A review on promising genetic biomarkers and therapeutic interventions for advancing precision medicine principles for autism

Autism spectrum disorder (ASD) is an etiologically and clinically complex neurodevelopmental disorder that is characterized by various developmental disorders that impact daily functioning of life. Affected individuals typically exhibit symptoms such as restricted or repetitive behaviors (hand flapping, rocking from side to side, full body movements), challenges with social interactions, hyperactivity or lack of attention or interest (which could also be related to attention deficit hyperactivity disorder [ADHD]), difficulty in non-verbal and verbal communication, etc. [1–4]. Over several decades, the prevalence of autism has been rising at an alarming rate. According to the recent estimates from the Centers for Disease Control and Prevention (CDC) Autism and Developmental Disabilities Monitoring (ADDM) Network, about 1 in 54 children has been identified with ASD in 2020, while the incidence was 1 in 150 in 2007 [5].

The clinical symptoms in ASD patients vary significantly, suggesting that the syndrome affects each person differently and perhaps a reflection of the underlying genetic heterogeneity [6]. This means that people with ASD may require unique treatment plans, and treatment for ASD is most successful when it is designed to meet a patient's specific needs [4]. Currently, ASD is primarily diagnosed using behavioral testing, all of which are not detected at least until three years of age. Therefore, there is a critical need for an earlier diagnosis of ASD risk children so that behavioral and therapeutic interventions could be initiated at an age where they are most effective [7,8].

To understand the underlying core molecular mechanisms and genetic perturbations (gene mutations, chromosome deletions or duplications, inherited variants) associated with ASD, scientists have been banking on the recent genetic and bioinformatic advances. A good understanding of the underlying mechanism for various clinical symptoms can lead to identification of potential biomarkers and development of effective drugs to treat these conditions.

In addition, various targeted behavioral therapies such as ‘Applied Behavior Analysis’ (ABA), psychological therapy, physical therapy and virtual reality and gaming interventions are at various stages of development. The adoption of these approaches may complement the therapeutic strategies and can contribute to improved quality of life for patients. This review is focused on summarizing some of the recent advances for deciphering the underlying genetic architecture of autism and understanding how these findings might promote precision medicine in the future and various therapeutic interventions that are attempting to improve ASD care. This review is by no means an exhaustive compilation and is a mere attempt to highlight some of the promising findings in the recent literature on ASD and therapeutic interventions.

Precision medicine

Over the past decade, principles of precision medicine and personalized medicine have been widely adopted for managing and/or treating patients with various complex disease conditions. These practices offer patient-specific care based on key biomarkers (such as perturbations in specific genes, proteins, mechanisms, etc.), aiding in accurate diagnosis and designing suitable treatment plans. Recent rapid advancements in molecular biology and computational technologies, as well as meta-analysis of the human genome, are paving the way for early identification of precise biomarkers for various complex disease conditions. These strategies are often used for various types of cancers (e.g., colorectal, breast, lung, stomach, ovarian, thyroid cancers, leukemia, melanoma) where two patients diagnosed with the same cancer type (e.g., melanoma skin cancer) may have different underlying gene mutations, and therefore may not always respond to treatment the same way. Therefore, physicians opt for testing cancer cells from patients with various drugs and observe for specific changes in the target gene or proteins to customize patient-specific treatment programs that may have a higher likelihood of success. Similarly, personalized medicine is routinely followed for managing Type-1 diabetes, where daily insulin therapy is customized based on an individual's unique physiological, emotional, and developmental state, dietary habits, physical activities, etc. Likewise, more recently, stratification of patient populations with underlying predisposition conditions (such as asthma, bronchiectasis, obesity, etc.) has been effective in developing personalized treatment plans for COVID-19 patients. A person-centered approach for early prediction and cure of ASD patients is an aspirational goal, and recent advances in understanding human genetic biomarker targets and complementary therapeutic interventions could help develop effective precision medicine tools in the near future.

