BSR and BHPR rheumatoid arthritis guidelines on safety of anti-TNF ...

RHEUMATOLOGY
Guidelines
doi:10.1093/rheumatology/keq249b
<strong>BSR</strong> <strong>and</strong> <strong>BHPR</strong> <strong>rheumatoid</strong> <strong>arthritis</strong> <strong>guidelines</strong> on
safety of anti-TNF therapies
Tina Ding 1 , Jo Ledingham 2 , Raashid Luqmani 3 , Sarah Westlake 4 , Kimme Hyrich 5 ,
Mark Lunt 5 , Patrick Kiely 6 , Marwan Bukhari 7 , Rikki Abernethy 8 , Ailsa Bosworth 9 ,
Andrew Ostor 10 , Kate Gadsby 1 , Frank McKenna 11 , Diana Finney 12 , Josh Dixey 13
<strong>and</strong> Chris Deighton 1 on behalf of the St<strong>and</strong>ards, Audit <strong>and</strong> Guidelines Working
Group of <strong>BSR</strong> Clinical Affairs Committee <strong>and</strong> <strong>BHPR</strong>
Key words: Rheumatoid <strong>arthritis</strong>, British Society for Rheumatology, Guidelines, Anti-TNF therapies, Drug safety,
Adverse event, Infliximab, Etanercept, Adalimumab.
Introduction <strong>and</strong> objectives
Why do we need to update the RA biologics
<strong>guidelines</strong> on the safety of anti-TNF therapies?
Inhibitors of TNF-a represent important treatment advances
for a number of inflammatory conditions, including
RA. TNF-a inhibitors offer a targeted strategy that contrasts
with the non-specific immunosuppressive agents
traditionally used to treat most forms of systemic inflammation.
Information on who benefits from these agents
<strong>and</strong> on their adverse effects continues to be collected
1 Rheumatology Department, Royal Derby Hospital, Derby,
2 Rheumatology Unit, Queen Alex<strong>and</strong>ra Hospital, Portsmouth,
3 Rheumatology Department, Nuffield Orthopaedic Centre, Oxford,
4 Rheumatology Department, Poole Hospital, Poole, 5 ARC
Epidemiology Unit, University of Manchester, Manchester,
6 Rheumatology Department, St George’s Healthcare, London,
7 Rheumatology Department, Royal Lancaster Infirmary, Lancaster,
8 Rheumatology Department, St Helens Hospital, St Helens, 9 National
Rheumatoid Arthritis Society, Maidenhead, 10 Rheumatology
Department, Addenbrooke’s Hospital, Cambridge, 11 Rheumatology
Department, Trafford General Hospital, Manchester, 12 Rheumatology
Unit, Worthing <strong>and</strong> Southl<strong>and</strong>s NHS Trust, Worthing <strong>and</strong>
13 Rheumatology Department, Robert Jones <strong>and</strong> Agnes Hunt
Orthopaedic Hospital, Oswestry, UK.
Submitted 19 May 2010; revised version accepted 29 June 2010.
Correspondence to: Tina Ding, Rheumatology Department,
Royal Derby Hospital, Derby DE22 3NE, UK.
E-mail: tina.ding@doctors.net.uk
through clinical studies, case series <strong>and</strong> reports <strong>and</strong>
through national registries.
In 2001 <strong>and</strong> 2005, the British Society for Rheumatology
(<strong>BSR</strong>) established <strong>and</strong> updated <strong>guidelines</strong> for the use of
anti-TNF drugs in RA [1, 2]. These <strong>guidelines</strong> have indicated
which adult patients with RA should be eligible for
treatment with anti-TNF therapies, precautions that need
to be taken in their use <strong>and</strong> action that should be taken in
the event of adverse effects. The previous <strong>guidelines</strong>
applied to the then-available anti-TNF therapies {etanercept
<strong>and</strong> infliximab in 2001 [1], <strong>and</strong> etanercept, infliximab
<strong>and</strong> adalimumab (first-generation anti-TNF agents) in 2005
[2]}. Due to the large volume of information now available
on these agents the <strong>BSR</strong> has, in 2010, produced separate
<strong>guidelines</strong> on eligibility for anti-TNF treatment in RA
(in press) These current <strong>guidelines</strong> cover the safety aspects
of anti-TNF treatment in RA <strong>and</strong> apply to the
first-generation products but also to the newly licensed
second-generation anti-TNF drugs, certolizumab pegol
<strong>and</strong> golimumab. There are relatively little safety data specifically
for these second-generation agents but there are
no data thus far to suggest that their side-effect profile
would differ significantly from the first-generation agents.
This is a rapidly changing field with new data emerging
each month, so it is vital that clinicians keep up to date
with this area of practice. These <strong>guidelines</strong> have
GUIDELINES
! The Author 2010. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org 1

Tina Ding et al.
incorporated information that was available to the authors
at the time of their completion (November 2009).
What are the objectives of these <strong>guidelines</strong>?
The scope for the <strong>guidelines</strong> was agreed by the <strong>BSR</strong> RA
Biologics Guidelines Group (<strong>BSR</strong>BG) in 2007, with a view
to producing recommendations on the safety aspects <strong>and</strong>
the appropriate use of anti-TNF drugs in RA.
As a principle, the group also agreed to review other
available <strong>guidelines</strong> (previous <strong>BSR</strong> [2], EULAR [3], ACR
[4] <strong>and</strong> other international <strong>guidelines</strong>), <strong>and</strong> endorse these
where appropriate. The group was mindful of the need not
to reproduce the considerable work of other evidencebased
<strong>guidelines</strong>, <strong>and</strong> to concentrate on those areas
that were most relevant to patients <strong>and</strong> health-care professionals
in clinical practice in the UK.
Who is the target audience for these <strong>guidelines</strong>?
