Introduction

Tumour necrosis factor (TNF) is a proinflammatory, multifunctional cytokine [1, 2] which is synthesised by various cells including activated monocytes, macrophages, and T-cells [2, 3]. TNF is initially produced as a transmembrane protein (tmTNF), which is cleaved to release the mature soluble cytokine sTNF [2, 4]. Finally it becomes biologically activated due to the aggregation of three TNF monomers to produce trimeric TNF, this then binds to one of the two cell-surface receptors which are expressed on many cells [24]:

  1. 1.

    Type 1 TNF receptors, p55

  2. 2.

    Type 2 TNF receptors, p75

In the normal body, TNF has an integral role in mounting the inflammatory response against invading pathogens [5]. Its functions are stated in Table 1.

Table 1 The function of TNF in the normal human body

Even though there is no invading pathogen, excessive amounts of TNF have been detected in patients with chronic inflammatory diseases. Therefore, it was concluded that TNF “plays an important role in the pathogenesis [7]” of these diseases. This led to the development of anti-TNF therapies as an alternative treatment.

Two types of anti-TNF therapies were created: monoclonal antibodies and soluble receptors. The Medicines and Healthcare products Regulatory Agency (MHRA) has approved three anti-TNF drugs, two are monoclonal antibodies namely infliximab and adalimumab, and one is a soluble receptor called etanercept [2], see Fig. 1.

Fig. 1
figure 1

Structures of the anti TNF agents [4]

Infliximab is a chimeric (human–mouse) monoclonal antibody, containing a constant region of the human IgG1 antibody and an antigen-binding region of a mouse antibody [3, 4, 8]. Adalimumab is a “humanised monoclonal antibody [4]”, therefore it has both constant and variable regions of the human IgG1 antibody. Etanercept is a type 2 TNF receptor fusion protein, created from two extracellular domains of the type 2 TNF receptor (p75) fused with part of the human IgG1 antibody. See Table 2 for the method of action of these anti-TNF agents and the inflammatory disorders which the National Institute of Clinical Excellence (NICE) has approved them for use.

Table 2 Anti TNF agents: their method of action and approved uses [8]

This year is the tenth anniversary of the first anti-TNF agent, infliximab to be approved; whilst in many ways these drugs have revolutionised the treatment of inflammatory diseases. Their increased clinical use has resulted in a concomitant increase in the reporting of their adverse effects. In this article the adverse effects associated with the immunosuppression caused by anti-TNF agents will be comprehensively analysed.

Methods

The search strategy was designed to retrieve references relating to anti-TNF agents, their mode of action and how this relates to the adverse effects caused by these drugs. Database searches included Scopus, Amed, Cochrane library and Medline through both Ovid and Pubmed. Important articles were also obtained by using reference lists from key reviews, titles and abstracts of all articles were assessed for inclusion/exclusion.

Internet searches were also utilised for information, examples of sites used are google scholar, NICE website, MHRA website and national library of medicines. Search terms for both internet and database searches included “anti-TNF”, “TNF antagonists” or the names of each of the anti-TNF agents plus one or more of the following words “adverse effects”, “side effects”, “complications”, “immunity”, “risks” and “infections”. No date restrictions were placed on any searches, however, priority was given to the most up-to-date trials and reviews to ensure that the information used was as relevant as possible.

Results and discussion

Due to the mechanism of action of all three anti-TNF drugs, immunosuppression is caused by blockage of the immune cascade TNF initiates. As a by-product of this, seven main types of adverse effects have been seen including:

  1. 1.

    Increased incidence of infections

  2. 2.

    Malignancies

  3. 3.

    Demyelinating syndromes

  4. 4.

    Worsening congestive cardiac failure (CCF)

  5. 5.

    Immunogenicity

  6. 6.

    Infusion/injection reactions and hypersensitivity

  7. 7.

    Hepatotoxicity

  8. 8.

    Haemological disorders

Reports of the above adverse effects have been associated with the use of all three anti-TNF therapies, however, there are slightly differing incidences for each drug, and the hypotheses for these differences are explained in Table 3 below.

