Genetically, Developmentally, and Clinically Distinct Cognitive Subtypes in Schizophrenia: A Tale of Three Trajectories

Cognitive impairment is a core symptom dimension in schizophrenia: it is common, often severe, and associated with functional impairment (1).Historically, cognitive impairment in schizophrenia has been thought to be a function of neurodevelopment starting well before illness onset.Studies across illness phases suggest that, at the group level, some degree of cognitive disruption is present prior to diagnosis, becoming more pronounced around the prodrome and first episode, and settling into a new, lower baseline by the time of established illness (2).However, substantial heterogeneity in cognitive presentation is well described (3, 4), creating challenges in characterizing cognition and linking cognitive symptoms to neurobiology and clinical outcomes.Given the marked variability in cognition after illness onset, it is likely that cognitive developmental trajectories are also highly heterogeneous.Recent cross-sectional studies of patients in different illness stages using estimates of premorbid and current cognitive functioning have identified empirically derived subgroups of patients with distinct cognitive trajectories (e.g., 5-7), suggesting that cognitive course in schizophrenia may not follow a unitary path.Different patterns of brain structure and connectivity have been associated with distinct cognitive subtypes and trajectories (5, 7-9), and evidence from two genetic association studies of cognitive subgroups found that genetic markers were differentially associated with cognitive profiles (10, 11).These findings suggest that variability in cognitive functioning may be leveraged to identify more homogeneous subgroups and trajectories reflecting distinct biological markers.The identification of genome-wide polygenic risk scores (PRSs) for a variety of constructs ranging from diagnostic classifications (e.g., schizophrenia, attention deficit hyperactivity disorder [ADHD]) to more dimensional features (e.g., cognition, education) based on very large samples has created a set of valuable tools for examination of genetic risk along specific constructs of interest and may provide key information about unique genetic contributions not only to illnesses themselves but to different pathophysiological processes that lead to them.To this end, the article by Dickinson and colleagues in this issue of the Journal (12) combines two novel approaches-empirical stratification of cognitive trajectories and examination of multiple polygenic risk indicators (schizophrenia, cognition, ADHD, and education)-to examine distinct cognitive pathways in schizophrenia that may be associated with unique combinations of genetic risk factors.