Version May 2024
INTRODUCTION — Inhibitors of tumor necrosis factor (TNF)-alpha represent important treatment advances for a number of inflammatory conditions, including rheumatoid arthritis, the seronegative spondyloarthropathies, psoriasis, and inflammatory bowel disease. TNF-alpha inhibitors offer a targeted strategy that contrasts with the nonspecific immunosuppressive agents traditionally used to treat most forms of systemic inflammation.
However, multiple adverse effects of TNF-alpha inhibition have been identified. These include:
●Infections
●Malignancies such as nonmelanoma skin cancers
●Injection site reactions
●Infusion reactions
●Induction of autoimmunity (eg, uveitis, psoriasis)
●Demyelinating disease
●Heart failure
Infections associated with bacterial, viral, and fungal organisms will be reviewed here. Other major complications of TNF-alpha therapy such as tuberculosis and nontuberculous mycobacterial infections, malignancy, the induction of autoimmunity, demyelinating disease, and heart failure are discussed separately. (See "Risk of mycobacterial infection associated with biologic agents and JAK inhibitors" and "Tumor necrosis factor-alpha inhibitors: Risk of malignancy" and "Tumor necrosis factor-alpha inhibitors: Induction of antibodies, autoantibodies, and autoimmune diseases" and "Tumor necrosis factor-alpha inhibitors: An overview of adverse effects".)
The biology of TNF-alpha and some of its effects on the immune system are reviewed elsewhere. (See "Tumor necrosis factor-alpha inhibitors: An overview of adverse effects".)
TNF-ALPHA INHIBITORS — Several inhibitors of TNF-alpha have been approved for the treatment of a variety of inflammatory illnesses (eg, rheumatoid arthritis, Crohn disease) by the US Food and Drug Administration. These medications are:
●Infliximab – A chimeric (mouse/human) anti-TNF-alpha monoclonal antibody
●Adalimumab – A fully human monoclonal anti-TNF-alpha antibody
●Etanercept – A soluble TNF-alpha receptor fusion protein
●Certolizumab pegol – A pegylated Fab fragment of a humanized monoclonal antibody
●Golimumab – A human monoclonal anti-TNF-alpha antibody
These agents are discussed in detail elsewhere. (See "Overview of biologic agents in the rheumatic diseases".)
TNF-ALPHA AND HOST DEFENSES — TNF-alpha is synthesized initially by activated macrophages and T cells as a transmembrane precursor protein. The cytoplasmic tail of this protein is then cleaved to release soluble TNF-alpha. TNF-alpha is important for macrophage activation, phagosome activation, differentiation of monocytes into macrophages, recruitment of neutrophils and macrophages, and granuloma formation and function [1].
Animal experiments have demonstrated the importance of TNF-alpha in protection against several pathogens, including Mycobacterium tuberculosis, Mycobacterium avium, Mycobacterium bovis, Bacillus Calmette-Guérin (BCG), Aspergillus fumigatus, Histoplasma capsulatum, Coccidioides species, Toxoplasma gondii, Cryptococcus neoformans, Candida albicans, and multiple viral pathogens [2-6]. These organisms are not killed readily by host defense mechanisms but rather are sequestered within granulomas, which are comprised of a central core of macrophages, multinucleated giant cells, and necrotic debris surrounded by macrophages and lymphocytes [7]. TNF-alpha is required for the orderly recruitment of these cells and for continued maintenance and function of the granuloma [8].
SCREENING FOR INFECTIONS — A variety of recommendations have been issued regarding screening for certain infections; prior to receiving a TNF-alpha inhibitor, patients should be screened for tuberculosis, hepatitis B, and hepatitis C [9-12]. Screening for these infections is discussed in detail separately. (See "Risk of mycobacterial infection associated with biologic agents and JAK inhibitors", section on 'Screening' and "Hepatitis B virus: Screening and diagnosis in adults" and "Screening and diagnosis of chronic hepatitis C virus infection".)
It is also important to obtain a history of possible recent (ie, in the last two years) infection with endemic fungi, as discussed below. (See 'Histoplasmosis' below and 'Coccidioidomycosis' below and 'Cryptococcus' below.)
CONTRAINDICATIONS — The American College of Rheumatology recommends that TNF-alpha inhibitors not be administered when any of the following infections are present [12,13]:
●Active bacterial infections or bacterial infections requiring antibiotic therapy
●Active tuberculosis (TB) or untreated latent TB (see "Risk of mycobacterial infection associated with biologic agents and JAK inhibitors")
●Active herpes zoster infection
●Active invasive fungal infections (eg, coccidioidomycosis, invasive aspergillosis)
●Nonhealed infected skin ulcers
●Acute infection with hepatitis B virus (HBV) or hepatitis C virus (HCV), untreated chronic HBV infection, or chronic HBV or HCV infection with significant liver injury, defined as Child-Pugh classes B or C (calculator 1)
We agree with these recommendations. If a TNF-alpha inhibitor is started in a patient with chronic HBV infection (ie, positive HbSag), antiviral therapy should be used concomitantly. (See 'Hepatitis B' below.)
Decisions about the appropriate timing for initiating a TNF-alpha inhibitor following resolution of the infections described above should be made on a case-by-case basis. For some infections (eg, coccidioidomycosis, pulmonary nontuberculous mycobacteria), resolution can be difficult to identify or document, and the decision regarding anti-TNF therapy resumption or initiation (potentially with concomitant antimicrobial therapy) should be made in conjunction with an infectious diseases specialist.
