Managing risks of TNF inhibitors: An update for the internist
ABSTRACTTumor necrosis factor (TNF) inhibitors have many beneficial effects, but they also pose infrequent but significant risks, including serious infection and malignancy. These risks can be minimized by judicious patient selection, appropriate screening, careful monitoring during treatment, and close communication between primary care physicians and subspecialists.
KEY POINTS
- Over the past 10 years, TNF inhibitors have substantially altered the management of autoimmune diseases such as rheumatoid arthritis and inflammatory bowel disease.
- Safety concerns include risks of infection, reactivation of latent infection (eg, fungal infection, granulomatous infection), malignancy, and autoimmune and neurologic effects.
- Before treating, take a complete history, including exposure to latent infections and geographic considerations, and bring patients’ immunizations up to date.
- Regular clinical and laboratory monitoring during treatment helps optimize therapy and minimize the risk of adverse effects.
- Physicians must be aware of atypical presentations of infection and understand how their treatment may differ in patients on biologic therapy.
DEMYELINATING DISEASE, INCLUDING MULTIPLE SCLEROSIS
Anti-TNF agents have been associated with the onset or exacerbation of clinical symptoms and radiographic evidence of central nervous system demyelinating disorders, including multiple sclerosis.37–40 Mohan et al38 identified 19 cases of demyelinating events occurring after administration of anti-TNF agents in early 2001. In most cases, symptoms improved or resolved after therapy was stopped.
Optic neuritis,41,42 bilateral optic neuropathy,43 and aseptic meningitis44 have also been reported, but these have occurred only rarely.
How common are these effects? Postmarketing surveillance in patients with rheumatoid arthritis yields an estimated incidence of demyelinating disorders of 1 per 1,000 patient-years with adalimumab therapy.45 Complicating the assessment is an observed slight increase in risk of demyelinating conditions associated with inflammatory bowel disease.
Symptoms that should heighten the physician’s suspicion of this adverse effect include confusion, paresthesias, and ataxia. Patients on anti-TNF therapy who develop new visual symptoms should be checked for painless visual loss as a sign of early demyelinating disease.
Although data that conclusively link anti-TNF agents to multiple sclerosis are lacking, these drugs should not be initiated in patients who have a history of demyelinating disease, and treatment should be stopped promptly if the diagnosis is suspected.
MALIGNANCY
Whether anti-TNF therapy is directly linked to development of malignancies is difficult to determine. There are many confounding factors, including the risk of malignancy in underlying inflammatory disease and the concomitant use of other medications such as thiopurines, which have a known association with lymphoma.46
The incidence of lymphoma is twice as high in rheumatoid arthritis patients as in the general population.47 The risk is higher in those with more aggressive joint disease—the subset of rheumatoid arthritis patients who are more likely to be given anti-TNF agents.5
In patients with inflammatory bowel disease, the risk of cholangiocarcinoma is four times higher, and the risk of small-bowel adenocarcinoma is 16 to 18 times higher, but no increased risk of lymphoma has been identified in this population.48,49
Data from six clinical trials of infliximab, including a long-term study of its safety in Crohn disease, suggest it poses no increase in overall risk of malignancy.50–52 Similar results have been reported in patients with other rheumatic diseases.8 Information on this topic is constantly evolving, and studies range from case series to clinical trials to large patient registries.
The decision to use a TNF inhibitor should be based on the patient’s clinical picture and risk factors. Discussion of the risks and benefits of therapy with the patient should be clearly documented.
Non-Hodgkin lymphoma
Evidence about the risk of lymphoma with anti-TNF use is mixed, as up to two-thirds of patients on anti-TNF therapy have received concomitant nonbiologic immunosuppressive medications, making it difficult to determine the true risk from the biologic agents alone.53 Current evidence both supports9,53–56 and refutes7,55,57–60 the idea that anti-TNF agents increase lymphoma risk.
In patients with inflammatory bowel disease, several population-based studies have not shown a clear increase in lymphoma risk with anti-TNF use.56,59,60 Pedersen et al,61 in a meta-analysis of eight studies, confirmed these findings by showing no overall lymphoma risk in patients with inflammatory bowel disease.
