Does Body Mass Index Affect Embryonic Mosaicism?

Abstract

Purpose: To examine the association between body mass index (BMI) and embryonic aneuploidy and mosaicism by trophectoderm biopsy with preimplantation genetic testing for aneuploidy (PGT-A).

Methods: This was a retrospective cohort study evaluating 602 women who underwent undergoing their first fresh autologous stimulation cycles wherein all embryos were biopsied for PGT-A analysis from January 2016 through December 2019 in a multicenter academic fertility practice. Associations between BMI (kg/m²) and rates of euploidy (2016-2019) and mosaicism (2018-2019), including low-level (LL), high-level (HL), whole chromosome, and segmental mosaicism, were determined using forward stepwise multivariable Poisson regression.

Results: 3,020 embryos from 602 cycles were available for analysis. Most patients were of normal weight (N=323) or overweight (N=143), or had class I obesity (N=80). Overall, 21% of patients in the cohort had obesity (BMI >30). There was no difference in overall euploid or mosaicism rate stratified by BMI. Compared with normal BMI, underweight (adjusted relative risk, aRR: 3.57 [1.31-9.78]), overweight (1.57 [1.01-2.43]), and obesity (1.55[1.03-2.35]) were each associated with significantly increased risks of LL mosaicism (6.6% vs. 20.8%, 10.0%, and 9.1%, respectively). Overweight BMI was associated with lower rates of HL (8.3% vs 3.7%, aRR: 0.39 [0.21-0.75]) and whole chromosome mosaicism (3.7% vs. 7.3%, aRR: 0.48 [0.26-0.91]) compared to normal BMI.

Conclusions: While the overall rate of mosaicism was similar among BMI categories, both low and high BMI were associated with increased rates of LL mosaicism, and overweight was associated with decreased rates of HL and whole chromosome mosaicism compared to normal BMI. 

Introduction

Overweight and obesity are increasingly common conditions in the United States. From 2000 through 2018, the prevalence of class I and II obesity (BMI 30–39 kg/m²) increased from 30–42%, and the prevalence of class III obesity (BMI ≥ 40 kg/m²) increased from 4.7–9.2% [1]. Obesity-related conditions include heart disease, stroke, type 2 diabetes mellitus, and certain types of cancer, which are many of the leading causes of preventable, premature death.

Most studies have linked obesity with unfavorable assisted reproductive technology (ART) outcomes [2]. Patients with obesity undergoing IVF require higher doses of gonadotropins, have fewer oocytes collected, more canceled cycles (for low or high response), higher miscarriage rates, and reduced pregnancy and live-birth rates. Data suggest that the detrimental effects of obesity are mitigated with the use of donor oocytes [3], which supports the theory that oocytes are directly affected by elevated BMI. However, prior research has not found a relationship between BMI and embryonic euploidy [4].

Mosaicism is defined as the co-presence of cells with two (or more) different chromosomal constitutions and typically results from mitotic error in the postzygotic state of development [5]. Embryonic mosaicism was initially detected in the 1990s, though testing technologies were limited. Now with the use of high-throughput methods and detailed resolution, the detection of mosaicism is routinely reported. Next-generation sequencing (NGS) has the capability of detecting mosaicism levels as low as 20%. The reported incidence of mosaicism detected by PGT-A is widespread in the literature, with most recent estimates ranging from 4–22% (mean 15%) [6]. While previous studies have demonstrated that events such as anaphase lag and cell-cycle dysregulation may influence development of mosaicism, the causes and significance of embryonic mosaicism remain somewhat elusive.

Previous studies have demonstrated that maternal metabolic disorders can affect oocyte mitochondrial function and expression of cell cycle regulator genes, as well as genes involved in chromosome alignment and/or separation [7–8]. This has been found to lead to anomalies in chromatid separation and alignment, which may play a role in mosaicism that occurs in cleavage stage embryos prior to total zygotic genome activiation. Our hypothesis, therefore, was that maternal factors such as body weight (measured by BMI) may influence the processes that impact chromosome separation during post-fertilization mitosis, increasing the rate of mosaicism. To date, no study has evaluated the effect of BMI on embryonic mosaicism.

