Search tips
Search criteria 


Logo of gutGutVisit this articleSubmit a manuscriptReceive email alertsContact usBMJ
Gut. 2007 May; 56(5): 725–732.
PMCID: PMC1942157

Risks and benefits of biologic therapy for inflammatory bowel diseases

The advent of biological drugs has had a significant impact on the management of inflammatory bowel diseases. For an important group of patients with Crohn's disease and ulcerative colitis, treatment with biologics has led to an improved quality of life, fewer admissions to the hospital and fewer side effects from corticosteroids and/or immunomodulators. Nonetheless, biologic agents are still being used with reluctance. The most frequently cited reasons for this are the high cost and uncertainty about long‐term safety. The idea that biologic agents interfere with key molecules in the human physiology justifies these appropriate concerns. Indeed, many unexpected toxicity problems have occurred and been recognised as related to biologic therapies. A few examples include deterioration of heart failure and reactivation of mycobacterial infections with anti‐tumour necrosis factor (TNF) agents,1 the rare occurrence of lethal viral encephalitis with the anti‐integrin antibody natalizumab,2 and the dramatic cytokine release syndrome that was observed in healthy volunteers treated with a monoclonal antibody directed against CD28.3 Fortunately, stringent post‐marketing surveillance programmes and registries have allowed early recognition of most of these problems and appropriate measures and guidelines have been developed to prevent and treat them. Given the facts that most biologic agents are immunogenic and could therefore lose their effect with time, that the ideal position of biologic agents in the management of Crohn's disease and ulcerative colitis is still poorly defined, that most patients treated with biologics need induction and maintenance therapy and that the cost of these agents is likely to remain high, it is of paramount importance that the balance between efficacy and safety is carefully considered for every individual patient.

Problems in toxicity assessment of biologic therapy

Many patients who receive treatment with biologic agents have received other therapies or continue to use concomitant therapies to which the biologics are added (box 1). It is important to realise that not all toxicity problems that occur in these patients are related to biologic therapy, but can be caused by the concomitant therapies in inflammatory bowel disease (IBD), most often corticosteroids and/or immunomodulators. Pancreatitis in patients with IBD, who started on anti‐TNF therapy in combination with azathioprine, for instance, is most often related to the use of azathioprine and not to anti‐TNF agents. In addition, toxicity can sometimes be related to a combination of therapies and not to each of the individual agents. This is typically the case in patients with opportunistic infections or malignancies such as lymphomas.

Furthermore, the underlying disease can be associated with certain complications by itself. Typical examples are dysplasia and cancer in longstanding ulcerative colitis, lymphoma in patients with rheumatoid arthritis and intestinal strictures in patients with Crohn's disease. Since IBDs can be complicated by a large variety of extraintestinal manifestations, it could not always be easy to determine the possible association between these conditions and the biologic therapy received by patients. Particular examples are arthralgias that occur in active Crohn's disease, but that can also be symptomatic for a (delayed) hypersensitivity reaction to biologic therapy, or pustular skin eruptions that can represent a Sweets' syndrome or psoriasis or, on the other hand, be a complication of anti‐TNF therapy.

Finally, not all adverse events associated with drugs are reported to the manufacturer or to the drug vigilance organisations. During clinical trials and in the first years of marketing, the level of suspicion for side effects and the rate of reporting are generally higher than after several years of widespread use. This could lead to under‐reporting or no reporting at all, of uncommon side effects.

Sources of information on toxicity associated with biologic therapy

The most stringent reporting of possibly and probably related adverse events takes place during controlled randomised trials (table 11).). These trials are important with regard to safety aspects, since they include a control population of patients receiving placebo or no experimental drugs at all. The limitation of safety reporting in clinical trials is that the number of patients under study is rather limited, and that the inclusion and exclusion criteria for the trial filters the patient population into a selected group of individuals without serious concomitant conditions (such as a history of malignancy), renal or hepatic insufficiency, pregnancy, prior surgery and so on. A typical example is the increased incidence of tuberculosis after anti‐TNF therapy for Crohn's disease, an event that was not detected in the registration trials with infliximab.

Table thumbnail
Table 1 Adverse event reporting and safety in drug development

The second level of safety data collection is organised by post‐marketing surveillance. Typically, doctors tend to report severe and life‐threatening problems, but there is a tendency towards under‐reporting of less severe health problems such as skin eruptions, fatigue and general malaise. With time, the motivation to report adverse events often diminishes. Nonetheless, post‐marketing surveillance is important because of the size of the patient population under surveillance. Adverse events that occur three times more frequently during certain therapies require a follow‐up of 750 000 exposed and non‐exposed individuals for a whole year.4

Thirdly, registries are helpful to determine the true incidence of certain problems in a somewhat selected, but above all, carefully observed patient cohort. Registries typically have a comparable control population with an identical medical profile but not treated with the therapy under study. The Therapy Resource Evaluation and Assessment Tool (TREAT) registry for patients with Crohn's disease in the USA allowed the investigators to ascertain that sepsis was associated with the use of concomitant corticosteroids and not with infliximab.5 The valuable information that can be drawn from these types of databases has led to the recommendation by medication agencies such as the Food and Drug Administration (FDA) and the European Medicines Agency (EMEA) to set up registration studies with biologics in the first years of marketing.

