Epidemiology

The prevalence of acute infusion reactions to infliximab varies widely in the literature, ranging from 3% to 20%. In the Crohn’s Therapy, Resource, Evaluation, and Assessment Tool (TREAT) registry, 3% of infliximab infusions resulted in an acute infusion reaction, most of which were mild. Similarly, Cheifetz and colleagues and Colombel and colleagues reported acute infusion reactions in 5% of patients at Mt Sinai Medical Center and 3.8% of patients treated at Mayo Clinic, respectively.

Classification and symptoms

Infusion reactions can also be classified based on their severity. Cheifetz and colleagues developed a classification scheme based on symptoms ( Table 1 ). In their cohort, most infusion reactions were classified as mild (51%), with 21% considered moderate, and 17% considered severe. However, in the TREAT study, serious infusion reactions were seen infrequently.

Prevention

Multiple factors have been shown to influence the development of infusion reactions. Utilization of an induction regimen (0, 2, 6 weeks), use of regular-scheduled maintenance dosing (nonepisodic), and concomitant immunomodulator administration all have been shown to decrease the risk of infusion reactions to infliximab by reducing the incidence of antibody formation. Some experts advocate the use of combination therapy for at least 6 months to help mitigate the immune response and prevent antibody formation. When possible, episodic therapy should be avoided, because it appears to increase the risk of antibody formation. Routine premedication with corticosteroids and antihistamines has not been shown to decrease the risk of infusion reactions.

Treatment

Although infusion reactions can be severe and life-threatening, the need to discontinue anti-TNF therapy is rare. In the ACCENT 1 trial, less than 3% of patients with infusion reactions stopped infliximab. Similarly, Schaible and others reported cessation of infliximab in less than 2% of patients.

The treatment of acute infusion reactions depends on the cause and severity of the reaction. Fig. 1 details an algorithm for managing acute infusion reactions. As most reactions are nonimmune and rate-related, the infusions typically can be successfully completed. Any patients experiencing an infusion reaction should be administered intravenous fluids (normal saline) and have frequent monitoring of vital signs. Mild reactions can be managed by temporarily reducing the infusion rate. Once the symptoms resolve, the infusion rate can be increased slowly back to baseline. However, some physicians prefer to temporarily discontinue the infusion until all symptoms have abated. If necessary, acetaminophen or antihistamine can also be used. In more moderate reactions, the infusion should be discontinued, and acetaminophen, an antihistamine, and a systemic steroid should be administered. The infusion can then be restarted at a lower rate and increased slowly as long as symptoms do not recur. When severe reactions develop, the infusion should be discontinued. Immediate treatment with acetaminophen, a histamine blocker, and systemic steroids are warranted. If there is no evidence of anaphylaxis (wheezing, angioedema), then the infusion can be restarted if necessary. If anaphylaxis is present, epinephrine is necessary and should be given before corticosteroids. Advanced life support measures may be needed. No attempt should be made to restart the infusion in the case of a true type I hypersensitivity reaction (anaphylaxis).

Acute infusion reaction treatment protocol.

Re-treatment

Once an infusion reaction to infliximab has occurred, premedication protocols can be used that allow for continued use of the medication. Fig. 2 provides an algorithm for pretreatment prophylaxis of patients following an infusion reaction. In the study by Cheifetz and colleauges, all patients with mild to moderate reactions who received prophylactic medication before subsequent infusions were successfully reinfused. Following a severe infusion reaction, 3 patients were re-treated, one of whom developed a recurrent severe reaction despite prophylaxis. Other groups have confirmed the efficacy of re-treatment protocols, with less than 3% of patients requiring discontinuation of infliximab because of infusion reactions. Given that recurrent infusion reactions may occur despite prophylaxis, in severe infusion reactions, a very careful risk-benefit assessment must be made before considering continuation of infliximab.

Acute infusion reaction prophylaxis protocol.

Currently, if a patient has an acute infusion reaction, it is assessed if ATI are present. If ATI are present in significant concentration, the patient is switched to another anti-TNF. If ATIs are not present, then the infliximab infusions using the prophylaxis protocols are often continued.

Epidemiology

Delayed infusion reactions typically occur 5 to 7 days (range, 1–14) after an infusion. They likely represent a mild type III immune complex reaction and have also been referred to as serum sickness-like reactions. Delayed infusion reactions have been reported in 1% to 3% of patients treated with infliximab. Although some older studies reported rates as high as 27%, this was likely attributable to use of a liquid formulation of infliximab that is no longer available and to the use of episodic dosing protocols. Similar to acute infusion reactions, delayed infusion reactions likely result from ATI formation. Risk factors include the episodic dosing of infliximab, delays in the routine infusion schedule of infliximab, and lack of a concomitant immunomodulator.