Table 3 Hypotheses for the differences in risk of adverse effects anti-TNF drugs [6, 7]

Infections

It is not surprising that there is a high incidence of infections seen as an adverse effect of patients being treated with anti-TNF drugs when you consider the integral part TNF plays as a “central role in the initial host response to infection [4]”. The incidence of infections is debatably the most important adverse effect of anti-TNF therapy. Many different types have been reported including, tuberculosis (TB), serious bacterial infections, listeriosis, atypical mycobacterial infections, histoplasmosis, coccidioldmycosis and pneumonia [9]. It is difficult to attribute these infections fully to the anti-TNF therapies because:

  • Patients with chronic inflammatory diseases such as rheumatoid arthritis (RA) are know to have an approximately twofold predisposition to infections in comparison to the general public irrespective of treatment [3].

  • Most patients who are candidates for anti-TNF treatment are usually already on corticosteroids and/or other immunosuppressive drugs which have already increased their infection risks.

However, due to the major role TNF plays in immunity against invading pathogens, it is biologically plausible that inhibition of TNF would cause an increase in infection incidence.

General bacterial infections are associated with all three anti-TNF agents; most of these are not serious and are resolved by antibiotic treatment and/or “temporarily stopping the drug [10]”. The incidence of serious bacterial infections is a debated issue; there have been sporadic reports of fatal cases with infections such as severe pneumonia, meningitis and sepsis [9]. Yet other reviews state that there is “no increased incidence of serious infections [8]”.

There is evidence that the risk of opportunistic infections is increased by the use of anti-TNF drugs. The major concern is the activation of latent TB [3]. It has been proven that TNF has a specific role in the defence against the TB pathogen causing the formation of granulomas and therefore containment of the disease. Therefore, screening for latent or active TB prior to treatment is highly recommended. In addition, on finding a positive test, the patient must undergo treatment for TB before initiation of any anti-TNF drugs.

Due to the essential role TNF has in granuloma formation; the reactivation of granulomatous fungal infections is another concern. There is no clear relationship between the formation of these infections and any one anti-TNF drug and their incidence is very low, therefore no guidelines have been produced to “screen and treat latent fungal infections [10]” before the start of anti-TNF drugs; however, clinicians still need to be vigilant.

Unfortunately, there is little data on the effects of anti-TNF agents on patients with existing viral infections [11]. However, the little data available suggest that their effect is dependant on the type of viral infection. Evidence suggests that patients with hepatitis c virus (HCV) have elevated levels of TNF in their bodies and this has been linked to the pathophysiology of the disease [12]. Small studies have shown that the anti-TNF drugs are safe and may even be beneficial to patients with HCV [12]. However, clinicians must be cautious when treating HCV patient with anti-TNF, monitoring of their serum aminotransferase and viral load is recommended [11, 12].

Conversely in patients with hepatitis B virus (HBV), the use of anti-TNF treatment is of greater concern as there have been reports of the reactivation of stable HBV [11, 12]. This may be due to the fact that TNF seems to “promote viral clearance [12]” and therefore controls the HBV; some studies recommend screening of patients for HBV before the initiation of anti-TNF therapy. In addition, the European Association for the Study of Liver Disease recommends that “antiviral treatment should be initiated 2–4 weeks before [8]” the start of the anti-TNF drugs and should continue throughout the patient’s treatment [8].

Reports of the use of anti-TNF therapies in patients with existing human immunodeficiency virus (HIV) have suggested that they have been well-tolerated; however, the data are limited. It is important that HIV treatment is “well-established before initiation of therapy with a TNF inhibitor [11]”. In addition, due to the immunocompromised nature of the disease, infection risk needs to be monitored closely whilst on anti-TNF therapy.

Malignancies

The oncogenic effect of all three anti-TNF drugs is of very real theoretical concern. The immune system, specifically TNF plays a large role in suppressing the development and spread of tumours by triggering apoptosis of tumour cells [9]. However, in reality the causal relationship between anti-TNF drugs and malignancies, specifically lymphoma, is a highly debated issue [3]. This is due to the fact that irrespective of medications used, the risk of lymphoma is known to be higher in patients with rheumatoid arthritis and Crohn’s disease [3]. Therefore, when assessing study results it is difficult to attribute the estimated 2.3–6.4 times higher incidence of lymphomas in patients using anti-TNF drugs, compared to the general public, to the treatment rather than to the disease itself [3, 9].

However, information presented to the Food and Drug Administration (FDA) showed that there is an increased risk of lymphomas in patients on anti-TNF compared to patients with RA not on this treatment [13]. Is it debatable whether the incidence of other cancers is raised by the use of anti-TNF therapies, some reviews state that “Apart from lymphoma, the incidence of cancer is not significantly altered [3]”, yet others talk about the increased risk of solid tumours and skin cancer [13]. Due to this uncertainty, continued vigilance is needed until the relationship is better understood [10].