Overview of the most common side effects caused by biologic therapies


Infections and septicaemia have undoubtedly been the most common cause of mortality associated with the therapy for IBD (box 2). The increased incidence of infections has long been recognised to be a potential complication of corticosteroids and immunomodulators, and biologic therapy appears to add to this elevated risk. Although common immunomodulatory therapy has predominantly been associated with viral infections such as cytomegalovirus and Epstein–Barr virus infections, biologic therapy is more frequently associated with mycobacterial infections (tuberculosis and histoplasmosis) and other opportunistic infections such as listeriosis, nocardiosis, invasive aspergillosis and others.1 A recent case–control study looked at 100 patients with IBD, with opportunistic infections, treated at the Mayo Clinics between 1998 and 2003.6 The occurrence of opportunistic infections was associated with the use of corticosteroids (odds ratio (OR) 3.4), azathioprine/6‐mercaptopurine (OR 3.1) and infliximab (OR 4.4) in the 3 months before infection. The OR for opportunistic infections in patients using two drugs rather than one rose from 2.6 to 12.9. The combination of infliximab with azathioprine/6‐mercaptopurine did not increase the risk significantly.

Box 1 Problems in toxicity assessment of biologics used in inflammatory bowel disease

  • Many patients receive multidrug treatment; it is not always clear to which agent toxicity can be attributed.
  • The underlying condition(s) can also be the cause of extraintestinal problems.
  • Rare adverse events are often not recognised during controlled trials and become apparent only after a period of widespread use.

The incidence of opportunistic infections in different groups of patients treated with infliximab was reported to vary between 0.3% and 0.9%.7 Serious infections were observed in 3–4% of patients treated with infliximab in clinical trials.8,9 In the trials, most patients having these serious infections received multiple concomitant therapies including immunosuppressives and corticosteroids. The TREAT registry for patients with Crohn's disease treated with infliximab revealed an increased risk of infections which was associated with the use of corticosteroids, and not with infliximab.5

The most common infections occurring in patients with IBD treated with anti‐TNF therapies are upper respiratory tract and urinary tract infections. It is now generally accepted that if a patient is having moderate‐to‐severe infection, it is advisable to postpone anti‐TNF therapy and to treat the infection properly first.

As already mentioned, an increased incidence of mycobacterial infections such as histoplasmosis and tuberculosis hominis has been associated with anti‐TNF therapies. Up to 2003, approximately 350 cases of tuberculosis were reported in 400 000 patients treated with infliximab.10 With the chimeric anti‐TNF agent, infliximab, 75% of documented cases presented within the first three infusions, 97% within the first six infusions and the vast majority of patients had a history of prior exposure to tuberculosis. As general awareness about this problem has increased in the medical community, and predominantly thanks to screening with chest x rays and purified protein derivative (PPD) skin tests in countries where bacillus Calmette‐Guerin (BCG) vaccination is not standard, the incidence of tuberculosis reactivation during anti‐TNF therapy has decreased dramatically (in Europe from 1.5 cases/1000 patients/6 months to 0.5 cases/1000 patients/6 months).10 Patients with signs of prior exposure to tuberculosis need to receive tuberculostatic therapy (isoniazide +/− rifampicine) during the first 3–6 months of anti‐TNF therapy, which is often started a few weeks before initiating treatment with the anti‐TNF agent (box 3). Patients with overt tuberculosis need triple anti‐tuberculosis treatment before anti‐TNF therapy can be started.

Even in patients with negative screening tests, tuberculosis can develop. Recently, a Japanese patient with rheumatoid arthritis who was receiving infliximab treatment was reported to have developed peritoneal tuberculosis. The infection was successfully treated and anti‐TNF therapy could be resumed shortly thereafter.11

The PPD skin test is prone to reader variability and its outcome is influenced by cross‐reactivity with environmental mycobacteria, previous BCG vaccination and anergy in immunosuppressed individuals. There is a recent trend to replace the skin test by the more reliable measurement of T‐cell‐based interferon (IFN) γ responses to mycobacterium tuberculosis, which displays a stronger association with exposure and is less biased. Further studies with this IFNγ‐gamma assay are going on, and will address the utility of the test in patients with IBD.12

Infections of the central nervous system are often difficult to diagnose. Cerebral toxoplasmosis was observed in 7 of 643 000 patients with IBD receiving infliximab therapy13; isolated reports have described cases of listeria meningitis/meningoencephalitis. Unfortunately, this diagnosis is often delayed and has led to considerable mortality.

Clear guidelines on the administration of vaccinations in patients with IBD receiving or scheduled to receive anti‐TNF agents are lacking. It is generally recommended that live vaccines should be avoided during anti‐TNF therapy. This could be relevant to travellers visiting Africa, where yellow fever is endemic. Paediatric patients with IBD should ideally receive all their vaccinations before starting anti‐TNF therapy (Remicade, Centocor, Pennsylvania, USA).

A particular infection, progressive multifocal leucoencephalopathy (PML) developed in 3 of 3819 patients treated with the monoclonal antibody, natalizumab, directed against the α4 integrin.2,14 PML is an opportunistic, infectious, demyelinating brain disorder associated with impaired T cell function and caused by a polyomavirus (JC virus, JCV). Of the three cases, two were fatal.