ASSESSING THE RISK OF SERIOUS INFECTION — Data from randomized trials and long-term open-label extension studies of biologic therapies have generally (but not always) suggested some increase in infectious risks associated with these compounds. While these studies have the benefit of being randomized with regard to drug exposure at the start of the study, they are often small, include relatively healthy patients (ie, lacking comorbidities), are underpowered to evaluate uncommon events, and often involve patients using concomitant immunosuppressants (eg, prednisone) that also can increase infectious risk. In addition, patients with rheumatoid arthritis and certain other inflammatory diseases have higher rates of pneumonia, skin and soft tissue infections, and other serious infections due to the underlying disease itself [14]. The majority of these infections are thought to be bacterial, although community-acquired viruses (eg, influenza) likely also play an important role.
A 2011 network meta-analysis that included 50,010 patients from 163 randomized trials and 46 long-term open-label extension studies (some of which involved biologic agents other than TNF-alpha inhibitors) showed an elevated risk of serious infections (ie, infections requiring hospitalization) for each TNF-alpha inhibitor, although the risk was only statistically significant for certolizumab (odds ratio [OR] 3.51, 95% CI 1.59-7.79) [15]. A 2015 meta-analysis that included 106 randomized trials of patients with rheumatoid arthritis who received biologic drugs found that standard-dose biologic drugs (OR 1.31, 95% CI 1.09-1.58) and high-dose biologic drugs (OR 1.90, 95% CI 1.50-2.39) were associated with an increased risk of serious infections compared with traditional disease-modifying antirheumatic drugs (DMARDs), but low-dose biologic drugs were not [16]. The risk was lower in patients who were methotrexate-naive compared with those who had previously received a traditional DMARD or a TNF-alpha inhibitor.
In general, observational studies have the benefit of evaluating risk in "real-world" populations and often have great statistical power. Given the lack of randomization, however, confounding can skew results, although several elegant studies have used propensity scores and other methods in an attempt to control for this. Many registry and population-based studies have documented "real-world" incidence rates of infection. Most, but not all, of these studies have suggested an increased risk in infection requiring hospitalization in patients receiving a TNF-alpha inhibitor, particularly during the first 6 to 12 months after drug initiation. Incidence rates from a selection of these population-based studies range from 2.9 to 8.2 infections requiring hospitalization per 100 patient-years of exposure to a TNF-alpha inhibitor [17]. Several studies suggest a higher risk of infection requiring hospitalization with infliximab than with adalimumab or etanercept, although it is unclear whether this is due to the drug itself or differences in concomitant immunosuppressive use or other factors [18-20]. A 2018 study from the British registry suggested similar risks between all of the TNF-alpha inhibitors and that risks could be higher with tocilizumab, a biologic agent that inhibits the interleukin 6 receptor [21].
Perhaps the most robust analysis with regard to controlling for differences between exposed populations before and after drug initiation comes from the German biologic registry. After controlling for concomitant prednisone use in time-varying fashion as well as for various comorbidities, a statistically significant increased risk (incidence rate ratio 1.8, 95% CI 1.2-2.7) for infection requiring hospitalization was found to be associated with TNF-alpha inhibitor (etanercept, adalimumab, and infliximab were studied) use [22]. The investigators developed a risk calculator and showed that the majority of infection risk is related to patient factors (ie, comorbidities and age) and concomitant use of prednisone.
Other population-based studies in the United States have had some variability in risk estimates. In a large retrospective population-based study using Medicare, health plan, and health maintenance organization data in patients with rheumatoid arthritis (RA), inflammatory bowel disease, psoriatic arthritis, or ankylosing spondylitis, no increase in the risk of hospitalization for infection was observed in patients initiating treatment with a TNF-alpha inhibitor compared with nonbiologic agents [23]. Among patients with RA, infliximab was associated with an increased risk of serious infections compared with etanercept (adjusted hazard ratio [aHR] 1.26, 95% CI 1.07-1.47) and adalimumab (aHR 1.23, 95% CI 1.02-1.48). An accompanying editorial pointed out that possible reasons for the lack of difference in the risk of hospitalization include multiple potential confounders in both treatment groups as well as the possibility that the infection rate in the comparator group may have been higher than in prior studies [24].
The US Food and Drug Administration has stated that patients receiving a TNF-alpha inhibitor who are greater than 65 years of age or who are taking concomitant immunosuppressants may be at even greater risk of infection [25]. The absolute risks are higher for older adults, although the relative risk between TNF-alpha inhibitors and nonbiologic DMARDs in this group remains similar [22,26]. Some comorbidities can also elevate the risk of serious infections substantially. As an example, in the United States population-based analysis described above, the crude incidence of infections requiring hospitalization for patients taking a TNF-alpha inhibitor without chronic obstructive pulmonary disease (COPD) was 7.2 per 100 patient-years (95% CI 6.5-8.1), whereas it was more than twice as high (15.0 per 100 patient-years, 95% CI 11.8-18.9) in those with COPD [23].
Most studies of infections in patients receiving TNF-alpha inhibitors have been conducted in adults. In a systematic review of studies that focused on children and adolescents receiving a TNF-alpha inhibitor for juvenile idiopathic arthritis or inflammatory bowel disease, the most frequently reported infections were upper respiratory tract infections and other mild infections [27]. Severe bacterial and fungal infections also occurred. Few patients developed tuberculosis, likely due at least in part to effective screening.
TIMING OF INFECTION RISK — Some data indicate that the risk of infection associated with TNF-alpha inhibitor use is highest in the period shortly after the initiation of therapy. In an analysis of the British Society for Rheumatology Biologics Register [28], the adjusted incidence ratio for infection was not elevated for patients receiving TNF-alpha inhibitors compared with those on disease-modifying antirheumatic drugs alone (ratio 1.22, 95% CI 0.88-1.69). However, when the analysis was limited to the first 90 days of follow-up after the initiation of a new therapy, the adjusted incidence rate ratio was 4.6 (95% CI 1.8-11.9) among patients treated with TNF-alpha inhibitors.