Symptoms and treatment

Symptoms of delayed infusion reactions can be quite variable. The most common complaints include joint pains, rash, arthritis, myalgias, jaw pain, fatigue, headaches, edema, sore throat, and fever. Delayed infusion reactions must be differentiated from viral syndromes, drug-induced lupus, and extra-intestinal manifestations of IBD.

Most delayed infusion reactions do not require specific treatment. For mild symptoms, acetaminophen can be administered. Patients with persistent, severe joint pains or arthritis may require a short course of corticosteroids. There are very little data regarding re-treatment of patients following a delayed infusion reaction, but some studies have shown that a delayed infusion reaction often leads to complete cessation of infliximab treatment. In one study, nearly 50% of patients who stopped infliximab did so because of a delayed infusion reaction. In a separate study, the same group showed a discontinuation rate of 92% following a delayed infusion reaction. As with acute infusion reactions, the authors’ practice is to assess for ATI in this setting. There are very few data regarding re-treatment of patients following a delayed infusion reaction. In the authors’ practice, they proceed with reinfusion so long as ATI are minimal or absent.

Prevention

Prevention of delayed infusion reactions is predicated on avoiding antibody development to infliximab (induction and maintenance infusions and use of concomitant immunomodulator) as described above.

Epidemiology

Injection site reactions are a frequent side effect of the self-injectable anti-TNF agents (adalimumab, golimumab, certolizumab pegol), with studies showing an incidence ranging from 1% to nearly 40%, and most occur during the induction phase. Higher rates of injection site reactions may be seen with adalimumab versus certolizumab pegol (2%–3%) or golimumab (3%–6%). The pathophysiology of these reactions is usually traumatic but may also be related to a mild delayed hypersensitivity reaction. Most of the injection site reactions are mild and do not lead to drug discontinuation.

Symptoms and treatment

Symptoms of injection site reactions include burning, pain, pruritus, erythema, swelling, bruising, irritation, and nonspecific complaints. Rarely, a hematoma may develop. Typically, the reactions last for 3 to 5 days and result in minor discomfort. Symptomatic reactions can be treated with topical lidocaine and ice. In persistent cases, topical corticosteroids may be effective. If injection site reactions recur, pretreatment with ice and/or topical lidocaine can be prescribed, and the injection site location should be varied. Even in recurrent cases, injection site reactions rarely require cessation of anti-TNF.

Antinuclear antibody

Antinuclear antibody (ANA) is elevated at baseline in many of the conditions for which anti-TNF therapy is indicated. However, the prevalence of ANA positivity increases with the use of an anti-TNF. In patients with rheumatoid arthritis (RA), ANA-positive tests increased from 40% before infliximab therapy to 80% after therapy. In seronegative spondyloarthropathies, the prevalence of ANA is only 8%, but treatment with infliximab has been shown to increase this prevalence to 46%. Similarly, Vermeire and colleagues reported a baseline prevalence of ANA in 7.2% of their 125 patients with IBD treated with infliximab. By 24 months of treatment with infliximab, 56.8% of patients had developed a positive ANA. In these cases, nearly 50% of the new ANA developed following the first infusion, and approximately 80% became ANA-positive within 3 infusions. Beigel and colleagues reported a prevalence of ANA positivity in 44% of patients with IBD treated with either infliximab or adalimumab. In contrast, the CLASSIC II study reported new ANA in 19% after 56 weeks of therapy with adalimumab. Schreiber and colleagues noted 8% of patients developed new ANA at 26 weeks on certolizumab pegol compared with 1% in the placebo arm. Golimumab may have a lower rate of ANA development of only 3.5%.

Anti-double-stranded DNA antibody

Anti-TNF therapy also appears to increase the rate of anti-double-stranded DNA antibody (anti-dsDNA) formation. Infliximab-associated anti-dsDNA ranges from 3% to more than 30%. Although the rates for adalimumab-induced anti-dsDNA was 19%, both certizolizumab pegol and golimumab reported rates of 0.5% to 1%.

Anti-cardiolipin and antihistone antibodies

Anticardiolipin antibodies (ACL) and antihistone antibodies may also be elevated because of anti-TNF therapy. In patients with RA treated with infliximab or etanercept, Jonsdottir and colleagues reported an increase in ACL from 16% to 26%.

In refractory spondyloarthropathies treated with infliximab, Sellam and colleagues found an increase in antihistone antibodies from 29% to 57%. In patients with IBD treated with either infliximab or adalimumab, 20.9% developed new-onset antihistone antibodies.

