Reproduction 2002;17:2362-7.

              7. Treff NR, Levy B, Su J, Northrop LE, Tao X, Scott RT, Jr. SNP microarray-based 24 chromosome aneuploidy screening is significantly more consistent than FISH. Mol Hum Reprod 2010;16:583-9.

              8. Fiorentino F, Bono S, Biricik A, Nuccitelli A, Cotroneo E, Cottone G et al. Application of next-generation sequencing technology for
              comprehensive aneuploidy screening of blastocysts in clinical preimplantation genetic screening cycles. Human Reproduction 2014;29:2802-13.

              9. Scott RT, Jr., Galliano D. The challenge of embryonic mosaicism in preimplantation genetic
              screening. Fertil Steril 2016;105:1150-2.

              10. Li M, DeUgarte CM, Surrey M, Danzer H, DeCherney A, Hill DL. Fluorescence in situ hybridization reanalysis of day-6 human blastocysts diagnosed with aneuploidy on day 3.
              Fertil Steril 2005;84:1395-400.

              11. Greco E, Minasi MG, Fiorentino F. Healthy Babies after Intrauterine Transfer of Mosaic
              Aneuploid Blastocysts. New Eng J Med 2015;373:2089-90.

              12. Bolton H, Graham SJ, Van der Aa N, Kumar P, Theunis K, Fernandez Gallardo E et al. Mouse model of chromosome mosaicism reveals
              lineage-specific depletion of aneuploid cells and normal developmental potential. Nat Commun 2016;7:11165.

              13. Huang A, Adusumalli J, Patel S, Liem J, Williams J, 3rd, Pisarska MD. Prevalence of chromosomal mosaicism in pregnancies from couples with infertility. Fertil Steril 2009;91:2355-60.

Translate »