Neurological

Since the approval of anti-TNF agents for treatment, there have been several reports of adverse effects on the neurological system, specifically reported are demyelinating syndromes and seizures. These adverse effects can be broadly placed into four main groups:

  1. 1.

    Exacerbation of Multiple Sclerosis(MS) [14]

  2. 2.

    New-onset MS [14]

  3. 3.

    Acute mental state changes for example encephalopathy

  4. 4.

    Cases of neurological disease excluding MS for example optic neuritis.

A wide variety of clinical symptoms have been attributed to these adverse effects, most commonly paresthesia, visual disturbances and confusion [3, 14]. Studies have proven that there is a “temporal relationship [15]” between neurological adverse effects and anti-TNF therapy; in addition it was found that all symptoms improved or disappeared completely when the therapy was discontinued [15]. The package leaflets of all three anti-TNF drugs state that nervous system disorders are an uncommon or rare side effect, see Appendices 1, 2 and 3; Table 4 for definitions.

Table 4 Definitions of risk attributions used in package leaflets

The mechanism of action of these adverse effects is still largely unclear; however, one hypothesis is that anti-TNF drugs may cause exacerbation of autoimmunity by blocking the normal activity of TNF in the immune system and this will cause the effects stated in Table 5.

Table 5 The effects of TNF inhibition

Due to the minor number of cases of seizures reported with the use of anti-TNF therapies a past history of seizure disorders does not seem to contraindicate the use of these drugs. In the case of patients with established demyelinating diseases such as MS, it is reasonable to avoid the use of these drugs, some clinicians will also be cautious when using them in patients with a family history of these diseases [8, 9, 15]. If any neurological symptoms are seen in patients on anti-TNF drugs they need to be discontinued immediately to avoid the progression of these symptoms [15].

Congestive cardiac failure

High levels of circulating TNF have been found in patients with congestive cardiac failure (CCF); it has been implicated in the pathogenesis of the disease and is thought to be the cause of the decreased cardiac contractility [3, 17]. Due to this finding, the therapeutic effects of anti-TNF drugs in patients with CCF were tested to see if the symptoms of the disease could be reduced. However, there was found to be no beneficial effects, and in some cases the anti-TNF drugs may even cause a worsening of the patient’s CCF, see Table 6. Due to the serious adverse effects reported in the trials of etanercept and infliximab, adalimumab never underwent any treatment trials.

Table 6 Anti-TNF trials in patients with heart failure [18]

The mechanism of this adverse effect is still very uncertain and continues to be the subject of large studies. The issue is further complicated because patients with inflammatory disorders such as rheumatoid arthritis are know to be at higher risk for cardiovascular disease compared to the general public [17]. Therefore, a causal relationship between anti-TNF and CCF is hard to prove, nevertheless results of postmarking case reports and spontaneous reporting systems have shown the very real, albeit rare, adverse effect of CCF with patients on anti-TNF therapies developing CCF or experiencing worsening symptoms of their existing CCF. Although the mechanism of this adverse effect is still largely unknown, studies have hypothesised that using anti-TNF agents may “decrease TNF-alpha levels below physiological levels for repair [18]”. In addition they have suggested that TNF may have beneficial effects on patients:

  1. 1.

    When the heart has acute ischemic injury, TNF may have a cytoprotective effect [18]

  2. 2.

    TNF may be able to “maintain peripheral blood flow in patients with heart failure [18]” by triggering the increased production of nitric oxide in the vascular system

  3. 3.

    TNF may have an “important role in tissue remodelling and repair” [18].

As anti-TNF drugs inhibit the majority of TNF in the body they will also halt of all these beneficial functions and therefore could lead to an exacerbation of the patients CCF. Given all this evidence, it is advisable to avoid using anti-TNF treatment in patients with symptomatic CCF. However, in patients with mild CCF, anti-TNF drugs can be utilised but care and strict vigilance are needed.

Immunogenicity

A number of studies have shown that anti-TNF drugs can lead to the formation of neutralising and non-neutralising antibodies [1]. Although there is evidence to suggest that these antibodies develop as a reaction to all three drugs, the risks vary slightly, see Table 7.

Table 7 Differing risks of formation of antibodies for the three anti-TNF drugs

The formation of neutralising antibodies is of concern for patients on anti-TNF therapies as their production may lead to allergic reactions to the drugs or change the efficiency of the agent.

