TY - JOUR
T1 - Genetic diagnostic yield by MRI pattern in children with cerebral palsy
T2 - a population-based study
AU - Berry, Jesia G.
AU - Taranath, Ajay
AU - Goetti, Robert
AU - Farrar, Michelle A.
AU - Fiori, Simona
AU - Pham, Huy-dat
AU - Mittinty, Murthy N.
AU - Corbett, Mark A.
AU - Palmer, Lyle J.
AU - Fornarino, Dani L.
AU - Harper, Kelly
AU - Gibson, Catherine S.
AU - Leishman, Shaneen J.
AU - Goldsmith, Shona C.
AU - McIntyre, Sarah J.
AU - Montgomerie, Alicia
AU - Pilkington, Rhiannon M.
AU - Lynch, John W.
AU - Russo, Remo N.
AU - Fahey, Michael C.
AU - Fripp, Jurgen
AU - Boyd, Roslyn N.
AU - Wright, Margaret J.
AU - Rice, James E.
AU - Waugh, Mary-Clare
AU - MacLennan, Alastair H.
AU - Gecz, Jozef
AU - van Eyk, Clare L.
PY - 2025/12
Y1 - 2025/12
N2 - Background: Neuroimaging abnormalities are detected in 80–86% of individuals with cerebral palsy (CP). Lesional white or grey matter injuries (WMI, GMI) are most common and typically attributed to environmental factors, while genetic causes are thought to underlie non-lesional injuries and normal brain imaging. This hypothesis has not been formally tested, and we aimed to evaluate it using the Australian CP biobank. Methods: This population-based study included 331 children with CP (195 males, 136 females), born between 1986 and 2018. Genomic DNA extracted from blood or saliva samples underwent sequencing, variant filtering, classification using ACMG-AMP criteria, and variant curation. Probands were classified as ‘resolved’, ‘candidate variant(s) identified’, or ‘no candidate variant identified’. Paediatric radiologists/neurologists coded brain MRI, CT, and ultrasound using the Magnetic Resonance Imaging Classification System (MRICS). Data analyses included descriptive statistics and multinomial logistic regression. Findings: A genetic aetiology was identified in 80 children (24%), while 165 (50%) had candidate variants and 86 (26%) had no candidate variant identified. Among children with predominant WMI or GMI (50% and 21%, respectively), 19% and 10% were genetically resolved. Children with maldevelopments, miscellaneous findings, or normal neuroimaging (10%, 10%, and 8%, respectively) were more often genetically resolved (41%, 48% and 39%) compared to those with lesional injuries (WMI or GMI), with relative risk ratios (RRR) of 3.54 (95% CI: 1.65–7.59), 4.75 (95% CI: 2.21–10.2), and 3.27 (95% CI: 1.42–7.52), respectively. Interpretation: These findings support the hypothesis that genetic aetiologies are more common in non-lesional CP. However, genetic diagnoses were observed across all MRICS categories, including 17% of lesional brain injuries. Notably, almost half of all genetically diagnosed children (39 of 80) were in the WMI and GMI groups. Therefore, we emphasise that neuroimaging should be used as a guide, not an exclusion criterion for genomic testing in CP. Funding: Channel 7 Children's Research Foundation, Cerebral Palsy Alliance Research Foundation, The Hospital Research Foundation, the NHMRC, and in-kind support from Illumina.
AB - Background: Neuroimaging abnormalities are detected in 80–86% of individuals with cerebral palsy (CP). Lesional white or grey matter injuries (WMI, GMI) are most common and typically attributed to environmental factors, while genetic causes are thought to underlie non-lesional injuries and normal brain imaging. This hypothesis has not been formally tested, and we aimed to evaluate it using the Australian CP biobank. Methods: This population-based study included 331 children with CP (195 males, 136 females), born between 1986 and 2018. Genomic DNA extracted from blood or saliva samples underwent sequencing, variant filtering, classification using ACMG-AMP criteria, and variant curation. Probands were classified as ‘resolved’, ‘candidate variant(s) identified’, or ‘no candidate variant identified’. Paediatric radiologists/neurologists coded brain MRI, CT, and ultrasound using the Magnetic Resonance Imaging Classification System (MRICS). Data analyses included descriptive statistics and multinomial logistic regression. Findings: A genetic aetiology was identified in 80 children (24%), while 165 (50%) had candidate variants and 86 (26%) had no candidate variant identified. Among children with predominant WMI or GMI (50% and 21%, respectively), 19% and 10% were genetically resolved. Children with maldevelopments, miscellaneous findings, or normal neuroimaging (10%, 10%, and 8%, respectively) were more often genetically resolved (41%, 48% and 39%) compared to those with lesional injuries (WMI or GMI), with relative risk ratios (RRR) of 3.54 (95% CI: 1.65–7.59), 4.75 (95% CI: 2.21–10.2), and 3.27 (95% CI: 1.42–7.52), respectively. Interpretation: These findings support the hypothesis that genetic aetiologies are more common in non-lesional CP. However, genetic diagnoses were observed across all MRICS categories, including 17% of lesional brain injuries. Notably, almost half of all genetically diagnosed children (39 of 80) were in the WMI and GMI groups. Therefore, we emphasise that neuroimaging should be used as a guide, not an exclusion criterion for genomic testing in CP. Funding: Channel 7 Children's Research Foundation, Cerebral Palsy Alliance Research Foundation, The Hospital Research Foundation, the NHMRC, and in-kind support from Illumina.
KW - Brain injury
KW - Cerebral palsy
KW - Genetic diagnoses
KW - Genetic testing
KW - Molecular diagnostic yield
KW - MRI
KW - Neuroimaging
UR - https://www.scopus.com/pages/publications/105020985294
U2 - 10.1016/j.ebiom.2025.106013
DO - 10.1016/j.ebiom.2025.106013
M3 - Article
C2 - 41202470
AN - SCOPUS:105020985294
SN - 2352-3964
VL - 122
JO - EBioMedicine
JF - EBioMedicine
M1 - 106013
ER -