利用全基因組關聯分析探討影響極低體重早產兒動作發展及心智之遺傳因素

Abstract

背景： 早產兒比足月兒有更高的機會面臨動作及心智發展遲緩的問題，發展遲緩不僅止於早產兒在嬰幼兒時期的表現，直至青少年或成人都有可能有一定程度的影響。由於嬰兒神經發展是一個連續的過程，發展軌跡比單一時間點的發展分數更能預測未來的結果。許多原因皆有可能讓相似週數和體重的早產兒之發展軌跡不同，包含少見的孟德爾遺傳疾病，但透過全基因組關聯分析找出的基因多型性，或是延伸出的多基因風險分數，是否扮演任何角色則未有定論，因此，本篇研究希望以出生體重、週數、自閉症、注意力不足及過動症及思覺失調症五種性狀或疾病所計算出的多基因風險分數，探討影響早產兒動作及心智發展軌跡的基因背景。 方法： 本研究納入478 位極低體重的早產兒，收案時間分別為2002 年至2004 年、2006 年至2008 年及2012 年至2014 年，收案來源來自台大醫院、聯合醫院婦幼院區、馬偕醫院及成大醫院，發展軌跡是利用貝萊嬰兒發展量表分析的結果，單核?酸變異則利用口腔細胞萃取出的核酸作為分析樣本來源，以Axiom Genome-Wide TWB 2.0 Array Plate 做基因型鑑定，將通過品質管控的單核?酸變異進行關聯分析和多基因風險分數的計算，計算出的多基因風險分數會進一步以線性回歸分析其預測動作及心智發展軌跡的能力。 結果： 共370 位同時具有基因資料和發展分數的早產兒納入分析，其中5.7%的動作發展軌跡為持續遲緩、26.8%是逐漸惡化、61.6%是穩定正常和5.9%是高於平均，3.2%的心智發展軌跡為持續遲緩、30.5%是逐漸惡化和66.2%是穩定正常。本研究並未從關聯性分析找出具有統計意義的單核?酸變異和發展軌跡相關，但較高的出生體重基因風險分數和較差的動作發展軌跡存在正相關，在動作發展較差的組別，其平均出生體重基因風險分數為0.000221，於在動作發展相較正常的組別，其平均出生體重基因風險分數為0.000217；注意力不足過動症基因風險分數也和動作發展軌跡相關，但動作發展較差的組別，其注意力不足過動症基因風險分數也較低，動作發展較差的組別，平均出生體重基因風險分數為-0.001270，動作發展相較正常的組別，其平均出生體重基因風險分數為-0.001243。 結論： 本研究初步結果顯示以出生體重計算出的基因風險分數，可以用來預測及低體重早產兒日後的動作發展，進而幫助辨識出高風險族群以提供早期發展介入治療。除此之外，早產兒的動作發展也和注意力不足過動症有部分共同的基因背景影響，此結果可呼應臨床注意力不足過動症的孩童早期發展的不同。未來將利用基因補值的方式納入外部已知的參照基因型，以更完整的基因資料提供更穩健的早產兒發展及基因背景的關聯性。

Background Preterm infants are vulnerable to developmental delay in both motor and mental aspects. Developmental outcome of preterm infants is of great importance owing to the long-term influences into adolescence and adulthood. Since infant developmental process is nonlinear, the trajectories in either motor or mental development in the first year of life has been found to be more predictive for later outcome than the assessment at single time point. Many factors might contribute to the differential developmental trajectories for the infants at similar gestational age or birth body weight, including some rare Mendelian genetic disorders. It remains unknown whether common genetic variants such as single nucleotide polymorphism (SNP), genotyped using genome-wide association studies (GWAS) array, contribute to the developmental trajectories in preterm infants in a form of polygenic risk score (PRS). Hence, this study aimed to explore the genetic influences on infant motor and mental developmental trajectories, respectively, in preterm infants using the PRS estimated from five traits or diseases, including birth weight, gestational age, autism spectrum disorder (ASD), attention deficit hyperactivity disorder (ADHD), and schizophrenia. Methods The study sample consisted of 478 very low birth weight (< 1500g) preterm (< 37 weeks) infants who were born in or admitted to four hospitals in Taiwan (National Taiwan University Hospital, Branch for Women and Children at Taipei City Hospital, and MacKay Memorial Hospital, and National Cheng Kung University Hospital) in three periods (2002–2004, 2006–2008, and 2012–2014). The development of each infant was assessed using the Bayley Scales of Infant Development. Developmental trajectories were conducted based on the motor and mental score on the Bayley scale, respectively. Participants’ DNA were extracted from buccal cell sampling of each child following standard quality control procedures. Genotyping for SNPs was conducted using the Axiom Genome-Wide TWB 2.0 Array Plate. SNPs identified from the genotyping results were further proceeded to association analysis and polygenic risk score estimation. The association between polygenic risk scores and developmental trajectories were verified by performing linear regression analyses. Results A total of 370 preterm infants who had both genotyping data and developmental measurement were included in this study. For motor trajectories, 5.7% of the infants were persistently delayed, 26.8% were deteriorating, 61.6% were stably normal, and 5.9% were above average. No significant SNPs associated with motor trajectories were identified in the association analysis of individual SNPs. Higher PRSs for birth weight were positively associated with suboptimal motor development. Mean PRS for birth weight was 0.000221 for infants with delayed motor development and 0.000217 for infants with normal development (p<0.05). Higher PRSs for ADHD were inversely related to optimal motor development that mean PRS for ADHD was -0.001270 for infants with delayed motor development and -0.001243 for infants with normal development (p<0.05) Conclusions Our preliminary findings suggest that the motor development outcomes at 12 month of very low birth weight preterm infants could be predicted using the PRSs for birth weight, which might help identify high-risk groups for early intervention. In addition, shared genetic susceptibility between ADHD and motor development could explain part of the clinical observations. Future work warrants to incorporate imputations for genotypes to examine the robustness of the relationship between genetic contributions and infant development.