TNF-α blockade is a successful treatment for several autoimmune diseases. However, in rare cases a paradoxical induction of an additional autoimmune disease occurs. According to the literature, the frequency of anti-TNF-induced autoimmunity ranges from 0.03% (demyelinating CNS events) to 10.7% (psoriasis) (1). With her informed consent, we report the case of a 33-year-old woman, who suffered from hidradenitis suppurativa (HS), an inflammatory skin disease, and then developed Vogt-Koyanagi-Harada-like disease (VKHLD) affecting her eyes and brain under adalimumab therapy. Uveitis has been described as a rare adverse event of anti-TNF-α therapy, and in a retrospective study four patients were described with a co-occurrence of HS and uveitis, two with anterior uveitis, one with iridocyclitis and one with panuveitis. These patients developed uveitis between one and twelve years before HS was diagnosed, notably, none of them were on anti-TNF-α therapy. Thus far, no cases of HS and VKHLD have been reported in the literature (2, 3). Our patient had experienced sudden onset of blurred vision in the left eye, which was diagnosed as acute posterior uveitis, and persistent headaches. Adalimumab was discontinued, and she was hospitalized for treatment with systemic corticosteroids, cyclosporine and mycophenolate mofetil. Following treatment, the patient gradually improved and achieved remission of her VKHLD during a 12-week follow-up period. For the treatment of her HS she was later switched to bimekizumab, a novel anti-IL-17A/F antibody. Here we discuss common pathomechanisms of HS and Vogt-Koyanagi-Harada disease (VKHD) and immune mechanisms of TNF-alpha blockade that might lead to autoimmunity. Additionally, we discuss the action of the new therapeutic antibody, which targets IL-17A and IL-17F, and the consequences of blocking IL-17F with respect to the gut microbiome and autoimmune diseases.

Hidradenitis suppurativa (HS) is an inflammatory, chronic skin disease characterized by recurrent abscesses, with a key role of tumor necrosis factor-alpha (TNF-α) in the pathogenesis (4). TNF-α inhibitors are used to treat various inflammatory and autoimmune diseases, including HS (5), but in rare cases, adalimumab has been associated with paradoxical inflammatory adverse events (PAEs), such as posterior uveitis (2). Uveitis comprises any type of intraocular inflammation and may occur as an isolated disease or be accompanied by other inflammatory disease. In addition to uveitis, TNF-alpha inhibition can also induce other autoimmune diseases such as demyelination, sarcoidosis, vasculitis, vitiligo, alopecia areata, psoriasis and thyroiditis (6–8) as well as VKH syndrome-like disease (VKHLD). VKHD is a rare inflammatory disorder that affects multiple organs, including the eyes, ears, meninges, and skin. The target of the immune response are melanocyte-specific proteins (7, 9, 10). VKHD typically begins with a prodromal phase characterized by symptoms such as headache and neck stiffness, followed by an acute uveitis phase with inflammation of the choroid, optic nerve, and retina. As the disease progresses and becomes chronic it often causes sequelae such as scarring and pigment changes in the affected parts of the retina and choroid and depigmentation of skin and hair (10).

A 33-year-old Asian woman with a history of hidradenitis suppurativa (HS) and ongoing adalimumab therapy for two years presented with sudden blurred vision in her left eye, along with severe headache and fever. She used to feel very stressed because of her hidradenitis, which was only slightly improved by adalimumab treatment. When she started adalimumab, her chest-X-ray was normal for Tb, but she received 2 x 300 mg rifampicin as a prophylactic anti-Tb treatment. One month before deterioration of vision the patient had a severe cold but was tested negative for COVID. At presentation best corrected visual acuity (BCVA) was normal in the right eye (logMAR 0.0) and count fingers (logMAR 1.9) in the left. Intraocular pressure was 19 and 21 mm Hg in the right and left eye, respectively, and slit-lamp examination of the anterior segment of both eyes was unremarkable. Her family history was negative for HS and VKH disease.

Most autoimmune diseases are driven by Th1 cells secreting IFN-γ, and/or Th17 cells characterized by IL-17 secretion. Both cell types also produce the pleiotropic cytokine TNF-α, which is the major mediator of destructive inflammation. Therefore, targeting TNF-α is thought to simultaneously downregulate both T cell types and the inflammatory responses of TNF-α-producing innate cells.

