11 July 2021 Episode 153 : Auto-immune phenomena associated with COVID and with vaccination

Dear colleagues,

Triggered by several observations, suggestions, hypotheses and speculations, I tried to find scientific literature on this topic.  It is a very scattered literature and part of it relies on case reports, of which the causal relation and pathogenesis remain to be established.  

1. Auto-immunity associated with COVID disease?

Episode 153-1: Auto-Ab against type 1 IFN

1. The well-known Science paper of Sept 2020 by Bastard et al. Showed that amongst 982 patients with severe COVID auto-antibodies against type 1 IFN, with a prevalence of at least 2.6% of women and 12.5% of men of all ages, whereas they were absent in mild and moderate COVID and rare (< 0.5 %) in healthy controls.

2. These data have been recently confirmed in a single site (Lyon), where 25 % of the critically ill COVID patients were positive for anti-IFN alpha 2 Ab, which again were absent in mild cases and also in healthy controls, but were positive in all patients with autoimmune polyendocrinopathy type 1 syndrome (APS-1) . Remarkably, the clinical characteristics were not different between critical patients with or without type 1 IFN Ab.  The authors provide 2 interesting suggestions:

• Use anti-type 1 IFN auto-Ab in risk assessment upon admission to hospital
• Use IFN-beta in therapeutic trials (as no auto-Ab against this subtype of IFN)

3. The suggestion that APS-1 syndrome is a predisposing factor for severe COVID was confirmed by Bastard: out of 22 young APS-1 (8-48 yrs) SARS-CoV-2 patients 21 had anti-type 1 IFN Ab. Of these 19 were hospitalized, 15 in ICU, 11 mechanical ventilation and 4 died!!

Episode 153-2: Anti-phospholipid antibodies and the pro-coagulation state in COVID-19.

As you probably know, the term “lupus-anti-coagulant” (LA) derives from the capacity of serum Ig from some patients to interfere with (i.e. prolong) the “activated pro-thrombin time” (aPTT). 

Paradoxically, this in vitro finding of anti-coagulation can be  associated in vivo with enhanced coagulation and thrombosis.  It is dependent on the presence of anti-phospholipid antibodies (aPL) with different specificities: anti-cardiolipin (aCL), anti-beta2 glycoprotein I (aβ2-GPI) of different isotypes (IgM, IgG, IgA) that essentially activate platelet aggregation.   

These aPL are permanently found in auto-immune diseases, but also transiently in various inflammatory and infectious conditions, which are not always clearly associated with enhanced thrombosis, but also in elderly with chronic disease.

The present review tries to summarize 48 studies on the occurrence of aPL in COVID. They are frequently found and it is quite possible that they are involved in the enhanced thrombotic event.  However, the authors feel that the study have not been done carefully enough.  Especially there is no follow-up to see if the aPL found during acute COVID disappear or persist after recovery: if they persist, they may be an underlying risk factor ,  but if they are transient, they could just be an inflammatory epiphenomenon.

It would be important to compare mild moderate and severe COVID, with and without thrombotic events

• Measure LA, IgG/IgM aCL, andIgG/IgM aβ2-GPI systematically
• Importantly check their persistence after 3 months.
• I would add: also check their affinity: high affinity more likely to be pathogenic.

Episode 153-A/B:  Guillain-Barré syndrome (GBS) and spinal/encephalo-myelitis

There is increasing evidence of an association between COVID and neuro-myelitis (inflammation of white matter) either peripheral (i.e. GBS) or more central (spinal of encephalomyelitis).  GBS can range from a very mild case with brief weakness to nearly devastating paralysis, leaving the person unable to breathe independently. Peroipheral or central myelitis has been found in association, usually after, infection with various agents (e.g. Zika, Chikungunya, West-Nile…) and is classically caused by molecular mimicry, eliciting anti-ganglioside Ab, leading to inflammation of the white matter (myelitis) and sometimes demyelination.  These phenomena lead to neurological symptoms: parestesias, weakness or more central: confusion, behavioral changes etc…

In case of SARS-CoV-2, there is a growing number of case reports, indicating that GBS or other forms of myelitis may not only occur post-infection (possibly contributing to long COVID), but also during the acute phase (para-infectious), even as the sole symptom in patients with PCR(+) SARS-CoV-2.