Epidemiology

DILE has been reported in patients treated with infliximab, etanercept, adalimumab, and golimumab. In the BIOGEAS Spanish registry of biologics used to treat rheumatologic conditions, DILE was reported in 105 patients, of whom 43% were on infliximab, 35% on etanercept, and 21% on adalimumab.

The incidence of DILE associated with anti-TNF is rare, ranging from 0.19% to 1.6%. In Vermeire’s cohort of 125 patients with IBD and a positive ANA, only 1.6% (n = 2) of patients on infliximab developed DILE. In Colombel’s series of 500 patients with Crohn disease on infliximab, DILE was reported in 0.6% (n = 3) of patients. Recently, Yanai and colleagues reported a higher rate of new-onset DILE (6.9%) in a cohort of 289 patients with IBD treated with infliximab or adalimumab. It is unclear if the higher rate is due to better reporting of DILE or to a referral bias.

The mean time of onset of DILE is 14 to 16 months (range, 1–52 months). DILE appears to be more prevalent in women. Subramanian and coworkers reported on 13 patients with IBD and DILE, of whom most were women with high titers of ANA and anti-dsDNA.

Symptoms

Symptoms of DILE include polyarthralgias, myalgias, serositis, fever, fatigue, or nonspecific rashes. The most common presenting symptoms are symmetric polyarthralgias. In contrast, typical lupus most commonly presents with a classic malar rash, discoid rash, alopecia, photosensitivity, and oral ulcers and may have systemic involvement of the kidneys and/or central nervous system (CNS).

Diagnosis

Most patients with DILE do not fulfill the American College of Rheumatology diagnostic criteria for systemic lupus erythematosus, and there are no official diagnostic criteria for anti-TNF-induced lupus. However, De Bandt and colleagues proposed a set of diagnostic criteria for DILE secondary to anti-TNF therapy ( Table 2 ). In Subramanian’s cohort of patients on anti-TNF with DILE, positive autoantibodies and polyarthralgias followed by serositis were the most frequently present criteria. The most commonly detected autoantibodies were ANA and anti-dsDNA; less frequently, anti-histone Ab titers are elevated, a finding typically associated with classic DILE.

Treatment

The mainstay of therapy for DILE is cessation of the anti-TNF agent. If symptoms persist, use of steroids or immunosuppressive therapy is effective. Once symptoms resolve, re-treatment with another anti-TNF seems reasonable, although this recommendation is based on a small number of cases. In the Mayo Clinic series, 5 patients were rechallenged with an anti-TNF (adalimumab or etanercept) and 80% did well. The one patient who developed recurrent DILE was rechallenged with infliximab rather than with a different anti-TNF agent. In Subramanian’s cohort of patients, 8 were rechallenged with a second anti-TNF, 6 with certolizumab pegol, and 2 with adalimumab. Of the 6 patients on certolizumab pegol, only one developed a recurrence of DILE, whereas 1 of the 2 patients treated with adalimumab developed a recurrence. Similarly, Yanai’s study reported rechallenging 14 patients with a different anti-TNF, of whom only one patient developed recurrent DILE. Nevertheless, the follow-up in these studies was short. Thus, it remains unclear if DILE is a true class effect or a reaction to a specific anti-TNF agent. If, after discussion with the patient regarding the risks and benefits, a decision is made to restart an anti-TNF, then a second anti-TNF agent should be tried. However, the agent that led to the initial DILE should be avoided. Based on Subramanian’s findings, certolizumab pegol may be the anti-TNF agent of choice for rechallenge following a DILE.

Epidemiology

Cutaneous vasculitis is the most frequent form of anti-TNF-associated vasculitis, but is extremely rare. Approximately 90% of the vasculitis associated with anti-TNF is a cutaneous leukocytoclastic vasculitis. The mean duration of anti-TNF therapy before developing vasculitis ranges from 9 to 38 weeks.

Symptoms

Palpable purpura is the most common symptom and occurs in 80% of cases. Other symptoms include ulcerated skin lesions, nodules, and a maculopapular rash.

Treatment

Cessation of anti-TNF leads to resolution of the vasculitis in more than 60% to 70% of cases. If symptoms persist, most cases appear to be responsive to systemic steroids with or without immunosuppressive therapy.

It is unclear if anti-TNF therapy can be reinstituted after an anti-TNF-associated vasculitis. In one study, 67% (n = 6/9) of patients rechallenged with the same anti-TNF developed a recurrent vasculitis. In contrast, in the cohort of Saint Marcoux and De Bandt of 9 patients who were rechallenged with a different anti-TNF, 67% tolerated the second agent without difficulty. As with DILE, any attempt at retreatment should be done with a different anti-TNF agent.