These <strong>guidelines</strong> are directed at all UK health-care professionals
who are involved in assessing patients for
anti-TNF therapies <strong>and</strong> responsible for the monitoring of
these drugs for efficacy <strong>and</strong> toxicity. They may also provide
a model for other countries in their approach to
managing anti-TNF safety issues.
Clinical situations covered by these <strong>guidelines</strong>
These <strong>guidelines</strong> provide a best-practice approach to the
evidence-based use of anti-TNF therapies in RA.
What are the areas that the present <strong>guidelines</strong> do
not cover?
These <strong>guidelines</strong> do not cover the use of anti-TNF drugs in
other inflammatory arthropathies or diseases. They do not
cover anti-TNF drugs that are still at an experimental
stage of development.
How have the patients’ views been incorporated
into the <strong>guidelines</strong>?
The National Rheumatoid Arthritis Society provided patient
representation on the <strong>BSR</strong>BG. Final drafts of the
<strong>guidelines</strong> have been circulated to patient representative
bodies for comment, <strong>and</strong> feedback has been incorporated
where appropriate. Drafts of the <strong>guidelines</strong> were also circulated
at the <strong>BSR</strong> Annual General Meeting in Glasgow in
April 2009, <strong>and</strong> feedback considered, <strong>and</strong> incorporated
where appropriate.
What is the evidence to support these <strong>guidelines</strong>?
The <strong>guidelines</strong> are referenced throughout, <strong>and</strong> where new
research has been performed to inform the <strong>guidelines</strong>
this is included in detail in the text. Comprehensive literature
searches were performed by T.D. <strong>and</strong> J.M.L., using
relevant search terms to seek evidence to determine what
changes to the previous <strong>BSR</strong> safety <strong>guidelines</strong> were
required. Searches were conducted using primarily
MEDLINE <strong>and</strong> PubMed. A manual search from the references
cited by generated articles was also used. All
searches were performed for literature up to October
2009. Abstracts were read for relevant evidence from
the <strong>BSR</strong>, EULAR <strong>and</strong> ACR annual conferences up to
October 2009. Evidence was graded according to the
strength of literature to support each statement, using
the grading suggested by the Royal College of
Physicians of London (National Clinical Guidelines for
Stroke by the Intercollegiate Stroke Working Party, June
2004 http://www.rcplondon.ac.uk/pubs/books/stroke/
stroke_<strong>guidelines</strong>_2ed.pdf, 22 June 2010, date last accessed;
Appendix I) <strong>and</strong> the document was prepared in
accordance with the principles outlined in the Appraisal of
Guidelines Research <strong>and</strong> Evaluation (AGREE) <strong>guidelines</strong>
(www.agreecollaboration.org).
How will these <strong>guidelines</strong> be piloted <strong>and</strong>
introduced?
These <strong>guidelines</strong> are an update of previous successful
<strong>guidelines</strong> <strong>and</strong> will be publicized via the <strong>BSR</strong>.
How often will these <strong>guidelines</strong> be reviewed?
The <strong>guidelines</strong> will be revisited either in 3 years or sooner
if major changes are required in the light of future published
anti-TNF safety information.
How will these <strong>guidelines</strong> be publicized <strong>and</strong>
implemented?
The full <strong>guidelines</strong> will be published on the <strong>BSR</strong>
website, <strong>and</strong> sent to all <strong>BSR</strong> members <strong>and</strong> primary care
trusts. A summary of the <strong>guidelines</strong> will be published in
Rheumatology, with web links to the full <strong>guidelines</strong>.
Cost implications <strong>and</strong> conflicts of interest
No funding has been received to assist with the development
of these <strong>guidelines</strong>. These <strong>guidelines</strong> have
been developed with complete editorial independence.
The conflicts of interest of contributors are listed at the
end of this document.
Infection
Infection in general
Recommendation 1: anti-TNF therapy should not be
initiated in the presence of serious active infections.
(Level III evidence, Grade of recommendation B.)
Recommendation 2: anti-TNF therapy should be
discontinued in the presence of serious infections,
but can be recommenced once the infection has
resolved clinically. (Level III evidence, Grade of
recommendation B.)
Recommendation 3: use anti-TNF therapy with caution
after discussing the relative risks (RRs) <strong>and</strong> benefits in the
following circumstances:
. chronic infected leg ulcers;
. septic <strong>arthritis</strong> of a native joint within the last
12 months;
. sepsis of a prosthetic joint within the last 12 months,
or indefinitely if the joint remains in situ;
. persistent or recurrent chest infections;
. indwelling urinary catheter;
2 www.rheumatology.oxfordjournals.org

RA <strong>guidelines</strong> on safety of anti-TNF therapies
. bronchiectasis; <strong>and</strong>
. hypogammaglobulinaemia.
(Level IV evidence, Grade of recommendation C.)
Serious infection (leading to hospital admission, i.v.
antibiotics or death) is a major contributor to the increased
mortality seen in RA patients—the risk is estimated to be
twice that of the general population [5]. RA severity <strong>and</strong>
extra-articular disease are also independently associated
with infection [5]. Therefore, when deciding whether any
drug raises infection risks, greater background infection
rates in RA need to be taken into account. More severely
affected patients are more likely to be on anti-TNF therapy
while at the same time having an elevated background
risk of infection.
Post-marketing surveillance <strong>and</strong> observational studies
provided the first indications that TNF inhibitors are associated
with an increased risk of serious infections [6, 7].
Data from individual r<strong>and</strong>omized trials <strong>and</strong> observational
studies were, however, inconsistent on this issue with reports
of both increased risk [8–12] <strong>and</strong> of no increased risk
[13–17]. The limitation of small numbers of patients in
r<strong>and</strong>omized controlled trials (RCTs) was addressed in a
systemic review <strong>and</strong> meta-analysis of nine RCTs of infliximab
<strong>and</strong> adalimumab that included 3493 patients who
received TNF inhibitors <strong>and</strong> 1512 who received placebo
[18]. The following observations were reported: pooled
odds ratio (OR) for serious infections among patients
treated with a TNF inhibitor was 2.0 (95% CI 1.3, 3.1)
<strong>and</strong> for serious non-granulomatous infections was
1.9 (95% CI 1.2, 2.9) in RA patients receiving infliximab
or adalimumab.
