Tumor necrosis factor alpha (TNFα) was the first cytokine targeted by a biological agent in patients with inflammatory arthritis. It is secreted by several immune cell types, such as neutrophils, T cells, and macrophages, recruiting further inflammatory cells and stimulating the production of other cytokines. The central role of TNFα in the pathogenesis of immune arthritis was historically elucidated decades ago, when it was demonstrated that TNFα and its receptors are abundant in the synovial membrane of patients with RA ^[1][32]. TNFα produced by cultured mononuclear cells from the joints of patients with RA is also a key mediator to stimulate other pro-inflammatory cytokines ^[2][33]. Many cell types in the joint, including macrophages, fibroblasts, T lymphocytes, and the vascular endothelium, could also produce and react to TNFα in both an autocrine and paracrine way ^[3][34]. Such findings formed the basis for the clinical use of monoclonal antibodies directed against TNFα for the treatment of RA and other forms of immune arthritis, starting from infliximab, the first chimeric anti-TNFα, and further arriving at similar biological agents, such as adalimumab, golimumab, certolizumab, and etanercept. All these anti-TNFα are currently approved for the treatment of RA, axial spondyloarthritis, and psoriatic arthritis, with similar long-term efficacy rates and safety profile. TNFα interacts also with non-inflammatory cells, such as joint fibroblasts, contributing to chronic tissue damage and fibrosis: at this regard, the early use of anti-TNFα demonstrated to be able to prevent radiological progression and to protect joint integrity in patients with RA ^[4][35]. Moreover, TNFα is an important activator of osteoclasts, leading to bone erosions in chronic arthritis ^[5][36] and contributing to osteoporosis; therefore, its blockage in this clinical setting may ameliorate bone metabolism and further reduce tissue damage ^[6][37].

In the pathogenesis of SpA, and in particular AS, the main key player is the activation of ciclooxygenase-2 after mechanical stress response ^[7][38], which increases the prostaglandin-E levels and then induces the overproduction of interleukin (IL)-17A, the principal cytokine involved in AS. IL-17A is produced by several immune cell types, including CD4 + Th17 lymphocytes, CD8+ cytotoxic T17 cells, T γ/δ lymphocytes, and innate lymphoid cells type 3 (ILC3), which are all present in the tendons and in the entheses of the spine cord and whose interactions determine chronic inflammation ^[8][39]. In addition to TNFα, IL-17A also stimulates osteoblast activity, leading to an exacerbated bone response to pro-inflammatory stimuli and causing subsequent tissue damage ^[9][40]. Moreover, IL-17 is a pain mediator that significantly contributes to symptoms in patients with AS ^[10][41]. The use of monoclonal antibodies directed against IL-17A, such as secukinumab or ixekizumab, have been demonstrated to be effective in the treatment of AS, and their use is currently widely approved ^[11][12][42,43].

Psoriatic arthritis (PsA) is a further immune articular disease that is usually associated with psoriasis, but is also common in other inflammatory conditions, such as IBD. It mainly affects peripheral joints, differently from AS, although an axial involvement may occasionally be present. Patients with PsA may be successfully treated with IL-17A inhibitors, as well as those with AS, confirming an important pathogenetic role of IL-17A in this setting ^[13][14][44,45]. However, in PsA, the IL-17A pro-inflammatory pathway seems to be dependent on IL-23, which comes from inflamed skin or gut and activates the cells that secrete IL-17 (T17, ILC3, and T γ/δ lymphocytes) leading to articular damage ^[15][29], whereas in AS, the production of IL-17A is stimulated directly by COX-2 and prostaglandin E, independently of IL-23. Such differences may justify the fact that IL-23 antagonists (i.e., ustekinumab, a monoclonal antibody against the subunit p40 of IL-12 and IL-23) show low rates of success in patients with predominantly axial involvement and are efficacious in those with peripheral PsA ^[16][17][46,47].

