Autologous cord blood vs individualized supplements in autistic spectrum disorder: CORDUS study results

Abstract

BACKGROUND

Cellular therapies have started an important new therapeutic direction in autistic spectrum disorder (ASD), and the ample diversity of ASD pathophysiology and the different types of cell therapies prompt an equally ample effort to employ clinical studies for studying the ASD causes and cell therapies. Stem cells have yielded so far mixed results in clinical trials, and at patient level the results varied from impressive to no improvement. In this context we have administered autologous cord blood (ACB) and a non-placebo, material intervention represented by an individualized combination of supplements (ICS) to ASD children.

AIM

To compare the efficacy of ACB vs ICS and find markers correlated with the child's progress in order to better predict ACB efficacy.

METHODS

CORDUS clinical study is a crossover study in which both oral ICS and intravenous ACB were sequentially administered to 56 children; ACB was infused as an inpatient procedure. Treatment efficacy was evaluated pre-treatment and post-treatment at 6 months by an independent psychotherapist with Autism Treatment Evaluation Checklist, Quantitative Checklist for Autism in Toddlers and a 16-item comparative table score, after interviewing the children’s parents and therapists. Before and after each intervention participants had a set of blood tests including inflammatory, metabolic and oxidative markers, and the neuronal specific enolase.

RESULTS

No serious adverse reactions were noted during and after cord blood or supplement administration. ACB improved evaluation scores in 78% of children with age 3–7-years (n = 28), but was much less effective in kids older than 8 years or with body weight of more than 35 kg (n = 28; only 11% of children improved scores). ICS yielded better results than ACB in 5 cases out of 28, while in 23 kids ACB brought more improvement than ICS (P < 0.05); high initial levels of inflammation and ferritin were associated with no improvement. Ample individual differences were noted in children's progress, and statistically significant improvements were seen after ACB on areas such as verbalization and social interaction, but not on irritability or aggressive behavior.

CONCLUSION

ACB has superior efficacy to ICS in ASD; high inflammation, ferritin, age and body weight predict less improvement; more clinical studies are needed for studying ACB efficacy in ASD.

Key Words: Autistic spectrum disorder; Cord blood; Stem cells; Neuroinflammation; Autologous stem cell treatment; Neuronal specific enolase

Core Tip: CORDUS clinical study aimed to find factors correlated to therapeutic efficacy in autistic spectrum disorder (ASD) by comparing sequential administration of autologous cord blood (ACB) and individualised supplements and testing both the ASD children and the ACB. High initial levels of inflammation markers (erythrocyte sedimentation rate, C-reactive protein, tumor necrosis factor-alpha, α-2 globulins) or ferritin, but not high initial levels of neuronal specific enolase (NSE), were correlated with little or no improvement after cord blood administration. In some patients NSE was lowered after ACB. In a few patients improvement on specific supplements was superior to cord blood. Further studies are needed for finding predictors of cord blood efficacy in ASD.

INTRODUCTION

Since the first autistic spectrum disorder (ASD) patient was diagnosed 80 years ago[1] important progress was made, yet many aspects remain unknown on both its diagnosis and treatment. Children with ASD present with various manifestations-repetitive gestures, irritability, anxiety, sometimes aggressive behavior-and different degrees of impairment in verbalization, focus, understanding, interaction, initiative, following commands, overall intelligence quotient, corresponding to various areas of cortex affected[2,3]. Therapeutic options in ASD are limited both in variety and effectiveness, with psychotherapy being the main therapeutic intervention. In addition, antipsychotic medications such as risperidone and aripiprazole are prescribed for children exhibiting heightened agitation; atomoxetine is used to enhance focus in children with hyperactivity; and nootropics such as piracetam, pyritinol, cerebrolysin (a porcine brain protein hydrolysate), and actovegin (a deproteinized hemoderivative from calf blood) are sometimes employed, though results are often modest.

Cellular therapies are offering a new therapeutic perspective in ASD[4-6] by acting via different mechanisms[7,8], however their results vary greatly as it was observed in two recent meta-analyses[9-11] and this is likely due to both the polymorphic nature of the ASD pathogenesis, and the difference in the characteristics of the individual autologous cord blood (ACB) or stem cell products, aspects which will be discussed in the respective section below.

