A response to Mastenbroek PGD study.
11 July 2007
On Friday, 6 July, 2007, the New England Journal of Medicine (NEJM) published a study by Mastenbroek and co-workers, describing the results of a collaborative randomized clinical trial of PGS (Preimplantation Genetic Screening) that was recently conducted in The Netherlands. The study purports that PGD for infertility or advanced maternal age (AMA) is not just ineffective but detrimental to pregnancy outcome. We would like to make you aware of several serious flaws in this work.
A primary concern is that PGD was performed by groups with little or no previous experience with or knowledge of this multifaceted and complex technology. As such, the study was designed without consideration of previously published crucial details in methodology that can determine success or failure of PGD. By Reprogenetics' standards, aspects of biopsy, fixation, patient selection and FISH as described in this Dutch study were not optimized for success but for inevitable failure.
In a significant number of cases (>50), the authors decided to transfer embryos that were biopsied but had no result or diagnosis; this is quite disconcerting not only because of the questionable medical decision, but because the test failed in 20% of all embryos biopsied – an extraordinarily high proportion compared to previously published data by others, including Reprogenetics (<5%). The authors point to "biopsy failure" having occurred in 2.8% and "incomplete nucleus" in 4.9% of embryos. Yet such categories are non-existent in experienced PGD laboratories such as Reprogenetics and only reflect poor biopsy and fixation techniques used by these inexperienced groups.
Once undiagnosed embryos were transferred, these embryos showed a poor implantation rate of 6%. This figure was significantly lower than the implantation rate for embryos that were diagnosed as 'normal' (16.8%) or embryos that were in the control arm of the study and were not biopsied (14.7%). This discrepancy demonstrates that the biopsy procedure as was performed at the participating clinics in the trial was detrimental to embryo viability and reduced the implantation potential of the embryos by as much as 59%. The authors do not mention this correlation. Obviously, even when only normal embryos were replaced, the selection advantage of PGD could not compensate in excess for a 59% loss in implantation potential. If damage was reduced to an acceptable range, it could have probably shown that PGD is beneficial.
Indeed, if such a tremendous reduction in viability actually were to occur in PGD for AMA, one could reconsider not only PGD for AMA, but all other applications of PGD, including those that the authors of the study consider clinically useful and necessary. But in fact, such a reduction in viability is unprecedented in the PGD literature and is an observation unique to the centers involved in this trial. Even in cryopreservation models estimating the impact of blastomere loss such detrimental effects on viability have never been reported.
Regarding the FISH procedure, no data on error rates were presented and no attempt was made to correct for unclear or missing results - as is routinely done in Reprogenetics laboratories. Moreover, the chromosome panel did not include probes for chromosomes 15 and 22 while it is known that these two chromosomes account for more than 10% of abnormalities in cleavage stage embryos.
The patients were selected from a population of 35-41 year old women with infertility, however, the majority of these patients had fewer than 6 embryos. Our published data clearly show that the efficacy of PGD for AMA is reduced in such patients.
In summary, a careful evaluation of the study by Mastenbroek and co-workers indicates that very poor biopsy technique, suboptimal fixation and FISH methods, and inappropriate patient selection rather than problems inherent to PGD are to blame for the poor results obtained in the Dutch study.
You can contact us for more information by calling Laurie Ferrara, Client Services Associate, at 973-436-5013.
write a review
(No reviews found. You may write the first one!)