Other cytokines seem to have specific pathogenetic roles in articular diseases. For example, IL-6 represents a crucial cytokine in patients with RA, having regulatory effects on the intra-articular leucocyte infiltrate and potentially influencing B lymphocytes and auto-antibody production ^[15][29]. IL-6 is overexpressed in both synovial fluid and membranes of patients with RA ^[18][48], and interacting with its receptor can activate neutrophils, T cells, B cells, monocytes, and osteoclasts. Tocilizumab and sarilumab, two monoclonal antibodies directed against the receptor of IL-6, showed to be efficacious for the treatment of patients with RA ^[19][20][49,50], whereas they were not effective in AS ^[21][51] and were not tested in PsA. Similarly, IL-1 is a different cytokine that is involved in the pathogenesis of RA and the inhibition of its pathway using a monoclonal antibody (canakinumab) ^[22][52] or anakinra (a human recombinant IL-1Rα antagonist) ^[23][53] has been demonstrated to be effective in this clinical setting. IL-1 inhibition showed significant efficacy in other rheumatological disorders as well, such as systemic juvenile idiopathic arthritis and adult-onset Still’s disease; however, no data are available in patients with PsA or AS so far.

Apart from the direct cytokine inhibition through specific biological agents, a valid therapeutic strategy in several IMID is the blockage of intracellular signal transduction pathways downstream of cytokines, which represents the mechanism of action of Janus kinase (JAK) inhibitors, such as tofacitinib, baricitinib, filgotinib, and upadacitinib, which have been already approved for the treatment of rheumatological conditions. In particular, JAK inhibitors target the intracellular signal transduction shared by many pro-inflammatory pathways, with a final simultaneous inhibition of several cytokines. At the same time, depending on the principal JAK inhibited (JAK1, JAK2, JAK3, or tyrosine kinase (TYK) 2), each drug may primarily impact specific cytokines; for example, IL-6 is mainly targeted by JAK1, JAK2, and TYK2 inhibition, but not JAK3; also, IL-23 is mainly affected by JAK2 and TYK2 inhibition ^[24][54]. In patients with immune arthritis, the reduction of joint pain represents an important clinical outcome: interestingly, a modulation of nociception has been demonstrated with the use of JAK inhibitors in patients with RA and PsA, in particular with tofacitinib and baricitinib, and this effect also seems to be due to cytokine modulation ^[25][55].

Several cytokines are crucial in the activation and perpetuation of pro-inflammatory pathways in the intestinal mucosa of IBD patients. TNFα was one of the first cytokines that was shown to have a main role in the induction of intestinal inflammation. Animal models of colitis that were genetically modified to have an enhanced production of TNFα (TNF^ΔARE mutant mice) developed transmural chronic inflammation with non-caseating granulomas, mimicking macroscopic and histological alterations observed in patients with CD ^[26][56]. TNFα is a pleiotropic cytokine that interacts with several intestinal cell types (endothelial, epithelial, and immune cells) and induces inflammation through multiple pathways, modulating, for example, intestinal epithelial cell growth and apoptosis, gut permeability, and mucosal integrity via matrix metalloproteinase (MMP) production ^[27][57]. Together with IL-2 and interferon (IFN) γ, TNFα also drives pathogenic T helper (Th)1-like responses that characterize the immune profile of active CD ^[28][58]. The publication of almost all findings on the pathogenetic role of TNFα occurred in the nineties, during the same period of the first demonstrations of infliximab efficacy in the treatment of IBD patients. Since then, adalimumab, certolizumab (only in the United States) and golimumab (only for UC) have also been approved for the treatment of IBD. Interestingly, the use of etanercept, a dimeric fusion protein that also targets TNFα and is currently approved for the treatment of several articular immune disorders, is not effective in CD patients ^[29][59] and seems to have negative effects on the intestine, inducing or worsening gut inflammation ^[30][60]. Etanercept, differently from other biologics, targets only soluble TNFα and not that bound on the membrane of immune cells, such as macrophages, which was demonstrated to stimulate both T-cell cytokine production and T-cell survival in patients with IBD ^[31][61]. Therefore, the blockage of the same cytokine, such as TNFα, may potentially induce different effects in each IMID, depending on which specific pathway is affected, with consequent important therapeutic implications.