Designing a clinical study which involves the administration of umbilical cord blood (CB) is complicated by the presence of the obligatory cryopreserving agent dimethyl sulfoxide (DMSO), a special substance with unique properties. Known as a universal aprotic solvent for organic molecules including DNA, it also protects cells against the ice crystals which form intracellularly while freezing-the main factor damaging cellular architecture and organelles. DMSO is metabolized after intravenous infusion into dimethyl sulfide and dimethyl sulfone[12]; both metabolites contain sulfur, giving a characteristic odor detectable in the patient’s breath a few minutes post-infusion and lasting 24-48 hours. This distinctive sign complicates the blinding of a clinical study, an aspect which can be partially addressed by employing a crossover study design.

In addition to its vital role as a cryoprotectant, DMSO exhibits numerous actions at the cellular and molecular levels: (1) It acts as an antioxidant[13,14], prevents the accumulation of DNA breakage, and facilitates DNA damage repair[15-17]; (2) It has neuroprotective effects by reducing glutamate and NMDA activation[18], and positively effects neurons[19]; (3) It modifies blood-brain barrier permeability, reduces thrombosis risk, and performs other vascular actions[20-24]; (4) It induces stem cell differentiation into neuronal precursors through the activation of necessary intracellular pathways[25-29]; and (5) It inhibits the pro-inflammatory functions of leukocytes, including the secretion and actions of the tumor necrosis factor-alpha (TNF-α)[30], along with other immunomodulatory effects[31-35]. Due to its inhibitory action on leukocytes, DMSO is typically “washed” from the cryopreserved grafts used in hematology-oncology transplants, which aim to promptly reconstitute the hematopoietic function after high doses of chemotherapy.

In ASD cord blood is administered to patients who are not conditioned for transplant, so the possible deleterious DMSO action on white cells is not a factor; moreover, it is known that in a majority of ASD kids there is an overactive immune system (pro-inflammatory or pro-allergic). Considering this and because of its beneficial actions on neurons and neuronal precursors, we have chosen to not wash the DMSO out of the cord blood graft, and we have used the dilution technique for the decryogenation and preparation of cord blood for infusion. This technique has the added advantages of preserving the other components of cord blood (exosomes, growth factors, cells, etc.) which can be lost after centrifugation and removal of supernatant, and also keeps manipulation of the graft to a minimum, which optimizes the viability and vitality of cells infused, and aligns with the early work of Dr. Broxmeyer on cord blood manipulation.

The CORDUS clinical study received ethics approval from the National Bioethics Committee of the Romanian Medicine Agency ANM, No. IS/4/12.02 and is registered at www.clinicaltrials.gov (NCT04007224). Informed consent was obtained from the parents of all participating children prior to enrollment, with both parents required to their child's participation.

MATERIALS AND METHODS

Participants in this study included children diagnosed with ASD, aged 3 years to 7 years, with a body weight of 15 kg to 30 kg, who had previously undergone multiple treatments, including psychotherapy, without achieving significant progress.

To evaluate differences in efficacy and safety, exceptions were made for children aged 8 years or older and/or those weighing more than 30 kg; these children also received treatments but were not included in the final statistical analysis, which only considered ASD children meeting both inclusion criteria-age and body weight. A flowchart summarizing the study design is represented in Figure 1.

Figure 1  CORDUS study design.

Before and after the two treatments (supplements and cord blood), CORDUS study participants underwent the following blood tests: (1) Blood type/rhesus factor (Rh); (2) Human leukocyte antigen (HLA)-A, HLA-B, and HLA-DRB1 typed from both peripheral blood and the child’s cryopreserved cord blood; (3) Complete blood count with differential; (4) Liver function tests—aspartate transaminase (AST), alanine transaminase (ALT), and/or gamma-glutamyl transferase; renal function (serum urea, serum creatinine); (5) Electrolytes—sodium (Na), potassium, calcium; (6) Neuronal specific enolase (NSE); (7) Markers of inflammation [erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), TNF-α, ferritin, α-2 globulins on serum protein electrophoresis]; (8) Anti-neuronal antibodies (Ac. Anti-amphifizin: Ac. Anti-CV2:Ac. Anti-PNMA2; Ac. Anti-Ri:Ac. Anti-Yo:Ac. Anti-Hu:Ac. Anti-recoverin; Ac. Anti-SOX1:Ac. Anti-titin); and (9) Individually (only some kids) [interleukin (IL)-1 beta, IL-6, IL-8, neopterin, procalcitonin, cationic protein of eosinophils, homocysteine, thyroid stimulating hormone (TSH), growth hormone (GH), insulin-like growth factor (IGF)-1, etc.].

