Browsing by Author "Jacques, Thomas S."

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    MRI-based radiomics for prognosis of pediatric diffuse intrinsic pontine glioma: an international study
    (Oxford University Press, 2021-03-05) Tam, Lydia T.; Yeom, Kristen W.; Wright, Jason N.; Jaju, Alok; Radmanesh, Alireza; Han, Michelle; Toescu, Sebastian; Maleki, Maryam; Chen, Eric; Campion, Andrew; Lai, Hollie A.; Eghbal, Azam A.; Oztekin, Ozgur; Mankad, Kshitij; Hargrave, Darren; Jacques, Thomas S.; Goetti, Robert; Lober, Robert M.; Cheshier, Samuel H.; Napel, Sandy; Said, Mourad; Aquilina, Kristian; Ho, Chang Y.; Monje, Michelle; Vitanza, Nicholas A.; Mattonen, Sarah A.; Radiology and Imaging Sciences, School of Medicine
    Background: Diffuse intrinsic pontine gliomas (DIPGs) are lethal pediatric brain tumors. Presently, MRI is the mainstay of disease diagnosis and surveillance. We identify clinically significant computational features from MRI and create a prognostic machine learning model. Methods: We isolated tumor volumes of T1-post-contrast (T1) and T2-weighted (T2) MRIs from 177 treatment-naïve DIPG patients from an international cohort for model training and testing. The Quantitative Image Feature Pipeline and PyRadiomics was used for feature extraction. Ten-fold cross-validation of least absolute shrinkage and selection operator Cox regression selected optimal features to predict overall survival in the training dataset and tested in the independent testing dataset. We analyzed model performance using clinical variables (age at diagnosis and sex) only, radiomics only, and radiomics plus clinical variables. Results: All selected features were intensity and texture-based on the wavelet-filtered images (3 T1 gray-level co-occurrence matrix (GLCM) texture features, T2 GLCM texture feature, and T2 first-order mean). This multivariable Cox model demonstrated a concordance of 0.68 (95% CI: 0.61-0.74) in the training dataset, significantly outperforming the clinical-only model (C = 0.57 [95% CI: 0.49-0.64]). Adding clinical features to radiomics slightly improved performance (C = 0.70 [95% CI: 0.64-0.77]). The combined radiomics and clinical model was validated in the independent testing dataset (C = 0.59 [95% CI: 0.51-0.67], Noether's test P = .02). Conclusions: In this international study, we demonstrate the use of radiomic signatures to create a machine learning model for DIPG prognostication. Standardized, quantitative approaches that objectively measure DIPG changes, including computational MRI evaluation, could offer new approaches to assessing tumor phenotype and serve a future role for optimizing clinical trial eligibility and tumor surveillance.
  • Thumbnail Image
    Item
    The clinical and genetic spectrum of inherited glycosylphosphatidylinositol deficiency disorders
    (Oxford University Press, 2024) Sidpra, Jai; Sudhakar, Sniya; Biswas, Asthik; Massey, Flavia; Turchetti, Valentina; Lau, Tracy; Cook, Edward; Alvi, Javeria Raza; Elbendary, Hasnaa M.; Jewell, Jerry L.; Riva, Antonella; Orsini, Alessandro; Vignoli, Aglaia; Federico, Zara; Rosenblum, Jessica; Schoonjans, An-Sofie; de Wachter, Matthias; Alvarez, Ignacio Delgado; Felipe-Rucián, Ana; Haridy, Nourelhoda A.; Haider, Shahzad; Zaman, Mashaya; Banu, Selina; Anwaar, Najwa; Rahman, Fatima; Maqbool, Shazia; Yadav, Rashmi; Salpietro, Vincenzo; Maroofian, Reza; Patel, Rajan; Radhakrishnan, Rupa; Prabhu, Sanjay P.; Lichtenbelt, Klaske; Stewart, Helen; Murakami, Yoshiko; Löbel, Ulrike; D'Arco, Felice; Wakeling, Emma; Jones, Wendy; Hay, Eleanor; Bhate, Sanjay; Jacques, Thomas S.; Mirsky, David M.; Whitehead, Matthew T.; Zaki, Maha S.; Sultan, Tipu; Striano, Pasquale; Jansen, Anna C.; Lequin, Maarten; de Vries, Linda S.; Severino, Mariasavina; Edmondson, Andrew C.; Menzies, Lara; Campeau, Philippe M.; Houlden, Henry; McTague, Amy; Efthymiou, Stephanie; Mankad, Kshitij; Radiology and Imaging Sciences, School of Medicine
    Inherited glycosylphosphatidylinositol deficiency disorders (IGDs) are a group of rare multisystem disorders arising from pathogenic variants in glycosylphosphatidylinositol anchor pathway (GPI-AP) genes. Despite associating 24 of at least 31 GPI-AP genes with human neurogenetic disease, prior reports are limited to single genes without consideration of the GPI-AP as a whole and with limited natural history data. In this multinational retrospective observational study, we systematically analyse the molecular spectrum, phenotypic characteristics and natural history of 83 individuals from 75 unique families with IGDs, including 70 newly reported individuals; the largest single cohort to date. Core clinical features were developmental delay or intellectual disability (DD/ID, 90%), seizures (83%), hypotonia (72%) and motor symptoms (64%). Prognostic and biologically significant neuroimaging features included cerebral atrophy (75%), cerebellar atrophy (60%), callosal anomalies (57%) and symmetric restricted diffusion of the central tegmental tracts (60%). Sixty-one individuals had multisystem involvement including gastrointestinal (66%), cardiac (19%) and renal (14%) anomalies. Though dysmorphic features were appreciated in 82%, no single dysmorphic feature had a prevalence >30%, indicating substantial phenotypic heterogeneity. Follow-up data were available for all individuals, 15 of whom were deceased at the time of writing. Median age at seizure onset was 6 months. Individuals with variants in synthesis stage genes of the GPI-AP exhibited a significantly shorter time to seizure onset than individuals with variants in transamidase and remodelling stage genes of the GPI-AP (P = 0.046). Forty individuals had intractable epilepsy. The majority of individuals experienced delayed or absent speech (95%), motor delay with non-ambulance (64%), and severe-to-profound DD/ID (59%). Individuals with a developmental epileptic encephalopathy (51%) were at greater risk of intractable epilepsy (P = 0.003), non-ambulance (P = 0.035), ongoing enteral feeds (P < 0.001) and cortical visual impairment (P = 0.007). Serial neuroimaging showed progressive cerebral volume loss in 87.5% and progressive cerebellar atrophy in 70.8%, indicating a neurodegenerative process. Genetic analyses identified 93 unique variants (106 total), including 22 novel variants. Exploratory analyses of genotype-phenotype correlations using unsupervised hierarchical clustering identified novel genotypic predictors of clinical phenotype and long-term outcome with meaningful implications for management. In summary, we expand both the mild and severe phenotypic extremities of the IGDs, provide insights into their neurological basis, and vitally, enable meaningful genetic counselling for affected individuals and their families.