An increased risk of infection has also been noted in an
observational study from a German registry. An RR of serious
infections of 2.2 (95% CI 0.9, 5.4) for etanercept
(64 per 1000 pyrs) <strong>and</strong> 2.1 (95% CI 0.8, 5.5)
for infliximab (62 per 1000 pyrs) was found for 512 RA
patients on etanercept <strong>and</strong> 346 on infliximab when compared
with patients on DMARDs only (23 per 1000 pyrs)
[19]. The registry data included viral <strong>and</strong> fungal infections,
tuberculosis (TB) <strong>and</strong> bacterial infections. The Spanish
registry also showed a 1.6-fold increase in rate of serious
infection of 2868 pyrs for patients on anti-TNF drugs compared
with 2433 pyrs for controls [20].
In contrast, the British Society for Rheumatology
Biologics Registry (<strong>BSR</strong>BR) found no increase in serious
infections in patients treated with anti-TNF therapies when
compared with patients on conventional DMARDs [21].
This was an observational study of 7664 anti-TNF- <strong>and</strong>
1354 DMARD-treated RA patients. The incidence of
serious infections in the anti-TNF group was 53.2 per
1000 person-years (pyrs) vs 41.2 in the control group.
However, the frequency of serious skin <strong>and</strong> soft-tissue
infections was increased in the anti-TNF-treated patients
[incidence rate ratio (IRR) of 4.2]. Nineteen serious bacterial
intracellular infections occurred, all in patients in the
anti-TNF-treated cohort. Further analysis of the <strong>BSR</strong>BR
data has, however, suggested that the way in which
UK rheumatologists select patients for starting <strong>and</strong>
discontinuing anti-TNF therapy may have influenced the
lack of observed increased risk of infection in this study,
<strong>and</strong> that important increases in true risk may exist,
particularly in the early stages of treatment [22]. The
most recent data presented from the <strong>BSR</strong>BR (2009)
have shown that exposure to anti-TNF therapy is associated
with an increased risk of septic <strong>arthritis</strong>, with
incident rates of 5.0/1000 pyrs (95% CI 4.3, 5.8) in anti-
TNF-treated RA patients compared with 1.9/1000 pyrs
(95% CI 1.1, 3.0) in the DMARD control group [23].
In summary, although data from individual r<strong>and</strong>omized
clinical trials <strong>and</strong> the <strong>BSR</strong>BR failed to demonstrate a
consistent increase in treatment-related infections, data
from the meta-analysis of r<strong>and</strong>omized clinical trials, the
German <strong>and</strong> Spanish registries, <strong>and</strong> a re-analysis of
<strong>BSR</strong>BR data, are consistent with the concept that TNF
inhibitor use is associated with an approximate doubling
of risk of serious infection, particularly early on in the
course of therapy [22, 24].
Should infection develop in a patient treated with
anti-TNF there is some evidence to suggest that infections
may become more severe if patients continue with
anti-TNF therapy [25]. Patients, therefore, need to be
aware of the increased risk of infection, <strong>and</strong> that they
should stop anti-TNF therapy until the infection has clinically
resolved. Clinicians also need to be vigilant in assessing
for infection, including joint infection, in patients
treated with anti-TNF.
Anti-TNF therapies may also pose unacceptable
additional infection risk to subgroups of RA patients with
high background risk of infection (e.g. bronchiectasis).
Unfortunately these risks have not been addressed adequately
in the literature to date. This was highlighted by
Lieberman-Maran et al. [26], when they reported on two
RA patients who, having developed pneumonia while on
infliximab or adalimumab, were diagnosed with bronchiectasis.
Their literature review highlighted that data on any
potential association between increased infection risk with
anti-TNF <strong>and</strong> bronchiectasis were limited as many
anti-TNF therapy trials have excluded patients with a history
of bronchiectasis <strong>and</strong> most studies assessing infection
risk have not included subgroup analysis for
bronchiectasis.
Mycobacterial infections
Recommendation 4: prior to commencing treatment with
anti-TNF therapy, all patients should be screened for
mycobacterial infection in accordance with the latest
National <strong>guidelines</strong>. Active mycobacterial infection needs
to be adequately treated before anti-TNF therapy can be
started. (Level IIb evidence, Grade of recommendation B.)
Recommendation 5: prior to commencing anti-TNF
therapy, consideration of prophylactic anti-TB therapy
(as directed by the latest National <strong>guidelines</strong>) should be
given to patients with evidence of potential latent disease
(past history of TB or abnormal chest X-ray). (Level IIb
evidence, Grade of recommendation B.)
Recommendation 6: all patients commenced on
anti-TNF therapies should be closely monitored for
www.rheumatology.oxfordjournals.org 3

Tina Ding et al.
mycobacterial infections. This should continue for at least
6 months after stopping treatment due to the prolonged
elimination phase of the drug. (Level IV evidence, Grade of
recommendation C.)
Recommendation 7: patients on anti-TNF therapy who
develop symptoms suggestive of mycobacterial infections
should receive full anti-mycobacterial chemotherapy, but
may continue with their anti-TNF therapy if clinically indicated.
(Level IV evidence, Grade of recommendation C.)