Adalimumab is a widely used biologic anti-TNF-α-agent for moderate to severe hidradenitis suppurativa and many other autoimmune diseases, including rheumatoid arthritis, ankylosing spondylitis, Crohn’s disease, and uveitis of the posterior segment. It works by blocking the binding of TNF-α to its receptors (4, 5). TNF-α exists in two forms, soluble and membrane-bound, both of which signal through TNFR1 and TNFR2 receptors.

The soluble form of TNF-α plays a central role in systemic inflammation by driving cytokine production, recruiting immune cells, and activating endothelial signaling pathways. The membrane-bound TNF-α can be cleaved off to generate soluble TNF-α but also serves as a receptor to promote cell-to-cell communication and regulation of the immune response. Adalimumab binds primarily to soluble TNF-α, preventing it from binding to TNF receptors, and even can scavenge already receptor-bound TNF-α (13, 14).

There are two receptors for TNF-α. TNFR1 is ubiquitously expressed at low levels on most cell types, whereas TNFR2 is inducible on endothelial and immune cells, such as monocytes, macrophages, natural killer cells, and T and B lymphocytes (13). While membrane-bound TNF-α activates both, TNFR1 and TNFR2, soluble TNF-α only activates TNFR1 (15, 16). The two TNF receptors trigger different signal transduction cascades and therefore have different effector functions. TNFR1-activation is pro-inflammatory and induces apoptosis or even necroptosis via its intracellular death domain. In contrast, TNFR2 binding leads to cell survival, tissue regeneration and activation by interaction with the cytoplasmic TNF-receptor associated factor (TRAF2), finally leading to antiapoptotic stimulation of various kinases (17).

When TNF-α binds TNFR2, IL-17A production is inhibited, because TNFR2-binding activates the anti-inflammatory TNF-α-induced protein 3 (TNFAIP3/A20). A20, the gene product of TNFAIP3, blocks TNF-α-induced NF-kB signaling via degradation of the receptor interacting protein kinase 1 (RIP1), resulting in decreased IL-17A production. However, blocking the interaction of TNF-α with TNFR2 inhibits TNFAIP3/A20 and thus promotes IL-17 production (18). Since many mutations resulting in TNFAIP3 dysfunction have been described that cause severe inflammatory syndromes and diseases (reviewed in (19)), many of the patients not responding to anti-TNF-α therapy may have a mutation in their TNFAIP3 gene. Anti-TNF-α-mediated prevention of TNFR2-binding disrupts anti-inflammatory processes mediated by nuclear factor-kappa B (NF-κB), which results in the production of other pro-inflammatory cytokines such as IL-6, IL-1, IL-2, granulocyte-macrophage colony-stimulating factor (GM-CSF), and interferon-γ, upregulating inflammatory responses via TNFR1 activation. On the other hand, TNF-α also induces negative feedback regulators, such as IL-10 and corticosteroids, thus playing a key role in maintaining the balance between pro- and anti-inflammatory responses (20, 21). Inhibiting TNF-α can disrupt this balance, potentially enhancing the activity of pro-inflammatory mediators such as IL-6 or IL-17 and type I interferons via TNFR1 while downregulating anti-inflammatory cytokines via TNFR2. This may lead to a dysregulation and promotion of autoreactive T cells resulting in pro-inflammatory paradoxical adverse events (PAEs) (5, 6, 8, 13, 22). The “severe cold” the patient reported before onset of her VKH-like symptoms might have contributed to the disbalance by upregulating type-1-IFNs and other inflammatory cytokines to fight her infection.

Both, hidradenitis suppurativa and VKHD are driven by IL-17A and IL-17F (23–29). Th17 cells co-produce IL-17 and TNF-α, which act synergistically on melanocytes by inducing IL-6, IL-8, IL-1b, and CXCL1 and inhibit the pigment production of melanocytes in the skin. This results in vitiligo, one of the most common signs seen in VKHD, which, however, was not observed in our patient (30–32). In addition to the melanocytes of skin and choroid also the meninges in the brain and pigmented cells of the inner ear are affected by VKHD, causing decreased vision, headaches, tinnitus and hearing loss. Melanocyte-specific proteins of the tyrosinase family are thought to be the targeted autoantigens in VKHD, as they are recognized by T cells from VKH patients and can induce VKH disease-like symptoms in rats (10).

