A Normative Model Representing Autistic Individuals Amidst Autism Spectrum Phenotypic Heterogeneity.

Area of Science:

• Neurodevelopmental disorders
• Quantitative psychology
• Biostatistics

Background:

• Autism spectrum diagnosis relies on social communication and restricted behaviors, but requires additional modifiers like IQ and comorbidities for comprehensive assessment.
• Existing methods for identifying autism subgroups have limitations in evaluating individuals against normative samples.
• A quantitative score based on phenotypic variability is needed for individual comparison, subgroup homogeneity evaluation, and trajectory tracking.

Purpose of the Study:

• To develop a quantitative approach for assessing autism spectrum heterogeneity.
• To combine measures of phenotypic variability with normative modeling for individual assessment.
• To create a tool for comparing autistic individuals and evaluating clinical trajectories.

Main Methods:

• Principal Component Analysis (PCA) was used on phenotypic data from the Simons Simplex Collection (n=2744).
• Key features contributing to clinical heterogeneity were assessed using ADI-R, VABS, and CBCL.
• Total Intelligence Quotient (IQ) was used for cognitive assessment.

Main Results:

• Three principal components (PCs) explained 72% of the normative sample variance.
• PCA-projected dimensions facilitated normative modeling using a multivariate normal distribution to calculate percentiles.
• A Multidimensional General Functionality Score (MGFS) was developed based on these percentiles for evaluating individuals.

Conclusions:

• The proposed approach provides a basis for comparing individuals over time.
• It enables the evaluation of phenotypic clinical presentation homogeneity.
• This method can potentially guide research sample selection for clinical trials.