The pathogenic mechanisms have not been investigated thoroughly:

• Viral RNA is not usually found in the neural system (cerebrospinal fluid)
• Anti-Ganglioside Ab have not been really investigated,
• There is also a theoretical possibility of molecular mimicry between SARS-COV-2 proteins and human heat shock proteins with ensuing auto-immunity
• Some authors believe that the ”hyper-cytokinemia” could be involved.    

Clearly, lots of unresolved questions.

Episode 153-4: Primary biliary cholangitis (PBC) and auto-immune hepatitis

1. A case report of a 44 yrs old women with hypertension and auto-immune (Hashimoto) thyroiditis, who after critical COVID pneumonia (with intubation) developed both Guillain-Barré and very typical PBC (with cholestasis and anti-mitochondrial Ab).
2. Another case report of a  57-year-old man with a medical history of hypertension, prediabetes and beta thalassemia minor, who was diagnosed with COVID-19 and subsequently developed fatigue and arthralgias, and whose blood showed hyperferritinemia, elevated liver enzymes (AST/ALT/GGT), hypergammaglobulinemia, anti-smooth muscle antibody, anti-mitochondrial antibody, and anti-double-stranded DNA antibodies. The patient was diagnosed with autoimmune hepatitis–primary biliary cholangitis overlap syndrome triggered by COVID-19   
3. Several cases of “secondary sclerosing cholangiopathy”, after critical COVID, where direct liver damage by the virus, rather than auto-immunity is suspected.

Episode 153-5: Auto-Ab against ACE-2

1.  Early 2020 Jean Amiral et al. proposed the following hypothesis in Transfusion and Apherisis Science: 

Formation of a strong complex between ACE-2, a self-component, and viral S1 (or S Protein), constitutes the basic context for developing allo-antibodies and generating a delayed autoimmune response, with antibodies

• first targeted to viral antigens,
• but which can extend to the associated self-component through epitope spreading

In view of the physiology of ACE-2 (see Fig. 1 p. 2)….this delayed autoimmune response could contribute to the cytokine storm and generate tissue injury and destruction.

This can activate hemostasis, beyond acute hypercoagulability, stimulate tissue injury, and totally impair the body’s regulation of hypertension and physiological defense mechanisms such as existing pathways of haemostasis/thrombosis and inflammatory processes

This hypothesis summarizes indeed some characteristic inflammatory and potentially auto-immune manifestations and is easily testable. 

2. A follow-up preprint in J Appl Lab Med shows that amongst 98 COVID patients (nicely distributed over mild-moderate-severe, 12 had antibodies against ACE-2.  However, a careful comparison of clinical and various lab characteristics (including in dept analysis of hemostasis parameters) failed to find any association

2. Immune-mediated side effects of COVID vaccination

Ep 153-6:  Comparison between VITT (Vaccine Induced Thrombotic Thrombocytopenia) , HIT (Heparin-Induced Thrombocytopenia) and PIC  (Pulmonary Intravascular Coagulopathy)

PIC = very common COVID-associated hyper-coagulopathy, mostly in older people (men>women). The term PIC is used, because of the pulmonary origin.  Whereas many authors point to “endotheliopathy” (infection and/or damage of pulmonary endothelial cells by SARS-CoV-2)  as a cause and others suspect anti-phospholipid Ab (see above) , these authors rather think the viral RNA itself activates coagulation and is responsible for thrombosis.  In a first time, local thrombosis may help to limit the infection to the lung, but after breakdown of the barrier, the RNA in circulation (RNaemia) will directly activate the coagulation cascade and also the platelets. The widespread predominantly arterial thrombosis will result in ischemic damage of multiple organs. 

(Clearly, platelet factor 4 antibodies are NOT involved)

HIT = heparin-induced thrombocytopenia, caused by formation of complexes between negative heparin and positive PF4, approx. 5 days after heparin therapy.  These complexes will exceptionally elicit anti-PF4 auto-Ab  Following heparin therapy for thromboembolic disease, heparin sequestration and further PF4 entrapment may cause clot extension and new thromboembolism and also thrombocytopenia.