The active involvement of IL-23 in IBD pathogenesis was first demonstrated at a genetic level. In 2006, a genome-wide study found a strong association among coding and non-coding IL-23R variants and both CD and UC, suggesting that such signaling pathways could be further therapeutic targets in IBD ^[32][62]. Subsequent experiments on an animal model of colitis confirmed that neutralization of IL-23 significantly ameliorates colonic inflammation ^[33][63]. IL-23 is mainly produced by macrophages and dendritic cells in the intestinal mucosa and may alter the balance between pro- and anti-inflammatory pathways, stimulating Th1 and Th17 cells and inhibiting regulatory T cells ^[34][64]. The use of ustekinumab, which targets the subunit p40 of both IL-23 and IL-12, is efficacious in the treatment of IBD, although the role of IL-12, which is normally involved in Th1 differentiation, seems to be marginal in this setting. The predominant function of IL-23 compared to IL-12 in IBD was confirmed by recent data about the use of new selective IL-23 antagonists. Risankizumab, for example, a monoclonal antibody directed against the subunit p19 of IL-23, proved to be efficacious in inducing and maintaining clinical remission in CD patients ^[35][36][65,66]. Successful results were obtained even with guselkumab, another selective p19 IL-23 antagonist, as induction therapy for patients with CD ^[37][67]. IL-23 is strictly related to the IL-17 family, especially in immune arthritis. However, the IL-17A seems to have a protective role in the gut, principally due to its ability to promote epithelial integrity and antimicrobial defense in the intestinal mucosa. Maxwell et al. demonstrated that the selective inhibition of IL-17A leads to colitis exacerbation in animals, mainly due to altered epithelial permeability and function; in the same model of colitis, treatment with inhibitors of p40 or p19 subunits showed a significant and comparable reduction of the intestinal inflammation ^[38][68]. Moreover, the mucosal production of IL-17A is provided by γ/δ T cells but independently from IL-23, leading to a normal level of intestinal IL-17A even during therapy with antagonists of IL-23 ^[39][69]. In line with these data, the treatment of CD patients with IL-17 antagonists, such as secukinumab, showed negative results in terms of both efficacy and safety ^[40][70], and several cases of new onset UC were reported during therapy with IL-17 inhibitors for rheumatological indications ^[41][71]. The axis IL-17/23 represents a further example of different organ-selective signaling pathways of the same cytokine, based on specific environmental interactions.

Other cytokines, such as IL-6, currently have a minor role in the clinical management of IBD, despite their active involvement in the pathogenesis of the diseases. The level of IL-6 is significantly increased in the intestinal mucosa of IBD patients, and its blockage by a specific monoclonal antibody led to good rates of clinical response in refractory CD but increased the incidence of serious adverse events, probably due to a strong immunosuppression ^[42][72]. Recently, a new IL-6 inhibitor has been described by Schreiber et al.: olamkicept is assembled by two complete extracellular domains of gp130 (the IL-6/IL-6R signal transducer), which are dimerized by fusion to the fragment crystallizable (Fc) region of a human IgG1 and act on IL-6 trans-signaling, therefore specifically blocking chronic inflammation without interfering with normal defense activities involving overall IL6 signaling ^[43][73]. The results of a 12-week open-label phase 2 trial demonstrated good rates of clinical response and remission in a patient with IBD treated with olamkicept, which was also well-tolerated ^[43][73]; however, larger controlled studies are needed to confirm these preliminary findings.

Differently from CD, UC is characterized by the activation of Th2 pathway in the colonic mucosa, also recruiting Th9 cells, ILC2, and eosinophils under the stimulus of IL-33 and with subsequent production of other specific cytokines, such as IL-5, IL-6, IL-9, and IL-13 ^[44][74]. However, a therapeutic attempt with selective IL-13 inhibition did not show any benefit in patients with UC ^[45][46][75,76].