Human Reproduction Update Advance Access published April 29, 2011 Human Reproduction Update, Vol.0, No.0 pp. 1 –13, 2011 doi:10.1093/humupd/dmr003

Preimplantation genetic screening: a systematic review and meta-analysis of RCTs S. Mastenbroek*, M. Twisk, F. van der Veen, and S. Repping Center for Reproductive Medicine, Academic Medical Center, University of Amsterdam, Meibergdreef 9, 1105 AZ, Amsterdam, The Netherlands Correspondence address. Tel: +31-20-5663090; E-mail: [email protected]

*

Submitted on December 31, 2009; resubmitted on January 10, 2011; accepted on January 31, 2011

table of contents

........................................................................................................................... † † †

†

†

Introduction Methods Results Results of the search Trial characteristics Extracted data Advanced maternal age Good prognosis patients Repeated implantation failure Developmental stage Method of analysis Discussion Interpretation Reasons for inefficacy Technical aspects Mosaicism New developments Conclusions

background: Preimplantation genetic screening (PGS) has increasingly been used in the past decade. Here we present a systematic review and meta-analysis of RCTs on the effect of PGS on the probability of live birth after IVF. methods: PubMed and trial registers were searched for RCTs on PGS. Trials were assessed following predetermined quality criteria. The primary outcome was live birth rate per woman, secondary outcomes were ongoing pregnancy rate, miscarriage rate, multiple pregnancy rate and pregnancy outcome. results: Nine RCTs comparing IVF with and without PGS were included in our meta-analysis. Fluorescence in situ hybridization was used in all trials and cleavage stage biopsy was used in all but one trial. PGS significantly lowered live birth rate after IVF for women of advanced maternal age (risk difference: 20.08; 95% confidence interval: 20. 13 to 20.03). For a live birth rate of 26% after IVF without PGS, the rate would be between 13 and 23% using PGS. Trials where PGS was offered to women with a good prognosis and to women with repeated implantation failure suggested similar outcomes. conclusions: There is no evidence of a beneficial effect of PGS as currently applied on the live birth rate after IVF. On the contrary, for women of advanced maternal age PGS significantly lowers the live birth rate. Technical drawbacks and chromosomal mosaicism underlie this inefficacy of PGS. New approaches in the application of PGS should be evaluated carefully before their introduction into clinical practice. Key words: preimplantation genetic screening / IVF / live birth / aneuploidy / meta-analysis & The Author 2011. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved. For Permissions, please email: [email protected]

2

Mastenbroek et al.

Introduction

Methods

In 1990, 12 years after the birth of the first child that was conceived by IVF (Steptoe and Edwards, 1978), the first pregnancies were announced that were obtained after transfer of embryos that had been tested for a specific genetic disorder (Handyside et al., 1990). In couples known to be at risk for a genetic disease, only unaffected embryos were transferred after IVF, aspiration of a single cell at the 6–8-cell stage, and subsequent genetic analysis of the aspirated cell, in order to prevent affected children. This new technology was termed PGD (Sermon et al., 2004). At first, only DNA amplification techniques were used for this analysis, but not much later fluorescence in situ hybridization (FISH) techniques were introduced, allowing the detection of abnormal copy numbers of chromosomes, or aneuploidies (Griffin et al., 1992; Delhanty et al., 1993). Further technical developments made it possible to simultaneously analyse more than two chromosomes with the use of FISH (Munne et al., 1993). In 1995 the first deliveries were reported after transfer of embryos that had been screened for aneuploidies (Verlinsky et al., 1995). The underlying rationale for this screening was no longer prevention of a genetic disease, but an expected increase in live birth rates after IVF, because embryos containing aneuploidies were thought not to implant or develop to term and hence to contribute to low live birth rates in specific groups of patients. This was called preimplantation genetic screening (PGS) (Wilton, 2002). The beneficial effect of PGS was expected to be greatest in women of advanced maternal age, since aneuploidies in clinically recognized pregnancies occur more frequently when a woman passes 35 years of age (Hassold and Hunt, 2001) and it is in these women that pregnancy chances decline sharply both in normal conception and after IVF (Lintsen et al., 2007). Next to women of advanced maternal age, PGS has been offered to women with a history of recurrent miscarriage, women with a history of repeated implantation failure (i.e. several failed IVF cycles) and women with a partner with low sperm quality (severe male factor), mainly since high percentages of aneuploidies have been found in the embryos of these women (Munne et al., 1995; Marquez et al., 2000; Silber et al., 2003; Werlin et al., 2003; Munne et al., 2004; Kahraman et al., 2004; Wilding et al., 2004; Platteau et al., 2005; Rubio et al., 2005; Baart et al., 2006). More recently PGS has been offered to younger women (under 35 years of age) too, as high aneuploidy rates were found in their embryos as well (Baart et al., 2006; Goossens et al., 2009). There has been an increase in the use of PGS in the last decade. The European Society for Human Reproduction and Embryology (ESHRE) PGD Consortium reported on 116 cycles of PGS performed worldwide in 1997–1998, steadily increasing to 3900 cycles in 2006 (Goossens et al., 2009). This number is probably an underestimation of the actual number of PGS cycles performed, as this report deals with a voluntary data collection. The only other data collection is a survey among all US-based infertility centers showing that 2197 cycles of PGS were performed in the USA alone in 2005 (Baruch et al., 2008; Goossens et al., 2008a). The efficacy of PGS is, however, still debated. Here we present a systematic review and meta-analysis of RCTs dealing with the effect of PGS on the probability of live birth after IVF.

Search strategy We searched PubMed (www.pubmed.gov) and databases for registration of RCTs (www.clinicaltrials.gov) in December 2009 using the following search criteria: ‘(preimplantation genetic screening OR PGS OR preimplantation genetic diagnosis OR PGD OR aneuploidy screening OR PGD-AS OR screening for aneuploidies OR preimplantation testing OR embryo screening) AND (randomized trial OR randomized trial OR randomized study OR randomized study)’. The reference lists of included studies were searched for relevant studies.

Study selection Two authors (S.M., M.T.) independently examined the electronic search results for reports of possibly relevant trials and those reports were retrieved in full. Published RCTs were eligible for inclusion if they compared women undergoing IVF or ICSI with and without PGS. All trials were assessed following predetermined quality criteria. Validity was assessed in terms of method of randomization, presence of a power calculation, unit of analysis used, use of an intention-to-treat analysis and presence or absence of blinding. If necessary, we contacted the corresponding author of a report in an attempt to retrieve missing data. Trials were grouped based on the indications for PGS: advanced maternal age, recurrent miscarriage, repeated implantation failure, severe male factor and good prognosis patients.

Statistical analysis The effect of PGS was assessed for each indication separately. The primary outcome was live birth rate per woman, secondary outcome measures were ongoing pregnancy rate, miscarriage rate, multiple pregnancy rate and pregnancy outcome per woman. We calculated risk differences (RD) with 95% confidence intervals (CI) for each individual trial. The fixed-effect model was used to combine data for each indication separately and to combine all included studies. Statistical heterogeneity between results of studies was examined by inspecting the scatter in the data points on the graphs and the overlap of CIs, and by checking the I 2 statistic. A value of ≥50% was considered to indicate substantial heterogeneity. In case of substantial heterogeneity the random effects model was used instead of the fixed-effect model. Data were analyzed according to the intention-to-treat principle. Revman Software (Version 5, The Cochrane Collaboration) was used to combine data for meta-analysis.