Most published studies have compared the incidence of
TB in patients on TNF inhibitors with the baseline population
risk but there is now a well-established increased risk
of TB associated with anti-TNF therapy. The particularly
high anti-TNF-associated TB case rates reported from the
Spanish registry {1113 per 100 000 (post TB screening) for
infliximab, [6]}, Korea (2558 per 100 000 for infliximab,
[27]), Japan (325 per 100 000 for infliximab, [28]) <strong>and</strong>
Portugal (1500 per 100 000 for infliximab, [29]) are most
likely to reflect the high incidence of latent TB infection in
these countries; lower rates are reported in other countries
such as the USA (61.9 per 100 000 for infliximab, [30])
<strong>and</strong> Sweden (145 per 100 000 for infliximab, [31]).
Important differences in the risk of latent TB reactivation
exist among the first-generation drugs, with the
risk being higher with infliximab <strong>and</strong> adalimumab than
with etanercept [6, 7, 27, 31, 32]—a finding confirmed
with recently published data from the French <strong>and</strong>
British biologic registries [33, 34]. Data from the
<strong>BSR</strong>BR have shown that the rate of TB was higher
with the mAbs adalimumab (144 events/100 000 pyrs)
<strong>and</strong> infliximab (136 events/100 000 pyrs) than with
etanercept (39 events/100 000 pyrs). After adjustment,
the IRR compared with etanercept-treated patients
was 3.1 (95% CI 1.0, 9.5) for infliximab <strong>and</strong> 4.2
(95% CI 1.4, 12.4) for adalimumab [34]. The risk of
TB reactivation from certolizumab pegol is uncertain,
as experience with this newer drug is limited. Several
cases of TB were reported in clinical trials of certolizumab
pegol for the treatment of RA, whereas no cases
occurred in the control arms [35, 36].
TB has been shown to occur sooner after starting infliximab
than etanercept. Forty-three per cent of infliximabassociated
cases occurred during the first 90 days of
treatment, a pattern consistent with reactivation of latent
infection. In contrast, etanercept-associated TB cases
were distributed evenly throughout the reporting period,
with only 10% occurring during the first 90 days of treatment
[7].
TB occurring in association with TNF inhibitors has a
higher likelihood of involving extrapulmonary sites <strong>and</strong> of
being disseminated at presentation when compared with
other TB cases. The majority of these cases have had a
known history of previous TB, again suggesting reactivation
of latent disease [37].
Appropriate screening for evidence of previous exposure
or for latent TB, prior to commencing anti-TNF therapy
can minimize the risk of developing TB. The efficacy of
screening was demonstrated in the Spanish registry of
patients with rheumatic disease treated with TNF
inhibitors, Biobadaser. The impact of m<strong>and</strong>ated recommendations,
implemented in 2002, regarding screening
for <strong>and</strong> treatment of latent TB infection [38] was investigated.
A 78% reduction in TB case rates was noted following
m<strong>and</strong>ated screening but the risk still appeared to
be higher than that of the general population.
Furthermore, Gomez-Reino et al. have shown that failing
to follow recommendations on screening is associated
with a 7-fold increase in risk of TB [39].
New screening tests for latent TB (Quantiferon <strong>and</strong> the
T-spot) have been developed over recent years but their
role has not been fully validated in the <strong>rheumatoid</strong> population
<strong>and</strong> they are not widely available to clinicians.
However, they may have improved sensitivity in the immunosuppressed
host (including patients with HIV infection)
<strong>and</strong> might prove useful in those who have received
Bacille Calmette-Guérin (BCG) immunization [40]. The
cost of Quantiferon (£35/test) <strong>and</strong> T-spot (£100/test)
screening of all UK patients prior to treatment with
anti-TNF would prove prohibitive (personal communication:
P. Ormerod, Blackburn Royal Infirmary). At the age
of 50 years the annual incidence of active TB disease in
the UK White Caucasian population is 5/100 000 <strong>and</strong> the
incidence of latent TB is approximately 10 times this level
(50/100 000). Therefore, the number of tests needed to
detect one case of latent infection would be 2000, at a
cost of £70 000 with Quantiferon <strong>and</strong> £200 000 with the
T-spot test. The economics improve slightly in older patients
but are much worse (more than double the cost) for
patients under the age of 35 years. The cost-effectiveness
is better for patients from ethnic minorities known to be
at higher risk of TB. For patients from South Asia, the
incidence of TB is 120/100 000 <strong>and</strong> in black Africans it is
240/100 000, <strong>and</strong> this reduces the numbers needed to test
to detect one positive case to 82 <strong>and</strong> 41, respectively. The
true-negative predictive value of a negative test or the
true-positive predictive value of a positive test is still not
known, however, <strong>and</strong> these tests are currently therefore
not recommended by the British Thoracic Society as
TB-screening tools. National Institute for Clinical
Excellence (NICE) are currently developing <strong>guidelines</strong> on
the use of these IFN-g tests for diagnosing latent TB with a
planned publication date of November 2010.
Atypical mycobacterial infections
The numbers are too small to draw any significant conclusions
but there are now many case reports of various
atypical mycobacterium infections associated with all
three first-generation anti-TNF drugs (Table 1). In a
recent review of the US Food <strong>and</strong> Drug Administration
(FDA) MedWatch database for reports of non-tuberculous
mycobacteria (NTM) infections in patients receiving
anti-TNF-a therapy, 239 reports were recorded. Of those
reports, 105 (44%) met NTM disease criteria [41]. The
median age of patients was 62 years; the majority of patients
were females [66 (65%)] had RA [73 (70%)] <strong>and</strong>
were taking prednisolone [68 (65%)] or MTX [58 (55%)]
concurrently. NTM infections were associated with infliximab
(n = 73), etanercept (n = 25) <strong>and</strong> adalimumab (n =7)
4 www.rheumatology.oxfordjournals.org

RA <strong>guidelines</strong> on safety of anti-TNF therapies
TABLE 1 Mycobacteria other than TB (MOTT)-associated
infections reported in patients receiving anti-TNF therapy
for RA
MOTT infections
Anti-TNF
therapy
References
Mycobacterium abscessus Infliximab <strong>and</strong> [43, 44]
etanercept
Mycobacterium avium Infliximab <strong>and</strong> [42, 45, 46]
complex
etanercept
Mycobacterium chelonae Adalimumab [47, 48]
Mycobacterium fortuitum Infliximab [49]
Mycobacterium mucogenicum Etanercept [50]
Mycobacterium szulgai Etanercept [51]
Mycobacterium xenopi Etanercept [52]
treatment. Mycobacterium avium [n = 52 (50%)] was most
commonly implicated, <strong>and</strong> nine patients (9%) had died at
the time their infections were reported. A high rate of
extrapulmonary manifestations [n = 46 (44%)] was also
reported.