VITT = typically after SARS-CoV-2 Adenoviral vector vaccination. Negative DNA, complexed with positive PF4 also breaks immune tolerance, mainly in young women. The anti-PF4 Ab lead to platelet and neutrophil activation, increased platelet consumption and thrombosis.

The presentation of the 10 first cases after the Astra-Zeneca vaccine is particular and very different from PIC and HIT: 

• 9 patients had cerebral venous thrombosis, 3 patients were diagnosed with pulmonary embolism, 3 patients had splanchnic vein thrombosis, 4 patients had other thromboses,
• 5 patients developed disseminated intravascular coagulation (DIC)
• there were 6 fatalities

There are nice illustrative figures and a clear table.

Ep 153-7: A case report of cerebral venous sinus thrombosis 2 weeks after the first dose of mRNA SARS‑CoV‑2 Pfizer vaccine.  This Malaysian patient had a cardiovascular history and had abandoned anti-platelet therapy a few months before vaccination.  A causal relation with the vaccine is very uncertain.

Ep 153-8 A/B:  Three cases of auto-immune hepatitis in associated with either Pfizer or Covishield (= Indian Astra-Zeneca).  Temporal association is clear, but causality not proven. 

Ep 153-9: A compilation of 27 cases of various auto-immune diseases after either RNA (23 Pfizer, 2 Moderna) or Adenovirus vaccines (2 Astra-Zeneca) in Israel, UK and US.

• In 6 cases, this was a new disease, in 21 it was a flare of an existing condition.
• The rheumatic-musculoskeletal diseases included; rheumatoid arthritis, Behcet disease, Systemic Lupus, vasculitis, polymyalgia rheumatica and seronegative symmetrical synovitis with pitting edema.
• The non-rheumatic included myasthenia gravis, multiple sclerosis, neuro-sarcoidosis, pericarditis.

The cause-effect relation was not sure, in view of the rarity (I could not find the “denominator”: how many vaccines were administered in total)

The authors conclude: Despite the high population exposure in the regions served by these centers, immune-mediated diseases flares or onset temporally-associated with SARS-CoV-2 vaccination appear rare. Most are moderate in severity and responsive to therapy although some severe flares occurred.

Ep 153-10:   As could be expected from experience with other vaccines, also cases of Guillain-Barré have been reported after vaccination. Remarkably,

• All cases that I found are after the ChAdOx1 (Astra-Zeneca-Covishield).
• It is often a “variant” form with bifacial weakness with paresthesia’s
• The prevalence is so low that definite causal association has not been confirmed.

Ep 153-11: An information leaflet of the children’s Hospital in Philadelphia, where the rare side effects of common vaccines (MMR, polio, Yellow Fever, Flu) and those of COVID vaccines are briefly compared with the risks of the disease. The risk/benefit ratio is rather  clear !

In summary:

1. Auto-antibodies against type-1 IFN have a clear causal relation with severe COVID. The occurrence of anti-IFN may be part of a poly-endocrine syndrome.
2. The pathogenesis of the very frequent thrombotic events in severe COVID remains a matter of debate: endotheliopathy, anti-phospholipid Ab, RNaemia?
3. The pathogenesis of the very rare Vaccine Induced Thrombosis and Thrombopenia is supposedly due to anti-Platelet Factor 4. Until now mostly related to Adenoviral vectors (only 1 case report after mRNA).
4. Neurological complication during severe COVID may be related to various factors (low oxygen saturation, cardiovascular events), but there is emerging evidence of peripheral (Guillain-Barré) and spinal/cerebral myelitis/demyelination.  Both para-infectious and post-infectious (long COVID).  Pathogenesis not yet elucidated.
5. Guillain-Barré has also been described after vaccination with ChAdOx1-SARS-CoV-2.   
6. Other forms of presumed or proven auto-immune diseases have been observed sporadically during COVID or after vaccination: auto-immune hepatitis and cholangitis, rheumatic and non-rheumatic immune diseases.  The causal relation has not yet been established. Remarkably, these phenomena were more associated with mRNA than with Adenoviral vaccines.
7. It is very evident that serious complications of the disease are much more frequent than proven and presumed serious side effects of the vaccines.  

I will be on holiday until 20 July and will come back with more episodes in 2 weeks.

Best wishes,

Guido