One of these infections was observed in a patient with Crohn's disease who presented with what was believed to be an atypical astrocytoma.2 In this patient, JCV DNA was detected (post hoc) in the serum while he was receiving monthly infusions of natalizumab, with a steep increase in viral load just before the onset of neurological symptoms. The viral DNA was also present in the resected brain tumour. Extensive re‐evaluation of all patients treated with natalizumab has since been carried out. No further cases of PML were discovered. At present there is no useful screening strategy available to screen for or to prevent JCV infection and the treatment is available for patients with multiple sclerosis under strict surveillance.15

Box 2 Toxicity associated with biologic therapies

  • Infections
  • Immunogenicity and loss of response
  • Injection and cytokine release reactions
  • Auto‐immunity
  • Malignancies
  • Liver function abnormalities
  • Heart failure
  • Demyelination
  • Skin eruptions

Also with natalizumab, an increased incidence of influenza infections was observed when compared to placebo (12% vs 5%), but in the majority of mild‐to‐moderate infections, the outcome of these infections was favourable.16


Virtually all monoclonal antibodies, even the so‐called humanised or fully human molecules, are immunogenic. It is now commonly accepted that antibody formation can be prevented by systematic administration schedules rather than episodic (on demand) therapy, by pretreatment with corticosteroids and by concomitant use of immunosuppressives (box 4). In addition, maintenance therapy following an induction regimen has been associated with better clinical outcome and fewer complications, albeit at a higher direct cost.17

An episodic or on demand treatment schedule led to antibody formation in >70% of patients treated with infliximab.18 The antibody titres correlated significantly with the occurrence of infusion reactions and loss of response to therapy. Also, the duration of response after an infliximab infusion was significantly reduced if an infusion reaction had occurred and antibody levels were high (cut‐off level 8 μg/ml). Suppression of antibody formation was significant with immunomodulators, without significant difference between patients cohorts treated with azathioprine/6‐MP or methotrexate.19

For how long concomitant immunosuppression needs to be continued has recently been studied in a randomised prospective trial where continued immunosuppression was compared to discontinuation in a cohort of patients on systematic eight‐weekly infliximab treatment for at least 6 months. The investigators did not find differences in the proportions of patients with allergic reactions or loss of response, although the infliximab serum levels tended to decrease in patients who discontinued immunosuppression.20

Box 3 Tuberculosis in patients receiving anti‐TNF therapy

  • All patients must be screened for prior exposure to tuberculosis. If an active disease is discovered, this needs to be treated before anti‐TNF can be initiated. If a latent disease is discovered, tuberculostatic agents need to be started before anti‐TNF therapy and continued for at least several months.
  • The vast majority of tuberculosis relapses occur during the first 6 months of anti‐TNF therapy.

Antibody formation was not only reported with infliximab but also with the fully human anti‐TNF antibody adalimumab (0.04% in the Classic‐1 trial at 4 weeks, 3.7% in the Classic‐2 trial and 5% in the rheumatoid arthritis trials). Circulating antibodies to adalimumab were higher in patients not using immunomodulators (in rheumatoid arthritis 12% in patients not on methotrexate, 1% in patients on methotrexate), but they were not associated with an increased incidence of adverse events.21,22 With certolizumab‐pegol, a pegylated Fab' fragment of a humanised anti‐TNF antibody, 8% of patients developed antibodies in the Precise‐2 trial (no Δ efficacy was observed).23 With fontolizumab, an antibody against IFN γ, 7 of 90 (8%) patients developed antibodies without clinical consequences.24

Antibody formation was also frequently observed with antibodies against adhesion molecules. A humanised monoclonal antibody directed against the α4β7 receptor was evaluated under the name MLN02 (Millennium Pharmaceuticals, London, UK) in 181 patients with active ulcerative colitis. It was notable that 44% of the patients treated with MLN02 developed antibodies against the drug, and those with higher antibody titres had response rates similar to placebo.25 In this trial most patients were not treated with immunomodulators. In the Evaluation of Natalizumab as Continuous Therapy (ENACT) trials with natalizumab for Crohn's disease, antibodies to natalizumab were measurable in 8–9% of patients.16 In the induction phase (ENACT‐1), infusion reactions occurred in 45% of patients if antibodies were present versus 9% if they were absent. In ENACT‐2, the maintenance phase, infusion reactions were observed in 19% of cases if antibodies were present compared to 7% if antibodies were absent. The persistent presence of antibodies was associated with loss of response.16

Patients who develop an infusion reaction should be managed properly and instantly with antihistamines, acetaminophen/paracetamol intravenously, and sometimes corticosteroids and even epinephrine. When the infusion is started again, it should be carried out at a very slow rate which can gradually be increased later on. Patients who have experienced an infusion reaction often benefit from a dose of hydrocortisone (100–500 mg) intravenously 1 h before further infusions. The same approach is often advocated in patients who are scheduled to receive an antibody therapy after a prolonged drug holiday (>4–6 months), since these patients are likely to have antibodies and, hence, are at risk for infusion reactions.

