Polygenic Risk Score of Adolescent Idiopathic Scoliosis for Potential Clinical Use

Nao Otomo, Hsing Fang Lu, Masaru Koido, Ikuyo Kou, Kazuki Takeda, Yukihide Momozawa, Michiaki Kubo, Yoichiro Kamatani, Yoji Ogura, Yohei Takahashi, Masahiro Nakajima, Shohei Minami, Koki Uno, Noriaki Kawakami, Manabu Ito, Tatsuya Sato, Kei Watanabe, Takashi Kaito, Haruhisa Yanagida, Hiroshi TaneichiKatsumi Harimaya, Yuki Taniguchi, Hideki Shigematsu, Takahiro Iida, Satoru Demura, Ryo Sugawara, Nobuyuki Fujita, Mitsuru Yagi, Eijiro Okada, Naobumi Hosogane, Katsuki Kono, Masaya Nakamura, Kazuhiro Chiba, Toshiaki Kotani, Tsuyoshi Sakuma, Tsutomu Akazawa, Teppei Suzuki, Kotaro Nishida, Kenichiro Kakutani, Taichi Tsuji, Hideki Sudo, Akira Iwata, Kazuo Kaneko, Satoshi Inami, Yuta Kochi, Wei Chiao Chang, Morio Matsumoto, Kota Watanabe, Shiro Ikegawa, Chikashi Terao

Research output: Contribution to journalArticlepeer-review

5 Citations (Scopus)


Adolescent idiopathic scoliosis (AIS) is a common disease causing three-dimensional spinal deformity in as many as 3% of adolescents. Development of a method that can accurately predict the onset and progression of AIS is an immediate need for clinical practice. Because the heritability of AIS is estimated as high as 87.5% in twin studies, prediction of its onset and progression based on genetic data is a promising option. We show the usefulness of polygenic risk score (PRS) for the prediction of onset and progression of AIS. We used AIS genomewide association study (GWAS) data comprising 79,211 subjects in three cohorts and constructed a PRS based on association statistics in a discovery set including 31,999 female subjects. After calibration using a validation data set, we applied the PRS to a test data set. By integrating functional annotations showing heritability enrichment in the selection of variants, the PRS demonstrated an association with AIS susceptibility (p = 3.5 × 10−40 with area under the receiver-operating characteristic [AUROC] = 0.674, sensitivity = 0.644, and specificity = 0.622). The decile with the highest PRS showed an odds ratio of as high as 3.36 (p = 1.4 × 10−10) to develop AIS compared with the fifth in decile. The addition of a predictive model with only a single clinical parameter (body mass index) improved predictive ability for development of AIS (AUROC = 0.722, net reclassification improvement [NRI] 0.505 ± 0.054, p = 1.6 × 10−8), potentiating clinical use of the prediction model. Furthermore, we found the Cobb angle (CA), the severity measurement of AIS, to be a polygenic trait that showed a significant genetic correlation with AIS susceptibility (rg = 0.6, p = 3.0 × 10−4). The AIS PRS demonstrated a significant association with CA. These results indicate a shared polygenic architecture between onset and progression of AIS and the potential usefulness of PRS in clinical settings as a predictor to promote early intervention of AIS and avoid invasive surgery.

Original languageEnglish
Pages (from-to)1481-1491
Number of pages11
JournalJournal of Bone and Mineral Research
Issue number8
Publication statusPublished - 2021 Aug



ASJC Scopus subject areas

  • Endocrinology, Diabetes and Metabolism
  • Orthopedics and Sports Medicine


Dive into the research topics of 'Polygenic Risk Score of Adolescent Idiopathic Scoliosis for Potential Clinical Use'. Together they form a unique fingerprint.

Cite this