Results Results of the search The literature search resulted in the identification of fourteen potentially eligible trials (Fig. 1) (Gianaroli et al., 1997; Gianaroli et al., 1999; Werlin et al., 2003; Staessen et al., 2004; Stevens et al., 2004; Mastenbroek et al., 2007b; Staessen et al., 2008; Hardarson et al., 2008; Mersereau et al., 2008; Blockeel et al., 2008; Jansen et al., 2008; Meyer et al., 2009; Debrock et al., 2009; Schoolcraft et al., 2009a). Five trials were excluded after further assessment. Two trials were excluded since couples were allocated to the treatment or control group on the basis of their volunteer decision, instead of random allocation (Gianaroli et al., 1997, 1999). PGS in these trials was performed for advanced maternal age and repeated IVF

3

Preimplantation genetic screening and live birth rate

Figure 1 Flow chart of search for RCTs on PGS.

failure. One trial was excluded (Stevens et al., 2004) since there was data overlap with another publication (Schoolcraft et al., 2009a). One study was excluded since it contained insufficient data to assess the methodological quality (Werlin et al., 2003). In this trial PGS was performed for advanced maternal age, recurrent miscarriage and repeated implantation failure. Baseline characteristics of patients were not provided and method of randomization, concealment of allocation and blinding status were not mentioned. This trial reported no outcomes beyond preclinical pregnancy. One trial was excluded since it was an ongoing trial and since data were reported only as percentages without mentioning the unit of analysis (Mersereau et al., 2008). PGS was offered in this trial to women with a good prognosis. Nine trials, representing 1589 treated women, were included in the meta-analysis (Staessen et al., 2004; Mastenbroek et al., 2007b; Blockeel et al., 2008; Hardarson et al., 2008; Jansen et al., 2008; Staessen et al., 2008; Debrock et al., 2009; Meyer et al., 2009; Schoolcraft et al., 2009a). In addition, three RCTs that are currently recruiting patients were identified by searching www.clinicaltrials.gov. In two trials (NCT00646893; NCT00795795) PGS is offered to women of

advanced maternal age. In one trial (NCT00547781) PGS is offered to women with repeated implantation failure. In these three trials biopsy will take place at cleavage stage (Day 3 of embryo development) and FISH will be used for the analysis. One trial for advanced maternal age will apply assisted hatching to the embryos in the control group, the other two trials use IVF without PGS or assisted hatching as control treatment.

Trial characteristics Main characteristics and quality features of the nine included trials are presented in Table I. In five studies PGS was offered to women of advanced maternal age (Staessen et al., 2004; Mastenbroek et al., 2007b; Hardarson et al., 2008; Debrock et al., 2009; Schoolcraft et al., 2009a). In three studies PGS was offered to good prognosis patients (Jansen et al., 2008; Staessen et al., 2008; Meyer et al., 2009). In one study PGS was offered to women with repeated implantation failure (Blockeel et al., 2008). In one trial, a maximum of three treatment cycles was offered to each participating couple (Mastenbroek et al., 2007b). In one trial a couple could participate several times with independent

4 randomization for each cycle (Debrock et al., 2009). All other trials offered one treatment cycle to each participating couple. One trial analyzed two to nine trophectoderm cells from the blastocyst stage of embryo development, aspirated from the embryo at the fifth or sixth day after fertilization (Jansen et al., 2008). All other trials analyzed blastomeres from the cleavage stage of embryo development, aspirated from the embryo at the third day after insemination. In the majority of these trials preferably one blastomere was aspirated for analysis, and in two trials two blastomeres were aspirated (Staessen et al., 2004; Debrock et al., 2009). All trials used FISH for the analysis of the aspirated cells. There were no trials that analyzed all chromosomes. One embryo was transferred to the uterus in two trials (Jansen et al., 2008; Staessen et al., 2008), two embryos were transferred if these were available in two trials (Mastenbroek et al., 2007b; Hardarson et al., 2008), and up to three embryos were transferred in one trial (Blockeel et al., 2008). In one trial the number of embryos transferred depended on the age of the woman, since up to three embryos were transferred when the recipient was between 37 and 39 years old and up to six embryos were transferred when the recipient was 40 years or older (Staessen et al., 2004). In one trial a maximum of three embryos were transferred in the first period of the trial and only one embryo in the second part of the trial for first treatment cycles of women younger than 36 years (Debrock et al., 2009). In two trials transfer policy was not described (Meyer et al., 2009; Schoolcraft et al., 2009a). In one trial there was a difference in embryo transfer policy between the PGS treatment group and the control group. In the PGS treatment group embryo transfer was performed on Day 5 while in the control group transfer was performed on Day 3 (Hardarson et al., 2008). Data on cryopreservation cycles and pregnancies resulting from those cycles were available from two trials (Mastenbroek et al., 2007b; Debrock et al., 2009). The included trials used embryo implantation rate (Staessen et al., 2004; Blockeel et al., 2008; Debrock et al., 2009) (clinical) pregnancy rate (Hardarson et al., 2008), ongoing pregnancy rate (Mastenbroek et al., 2007b) or live birth rate (Jansen et al., 2008; Staessen et al., 2008; Meyer et al., 2009) as primary outcome measure. For one trial the primary outcome was not specified (Schoolcraft et al., 2009a). Live birth rate, the primary outcome measure of our meta-analysis, was reported in nine trials (Table II) (Staessen et al., 2004; Mastenbroek et al., 2007b; Staessen et al., 2008; Blockeel et al., 2008; Hardarson et al., 2008; Jansen et al., 2008; Debrock et al., 2009; Meyer et al., 2009; Schoolcraft et al., 2009a). For two trials the authors kindly provided missing or additional data on live birth rates upon request (Staessen et al., 2004; Debrock et al., 2009). The ongoing pregnancy rate was available for five trials (Staessen et al., 2004; Mastenbroek et al., 2007b; Blockeel et al., 2008; Staessen et al., 2008; Debrock et al., 2009), the miscarriage rate was available for nine trials (Staessen et al., 2004; Mastenbroek et al., 2007b; Blockeel et al., 2008; Hardarson et al., 2008; Jansen et al., 2008; Staessen et al., 2008; Debrock et al., 2009; Meyer et al., 2009; Schoolcraft et al., 2009a), multiple pregnancy rate was available in six trials (Staessen et al., 2004; Mastenbroek et al., 2007b; Blockeel et al., 2008; Hardarson et al., 2008; Staessen et al., 2008; Debrock et al., 2009, personal communication), and information on congenital malformations was available from three trials (Staessen et al., 2004; Mastenbroek et al, 2007b, personal communication; Hardarson et al., 2008 personal communication).

Mastenbroek et al.

Power calculation was used to calculate sample size in seven trials (Staessen et al., 2004; Mastenbroek et al., 2007b; Blockeel et al., 2008; Hardarson et al., 2008; Jansen et al., 2008; Staessen et al., 2008; Meyer et al., 2009). Two trials were terminated prematurely since continuing was considered not useful after interim analysis (Hardarson et al., 2008; Staessen et al., 2008). One trial was terminated prematurely since interim analysis showed such a large difference that it was considered unethical to continue (Meyer et al., 2009), and one trial was terminated prematurely since the trend was opposite to what was expected and cryopreserved blastocysts were accumulating in the control group (Jansen et al., 2008). In two studies the power calculation was based on embryos instead of women (Staessen et al., 2004; Blockeel et al., 2008). This results in a underpowered sample size when analyzing an outcome where women are the unit of analysis, as is done in this meta-analysis by using live birth rate per woman as primary outcome (Mastenbroek et al., 2005).

Extracted data In one trial randomization assigned 200 women to the PGS treatment group and 200 women to the control group (Staessen et al., 2004). However, in the PGS group one woman did not fulfill the inclusion criteria, and in the control group 10 women did not, so therefore we included 199 and 190 women, respectively, in our meta-analysis (Table II). In one trial, randomization assigned 95 women to the PGS treatment group and 105 women to the control group (Blockeel et al., 2008): among these were 8 and 10 women, respectively, that were not fulfilling the inclusion criteria and we excluded those from the meta-analysis. The authors excluded 15 women in the PGS group and 28 women in the control group owing to ‘insufficient ovarian response’, ‘stop further fertility treatment’, ‘wrong allocation’ and ‘spontaneous pregnancy’. No results were provided for these women, so we could not include them in our meta-analysis following the intention to treat principle. Thus, we included 72 and 67 women for the PGS and control group, respectively, for this trial in our meta-analysis. One trial reported only percentages and used transfers and pregnancies as units of analysis, therefore we recalculated the numbers per patient for the various outcomes (Schoolcraft et al., 2009a). One trial included drop-outs in their analysis (Mastenbroek et al., 2007b), four trials did not or did not correctly do so (Staessen et al., 2004, 2008; Debrock et al., 2009; Meyer et al., 2009) and for the other trials nothing was mentioned about drop-outs. Missing information about drop-outs was kindly provided on request by the authors of one trial (Debrock et al., 2009).We analyzed all trials following the intention to treat principle. All drop-outs, where available, were included in our analysis.

