TNF is one of the most critical pro-inflammatory cytokines of the innate immune response and mediates pleiotropic effects, which implies action on diverse cells subpopulations to mediate a wide range of activities such as the production of inflammatory mediators, cell proliferation, and cell death. TNF is produced by macrophages, T-, B- NK-, dendritic cells, and fibroblasts [75]. TNF is a versatile cytokine that acts as an alarm system in host defense, appearing in the first few minutes of damage [76]. TNF is a trimeric molecule that can be found as a transmembrane (tmTNF) or soluble form (sTNF) by the action of ADAM17 [77,78]. Both forms are bioactive molecules after interacting with one of its receptors, TNFR1 or TNFR2 [77,79]. Importantly, the soluble formats of both receptors can also be generated by the sheddase activity of ADAM17. Transmembrane TNFR1 and TNFR2 signaling to active nuclear factor-kappa B (NF-κB) or MAP kinase family inducing cell survival or cell death [80]. When tmTNF interacts with soluble TNFRs may trigger an activation called reverse signaling, which is activated in the tmTNF expressing subpopulation; in particular, sTNFR1 induces apoptosis through tmTNF by reverse signaling [78].

In addition to the ADAM17 activity, the soluble formats of TNFR1 and TNFR2 may result from alternative splicing [81]. The soluble receptor variants have been associated with different biological effects. For example, in viral and bacterial infections, some pathogens can encode soluble homologs of receptors as a mechanism of immune evasion throughout the sequestration of the corresponding cytokine [82,83,84].

Anti-TNF therapy was approved by the U. S. Food & Drug Administration (FDA) in 1998, and currently, it is efficiently administered as a treatment for diverse inflammatory processes like rheumatoid arthritis, juvenile idiopathic arthritis, ankylosing spondylitis, Crohn´s disease, plaque psoriasis, among others [15,76,85,86].

There are five types of biological TNF inhibitors approved for use; the difference between them is the nature of their origin. The first of these TNF inhibitors is infliximab, approved in August 1998, which is an anti-human mAb chimeric mouse-human IgG that inhibits soluble TNF in both the monomeric and trimeric form [77,87]. The second TNF inhibitor, etanercept, approved in November 1998, is a fully human recombinant molecule consisting of two subunits of the TNFR2 linked to a human IgG1 Fc and can block the trimeric form of TNF [77,88]. Third, Adalimumab and Golimumab are fully humanized IgG1, the first was FDA-approved in 2002, the second in 2009, and both neutralize tmTNF and sTNF [77,89]. Finally, Certolizumab-pegol is a PEGylated Fab fragment, which means an IgG1 mAb with a hinges region linked to two cross-linked chains of a 20-kDa of polyethylene glycol (PEG) but has lacked the Fc region. It was approved in 2008 and also can neutralize both tmTNF and sTNF [85,89,90].

The anti-TNF biological agents target both tmTNF and sTNF, and all of them trigger diverse effects such as neutralization, apoptosis, and modulation of the immune system; several facts may affect the efficacy of every anti-TNF inhibitor as pharmacokinetic, tissue penetration, affinity, and avidity, among others [91], reasons by which more systematic reviews are required to understand the global impact of anti-TNF therapy in COVID-19. Despite the benefit of anti-TNF therapy in a broad panel of inflammatory diseases, it has been well documented that the TNF inhibitors may present controversial effects in non-responder patients; this means patients who previously responded to the treatment but posteriorly, they are treatment-refractory and susceptible to infection from some bacteria such as Legionella pneumonia and Listeria monocytogenes [15,85]. Even more, the side effects in some cases may also include Mycobacterium tuberculosis and hepatitis B reactivation, anti-TNF inhibitor antibodies, and neurological impact [15,92,93]. In some cases, the side effects may be overcome by stopping the administration or changing the inhibitor agent.

Various pronouncements have been made to support the administration of anti-TNF agents in COVID-19 patients; they are based on the principle that the increase of this cytokine has severe effects on diverse cell subpopulations and its function blocked with the anti-TNF therapy is efficient in diverse autoimmune diseases [18,19,48,94]. However, the first problem found is the discrepancy in the TNF levels identified in COVID-19 by several groups; some authors showed that COVID-19 patients display increased levels of this cytokine, but others cannot find it [44,95,96,97].

At present, there are ongoing clinical protocols to identify the efficiency of the anti-TNF therapy in COVID-19, some of them suggesting that COVID-19 patients treated with anti-TNF agents have a better prognosis [98,99]. In this regard, recently, it was suggested that patients with any inflammatory disease using TNF inhibitors had a lower probability of hospitalization or the development of severe COVID-19 compared to patients diagnosed with an inflammatory disease but another treatment [98,100]. However, despite the evidence, the use of these drugs is debatable in some cases, and probably it is favored because the knowledge about levels and regulation mechanisms of TNF, TNFR1, and TNFR2 in COVID-19 is limited, and other important factors call into question what side effects can leave the use of TNF immune modulators in viral infections [99].

In 2020, the Crohn´s & Colitis Foundation of America recommended that patients with Inflammatory Bowel Disease (IBD) that develop COVID-19 stop the anti-TNF administration (biologics or biosimilars) until they recover. It was supported by the National Scientific Advisory Committee of the USA, the International Organization for the Study of Inflammatory Bowel Diseases (IOIBD), and the American Gastroenterological Association [101,102,103,104]. These recommendations were made because patients with chronic inflammation receive immunosuppressive or immune-modulator therapies that may represent a risk of viral infections. Additional to the lack of knowledge about what side effects can leave the use of TNF immune modulators in viral infections [103,104].

