Among children, schizophrenia is a rare neuropsychiatric disorder. The best prevalence estimates for schizophrenia in patients younger than 15 years is 0.05%, and only 2% of adult patients are estimated to have experienced the onset of their psychosis before the age of 13 years. According to the data and criteria provided by the National Institute of Mental Health (NIMH), in childhood and adolescence, two types of schizophrenia are commonly described, depending on the age of onset: early-onset schizophrenia (EOS), which presents between the ages of 13 and 17 years; and very-early-onset schizophrenia (VEOS), which presents at or before the age of 12 years. Although they are less common than adult-onset schizophrenia (AOS, presenting from the age of 17 years), VEOS and EOS tend to be more severe and disabling (1, 2). Concerning their clinical profile, both show higher rates of auditory hallucinations (3), negative symptoms, and bizarre behavior (4), as well as more severe neurocognitive difficulties (5), relative to AOS. Evidence of more severe neurocognitive difficulties among VEOS/EOS patients has been provided by anatomical brain MRI studies showing a greater progressive loss of cortical gray matter (6, 7) and progressive increases in ventricular volume (7) over time. In addition, compared with AOS patients, VEOS/EOS patients usually experience a longer duration of untreated psychosis (DUP) (8), worse outcomes (9), and more premorbid neurodevelopmental disorders (1).

Previous studies (10) have shown that premorbid neurodevelopmental difficulties are frequently present in children and adolescents who develop schizophrenia, especially those who experience an early onset of psychotic symptoms (e.g., VEOS patients). Within the NIMH cohort, Driver et al. (1, 11) observed that 67% of VEOS patients registered premorbid social, motor, language, and learning difficulties. This finding supports the results obtained by the same research group on a cohort of 118 participants with childhood-onset schizophrenia (11), of whom 55% (n = 65) showed premorbid academic difficulties, 72% (n = 85) had premorbid social difficulties, and 44% (n = 52) had premorbid motor difficulties; in addition, 20% (n = 24) screened positive for a pervasive developmental disorder (e.g., autism, Asperger syndrome) according to DSM-IV-TR criteria. Building on these findings, other longitudinal studies (12) have shown an overlap between early autistic symptoms and psychotic symptoms during childhood and adolescence. Specifically, in one study, 20–50% of VEOS participants were found to meet the criteria for premorbid autism spectrum disorder, according to DSM-5 criteria. Finally, some studies (11, 13, 14) have found the outcome and prognosis of VEOS/EOS patients to be positively correlated with the presence and severity of premorbid difficulties. Kincaid et al. (15) reported that VEOS/EOS patients with autistic features had higher symptom severity and poorer long-term outcomes compared with VEOS/EOS patients without autistic symptoms. Overall, these findings lend support to the neurodevelopmental continuum model, which holds that childhood neurodevelopmental difficulties and psychiatric disorders (e.g., schizophrenia) fall on an etiological and neurodevelopmental continuum, and should not be considered discrete entities (16). Recent studies (16, 17) have indicated a genetic overlap between schizophrenia and other syndromes that commonly arise during the developmental period. In the DSM-5, these syndromes are classified as “neurodevelopmental disorders,” and include intellectual disability (ID), communication disorders (CDs), autism spectrum disorder (ASD), neurodevelopmental motor disorders (including tic disorders), and specific learning disorders (SLDs).

Over the years, many empirical studies have shown that childhood neurodevelopmental disorders such as ID, ASD, and ADHD share specific genetic risk alleles, both with each other and with psychiatric disorders, particularly schizophrenia (16, 18). For example, Singh et al. (17) found that copy number variants associated with ID were significantly enriched in individuals with schizophrenia, concluding that many additional ID-related variants may be a risk factor for schizophrenia. Further support for this theory is provided by studies showing that individuals with ID have higher rates of schizophrenia than the general population (19, 20). More recently, Stanfield et al. (21) confirmed the relationship between ID and risk for psychotic positive symptoms in a large sample of adolescents (N = 168).

In addition, neuroimaging studies (22) on brain alteration in VEOS/EOS patients have found less intracranial, hippocampal, and amygdala volume and higher caudate, pallidum, and lateral ventricle volume compared with healthy controls. This result is relatively consistent with data reported for adult psychosis patients (23), with the exception of the lower intracranial volume (ICV). This lower ICV in VEOS/EOS patients, compared with adult psychosis patients, suggests more severe disruption of brain neurodevelopment and at an earlier age. Of note, previous studies have found lower ICV in children and adolescents with attention deficit hyperactivity disorder (ADHD) (24) and higher ICV in patients with ASD (22), highlighting the importance of considering ICV in neurodevelopmental disorder imaging studies. In addition, both children and adolescents with ADHD and adults suffering from schizophrenia and bipolar disorder have been found to have lower hippocampal volume (25, 26). Thus, lower hippocampal volume could be a feature of several neurodevelopment disorders, reflecting both shared and distinct illness mechanisms. Overall, these results provide further support for the neurodevelopmental continuum model, which considers neurodevelopmental disorders and psychiatric disorders (e.g., schizophrenia) diverse outcomes on the range of disrupted or deviant brain development (27).