Material And Methods

Ethical approval

The study was reviewed by the Institutional Review Board at Northwell Health and was determined to be exempt.

Patients, study design and study groups

This was a retrospective cohort study of women who underwent their first fresh autologous stimulation cycles wherein all embryos were biopsied for PGT-A analysis between January 2016 and December 2019. BMI (kg/m²) was recorded prior to cycle start and patients were stratified into World Health Organization (WHO) BMI categories: underweight (BMI <18.5), normal weight (BMI 18.5-24.9), overweight (BMI 25-29.9), or obesity class I (BMI 30-34.9), II (BMI >35-39.9), or III (BMI >40). Patients were excluded if the ovarian stimulation was cancelled for any reason, if there were no embryos available for trophectoderm biopsy, or if a fresh embryo transfer was performed.

Patients were stimulated according to protocols determined by their individual providers, which included GnRH antagonist protocols, microdose Lupron protocols [9], and estrogen priming protocols [10], as described in the literature. Patients received either human chorionic gonadotropin, leuprolide acetate, or a combination trigger for final oocyte maturation, based on provider discretion of baseline characteristics and response to controlled ovarian stimulation. An oocyte retrieval was performed 35-37 hours after trigger. Oocytes were injected with sperm using intracytoplasmic sperm injection (ICSI). Embryos were analyzed on day one following oocyte retrieval for normal fertilization and again on days two and three, per laboratory protocol. Laser-assisted breaching of the zona pellucida was performed on embryos that met criteria at the cleavage-stage. Embryos were examined on day five and six (if applicable) for biopsy eligibility.

Embryo biopsy and PGT-A analysis

Embryos that met biopsy criteria were biopsied on days 5 and/or 6. Embryos were vitrified after biopsy, and the biopsy specimen was sent to a single laboratory for analysis. PGT-A was performed using array comparative genomic hybridization (aCGH) (2016 and 2017) or next generation sequencing (NGS) (2018 and 2019, during which mosaicism was additionally reported). Mosaicism was defined as an abnormality detected in 20-80% of cells, in line with the Preimplantation Genetic Diagnosis International Society (PGDIS) guidelines [11]. The PGT laboratory defined embryos with 20-40% mosaic cells to be low-level mosaic, and >40-80% mosaic cells to be high-level mosaic. Segmental aneuploidy or mosaicism was defined as a duplication or deletion of a specific chromosome segment, but not the whole chromosome.

Statistical Analysis

Demographic variables were compared between the BMI categories. Wilcoxon rank sum test was used to compare the distribution of continuous variables between groups and Fisher’s exact test was used to compare categorical outcomes.

The associations between BMI and rates of euploidy and mosaicism, including low level (LL), high level (HL), whole chromosome, and segmental mosaicism, were determined using forward stepwise multivariable Poisson regressions, controlling a priori for age and AMH, and retaining other patient- and cycle-specific variables (e.g., diagnosis) that changed the effect estimates by >10%. For the regression analyses evaluating rates of euploidy, we assumed that LL mosaic embryos diagnosed in 2018 and 2019 would have been reported as euploid if tested using aCGH prior to mosaicism reporting. LL mosaic embryos were therefore re-classified as euploid for this analysis. A sensitivity analysis evaluating euploid rate was then performed restricting the data to embryos diagnosed using NGS only. Regression analyses evaluating rates of mosaicism used data from 2018-2019. Relative risks with 95% confidence intervals (RR [CI]) were determined, using normal BMI as the reference value.

Results

Three thousand twenty diagnosed embryos were available for analysis from 602 patients who underwent their first fresh autologous IVF cycle with PGT-A by aCGH or NGS. Patients were stratified into BMI category: underweight (N=10; 47 embryos), normal weight (N=323; 1,677 embryos), overweight (N=143; 672 embryos), obesity class I (N=80; 407 embryos), II (N=33; 154 embryos), or III (N=13; 63 embryos). Overall, 21% of patients in the cohort had obesity (N=126). The average patient age was 35.9 years, and age and BMI were positively correlated (spearman rank correlation coefficient = 0.2, p<0.01). There were no statistically significant differences in serum baseline FSH, AMH, infertility diagnosis, number of oocytes retrieved, number of 2PN embryos, or number of blastocysts biopsied or diagnosed between the WHO BMI classes (Table I).