Increased risk of tuberculosis in the period immediately following the institution of TNF-alpha inhibition has also been reported. (See "Risk of mycobacterial infection associated with biologic agents and JAK inhibitors".)
MYCOBACTERIAL INFECTIONS — All TNF-alpha inhibitors increase the risk of tuberculosis (TB), and all patients should be evaluated for TB, both by history and screening assays, prior to starting therapy. (See "Risk of mycobacterial infection associated with biologic agents and JAK inhibitors", section on 'Quantifying risk for TB'.)
The risk of nontuberculous mycobacterial (NTM) infections is also increased with TNF-alpha inhibitor therapy [29,30]. In the United States, pulmonary and extrapulmonary NTM infections are several-fold more common than TB. The approach to diagnosis and management of NTM infections for patients receiving TNF-alpha inhibitor therapy is discussed separately. (See "Risk of mycobacterial infection associated with biologic agents and JAK inhibitors", section on 'Nontuberculous mycobacterial disease'.)
BACTERIAL INFECTIONS
Risk of bacterial infections — The majority of infections requiring hospitalization described in the clinical trial and population-based studies above have been evaluated or reported without regard to specific site of infection or pathogen and have been presumed to be due to common bacterial pathogens (see 'Assessing the risk of serious infection' above). However, some studies have reported either site-specific infections or specific bacterial pathogens, as discussed below.
Septic arthritis — The British Society for Rheumatology Biologics Register, a prospective observational study of patients with rheumatoid arthritis, compared the risk of septic arthritis in 11,881 patients who received a TNF-alpha inhibitor and 3673 patients who received nonbiologic disease-modifying antirheumatic drugs (DMARDs) [31]. One hundred seventy-nine patients who received a TNF-alpha inhibitor had at least one episode of septic arthritis compared with 20 patients who received DMARDs. The adjusted hazard ratio for septic arthritis in patients who received a TNF-alpha inhibitor was 2.3 (95% CI 1.2-4.4). The risk was highest in the early months following initiation of TNF-alpha inhibitor therapy and did not differ based upon which TNF-alpha inhibitor was used.
Staphylococcus aureus was the most common cause of septic arthritis in both cohorts, occurring in 57 percent of patients who received a TNF-alpha inhibitor and 43 percent of patients who received DMARDs. In the patients who received a TNF-alpha inhibitor, several cases were caused by organisms that rarely cause septic arthritis, including five cases caused by intracellular pathogens (three cases caused by Salmonella spp and two cases caused by Listeria spp) and four cases caused by Pseudomonas aeruginosa. (See "Septic arthritis in adults".)
Listeriosis — Listeria monocytogenes is a gram-positive intracellular bacterium that is associated most often with meningoencephalitis and/or septicemia. The disease is linked to the ingestion of contaminated dairy and meat products. (See "Clinical manifestations and diagnosis of Listeria monocytogenes infection".)
Listeriosis has been described in patients receiving TNF-alpha inhibitors [25,30,32-34], and, in 2011, the US Food and Drug Administration (FDA) added a boxed warning about the risk of listeriosis for the entire class of TNF-alpha inhibitors [25]. In a report on granulomatous infectious diseases from the FDA Adverse Events Reporting System from 1998 to 2002, there were 38 cases of Listeria infection: 36 with infliximab and 2 with etanercept (15.5 versus 1.8 per 100,000 patients, a difference that was highly significant) [30]. A search of the medical literature in 2011 identified 26 published cases of listeriosis in patients who had received a TNF-alpha inhibitor, including seven deaths [25]. (See "Epidemiology and pathogenesis of Listeria monocytogenes infection".)
The acquisition of Listeria through the ingestion of contaminated foods emphasizes the importance of educating patients regarding food preparation and safety. Preventive measures to minimize the risk of foodborne Listeria infection are discussed separately. (See "Treatment and prevention of Listeria monocytogenes infection", section on 'Prevention of foodborne infection'.)
Legionella pneumonia — In 2011, the FDA added a boxed warning about the risk of Legionella infection for the entire class of TNF-alpha inhibitors [25]. This warning was issued because an increased risk of Legionella infection was detected among patients who had received a TNF-alpha inhibitor. Between 1999 and 2010, 80 cases of Legionella pneumonia, including four deaths, were reported to the FDA's Adverse Events Reporting System in patients receiving a TNF-alpha inhibitor.
A population-based study conducted in France suggested that Legionella pneumonia risk with TNF-alpha inhibitors is elevated as compared with the general population and that higher risk exists with adalimumab and infliximab than with etanercept [35].
Legionella infection is discussed in greater detail separately. (See "Clinical manifestations and diagnosis of Legionella infection" and "Treatment and prevention of Legionella infection" and "Microbiology, epidemiology, and pathogenesis of Legionella infection", section on 'Host risk factors'.)
Other important pathogens — Other serious bacterial infections have been reported across the class of TNF-alpha inhibitors, including but not limited to nocardiosis, actinomycosis, and salmonellosis [30,36].
VIRAL INFECTIONS — There are few data related to the impact of TNF-alpha inhibitor use on patients with viral infections. The effects of TNF-alpha inhibition on the course of viral infections may vary according to the specific type of infection. As an example, inhibition of TNF-alpha appears to delay the clearance of hepatitis B but may attenuate hepatic injury with hepatitis C [37-39]. For severe acute respiratory syndrome coronavirus 2, observational data suggest patients using TNF-alpha inhibitors at the time of infection might have a lower or neutral risk of severe coronavirus disease 2019 outcomes [40,41].