Administration of supplements

The administration of supplements was individualized for each ASD child, considering data from pre-clinical studies[36-39] and ASD patients[40-42], as well as blood test results. The following guidelines were used: (1) For immune dysfunction type I: Over-activation of pro-inflammatory actions (type M1 macrophages, etc.) resulting in subacute inflammation (increased TNF-α or α-2 globulin fraction which occurred in about 70% of the 56 ASD kids tested[43]: Uridine or Boswellia (extracted from Boswellia serrata) or curcumin (from Curcuma longa); (2) For immune dysfunction type II: Over-activation of allergic-type immune reaction (activation of type M2 macrophages, etc.) resulting in eosinophilia and/or increased immunoglobulin E and/or cationic protein of eosinophils (about 30% of ASD kids): Blackcurrant extract (Ribes nigrum contains a natural steroid) and/or Viola tricolor extract; (3) For high ferritin (in a few cases is associated with hemoglobinopathies): One or more antioxidants-glutathione; polyphenols from blueberry (Vaccinium mirtillus extract), ascorbate, N-acetylcysteine, etc.; (4) For high homocysteine (most likely due to suboptimal folate metabolism or receptor issue): Methylcianocobalamin (vitamin B12 conjugate) and folate or leucovorin (folinic acid); (5) For low homocysteine: Antioxidants, resveratrol or antioxidants which cross the blood-brain barrier–proanthocyanidins from blueberry, luteolin (flavonoid, the main yellow pigment in Reseda luteola), astaxanthin (carotenoid red pigment from algae), zeaxanthin (carotenoid alkaloid from plants); (6) For low ferritin: An antioxidant and a cell membrane stabilizer-fish or vegetable oil, omega 3, docosahexaenoic acid or eicosapentaenoic acid, and vitamin D3; (7) For intestinal dysbiosis: Recommended supplements include probiotics (particularly Bifidobacterium salivarius) and prebiotics or inulin (fructose-containing oligosaccharides) to reduce intestinal inflammation, Candida sp proliferation, inflammatory cytokine formation, amines, and balance serotonin levels; (8) For high lactate and/or lactate dehydrogenase: Mitochondrial enhancers such as Pyrroloquinoline quinone, Uridine, luteolin to improve aerobic glycolysis and the pyruvate/lactate imbalance; (9) For metabolic or liver issues (increased AST, ALT, bilirubin): Astragalus (Astragalus lentiginosus), antioxidants; (10) For low GH, IGF-1: L-arginine; (11) For high TSH-spirulina (Arthrospira platensis) or Kanchanar guggul extracts; (12) If demyelination on magnetic resonance imaging: Bacopa (from Bacopa monnieri), citicoline (cytidine diphosphate choline), plus an anti-inflammatory–Boswellia, curcuma; (13) For high NSE values (especially above 30 pg/mL): Administration of supplements which stimulate and support neurogenesis, natural anti-inflammatory agents and anti-oxidants; and (14) For anxiety, agitation and focus deficits: Supplements containing combinations of Passiflora, Humulus lupulus, Valerian (Valeriana officinalis), chamomile (Chamomilla recutita and Chamomilla nobile), trillium (from Trillium species), echinacea (Echinacea purpurea), gamma-aminobutyric acid (GABA), theanine (extracted from Camellia sinensis), or Rhodiola rosea extracts.

Administration of ACB

The administration of ACB was performed following confirmatory blood group, Rh, and HLA typing, ensuring that each graft contained a minimum of 5 × 10⁶ total nucleated cell count (TNC)/kg. Two grafts showed microbiological contamination but were successfully decontaminated prior to infusion, preserving stem cell viability[44]. Each cord blood graft was used integrally. Both contiguous compartments (20 mL and 5 mL) were processed by adding Dextran 40 and albumin to minimize osmotic modifications of the cells post-decryogenation. The Rubinstein decryogenation protocol[45] was applied, employing minimal handling of the decryogenated graft and using the dilution technique to minimize cellular stress and prevent loss of cells and exosomes. After decryogenation and while preparing the cord blood for infusion, cell viability was analyzed extemporaneously with Trypan Blue 0.4% under a brightfield microscope at 200 × magnification. The prepared CB was administered intravenously (IV) after pre-medicating the child with antihistamines and hydrocortisone hemi-succinate[5]; IV sedation with midazolam and ketamine was also administered as needed.