A previous review of US data on more than 233 000
patients treated with infliximab from 1998 to 2002, [7]
showed that 30 patients developed unspecified mycobacterial
species infection. Mycobacteria other than TB
(MOTT) is rare <strong>and</strong> it remains difficult to determine
whether cases associated with anti-TNF therapy are due
to reactivation or primary infection. Cases of MOTT associated
with TNF inhibitors may progress rapidly but have
to date responded well to treatment <strong>and</strong> withdrawal of
anti-TNF treatment [42].
Other opportunistic bacterial <strong>and</strong> fungal infections
Recommendation 8: patients on anti-TNF should be
informed of appropriate food hygiene (see Arthritis
Research Campaign patient information leaflets on etanercept
[53], infliximab [54] <strong>and</strong> adalimumab [55]). (Level
III evidence, Grade of recommendation B.)
Recommendation 9: health-care professionals managing
patients on anti-TNF therapy should be aware of
the risk of opportunistic infections in patients on
anti-TNF therapy. They should have a high index of suspicion
for atypical <strong>and</strong> opportunistic infections, <strong>and</strong>
anti-TNF therapy should be promptly stopped in suspected
cases <strong>and</strong> patients should have rapid access to
specialist health care for consideration of early
anti-bacterial/anti-fungal treatment. (Level IIII evidence,
Grade of recommendation B.)
Food-borne infections from both Listeria <strong>and</strong>
Salmonella have been reported with anti-TNF therapy.
Listeria can be found in uncooked meat, vegetables, unpasteurized
milk or foods prepared from raw milk. One
report on 15 patients with Listeria infection while on
anti-TNF therapy [56], indicated that 14 of the 15 cases
were associated with infliximab with the remaining case
occurring with etanercept. A further case of Listeria sepsis
with infliximab [57], two cases of Listeria meningitis with
etanercept [58] <strong>and</strong> infliximab [59] <strong>and</strong> a case of joint
sepsis with adalimumab [21] have also been reported.
Wallis et al. [7] reported on data collected by the
Adverse Event Reporting System (AERS) of the US FDA.
Between January 1998 <strong>and</strong> September 2002, 323 cases
of granulomatous infections were found in patients treated
with infliximab or etanercept, including 38 cases with listeriosis
(36 cases associated with infliximab <strong>and</strong> 2 with
etanercept).
Salmonella can be spread through contaminated raw
eggs, unpasteurized milk <strong>and</strong> undercooked meat. Wallis
et al. [7] have identified 11 cases of anti-TNF-associated
Salmonella infection (7 cases with infliximab <strong>and</strong> 4 cases
with etanercept) from the FDA’s AERS. Two cases of
Salmonella septic <strong>arthritis</strong> have been reported with etanercept
<strong>and</strong> a further two have been reported with infliximab
[57, 60]. Three cases of Salmonella septicaemia have
been reported with anti-TNF therapy [61].
Invasive fungal infections (IFIs) are also reported with
anti-TNF therapy. A recent review by Tsiodras et al. [62]
reported on MEDLINE <strong>and</strong> PubMed database searches
for reports on IFIs associated with the three available
anti-TNF therapies. Of the 281 cases of IFI associated
with TNF-a inhibition, 226 (80%) were associated with
infliximab, 44 (16%) with etanercept <strong>and</strong> 11 (4%) with
adalimumab. The most prevalent IFIs were histoplasmosis
[n = 84 (30%)], c<strong>and</strong>idiasis [n = 64 (23%)], aspergillosis
[n = 64 (23%)] <strong>and</strong> Coccidioides species [n = 29 (10%)].
Pneumonia was the most common pattern of infection.
Pneumocystis jiroveci pneumonia [formerly known as
Pneumocystis carinii pneumonia (PCP)] is also reported
with each of the anti-TNF agents. Eighty-four cases of
PCP pneumonia with infliximab were reported from the
FDA safety database; the majority of these patients were
also on other immunosuppressants [63]. There are further
case reports of PCP infection following infliximab [64–66],
etanercept [67] <strong>and</strong> adalimumab [68] therapy. The risk of
PCP has been reported to correlate with high-dose glucocorticoid
use [69] but, to date, there are inadequate data
to support the use of primary prophylaxis.
Varicella infections
Recommendation 10: if a patient on anti-TNF treatment,
or one of their household contacts, develops primary varicella
(chickenpox), <strong>and</strong> if the risks from infection are perceived
to be significant, the patient should be considered
for varicella zoster immune globulin (VZIG). Shingles
should be treated conventionally. (Level IIb evidence,
Grade of recommendation B.)
Experience with primary varicella infection is limited as
>90% of the adult population have evidence of serological
immunity <strong>and</strong> all bar a very few cases of varicella infection
are due to reactivation [70]. Several case reports
highlight risks from varicella zoster reactivation, in patients
treated with anti-TNF therapy. Infliximab has had the highest
rate of reports [71, 72] but cases have also been reported
with etanercept [73] <strong>and</strong> adalimumab [74].
In a prospective study that included 5040 patients
receiving a variety of DMARDs for RA, TNF-a inhibitors
were associated with a moderately increased risk of
herpes zoster [75]. There were 86 cases of herpes
www.rheumatology.oxfordjournals.org 5

Tina Ding et al.
zoster among 82 patients. Fifteen patients had multidermatomal
zoster, <strong>and</strong> four had herpes zoster ophthalmicus.