Hepatitis B — The potential exists for reactivation of certain viral infections by TNF-alpha inhibitor therapy either directly due to lack of TNF-alpha or indirectly via effects on T cell activation and interferon (IFN) production. This appears most relevant in the case of hepatitis B virus (HBV), an intracellular infection that is controlled by cytokine and cellular mechanisms [37,38,42]. There are a number of reports of patients with HBV infections that reactivated during treatment with TNF-alpha inhibitors or other biologic agents [43-48].
The American College of Rheumatology recommends that patients with active HBV infection who are receiving or have received effective antiviral therapy should be treated for rheumatoid arthritis (RA) according to recommendations for those without hepatitis B infection (ie, they should receive a TNF-alpha inhibitor if indicated) [12]. In contrast, they recommend that TNF-alpha inhibitors be avoided in patients with untreated chronic HBV infection or with chronic HBV infection with significant liver injury, defined as Child-Pugh classes B or C (calculator 1) [13]. We agree with these recommendations. (See 'Contraindications' above.)
In patients with untreated HBV infection who do not have Child-Pugh class B or C liver injury, a TNF-alpha inhibitor can be given, although some patients require antiviral therapy to prevent HBV reactivation. We use the following approach [49-51]:
●In HBsAg-positive patients (ie, patients with chronic HBV) starting a TNF-alpha inhibitor, we recommend starting an antiviral agent (eg, tenofovir or entecavir) prior to initiating the TNF-alpha inhibitor.
●For patients who are HBsAg negative but HBcAb positive (resolved HBV), we monitor for viral reactivation during TNF-alpha inhibitor therapy, particularly in those lacking evidence of immunity (HBsAb negative). In such patients, we check HBV DNA and transaminase levels before starting therapy and every three months thereafter. Some experts monitor HBV DNA and transaminases at more frequent intervals (ie, every one to two months) [52]. We initiate antiviral therapy if reactivation occurs. This issue is discussed in detail separately. (See "Hepatitis B virus reactivation associated with immunosuppressive therapy", section on 'Categorizing level of risk' and "Hepatitis B virus reactivation associated with immunosuppressive therapy", section on 'Summary and recommendations'.)
Hepatitis C — All patients should be screened for hepatitis C infection before starting TNF-alpha inhibitor therapy. Patients with chronic hepatitis C virus (HCV) infection should be treated given the wide availability of safe and effective antiviral therapy. (See "Overview of the management of chronic hepatitis C virus infection".)
The optimal approach to treatment for patients with chronic HCV infection who are starting or receiving TNF-alpha inhibitor therapy has not been determined. We generally tailor our approach based on patient characteristics (eg, degree of liver injury, acuity of HCV infection, and need for the TNF-alpha inhibitor). In general, we prefer to start antiviral therapy and assess virologic response before starting anti-TNF inhibitor therapy. However, case reports suggest that concurrent treatment with TNF-alpha inhibitor and antiviral therapy may be safe for patients without liver injury [53]. One randomized trial of etanercept as adjunctive therapy to IFN-alpha plus ribavirin (a regimen no longer used) found that etanercept improved the virologic response rate and decreased the rate of adverse events [54]. Even in the absence of antiviral treatment, worsening of HCV infection in patients receiving etanercept or infliximab appear to be uncommon [55-63]. The American College of Rheumatology suggests etanercept as the preferred TNF-alpha inhibitor if such therapy is used in patients with HCV infection [64]. We agree with this approach.
By contrast, we avoid TNF-alpha inhibitors in patients with acute HCV infection or with chronic HCV infection with significant liver injury, defined as Child-Pugh classes B or C (calculator 1), because progression of HCV has been observed in some patients receiving a TNF-alpha inhibitor [13,64]. (See 'Contraindications' above.)
Herpes zoster — There is an increased risk of herpes zoster in patients with underlying autoimmune diseases (eg, rheumatoid arthritis, inflammatory bowel disease), particularly in those receiving immunosuppressive therapies including glucocorticoids and disease-modifying antirheumatic drugs (DMARDs). (See "Epidemiology, clinical manifestations, and diagnosis of herpes zoster", section on 'Autoimmune disease'.)
There are conflicting data regarding the question of whether TNF-alpha inhibitors confer a greater risk of herpes zoster compared with nonbiologic therapies for autoimmune diseases. Several large population-based studies have been conducted, with those in the United States not finding an increased risk with TNF-alpha inhibitors, but those in Europe generally showing an increased risk. The largest study to address this issue was a multicenter cohort study performed in the United States that compared the incidence of herpes zoster in 33,324 patients initiating a TNF-alpha inhibitor and 25,742 patients initiating nonbiologic DMARDs (eg, methotrexate, leflunomide); underlying diseases included rheumatoid arthritis, inflammatory bowel disease, psoriasis, psoriatic arthritis, and ankylosing spondylitis [65]. Patients who initiated a TNF-alpha inhibitor were not at higher risk for herpes zoster than patients initiating nonbiologic DMARDs, and no differences in risk were observed among infliximab, etanercept, and adalimumab. A large retrospective study of patients with rheumatoid arthritis among United States veterans also showed that TNF-alpha inhibitors increased the risk of herpes zoster to a similar degree as other nonbiologic therapies [66].