Cord blood was infused via gravitational drip; volumes infused ranged from 47 mL to 75 mL, with infusion times between 20 minutes and 35 minutes, followed by saline flushing of the IV line with 75-150 mL of saline solution. A small sample (1-2 mL) from the infused cord blood was taken from the administration bag for future testing. Approximately 2-6 hours post-infusion, flow cytometry was performed on the sample prepared for infusion. This analysis was performed at an independent laboratory—The Biochemistry Institute of The Romanian Academy—where stem cell viability was analyzed using 7-actinomycin D and the presence of surface cellular markers CD34⁺, CD45⁺, CD271⁺, and CD133⁺ was examined. These results will be detailed in a separate article.

Psychometric evaluation

Psychometric evaluation was conducted by a clinical psychologist, who interviewed both the parents and the therapist of each child regarding the child's behavior before and after the treatments. Assessments included the Autism Treatment Evaluation Checklist (ATEC), Quantitative Checklist for Autism in Toddlers, and CAST questionnaires, along with a 16-item symptom table, to evaluate initial behavior and post-treatment behavior after 4-8 weeks of supplement use and approximately 6 months after CB infusion.

The psychometric evaluation by the clinician during office visits may be influenced by various factors, including unexpected changes in the child's health on that specific day (e.g., a “cranky” child with acute infections or pain), as well as day-to-day mood fluctuations (e.g., “our child has good days and bad days”), or the parental mood that the child may mirror (“having a bad day”). Such factors may increase anxiety and reduce the child's cooperation, thereby introducing potential bias into evaluations. This unintended bias might explain why some studies report statistically significant differences when using the Autistic Behavior Checklist and Childhood Autism Rating Scale but not with the Vineland Adaptive Behavior Scale[7,11].

The evaluation table used by the psychotherapist comprised 16 items, where parents and therapists were prompted to assign grades from 0 (indicating complete absence of a symptom/behavior) to 10 (indicating maximum intensity for the respective symptom/behavior). Parents and therapists graded 6 positive items (attention, understanding, verbalization, social interaction, initiative, and maintaining focus) and 10 negative items (irritability, hyperactivity, stereotypies, inadequate verbalization, aggressivity, withdrawal, somnolence, headaches, constipation/diarrhea), along with one general health item (specific manifestations such as food preferences in type, shape, texture, color, taste, smell; fears/phobias, nail biting, etc.). Scores were analyzed individually, comparing scores for the same items before and after treatments, as well as total scores for both positive and negative items.

Statistical analysis

All resulting data from the study were organized in Microsoft Excel, and statistical analyses were performed using the same software by employing descriptive statistics (means, SD), Pearson correlation for parametric data, Spearman correlation for non-parametric data, and the two-tailed paired t-test, with P < 0.05 considered statistically significant.

RESULTS

Cord blood and supplements were administered to a total of 56 children with ASD, of whom only 28 met the inclusion criteria of being younger than 8 years, weighing less than 30 kg, and available for psychometric evaluations. These 28 participants were included in the final data analysis, while psychometric data from the children not meeting the study's inclusion criteria (28 children aged > 8 and weighing > 35 kg) were excluded from the psychometric analysis. For cord blood administration, the mean dose of TNC per body weight was 25.01 × 10⁶/kg ± 13.84 × 10⁶/kg, with a respective mean CD34⁺ count of 68.9 × 10³/kg ± 64.3 × 10³/kg.

Blood test data from all 56 children, along with the respective results and analysis, were published in a recent article[43], which discussed all values obtained following blood tests. Exceptions were made for enrollment to determine whether age and body weight were important determinants of treatment efficacy; findings indicated that these were indeed significant and their inclusion helped increase the study's ability to identify blood markers with predictive value for treatment efficacy.