Advanced maternal age For the primary outcome live birth rate we combined the outcomes of five trials, reporting on 537 women in the PGS group and 525 women in the control group (Fig. 2). Live birth rate was significantly lower after PGS (18%) compared with the control group (26%) (n ¼ 1062; RD: 20.08; 95% CI: 20.13 to 20.03). This suggests that for this group of women with a live birth rate of 26% after IVF without PGS, the rate would be between 13 and 23% when using PGS. There was no indication for statistical heterogeneity.

Study

Patients

Cycles

Indication for PGS

Inclusion criteria

ART and laboratory

Biopsy and FISH

Quality features

.......................................................................................................................................................................................................................................................... Staessen et al. (2004)

389

389

AMA

≥37 years, normal karyotype of both partners, need for ICSI with motile sperm

ICSI; maximum three embryos for transfer if age 37– 39, maximum six embryos if age ≥40;

Laser; mostly two blastomeres removed; biopsy on embryos with at least five blastomeres and with a maximum of 50% fragmentation; X, Y, 13, 16, 18, 21, 22

Method of randomization nr; concealment of allocation; not blind; power calculation performed; two centers; full paper; no intention to treat

Mastenbroek et al., 2007b

408

836

AMA

35– 41 years, no previous failed IVF cycles, no objection to DET

IVF and ICSI; maximum two embryos for transfer

Laser; mostly one blasomere removed; biopsy on embryos with at least four blastomeres and with a maximum of 50% fragmentation; X, Y, 1, 13, 16, 17, 18, 21

Randomization by computer; concealment of allocation; double blind; power calculation performed; four centers; full paper; analysis on intention-to-treat basis; data on cryo-cycles present

Hardarson et al. (2008)

109

109

AMA

≥38 years, at least three embryos of good morphological quality if DET, at least two embryos of good morphological quality if SET

IVF and ICSI; maximum two embryos for transfer

Laser; mostly one blasomere removed; biopsy on embryos with at least six blastomeres and with a maximum of 20% fragmentation; X, Y, 13, 16, 18, 21, 22

Randomization by computer; concealment of allocation nr; not blind; power calculation performed; two centers; full paper; intention-to-treat nr

Schoolcraft et al., 2009a

62

62

AMA

≥35 years, presence of at least five embryos with ≥6 cells and ≤15% fragmentation on Day 3

IVF and ICSI; maximum number of embryos transferred nr

Laser or acidic tyrodes; mostly one blastomere removed; embryos for biopsy nr.; X, Y, 13, 15, 16, 17, 18, 21, 22

randomization by computer; concealment of allocation nr; blinding nr; power calculation nr; centers nr; full paper; intention-to-treat nr

Debrock et al. (2009)

94

94

AMA

≥35 years with at least two fertilized oocytes available on Day 1 after oocyte retrieval, and with at least two embryos consisting of six or more cells at Day 3 after oocyte retrieval

IVF, ICSI and TESE; maximum two.to three embryos transferred, after July 2003 maximum one embryo transferred in first cycle of patients ,36 years

Laser; two blastomeres removed; biopsy on embryos with al least six blastomeres; X, Y, 13, 16, 18, 21, 22

randomization by blinded envelopes; concealment of allocation nr; not blind; power calculation nr; single center; abstract; no intention to treat; data on cryo-cycles present

Staessen et al. (2008)

240

214

Good prognosis

,36 years, need for ICSI with motile sperm, both partners normal karyotype

ICSI; single embryo transfer

Laser; mostly one blastomere removed; biopsy on embryos with al least five blastomeres with a maximum of 50% fragmentation X, Y, 13, 16, 18, 21, 22

Randomization by computer; concealment of allocation nr; not blind, power calculation performed, but study was stopped after interim analysis because of futility; one center; full paper; intention-to-treat nr

Jansen et al. (2008)

101

101

Good prognosis

,38 years, agreement to SET, first or second IVF, no cancelled cycles because of poor response, eight or more follicles .1 cm on Days 8– 10 of stimulation, four or more embryos seven or more cells on Day 3, two or more blastocysts for biopsy

IVF/ICSI nr; single embryo transfer

Laser; 2 –9 trophectoderm cells removed; biopsy only on blastocysts; X, Y, 13, 18, 21

Randomization by sealed envelopes; concealment of allocation nr; blinding nr; power calculation performed but not reached; one center; full paper; intention to treat

Preimplantation genetic screening and live birth rate

Table I Characteristics of studies included in a systematic review and meta-analysis of RCTs on the effect of preimplantation genetic screening (PGS) on the probability of live birth after IVF.

Continued

5

6

AMA, advanced maternal age; DET, double-embryo transfer; nr, not reported; RIF, repeated implantation failure; SET, single-embryo transfer; TESE, testicular sperm extraction.

Randomization by computer; no concealment of allocation; blinding nr; power calculation performed; one center; full paper, intention-to-treat nr Method for zona drilling nr; between 2001 and October 2005 two blastomeres were removed from embryos with at least six blastomeres, from November 2005 onwards, one blastomere was removed; X, Y, 13, 16, 18, 21, 22 162 Blockeel et al. (2008)

139

RIF

Three or more failed IVF/ICSI attempts with embryos of good morphological quality, maternal age ,37 years, normal karyotype of both partners, motile sperm

ICSI; maximum three embryos for transfer

Randomization by computer; concealment of allocation nr; not blind; power calculation performed, but study was stopped after interim analysis because of large difference between the two groups; one center; full paper; no intention to treat Acidic tyrodes; mostly one blastomere removed; embryos for biopsy nr; X,Y, 13, 16, 17, 18, 21, 22 ICSI; maximum embryos for transfer nr ,39 years, normal ovarian reserve, body mass index 30 kg/m2, no smoking history, no hydrosalpinx, presence of ejaculated sperm, normal uterus, two or more previous failed IVF cycles, four or more embryos containing at least 5-cells with ,40% fragmentation Good prognosis 47 47 Meyer et al. (2009)

..........................................................................................................................................................................................................................................................

Patients Study

Table I Continued

Cycles

Indication for PGS

Inclusion criteria

ART and laboratory

Biopsy and FISH

Quality features

Mastenbroek et al.

Ongoing pregnancy rate was significantly lower in the PGS group (18%) than in the control group (26%) (n ¼ 891; RD: 20.08, 95% CI: 20.13 to 20.02) without indication for statistical heterogeneity. The miscarriage rate was not significantly different between the PGS group (11%) and the control group (12%) (n ¼ 1062; RD: 20.01, 95% CI: 20.05 to 0.02) without indication for statistical heterogeneity. The multiple pregnancy rate was not significantly different between the PGS and the control group; 14 twin and one triplet pregnancies were reported in the PGS group, with one twin pregnancy resulting from a vanishing triplet pregnancy, and 23 twin and 3 triplet pregnancies were reported in the control group. Two triplet pregnancies were reduced to a singleton pregnancy and one to a twin pregnancy, which ended in a singleton pregnancy. This results in an RD of 0.00 (n ¼ 199; 95% CI: 20.12 to 0.11; no indication for statistical heterogeneity). In the PGS group one trisomy 18 (after spontaneous conception) was reported, one intrauterine death (due to abruptio placentae) and one premature delivery of twins at 24 weeks of gestation, resulting in the post-partum death of both children (Mastenbroek et al., 2007b). In the control group one trisomy 21 was reported (Staessen et al., 2004, personal communication), one trisomy 18, one conception with a cleft lip and palate and one intrauterine death of a fetus (Mastenbroek et al., 2007b). In one trial no congenital malformations were detected (Hardarson et al., 2008 personal communication). The other studies did not report on pregnancy outcome.

