Plan B · Learn
reading the test

Reading an anti-neuronal antibody panel

A child gets a full autoimmune encephalitis workup. The hospital tests come back negative. A commercial panel shows fourteen elevated antibodies. Both results are real. Here is how to hold them.

They are not the same test

Gold-standard assays (cell-based assays and immunofluorescence, usually run through a major reference lab, often on both serum and spinal fluid) test for the recognized autoimmune encephalitis antibodies: NMDA receptor, LGI1, CASPR2, GAD65, AQP4, GFAP, AMPA, GABA-B, DPPX, MOG.

Commercial neural panels are semi-quantitative, run many more targets, and use a lower cutoff. They are not validated as diagnostic tests, and mainstream neurology does not accept them as such.

A negative gold-standard panel genuinely rules out classic autoimmune encephalitis. That is real, reassuring information.

How to read a wall of small positives

When a commercial panel shows twelve or fourteen results barely over the cutoff, that pattern is more consistent with general immune activation — or assay noise — than with a dozen separate antibody syndromes.

What is worth attention is a result that stands alone — one value at two or three times the cutoff while the rest hover at the line. That is a signal; the crowd around the cutoff usually is not.

The honest evidence tiers

  • Established: post-streptococcal autoimmunity is real — Sydenham chorea is the proof. Anti-neuronal antibodies can cause neuropsychiatric disease.
  • Emerging and contested: commercial anti-neuronal panels as diagnostic tests. The antibodies are real; their clinical utility as a test is disputed.
  • The honest framing: a panel result is a supporting signal, not proof. The clinical story — infection, then a two-day-later collapse, over and over — is stronger evidence than any antibody number.

What it can unlock

Even a contested result can be useful when it lines up with the story. Combined with elevated strep markers and an infection-triggered pattern, it supports a coherent case — and cases are what open doors to anti-inflammatory trials, immune workups, and, where warranted, immune-directed treatment.

Repeating a panel in 8–12 weeks is often more informative than the first result. If a value tracks with your child’s clinical state, that means considerably more than a single snapshot.

Want this answered for your own child? A Plan B synthesis reads every lab and note you have — together — and tells you what is driving the picture, what has never been checked, and what to ask for next.

The first one is free, and you keep the list either way. Phone photos of labs are fine.

Start a free synthesis
Keep reading
Your Lyme tests disagree. Here is how to read that.
Specialty tick-borne panels and standard two-tier testing often disagree. Why a negative Quest or LabCorp result does not mean your child was never infected.
The immune workup almost nobody orders
Immunoglobulins, IgG subclasses and pneumococcal titers — the standard, covered blood tests that explain why some children cannot stop getting reinfected.
How old is too old for a lab?
How to tell which old results are still useful, which need rechecking, and why a stale abnormal result should never be treated as today’s emergency.

Plan B for PANS is written by a parent, not a physician. Nothing here is medical advice or a substitute for your child's clinicians — it is meant to make you the best-informed person in the room when you talk to them.