ABSTRACT Several recent studies have suggested an association between children conceived by assisted reproductive technology (ART) and a distinct growth pattern in early life, which has been linked to diseases of older age. Telomere shortening, specifically a shorter leukocyte telomere length (LTL), has also been linked to age-related diseases including cardiometabolic diseases and cancer. Initial telomere length is shaped by a telomere reset process during gamete fertilization and the early stages of preimplantation development. Given that ART involves the in vitro manipulation of oocytes and embryos, this study aimed to evaluate the association between ART-related factors and LTL in children. Whole genome-sequencing (WGS) data from the Nanjing and Suzhou centers of the China National Birth Cohort (CNBC) were obtained from 1137 individuals from 365 parent-children families, including 202 children conceived using ART and 205 conceived spontaneously. The association was determined between blastocyst-stage transfer and shorter telomere length in 180 children conceived by ART in the same centers. In addition, qPCR was used to perform validation on fingerstick blood samples from 406 children conceived using ART at 3 different centers in China. Data from qPCR were then compared with data obtained from WGS in 70 children in the discovery cohort with sufficient DNA samples. Associations between parental factors and demographics with LTL were examined in the discovery cohort. LTL attribution with aging was calculated using data from the discovery cohort, 1185 individuals aged 40–69 years from healthy controls in the Nanjing Lung Cancer Cohort with LTL measured using WGS, and 1452 East Asian adults aged 40–69 years in the UK biobank with LTL measured using qPCR. Results of association analyses found paternal LTL (β = 0.27, P = 3.26 × 10−8), maternal LTL (β = 0.26, P = 2.41 × 10−7), plurality (twins vs singletons; β = −0.34, P = 0.010), sex of children (male vs female; β = −0.22, P = 0.032), gestational age (β = 0.05, P = 0.026), and conception type (ART-conceived pregnancy vs spontaneously conceived, β = −0.35, P = 5.98 × 10−4) were all significantly associated with LTL in children. Multivariate regression analysis revealed that children conceived by ART had a significantly shorter LTL than those conceived spontaneously, even when adjusting for parental age at conception (adjusted β = −0.41, P = 3.33 × 10−4). Transfer of blastocyst-stage embryos was found to be significantly associated with shorter LTL (β = −0.54, P = 2.69 × 10−3). In addition, when LTL in children from spontaneous pregnancies was compared with LTL in children from ART pregnancies through cleavage-stage versus blastocyst-stage transfer, the LTL of the cleavage-stage group was comparable to the spontaneous pregnancy group, whereas the LTL of the blastocyst-stage group was shorter (β = −0.67, P = 7.81 × 10−8). This result suggests that the difference in LTL between transfer stages might explain the difference between the ART-conceived and spontaneous pregnancy groups. No associations between COS protocols, fertilization methods, or embryo transfer cycles on shorter LTL in children were observed. The multicenter validation cohort had similar associations between blastocyst-stage transfer and shorter LTL. Shortened LTL associated with blastocyst transfer was equivalent to 13.93 years (95% CI, 4.64–23.22 years), 13.42 years (95% CI, 4.52–22.21 years), and 10.85 years (95% CI, 1.03–20.67 years) of aging between 40–69 years in the discovery and 2 validation cohorts, respectively. The results of this study demonstrate that children conceived by ART are associated with a shorter LTL compared with those conceived spontaneously, and transfer of blastocyst-stage embryos was associated with shorter LTL in children than is the transfer of cleavage-stage embryos.
Read full abstract