However, a Canadian population-based study found a statistically significant increase in non-Hodgkin lymphoma in males with Crohn disease, with an incidence ratio of 3.63 (95% confidence interval [CI] 1.53–8.62).61 Additionally, Siegel et al62 found a significantly higher risk (6.1 cases per 10,000 patients) in patients treated with anti-TNF agents and thiopurines than in the general population (1.9 cases per 10,000 people). Although the difference was statistically significant, the overall risk is still very low.
Patients with rheumatoid arthritis seem to have a risk of lymphoma two to three times higher than in the general population. However, large population-based studies have not shown a statistically significant increase in the risk of lymphoma with anti-TNF therapy.63
Hepatosplenic T-cell lymphoma is a rare subtype of peripheral T-cell non-Hodgkin lymphoma; 25 cases have been reported in patients receiving anti-TNF therapy.64 Although the risk is extremely low (< 0.05%), physicians must carefully consider the risks and benefits of combination therapy, especially in young male patients with inflammatory bowel disease, since death is the usual outcome of this disease.65–67
Skin cancers
Wolfe and Michaud8 evaluated malignancy risk in rheumatoid arthritis patients being treated with biologic agents, including TNF inhibitors, using a large longitudinal database. These data were compared with those of the US Surveillance, Epidemiology, and End-Results (SEER) national database. No increase in the overall cancer rate was seen in rheumatoid arthritis patients (standardized incidence ratio [SIR] 1.0, 95% CI 1.0–1.1).
However, melanoma was more common in rheumatoid arthritis patients compared with SEER rates (SIR 1.7, 95% CI 1.3–2.3).8 In addition, biologic therapy was associated with a higher (but not statistically significant) risk of melanoma (odds ratio [OR] 2.3, 95% CI 0.9–5.4) and a higher risk of nonmelanoma skin cancer (OR 1.5, 95% CI 1.2–1.8), but not of other types of cancer.8
INFECTION
Patients on anti-TNF therapy are at a higher risk of infection, ranging from minor to life-threatening bacterial infections, and including the reactivation of granulomatous and fungal infections. More importantly, these agents are similar to steroids in blunting signs of infection, which may delay diagnosis and treatment.
The management of infection in patients on anti-TNF medications varies from case to case. In general, patients with a minor infection that does not require hospitalization or intravenous antibiotics can continue the biologic therapy while taking oral antibiotics. TNF inhibitors must be held in the event of a major infection.
Consultation with an infectious disease specialist is recommended, especially in complex cases.
Bacterial infections
An increased risk of minor bacterial infections such as urinary tract and respiratory infections has been well documented in several randomized control trials of anti-TNF agents, though other studies have shown no such increase in risk.33,51–59
The threshold for using antibiotics for a suspected bacterial infection is somewhat shifted in favor of treatment in patients on anti-TNF therapy. The reason is twofold: as previously noted, infections may be worse than they appear, because anti-TNF drugs can mask the signs and symptoms of a serious infection, and in patients on these drugs, an untreated bacterial infection may rapidly become life-threatening.
In general, broad-spectrum antibiotics are not warranted unless the source of infection is unclear or the patient is in danger of hemodynamic compromise.
Opportunistic infections
The association of anti-TNF agents with opportunistic infections could be viewed as an extension of their normal and intended therapeutic activity as potent immunosuppressive agents.68 Rheumatoid arthritis and inflammatory bowel disease are usually associated with conditions and situations that predispose patients to opportunistic infections, such as decreased immune response, malnutrition or malabsorption, surgeries, and concomitant immunosuppressive medications.7 Combination therapy with other immunosuppressive drugs and older age appear to markedly increase the risk of opportunistic infections, including mycobacterial and fungal infections, in patients with inflammatory bowel disease.7
Overall, opportunistic infections represent a measurable risk of anti-TNF therapy, and awareness and vigilance are important, especially in areas where opportunistic infections such as histoplasmosis and coccidiomycosis are endemic.50 Furthermore, physicians must be aware of the higher risk of opportunistic infections when multiple immunosuppressive drugs are used concurrently.