The evidence of the development of autoantibodies in patients treated with anti-TNF drugs [19] is of greater concern; specifically there is evidence of the formation of antinuclear antibodies and anti-DNA antibodies. As with the risk of formation of neutralising antibodies, infliximab may be more likely to form these autoantibodies in comparison to adalimimab and etanercept [9, 19]. Even though this is a concern for health care professionals, the formation of these autoantibodies have been shown not to be a presage to autoimmune disorders [9].

Several studies have also found autoimmune diseases to be an adverse effect of anti-TNF agents. These studies have shown

  1. 1.

    There is a “temporal association [19]” between anti-TNF therapies and patients developing symptoms of an autoimmune disease [19]

  2. 2.

    After the anti-TNF drug is stopped, there is a regression of the autoimmune condition [19]

  3. 3.

    If the drug is re-given to a patient after being stopped due to formation of autoimmune symptoms, these symptoms will reappear and in some cases even become worse.

The most common autoimmune diseases seen with anti-TNF drugs are; drug induced lupus, vasculitis, interstitial lung disease, uveitis and psoriatic skin lesions.

Infusion/injection site reactions and hypersensitivity

Injection/infusion site reactions are the most common adverse reactions seen with anti-TNF drugs. The symptoms usually very mild and are therefore of “little clinical significance [8]”, they include:

  • minor itching and redness around the site

  • headache

  • dizziness

  • nausea

These reactions are thought to be mediated by T-cells and are described as “delayed-time hypersensitivity reactions [9]”. They would usually occur in the first month of treatment and decrease in severity over time, this is possibly due to an induced tolerance of the drug.

Injection site reactions are more common in etanercept and adalimumab. Infusion reactions on the other hand are more common with infliximab; these are also usually very mild and are remedied by slowing the infusion rate or prescribing either antihistamines or glucosteroids.

Hepatotoxicity

Anti-TNF agents have been associated with a small risk of hepatotoxicity; in 2004 the FDA issued a warning to healthcare professionals about the risks of hepatic disease [8, 15]. The warning was issued after FDA receiving approximately 35 case reports of hepatic disease from their MedWatch programme [8, 15]. Examples of the adverse effects reported were acute liver failure, hepatitis, and cholestasis; some of which were very severe leading to the need-to-need of transplantation or mortality. Infliximab seemed to be the cause in the majority of these cases, however, all three anti-TNF drugs have raised liver function tests as an uncommon or rare side effect on their package inserts, see Appendices 1, 2 and 3.

Haematological

There have been reports of haematological dyscrasias, such as aplastic anaemia and pancytopenia due to anti-TNF therapies, however, these adverse reactions are very rare. Nevertheless these disorders can be very serious, therefore, clinicians and patients both need to be aware of the symptoms for example “pallor, gum bleeding, easy brusibility, general bleeding and persistent fever or infection [12]”. Due to the isolated nature of these cases, it is unnecessary to formally monitor patients for haematological dycrasias, however if symptoms are noticed patients should be taken off the anti-TNF drug and should be assessed for any underlying diseases [9].

Conclusion

The decision to start a patient on anti-TNF therapy is a very complicated one; these drugs cause immunosuppression which in turn leads to many potentially serious adverse effects. However, they are arguably “among the most effective treatments [3]” for chronic inflammatory diseases. Due to this, the risk-benefits analysis is complex, nevertheless when used correctly and when the adverse effects detection and prevention methods are put in place the benefits outweigh the risks, see Table 8.

Table 8 British Society for rheumatology guidelines for anti-TNF-α therapy [20]

There are many different adverse effects seen with anti-TNF drugs, all of which are still of great concern to clinicians prescribing these drugs. The incidence of infections has to be monitored closely to ensure that they do not become fatal. The lymphoma rate is also a serious concern; however, a causal relationship has yet to be found; more long-term studies are needed to clarify the importance of this relationship. In addition, the development of neurological disorders, CCF onset or worsening, immunogenicity, infusion/injection reactions, hepatotoxicity and haematological disorders are all very uncommon and seem to be easily reversed with discontinuation of treatment.

Finally, the most important thing to consider when using anti-TNF treatment is that all healthcare professionals and patients are aware of the adverse effects to ensue early detection, it is also important that the decision to start treatment should be individualised to the patient’s circumstances and risk profile [17, 20].