Regarding safety, there were no major adverse reactions to either the supplements or the administered cord blood. The most common side effect encountered after both treatments was temporary agitation, seen only in children who had previously experienced such episodes (around 30%). This agitation was managed with natural anxiolytics/sedatives (Passiflora, Chamomile, GABA) or with risperidone/aripiprazole, which the child had previously been taking. Three children were subsequently able to discontinue risperidone or aripiprazole following cord blood administration. Agitation typically occurred the day after either new supplements or cord blood administration, likely due to nonspecific, global stimulation of neuronal activity; some episodes lasted for hours and were intense but were manageable and reduced as supplements or medication were tapered. Agitation was more common following CB administration (15 patients) than after supplements (4 patients). Additionally, side effects related to cord blood administration included red urine during the first micturition after CB infusion in 6 children, which was self-limiting, resolving spontaneously. This is likely explained by the elimination of hemoglobin released from some of the red cells infused and destroyed shortly after infusion. Four children experienced vomiting after post-infusion, likely due to gallbladder spasm; this also resolved without treatment. One child, with a history of allergies and atopic dermatitis, experienced brief facial flushing during infusion, which also resolved spontaneously.

Following cord blood administration, improvements in ASD symptoms varied widely, from truly transformative effects (in about 10% of children aged 3-7, with parents stating, “it was like we saw a different kid”) to no observed effects (also around 10% of parents). For children older than 8 years and weighing over 35 kg, improvements were significantly reduced in both occurrence and amplitude, with only one in ten children (around 10%) showing improvement after cord blood administration.

Analysis of the scores on the 16-item table for all 28 children showed that the mean scores before (i) and after (f) CB administration revealed statistically significant improvements across all 6 “positive” areas (items 1-6): (1) Attention; (2) Understanding; (3) Verbalization; (4) Social interaction; (5) Initiative; and (6) Maintaining focus. The two-tailed t-test yielded P < 0.05 when comparing the respective scores before and after CB administration (Figure 2A).

Figure 2 Initial (i) and final (f). A: Post-treatment psychometric scores on the positive subscales; B Initial: and final psychometric scores on the negative subscales after autologous cord blood administration.

Comparing the mean scores for all 28 children before (i) and after (f) CB administration on the 10 “negative” items, we have seen significant improvements in 4 areas (hyperactivity, stereotypical behavior, inadequate verbalization and varia). However, no significant improvements were noted in the remaining 6 items: (1) Irritability; (2) Aggressivity; (3) Withdrawal; (4) Somnolence; (5) Headaches; and (6) Constipation/diarrhea. Initial and final psychometric scores on subscales 7-16 before and after CB infusion (Figure 2B).

When considering all 6 positive items, their combined mean scores were significantly better after CB infusion, with an average score of 29.11 ± 10.14 SD before infusion and 38.16 ± 9.83 after infusion (P < 0.05). Surprisingly, the combined scores for the 10 negative items also showed a significant improvement after infusion, with scores of 29.17 ± 9.96 after CB infusion compared to 32.62 ± 11.09 before infusion (P < 0.05).

The ATEC total score for all children before and after CB administration indicated significant improvement (two-tailed t-test, P = 0.0051). The t-tests also showed statistically significant improvements on the subscales for speech (P = 0.0043) and socialization (P = 0.012), but not on the sensory/cognitive (P = 0.09) or overall health/behavior (P = 0.18) subscales. In the case of supplement administration, after analyzing the total scores on both ATEC and the 16-item table scores before and after administration of supplements, neither showed a statistically significant improvement, making the overall CB efficacy superior to that of individualized supplements. However, in 5 of 28 kids (age < 7 years, body weight < 35 kg) the improvement obtained with supplements was bigger than that observed after CB administration.

While analyzing evaluations conducted by the psychotherapist, which were all recorded and transcribed, we noted discrepancies in some cases between the very enthusiastic statements made by some parents during interviews and the minimal scores resulting from the completion of the ATEC and the 16-item table by the same parents. For example, one parent (P1) remarked, “The second day after transplant ..., it was like he was another kid”, while the respective improvement on the 16-item table was only 8 points. Another parent (P2) stated, “The kid had an explosion for the better after both treatments with stem cells”, yet only a 2-point improvement on the total score of the 16-item table was recorded. A third parent (P3) noted, “Extraordinary good evolution after cell administration. He started saying words”, with a respective improvement on the 16-item table totaling just 1 point. These discrepancies may be due to unidirectional progress in the child, where only one aspect improved, such as verbalization, or possible differences in the perceptions of parents compared to the quantification by the questionnaires. This underscores the subjective nature of evaluations and the importance of interviewing the therapist who worked with the child before and after treatments. In our study, a strong correlation was observed between the evaluation scores given by parents and those provided by the respective psychotherapists, with a Spearman’s rho = 0.7.