Good prognosis patients For the primary outcome live birth rate we combined the outcomes of three trials, reporting on 198 women in the PGS group and 190 women in the control group (Fig. 2). The live birth rate was lower after PGS (32%) compared with the control group (42%) (n ¼ 388; RD: 20.10, 95% CI: 20.20 to 20.01), however there was substantial statistical heterogeneity (I 2 ¼ 76%). When pooling the data using the random effects model because of this substantial statistical heterogeneity, no difference in live birth rate could be proved (n ¼ 388; RD: 20.17, 95% CI: 20.39 to 0.04). This suggests that for this group of women with a live birth rate of 42% after IVF without PGS, the rate would be between 3 and 46% when using PGS. Ongoing pregnancy rate was only reported in one trial (Staessen et al., 2008), therefore no meta-analysis could be performed. The ongoing pregnancy rate was not significantly different between the PGS (31%) and the control group (33%) (n ¼ 240; RD: 20.02, 95% CI: 20.13 to 0.10). The combined miscarriage rate was 10% in the PGS group and 9% in the control group (n ¼ 388; RD: 0.01, 95% CI: 20.04 to 0.07). There was substantial statistical heterogeneity (I 2 ¼ 69%). Pooling the data using the random effect model showed no difference in miscarriage rate (n ¼ 388; RD: 0.05, 95% CI: 20.06 to 0.16). In one trial three multiple pregnancies occurred even though single embryo transfer was applied (Staessen et al., 2008); one in the PGS group and two in the control group (RD: 20.03; 95% CI: 20.12 to 0.06). In one trial multiple pregnancy rate was not mentioned (Jansen et al., 2008) and in one trial it was only mentioned as a percentage and it was not possible to calculate this into exact numbers (Meyer et al., 2009).

Study/indication

Patients

Live births

Ongoing pregnancies

Clinical pregnancies

Pregnancies

PGS

PGS

PGS

Miscarriages

Multiple pregnancies

Embryo transfers

Embryos per transfer

PGS

PGS

PGS

...................... ...................... ...................... ...................... ...................... ...................... ...................... ...................... ...................... PGS

Control

PGS

Control

Control

Control

Control

PGS

Control

Control

Control

Control

.......................................................................................................................................................................................................................................................... Advanced maternal age Staessen et al. (2004)

199

190

21

29

22

29

22

30

29

39

7

10

4

6

81

121

2.0

2.8

Mastenbroek et al., 2007ba

206

202

49

71

52

74

61

88

81

106

37

36

10

14

367

364

1.8

1.9

Hardarson et al. (2008)

56

53

3

10

5

13

10

16

7

6

0

2

45

53

1.5

1.8

Schoolcraft et al., 2009a

32

30

16

16

21

23

5

7

31

30

2.2

2.7

Debrock et al. (2009)

44

50

6

10

6

10

8

13

8

15

2

5

1

1

39

47

1.6

2.0

120

120

37

37

37

39

37

44

47

52

10

15

1

2

85

89

1

1

Jansen et al. (2008)

55

46

20

27

22

27

25

28

5

1

54

46

Meyer et al. (2009)

23

24

6

15

11

16

5

1

21

22

1.8

1.8

67

15

26

18

27

10

5

55

63

1.4

2.1

Preimplantation genetic screening and live birth rate

Table II Outcomes of included studies.

Good prognosis Staessen et al. (2008)

Repeated implantation failure Blockeel et al. (2008)

72

15

26

25

31

3

10

a

Multiple cycles per patient; outcomes reported as women with at least one occurrence of that outcome; number of embryo transfers is total of all cycles.

7

8

Mastenbroek et al.

Figure 2 The effect of PGS on the live birth rate per patient.

In one trial no major or minor congenital abnormalities were detected (Meyer et al., 2009). The other two trials did not report on congenital abnormalities.

Repeated implantation failure The single trial in which PGS was offered to women with repeated implantation failure showed a significant difference in live birth rate between the PGS group (21%) and the control group (39%) (n ¼ 139; RD: 20.18, 95% CI: 20.33 to 20.03) (Fig. 2). This suggests that for this group of women with a live birth rate of 39% after IVF without PGS, the rate would be between 6 and 36% when using PGS. The ongoing pregnancy rate per woman was significantly different between the PGS group (21%) and the control group (39%) (n ¼ 139; RD: 20.18, 95% CI: 20.33 to 20.03). The miscarriage rate was not significantly different between the PGS (14%) and the control group (7%) (RD: 0.06, 95% CI: 20.04 to 0.17). In the PGS group three twin pregnancies were reported and in the control group 10 twin pregnancies were reported (RD: 20.18; 95% CI: 20.46 to 0.09). All children in the PGS group as well as the control group were born healthy.

Developmental stage There was only one trial that analyzed embryos at blastocyst stage (Jansen et al. 2008). This was a trial with good prognosis patients

and live birth rates were significantly lower after PGS (n ¼ 101; RD: 20.22, 95% CI: 20.41 to 20.03). When PGS was performed at the cleavage stage for the same indication the live birth rate was also lower after PGS, but this difference was not significant (n ¼ 287; Fixed-effect model: RD: 20.06, 95% CI: 20.17 to 0.05, I 2 84%; random effects model: RD: 20.16, 95% CI: 20.52 to 0.19).

Method of analysis All trials used FISH for the analysis of aspirated cells, thus no comparison could be made with other methods of analysis.

Discussion Interpretation Our review and analysis of the literature clearly shows that there is no evidence of a beneficial effect of PGS as currently conducted on live birth rates after IVF. In women of advanced maternal age PGS significantly lowered live birth rate after IVF (RD: 20.08; 95% CI: 20.13 to 20.03). In women with a good prognosis (RD: 20.17, 95% CI: 20.39 to 0.04) and in women with repeated implantation failure (RD: 20.18, 95% CI: 20.33 to 20.03) a similar effect after PGS was seen.

9

Preimplantation genetic screening and live birth rate

It has been argued that the inclusion criteria for the trials included in this review were not targeted specifically enough to women with a high risk of embryonic aneuploidy (Handyside and Thornhill, 2007). Data from one of the trials included in this review were analyzed to assess whether the effect of PGS indeed differs with variable risk for embryonic aneuploidy, but no beneficial effect of PGS was found regardless of the risk for embryonic aneuploidy (Twisk et al., 2008). It has also been argued that the trials included in this review lack technical prowess causing them to be neither valid nor generalizable (Cohen and Munne, 2005; Cohen and Grifo, 2007; Munne et al., 2007; Simpson, 2008; Munne et al., 2010), but all trials from multiple independent established groups showed the same negative effect of PGS and therefore it seems particularly justified to generalize the outcome of these studies and to conclude that there is no beneficial effect of PGS in terms of increased live birth rates.

Reasons for inefficacy There are several possible explanations for the inefficacy of PGS. These concern technical aspects of the PGS procedure, such as possible harm from the biopsy procedure, the failure rate of the technique and the limitations of the FISH analysis, in addition to an intrinsic biological feature of the analyzed embryos, which is their mosaic nature.

Technical aspects Biopsy A biopsy entails taking an embryo out of the incubator for at least a couple of minutes, making a hole in the zona pellucida—by mechanical dissection, by using acidic Tyrode’s solution or by using a laser—and aspirating one or more cells (De Vos and Van Steirteghem, 2001). It seems plausible to assume that an embryo would at least not benefit from this, but the extent to which this could be harmful for an embryo is still a moot point as data on this topic are limited. It has been shown that aspiration of cells at the 2 –4-cell stage, or the removal of a quarter of the cell mass, influences embryo development by lowering the ratio of inner cell mass cells to trophectoderm cells (Tarin et al., 1992). Embryo development was reported to be less affected when biopsy was performed at the 8-cell stage (Hardy et al., 1990; Hardy and Handyside, 1993). However, the effect of biopsy alone on pregnancy rates has never been properly studied. It has been argued that the removal of two blastomeres at Day 3 of development is likely to negatively influence pregnancy rates (Cohen et al., 2007), similar to the negative effect of cell loss after cryopreservation on pregnancy rates (Edgar et al., 2000a, b), which would be relevant to two trials included in our review (Staessen et al., 2004; Debrock et al., 2009). A prospective controlled trial showed aspiration of two blastomeres instead of one to negatively influence embryo development without an effect on live birth rates (Goossens et al., 2008b), although more recently a cohort study by the same research group did show a detrimental effect of the aspiration of two blastomeres on live birth rates (De Vos et al., 2009).

