IVIG and IMIG as upstream neuroinflammation modulators — where the evidence stands, why a new delivery route may be mechanistically preferable, and what a clinical trial would need to show.
See framework disclaimer belowEvery intervention in the Biology of Autism framework addresses a node somewhere along the cascade — from IDO1 inhibitors that reduce kynurenine shunting, to NAD⁺ precursors that restore SIRT1 function, to compounds that target the somatostatin brake directly. Immunoglobulin therapy is different. It acts before the cascade loads — at the level of the cytokine environment that drives IDO1 upregulation in the first place.
The significance of this entry point is not merely positional — it is mechanistic. Treating downstream consequences while the upstream inflammatory driver remains active is like addressing individual symptoms of a leaking pipe without closing the valve. Immunoglobulin therapy targets the valve. For children with documented immune dysregulation, it may be the most direct available intervention.
IVIG has been studied in autism longer than almost any other biological intervention. A 2021 systematic review and meta-analysis by Rossignol and Frye, published in the Journal of Personalized Medicine, examined 27 publications reporting the therapeutic use of IVIG in individuals with ASD — including four prospective controlled studies, six prospective uncontrolled studies, and fifteen retrospective case series and reports.
"IVIG is a promising and potentially effective treatment for symptoms in individuals with ASD; further research is needed to provide solid evidence of efficacy and determine the subset of children with ASD who may best respond to this treatment as well as to investigate biomarkers which might help identify responsive candidates."
Rossignol & Frye, Journal of Personalized Medicine, 2021The consistent finding across studies is that gains are most pronounced — and most durable — in children with identified immune abnormalities: low total IgG, IgG subclass deficiency, elevated pro-inflammatory cytokines, specific polysaccharide antibody deficiency, or autoantibodies to brain proteins. This is not a treatment for autism broadly defined. It is a treatment for a biologically defined subpopulation in which immune dysregulation is a measurable, active driver.
One critical and consistent finding across multiple studies: when IVIG was stopped, gains were frequently lost. This is not an isolated observation from a single study — it appears across the literature wherever investigators followed outcomes after discontinuation. Boris et al. (2005, n=26) reported that 85% of treated children lost some improvements after stopping. Maltsev & Yevtushenko (2016, n=78) found that 50% of children with mild-to-moderate improvement lost their gains within 2–4 months of completing therapy. Plioplys (1998) documented complete regression in the one child who had shown a remarkable response when treatment was stopped. Knutsen & Fenton (1998) reported worsening of seizures when IVIG was discontinued in a child who had previously improved. The consistency of this pattern across studies, patient populations, and dosing protocols is itself informative: it is not a treatment failure. It is a signal that the underlying inflammatory process continues without ongoing modulation, and that durable benefit requires sustained immunomodulation rather than a completed course. The implication for trial design — and for understanding why a steady-state delivery route may matter — is significant.
IVIG and IMIG are not the same formulation. They differ in concentration, volume, route of administration, cost, and critically — in how serum IgG levels behave over time. IVIG is formulated at 5–10% IgG for intravenous infusion and produces a rapid, high peak in serum IgG within 24 hours. IMIG is formulated at 16.5% IgG for intramuscular injection in small volumes (1–5 mL) and produces a slower, lower, but more sustained IgG elevation.
IVIG vs IMIG — a pharmacokinetic distinction with clinical implications. IVIG produces a high serum IgG peak within 24 hours, then declines over 3–4 weeks, creating oscillating inflammatory suppression. IMIG delivers a slower, sustained IgG elevation — a steady-state profile that may be mechanistically preferable for cascade biology, where SIRT1 recovery, CREB/BDNF-dependent plasticity, and SST normalisation all require weeks of consistently reduced neuroinflammation, not periodic windows of relief.
| Feature | IVIG — Intravenous | IMIG — Intramuscular |
|---|---|---|
| IgG concentration | 5–10% (large volume) | 16.5% (small volume, 1–5 mL) |
| Serum IgG kinetics | Rapid high peak, then 3–4 week decline back toward baseline | Slower rise, lower peak, sustained steady-state elevation |
| Administration setting | Infusion centre, IV access, medical supervision for hours | Outpatient IM injection — clinic or practice, monthly |
| Cost per course | $10,000–$25,000 per infusion (US); R50,000–100,000 (SA) | Materials cost approximately $50 per treatment (NBI Intragam, South Africa — repriced from prior below-cost supply). Comparable global product Beriglobin (CSL Behring, Switzerland) is $80 per equivalent dose. Both represent a fraction of IVIG at $10,000–$25,000 per infusion — accessible in resource-limited settings globally. |
| Anxiety burden | IV access, clinic environment — significant anxiety for many ASD children | Less invasive environment; easier for sensory-sensitive patients |
| Regulatory status (ASD) | Off-label — not approved for ASD in any jurisdiction | Investigational — no published RCT in ASD as of 2024 |
In 2024, the first published case report of intramuscular immunoglobulin specifically in ASD and PANS (Pediatric Acute-onset Neuropsychiatric Syndrome) appeared in Medical Research Archives (European Society of Medicine). The report describes a monthly IMIG protocol using Intragam (National Bioproducts Institute, South Africa, 16% IgG) administered at 0.2 mL/kg body weight intramuscularly on a 4–6 weekly basis.
The more modest ASD response compared to PANS is interpreted by the authors as a timing effect. In PANS, the autoimmune trigger is often acute and relatively recent. In ASD, neuroinflammation typically begins in early development — and by the time IMIG is initiated, structural neural changes may have already accumulated. The logical implication is that earlier intervention should produce a stronger response. This is a hypothesis that can be tested. It is also the reason why biomarker-guided patient selection — using the cascade framework's immune dysregulation markers — is central to any future trial design.
The existing evidence base — Rossignol & Frye's meta-analysis establishing IVIG's signal, and Fourie & Armstrong's first published IMIG case report — establishes proof of mechanism and preliminary clinical signal. What it does not yet provide is the controlled, biomarker-stratified evidence that would allow IMIG to move from investigational to accessible.
For context on the opportunity cost: a phase II trial of this scope represents a fraction of the cost of a failed late-stage pharmaceutical trial — and produces data that could support regulatory approval for the first biologically targeted, affordable treatment for a defined ASD subpopulation. No such approved treatment currently exists anywhere in the world.