Given the variability of results following CB administration, we sought to identify markers that could predict the efficacy of CB administration, particularly correlations between various blood markers and the psychometric evaluations.

When analyzing all blood test values (including children older than 7 years and/or those weighing more than 30 kg, n = 56) for NSE, TNF-α, and α-2 albumin, we found a strong correlation between inflammation markers—specifically, TNF-α and α-2 albumin (rho = 0.883). However, there was only a moderate correlation between NSE levels and TNF-α (rho = 0.508) and between NSE and α-2 albumin (rho = 0.502), suggesting that factors beyond inflammation, such as genetic, metabolic, and oxidative stress, contribute to the neuronal apoptosis observed in children with ASD.

To evaluate the predictive value of blood markers for the success of CB administration, we analyzed the psychometric score modifications (dPS) and focused on the group of children meeting both inclusion criteria (age and body weight, n = 28).

There was no correlation between the levels of TNF-α and NSE (r = 0.052). When considering only the ATEC scores, a very small correlation (r = 0.138) was found between initial TNF-α levels and improvement in ATEC scores (initial vs final score difference; dATEC = initial ATEC – final ATEC), while a better correlation was observed between initial NSE levels and dATEC (r = 0.401) (Figure 3). This suggests that treatment may be more efficacious in children in whom neuronal destruction is more pronounced—specifically those with higher initial NSE values.

Figure 3  Initial and final Autism Treatment Evaluation Checklist scores–before and after cord blood infusion.

It is possible that the baseline presence of both inflammation (abnormal TNF-α and/or α-2 globulins) and neuronal destruction (NSE > 20 pg/mL) may serve as good prognostic factors for the therapeutic success of stem cells in ASD; however, the CORDUS study was not designed or powered to study this correlation. We plan to investigate the predictive value of these two markers taken together, as well as other potential markers (such as miRNA profiles in both the ASD child and the administered cord blood), in a future clinical study.

DISCUSSION

In a previous study it was found that a good predictor for the efficacy of cord blood in ASD was the posterior beta power on electroencephalogram (EEG) recorded prior to cord blood infusion, so that higher values were associated with increased alpha and beta and decreased theta power on EEG 12 months post-infusion[46]; however, the validation of this predictive marker was not pursued in subsequent studies.

Considering the data from our study and the correlation values between dPS and various markers from Table 1, it can be inferred that the cord blood treatment was more efficacious in children with higher initial NSE values. Conversely, high initial values of inflammatory markers—CRP, ESR, TNF-α, α-2 globulins, and ferritin—were associated with poorer outcomes following cord blood administration.

Table 1 Correlation of psychometric score improvement with initial markers.

Correlation of psychometric score improvement	C-reactive protein (initial)	erythrocyte sedimentation rate (initial)	Tumor necrosis factor-alpha (initial)	α-2 globulins (initial)	Ferritin (initial)	Neuronal specific enolase (initial)	Total nucleated cell count/kg	CD34+/kg
Rho coefficient	-0.34	- 0.63	-0.53	-0.4	-0.57	0.4	0.08	0.53

Considering the big picture, ASD is linked to a wide variety of causative factors, which can be categorized based on their levels of action into genetic, epigenetic, and allogeneic/environmental factors (e.g., infections, nutrient imbalances, toxins)[47]. These factors may act alone or in combination, making it challenging to identify and treat the predominant factor. To date, more than 800 genes have been associated with ASD pathogenesis, and these genes are involved in pathways active in cell cycle, metabolism, adhesion and signaling, inflammation, and cancer[48-50].

The phenotypic expression of these genes results in abnormal levels of various proteins; proteomics analysis has shown that more than 100 proteins are dysregulated in ASD, including cytokines, markers of mitochondrial dysfunction, oxidative stress, and epigenetic markers indicating deficient methylation[51].