The percentage of embryos diagnosed as euploid was 54.5% for the entire cohort. There were no statistically significantly differences in euploidy rate by BMI category when considering the entire cohort (aCGH + NGS tested embryos) as well as for NGS-tested embryos only. The rates of embryonic euploidy by BMI class for the full cohort are depicted in Figure I.

The overall rate of mosaicism for patients with elevated BMI in our cohort ranged from 10.7%-16.1%. There was no significant difference in overall mosaicism rate stratified by BMI, before and after adjusting for age, AMH, and a diagnosis of recurrent pregnancy loss (RPL), which remained a significant variable in the stepwise model (Table II).

The average low-level mosaicism rate for patients with obesity (BMI classes I-III) was 9.1% (38/419) and the high-level mosaicism rate was 5.5% (23/419). Compared with normal BMI, underweight (aRR: 3.57 [1.31-9.78]) and overweight (1.57 [1.01-2.43]) were both associated with significantly higher risks of LL mosaicism (6.6% vs. 20.8% and 10.0%, respectively) (Table III). Obesity (BMI>30) was also associated with a statistically significantly higher risk of LL mosaicism compared with normal BMI (9.1% vs. 6.6%; aRR: 1.55 [1.03-2.35]), though comparisons with individual obesity classes I-III did not reach the thresholds for statistical significance. Compared with normal BMI, overweight was associated with a lower risk of HL mosaicism (8.3% vs 3.7%, aRR: 0.39 [0.21-0.75]).

There were no statistically significant differences in segmental mosaicism among groups. Patients with overweight BMI had lower rates of whole chromosome mosaicism as compared to those with normal BMI (3.7% vs. 7.3%, respectively; aRR: 0.48 [0.26-0.91]) (Table III).

Discussion

Our results suggest that while the overall rate of mosaicism does not differ by BMI category, both low and high BMI may be associated with increased risks of LL mosaicism compared with normal BMI, and overweight may be associated with decreased rates of HL and whole chromosome mosaicism. These findings indicate that body weight may influence the early embryonic mechanisms associated with mosaicism.

There are various mechanisms thought to contribute to oocyte abnormalities in patients with elevated BMI, leading to poorer reproductive outcomes. For example, a study by Robker, et al. [12] found increased follicular concentrations of insulin, triglycerides and inflammatory markers in obese women, which may lead to impaired folliculogenesis and oocyte competence. Abnormalities of oxidative balance and steroidogenesis have also been observed in patients with obesity [13]. However, whether BMI impacts the chromosomal complement of oocytes is an avenue of ongoing investigation.

A study by Goldman, et al. [4], demonstrated similar findings to our results, in which no significant relationship between BMI and the number or percentage of euploid embryos was found. Their study evaluated 279 women who underwent IVF with PGT using aCGH between 2010 and 2013 and was limited by sample size, as there were only 50 patients with overweight and 22 patients with obese BMIs. Additionally, mosaicism was not reported over that study’s timeframe. To date, there are no published studies evaluating whether BMI affects euploid or mosaicism rates using embryos subjected to testing by NGS.

Importantly, our analysis of euploid rate by BMI category used re-classified embryos that were originally designated as “LL mosaic” to “euploid” (2018-2019 data). This assumption was made because prior to mosaicism reporting, those embryos would have been routinely classified as euploid. A sensitivity analysis evaluating the true “euploid” rate in the subset of NGS-tested embryos (i.e., without re-classification of LL mosaic to euploid) demonstrated similar findings.