Epidemiology

De novo joint pains without any other criteria for DILE have been reported in patients receiving anti-TNF therapy without an obvious flare of the underlying IBD. In a series of 1300 patients who received anti-TNF, 21 developed new disabling polyarthralgias. The mean onset of arthralgias was 12 months from onset of therapy. However, it remains unclear if these represent unique symptoms or if they are related to extraintestinal manifestations of IBD, delayed infusion reactions, or DILE.

Symptoms

Symptoms include morning stiffness of the hands and wrists, but no synovitis. In 11 of the patients, ANA titer was greater than 1:280, but other criteria for DILE were lacking.

Treatment

Treatment involves changing to a different anti-TNF, corticosteroids, or immunomodulator therapy.

Epidemiology

There are numerous reports of psoriasis developing during anti-TNF therapy. Because patients with IBD are at an increased risk of psoriasis, it may be difficult to ascertain how many cases of psoriasis are truly secondary to the anti-TNF therapy versus the underlying disease. The reported prevalence of psoriasis on anti-TNF ranges from 0.6% to 5.3%. In one study from France, 2% of patients with IBD on anti-TNF developed new-onset psoriasis over a period of 4 years.

Psoriasis can occur at any time during anti-TNF therapy. The onset of psoriasis ranges from several days to up to 4 years after starting therapy. In the study of patients with IBD by Rahier and colleagues, the median time to onset of psoriasis was 17 months for infliximab, 12 months for adalimumab, and 4.5 months for certolizumab pegol. In contrast, in Cullen’s review of the literature of anti-TNF-induced psoriasis in patients with IBD, the median time to onset of rash was 5 months for infliximab, 2.5 months for adalimumab, and 5 months for certolizumab pegol. Patients who developed anti-TNF-related psoriasis were more likely to have a personal history of psoriasis, family history of psoriasis, or family/personal history of atopy, and Crohn disease and be female.

Pathophysiology

The exact pathogenesis of anti-TNF-induced psoriasis is unknown. Given that changing to a different anti-TNF does not improve the psoriasis in most cases, it is likely a class effect. One hypothesis is that in genetically predisposed individuals, anti-TNF therapy may act as the “second hit” phenomenon. Studies have shown that TNF inhibition causes an increase in the production of interferon-α by dendritic cells, ultimately resulting in the development of psoriatic lesions.

Symptoms

Although the appearance of idiopathic psoriasis is typically characterized by plaques symmetrically involving the scalp and extensor surfaces, anti-TNF-induced psoriasis has been described as pustular and localized to the palms and soles of the feet in more than 40% of cases. Rahier and colleagues and Cullen and colleagues report other areas of involvement, including the scalp and the flexures. The psoriatic lesions can be painful and pruritic. They can appear as scaly erythematous plaques or pustules on examination. Multiple other forms of psoriasis have also been reported, including plaque-type, guttate, and forms with nail involvement.

Treatment

We recommend dermatologic consultation for cases of anti-TNF-induced psoriasis. Initial therapy involves the use of topical therapies: steroids, keratolytics, emollients, or ultraviolet lights. However, topical therapies only are effective in 25% to 50% of cases. If these therapies fail and the psoriasis is severe, cessation of anti-TNF is warranted. Following anti-TNF withdrawal, the median time to complete resolution of psoriasis is 3 months. Although changing to a different anti-TNF can be attempted, this often leads to recurrence of the psoriasis and is only successful in 15% of cases.

Epidemiology

In a review of patients with RA on anti-TNF therapy, 7% of patients developed new-onset eczema. In patients with IBD, the incidence of anti-TNF-induced eczema was 3%. Eczema occurred a median 12 months after initiation of infliximab and 6 months after adalimumab. It appeared to be more common in men, active smokers, and patients with a history of atopy. The most common types of eczema included dyshidrotic eczema, contact dermatitis, nummular eczema, atopic dermatitis, or nonspecific.

Pathophysiology

The exact pathogenesis of anti-TNF-induced eczema is unknown but presumed to be the same as that described for psoriasis.

Symptoms

The lesions are typically red, scaly, and crusted and involve the scalp, trunk, genitals, face, and flexures. Symptoms are usually pruritus and pain as classically seen with eczema.

Treatment

Most cases of eczema appear to respond to topical therapy. If this fails, then steroids with or without oral immunosuppressants are often effective in inducing remission. However, in severe cases of eczema, cessation of the anti-TNF may be necessary.