Eighteen events were serious with 12 requiring hospitalization
due to either severe multidermatomal disease
(n = 8), eye involvement (n = 1) or other reasons (n = 3).
Complications were reported in three patients; postherpetic
neuralgia occurred in two patients (one while
receiving etanercept <strong>and</strong> one while receiving adalimumab)
<strong>and</strong> multidermatomal zoster with oesophagitis <strong>and</strong>
pulmonary involvement occurred in one patient (while
receiving infliximab). A statistically significant association
between infliximab or adalimumab use <strong>and</strong> herpes
zoster was observed after adjustment for age, disease severity
<strong>and</strong> prednisolone use [hazard ratio (HR) 1.82; 95%
CI 1.05, 3.15]. The risk was also increased with etanercept,
although it did not achieve statistical significance.
Similarly, Smitten et al. [76], analysed two large databases
in the USA <strong>and</strong> UK <strong>and</strong> reported an increased
risk of herpes zoster in patients receiving any diseasemodifying
treatment—the OR, when compared with RA
patients not receiving DMARDs or glucocorticoids, was
1.54 (95% CI 1.04, 2.29) for patients receiving anti-TNF,
1.37 (95% CI 1.18, 1.59) for patients receiving conventional
DMARDs alone <strong>and</strong> 2.51 (95% CI 2.05, 3.06) for patients
receiving glucocorticoids.
In contrast to the studies described above, two large
retrospective studies of patients with RA did not demonstrate
an increased risk of herpes zoster in patients receiving
TNF-a inhibitors [77, 78]. Most of the reported cases of
varicella zoster reactivation did well with conventional
anti-viral treatment.
Little data are available on the effectiveness of VZIG in
preventing infection in immunosuppressed patients. One
study of immunosuppressed children who had household
exposure to varicella [79] demonstrated a moderate response
to VZIG; however, 49 of 81 recipients of VZIG
(60%), including some patients who had serological evidence
of past varicella exposure, still developed varicella
infection. The <strong>BSR</strong> has previously produced <strong>guidelines</strong>
(2002) on immunizations in immunosuppressed patients,
which recommend the use of VZIG if contact with varicella
is significant [70]. Until more data are available on
anti-TNF-treated patients we feel it appropriate to recommend
that VZIG is considered for such patients exposed
to varicella infection in whom the risks of infection are
perceived to be high.
Hepatitis
Small numbers of case reports <strong>and</strong> case series continue
to be published on anti-TNF therapy in patients with hepatitis
B <strong>and</strong> C.
HBV
Recommendation 11: screening for risk factors for HBV
infection should be performed prior to commencing
anti-TNF therapy <strong>and</strong> HBV tests should be performed in
patients with risk factors. In patients who are HBV positive,
a risk : benefit assessment should be undertaken, as
anti-TNF treatment may be safe if appropriate anti-viral
treatment is given. (Level IV evidence, Grade of recommendation
C.)
Recommendation 12: close monitoring of serum aminotransaminases
<strong>and</strong> HBV DNA load during therapy should
be considered in patients with HBV treated with anti-TNF
therapy <strong>and</strong> concomitant anti-viral treatment would be
recommended. (Level IV evidence, Grade of recommendation
C.)
Recommendation 13: patients with serological evidence
of cleared past infection [HBsAg negative/core antibody
(anti-HBcAb) positive] should have their HBV serology
monitored during therapy <strong>and</strong> may require concomitant
anti-viral treatment if detrimental changes develop.
(Level IV evidence, Grade of recommendation C.)
When anti-TNF therapies have been used in HBVinfected
patients who have also received concomitant
anti-viral treatment the outcome from case reports has
been good for infliximab, etanercept <strong>and</strong> adalimumab
with no evidence of viral reactivation [80–83]. Serum aminotransaminases
remained normal <strong>and</strong> there was no increase
in viral load during treatment with each of the three
anti-TNF agents. A further report on two HBV-infected RA
patients <strong>and</strong> one HBV-infected SpA patient also indicated
a good outcome of treatment with anti-TNF agents when
concomitant lamivudine is used [82].
The outcome from case reports of HBV-infected patients
with RA <strong>and</strong> seronegative <strong>arthritis</strong> treated with
anti-TNF therapy has been less favourable, however,
when concomitant anti-viral treatment is not prescribed
[84–88]. Reactivation of HBV infection in these cases
was controlled with the introduction of lamivudine.
Although there have been reports of a few patients
doing well on anti-TNF without anti-viral prophylaxis [89],
the European Association for the Study of the Liver recommends
that anti-viral therapy begins 2–4 weeks prior to
the start of a TNF inhibitor in patients with inactive hepatitis
B [80]. The duration of treatment required is currently
unknown but may need to be life long.
There are some reports indicating that anti-TNF may be
safe in patients with serological evidence of cleared
past infection with hepatitis B (i.e. HBsAg-negative/anti-
HBcAb-positive) without the need for anti-viral prophylaxis
[90, 91]. However, Charpin et al. [90] have reported
a significant decrease in anti-HBsAb titre in a proportion
of patients treated with anti-TNF <strong>and</strong> recommended that
HBV serology titres should be monitored in these patients,
especially in those with a low anti-HBsAb titre at baseline.
HCV
Recommendation 14: screening for risk factors for HCV
infection should be performed prior to commencing
anti-TNF <strong>and</strong> HCV tests should be performed in patients
with risk factors. Although studies to date suggest that
anti-TNF therapies do not have a detrimental effect on
HCV infection, anti-TNF should continue to be used
with caution in such patients. (Level III evidence, Grade
of recommendation B.)