In contrast, in a prospective cohort study that included 5040 patients with rheumatoid arthritis in Europe, after adjustment for age, disease severity, and prednisone use, use of infliximab or adalimumab was associated with a moderately increased risk of herpes zoster (hazard ratio [HR] 1.82, 95% CI 1.05-3.15) compared with nonbiologic DMARDs [67]. There was a trend toward an increased risk with etanercept, although it did not achieve statistical significance. A later analysis from the British Biologic Registry identified a significantly increased risk (HR between 1.5 [95% CI 1.1-2.0] and 2.2 [95% CI 1.4-3.4]) for the three TNF-alpha inhibitors evaluated (etanercept, infliximab, and adalimumab) [68]. It is unclear why European studies have found increased risk but United States studies have not. Differential use of prednisone (a strong risk factor for disease identified in all of the above studies) between European and United States studies might help explain the difference in findings.
The risk of herpes zoster has also been compared in those receiving a TNF-alpha inhibitor or another biologic DMARD. A cohort analysis of the Medicare population compared the risk of herpes zoster associated with TNF-alpha inhibitor therapy with other biologic DMARDs (eg, abatacept, tocilizumab, rituximab) and suggested that the risk was similar between TNF-alpha inhibitors and other biologic DMARDs used to treat rheumatoid arthritis [69].
Human immunodeficiency virus — There are several case reports describing the use of TNF-alpha inhibitors to treat rheumatoid arthritis or other inflammatory conditions in the setting of HIV infection [70,71]. In general, these reports suggest that TNF-alpha inhibition can be tolerated well by HIV-infected patients, provided that the patient is on an effective antiretroviral regimen before starting a TNF-alpha inhibitor.
Other viruses — A number of other viral infections, such as those caused by cytomegalovirus, herpes simplex virus [72], and molluscum contagiosum, have been reported to occur in patients treated with TNF-alpha inhibitors, but the true nature of the association between these infections and TNF-alpha inhibition is not clear.
FUNGAL INFECTIONS — The critical role of TNF-alpha in granuloma formation and its other contributions to host defense make granulomatous infections an anticipated complication of TNF-alpha inhibitors. Data from human studies have confirmed the heightened risk of granulomatous infections, including fungal infections, associated with TNF-alpha inhibitor use. (See "Tumor necrosis factor-alpha inhibitors: An overview of adverse effects", section on 'TNF-alpha and host defenses'.)
In addition to fungal infections usually associated with granulomatous inflammation, cases of Pneumocystis pneumonia (PCP) have also been reported among patients treated with targeted TNF-alpha inhibition [73,74].
Granulomatous infections — Fungi such as H. capsulatum and Coccidioides spp are found readily in the environment in endemic areas. Both reactivation of latent disease and new infections with histoplasmosis and coccidioidomycosis have been described with the use of TNF-alpha inhibitors, particularly infliximab [75]. Blastomycosis is described less frequently as a complication of TNF-alpha inhibitor therapy. (See "Diagnosis and treatment of disseminated histoplasmosis in patients without HIV" and "Diagnosis and treatment of pulmonary histoplasmosis" and "Management considerations, screening, and prevention of coccidioidomycosis in immunocompromised individuals and pregnant patients" and "Mycology, pathogenesis, and epidemiology of blastomycosis", section on 'Epidemiology'.)
In 2008, the US Food and Drug Administration (FDA) issued a warning about the risk for pulmonary and disseminated histoplasmosis, coccidioidomycosis, blastomycosis, and other invasive fungal infections in patients undergoing treatment with infliximab or other TNF-alpha inhibitors [76]. The warning is based on reports of 240 cases of histoplasmosis in patients treated with TNF-alpha inhibitors. In at least 21 of these reports, histoplasmosis was not recognized initially, which led to a delay in antifungal therapy; there were 12 fatal outcomes.
Clinicians should have a high index of suspicion for invasive fungal infections, particularly in patients who live in or have traveled to endemic regions, and investigate for these infections and treat promptly. Typical signs and symptoms of histoplasmosis and other invasive fungal infections include fever, malaise, weight loss, sweats, cough, dyspnea, pulmonary infiltrates on radiograph, or serious systemic illness including shock. (See "Pathogenesis and clinical manifestations of disseminated histoplasmosis" and "Pathogenesis and clinical features of pulmonary histoplasmosis".)
Histoplasmosis — H. capsulatum is acquired through inhalation of mycelial fragments and microconidia, typically from soil. (See "Pathogenesis and clinical manifestations of disseminated histoplasmosis" and "Pathogenesis and clinical features of pulmonary histoplasmosis".)
Ten cases of H. capsulatum infections were reported to the FDA Adverse Events Reporting System (AERS) from the late 1990s through July 2001 [77]. By September 2008, 240 additional cases of histoplasmosis associated with TNF-alpha inhibitor were reported to AERS, and since then this has been a well-recognized complication of TNF-alpha inhibitor therapy [76]. It is likely that these numbers are underestimates as the infection is not typically reportable to public health agencies. The largest series to date of histoplasmosis cases occurring in patients receiving a TNF-alpha inhibitor included 98 patients [78]. These cases occurred within an endemic region in the Midwest of the United States, and many were also using methotrexate (44 percent) or prednisone (34 percent). TNF-alpha inhibitor therapy was initially discontinued in 95 of 98 patients (96.9 percent) and later resumed in 25 of 74 (33.8 percent) either during or after successful antifungal therapy (median time to resumption was 12 months after diagnosis). Overall, 3.2 percent of patients died and nine (9 percent) developed immune reconstitution inflammatory syndrome (IRIS) during therapy, of whom three were given steroids for IRIS management. No patients reinstituted TNF-antagonists for IRIS. (See "Immune reconstitution inflammatory syndrome".)
