Current evidence suggests the phenotype of autism spectrum disorder to be driven by a complex interaction of genetic and environmental factors impacting onto brain maturation, synaptic function, and cortical networks. However, findings are heterogeneous, and the exact neurobiological pathways of autism spectrum disorder still remain poorly understood. The co-twin control or twin-difference design is a potentially powerful tool to disentangle causal genetic and environmental contributions on neurodevelopment in autism spectrum disorder. To this end, monozygotic twins discordant for this condition provide unique means for the maximum control of potentially confounding factors. Unfortunately, only few studies of a rather narrow scope, and limited sample size, have been conducted. In an attempt to highlight the great potential of combining the brain connectome approach with monozygotic twin design, we first give an overview of the existing neurobiological evidence for autism spectrum disorder and its cognitive correlates. Then, a special focus is made onto the brain imaging findings reported within populations of monozygotic twins phenotypically discordant for autism spectrum disorder. Finally, we introduce the brain connectome model and describe an ongoing project using this approach among the largest cohort of monozygotic twins discordant for autism spectrum disorder ever recruited.