Autism is a neurodevelopmental condition associated with a multifaceted presentation. This disorder is typically characterized by; deficits in communication and social reciprocity, hyperactivity, attention deficit, repetitive behavior and restricted interests. It usually begins in infancy and a definitive diagnosis is typically achieved at around 24 - 36 months [1]. According to the March 2014 report released by the United States Centres for Disease Control and Prevention, the prevalence of autism is an estimated 1 in 68 children [2]. This is a sharp increase from the previous report released in March 2012, which suggested the prevalence to be 1 in 88 children [2,3]. Various pathomechanisms have been implicated in the onset of autism which includes genetic factors, immune dysregulation and hypoperfusion. Current treatment options available for autism comprise of occupational therapy, behavior therapy, speech therapy, nutritional guidance and pharmacological treatment. However, these treatment options are directed towards achieving symptomatic changes in the affected individual but the underlying brain damage remains unaddressed.

In autism, MRI of brain is normal hence, functional neuroimaging techniques such as PET - CT scan of the brain should be explored to study the effect of intervention. MRI of the brain gives information about the structural abnormalities of the brain. It does not record the cellular dysfunctions at the metabolic level. Whereas, PET CT scan uses biomarkers to record the metabolic changes at the cellular level [4]. Hence, PET CT scan brain is more sensitive in analyzing the effects of cell therapy on the function of the brain as compared to MRI. In this case report, 18 FDG PET CT scan has been used as a monitoring tool to study the effect of intrathecal administration of autologous bone marrow mononuclear cells (BMMNCs) in autism.

Herein, we present a case of autism that despite being treated with conventional medical therapies and treatments continued to experience major communication and behavioral deficits. Cell therapy has been widely studied and has shown to have potential as a novel treatment for various neurological disorders including autism [5-10]. These cells are safe, easily obtainable (involving less invasive techniques), non - teratogenic and have no ethical issues.

Case presentation

In this report, we present the case of a 6 year old boy with autism who, after showing limited response to other medical and rehabilitation interventions underwent autologous BMMNC transplantation. The boy was born by normal delivery at full term, cried immediately after birth, with normal birth weight and had no neonatal complications. He had delayed developmental milestones and his speech was not developed. He was diagnosed with autism at the age of two years. No family history of autism was reported. He had normal vision and hearing. Clinical manifestations included hyperactivity with attention and concentration deficit. He also presented with stereotypical and self - stimulating behavior like finger counting, rocking and swinging. He had fair eye contact, but had difficulty maintaining it and had poor sitting tolerance. There was presence of self - injurious behavior like head banging along with behavioral issues like screaming and throwing things. His appetite was appropriate while there were occasional disturbances in his sleep patterns. Functionally, he was dependent for all Activities of Daily Living (ADLs), with a lack of complete toilet - training. He had poor social interaction and primarily used need based communication as his linguistic abilities were severely deficient. He exhibited abnormal emotional responses like irrelevant laughter, poor imitation skills, and inappropriate object use. He also exhibited sensory issues like hypersensitivity to noise and tactile hypersensitivity (preferred soft articles and liked the touch of rugs). He also presented with poor pragmatic skills, delayed response time, poor proximal joint stability, hyporegistration for sensory stimulation and poor awareness of surroundings and body.

He scored 113 on Indian Scale for Assessment of Autism (ISAA), 42 on Functional Independence Measure (FIM) and 42.5 on Childhood Autism Rating Scale (CARS). Severity of illness on the Clinical Global Impression scale (CGI - I) was 5, categorizing him as markedly ill.

Magnetic Resonance Imaging (MRI) of the brain did not reveal any significant morphological abnormality. An electroencephalography (EEG) showed epileptiform activity from the bilateral centro - parietal regions, with no history of seizures. A Positron Emission Tomography–Computed Tomography (PET - CT) scan of the brain using the radioisotope 18 - F FDG (fluorodeoxyglucose) was performed on a Siemens Biograph HD MDCT with LSO detector technology. The PET - CT scans (Figure 1) revealed moderate to severe hypometabolism (reduced FDG uptake) in bilateral cerebellar hemispheres, along with mild hypometabolism in the bilateral thalami, amygdale, mesial temporal lobe. Hypermetabolism was observed in frontal lobe and middle frontal gyrus.

Figure 1: (A) PET - CT scans of the patient carried out pre - intervention. (B) Post - intervention PET - CT scans of the patient carried out after 6 months. Previously hypometabolic (hypofunctioning) brains areas depicted as blue areas show improved metabolism post - intervention, represented by green areas. Red areas representing hypermetabolism (hyperfunctioning) have improved to more normalized green areas post - intervention. The images reveal a balancing effect on overall brain metabolism.