Screening for histoplasmosis involves questioning TNF-alpha inhibitor candidates about potential exposures with specific reference to high-risk activities (eg, demolition of old buildings, cleaning chicken coops, spelunking) in endemic areas and symptoms of active or recent histoplasmosis [79]. Areas of endemicity include parts of the United States (particularly Midwestern and central states), Mexico, Central and South America, Africa, and Asia. In patients with a history of high-risk activities, symptoms of active histoplasmosis, or a history of histoplasmosis within the past two years, a chest radiograph should be obtained. The routine use of Histoplasma serologies or antigen testing is not recommended for screening, as these tests are insensitive in this setting; however, these tests are indicated as part of the diagnostic evaluation in patients suspected of having histoplasmosis. As current diagnostic tests perform poorly in immunocompromised hosts, an empiric antifungal therapy may be needed when there is a high suspicion of disease [80]. The diagnosis and treatment of histoplasmosis are discussed in greater detail separately. (See "Pathogenesis and clinical features of pulmonary histoplasmosis", section on 'Epidemiology' and "Diagnosis and treatment of pulmonary histoplasmosis" and "Diagnosis and treatment of disseminated histoplasmosis in patients without HIV".)
The FDA suggests that clinicians generally discontinue the TNF-alpha inhibitor in patients who develop histoplasmosis or other invasive fungal infections during therapy [76]. However, as discussed above, some patients have experienced a paradoxical clinical worsening following discontinuation of the TNF-alpha inhibitor despite evidence of improvement of the fungal infection; such a pattern is suggestive of IRIS [79]. In some cases, this has necessitated reinstitution of immunosuppressive therapy, typically with prednisone. Similar reactions can occur during treatment of tuberculosis and other opportunistic infections. (See "Immune reconstitution inflammatory syndrome".)
Coccidioidomycosis — Coccidioides spp are dimorphic fungi endemic to the southwestern United States and parts of Central and South America. (See "Primary pulmonary coccidioidal infection" and "Management considerations, screening, and prevention of coccidioidomycosis in immunocompromised individuals and pregnant patients".)
There are limited data on the occurrence, prophylaxis, and management of Coccidioides immitis infections among patients treated with TNF-alpha inhibitors. In a study of patients from clinics in endemic areas in California, Arizona, and Nevada, the following findings were noted [81]:
●Thirteen cases were reported over a five-year period; 12 were associated with infliximab, and 1 was associated with etanercept. Two of the infections were believed to be reactivation of previous coccidioidal infections.
●Among the 247 patients treated with infliximab and the 738 patients treated with other therapies for inflammatory arthritis, the rate of symptomatic coccidioidomycosis over the five-year observation period was significantly higher with infliximab (2.8 versus 0.5 percent, relative risk 5.2, 95% CI 1.5-17.7).
●Eleven of the 12 patients receiving infliximab were also receiving methotrexate, and two were also taking additional immunosuppressive medications.
●All 13 patients had pneumonia at presentation, and 4 had documented dissemination of the Coccidioides infection to the blood, spleen, peritoneum, or central nervous system.
●Two patients died (one from staphylococcal line sepsis during the course of therapy for coccidioidomycosis).
In a more recent case series report detailing 35 patients using anti-TNF therapy, most patients with pulmonary disease were able to successfully resume anti-TNF therapy during or after antifungal treatment [82].
In areas in which Coccidioides spp are endemic, recommendations prior to the institution of a TNF-alpha inhibitor include a chest radiograph and consideration of a coccidioidal serologic test for immunoglobulin (Ig)M and IgG. However, baseline testing may be of limited utility because, as shown in the prior study, most coccidioidomycosis cases after anti-TNF-alpha therapy represent acute infection rather than reactivation [81]. On the other hand, one retrospective study suggested that screening patients receiving a TNF-alpha inhibitor in an endemic area might reduce the risk of subsequent symptomatic coccidioidomycosis [83].
The possible role of antifungal prophylaxis to prevent coccidioidal infection is discussed separately. (See "Management considerations, screening, and prevention of coccidioidomycosis in immunocompromised individuals and pregnant patients", section on 'Patients taking immunomodulatory agents'.)
Cryptococcus — C. neoformans is an encapsulated fungus that is distributed worldwide. The organism is typically found in soil and in especially high concentrations near bird roosts. Inhalation of the organism frequently occurs after disruption of soil. (See "Microbiology and epidemiology of Cryptococcus neoformans infection".)
In animal models of Cryptococcus infection, TNF-alpha inhibition is associated with impaired recruitment of activated CD4+ T cells and Langerhans cells to the area of infection and with an increased risk of dissemination [4,84].
Cryptococcal infections were described in 19 patients reported to AERS between 1998 and 2002 [30]. Eleven of the patients had received infliximab, and eight had received etanercept (4.7 and 7.1 infections per 100,000 treated patients, respectively, a difference that was not significantly different). Most patients in this report and others were receiving additional immunosuppressive agents, particularly methotrexate [85-87].
Among patients with AIDS who have experienced cryptococcal infections, secondary prophylaxis is recommended until immune reconstitution occurs. There are no data about whether secondary prophylaxis is necessary or optimal in patients receiving a TNF-alpha inhibitor who develop cryptococcosis. (See "Cryptococcus neoformans meningoencephalitis in persons with HIV: Treatment and prevention", section on 'Preventing symptomatic disease'.)
Aspergillosis — Aspergillus species (eg, A. fumigatus) are ubiquitous environmental fungi capable of causing a wide variety of diseases in both healthy and immunocompromised hosts. (See "Epidemiology and clinical manifestations of invasive aspergillosis".)