Recommendation 15: close monitoring of serum aminotransaminases
<strong>and</strong> HCV RNA during therapy should be
6 www.rheumatology.oxfordjournals.org

RA <strong>guidelines</strong> on safety of anti-TNF therapies
performed in patients with HCV treated with anti-TNF
therapy. (Level III evidence, Grade of recommendation B.)
There are a number of case reports <strong>and</strong> series demonstrating
a good response to anti-TNF treatment, with no
reactivation of HCV infection, regardless of whether concomitant
anti-viral treatment is prescribed. Peterson et al.
[92] reported no significant adverse events or significant
changes in liver aminotransaminase levels among
24 patients with chronic HCV infection <strong>and</strong> RA who
received infliximab or etanercept therapy. They also reported
no change in HCV RNA viral load in six patients
but some reduction in HCV RNA viral load in the remaining
16 patients.
Zein [93] reported on a small, r<strong>and</strong>omized, placebo
controlled trial (25 patients in each arm) of etanercept as
adjuvant therapy to ribavirin in HCV-infected patients.
More patients in the etanercept arm met the primary
endpoints of normal alanine transaminase, absence of
serum HCV RNA <strong>and</strong> absence of serious adverse side
effects.
Further case reports/series for each of the anti-TNF
agents have also reported good outcomes from anti-
TNF treatment when used in HCV-infected patients with
RA or seronegative arthropathies [81, 82, 88, 94–99].
There has been just one case report of HCV reactivation
with etanercept [100].
HIV
Recommendation 16: risk factors for HIV infection should
be documented prior to commencing anti-TNF therapy
<strong>and</strong>, if present, an HIV test should be done. (Level III evidence,
Grade of recommendation B.)
Recommendation 17: in considering anti-TNF use in
HIV-positive patients, the following should influence
decision-making, that a reasonable benefit to risk ratio
exists for HIV patients if: (i) HIV infection is controlled
<strong>and</strong> patients are not severely immunosuppressed
(e.g. CD4 count >200 <strong>and</strong> HIV viral load 200
<strong>and</strong> a HIV viral load

Tina Ding et al.
undertaken were orthopaedic (85%). The complication
rates for orthopaedic procedures <strong>and</strong> for abdominal procedures
were 13 <strong>and</strong> 43%, respectively—higher than the
reported rates for surgery in RA patients [7% (orthopaedic)
<strong>and</strong> 13% (abdominal)]. Dixon et al. [110] have analysed
data from the <strong>BSR</strong>BR <strong>and</strong> found that the risk of serious
post-operative infection is nearly 2-fold higher in patients
who received anti-TNF therapy in the 28 days prior to
surgery than in those who were not exposed.
Pappas <strong>and</strong> Giles [111] assessed peri-operative infection
risks of anti-TNF therapy associated specifically
with orthopaedic surgery. All studies assessed involved
retrospective outcome data <strong>and</strong> the source populations
differed greatly between studies making direct comparison
difficult. Of the six published major studies, only one
study [108] identified an increased risk associated with
peri-operative anti-TNF exposure but the authors recommended
that anti-TNF therapy should be stopped
for a duration of three to five times their half-lives
pre-operatively <strong>and</strong> restarted once the wound healing is
deemed satisfactory (10–14 days).
Immunization
Recommendation 21: when considering primary immunizations
<strong>and</strong> live attenuated immunizations in patients on
anti-TNF therapy, it should be acknowledged that data on
the effects of anti-TNF therapies are limited. Until further
evidence is available, the <strong>BSR</strong> recommendations on the
use of immunizations in patients on immunosuppressive
therapy should be adhered to in patients on anti-TNF therapy
[70]. (Level IV evidence, Grade of recommendation C.)
Recommendation 22: although there may be an attenuated
response (particularly if MTX is being co-prescribed),
patients on anti-TNF therapy should receive both influenza
(including vaccines generated for specific flu outbreaks
such as the recent Swine flu) <strong>and</strong> pneumoccocal immunizations
unless there are contraindications. (Level IV evidence,
Grade of recommendation C.)
Recommendation 23: prior to commencing anti-TNF
therapy, hepatitis B immunization should be considered
for at-risk patients. (Level IV evidence, Grade of recommendation
C.)
Patients receiving immunosuppressive therapy are
advised to have both influenza <strong>and</strong> pneumococcal immunization.
There have been only a few studies investigating
the administration of vaccines in <strong>rheumatoid</strong> patients
with assessments made on RA patients treated either with
MTX alone or in combination with anti-TNF. A study [112]
in SLE has shown that these patients had slightly lower
response rates than expected for the healthy population
but the disease itself could be an additional factor for
decreased responsiveness. There are now some data,
as detailed below, on the titre response to specific immunizations
in patients receiving anti-TNF therapy. The
effects of live attenuated vaccines in patients receiving
anti-TNF agents remain unknown but these vaccines
are not recommended in patients receiving anti-TNF
treatment.
Influenza immunization
There is some evidence that there are differences
between each of the first-generation anti-TNF agents
in terms of response to influenza immunization.
Adalimumab has been shown not to reduce responsiveness
to immunization, whether used as monotherapy or in
combination with MTX [113]. However, study findings for
etanercept <strong>and</strong> infliximab [114, 115] indicate that these
drugs should be considered as factors that could lead
to decreased response rates. A recent review [114] of influenza
studies has indicated that this immunization is well
tolerated <strong>and</strong> effective with or without the administration
of MTX. However, a subset of patients may remain
unprotected.
Pneumococcal immunization
MTX is associated with lower response rates to pneumococcal
vaccines in most studies in patients with RA, SLE
<strong>and</strong> other rheumatological diseases [113, 116–119].
In contrast, the use of etanercept or infliximab with
MTX has not been associated with poor response
[113, 116, 117, 119]. The theory that certain anti-TNF
agents can lead to enhanced immune responses was
put forward from one study [116], where patients with
MTX had lowered responsiveness, but those with the
combination of infliximab or etanercept with MTX had
even better response rates than the healthy controls.