Autism & underlying genetic factors

Research over the past decade has shown that genetic perturbations play an important role in autism. The spectrum of genetic risk for ASD includes new spontaneous or inherited gene mutations, chromosomal deletions or duplications [6]. Earlier research into the genetic factors for ASD primarily focused on the identification of spontaneous gene mutations in protein-coding regions of the DNA because these mutations constitute about 25–30% of ASD cases and have been found to be associated with various symptoms, including lower non-verbal IQs [6,9]. However, only about 2% of the genome consists of protein-coding regions. The recent advances in sequencing and statistical methods are enabling scientists to cast a wider net by expanding the scope to non-coding regions of the DNA (e.g., genomic regions that regulate transcription and translation) in larger population samples to identify autism-susceptible genes [6,10,11]. Recently, Trost et al. [12] published the recent release of the Autism Speaks MSSNG resource, consisting of whole genome sequencing data from 11,312 individuals (5100 individuals with ASD and 6212 non-ASD parents and siblings). This database serves as a great resource for understanding genotype-phenotype correlations, etiology, rare variants in ASD population. Similarly, Satterstrom et al. [13] analyzed 35,584 samples, including 11,986 with ASD, and identified many gene biomarkers that play an essential role in neurodevelopment processes, suggesting an important role of these gene perturbations in ASD clinical manifestation. Other similar studies have identified ASD-susceptible genes, including some that serve essential functions to regulate child development (such as cell proliferation, transcription, translation mechanisms, neurodevelopment, ion channel activity, etc.) (Table 1) [6]. Mutations in Chd8, Pten, Tsc1, Tsc2, Nf1 genes are associated with macrocephaly and associated symptoms such as seizures [14], while mutations in Dyrk1a and Dkl5 are associated with microcephaly [15–17]. Similarly, Bckdk gene mutations are associated with neurobehavioral defects [18]. Mutations in single critical genes associated with transcription and translation in neurodevelopment networks (e.g., Tbr1 and Chd8 encode transcription factors) impact the function of other neurodevelopment genes and proteins and thus affect the function of broader gene regulatory networks in the developing brains of ASD children [6,9,19]. While these well-characterized biomarkers show great promise, the specific role of the majority of the identified genes in developing ASD clinical signs is yet to be fully characterized.

In addition to the gene mutations, deletions (DEL) or duplications (DUP) of a large genomic region on human chromosome 15 (15q11–13) and 16 (16p11.2) have been identified as one of the stronger risk factors for developing various ASD symptoms. These larger genetic perturbations impact many important pathways, including cell proliferation (such as mTOR, MAPK and Wnt signaling) and could contribute to the observed alterations in anatomical and physiological abnormalities in ASD patients, including the development of craniofacial structure and growth of the skull, attributed to impairment of social behavior [6]. Other effects include effects on critical neurobiological mechanisms during early brain development, including a reduced number of neural progenitor cells, disruption of neuron migration, maturation and morphology. In addition, multiple pathways, including cell motility, ion channel activity, and actin cytoskeleton, are disrupted by the 16p.11.2 deletion/duplication [20].

Epigenetic investigations over the past two decades show promising insights toward improving early prediction of ASD. Epigenetic studies including DNA methylation using human brain, placenta cord blood and other surrogate models have identified transcripts that are epigenetically modified in ASD, which could provide promising clues toward improved early prediction of ASD [8]. For example, Wong et al. [21] analyzed post-mortem brain tissues from 43 ASD and 38 control donors and identified distinct DNA methylation signatures associated with ASD. They reported altered methylation patterns in 15q11-q13 loci and other genes associated with immune system, synaptic signaling and neuronal regulation in brain tissues from ASD patients.

Further understanding of the causative relationship between various genetic perturbations and ASD clinical signs may further improve our understanding of the key mechanisms involved and thus help to develop promising biomarkers for efficient screening of patients at earlier stages of child development and for potential drug development.

Precision medicine in autism care

The first few years of life are critical to a child's development and are particularly important for children with ASD. Early diagnosis and adopting precision medicine approaches could result in significant improvements in a child's life and their ability to overcome functional challenges at home, school and social settings and help to reach developmental milestones [4,8]. Various innovative approaches are being proposed to improve diagnosis procedures for ASD. Recently, Chan et al. [22] used minor physical, major congenital anomalies, and patient-level whole genome-wide rare variant scores for the categorization of 325 children with ASD. With improvements in diagnostic tools, and increased awareness among parents, it's feasible to encourage families with ASD history to have their children undergo genetic testing to determine if they have inherited genetic perturbations that puts them at a higher risk of certain developmental defects. For example, screening children's DNA for 16p11.2 deletions or duplications may detect susceptibility for cognitive, linguistic skills, impact on IQ. This approach might help diagnose ASD at earlier stages of child development, and further develop specialized developmental, physical, pediatric speech, and psychological interventions to help children navigate daily challenges. To add complexity, children were shown to develop ASD due to inheritance of genetic alterations (in noncoding DNA regions) from parents (especially, fathers), who do not meet the criteria of ASD (exhibiting mild or no symptoms) [6]. These findings warrant expediting the identification of biomarkers covering wider genetic perturbations in diverse populations (and represent various geographies) to identify ASD susceptibility more efficiently.