Injection and cytokine release reactions

Injection‐site reactions typically appear with subcutaneous preparations such as adalimumab (24–38% vs 16% with placebo in Classic‐1)21 and certolizumab pegol (overall incidence 5%).23,26 With adalimumab, the occurrence of injection‐site reactions increased with higher dose regimens and included burning, pain, erythema, bruising and pruritus.21 A more experimental biologic therapy for Crohn's disease, sargramostim (granulocyte‐macrophage colony‐stimulating factor, GM‐CSF), was associated with 90% injection‐site reactions (vs 12% with placebo).27 Anti‐interleukin (IL)12 antibody treatment was associated with injection‐site reactions in 77–88% of the patients.28

Box 4 Measures to prevent immunogenicity with biologics

  • Association of immunomodulatory drugs (azathioprine, 6‐mercaptopurine, methotrexate).
  • Hydrocortisone before infusions.
  • Systematic rather than episodic treatment.
  • Three‐infusion induction regimen for infliximab (week 0, 2 and 6).

Cytokine release syndromes (CRSs) typically occur within the first hours after an infusion/injection and can even be hyper‐acute with fever, hypotension, muscle aches and shock, or somewhat slower with arthralgias and general malaise (box 5). The phenomenon is most commonly related to the release of cytokines from activated and apoptotic cells into the blood circulation and subsides after 4–6 h. Serious cytokine release problems occurred in healthy volunteers during a phase 1 trial with an antibody against CD28.3 In IBD, the problem has frequently been observed with visilizumab, an anti‐CD3 antibody (directed against activated T cells). In a trial with this drug, most patients developed mild‐to‐moderate cytokine release syndrome (CRS), characterised by fatigue, nausea, chills, headache, arthralgia, fever, emesis, dehydration, dizziness and diaphoresis, which lasted typically 1–2 h after the first infusion. The symptoms were less pronounced at the second infusion on day 2 than on day 1.29

Delayed hypersensitivity reactions occur typically >3 days after drug administration. Typical symptoms are joint pain and muscle stiffness (with even dysphagia and inability to swallow). These reactions tend to be more common after longer drug holidays and are probably related to immune complex deposition in joints and muscles. The condition was first observed in 9 of 40 patients with Crohn's disease who had been treated successfully in clinical trials with infliximab and were re‐exposed to the drug after an interval of 3–5 years.30 Patients with severe delayed hypersensitivity reactions require treatment with corticosteroids, paracetamol and antihistamines, most often leading to complete resolution of symptoms after 1–2 weeks.


Biologics that induce cell lysis could lead to massive release of cellular debris including DNA in the circulation. This phenomenon triggers autoimmunity with the formation of antibodies against nuclear factors and DNA. In clinical trials, antinuclear antibodies (ANAs) have been measured in 46–56% of patients treated with infliximab compared to 18–35% with placebo. Cumulative ANA formation was studied in a series of 125 consecutive patients with Crohn's disease treated with infliximab.31 Almost half of these patients developed ANAs after the first infusion, and >75% became ANA‐positive after <3 infusions. Of the 43 ANA‐positive patients who were further subtyped, 14 of 43 (32.6%) had double‐stranded DNA, 17 (39.5%) had single‐stranded DNA, 9 (20.9%) had antihistone and 0% were extractable nuclear antigens (ENA)‐positive. Two patients (0.2%, double‐stranded DNA‐positive) developed drug‐induced lupus (rash, pleuritis) without major organ damage, and one developed autoimmune haemolytic anaemia. ANAs were associated with female sex and with papulosquamous or butterfly rash. The cumulative incidence of ANAs was 56.8% after 24 months. ANAs persisted for up to 1 year after the last infusion, and only a few patients became seronegative.

Box 5 Cytokine release syndrome

  • Typically, begins shortly after the infusion and lasts 4–6 h, maximally 24 h.
  • Symptoms consist of fever, chills, hypotension, shortness of breath, emesis, athralgia.
  • Treatments to be given are intravenous fluids, antihistamines, acetaminophen and sometimes corticosteroids.

Box 6 Risk–benefit assessment of biologic therapies in IBD

  • The risk–benefit of biologic therapy in IBD is greatly in favour of these powerful drugs.
  • Careful selection of the patients who need these therapies is required.
  • Reduction of toxicity is possible through awareness of specific drug‐related problems.
  • Monotherapy should be considered; combined regimens could be limited in duration.
  • Prolonged association with corticosteroids and deep immunosuppression should be avoided.

In most cases with overt infliximab‐induced lupus, treatment with systemic corticosteroids is indicated and infliximab is to be discontinued. It is unclear whether symptoms always recur when infliximab treatment is resumed. It is also unknown whether the presence of autoantibodies can lead to future autoimmune disease, even years later. For the time being, the presence of ANA without clinical symptoms should not affect therapeutic strategies.


The most important concerns with prolonged use of biologic agents are related to cancer development. It is believed that molecules such as TNF do play a role in the preclinical suppression of cancer development. Hence, it is warranted to be extremely cautious in this regard. Suspicion was first raised by the occurrence of two lymphomas in the early infliximab trials, and by several uncontrolled reports of intestinal adenocarcinomas in infliximab‐treated patients. It is of paramount importance, however, to realise that the genuine cancer risk induced by biologic agents cannot be ascertained within the frame of controlled clinical trials, given the relatively small size of the patient cohorts under study. Therefore, registries and post‐marketing surveillance programmes are particularly useful in studying cancer incidence.