However, most of the studies cited so far consider small samples of VEOS/EOS and do not deepen the differences in the clinical presentation of psychosis between VEOS/EOS with a neurodevelopmental disorder or neurodevelopmental difficulties and VEOS/EOS without a neurodevelopmental disorder or neurodevelopmental difficulties.

Studying clinical differences at the onset of psychosis in these clinical populations could allow us to better understand the overlap between neurodevelopmental disorders and schizophrenia, and thereby improve the diagnostic and treatment process. For example, previous studies (15) suggest that a longer duration of illness is associated with poorer long-term outcomes and higher symptom severity in VEOS/EOS with neurodevelopmental disorders or neurodevelopmental difficulties (i.e., autistic features). Therefore, an early screening of psychosis in children and adolescents with neurodevelopmental difficulties or diagnosed neurodevelopmental disorder could allow starting earlier both psychological and pharmacological treatments and possibly modify the clinical outcome. On this basis, the present study aimed to investigate, in a large sample (N = 230) of VEOS/EOS, the clinical differences, at the onset of psychosis, between VEOS/EOS with neurodevelopmental difficulties or diagnosed neurodevelopmental disorder as proposed in DSM-5 (ID, CDs, ASD, neurodevelopmental motor disorders, and SLDs) and VEOS/EOS with no diagnosed neurodevelopmental disorder or neurodevelopmental difficulties. In particular, we have focused on the age of psychosis onset, clinical profile at the onset of psychosis (e.g., positive, negative, disorganized, and general symptoms), and functional impairment at the onset of psychosis.

The main aim of the present study was to explore, in a large sample of VEOS/EOS, the clinical differences, at the onset of psychosis, between VEOS/EOS with neurodevelopmental difficulties or diagnosed neurodevelopmental disorder as proposed in DSM-5 (ID, CDs, ASD, neurodevelopmental motor disorders, and SLDs) and VEOS/EOS with no diagnosed neurodevelopmental disorder or neurodevelopmental difficulties.

We have focused on the age of psychosis onset, clinical profile at the onset of psychosis (e.g., positive, negative, disorganized, and general symptoms), and functional impairment at the onset of psychosis.

The first finding suggested that neurodevelopmental disorders may be common in our sample VEOS/EOS children and adolescents. Indeed, in our large sample (N = 176), 65 (37%) participants had a diagnosed neurodevelopmental disorder. Most frequently, the diagnoses were for ID (n = 25; 38%) and ADHD (n = 21; 32%). This finding is consistent with the results of previous studies showing that ID is common in individuals who develop schizophrenia (19, 32). Also, a history of ADHD symptoms is common in children and adolescents who develop schizophrenia (39), and ADHD is diagnosed in a high proportion of children at genetic risk for schizophrenia (40).

In addition, 22% of our participants had neurodevelopmental difficulties (e.g., motor, social, linguistic difficulties) that did not meet the criteria for a frank neurodevelopmental disorder. Specifically, 58% showed social difficulties, 39% showed language difficulties, 24% showed psychomotor difficulties, and 24% showed learning difficulties. These results are consistent with previous research finding that motor, linguistic, and social difficulties are present in children and adolescents who develop schizophrenia (1, 11).

Concerning the age of psychosis onset, in our sample, children and adolescents with a diagnosed neurodevelopmental disorder or neurodevelopmental difficulties showed an earlier onset of psychosis compared with participants with no neurodevelopmental diagnosis or difficulties. This result also aligns with the previous studies (10, 14, 41–43), which reported an association between earlier psychosis onset and the presence of language, motor, and social difficulties. For example, Petruzzelli et al. (44) found early onset of schizophrenia in 36 patients (age range: 7–17 years), of whom 70.6% presented neurodevelopmental difficulties, as well as difficulties in school learning or difficulties in sphincter control (enuresis).

Regarding clinical profile at the onset of psychosis, based on the SIPS/SOPS domains, we distinguished between positive symptoms, negative symptoms, disorganized symptoms, and general symptoms. Regarding positive symptoms, VEOS/EOS patients with a diagnosed neurodevelopmental disorder scored higher on grandiose ideas, perceptual abnormalities/hallucinations, and disorganized communication than VEOS/EOS patients with neurodevelopmental difficulties and VEOS/EOS patients with neither a neurodevelopmental disorder nor neurodevelopmental difficulties. The same pattern was found for disorganized symptoms, with VEOS/EOS patients with a diagnosed neurodevelopmental disorder scoring higher in odd behavior or appearance and bizarre thinking.

