The sincerest form of flattery

A reader just commented, “What other diseases can mimic PSP?” Below is a pretty exhaustive list of things that can cause vertical gaze palsy*, the most specific** diagnostic hallmark of PSP. Most of these disorders don’t mimic the whole classic PSP syndrome, but even PSP doesn’t do that in many cases. Keep in mind that most of these mimics have other features besides the gaze palsy, occur at much younger ages than PSP, or are exceedingly rare. For all those reasons, a good neurologist is unlikely to confuse these conditions with PSP in practice.

The disorders with specific treatment (though maybe not cures) have three asterisks ***.

*”Palsy” in general means weakness (not tremor, as popularly thought). In the setting of PSP, palsy refers to a limitation of the range of voluntary eye movements.

**The “specificity” of a diagnostic sign is technically the fraction of the people without the disease who don’t have the sign. In other words, specificity = [true negatives] divided by [true negatives + false positives].




  • Amyotrophic lateral sclerosis
  • Corticobasal degeneration
  • Dementia with Lewy bodies
  • Frontotemporal dementia with tau staining
  • Frontotemporal dementia with ubiquitin staining
  • Globular glial tauopathy
  • Lytico-bodig
  • Motor neuron disease with congophilic angiopathy
  • Multiple-system atrophy
  • Pallidal degeneration
  • Parkinson disease (only upgaze affected)***
  • PSP


  • Normal-pressure hydrocephalus***
  • Pineal region masses***
  • Third ventricular enlargement***

Metabolic / Genetic          

  • B-12 deficiency***
  • Huntington disease
  • Neuronal intranuclear inclusion disease
  • Niemann-Pick disease type C***
  • Spinocerebellar ataxia type 8
  • Tay-Sachs disease, adult-onset (hypometric vertical saccades)
  • Wernicke encephalopathy***
  • Wilson disease***


  • Anti-phospholipid syndrome***
  • Anti-IgLON4 disease***
  • Paraneoplastic syndromes***
  • Postencephalitic parkinsonism***


  • Cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL)
  • Lacunar states (“vascular PSP”)***
  • Post-aortic surgery


  • Whipple disease***
  • Neurosyphilis***


  • Guadeloupean tauopathy


  • Creutzfeldt-Jakob disease

A catalogue of autoimmune mimics

A post from two weeks ago describes a PSP-like condition caused by antibodies against a normal protein called IgLON5. Like most autoimmune disorders, it responded to immune-modulatory treatment.  Now, six cases of an MSA-like picture with antibodies against a different protein called Homer-3 has just been published by a research group in China. (See below for a downloadable pdf.)

Those patients’ conditions looked something like MSA in featuring ataxia, slowed movement and autonomic loss, and most of them responded to steroids or other immune modulation to some degree.  That report prompted an editorial by Eoin Mulroy, Bettina Balint and Kailash Bhatia of University College London summarizing what we know about immune disorders mimicking PSP, MSA and CBD.  I subscribe to that journal, Movement Disorders Clinical Practice, but I can’t upload the editorial to this public space without paying a hefty fee to the publisher.  So I’ll summarize it for you.

Drs. Mulroy and colleagues list 13 kinds of antibodies found in blood and/or spinal fluid that can produce some of the diagnostic features of one or more atypical parkinsonian syndrome. Each of the 13 antibodies attacks one of these proteins: CASPR2, CRMP5, DPPX, GAD, glycine receptor, Homer-3, Hu, IgLON5, Kelch-like protein 11, LGI1, Ma2, Ri and Sez612. 

  • The disorders mimicked by these syndromes are PSP (12 of 13), MSA (11 of 13) or CBD (3 of 13).
    • The one not reported to cause PSP symptoms is anti-Homer-3.
    • The two not reported to cause MSA symptoms are anti-DPPX and anti-MA2.
    • The three that have been reported to cause CBD symptoms are anti-GAD, anti-glycine receptor, and anti-IgLON5.
  • Only 4 of the 13 always start before age 60. 
  • Nine of 13 show evidence of inflammation on MRI in the form of high T2 signal in various spots.  
  • Routine tests on the spinal fluid (i.e., cell count, total protein, glucose) are normal in about half of cases.
  • A malignant tumor is commonly present in an organ outside the brain in 5 of the 13 and are “uncommon” in another 6.  Detection and removal of that tumor sometimes helps the neurological issue.
  • The good news is that in only 1 of the 13 is the response to immunomodulatory treatment described as “poor.”  It’s “good” in 2 and “partial” or “variable” in the other 10.

An important piece of advice is that the symptoms of these disorders develop from nothing to a state of some disability over weeks to months.  That’s called a “subacute” course. So there’s no point in doing these tests when the symptoms have developed over a period of years, a chronic course. 

None of the lab testing companies offers all of the antibody tests listed, and not everyone’s insurance covers tests from every company.  So the doctor (or patient/family/caregiver) will have to do some homework to figure out which tests are applicable to the specific symptoms, which companies offer the required tests and which testing companies work with that person’s insurance.