A critical question regarding ACB administration is whether a child with ASD and genetic problems can benefit from their own cord blood, which is likely affected by the same genetic issues, and we considered that the neurotrophic and immunomodulatory actions of stem cells and exosomes are still possible. This approach led us to administer autologous CB in a child with a GABBR2 mutation, resulting in significant improvements in both the child’s NSE values and ATEC score. Nonetheless, this success needs confirmation in other cases.

Treatment-wise, due to the complexity of the genetic landscape in ASD, it is improbable that a single common ASD determinant in the form of a target molecule [e.g., nuclear factor kappa B (NF-κB), mammalian target of rapamycin, or sirtuin 1] could be successfully modulated by a therapeutic agent to achieve a cure for ASD.

Cellular therapies, particularly cord blood, have inherent advantages over other therapeutic modalities, as they allow for the simultaneous administration of a wide range of molecules with various biological actions contained in extracellular vesicles (EVs) or exosomes. This characteristic complements the actions of the administered stem cells, which can replace damaged neurons beyond innate neurogenesis.

Genetic brain anomalies associated with ASD—such as brain mosaicism due to copy number variations, mosaic chromosomal alterations, single nucleotide polymorphism (SNPs), etc.—can potentially be ameliorated by cord blood and other stem cell treatments by providing neuronal precursors from the administered stem cells while neuronal destruction is increased (indicated by high NSE levels) and finally by supporting the formation of synapses integrated into functional neural networks via exosomes, which improve mitochondrial function in neurons and modulate microglia primarily by decreasing its pro-inflammatory functions [e.g., increased glial fibrillary acidic protein (GFAP)].

Compared to adult stem cells, cord blood has a number of advantages: (1) It contains about eight times more colony-forming cells with high proliferative potential than bone marrow; (2) It has a higher proportion of more primitive hematopoietic stem cells, with CD34⁺ cells exhibiting more adhesion molecules such as CD44 proteoglycan and integrins like CD49d or CD49f, and longer telomeres—capable of more self-renewal; and (3) It also has a higher proportion of immature B cells (CD19⁺ and CD5⁺) and a characteristic precursor of T cells (CD3⁻/CD8⁻). Moreover, these cells are less sensitive to potential toxic environmental substances, and CD34⁺ cells from cord blood home better to injury sites due to a higher affinity for stromal cell-derived factor[52].

Additionally, cord blood contains mesenchymal precursor cells that are more primitive and capable of generating cells from all three germ layers[53], and CB has more proteins with different properties than adult-derived cells. Proteomic analysis has shown that perinatal-derived EVs differ from adult mesenchymal stem cells (MSCs) in their enrichment of key proteins involved in immune, metabolic, and regenerative pathways[54]. Generally, perinatal stem cells generally exhibit a better overall genetic status than adult stem cells, as each cell division produces an average of 2-3 genetic mutations per generation, resulting in approximately 80 SNPs in more than 2% of somatic cells. Furthermore, about half of all adults harbor at least one somatic mutation with function-altering potential in the brain[55].

Given that subacute systemic inflammation is present in a majority of ASD children[56-58], alongside neuroinflammation[59,60], the administration of naïve, non-opsonized leukocytes and their precursors from cord blood can replenish T and B lymphocytes with unaffected ones and re-establish the balance between pro-inflammatory and anti-inflammatory cytokines affecting microglia in ASD. Increased, sustained immune cell activation is followed by impairment in their function, associated with reduced thymic production of T cells and impaired antigen presentation and production[61]. Moreover, the activation and proliferation of immune cells depend on mitochondrial energy production, with various roles for the tricarboxylic cycle and oxidative phosphorylation; thus, an extended inflammatory state and mitochondrial exhaustion may lead to alterations in immune function, both innate and acquired[62].

As previously mentioned, the significant variability in the effects of CB infusions is likely due to qualitative differences in the cord blood administered, contributing to the sometimes opposing results observed in clinical studies analyzing outcomes[63-65]. Cord blood contains both hematopoietic and MSCs in varying proportions, and indications suggest that the ratio of these two populations significantly impacts the site of stem cell implantation (lung vs brain) after intravenous administration[66]. The capability of MSCs to generate new neurons in the central nervous system (CNS) can thus be greatly influenced by the properties of the cord blood. In the CORDUS study, we analyzed the CD133⁺/CD271⁺ ratio, as well as the numbers and ratios of CD45⁺ and CD34⁺ cells; the results of the flow cytometry testing and correlations with psychometric evaluations will be published in a subsequent paper.