Regarding negative symptoms, VEOS/EOS patients with neither a neurodevelopmental disorder nor neurodevelopmental difficulties scored higher on social anhedonia, avolition, expression of emotion, experience of emotions and self, and ideational richness, relative to VEOS/EOS patients with a neurodevelopmental disorder or neurodevelopmental difficulties. Regarding general symptoms, we found no significant differences between groups.

Finally, regarding functional impairment at the onset of psychosis, in our sample, VEOS/EOS patients with a neurodevelopmental disorder or neurodevelopmental difficulties reported major global functional impairment, whereas VEOS/EOS patients without these two conditions did not. To understand the effect of neurodevelopmental disorders and difficulties on functioning in our VEOS/EOS sample, we also investigated two specific functioning domains: role and social. Both of these domains were found to be more compromised in VEOS/EOS patients with a neurodevelopmental disorder or neurodevelopmental difficulties than in VEOS/EOS patients with neither of these two conditions. We examined the relationship between neurodevelopmental disorders and difficulties and global, social, and role functioning, controlling for IQ (in the subgroup with neurodevelopmental disorders we included patients with ID) and number of neuropsychiatric disorders (neurodevelopmental disorders plus VEOS/EOS). The results showed that the poorer global, social, and role functioning of VEOS/EOS patients with a neurodevelopmental disorder or neurodevelopmental difficulties was not affected by any of these variables. In other words, neurodevelopmental disorders and difficulties tended to determine poorer global, social, and role functioning, regardless of cognitive functioning and the number of neuropsychiatric disorders.

Overall, based on our findings, we propose that the clinical picture of VEOS/EOS patients may differ at the onset of psychosis according to the presence (or lack) of a neurodevelopmental disorder or neurodevelopmental difficulties. In more detail, in children and adolescents with a neurodevelopmental disorder or neurodevelopmental difficulties, psychosis onset tends to occur at an earlier age and is associated with more severe (global, social, and role) functional impairment. In addition, in these patients, the onset of psychosis is likely to be characterized by positive symptoms (e.g., grandiose ideas, perceptual abnormalities, disorganized communication) and disorganized symptoms (e.g., odd behavior or appearance, bizarre thinking). Instead, in children and adolescents without a neurodevelopmental disorder or neurodevelopmental difficulties, the onset of psychosis is likely to be characterized by negative symptomatology (e.g., social anhedonia, avolition, expression of emotion, experience of emotions and self, ideational richness). These results should be replicated in future studies, with caution, to gain further information on the clinical and neurodevelopmental profiles of children and adolescents with suspected VEOS/EOS. Such research is essential for the early recognition of symptoms at the onset of psychosis and the preparation of a therapeutic program tailored to each patient.

The present research was the first study to explore the differences on clinical presentation of psychosis between VEOS/EOS with a neurodevelopmental disorder or neurodevelopmental difficulties and VEOS/EOS without a neurodevelopmental disorder or neurodevelopmental difficulties. Among the strengths that can be considered are the size of the sample and the age range (7–18 years) adequately representative of the developmental age. The research also considered a rich neurodevelopmental profile—considering both neurodevelopmental disorders (according to the DSM-5) and neurodevelopmental difficulties—and assessed psychotic symptoms (e.g., SIPS/SOPS) and level of functioning (e.g., GF: Role, GF: Social) at the onset of psychosis using semi-structured interviews. The study has several limitations. The first relates to the high prevalence of ID in the group with a neurodevelopmental disorder. This may have overestimated the presence of psychotic positive symptoms in this group. The second relates to the lack of data analysis about patients' pharmacological or psychosocial treatment before the onset of psychosis. The third relates to the lack, in the literature, of standardized structured or semi-structured interviews that can be used for the assessment of psychotic symptoms in children and adolescents with IDs. Future research should examine this aspect through longitudinal studies focused on the clinical overlap between neurodevelopmental disorders, neurodevelopmental difficulties, and VEOS/EOS.

The study data are not available due to ethical concerns. Patient privacy and security are protected, according to the ethical rules of our institutions and their restriction on data sharing.

The studies involving human participants were reviewed and approved by Children Hospital Bambino Gesù. Written informed consent to participate in this study was provided by the participants' legal guardian/next of kin.

MP and RA conceived the research study. MCT and MLP collected data. MP and MCT contributed to the data analysis/interpretation. MP, MCT, FC, and MLP contributed to the writing of the article. MP, RA, and SV critically reviewed the final draft of the article. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.