blastomeres, and 92.3 and 92.9% blastomeres with successful FISH analysis for the centers in Amsterdam and Brussels, respectively (Michiels et al., 2006; Mastenbroek et al., 2007b): this led to 20.1% of embryos without a diagnosis in the former trial and 2.4% in the latter trial (instead of a cumulative 19.3%), since in the latter trial a second cell was available for the analysis of each embryo (Staessen et al., 2004; Mastenbroek et al., 2007b). The Staessen et al. (2004) and Mastenbroek et al., (2007b) trials have been fiercely criticized for their failure rates (Cohen and Munne, 2005; Cohen and Grifo, 2007; Munne et al., 2007). The ESHRE PGD consortium reports a failure rate of 8.8% per embryo for all indications of PGS and 9.9% for the indication advanced maternal age in their latest data collection [data collection I –IX (Goossens et al., 2009)]. This could reflect underreporting, since the data collection is voluntary, not based on an intention to treat analysis, and incomplete reporting has previously been noted (Sermon et al., 2007). Another reason for the reported difference in diagnostic efficiency could be the way PGS was performed. Failure rate in the criticized trials was especially high in embryos with four or five blastomeres with up to 50% fragmentation (Mastenbroek et al., 2007a), possibly related to the arrested nature of these embryos and the resulting compromised DNA integrity. In more recent years, it became common to perform biopsy only on better quality embryos, which obviously limits the applicability of PGS, but possibly increases diagnostic efficiency (Munne et al., 2003; Summers and Foland, 2009). One trial on advanced maternal age that biopsied only better quality embryos, i.e. embryos with at least six cells and ,20% fragmentation, had a failure rate of 10.6% (Hardarson et al., 2008). Another trial on advanced maternal age that used even stricter inclusion criteria, i.e. the presence of at least five embryos with at least six or more cells and ,15% fragmentation, had a failure rate of 5.0% (Schoolcraft et al., 2009a). The last trial included in our review for advanced maternal age performed 2-cell biopsy and diagnostic efficiency could not be corrected for this (Debrock et al., 2009). Performing PGS on lower quality embryos thus seems to have lowered diagnostic efficiency in the two largest available trials; it is important to note that this does not necessarily translate to decreased PGS efficacy, as these lower quality embryos are simply excluded in the other trials. Indeed, removing the trials with the highest failure rates from our analysis did not result in a different RD for our main outcome (RD: 20.09; 95% CI: 20.18 to 0.01 versus RD: 20.08; 95% CI: 20.13 to 20.03). This result is not surprising because statistical heterogeneity was 0% in our initial analysis. Other adjustments to the PGS technique that have been suggested in recent years, such as the reanalysis of a nucleus with a different probe set in case of a failed FISH analysis (Colls et al., 2007) or the use of an increased number of probes (Colls et al., 2009), could also contribute to lower PGS failure rates, but prospective comparative data to substantiate this are not available.

FISH Failure rate PGS has not been without failure. The success rates for the separate steps of PGS, i.e. biopsy, fixation and FISH analysis, seem to be similar between the largest RCTs included in this review: 97.2 and 94.2% successfully biopsied embryos, 89.1 and 93.6% successfully fixated

All trials included in our review used FISH for the analysis of aspirated cells. The intrinsic limitations of FISH could well in part underlie the inefficacy of PGS. Owing to technical limitations, i.e. the available number of fluorochromes and the time available for the analysis, one can only test for a limited number of chromosomes in PGS

10 using FISH. PGS therefore fails to detect embryos that are aneuploid for the chromosomes that are not tested. In addition, it is not exactly known which combination of probes is best. The selection of probes is in fact quite arbitrary since there are no large detailed studies on the frequency of numerical abnormalities of all 23 chromosome pairs in cleavage stage embryos. Prospective studies comparing different probe sets in relation to live birth rates after PGS have not been conducted. Furthermore, scoring errors of FISH signals may arise from loss or damage of nuclear material, overlapping signals, split signals, diffused signals, hybridization failure and probe inefficiency. FISH analysis has a 92 –99% accuracy per probe, so when using a multi-probe panel on 1 cell, the risk of misdiagnosis is significant (Ruangvutilert et al., 2000; Michiels et al., 2006; Deugarte et al., 2008; Scriven and Bossuyt, 2010). To illustrate this: an accuracy of 98% per probe will result in an estimated 15% error rate for an eight-probe panel (0.98^8). The low positive predictive value of the test will result in the exclusion of embryos for consideration for transfer, thereby worsening PGS efficacy (Michiels et al., 2006; Deugarte et al., 2008; Northrop et al., 2010; Treff et al., 2010).

Mosaicism Embryo mosaicism, the condition that the chromosomal constitution differs between blastomeres of the same embryo, could well be an important reason for the inefficacy of PGS. Clinical studies on the effect of mosaicism on PGS commonly focus on ‘misdiagnosis’ rates, i.e. false-negative and false-positive-rates of FISH analysis, rather than on the impact of mosaicism on pregnancy or live birth rates (Los et al., 2004; Staessen et al., 2004). It is diploid –aneuploid mosaicism, i.e. the condition that an embryo consists of both normal and aneuploid blastomeres, which could be causal to the lower live birth rates after PGS. Aspiration of aneuploid blastomeres from diploid– aneuploid embryos leads to the discarding of these embryos, although they are potentially viable as these embryos contain normal blastomeres. Conversely, aspiration of normal blastomeres from diploid – aneuploid embryos results in transfer or cryopreservation of these embryos, although the proportion of normal blastomeres was reduced, thereby potentially hampering the embryos potential. A recent systematic review of the literature reported that 46% of 107 developing cleavage stage embryos that were analyzed for eight or more chromosomes were diploid –aneuploid mosaic, with a mean of 47% normal blastomeres in these embryos (van Echten et al., 2011). This means that in 24% of all embryos (0.46 × 0.53) an aneuploid blastomere will be aspirated, thus leading to discarding of these potentially viable embryos and in 22% of all embryos (0.46 × 0.47) a normal blastomere is aspirated during the PGS procedure, thus leading to transfer or cryopreservation of embryos with a relatively lower number of normal blastomeres. It must be noted that these data were obtained from spare, and thus lower quality embryos after IVF, which have been suggested to contain higher percentages of abnormalities (Ziebe et al., 2003; Magli et al., 2007). Furthermore, it is unknown to what extent FISH errors contribute to the prevalence of mosaicism and thus to the percentages mentioned here (Treff et al., 2010), although recently support for the above reasoning was obtained with the use of genome-wide comparative genome hybridization arrays (CGH arrays) (Vanneste et al., 2009a, b). Notwithstanding

Mastenbroek et al.

the uncertainty of the exact percentages, there seems to be a clear role for mosaicism in the inefficacy of PGS.

New developments The publication of an increasing number of trials that was unable to show any benefit of PGS led to a renewed and increasing interest in further development of, and adjustments to, the PGS technique in recent years. The focus of these developments in technique optimization is on the timing of biopsy, i.e. the developmental stage of the embryo when PGS is performed, and on the method that is used to analyze the aspirated cells.

Developmental stage Of the 10 trials included in our review, 9 performed PGS at cleavage stage (generally at Day 3 of embryo development) and one performed PGS at the blastocyst stage (generally at Day 5 or 6 of embryo development). Blastomere aspiration at Day 3 has been most widely used in PGS (Goossens et al., 2009), because it is less harmful for the embryo than at earlier stages (Hardy et al., 1990; Tarin et al., 1992), and it provides enough time for the analysis of cells before transfer (De Vos and Van Steirteghem, 2001). Although the single trial included in our review that performed PGS at the blastocyst stage showed lower live birth rates after IVF/ICSI with PGS (Jansen et al., 2008), there is an increased interest in blastocyst biopsy. In blastocyst biopsy more cells can be aspirated and mosaicism seems to be less frequent (Bielanska et al., 2002). It is possible to remove only the extraembryonic trophectoderm cells, leaving the inner cell mass, which contains the cells the future child derives from, intact (Dokras et al., 1990). However, it is unknown whether trophectoderm cells are representative of the cells the future child derives from and blastocyst PGS leaves only limited time for the analysis of the aspirated cells. Although the first reported children after PGS were conceived with the use of polar body aspiration (Verlinsky et al., 1995; Verlinsky et al., 2004), its use remained limited compared with cleavage stage aspiration (Goossens et al., 2009). With the current evidence for inefficacy of PGS using cleavage stage aspiration, there is a renewed interest in the analysis of polar bodies for PGS purposes (Geraedts et al., 2009). An advantage is that both polar bodies are extra-embryonic material and are expected to have no biological role in the further development of the embryo. A disadvantage is that only maternal meiotic errors and no abnormalities from paternal origin or abnormalities arising during mitosis (mosaicism) can be evaluated (De Vos and Van Steirteghem, 2001; Kuliev and Verlinsky, 2004; Montag et al., 2009). Yet it is assumed that the extent of paternal contribution to embryonic aneuploidy is limited and surely less than the maternal contribution (Nicolaidis and Petersen, 1998; Magli et al., 2009; Montag et al., 2009). A preclinical study recently conducted by the ESHRE PGS Task Force has shown that PGS using polar body biopsy and array technology for the analysis of the polar bodies is technically feasible and now requires a prospective RCT to evaluate its efficacy (Geraedts, 2010).