Currently, there are no medications available to completely cure all ASD symptoms. However, the identified genetic or molecular biomarkers are being evaluated for developing and advancing therapeutic drug development strategies. For example, identification of the gene perturbations in cell proliferation pathways led to evaluation of mTOR pathway (and other related proliferation pathways) inhibitors as a potential treatment option for certain (social deficit and repeated behaviors) ASD symptoms [23]. Everolimus is one such drug that was developed and has received FDA approval for treating seizures [24]. Similarly, mechanistic understanding of ASD has contributed to other medications for managing symptoms such as inability to focus, self-harming behavior, anxiety, depression, seizures and gastrointestinal problems. Further understanding of various genetic perturbations may further decipher key molecular biomarkers and contribute to identifying additional key biomarkers and for developing targeted drug interventions. Some of the prominent ASD symptoms overlap with the symptoms of other neurodevelopment diseases such as intellectual disability, ADHD and schizophrenia. Clustering of patients with similar etiologies and neurodevelopmental symptoms may allow mechanism-based precision medicine strategies for these conditions. Similarly, advancements in physical interventions and virtual reality approaches to meet each individual's unique requirements may complement understanding of genetic pathobiology and further improve the ASD management.

Traditional & emerging therapeutic interventions

In addition to the drug-based treatments, other therapeutic interventions are often applied, which can be broadly categorized into traditional or virtual reality (VR) procedures. These interventions generally attempt to improve or manage symptoms associated with cognitive behaviors, specific phobias, or to motivate children with ASD to improve physical activity and quality of life. These therapeutic interventions primarily focus on improving social skills, emotions, daily living activities, communication, cognitive training and improvements of physical activity or motivation [1].

Behavioral treatments such as ‘Applied Behavior Analysis’ (ABA) are widely used to improve a variety of skills by encouraging desired behavior (e.g., initiate communication with others) via step-by-step instructions in either a clinical or neutral setting and discourage undesired behaviors, with the progress being regularly tracked. Developmental therapies such as speech and language therapy, psychological therapy, occupational therapy, sensory integration therapy and physical therapy are advocated to develop skills to aid leading independent daily activities. Therapies such as ‘Early Start Denver Model’ are initiated in children at 1–4 years of age, using play, social exchanges, and shared attention in natural settings to improve social, language and learning skills. Similarly, relationship-development intervention therapies are focused on improving motivation, interest, and social interactions [25]. Previous studies have reported improvements in ASD symptoms, executive function deficits and social functioning after receiving yoga and mindfulness-based interventions [26,27].

The VR procedures typically focus to merge real and virtual worlds by utilizing various gadgets, including video gaming, stereoscopic displays, user tracker technologies, augmented reality (AR) etc. These procedures represent a step forward in the ASD treatment because of their ubiquitous nature, simulation of everyday life situations (learnings acquired could be easily transferred to real life) and flexibility to provide therapist-controlled training remotely under safe/comfortable/playful environments (including simulating home or school conditions), which could be controlled and personalized based on objectives and improvements of the intervention. The VR tools are particularly advantageous for treatments focused on improving social skills and interactions, emotional and non-verbal communication skills, executive functions, training in daily living activities and cognitive abilities. Generally, the VR-based technologies could increase motivation and adherence of ASD children as they can customize the characteristics of the avatars. It is relatively easy for the therapist/parents to quantitatively monitor progress and customize the tasks, measures, challenges, situations and environmental stimuli [1].

Mesa-Gresa et al. [1] conducted a thorough review of the literature (published between 2010 and 2018) on the application of VR programs in the intervention and treatment of ASD children and published their findings, which are summarized in this review. The majority of the VR studies in ASD have focused on exposing ASD children to various social situations in virtual scenarios using personalized avatars, and these studies have shown improvements in social skills (such as initiation of play, social response and conversational skills) and emotional recognition [1]. The virtual characters contributed to the training of non-verbal communication (such as recognition of facial expressions and body gestures) and contributed to improvements in cognitive abilities (such as attention, executive functions and contextual processing of objects) [1].

The VR technologies have also been used to teach life skills, such as teaching, driving or shopping in a supermarket under safer and controlled environments [1]. While some advanced VR tools may not be available in remote parts of the world, many of the recent interventions are conducted using lower-cost technologies, which allow wider adoption and permit children to continue home-based treatment plans, especially for those who may have restrictions on mobility [1]. Several studies thus far have shown positive outcomes from VR-based therapeutic interventions among ASD children, and additional studies with larger group sizes could shed light on further improvements in these technologies to personalize therapies and become more interactive and effective.

Conclusion & future steps

Recent advances in identification of genetic biomarkers show promise for the early identification of ASD. Further deciphering the specific causative role of various genetic perturbations may enhance mechanistic understanding of the disease syndrome and further improve potential treatment options. The traditional and VR-based therapeutic interventions are showing great promise among ASD children. Additional studies focusing on applying these advancements and following precision medicine principles may significantly improve care for ASD patients.