In the TREAT registry, for instance (6273 patients, 3272 treated with infliximab), all types of cancer taken together were observed at an incidence of 0.58 per 100 patient‐years compared to 0.53 per 100 patient‐years in the control population. This brought the relative risk (RR) of cancer under infliximab treatment to 1.1, with 95% CI 0.71 to 1.63, meaning that the increased risk was not statistically significant. Looking at lymphoma in particular, the incidence was 0.06 per 100 patient‐years in patients treated with infliximab versus 0.05 per 100 patient‐years in control patients (RR 1.3, 95% CI 0.36 to 5.03), also not statistically significant.32

Is spite of these data (even with over 6000 patients, the sample size was still relatively small for rare adverse events), the development of lymphoma has been a nightmare for patients and doctors using biologic therapy, particularly since six patients with Crohn's disease were reported to have developed a hepatosplenic T‐cell lymphoma (HSTL) during treatment with infliximab and immunomodulators.33 These young patients (aged 12–31 years) had received 2–21 doses of infliximab. Five of them have died. Hepatosplenic T cell lymphoma is a subtype of non‐Hodgkin's lymphoma which most frequently develops in young adult males, both immunocompromised and with normal immunity (de novo). With only 120 cases reported in the literature, the true incidence of the condition is unknown. In most cases, however, the disease is very aggressive and often fatal. A literature search revealed three additional cases of HSTL in patients with Crohn's disease who were on azathioprine/6‐mercaptopurine (duration 4–6 years), who had never been exposed to infliximab.34,35,36

In addition, a recent metanalysis pooled data from nine selected placebo‐controlled trials with infliximab and adalimumab for the treatment of rheumatoid arthritis, comprising over 5000 patients. Of these, 3493 patients had received at least one dose of an anti‐TNF agent. In this cohort, 29 malignancies were discovered (10 lymphomas). In the control group, three malignancies were reported (no lymphomas). The pooled OR for malignancy in this analysis was 3.3 (number needed to harm = 154) and cancer developed more frequently with higher doses.37 With rheumatoid arthritis, however, an elevated risk of lymphoma has been associated with the disease itself even without immunomodulatory treatment and this is probably not the case for Crohn's disease.38

The issue was also examined in a recently published study by Siegel and colleagues,39 in which a decision analytic model was constructed to determine the risk of infliximab when compared with standard therapy. In a hypothetical cohort of 100 000 patients with Crohn's disease receiving infliximab compared to standard therapy, 201 more lymphomas would develop in the former group after 1 year.

Taking the available data together, it is fair to state that the increased incidence of lymphoma in patients treated with anti‐TNF agents is most likely a true phenomenon which must be taken seriously. Therefore, biologics ought to be reserved for patients who really need it. As most malignancies occur under combined therapy with biologics and immunosuppression, single‐drug strategies need to be revisited. At this point, however, it remains unclear whether it is the best strategy to discontinue immunosuppressors after a certain period of combined treatment (eg, 6 months), or rather to stop the biologic therapy and try to treat the patient with immunosuppression alone (the bridging concept).

A few studies have recently addressed the increased incidence of abnormal pap smears, suggestive of cervical dysplasia, in women with IBD. Although this problem seems to be associated with the use of immunosuppressive therapy, it is unclear if the risk is even higher in patients receiving infliximab therapy.40

Other adverse events

Neurological disorders caused by demyelinisation have been reported several times in patients treated with anti‐TNF agents. In the ACT trials with infliximab for ulcerative colitis, one patient developed optic neuritis in the 5 mg/kg dose group and one patient developed multifocal motor neuropathy in the 10 mg/kg treatment group by the 30th week.41 In a cohort of 500 patients treated with infliximab at the Mayo Clinics, one patient developed signs of demyelinisation.42 In the recent top‐down trial with infliximab and azathioprine in patients with newly diagnosed Crohn's disease, one patient developed optic neuritis after treatment with infliximab. Symptoms abated completely after cessation of the infusions.43 It is recommended that neurological signs and symptoms, including disturbed or diminished vision, be taken seriously in patients receiving anti‐TNF agents. Magnetic resonance scanning can reveal signs of demyelinisation of the central nervous system in case of doubt. Discontinuation of anti‐TNF therapy is mandatory in case of persistent or progressive neurological symptoms.

Neurological problems have also been associated with natalizumab treatment, as mentioned above. PML was found to be induced by the infection of the JCV and can be lethal if not recognised in time.

Serious liver‐function abnormalities can be observed in association with infliximab therapy.44 Two cases of hepatic failure have been reported. Chronic viral hepatitis (B and C) needs to be excluded in patients who are considered for anti‐TNF therapy, since a few cases of viral hepatitis exacerbation have been observed following cessation of infliximab therapy.45 Antiviral therapy (eg, with lamivudine for hepatitis B) could prevent this problem. It is now recommended that patients be screened for viral hepatitis and that liver function tests are monitored during treatment with biologic agents.

