Effect of Uterine Leiomyomas on Noninvasive Prenatal Testing Parameters in the First Trimester

Abstract

Uterine leiomyomas may affect the performance of cell-free DNA (cfDNA) based non-invasive prenatal testing (NIPT). We conducted a retrospective cohort study of pregnant individuals with and without leiomyomas undergoing first trimester cfDNA-based NIPT. NIPT characteristics in those with leiomyomas (n=122) were compared to those without (n=937). Mean fetal fraction (FF) was lower in those with leiomyomas compared to those without (10.0% vs. 11.5%; P=0.001), however the rate of indeterminate results was different only in patients without obesity (BMI<30 kg/m²) (5.3% vs 1.5%; P=0.03). Total cfDNA concentration was higher in those with leiomyomas (P=0.002), suggesting possible dilution of the FF. Leiomyoma size did not affect NIPT metrics. In conclusion, uterine leiomyomas are associated with lower FF and, in patients without obesity, with a higher rate of indeterminate results independent of leiomyoma size.

Precis

Uterine leiomyomas are associated with lower fetal fraction on NIPT and, in patients without obesity, with a higher rate of indeterminate results independent of leiomyoma size.

Introduction

Non-invasive prenatal testing (NIPT) is the preferred aneuploidy screening modality during pregnancy¹. NIPT’s accuracy is dependent on the fetal fraction (FF), the proportion of circulating cfDNA derived from the placenta, while the remainder of circulating cfDNA derives from maternal sources. Multiple maternal and fetal characteristics affect the FF including maternal body mass index (BMI), gestational age, fetal aneuploidy, autoimmune disease, anticoagulation, and neoplasms ^2-6. We sought to determine how the presence of leiomyomas affects NIPT metrics. We hypothesized that uterine leiomyomas are associated with a lower FF and a higher indeterminate result rate.

Methods

University of Washington IRB approval was obtained (STUDY00005540). Individuals with singleton pregnancies who completed our internally developed NIPT assay during the first trimester (10 0/7-14 0/7 weeks gestation) between 2017 and 2019 were included. Largest leiomyoma diameter was obtained from ultrasound reports as leiomyoma volume was not consistently documented. Patients without available imaging reports, with autoimmune disease, on anticoagulation therapy, or those with abnormal NIPT results (e.g., aneuploidy) were excluded.

Low pass whole genome sequencing NIPT was performed as previously described². The FF was calculated either by the percent of reads aligning to the Y chromosome, or, for female fetuses, a custom bioinformatic algorithm based on aggregate length distribution. An indeterminate result was defined as FF <4%.

The primary outcome was differences in FF and secondary outcomes included differences in the indeterminate result rate and total cfDNA concentration. Parametric and nonparametric univariate analyses were utilized as appropriate. Multivariable logistic regression was used to determine the risk of an indeterminate result based on leiomyoma status. Linear regression was used to evaluate the association between leiomyoma size and NIPT metrics. Prior observational data suggests that leiomyomas are associated with 2% lower FF⁷. To detect a 2% difference in the mean FF with an α = 0.05 and a power of 0.90, a cohort of 132 (66 in each group) was needed.

Results

A cohort of 1059 met inclusion criteria: 122 with uterine leiomyomas and 937 without (Appendix 1, available online at http://links.lww.com/xxx). Those with leiomyomas were older and more likely to undergo cesarean delivery, without differences in other demographics (Appendix 2, available online at http://links.lww.com/xxx).

Those with leiomyomas had a lower mean FF (10.0 ± 4.5% vs 11.5 ± 4.6%; P=0.001) and a higher total cfDNA concentration (98.7 vs 83.8 pg/uL; P=0.002) (Table 1). However, this did not result in a higher indeterminate result rate (5.7% vs 3.4%; P=0.200). Because elevated maternal body mass index (BMI) strongly associates with low FF and indeterminate results², we examined the same NIPT parameters in individuals without obesity (BMI <30 kg/m²). FF remained significantly different between groups as did total cfDNA concentration (Table 1). Indeterminate results were significantly more frequent in individuals without obesity who had leiomyomas compared to those without (5.3% vs 1.5%; P=0.030). Logistic regression adjusting for the gestational age at NIPT, maternal age, and fetal sex noted that the presence of leiomyomas was associated with an increased likelihood of an indeterminate result in the cohort without obesity (aOR 4.35; 95% CI 1.26-15.0; P=0.020).

Table 1:

Noninvasive Prenatal Testing Metrics in Those With and Without Leiomyomas

	Entire Cohort			No Obesity (BMI <30 kg/m2) Cohort
NIPT metrics	(−) Leiomyomas n = 937	(+) Leiomyomas n = 122	P-value	(−) Leiomyomas n = 711	(+) Leiomyomas n = 94	P-value
Fetal fraction (%)	11.5 ± 4.6	10.0 ± 4.5	0.001	12.2 ± 4.4	10.6 ± 4.5	0.001
Indeterminate result	32 (3.4%)	7 (5.7%)	0.200	11 (1.5%)	5 (5.3%)	0.030
Total cfDNA concentration (pg/uL)	83.8 [78.2–84.2]	98.7 [88.7–109.8]	0.002	82.2 [78.9–85.6]	98.7 [87.0–111.9]	0.003

Data represented as mean ± standard deviation, n (%), or geometric mean [95% confidence interval]

BMI, body mass index; cfDNA, cell-free DNA.

To determine whether the observed differences in NIPT metrics correlated with leiomyoma size, the leiomyoma group (n=122) was dichotomized using a diameter cut off of 5 cm, chosen given its association with adverse obstetric outcomes⁸. There were no differences in test metrics based on leiomyoma size (Appendix 3, available online at http://links.lww.com/xxx) and no correlation with FF or cfDNA concentration (both P>0.1; Figure 1A and 1B). Although excluded from the primary analysis, we summarize the types of abnormal NIPT results observed in those with leiomyomas in Appendices 4 and 5, available online at http://links.lww.com/xxx.

Figure 1:

Noninvasive prenatal testing (NIPT) metrics and leiomyoma diameter. A. Relationship between NIPT fetal fraction and largest leiomyoma diameter (P=.33). B. Relationship between NIPT total cell-free DNA (cfDNA) concentration and largest leiomyoma diameter (P=.14).

Discussion

The presence of uterine leiomyomas in the first trimester is associated with lower FF and higher total cfDNA concentrations, independent of leiomyoma size; and in those with a BMI <30 kg/m², a higher indeterminate result rate.

An Australian study demonstrated that leiomyomas were associated with lower FF although they excluded those with very low FF and indeterminate results⁷. Our findings are concordant and further reveal that leiomyomas are also associated with elevated total cfDNA concentration suggesting a dilutional effect on the FF due to an increase in contribution from maternal sources. The additional circulating maternal cfDNA may derive from mitotically active leiomyomas, which if true, is independent of leiomyoma size. Alternatively, the lower FF may represent an inherent difference in trophoblast activity in those with leiomyoma uteri. Further investigation into the mechanisms of cfDNA release and clearance is needed.

In summary, our study demonstrates that uterine leiomyomas are associated with altered NIPT metrics. As uterine leiomyomas are present in approximately 10% of pregnant individuals, clinicians, geneticists, and NIPT laboratories should be aware of the association of leiomyomas with altered NIPT metrics, specifically lower FF, independent of leiomyoma size.