TNF-alpha inhibition may increase the risk of aspergillosis through the inhibition of neutrophil recruitment. Case reports and small cases series exist of A. fumigatus infections associated with TNF-alpha inhibition [30,88]. However, the interpretation of many cases is complicated by the simultaneous use of several immunosuppressive agents.
Aspergillosis was noted in 39 patients in the AERS report cited above: 29 with infliximab and 10 with etanercept (12.4 versus 8.8 per 100,000 patients, a difference that was not significant) [30].
Pneumocystis pneumonia — A review from AERS from 1998 to 2003 included 84 patients with PCP associated with infliximab use [73]. The following findings were noted:
●The mean time between the first infliximab infusion and the onset of PCP symptoms was 21 days.
●Death occurred in 23 of the 84 patients (27 percent).
●Concomitant immunosuppressive medications included methotrexate, prednisone, azathioprine, 6-mercaptopurine, and cyclosporine.
●Prednisone doses and CD4 lymphocyte counts at the time of PCP infection were not reported.
In a case-control study, risk factors for PCP in patients receiving infliximab included older age, coexisting pulmonary disease, and high doses of glucocorticoids [89]. Because the risk of PCP correlates with high-dose glucocorticoid use in some studies, PCP prophylaxis should be considered among patients treated with TNF-alpha inhibitors if they are also receiving high-dose glucocorticoids (eg, ≥20 mg prednisone equivalents daily for ≥4 weeks) or other intensive immunosuppression [90]. Otherwise, routine PCP prophylaxis for those using TNF therapy is likely of little benefit. (See "Treatment and prevention of Pneumocystis pneumonia in patients without HIV", section on 'Prophylaxis'.)
A later United States population-based study of nonviral opportunistic infections identified a significantly elevated crude incidence rate of 270 cases per 100,000 patient-years among those starting a TNF-alpha inhibitor compared with 170 cases per 100,000 for patients using nonbiologic disease-modifying antirheumatic drugs, and PCP was the most common opportunistic infection identified (crude incidence 56 cases per 100,000 patient-years) [36].
OTHER INFECTION ISSUES — As noted by the preceding findings, there is clear evidence for an increased risk of a variety of infections among patients treated with TNF-alpha inhibitors. However, the risk of these infections must be interpreted in the context of the potential benefits and of the possible adverse effects of conventional therapies. The decision to use an anti-TNF-alpha agent is an individual one, based upon the specific clinical features and unique risk profile of a given patient.
Patients should be educated about the importance of recognizing symptoms and signs of infections, particularly fever, because of the heightened risk of these complications and the greater likelihood of disseminated infections among immunocompromised hosts. Patients with symptoms or signs of infections should be evaluated without delay.
Perioperative use of TNF-alpha inhibitors — The use of TNF-alpha inhibitors has been associated with an increased risk of perioperative infections in some, but not all, studies. A meta-analysis that included eight observational studies and three case-control studies found an increased risk of surgical site infection for those receiving a TNF-alpha inhibitor at the time of surgery versus those without recent exposure (pooled odds ratio [OR] 2.47, 95% CI 1.66-3.68) [91]. The studies were small, however, and many had few infectious events. No randomized controlled trials were identified. Further, a real-world population-based analysis evaluated infliximab in the perioperative setting and found no difference in postoperative infection in those stopping infliximab within four weeks of surgery compared with those stopping earlier [92-94]. Several other population-based studies found no difference in risk among those using TNF-alpha inhibitors just prior to surgery as compared with those using methotrexate and similar risks among patients using TNF-alpha inhibitors as compared with other biologics.
Although the data are mixed, based upon recommendations from the American College of Rheumatology and the American Association of Hip and Knee Surgeons, TNF-alpha inhibitors should generally be withheld for several weeks before orthopedic surgery (as close to one dosing cycle as scheduling allows) and can be resumed after evidence of wound healing, typically after 14 days [95]. However, the potential benefit of preventing some postoperative infections must be balanced against the risk of a perioperative flare in rheumatoid arthritis activity. If a TNF-alpha inhibitor is discontinued before a surgical procedure, close monitoring of the patient for a rheumatoid arthritis flare is necessary.
Combination therapy with anakinra — A higher rate of serious infections has been observed in patients receiving the combination of etanercept and anakinra (a recombinant human interleukin 1 receptor antagonist) compared with trials of either agent alone [96]. The use of TNF-alpha inhibitors in combination with anakinra is not recommended.
VACCINES — The approach to vaccination for patients receiving immunosuppressive therapy for autoimmune inflammatory rheumatic disorders is discussed separately. (See "Immunizations in autoimmune inflammatory rheumatic disease in adults".)
SUMMARY AND RECOMMENDATIONS
●Assessing the risk of serious infection – Tumor necrosis factor (TNF)-alpha inhibitor use is associated with an increased risk of serious infections. The increase in risk is seen with a variety of infections, including tuberculosis (TB) and bacterial, viral, and fungal infections. This risk must be interpreted in the context of the potential benefits and of the possible adverse effects of conventional therapies. (See 'Assessing the risk of serious infection' above.)
●Pretreatment screening for infections – Prior to initiating a TNF-alpha inhibitor, all patients should be screened for TB, hepatitis B, and hepatitis C. It is also important to obtain a history of possible exposure to endemic fungi. (See 'Screening for infections' above and "Risk of mycobacterial infection associated with biologic agents and JAK inhibitors", section on 'Screening' and 'Granulomatous infections' above.)