Skin lesions are a common phenomenon during anti‐TNF therapy. The problem has been relatively under‐reported as most lesions are limited, and a causal relationship with the treatment is often unclear. A lot of different types of eruptions can develop such as leucocytoclastic vasculitis, lichenoid drug reaction, perniosis‐like eruption, superficial granuloma annulare and acute folliculitis, which were observed in a cohort of patients with Crohn's disease and rheumatoid arthritis receiving infliximab.46 Although anti‐TNF therapies are usually effective in treating psoriasis,47 all TNFα‐blocking agents have been reported to lead to or exacerbate psoriasis.48 In some cases, skin changes were severe enough to discontinue the medication. In addition, many patients develop transient or mild skin eruptions that disappear spontaneously or with topical therapy alone (fig 11).

figure gt103564.f1
Figure 1 Typical examples of skin toxicity induced by anti‐TNF agents. The pictures on the left represent psoriasis‐like skin lesions, and the pictures on the right show acneiform skin eruptions in the face. Informed consent was ...

A final uncommon complication with anti‐TNF therapies is cardiac failure in patients who already have a diminished cardiac function at the start. The problem was recognised during a controlled clinical trial with infliximab for cardiac failure, in which patients receiving active therapy were found to have an elevated risk of mortality.49 It is now recommended that patients with functional class III or IV heart failure (New York Heart Association) should not receive anti‐TNF therapy.

Pregnancy during biologic therapy

Several reports have now been published documenting the relative safety of infliximab during pregnancy. In a cohort of 96 pregnancies where infliximab was given around the time of conception, 67% live births, 15% miscarriages and 19% therapeutic terminations were observed. These numbers are comparable with historical controls without infliximab therapy.50 Only a small number of patients intentionally received infliximab during pregnancy because of intractable Crohn's disease. One series reported on 10 women, eight of whom received maintenance infliximab infusions throughout their pregnancy. Two women received their first infliximab infusions during pregnancy. All 10 pregnancies ended in live births and none of the infants had congenital malformations, intrauterine growth retardation or were small for gestational age. Three infants were premature and one had low‐birth weight.51 In animal studies, no harm to the fetus has ever been observed. But in current practice, it is still recommended that infliximab treatment be avoided, on the basis of limited toxicity data, but in patients who really need the treatment it is probably safe to use it. The safety of other biologic therapies during pregnancy is unknown.

Risk–benefit analysis of biologic therapy for IBD

Side effects with biologic therapy for IBDs are common. Fortunately, most of these events are mild and transient, such as mild‐to‐moderate infections, skin eruptions or infusion/injection reactions. Clinicians need to remain cautious, however, to detect more severe problems such as opportunistic infections, cancer or lymphoma. In particular, patients who use biologic therapies in combination with classic immunomodulators appear to carry an elevated risk for these problems (box 6). Nonetheless, most experts agree that biologic therapies offer such important clinical benefits to patients with severe IBD, that their widespread use is definitely warranted.

The risk–benefit assessment was recently addressed in a model looking specifically at lymphoma formation and mortality. In the model comprising 100 000 patients an elevated risk of death (249 extra cases) and lymphoma (201 extra cases) was observed. However, 12 216 more patients were in remission and 4255 fewer surgeries needed to be performed in patients treated with infliximab. The authors concluded that the benefits of infliximab outweigh the risks in properly selected patients.39

It is clear that certain measures need to be taken into account when biologics are started. Toxicity can be significantly reduced by routine tuberculosis screening, by avoiding anti‐TNF agents in patients with heart failure and chronic infections, by careful timing of combination therapy with immunosuppressives and later switching back to single‐agent therapy, by exploring neurological symptoms whenever they develop and by timely discontinuation of treatment and so on. A firm recommendation is that doctors need to see and examine their patients (including regular blood checks) every 8–12 weeks.

General conclusions

Biologics have improved the quality of life of a significant proportion of patients with IBD. These agents are associated with certain toxicities and should therefore only be used in patients who really need the treatment. Physicians administering these drugs should be aware of the toxicity profile and of the appropriate guidelines to prevent and treat complications. They need to see and examine patients receiving biologic therapy on a regular basis. After all, it is not any more justifiable to withhold these potent therapies in patients with moderate and severe forms of IBD.


Competing of interests: Dr D'Haens has acted as a consultant, investigator and speaker for Centocor, Schering‐Plough, Abbott, UCB, Elan and PDL.