While whole chromosome mosaicism of embryos arises from mitotic errors, literature suggests that segmental aneuploidies may arise from mechanisms associated with genomic instability [14]. Our results demonstrate differences in rates of whole chromosome mosaicism by BMI category, but did not show any difference in the rate of segmental mosaicism between groups. This supports our hypothesis that BMI may affect factors in the oocyte that have an effect on chromatid separation and alignment more so than those impacting genomic stability.

To our knowledge, this is the first study to report on BMI and embryonic mosaicism. We identified a novel finding that both low and high BMI were associated with increased risks of LL mosaicism. An additional strength of our study is the patient population, with a relatively high proportion of patients with overweight (24%) or obesity (21%). This study used strict inclusion criteria, including only patients’ first fresh autologous cycles wherein all eligible embryos were biopsied for PGT-A testing. PGT-A testing was completed at a single laboratory, minimizing potential confounders from differences in testing technique or interpretation.

Our study is limited due to its retrospective nature with a single patient population. Despite having a larger study population than previous studies, there were still relatively few patients with underweight BMI, obesity class II and obesity class III. While obesity was associated with increased risks of LL mosaicism, individual obesity classes demonstrated non-statistically significant increases in the risks of LL mosaicism, likely due to limited sample size. Additionally, there was a change in technology from aCGH to NGS during our study timeframe, but this change affected all study subjects equally. There are inherent limitations to using BMI as an exposure, as BMI is an indirect and imperfect measurement, and fails to distinguish between body fat and lean body mass. However, BMI is easily calculable from height and weight, which are routinely recorded at patient visits, whereas waist circumference is not; BMI is therefore widely used as a health measurement in both clinic and community settings, as well as in large research studies.

As with any study reporting on mosaicism by PGT-A, it is important to consider the reproducibility of mosaic results by re-biopsy, or inner cell mass (ICM) biopsy. Mosaicism is often due to mitotic non-disjunction events and are not necessarily uniformly distributed throughout the blastocyst. A recent study investigated 32 blastocysts and calculated per chromosome concordances between trophectoderm biopsy and ICM biopsy, and found concordance rates of 99.5% for euploid results, 97.3% for aneuploid results, but only 35.2% for mosaic results [15]. Future research which incorporates re-biopsy data would strengthen our findings. Improvements in PGT-A technology to reduce the occurrence of technical artifacts that may contribute to erroneous mosaicism reporting would also allow for more refined results. Importantly, as embryonic euploidy rates may vary by fertility practice [16], and a prior study using cleavage-stage embryos and fluorescence in situ hybridization suggested that mosaicism rate may similarly vary by infertility center [17], we encourage repeat investigation with a distinct study population to further evaluate the relationship between BMI and mosaicism.

This is the first study to examine the effect of BMI on embryo mosaicism. Our findings support previous research demonstrating a lack of evidence that BMI is associated with the rate of embryonic euploidy. We introduce the possibility that a patient’s BMI may play a role in the rate of embryonic mosaicism following IVF with PGT-A. An increase in the risk of low-level mosaicism for patients with obesity versus normal BMI may help explain unfavorable ART outcomes demonstrated by these patients. Though a large proportion of our patient cohort had elevated BMI, further research is needed to clarify the impact of BMI on embryo mosaicism and to elucidate possible mechanisms.

Declarations

Funding: No funds, grants, or other support was received.

Conflicts of interest/Competing interests: The authors have no relevant financial or non-financial interests to disclose.

Ethics approval: The study was reviewed by the Institutional Review Board at Northwell Health and was determined to be exempt.

Consent to participate: Not applicable

Consent for publication: Not applicable

Availability of data and material: The datasets used analyzed during the current study are available from the corresponding author on reasonable request.