Due to the co-expression of IL-17 and TNF-α in Th17 cells, anti-TNF-α is expected to inhibit these cells and thus to downregulate also IL-17 production. Therefore, TNF-α blockers have been used to treat HS as well as VKHD and many other autoimmune diseases. However, under anti-TNF-α therapy some patients show increased disease activity or even the onset of new autoimmune diseases, such as our patient, who developed VKHLD. Patients with rheumatoid arthritis who developed paradoxical psoriasis under TNF-α inhibitor therapy showed an increase in Th17 cells and IL-17A production. Interestingly, in patients with inflammatory bowel disease and anti-TNF-α therapy the increase in Th17 and IL-17A levels indicates a better outcome, while the opposite is observed in MS patients, who worsen as IL-17 increases under TNF-α blockade (33–35).

As blocking TNF-α can lead to increased IL-17 production, which is counterproductive in several IL-17-driven autoimmune diseases (such as HS and VHK or VKHLD as described here), disease exacerbations or new autoimmune diseases during TNF-α blocker therapy should be carefully monitored. Anti-TNF-α should probably be replaced by anti-IL-17 in diseases where IL-17 is known to be the disease-driving cytokine. A new therapeutic antibody, bimekizumab, which blocks both IL-17A and IL-17F, is approved for the treatment of severe psoriasis and currently being tested for its effect on HS. IL-17A and IL-17F are very similar in structure and function and the differences between both are not yet clear. Former therapeutic antibodies that have only targeted IL-17A were not efficient in treating uveitis, since blocking IL-17A might have been compensated by IL-17F. Interestingly, the inhibition of IL-17F rather than IL-17A, has been shown to induce regulatory T cells. This suggests that the therapeutic efficiency of bimekizumab may extend beyond merely blocking a proinflammatory cytokine. This has been demonstrated in an experimental colitis model and may also apply to other autoimmune diseases (36). In mice with experimental colitis an increase in Prevotella species within the gut microbiome was observed, which correlated with the severity of the disease. A dominance of the “pathogenic” commensal bacterium Prevotella was also found in patients with inflammatory bowel disease and rheumatoid arthritis. Neutralizing IL-17F, but not IL-17A, decreased Prevotella in the gut. This allowed the expansion of “protective” commensals and ameliorated colitis (37). Interestingly, Prevotella spp. are also abundantly found in the skin lesion of patients with HS and are suspected to outcompete other skin commensals, as it was found in the guts of colitis mice (38). This may explain the therapeutic effect of bimekizumab due to its neutralization of IL-17F. Additionally, Prevotella dominance was also was found in the gut microbiome of VKH patients (39), suggesting a similar mechanism by which IL-17F neutralization ameliorates the disease. However, since we have no data of our patient regarding gut microbiota and/or Prevotella prevalence in the skin lesions, the aforementioned pathomechanism remains speculative. Nevertheless, the therapeutic success we have seen in our patient supports this thesis.

Our patient was switched to treatment with this antibody and her VKHLD was also controlled by bimekizumab. Therefore, targeting both IL-17A and -F might be considered as a novel therapy for VKHD and VKHLD, especially when TNF-α blockade is ineffective or induces paradoxical effects.

Although adalimumab plays a crucial role in the treatment of autoimmune diseases by specifically inhibiting TNF-α and is significantly improving outcomes for many patients, its use is not without risks. Infections or neoplasms due to immunosuppression, as well as rare but serious paradoxical reactions such as uveitis can occur. Since other autoimmune diseases are usually induced earlier under TNF-alpha blocking therapy, we speculate that the Patient’s flu-like infection could have initiated an unspecific stimulation and dysregulation of her immune system. This may have resulted in a bystander activation of a latent but so far controlled ocular autoimmunity, or antigenic mimicry of pathogen and ocular autoantigen epitopes, or both (40).

In cases where TNF-α inhibition leads to paradoxical inflammation, switching to an alternative approach, such as anti-IL-17A/F therapy, may be beneficial. However, further clinical studies are needed to prove that bimekizumab is an effective alternative to TNF-α blockade for VKH and VKHLD.