●Contraindications – TNF-alpha inhibitors should not be administered when any of the following infections are present (see 'Contraindications' above):
•Active bacterial infections or bacterial infections requiring antibiotic therapy
•Active TB or untreated latent TB (see "Risk of mycobacterial infection associated with biologic agents and JAK inhibitors")
•Active herpes zoster infection
•Active invasive fungal infections (eg, coccidioidomycosis, invasive aspergillosis)
•Nonhealed infected skin ulcers
•Acute infection with hepatitis B virus (HBV) or hepatitis C virus (HCV), untreated chronic HBV infection, or chronic HBV or HCV infection with significant liver injury, defined as Child-Pugh classes B or C (calculator 1)
●Bacterial infections – Because of the risk of listeriosis, patients on TNF-alpha inhibitors should follow recommendations to avoid foodborne infections. (See 'Bacterial infections' above and "Treatment and prevention of Listeria monocytogenes infection", section on 'Prevention of foodborne infection'.)
●Viral infections
•Hepatitis B – In patients with untreated HBV infection, there is increased risk of HBV reactivation in the setting of TNF-alpha inhibitor therapy. In patients with untreated HBV infection who do not have Child-Pugh class B or C liver injury, a TNF-alpha inhibitor can be given, although some patients require antiviral therapy to prevent HBV reactivation:
-For HBsAg-positive patients (ie, patients with chronic HBV) starting a TNF-alpha inhibitor, we recommend starting an antiviral agent prior to initiating the TNF-alpha inhibitor (Grade 1B). We use tenofovir or entecavir when an antiviral agent is indicated.
-In patients who are HBsAg negative but HBcAb positive (ie, patients with resolved HBV), we suggest monitoring for HBV reactivation rather than administering preventive antiviral therapy (Grade 2C). In such patients, we check HBV DNA and transaminase levels before starting therapy and every three months thereafter. We initiate antiviral therapy if reactivation occurs. (See 'Hepatitis B' above.)
•Hepatitis C – Patients with chronic HCV infection should be treated given the wide availability of safe and effective antiviral therapy. We generally individualize our approach based on patient characteristics (eg, degree of liver injury, acuity of HCV infection, and need for the TNF-alpha inhibitor). In general, we prefer to start antiviral therapy and assess virologic response before starting anti-TNF inhibitor therapy in patients with chronic infection. However, limited data suggest that concurrent treatment with TNF-alpha inhibitor and antiviral therapy may be safe for patients without liver injury. By contrast, we avoid TNF-alpha inhibitors in patients with acute HCV infection or with chronic HCV infection with significant liver injury, defined as Child-Pugh classes B or C. If TNF-alpha inhibitor therapy is used in patients with HCV infection, we suggest etanercept rather than another TNF-alpha inhibitor (Grade 2C). (See 'Hepatitis C' above.)
•Herpes zoster – There is an increased risk of herpes zoster in patients with underlying autoimmune diseases, particularly among those receiving immunosuppressive therapies including glucocorticoids. However, in United States studies (unlike those in Europe), the risk of herpes zoster in patients receiving a TNF-alpha inhibitor does not appear to be higher than in patients receiving nonbiologic disease-modifying antirheumatic drugs (eg, methotrexate, leflunomide), and the risk appears similar to other biologics used in autoimmune disease [97]. (See 'Herpes zoster' above.)
•Human immunodeficiency virus – TNF-alpha inhibition use in patients infected with HIV appears to be safe, provided that the patient is on an effective antiretroviral regimen before starting a TNF-alpha inhibitor. (See 'Human immunodeficiency virus' above.)
●Fungal infections
•Endemic fungi – Case series of histoplasmosis, coccidioidomycosis, cryptococcosis, and other invasive fungal infections indicate an increased risk through either new infections or reactivation of latent ones, although the former is likely much more common. Clinicians caring for patients in regions that are endemic for invasive fungal infections must maintain a high degree of suspicion for the occurrence of such complications. (See 'Fungal infections' above.)
•Pneumocystis pneumonia – Cases of Pneumocystis pneumonia (PCP) have been reported among patients treated with TNF-alpha inhibitors. Given the relatively low level of risk of PCP associated with TNF-alpha inhibitors, prophylaxis for patients receiving a TNF-alpha inhibitor is not recommended unless patients have other indications for prophylaxis (eg, receipt of high-dose glucocorticoids). (See "Treatment and prevention of Pneumocystis pneumonia in patients without HIV", section on 'Prophylaxis'.)
●Perioperative use of TNF-alpha inhibitors – The use of TNF-alpha inhibitors has been associated with an increased risk of perioperative infections in some, but not all, studies. TNF-alpha inhibitors should generally be withheld for several weeks before orthopedic surgery (as close to one dosing cycle as scheduling allows) and can be resumed after evidence of wound healing, typically after 14 days. However, the potential benefit of preventing some postoperative infections must be balanced against the risk of a perioperative flare in rheumatoid arthritis activity. (See 'Perioperative use of TNF-alpha inhibitors' above.)
●Combination therapy with anakinra – The combination of a TNF-alpha inhibitor and anakinra (a recombinant human interleukin 1 receptor antagonist) is not recommended because of an increased risk of infection. (See 'Combination therapy with anakinra' above.)
●Vaccinations – Clinicians should ensure that patients are up to date on vaccinations before starting a course of TNF-alpha inhibition (figure 1). Live virus vaccines generally should not be administered to patients receiving therapy with a TNF-alpha inhibitor. Initiation of TNF-alpha inhibitor therapy should be delayed for at least four weeks after administration of a live-virus vaccine. (See "Immunizations in autoimmune inflammatory rheumatic disease in adults".)
ACKNOWLEDGMENTS — The UpToDate editorial staff acknowledges John Stone, MD, MPH, and Kieren A Marr, MD, who contributed to earlier versions of this topic review.
آیا می خواهید مدیلیب را به صفحه اصلی خود اضافه کنید؟