Method of analysis Optimizing the FISH technique used for PGS has been proposed, such as the re-analysis of a cell with different FISH probes for the same chromosomes in case of FISH failure (Colls et al., 2007), or increasing the number of FISH probes by performing multiple rounds of FISH

11

Preimplantation genetic screening and live birth rate

(Colls et al., 2009). Whether these modifications indeed improve PGS efficacy, in terms of live birth rate per women, has not yet been shown. Alternatives to the use of FISH, such as methods based on CGH or the analysis of single nucleotide polymorphisms (SNPs), have also been proposed (Wells et al., 2008). The use of CGH-based methods for PGS is not new; the first report of a child born after IVF with PGS using CGH stems from 2001 (Wilton et al., 2001). As the time needed for the CGH analysis (3–4 days) exceeds the time available between biopsy and transfer (1 –2 days), cryopreservation of all embryos was necessary, which probably limited its widespread use because cryopreservation methods hampered the embryos potential for implantation, already compromised by the biopsy (Edgar et al., 2000b; Wilton et al., 2001; Stachecki et al., 2005; Magli et al., 2006). With recent advances in cryopreservation technology, the need for cryopreservation of embryos is considered to be less of a problem, resulting in an increased interest in PGS with the use of CGH (Kuwayama, 2007; Teramoto and Kato, 2007; El-Toukhy et al., 2009; Keskintepe et al., 2009; Schoolcraft et al., 2009b). At the same time, increased effort is put in the development of arraybased methods that allow for analysis within the available time frame between biopsy and transfer (Hellani et al., 2008). Cryopreservation is thus not required, and these methods potentially have the advantage of being less complex and easier to use than the above-mentioned CGH analysis. Some of these array-based methods use the principle of CGH, but now labeled DNA is hybridized with DNA spotted on a slide instead of hybridization with metaphase chromosomes, with the potential benefit of automated analysis and less time needed for the hybridization (Wells et al., 2008). Other array-based methods analyze SNPs, i.e. single nucleotide variations in the human genome to detect aneuploidies (Wells et al., 2008; Northrop et al., 2010; Treff et al., 2010). Currently, these developments are at the stage of method assessment or at the stage of pilot studies. No trials that investigate the value of these new approaches in PGS have yet been registered. Nonetheless, these developments have already been implemented in clinical practice at least in some clinics. Indeed, the theory behind these developments sounds plausible, the techniques used are attractive and provide potential, and the first results seem promising. However, a decade ago, this was exactly the same when PGS was developed and introduced into clinical practice. Any fair-minded person will conclude from this review that PGS as widely practiced has for many women been a significant additional cost without robust evidence of any benefit, and for some their chance of a live birth has probably been harmed. The same issues that are suggested to undermine PGS as reviewed here, i.e. possible harm of the biopsy procedure, a high failure rate, incorrect diagnosis by the technique used for analysis, and mosaicism, could very well also undermine these new approaches. Therefore, new approaches to PGS should not simply be introduced into clinical practice; it is essential that method assessment and pilot studies must first provide sufficient evidence to allow rigorous RCTs on the efficacy of these new techniques.

Conclusions There is no evidence of a beneficial effect of PGS as currently applied on the live birth rate after IVF. On the contrary, for women of advanced maternal age PGS significantly lowers the live birth rate.

New approaches in the application of PGS should be evaluated carefully before their introduction into clinical practice.

Authors’ roles S.M.: conception and design of study; acquisition, analysis and interpretation of data; drafting and revising the manuscript; final approval of manuscript. M.T.: design of study; acquisition, analysis, and interpretation of data; revising the manuscript; final approval of manuscript. F.V.: conception of study; interpretation of data; revising the manuscript; final approval of manuscript. S.R.: conception of study; interpretation of data; revising the manuscript; final approval of manuscript.

References Baart EB, Martini E, van den Berg I, Macklon NS, Galjaard RJ, Fauser BC, Van Opstal D. Preimplantation genetic screening reveals a high incidence of aneuploidy and mosaicism in embryos from young women undergoing IVF. Hum Reprod 2006;21:223– 233. Baruch S, Kaufman D, Hudson KL. Genetic testing of embryos: practices and perspectives of US in vitro fertilization clinics. Fertil Steril 2008;89:1053– 1058. Bielanska M, Tan SL, Ao A. Chromosomal mosaicism throughout human preimplantation development in vitro: incidence, type, and relevance to embryo outcome. Hum Reprod 2002;17:413 –419. Blockeel C, Schutyser V, De Vos A, Verpoest W, De Vos M, Staessen C, Haentjens P, Van der Elst J, Devroey P. Prospectively randomized controlled trial of PGS in IVF/ICSI patients with poor implantation. Reprod Biomed Online 2008;17:848 –854. Cohen J, Munne S. Comment 2 on Staessen et al. (2004). Two-cell biopsy and PGD pregnancy outcome. Hum Reprod 2005;20:2363 –2364. Cohen J, Grifo JA. Multicentre trial of preimplantation genetic screening reported in the New England Journal of Medicine: an in-depth look at the findings. Reprod Biomed Online 2007;15:365 –366. Cohen J, Wells D, Munne S. Removal of 2 cells from cleavage stage embryos is likely to reduce the efficacy of chromosomal tests that are used to enhance implantation rates. Fertil Steril 2007;87:496– 503. Colls P, Escudero T, Cekleniak N, Sadowy S, Cohen J, Munne S. Increased efficiency of preimplantation genetic diagnosis for infertility using ‘no result rescue’. Fertil Steril 2007;88:53 –61. Colls P, Goodall N, Zheng X, Munne S. Increased efficiency of preimplantation genetic diagnosis for aneuploidy by testing 12 chromosomes. Reprod Biomed Online 2009;19:532 –538. De Vos A, Van Steirteghem A. Aspects of biopsy procedures prior to preimplantation genetic diagnosis. Prenat Diagn 2001;21:767 –780. De Vos A, Staessen C, De Rycke M, Verpoest W, Haentjens P, Devroey P, Liebaers I, Van de V. Impact of cleavage-stage embryo biopsy in view of PGD on human blastocyst implantation: a prospective cohort of single embryo transfers. Hum Reprod 2009;24:2988– 2996. Debrock S, Melotte C, Spiessens C, Peeraer K, Vanneste E, Meeuwis L, Meuleman C, Frijns JP, Vermeesch JR, D’Hooghe TM. Preimplantation genetic screening for aneuploidy of embryos after in vitro fertilization in women aged at least 35 years: a prospective randomized trial. Fertil Steril 2009. advance access published 25 February 2009. PMID: 19249029. Delhanty JD, Griffin DK, Handyside AH, Harper J, Atkinson GH, Pieters MH, Winston RM. Detection of aneuploidy and chromosomal mosaicism in human embryos during preimplantation sex determination by fluorescent in situ hybridisation, (FISH). Hum Mol Genet 1993;2:1183 –1185. Deugarte CM, Li M, Surrey M, Danzer H, Hill D, Decherney AH. Accuracy of FISH Analysis in Predicting Chromosomal Status in Patients Undergoing Preimplantation Genetic Diagnosis. Fertil Steril 2008;90:1049 –1054. Dokras A, Sargent IL, Ross C, Gardner RL, Barlow DH. Trophectoderm biopsy in human blastocysts. Hum Reprod 1990;5:821–825.