Intrathecal administration of autologous BMMNCs was used as a treatment method. The rationale of this intervention is in accordance with the inclusion criterion of the World Medical Associations Helsinki declaration [11]. It has been reviewed and approved by the Institutional Committee for Stem Cell Research and Therapy (IC - SCRT). The parents of the patient were informed about the procedure and a duly filled informed consent form was obtained. Pre - procedure routine blood tests and chest x - ray were carried out to rule out active infection and assess fitness for anesthesia. MRI brain, PET CT scans brain and EEG was performed. 300 mcg of Granulocyte colony - stimulating factor (G - CSF) injections were administrated 72 hours and 24 hours prior to BMMNC transplantation, to stimulate CD34+ cells and increase their survival and multiplication [12]. A 100 ml volume of Bone marrow was aspirated from the iliac bone. The density gradient separation method was used to obtain Mononuclear Cells (MNCs). The MNCs were then evaluated for CD34+ by FACS analysis and viable count was calculated and found to be about 98%. Approximately 2.35×108 MNCs were administered intrathecally immediately post separation, in the space between L4 - L5 using a lumbar puncture needle. 1 gm Methylprednisolone in 500 ml Ringer’s Lactate (RL) was simultaneously injected intravenously to reduce local inflammation.

The patient also underwent rehabilitation therapy which included occupational therapy, psychological therapy and speech therapy. During rehabilitation sessions, effective motor learning strategies with task oriented training for real life environment were utilized and successful attainment of functional outcomes were achieved. The rehabilitation sessions were recorded and a CD with the exercises was given to the patient for home program, to be followed rigorously for next 6 months. No adverse events were reported throughout the course of the therapy.

The patient showed significant improvements over the period of 6 months. At the follow up assessment after three months of receiving the therapy, the patients’ hyperactivity had reduced along with a minimal improvement in sitting tolerance. Post - treatment, his awareness of surroundings, his understanding and eye contact had also improved.

On the follow up conducted six months after the therapy, all previously reported improvements were sustained. Hyperactivity had further reduced. Command following also improved and there was a considerable reduction in self-injurious behavior. His social interaction with other children improved. Reduction in tactile hypersensitivity was noted. He also developed efficiency in carrying out basic ADLs like eating. His ability to learn new tasks, problem solving and concept forming improved; making it possible to perform tasks like finishing his homework. He could also perform coloring activities more efficiently.

The score on ISAA (Table 1) reduced from 113 (moderate autism) to 97 (mild autism) at six months. The score on CARS improved from 42.5 to 35.5. His FIM score also improved from 42 to 50. The severity of illness score on CGI - I reduced from 5 (markedly ill) to 4 (moderately ill). Global improvement on CGI - II scale suggested ‘much improvement’ with a score of 2. The efficacy index on CGI–III scale was scored as 5 which was suggestive of moderate therapeutic effect. Moderate therapeutic effect was characterized as definite improvement with partial remission of symptoms, without any side effects.

Table 1: Table showing the patients score on different sub - components of Indian Scale. for Assessment of Autism (ISAA) pre - intervention and at 6 months follow up post - intervention.

A PET - CT scan conducted 6 months post cellular therapy revealed improved metabolism in Amygdala, Cerebellum, Thalamus and Mesial temporal lobe bilaterally; along with improvements in hypermetabolic areas like Middle frontal gyrus, Frontal lobe and Parietal lobe. The mean standardized uptake values (SUVs), depicting improved metabolism in the different areas of brain (Table 2(A) and 2(B)).

Table 2 (A): Comparison of pre and post treatment SUV and SD values of previously hypometabolic areas in 18 FDG PET - CT scans and their correlation with clinical symptom improvements in the case study.

Table 2 (B): Comparison of pre and post treatment SUV and SD values of previously hypermetabolic areas in 18 FDG PET - CT scans and their correlation with clinical symptom improvements in the case study.

PET CT brain imaging was used as a monitoring tool to study the effect of intervention at the cellular level as other imaging studies like MRI do not give information about the function of the brain tissue and in most cases of autism it does not reveal any significant abnormality. PET CT utilizes 18 - FDG, a dye that is analogous to glucose. The dye is entrapped in the brain cells, which can then be measured on the CT scan giving a diagrammatic representation of the function of cells. Function of the brain cells is directly proportional to the glucose uptake and metabolism. Thus, hypofunctioning areas will depict reduced FDG uptake and hypometabolism; while hyperfunctioning areas will depict increased FDG uptake and hypermetabolism [13,14]. Uptake of the FDG dye by cells of a particular area is measured and a mean value of that area is calculated and recorded as mean SUV (Standard Uptake Value). These values are then compared to the control group values by the software and the standard deviation (SD) is calculated and recorded as SD value.

In this study, a comparative PET - CT investigation done after six months of intervention showed increased uptake of FDG in amygdala, cerebellum, thalamus and mesial temporal lobe which were previously hypometabolic or hypofunctioning. A reduced FDG uptake was observed in the middle frontal gyrus, calcarine fissure, as well as in the frontal and parietal lobes which were previously hypermetabolic or hyperfunctioning (Table 2(A) and 2(B)). These findings correlated with the clinical improvements recorded in the patient on follow up.