Informed consent was obtained for the publication of fig 1


1. Blonski W, Lichtenstein G R. Complications of biological therapy for inflammatory bowel disease. Curr Opinion Gastroenterol 2006. 2230–43.43
2. Van Assche G, Van Ranst M, Sciot R. et al Progressive multifocal leukoencephalopathy after natalizumab therapy for Crohn's disease. N Engl J Med 2005. 353362–368.368 [PubMed]
3. Suntharalingam G, Perry M R, Ward S. et al Cytokine storm in a phase 1 trial of the anti‐CD28 monoclonal antibody TGN1412. N Engl J Med 2006. 3551018–1028.1028 [PubMed]
4. Sackett D L. Bias in analytical research. J Chronic Dis 1979. 3251–63.63 [PubMed]
5. Lichtenstein G R, Feagan B G, Cohen R D. et al Serious infections and mortality in association with therapies for Crohn's disease: TREAT registry. Clin Gastroenterol Hepatol 2006. 4621–630.630 [PubMed]
6. Toruner M, Loftus E V, Colombel J F. et al Risk factors for opportunistic infections in inflammatory bowel diseases: a case–control study. Gastroenterology 2006. 130A71
7. Sandborn W J, Faubion W A. Biologics in inflammatory bowel disease: how much progress have we made ? Gut 2004. 531366–1373.1373 [PMC free article] [PubMed]
8. Hanauer S B, Feagan B G, Lichtenstein G R. et al Maintenance infliximab for Crohn's disease: the ACCENT I randomised trial. Lancet 2002. 3591541–1549.1549 [PubMed]
9. Sands B E, Anderson F H, Bernstein C N. et al Infliximab maintenance therapy for fistulizing Crohn's disease. N Engl J Med 2004. 350876–885.885 [PubMed]
10. Rutgeerts P, van Assche G, Vermeire S. Optimizing anti‐TNF treatment in inflammatory bowel disease. Gastroenterology 2004. 1261593–1610.1610 [PubMed]
11. Matsumoto T, Tanaka T, Kawase I. Infliximab for rheumatoid arthritis in a patient with tuberculosis. N Engl J Med 2006. 355740–741.741 [PubMed]
12. Dheda K, Udwadia Z F, Huggett J F. et al Utility of the antigen‐specific interferon‐gamma assay for the management of tuberculosis. Curr Opin Pulm Med 2005. 11195–202.202 [PubMed]
13. Young J D, McGwire B S. Infliximab and reactivation of cerebral toxoplasmosis. N Engl J Med 2005. 3531530–1531.1531 [PubMed]
14. Langer‐Gould A, Atlas S W, Green A J. et al Progressive multifocal leukoencephalopathy in a patient treated with natalizumab. N Engl J Med 2005. 353375–381.381 [PubMed]
15. Yousry T A. et al Evaluation of patients treated with natalizumab for progressive multifocal leukoencephalopathy. N Engl J Med 2006. 354924–933.933 [PMC free article] [PubMed]
16. Sandborn W J, Colombel J F, Enns R. et al Natalizumab induction and maintenance therapy for Crohn's disease. N Engl J Med 2005. 3531912–1925.1925 [PubMed]
17. Rutgeerts P, Diamond R H, Bala M. et al Comparison of scheduled and episodic treatment strategies of infliximab in Crohn's disease. Gastroenterology 2004. 126402–413.413 [PubMed]
18. Baert F, Noman M, Vermeire S. et al Influence of immunogenicity on the long‐term efficacy of infliximab in Crohn's disease. N Engl J Med 2003. 348601–608.608 [PubMed]
19. Noman M, Vermeire S, Van Assche G. et al The effectiveness of immunosuppression to suppress the formation of antibodies to infliximab in Crohn's disease. Gastroenterology 2004. 126(Suppl)A54–A55.A55
20. Van Assche G, Paintaud G, D'Haens G. et al Continuation of immunosuppression is not required to maintain adequate infliximab efficacy in patients with Crohn's disease but may improve pharmacokinetics. Gastroenterology 2006. 130(Suppl)A142
21. Hanauer S B, Sandborn W J, Rutgeerts P. et al Human anti‐tumor necrosis factor monoclonal antibody (adalimumab) in Crohn's disease: the CLASSIC‐I trial. Gastroenterology 2006. 130323–333.333 [PubMed]
22. Weinblatt M E, Keystone E C, Furst D E. et al Efficacy and safety of the fully human anti‐tumour necrosis factor alpha monoclonal antibody for the treatment of rheumatoid arthritis in patients taking concomitant methotrexate: the ARMADA trial. Arthritis Rheum 2003. 4835–45.45 [PubMed]
23. Schreiber S, Khaliq‐Kareemi M, Lawrance I. et al Certolizumab pegol, a humanized anti‐TNF pegylated Fab' fragment, is safe and effective in the maintenance of response and remission following induction in active Crohn's disease: a phase III study (precise). Gut 2005. 37(Suppl)A82
24. Hommes D W, Mikhajlova T L, Stoinov S. et al Fontolizumab, a humanised anti‐interferon‐gamma antibody, demonstrates safety and potential clinical activity in patients with moderate‐to‐severe Crohn's disease. Gut 2006. 551131–1137.1137 [PMC free article] [PubMed]
25. Feagan B G, Greenberg G R, Wild G. et al Treatment of ulcerative colitis with a humanized antibody to the α4β7 integrin. N Engl J Med 2005. 3522499–2507.2507 [PubMed]
26. Sandborn W, Feagan B, Stoinov S. et al Certolizumab Pegol administered subcutaneously is effective and well tolerated in patients with active Crohn's disease: results from a 26‐week, placebo‐controlled phase III study (precise 1). Gastroenterology 2006. 130(Suppl)A107