12 van Echten-Arends J, Mastenbroek S, Sikkema-Raddatz B, Korevaar JC, Heineman MJ, van der Veen F, Repping S. Chromosomal mosaicism in human preimplantation embryos: a systematic review. Hum Reprod 2011;in press. Edgar DH, Bourne H, Jericho H, McBain JC. The developmental potential of cryopreserved human embryos. Mol Cell Endocrinol 2000a;169:69– 72. Edgar DH, Bourne H, Speirs AL, McBain JC. A quantitative analysis of the impact of cryopreservation on the implantation potential of human early cleavage stage embryos. Hum Reprod 2000b;15:175–179. El-Toukhy T, Kamal A, Wharf E, Grace J, Bolton V, Khalaf Y, Braude P. Reduction of the multiple pregnancy rate in a preimplantation genetic diagnosis programme after introduction of single blastocyst transfer and cryopreservation of blastocysts biopsied on day 3. Hum Reprod 2009;24:2642 –2648. Geraedts J. O-042 Results from the polar body proof of principle study. Hum Reprod 2010;25:i17 –i18. Geraedts J, Collins J, Gianaroli L, Goossens V, Handyside A, Harper J, Montag M, Repping S, Schmutzler A. What next for preimplantation genetic screening? A polar body approach! Hum Reprod 2009;25:575 –577. Gianaroli L, Magli MC, Munne S, Fiorentino A, Montanaro N, Ferraretti AP. Will preimplantation genetic diagnosis assist patients with a poor prognosis to achieve pregnancy? Hum Reprod 1997;12:1762– 1767. Gianaroli L, Magli MC, Ferraretti AP, Munne S. Preimplantation diagnosis for aneuploidies in patients undergoing in vitro fertilization with a poor prognosis: identification of the categories for which it should be proposed. Fertil Steril 1999;72:837 –844. Goossens V, Harton G, Moutou C, Scriven PN, Traeger-Synodinos J, Sermon K, Harper JC. ESHRE PGD Consortium data collection VIII: cycles from January to December 2005 with pregnancy follow-up to October 2006. Hum Reprod 2008a;23:2629 –2645. Goossens V, De Rycke M, De Vos A, Staessen C, Michiels A, Verpoest W, Van Steirteghem A, Bertrand C, Liebaers I, Devroey P et al. Diagnostic efficiency, embryonic development and clinical outcome after the biopsy of one or two blastomeres for preimplantation genetic diagnosis. Hum Reprod 2008b;23:481–492. Goossens V, Harton G, Moutou C, Traeger-Synodinos J, Van Rij M, Harper JC. ESHRE PGD Consortium data collection IX: cycles from January to December 2006 with pregnancy follow-up to October 2007. Hum Reprod 2009; 24:1786– 1810. Griffin DK, Wilton LJ, Handyside AH, Winston RM, Delhanty JD. Dual fluorescent in situ hybridisation for simultaneous detection of X and Y chromosome-specific probes for the sexing of human preimplantation embryonic nuclei. Hum Genet 1992;89:18– 22. Handyside AH, Thornhill AR. In vitro fertilization with preimplantation genetic screening. N Engl J Med 2007;357:1770 –1770. Handyside AH, Kontogianni EH, Hardy K, Winston RM. Pregnancies from biopsied human preimplantation embryos sexed by Y-specific DNA amplification. Nature 1990;344:768– 770. Hardarson T, Hanson C, Lundin K, Hillensjo T, Nilsson L, Stevic J, Reismer E, Borg K, Wikland M, Bergh C. Preimplantation genetic screening in women of advanced maternal age caused a decrease in clinical pregnancy rate: a randomized controlled trial. Hum Reprod 2008;23:2806– 2812. Hardy K, Handyside AH. Cell allocation in twin half mouse embryos bisected at the 8-cell stage: implications for preimplantation diagnosis. Mol Reprod Dev 1993; 36:16 –22. Hardy K, Martin KL, Leese HJ, Winston RM, Handyside AH. Human preimplantation development in vitro is not adversely affected by biopsy at the 8-cell stage. Hum Reprod 1990;5:708 –714. Hassold T, Hunt P. To err (meiotically) is human: the genesis of human aneuploidy. Nat Rev Genet 2001;2:280– 291. Hellani A, Abu-Amero K, Azouri J, El-Akoum S. Successful pregnancies after application of array-comparative genomic hybridization in PGS-aneuploidy screening. Reprod Biomed Online 2008;17:841 –847. Jansen RP, Bowman MC, de Boer KA, Leigh DA, Lieberman DB, McArthur SJ. What next for preimplantation genetic screening (PGS)? Experience with blastocyst biopsy and testing for aneuploidy. Hum Reprod 2008;23:1476 –1478. Kahraman S, Benkhalifa M, Donmez E, Biricik A, Sertyel S, Findikli N, Berkil H. The results of aneuploidy screening in 276 couples undergoing assisted reproductive techniques. Prenat Diagn 2004;24:307 –311. Keskintepe L, Sher G, Machnicka A, Tortoriello D, Bayrak A, Fisch J, Agca Y. Vitrification of human embryos subjected to blastomere biopsy for

Mastenbroek et al.

pre-implantation genetic screening produces higher survival and pregnancy rates than slow freezing. J Assist Reprod Genet 2009. Advance Access published 5 December 2009. PMID: 19967555. Kuliev A, Verlinsky Y. Meiotic and mitotic nondisjunction: lessons from preimplantation genetic diagnosis. Hum Reprod Update 2004;10:401– 407. Kuwayama M. Highly efficient vitrification for cryopreservation of human oocytes and embryos: the Cryotop method. Theriogenology 2007;67:73 –80. Lintsen AM, Eijkemans MJ, Hunault CC, Bouwmans CA, Hakkaart L, Habbema JD, Braat DD. Predicting ongoing pregnancy chances after IVF and ICSI: a national prospective study. Hum Reprod 2007;22:2455 –2462. Los FJ, Van Opstal D, van den Berg C. The development of cytogenetically normal, abnormal and mosaic embryos: a theoretical model. Hum Reprod Update 2004; 10:79 –94. Magli MC, Gianaroli L, Grieco N, Cefalu E, Ruvolo G, Ferraretti AP. Cryopreservation of biopsied embryos at the blastocyst stage. Hum Reprod 2006;21:2656 –2660. Magli MC, Gianaroli L, Ferraretti AP, Lappi M, Ruberti A, Farfalli V. Embryo morphology and development are dependent on the chromosomal complement. Fertil Steril 2007;87:534 –541. Magli MC, Gianaroli L, Ferraretti AP, Gordts S, Fredericks V, Crippa A. Paternal contribution to aneuploidy in preimplantation embryos. Reprod Biomed Online 2009;18:536 –542. Marquez C, Sandalinas M, Bahce M, Alikani M, Munne S. Chromosome abnormalities in 1255 cleavage-stage human embryos. Reprod Biomed Online 2000;1:17 –26. Mastenbroek S, Bossuyt PMM, Heineman MJ, Repping S, van der Veen F. Comment 1 on Staessen et al. (2004). Design and analysis of a randomized controlled trial studying preimplantation genetic screening. Hum Reprod 2005;20:2362– 2363. Mastenbroek S, van der Veen F, Repping S. In vitro fertilization with preimplantation genetic screening—the authors reply. N Engl J Med 2007a;357:1769 –1771. Mastenbroek S, Twisk M, van Echten-Arends J, Sikkema-Raddatz B, Korevaar JC, Verhoeve HR, Vogel NE, Arts EG, de Vries JW, Bossuyt PM et al. In vitro fertilization with preimplantation genetic screening. N Engl J Med 2007b; 357:9 – 17. Mersereau JE, Pergament E, Zhang X, Milad MP. Preimplantation genetic screening to improve in vitro fertilization pregnancy rates: a prospective randomized controlled trial. Fertil Steril 2008;90:1287 –1289. Meyer LR, Klipstein S, Hazlett WD, Nasta T, Mangan P, Karande VC. A prospective randomized controlled trial of preimplantation genetic screening in the ‘good prognosis’ patient. Fertil Steril 2009;91:1731 –1738. Michiels A, Van Assche E, Liebaers I, Van Steirteghem A, Staessen C. The analysis of one or two blastomeres for PGD using fluorescence in-situ hybridization. Hum Reprod 2006;21:2396 –2402. Montag M, van der V, Rosing B, van der V. Polar body biopsy: a viable alternative to preimplantation genetic diagnosis and screening. Reprod Biomed Online 2009; 18:6 –11. Munne S, Lee A, Rosenwaks Z, Grifo J, Cohen J. Diagnosis of major chromosome aneuploidies in human preimplantation embryos. Hum Reprod 1993;8:2185–2191. Munne S, Alikani M, Tomkin G, Grifo J, Cohen J. Embryo morphology, developmental rates, and maternal age are correlated with chromosome abnormalities. Fertil Steril 1995;64:382 –391. Munne S, Sandalinas M, Escudero T, Velilla E, Walmsley R, Sadowy S, Cohen J, Sable D. Improved implantation after preimplantation genetic diagnosis of aneuploidy. Reprod Biomed Online 2003;7:91 –97. Munne S, Sandalinas M, Magli C, Gianaroli L, Cohen J, Warburton D. Increased rate of aneuploid embryos in young women with previous aneuploid conceptions. Prenat Diagn 2004;24:638– 643. Munne S, Gianaroli L, Tur-Kaspa I, Magli C, Sandalinas M, Grifo J, Cram D, Kahraman S, Verlinsky Y, Simpson JL. Substandard application of preimplantation genetic screening may interfere with its clinical success. Fertil Steril 2007;88:781 –784. Munne S, Wells D, Cohen J. Technology requirements for preimplantation genetic diagnosis to improve assisted reproduction outcomes. Fertil Steril 2010; 94:408– 430. Nicolaidis P, Petersen MB. Origin and mechanisms of non-disjunction in human autosomal trisomies. Hum Reprod 1998;13:313 –319. Northrop LE, Treff NR, Levy B, Scott RT Jr. SNP microarray-based 24 chromosome aneuploidy screening demonstrates that cleavage-stage FISH poorly predicts

