The current study supports the findings of our two previously published case series [30, 31], in which the chronic use of risperidone or aripiprazole, along with standard supportive therapies and with or without ADHD medication (methylphenidate or atomoxetine), was shown to improve the core signs and symptoms of ASD in young children. This treatment approach was effective in improving the CGI-I score of 79/82 patients (96%) to 2 or 1, with 35/82 patients (43%) experiencing complete remission of their ASD symptoms (CGI-I score of 1 and CARS2-ST score of < 30) as a result of early and chronic therapy of at least 6 months in duration. This is the first large retrospective case series (N = 82) to indicate that ASD core signs and symptoms can be treated successfully. Our two earlier case series documented the adoption of a comparable treatment strategy for 18 and 10 children aged 2 to 10 years old and demonstrated the complete resolution of ASD symptoms in 56% and 60% of patients, respectively. Although several previous studies had reported some improvement in ASD core signs and symptoms in young children, none have reported complete remission (these studies are reviewed in our previous case series) [45–52].
Here, risperidone and aripiprazole were prescribed to children with ASD at a younger age (from 2 years) than what is currently approved by the FDA (5 and 6 years, respectively) to assist with challenging behaviors that were proving a major barrier to learning, were resistant to standard supportive therapies alone, and warranted pharmacological intervention. This treatment regimen was offered to patients whose parents or legal guardians consented to the regimen, and who understood the potential adverse effects and the off-label use of these medications. Thus, we combined non-biologic (e.g., ABA therapy) and biological therapies to treat children with autism in a region where the vast majority of families cannot afford regular treatment with the recommended range of supportive therapies (ABA therapy, speech therapy, and occupational therapy).
In addition to previous studies showing improvement in ASD core signs and symptoms in young children following treatment with risperidone or aripiprazole [45–52], a number of studies have shown that the use of these medications can aid learning and cognitive development in children with ASD [53–59]. Unfortunately, however, the majority of the studies performed on ASD in children thus far have been short-term, employed prescription drugs to chiefly regulate behavioral problems, have not combined antipsychotic medications with ADHD medications, and have not included a cross-over arm for non-responders [51, 53–58, 60–63].
Because attention is critical in the learning process, we augmented concentration with ADHD medications as required. The co-occurrence of ADHD in young children with ASD further impacts on their function and quality of life , and the beneficial effects of ADHD medications on anxiety, inattention, social communication, and self-regulation in this population is supported by a number of previous studies and reports [15–21].
In our patients, the side effects following medication were either slight and fleeting (such as drowsiness and drooling), asymptomatic (elevated prolactin), or treatable with adjunctive treatment such as dietary changes or the use of methylphenidate or atomoxetine for both inattention and excessive weight gain. However, methylphenidate and atomoxetine were omitted wherever possible since these medications may increase the risk of adverse effects and their consumption in very young children is unapproved. Both risperidone and aripiprazole have previously been linked to obesity and asymptomatic elevated prolactin [9, 39, 59, 64, 65], and these were the main side effects observed in our patients.
There are several limitations inherent in this retrospective case series: it was open-label without controls or randomization; the confounding effects of non-pharmacological interventions were not, therefore, controlled for; and the patients did not follow a unified treatment model. Despite the fact that no definitive conclusions can be drawn owing to the study’s limitations, we expect that our findings will inspire further research in this area.