Code availability: Not applicable

Authorship statement: All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Baruch Abittan, Robyn Frankel, and Randi Goldman. The first draft of the manuscript was written by Baruch Abittan and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

References

1. Hales CM, Carroll MD, Fryar CD, Ogden CL. Prevalence of obesity and severe obesity among adults: United States, 2017–2018. Hyattsville, MD: National Center for Health Statistics. February 2020. NCHS Data Brief, no 360.
2. Gambineri A, Laudisio D, Marocco C, et al. Female infertility: which role for obesity? Int J Obes Suppl. 2019; 9:65-72.
3. Jungheim ES, Schon SB, Schulte MB, DeUgarte DA, Fowler SA, Tuuli MG. IVF outcomes in obese donor oocyte recipients: a systematic review and metaanalysis. Hum Reprod 2013; 28:2720–7.
4. Goldman KN, Hodes-Wertz B, McCulloh DH, Flom JD, Grifo JA. Association of body mass index with embryonic aneuploidy. Fertil Steril. 2015; 103(3):744-8.
5. Viotti M. Preimplantation Genetic Testing for Chromosomal Abnormalities: Aneuploidy, Mosaicism, and Structural Rearrangements. Genes (Basel). 2020; 11(6):602.
6. Marin D, Scott Jr. RT, Treff NR. Preimplantation embryonic mosaicism: origin, consequences and the reliability of comprehensive chromosome screening. Curr Opin Obstet Gynecol. 2017; 29: 168-17.
7. Wood, J., Dumesic, D., Abbott, D., Strauss, III., J. Molecular Abnormalities in Oocytes from Women with Polycystic Ovary Syndrome Revealed by Microarray Analysis. The Journal of Clinical Endocrinology & Metabolism. 2007; 92(2): 705-713.
8. Snider, A., Wood, J. Obesity induces ovarian inflamaation and reduces oocyte quality. Reproduction. 2019; 158: R79-R90.
9. Scott RT, Navot D. Enhancement of ovarian responsiveness with microdoses of gonadotropin-releasing hormone agonist during ovulation induction for in vitro fertilization. Fertil Steril. 1994; 61(5):880-5.
10. Dragisic KG, Davis OK, Fasouliotis SJ, Rosenwaks Z. Use of a luteal estradiol patch and a gonadotropin-releasing hormone antagonist suppression protocol before gonadotropin stimulation for in vitro fertilization in poor responders. Fertil Steril. 2005; 84 (4):1023-6.
11. Cram DS, Leigh D, Handyside A, Rechitsky L, Xu K, Harton G, et al. PGDIS Position Statement on the Transfer of Mosaic Embryos 2019. Reprod Biomed Online. 2019; 39(Suppl 1):e1-e4.
12. Robker RL, Akison LK, Bennett BD, Thrupp PN, Chura LR, Russell DL, Lane M, Norman RJ. Obese women exhibit differences in ovarian metabolites, hormones, and gene expression compared with moderate-weight women. J Clin Endocrinol Metab. 2009; 94:1533–1540.
13. Broughton DE, Moley KH. Obesity and female infertility: potential mediators of obesity’s impact. Fertil Steril. 2017; 107 :840–884.
14. Escribà MJ, Vendrell X, Peinado V. Segmental aneuploidy in human blastocysts: a qualitative and quantitative overview. Reprod Biol Endocrinol. 2019; 17:76.
15. Sachdev N.M., McCulloh D.H., Kramer Y., Keefe D., Grifo J.A. The reproducibility of trophectoderm biopsies in euploid, aneuploid, and mosaic embryos using independently verified next-generation sequencing (NGS): A pilot study. J. Assist. Reprod. Genet. 2020; 37:559–571
16. Munné S, Alikani M, Ribustello L, Colls P, Martínez-Ortiz PA, McCulloh DH; Referring Physician Group. Euploidy rates in donor egg cycles significantly differ between fertility centers. Hum Reprod. 2017; 32(4):743-749.
17. Munné S, Magli C, Adler A, Wright G, de Boer K, Mortimer D, et al. Treatment-related chromosome abnormalities in human embryos. Hum Reprod. 1997; 12(4):780-4.

Tables

Table I. Baseline patient characteristics and IVF stimulation outcomes (N=602)

a. Polycystic ovarian syndrome, b. Diminished ovarian reserve, c. Recurrent pregnancy loss 

Table II. Overall mosaicism rates by World Health Organization BMI class

Table III. Rates of low-level, high-level, segmental and whole-chromosome mosaicism by WHO BMI class