Preimplantation genetic screening and live birth rate

aneuploidy in embryos that develop to morphologically normal blastocysts. Mol Hum Reprod 2010;16:590– 600. Platteau P, Staessen C, Michiels A, Van Steirteghem A, Liebaers I, Devroey P. Preimplantation genetic diagnosis for aneuploidy screening in patients with unexplained recurrent miscarriages. Fertil Steril 2005;83:393– 397. Ruangvutilert P, Delhanty JD, Rodeck CH, Harper JC. Relative efficiency of FISH on metaphase and interphase nuclei from non-mosaic trisomic or triploid fibroblast cultures. Prenat Diagn 2000;20:159 –162. Rubio C, Pehlivan T, Rodrigo L, Simon C, Remohi J, Pellicer A. Embryo aneuploidy screening for unexplained recurrent miscarriage: a minireview. Am J Reprod Immunol 2005;53:159–165. Schoolcraft WB, Katz-Jaffe MG, Stevens J, Rawlins M, Munne S. Preimplantation aneuploidy testing for infertile patients of advanced maternal age: a randomized prospective trial. Fertil Steril 2009a;92:157– 162. Schoolcraft WB, Fragouli E, Stevens J, Munne S, Katz-Jaffe MG, Wells D. Clinical application of comprehensive chromosomal screening at the blastocyst stage. Fertil Steril 2009b. Advance access published 23 November 2009. PMID: 19939370. Scriven PN, Bossuyt PM. Diagnostic accuracy: theoretical models for preimplantation genetic testing of a single nucleus using the fluorescence in situ hybridization technique. Hum Reprod 2010;25:2622– 2628. Sermon K, Van Steirteghem A, Liebaers I. Preimplantation genetic diagnosis. Lancet 2004;363:1633 –1641. Sermon KD, Michiels A, Harton G, Moutou C, Repping S, Scriven PN, SenGupta S, Traeger-Synodinos J, Vesela K, Viville S et al. ESHRE PGD Consortium data collection VI: cycles from January to December 2003 with pregnancy follow-up to October 2004. Hum Reprod 2007;22:323 –336. Silber S, Escudero T, Lenahan K, Abdelhadi I, Kilani Z, Munne S. Chromosomal abnormalities in embryos derived from testicular sperm extraction. Fertil Steril 2003;79:30 –38. Simpson JL. Randomized clinical trial in assessing PGS: necessary but not sufficient. Hum Reprod 2008;23:2179– 2181. Stachecki JJ, Cohen J, Munne S. Cryopreservation of biopsied cleavage stage human embryos. Reprod Biomed Online 2005;11:711 –715. Staessen C, Platteau P, Van Assche E, Michiels A, Tournaye H, Camus M, Devroey P, Liebaers I, Van Steirteghem A. Comparison of blastocyst transfer with or without preimplantation genetic diagnosis for aneuploidy screening in couples with advanced maternal age: a prospective randomized controlled trial. Hum Reprod 2004;19:2849– 2858. Staessen C, Verpoest W, Donoso P, Haentjens P, Van der Elst J, Liebaers I, Devroey P. Preimplantation genetic screening does not improve delivery rate in women under the age of 36 following single-embryo transfer. Hum Reprod 2008;23:2818 –2825. Steptoe PC, Edwards RG. Birth after the reimplantation of a human embryo. Lancet 1978;2:366–366. Stevens J, Wale P, Surrey ES, Schoolcraft WB, Gardner DK. Is aneuploidy screening for patients aged 35 or over beneficial? A prospective randomized trial. Fertil Steril 2004;82:S249 –S249.

13 Summers MC, Foland AD. Quantitative decision-making in preimplantation genetic (aneuploidy) screening (PGS). J Assist Reprod Genet 2009. Advance access published 22 October 2009. PMID: 19847639. Tarin JJ, Conaghan J, Winston RM, Handyside AH. Human embryo biopsy on the 2nd day after insemination for preimplantation diagnosis: removal of a quarter of embryo retards cleavage. Fertil Steril 1992;58:970– 976. Teramoto S, Kato O. Minimal ovarian stimulation with clomiphene citrate: a large-scale retrospective study. Reprod Biomed Online 2007;15:134 –148. 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– 589. Twisk M, Mastenbroek S, Hoek A, Heineman MJ, van der Veen F, Bossuyt PM, Repping S, Korevaar JC. No beneficial effect of preimplantation genetic screening in women of advanced maternal age with a high risk for embryonic aneuploidy. Hum Reprod 2008;23:2813 –2817. Vanneste E, Voet T, Melotte C, Debrock S, Sermon K, Staessen C, Liebaers I, Fryns JP, D’Hooghe T, Vermeesch JR. What next for preimplantation genetic screening? High mitotic chromosome instability rate provides the biological basis for the low success rate. Hum Reprod 2009a;24:2679– 2682. Vanneste E, Voet T, Le Caignec C, Ampe M, Konings P, Melotte C, Debrock S, Amyere M, Vikkula M, Schuit F et al. Chromosome instability is common in human cleavage-stage embryos. Nat Med 2009b;15:577 –583. Verlinsky Y, Cieslak J, Freidine M, Ivakhnenko V, Wolf G, Kovalinskaya L, White M, Lifchez A, Kaplan B, Moise J et al. Pregnancies following pre-conception diagnosis of common aneuploidies by fluorescent in-situ hybridization. Hum Reprod 1995; 10:1923– 1927. Verlinsky Y, Cohen J, Munne S, Gianaroli L, Simpson JL, Ferraretti AP, Kuliev A. Over a decade of experience with preimplantation genetic diagnosis: a multicenter report. Fertil Steril 2004;82:292– 294. Wells D, Alfarawati S, Fragouli E. Use of comprehensive chromosomal screening for embryo assessment: microarrays and CGH. Mol Hum Reprod 2008; 14:703– 710. Werlin L, Rodi I, DeCherney A, Marello E, Hill D, Munne S. Preimplantation genetic diagnosis as both a therapeutic and diagnostic tool in assisted reproductive technology. Fertil Steril 2003;80:467 –468. Wilding M, Forman R, Hogewind G, Di Matteo L, Zullo F, Cappiello F, Dale B. Preimplantation genetic diagnosis for the treatment of failed in vitro fertilizationembryo transfer and habitual abortion. Fertil Steril 2004;81:1302–1307. Wilton L. Preimplantation genetic diagnosis for aneuploidy screening in early human embryos: a review. Prenat Diagn 2002;22:512 –518. Wilton L, Williamson R, McBain J, Edgar D, Voullaire L. Birth of a healthy infant after preimplantation confirmation of euploidy by comparative genomic hybridization. N Engl J Med 2001;345:1537 –1541. Ziebe S, Lundin K, Loft A, Bergh C, Nyboe Andersen A, Selleskog U, Nielsen D, Grondahl C, Kim H et al. FISH analysis for chromosomes 13, 16, 18, 21, 22, X and Y in all blastomeres of IVF pre-embryos from 144 randomly selected donated human oocytes and impact on pre-embryo morphology. Hum Reprod 2003;18:2575 –2581.