27. Korzenik J R, Dieckgraefe B K, Valentine J F. et al Sargramostim for active Crohn's disease. N Engl J Med 2005. 3522193–2201.2201 [PubMed]
28. Mannon P J, Fuss I J, Mayer L. et al Anti‐interleukin‐12 antibody for active Crohn's disease. N Engl J Med 2004. 3512069–2079.2079 [PubMed]
29. Targan S, Salzberg B, Mayer L. et al A phase 1 study: multiple dose levels of visilizumab are well tolerated and produce rapid and sustained improvement in ulcerative colitis patients refractory to treatment with intravenous steroids. Gastroenterology 2005. 128(Suppl)A75
30. Hanauer S, Rutgeerts P, D'Haens G. et al Delayed hypersensitivity to infliximab (Remicade®) re‐infusion after a 2–4 year interval without treatment. Gastroenterology 1999. 116A731
31. Vermeire S, Noman M, Van Assche G. et al Autoimmunity associated with anti‐tumor necrosis factor alpha treatment in Crohn's disease: a prospective cohort study. Gastroenterology 2003. 12532–39.39 [PubMed]
32. Lichtenstein G R, Feagan B G, Cohen R D. et al Serious infections and mortality in association with therapies for Crohn's disease: TREAT registry. Clin Gastroenterol Hepatol 2006. 4621–630.630 [PubMed]
33. Centocor Hepatosplenic T‐cell lymphoma in patients treated with Remicade® (infliximab) Press release May 16 2006
34. Lémann M, Gérard de la Vallusière F, Bouhnik Y. et al Intravenous cyclosporine for refractory attacks of Crohn's disease : long‐term follow‐up of patients. Gastroenterology 1998. 114G4178
35. Mittal S, Milner B, Johnston P. et al A case of hepatosplenic γ–δ T‐cell lymphoma with a transient response to fludarabine and alemtuzumab. Eur J Haematol 2006. 76531–534.534 [PubMed]
36. Navarro J, Ribera J, Mate J. et al Hepatosplenic T‐gammadelta lymphoma in a patient with Crohn's disease treated with azathioprine. Leuk Lymphoma 2003. 44531–533.533 [PubMed]
37. Bongartz T, Sutton A J, Sweeting M J. et al Anti‐TNF antibody therapy in rheumatoid arthritis and the risk of serious infections and malignancies: systematic review and meta‐analysis of rare harmful effects in randomized controlled trials. JAMA 2006. 2952275–2285.2285 [PubMed]
38. Lewis J D, Bilker W B, Brensinger C. et al Inflammatory bowel disease is not associated with an increased risk of lymphoma. Gastroenterology 2001. 1211080–1087.1087 [PubMed]
39. Siegel C A, Hur C, Korzenik J. et al Risks and benefits of infliximab for the treatment of Crohn's disease. Clin Gastroenterol Hepatol 2006. 41017–1024.1024 [PubMed]
40. Kane S V, Khatibi B, Reddy D. Use of immunosuppressant results in higher incidence of abnormal PAP smears in women with inflammatory bowel disease. Gastroenterology 2006. 130(Suppl)A2
41. Rutgeerts P, Sandborn W J, Feagan B G. et al Infliximab for induction and maintenance therapy for ulcerative colitis. N Engl J Med 2005. 3532462–2476.2476 [PubMed]
42. Colombel J F, Loftus E V, Tremaine W J. et al The safety profile of infliximab in patients with Crohn's disease: the Mayo clinic experience in 500 patients. Gastroenterology 2004. 12619–31.31 [PubMed]
43. Hommes D, Baert F, Van Assche G. et al The ideal management of Crohn's disease: top down versus step up strategies, a randomized controlled trial. Gastroenterology 2006. 130(Suppl)A108
44. Tobon G J, Canas C, Jaller J J. et al Serious liver disease induced by infliximab. Clin Rheumatol. 2006[Epub ahead of print].
45. Esteve M, Saro C, Gonzalez‐Huix F. et al Chronic hepatitis B reactivation following infliximab therapy in Crohn's disease patients: need for primary prophylaxis. Gut 2004. 531363–1365.1365 [PMC free article] [PubMed]
46. Devos S A, Van Den Bossche N, De Vos M. et al Adverse skin reactions to anti‐TNF‐alpha monoclonal antibody therapy. Dermatology 2003. 206388–390.390 [PubMed]
47. Kleyn C E, Griffiths C E. Infliximab for the treatment of psoriasis. Expert Opin Biol Ther 2006. 6797–805.805 [PubMed]
48. Sari I, Akar S, Birlik M. et al Anti‐tumor necrosis factor‐alpha‐induced psoriasis. J Rheumatol 2006. 331411–1414.1414 [PubMed]
49. Chung E S, Packer M, Lo K H. et al Anti‐TNF therapy against congestive heart failure investigators. Randomized, double‐blind, placebo‐controlled, pilot trial of infliximab, a chimeric monoclonal antibody to tumor necrosis factor‐alpha, in patients with moderate‐to‐severe heart failure: results of the anti‐TNF therapy against congestive heart failure (ATTACH) trial, Circulation 2003. 1073133–3140.3140 [PubMed]
50. Katz J A, Antoni C, Keenan G F. et al Outcome of pregnancy in women receiving infliximab for the treatment of Crohn's disease and rheumatoid arthritis. Am J Gastroenterol 2004. 992285–2292.2292
51. Mahadevan U, Kane S, Sandborn W J. et al Intentional infliximab use during pregnancy for induction or maintenance of remission in Crohn's disease. Aliment Pharmacol Ther 2005. 21733–738.738 [PubMed]

Articles from Gut are provided here courtesy of BMJ Publishing Group