Recently, Keewan et al. suggested that PEG-, non-PEG-Certolizumab, and Adalimumab favor the infection of SARS-CoV-2 because the mode of action of these agents induces the Notch-1 signaling pathway promoting a pro-inflammatory environment [105]. In addition, previous reports documented that ADAM17 participates in the Notch signaling pathway, which also may lead to an inflammatory response [106,107]. In an in-vitro model of Mycobacterium avium subspecies paratuberculosis (MAP) infection has suggested that the Notch-1 signaling pathway induced by the anti-TNF therapy impacts IL-6 and MCL-1 expression in macrophages. Additionally, the authors showed that, in particular, adalimumab increases the TMPRSS2/ADAM17 ratio, suggesting the infection facilitated by TMPRSS2 [105].

Although successful cases of immunotherapy with TNF inhibitors have been well documented for years, indeed, some adverse events are also known. A meta-analysis documented that the use of adalimumab, golimumab, infliximab, certolizumab, and etanercept, significantly increases fungal, viral, and bacterial (including mycobacterial) infections [108]. Demyelinating disorders such as type I diabetes and psoriasis also have been reported as a side effect associated with the use of TNF inhibitors; it is the primary failure that occurs in one-third of patients [109].

Epigenetic control of the TNF gene might direct or predict the patient’s responsiveness to anti-TNF therapies, and TNF signaling pathways may be restricted to epigenetic regulation [110]. In particular, the presence of increased levels of H3K4me2 (histone H3 lysine 4 dimethylation) is a modification identified in promoters of expressed genes of TNF-producing cells [111]. The TNF-308G>A polymorphism in the promoter region could be associated with a poor response to TNF inhibitors [112]. At the level of the TNF locus, loss of CpG methylation has been reported as an age-associated factor in healthy donors [113]. In this region, a −238 position polymorphism disturbs a CpG motif, a fact that is also related to the severity in cases of arthritis rheumatoid [114].

In the case of COVID-19, there are two main unknowns related to TNF inhibitors: the risk of immunosuppression in the SARS-CoV-2 infection and the possibility that these agents could facilitate virus entry into target cells. Therefore, further molecular and clinical studies are necessary to understand the real potential of anti-TNF inhibitors in COVID-19. Meanwhile, novel treatment targets must be considered for evaluation, alone or combined with antiretrovirals [115]. In this regard, TNFRs deserve particular attention in COVID-19.

Scientific evidence shows that the TNF is not the alone-worker in COVID-19; data indicate that the couple TNF/IFN-γ synergizes to induce crosstalk of several types of cell death such as pyroptosis, necroptosis, and apoptosis (collectively named PANoptosis). It is regulated by signal transducer and activator of transcription 1 (STAT1), interferon regulatory factor 1 (IRF1), inducible nitric oxide synthase (iNOS), and nitric oxide (NO); and Caspase-8/RIPK3, all together induce an inflammatory cell death that is in detriment to the host. Under the COVID-19 context, it has also been identified that this TNF/IFN-γ synergism could be blocked with the mAbs co-administering [12].

At present, it is well known that TNFR1 signaling is related to apoptotic cell death, but additionally, this receptor also promotes the release of pro-inflammatory molecules, including cytokines, chemokines, adhesion molecules, and matrix metalloproteinases [116,117]. Thus, TNFR1 signaling is highly inflammatory, a high level of sTNFR1 in fluids has been previously proposed as a biomarker in several diseases with a pro-inflammatory profile as diabetic nephropathy, hyper-inflammation on the ocular surface, sepsis, ventricular dysfunction, and myocardial infarction, acute lung injury, smokers with microvascular complications, and acute graft-versus-host disease [116,118,119,120,121,122].

Efforts have been made to identify if sTNFR1 could represent a potential biomarker in COVID-19. A first report indicated that COVID-19 patients in an intensive care unit (ICU) have a higher level of a pro-inflammatory profile (IL-1β, IL-6, IL-8), including high sTNFR1 levels [123]. Posteriorly, Mortaz et al. [124] identified high levels of sTNFR1 in patients with severe COVID-19, suggesting for the first time that this molecule could represent a biomarker of severity and mortality. In accordance, other authors identified that sTNFR1 is increased in severe COVID-19 compared with mild and moderate illness, and also it was associated with mortality [97]. However, Bowman et al. [125] reported that both TNFR1 and TNFR2 increased independently of severity.

The use of anti-TNF therapy requires an integrative comprehension considering the roles of TNFR1 and TNFR2, as well as the soluble or transmembrane formats of the proteins [79]. Based on the discussed evidence, we suggest that the clinical trials to evaluate the efficiency of the anti-TNF therapy should do an integral association between levels of TNF and its receptors. Moreover, a relevant question that has not been clarified is whether COVID-19 TNF is found mainly in transmembrane or soluble forms. In other pathologies, it has been deeply described that the TNF form is fundamental to determining its function as an activator or regulator of the inflammatory response [79,126].

Although the current TNF inhibitors are highly efficient, it has been well documented that total TNF inhibition may lead to severe side effects such as tuberculosis reactivation [127,128]. An advantage is that TNFR1 and TNFR2 mediate different cellular events, the reason why the option of selectively blocking one of them may result in fewer adverse effects [129,130]. As TNFR1 promotes inflammation and cell death, the specific inhibition can avoid side effects associated with TNF inhibitors and therefore constitute the next generation of immunomodulators even in viral infection as COVID-19 [131]. Various efforts are committed to the selective modulation of TNF receptors, with TNFR1 receiving greater attention due to the roles in which it has been identified [132,133]. This positions TNFR1 as a potential therapeutic target during SARS-CoV-2 infection.