EPM 2025

My laboratory studies showed that host immune responses to S. neurona play a large role in defense against or succumbing to EPM. Initially, the Bayer studies(conducted in 2002) revealed a difference in acute and chronic S. neurona sarcocystosis and disease was biphasic. There is active interest in a serum test to detect EPM. A 2025 report (Analysis of IgG responses to Sarcocystis neurona in horses with equine protozoal myeloencephalitis authors Angwin, de Assis Rocha, Reed, Morrow, Graves, Howe) by Equine Diagnostic Solutions (EDS) in Ky. are looking into it. I say that because the authors Reed, Morrow, Graves and Howe declare a relationship with EDS. Their approach to develop a serum based EPM-detecting ELISA the Angwin paper examines IgG. By EPM they mean the presence of S. neurona in the central nervous system.

Their hypothesis is that IgG, adaptive immunity, is represented by four subtypes, two subtypes predominate in EPM and these are indicative of a Th1 cell mediated immune response and the other two are indicative of a Th2 humoral response. They are investigating the idea that EPM occurs due to an aberrant immune response in some horses. Finally! I have said that for years, the disease is a syndrome driven first by infection and then perpetuated by inflammation. They say that central nervous system pathology associated with EPM is, at least in part, immune based. I agree. They further examine ability to discern EPM by IgG subtype responses. And here I believe we part ways. I propose that the abiliity to regulated the innate immune response is where the bias in horses occurs.

The Angwin study collected sera and CSF from infected diseased horses and infected normal horses and used a modifed ELISA to quantify IgG subtypes 1/2, 4/7 (Th1) and 3/5 (Th2) that are detected in the sera.  Interestingly, they found 3/5 serum concentration and the ratio of  IgG 1/2:IgG 3/5 ratio were significantly different between diseased and normal horses. These data suggesting that the immune response to S. neurona in EPM horses is skewed towards a Th1, cell mediated response. They used a rigorous cutoff, at 25:1 for their diagnosis of EPM, a cutoff that was previously shown to provide a 98.8% specificity for EPM, according to Steve Reed in 2013.

After all was said and done the conclusion was there are insufficient differences for developing a serum-based immunoassay for EPM diagnosis.

The research is meaningful even if it doesn't provide an "EPM" test. These researchers tested the hypothesis that a Th1/Th2 imbalance is detectable in horses with EPM. When a type 1 cell mediated immune response is mounted, infection is controlled and there is a self-limiting infection; a type 2 response does not clear the infection and results in more severe disease.

This chart outlines important particulars of the experiment:

Th1 cell-mediated	IgG1/2 (IgGa)	IgG4/7 (IgGb)
Th2 humoral	IgG3/5 (IgG(T))	IgG6 (IgGc)
S. neurona-seropositive (116) control	Clinically healthy	EPM -negative (infected normal)
CSF EPM negative (30)	CVSM	EPM -negative (infected normal)
Serum (93) CSF (91) paired	Clinical EPM	Infected diseased

There are some red flags for us.  First, the Lewis et al. 2014 paper they cite was the experiment in which the in vitro assay of peripheral blood mononuclear cells from horses which were challenged with a SAG1 strain of S. neurona but the in vitro immune response assays were conducted with a SAG5 strain. The Lewis paper, while not in any way negating their conclusions, should be discounted as supporting or disqualifying a TH1/Th2 bias. Unless, of course, one is examining the responses of a horse continuously exposed to one serotype of S. neurona and suddenly exposed to another. In which case their data should including serotypeing the seropositive horses and the infected, diseased horses.

There is whole hearted agreement that immune responses generated against S. neurona in EPM horses are complex and difficult to characterize. One difficulty is defining a diseased population pre-mortem. That is something I discuss in developing clinical trials for drug effectiveness in previous blogs and in my book Natures Chamber of Secrets, release date pending. Nonetheless, the authors propose that despite the ambiguous data involving a delayed-type hypersensitivity response, lesions in the CNS of EPM horses could be due to an overly polarized Th1 immune response to S. neurona infections. They support their argument citing work from Peter Olias using S. calchasi infection of pigeons.

The pathogenicity of pigeon sarcocystosis was proposed by Dr. Peter Olias, Freie Universitat, Berlin in 2013. His paper was the beginning of my collaboration with him. After many conversations, his PhD student Kristina Maier visited me. Kristina camped out in Pathogenes lab for a few weeks and learned about our ELISA testing. I discussed the Olias pigeon results with Tom Kennedy (by then a trusted confidant) and David Lindsay, professor College of veterinary medicine Virginia-Maryland. The bottom line was that Sarcocystis produce CNS infections in the natural host and S. neurona was not uniquely neurotropic. At the time I was concerned with neurotrophism. The pigeon was not going to be a useful EPM model for several reasons, most important was that the Olias studies didn’t always detect parasites in the CNS in experimentally infected, neurologically abnormal birds. Disease and no parasites, not unique.