On follow up, significant clinical and functional improvements were observed and recorded on objective scales. The ISAA score reduced from 113 to 97 and correlated well with the observed clinical improvements. Significant improvements were observed in five out of six ISAA subcomponents; these were social relationship and reciprocity, emotional responsiveness, behavior patterns, sensory aspects and cognitive component. The CARS showed a significant improvement as the score reduced from 42 to 35.5, suggesting a shift from severe autism to mild - moderate autism. The child’s FIM score also improved from 42 to 50. This change was supported by the boy’s ability to become more independent in his ADLs. Severity of illness of the CGI scale showed an improvement from 5 (markedly ill) to 4 (moderately ill). Global improvement on CGI scale graded him with a score of 2 (much improved). The efficacy index on CGI was noted to be 5 i.e. moderate therapeutic effects. This indicates that there was decided improvement with partial remission of symptoms with no side effects.

Autologous BMMNCs have been extensively studied and shown to be a promising therapeutic approach to address the core neuropathology of autism. It has the potential to repair neural tissue damage at the molecular, functional and structural levels. These cells are a combination of hematopoietic cell populations such as CD34+ cells and non - hematopoietic cells like mesenchymal stem cells (MSCs), stromal cells and tissue specific progenitor cells [15]. The rationale behind this therapy lies in the knowledge that the functional effects of BMMNCs arise from a delicate balance between the beneficial effects of different cell types. These cells can be easily obtained and their safety and potency have been well defined with no ethical issues.

Inflammation, inflammatory response and immune activation have been strongly implicated in the pathogenesis of autism [3]. Several studies exploring this aspect have found abnormalities in the peripheral immune system, while related studies have revealed other immune disruptions involving activation of the innate neuroimmune system and microglial cells [16-18]. Evidence suggests that the function of the blood brain barrier (BBB), a crucial regulator of brain homeostasis, is altered in autism due to neuroinflammation, increased inflammatory cytokines and immune dysregulation in the brain [17,18]. An imbalance of various cytokines like IFN - γ and interleukins (ILs) has been reported in the peripheral blood, along with an increased activation of pathways involving Th1 and Th2 [19].

A crucial effect of BMMNCs is immunomodulation through an increase in anti - inflammatory TGF - β and IL - 10; along with an inhibition of pro - inflammatory cytokine production of TNF - α, IL - 1β and INF - γ. They also show paracrine effects and induce angiogenesis by the production of vital growth factors like fibroblast growth factor (FGF2), vascular endothelial growth factor and ciliary neurotropic factor (CNTF). Immunomodulation counterbalances the dysregulated immune system and restores normal functioning by reducing neural damage [20].

Hypoperfusion has also been recognized as a major underlying pathology of autism. It causes abnormal neurotransmitter or metabolite accumulation and hypoxia, further causing neural tissue damage [17]. The severity of symptoms observed in autism is believed to be directly proportional to the degree of hypoperfusion. One of the important effects of BMMNCs is neoangiogenesis. This reduces hypoxia by improving clearance of toxic metabolites and perfusion [21,22].

The neuroanatomy studies in autism have revealed the involvement of the cerebellum, frontal lobes and mesial temporal lobe (amygdala) along with an imbalance in the connectivity of the brain [23-25]. The imbalance in the excitation - inhibition pathways of neurons plays a crucial role in autism. A recent study suggests that during the development of an autistic brain, excitatory (glutamatergic) neurons remain unaffected while inhibitory (GABAergic) neurons increase. BMMNCs through its paracrine effect, neurorepair, synaptogenesis, axonal sprouting improves the neuronal connectivity and the excitatory - inhibitory balance [26,27].

Intrathecal administration of BMMNCs, a relatively less invasive procedure allows the cells to breach the BBB, migrate to and “home” into the affected brain areas through the CSF [28]. Since the intervention is autologous, BMMNCs avoid the risk of a rejection making it a safe and feasible therapeutic approach.

Hence, we hypothesize that these multitudes of reparative mechanisms restore the connectivity in the autistic brain resulting in clinical improvements. The improved cellular functions are reflected on the PET CT findings as a balancing effect of brain metabolism in autism.

Though a solitary case, the significant improvements with objective PET - CT findings demonstrate that autologous BMMNC transplantation may be a feasible and safe therapeutic approach in the treatment of autism. However, to ascertain the therapeutic benefits of cellular therapy in autism and to compare the benefits of different types of cells and different routes of administration; multicentre randomized controlled trials with larger sample size are necessary. PET - CT scan can be used as a more sensitive monitoring tool and outcome measure for assessing benefits of cellular therapy in combination with standard rehabilitation.