Cord blood of various origins also contains exosomes of differing quality and quantity, which may account for the discrepancies in results observed after its administration. The dilution technique used for decryogenation of the cord blood graft—employed in our study—offers the advantage of minimal manipulation, potentially increasing cell viability while preserving all contents of the graft, including exosomes and very small embryonic-like stem cells that would otherwise be excluded as supernatants after centrifugation. We also plan to test for differences in exosome content by profiling the microRNAs in both the cord blood and the ASD children receiving treatment, as significant differences have been observed in microRNA testing microRNAs in ASD children by various research teams[67-70].

Beyond genetic, epigenetic, and environmental causality, ASD causative factors can also be classified based on the primary organ affected—whether intrinsic or extrinsic to the CNS[43]. For example, a mutation directly affecting neurons (e.g., GABBR2) is considered intrinsic to the CNS (primary ASD), while a metabolic or immune system dysfunction resulting in systemic inflammation (evidenced by microglia activation as indicated by increased glial fibrillary acidic protein—GFAP, NF-κB, and other markers such as TNF-α and α-2 globulins) or, in rare cases, the presence of antibodies to neuronal proteins or folate receptors is considered extrinsic to the CNS (secondary ASD). In these secondary/extrinsic cases, specific treatments can yield good results.

Following blood tests in children with ASD, supplements can be administered more specifically and can serve as an effective initial treatment before considering CB or other stem cell treatments. In addition to the examples mentioned above, there are other useful observations: (1) In children with low ferritin, there is likely an association with ferroptosis, which occurs as a compensatory mechanism for increased lipid oxidation and neuronal membrane degradation. While iron supplementation provides noticeable short-term improvement, especially in sleep quality, an antioxidant or cell membrane stabilizer is more beneficial for long-term administration; (2) If marginally low serum Na is repeatedly observed (e.g., 137 mmol/L; normal is above 138 mmol/L) while other electrolytes remain normal, there may be a Na transporter defect, and the child may benefit from administering the diuretic bumetanide or torasemide[71,72]; (3) In children with associated hyperactivity, viloxazine may yield better results than the non-stimulant atomoxetine [73]; (4) For children with anxiety, agitation, and/or focus deficits that are not improved with the aforementioned supplements, the administration of an adaptogen—Rhodiola—along with magnesium citrate and vitamin B6 may help; and (5) Short-term administration of low-dose aripiprazole may also be necessary. However, epigenetic factors such as methylation or acetylation should also be considered in this situation, and administration of folate/cobalamin derivatives or S-adenosyl-methionine can improve underlying deficits.

Even though stem cell administration in ASD has not yet yielded consistent, solid results, making it difficult to recommend as standard second-line therapy in clinical practice[74], it is worth noting the exceptional outcomes that have followed cord blood administration in some children (approximately 1 in 10 children) and the fact that a majority of children (3 out of 4) aged 3-7 years and weighing less than 30 kg demonstrated improvements. At same time, the heterogenicity of ASD causal factors and also of the characteristics of ACB are prompting the need for pursuing a set of selection criteria and efficacy predictors for cord blood treatment in future clinical studies.

CONCLUSION

Results from the CORDUS study show that ACB administration in children with ASD is beneficial for those aged 3-7 years, with approximately 3 out of 4 children showing variable but clear improvements, particularly in verbalization, initiative, social interaction, and understanding. Children older than 8 years showed significantly less improvement, with only 1 out of 10 children older than 8 years demonstrating clear behavioral improvement. ACB administration presents a very good risk/benefit ratio due to the minimal risk involved (compared to unknown sources, which, although thoroughly tested, carry an undetermined risk for future complications). The benefits of cord blood appear to reside in immune modulation and regenerative capacity, both directly from the infused cells and from the exosomes, which influence the activity of other cells in the body, particularly immune cells, microglia, and neurons. Testing ASD children for a variety of inflammation, metabolic (including folate and methylation), and oxidative stress markers allows for the administration of specific supplements, which in some cases yielded better results than ACB administration.

Considering the excellent risk profile of administering autologous stem cells, we advocate for further studies on autologous stem cell administration in ASD through new clinical trials, particularly to identify prognostic factors associated with positive outcomes that will enable us to select patients likely to benefit from stem cell treatment and further understand and treat ASD.