However, I was not off base suggesting inflammation as driving force for disease in horses especially in chronic or relapsing cases. Both the authors of the Angwin study and I looked to the results of experiments conducted in severe combined immunodeficient (SCID) foals infected with S. neurona (Sellon 2004). The experiment added evidence to the role of  Th1 immunity in EPM when compared to infections in immunocompetent animals, another experiment published at the time. I have used that argument based on the Sellon work myself on several occasions. I developed a model to test the efficacy of drugs against Sarcocystis. I was able to block the entry of parasites into leukocytes chemically, the horse was now SCID foal-like. The model is used to examine the parasitemic phase of infections. This is an important concept we teach veterinarians. Treating horses with parasite-blocking chemicals when the animals are in an environment that ensures a gut infection results in a parasitemia. The obvious fall out from such treatment is changing the natural Th1/Th2 balance in the host. Another in vitro assay I presented to the EPM society was the Resazurin dye assay. The assay showed a clear difference in the ability of horses to be infected with S. neurona. One could block specific receptors on a horse's cells and then run the Resazurin assay to determine susceptability to infection. The specific path of resistance could be revealed, if one were interested.

The Angwin study concludes that the dissimilarity in immune responses between infected normal and infected diseased horses remain unclear and they could not determine if the Th1-biased immune response is a predisposing fact to EPM development or a consequence of S. neurona neuroinvasion. I suggest another possibility, the disease is due to innate immunity, an inflammatory neuropathy using a bystander mechanism. My hypothesis is parasite induced dysregulation of innate immunity.

Angwin et al. suggest a path forward to determine if there are differences in immune responses in horses that succumb to EPM that could contribute to susceptibility and suggest immunophenotyping studies including using cytokine analysis and lipid profiling to help determine the immune mechanisms contributing to EPM pathogenesis. Their path forward will depend on significant EPM funding, with the dissolution of the EPM society due to lack of funding it will be interesting to see what develops. I also suggest that the puzzle can’t be solved with natural infections, it will take a horse model.

I conducted a study for GeneTraks using my Trojan Horse Model. They didn't find what they were looking for, a gene signature for EPM. GeneTraks looked at specific chages in genes in acute and chronic EPM and compared the results from challenged (n=8) horses, the diseased population, to unchallenged horses (n=5), the normal population. All horses were seronegative prior to the study. Here is a quick view of their report, I have a more in depth disusson in my book.

I presented results to the EPM society over the years that may help them navigate their future investigations, perhaps someone saved the presentations. I presented the Trojan Horse model to experimentally induce EPM.  I pointed to the biphasic nature of acute EPM (the Bayer studies) and the rapidity of decline of thecal antibody production in experimental infections. These experiments may hint to the timeline of a Th1/Th2 shift. I suggest that they refine their populations to remove Sarcocystis fayeri infections (EMS) or at least account for this common infection. There is no reason to ignore S. fayeri, my expectation is that it would produce an initial immune response that is similar to S. neurona. And finally, the presence of polyneuritis (PNE) in their populations should be given credence. My view is PNE is the pathogenesis of inflammation set off by S. neurona.

While the Th1/Th2 paradigm is a useful model, it is an oversimplification. There are other cells that play crucial roles in the related inflammatory neuropathies EPM, EMS, and PNE. I examined the relationships in these three conditions in 2017, Defining Relapses Attributed to Equine Protozoal Myeloencephalitis. I fouund S. fayeri toxin and polyneuritis were key players in disease identified by veterinarians as chronic, relapsing EPM. The production of autoantibodies is supported by  Th2 cytokines and is a pathogenetic mechanism that should not be overlooked. Horses in the Bayer studies developed autoimmune antibodies in the chronic phase of EPM. A large contribution to this line of research is providing a drug known to turn off specific pathways and will help future researchers answer their questions. The down side of my work is the oversimplification of sarcocystosis in social media.

I am pleased that Angwin and co-workers support inflammation as significant in EPM. The research will be hampered by acute versus chronic disease and the differing degree of responses by individuals.  In most cases, antibody levels do not correlate with degree of disease but rather presence of disease, a positive/negative result may be better. However, I would put my money on innate responses contributing to the pathogenesis of EPM rather than adaptive responses.

©2025 Pathogenes Inc. All rights reserved.