The use of adalimumab alone has not been shown to
affect antibody titre response to pneumococcal immunization,
but the combination of MTX <strong>and</strong> adalimumab has
resulted in decreased response rates when compared
with MTX alone [113]. In order to increase the chance of
developing protective antibody titres, it has been suggested
that pneumococcal immunization should be
considered before starting MTX in combination with adalimumab
[118].
Hepatitis B immunization
There are limited data with regard to response to hepatitis
B immunization in rheumatic patients on anti-TNF therapy.
A study by Ravikumar et al. [120] suggests that TNF
blockade with etanercept decreases responses to the
hepatitis B vaccine. Patients with RA, who were treated
with MTX, had very good response rates, whereas those
treated with an MTX–etanercept combination or etanercept
alone had poor response rates. Hepatitis B immunization
is recommended for all unvaccinated adults at risk
for HBV infection, ideally prior to starting anti-TNF
treatment.
Duration of immunization response
Antibody levels in <strong>rheumatoid</strong> patients on DMARD therapy
have been shown to decline faster than in those not on
DMARD therapy <strong>and</strong> this raises the question of whether
such patients should be revaccinated more frequently
than st<strong>and</strong>ard recommendations [121]. No data to support
specific recommendations are yet available on this issue
for patients on anti-TNF therapy.
8 www.rheumatology.oxfordjournals.org

RA <strong>guidelines</strong> on safety of anti-TNF therapies
Malignancy
Recommendation 24: patients commencing anti-TNF
therapy should be informed that overall there is no conclusive
evidence for an increase in risk of solid tumours or
lymphoproliferative disease with the anti-TNF therapies
above that which would be expected for the rest of the
RA population, but ongoing vigilance is required. (Level Ia
evidence, Grade of recommendation A.)
Recommendation 25: patients should be investigated
for potential malignancy if clinically suspected, <strong>and</strong> anti-
TNF treatment should be stopped if malignancy is confirmed.
(Level III evidence, Grade of recommendation B.)
Recommendation 26: caution should be exercised in
the use of anti-TNF therapies in patients with previous malignancy.
(Level III evidence, Grade of recommendation B.)
Recommendation 27: the effect of anti-TNF therapies
on pre-malignant conditions such as Barrett’s oesophagus,
cervical dysplasia <strong>and</strong> large bowel polyps is unknown.
Caution should be exercised in the use of
anti-TNF therapies in such patients. (Level IV evidence,
Grade of recommendation C.)
Recommendation 28: patients should be advised that
there appears to be an increased risk of some skin cancers
with anti-TNF therapy <strong>and</strong> on preventative skin care
<strong>and</strong> skin surveillance. Patients should be encouraged to
promptly report any new persistent skin lesions. (Level IIb
evidence, Grade of recommendation B.)
Overall cancer rates
The possibility of an increase in the occurrence of cancer
associated with anti-TNF therapy has been raised by
some RCTs <strong>and</strong> a recent meta-analysis [18]. This
meta-analysis of nine studies identified a 3-fold increase
in the rate of malignancy with infliximab <strong>and</strong> adalimumab
(OR 3.3; 95% CI 1.2, 9.1). High-dose therapy (defined as
56 mg/kg infliximab every 8 weeks or 540 mg adalimumab
every other week) was associated with the greatest
risk (OR 4.3; 95% CI 1.6, 11.8); there was no significant
increased risk with lower doses (OR 1.4; 95% CI 0.3, 5.7).
All nine clinical trials included in this meta-analysis
excluded patients with pre-existing malignancies, <strong>and</strong>
other factors could have lead to bias towards detection
of malignancy in the anti-TNF arms of these studies.
In contrast, most observational studies have failed to
confirm an increased overall risk of cancer among patients
treated with TNF inhibitors. Setoguchi et al. [122] undertook
a retrospective cohort study of US databases
<strong>and</strong> controls <strong>and</strong> reported no increased risk of overall
cancers [pooled adjusted HR for biologic users was 0.99
(95% CI 0.71, 1.36)]. Two reports from the Swedish registries
[123, 124] indicated no increase in overall cancer risk
in patients receiving anti-TNF therapy compared with
those that are not [st<strong>and</strong>ardized incidence ratio (SIR) 1.1
(95% CI 0.6, 1.8); adjusted RR 0.93 (95% 0.76, 1.13)].
Cancer events from Spanish registries [20] are reported
to be lower in anti-TNF-treated RA patients; the RR in
RA patients not on anti-TNF therapy (n = 789) was 2.9
(95% CI 2.4, 3.5) when compared with RA patients treated
with TNF blockers (n = 4459). The <strong>BSR</strong>BR [125] have
reported no difference in the incidence of new malignancy
in their anti-TNF cohort when compared with the DMARD
cohort (adjusted IRR 0.7; 95% CI 0.4, 1.2). A large observational
study from the USA [126] has recently also concluded
that there is no increased overall risk of any
malignancy for RA patients treated with anti-TNF therapy
(SIR for all cancer was 1.0; 95% CI 1.0, 1.1).
Lymphoma
Patients with RA, irrespective of any treatments they
receive, are known to have an increased incidence of
lymphoma with reported risk ratios, compared with
the general population, ranging from 1.9 to 2.7
[127, 133]. Post-marketing surveillance has identified patients
who have developed lymphoma while being treated
with anti-TNF therapies. A 2002 study based upon reports
to the FDA noted 26 cases of lymphoma (mostly non-
Hodgkin’s lymphoma) among patients treated with etanercept
or infliximab [134]. Fifty-six per cent of the lymphomas
were detected within 8 weeks of initiation of
anti-TNF treatment <strong>and</strong> two patients had regression of
their lymphoma when anti-TNF therapy was discontinued.
Clinical data on the risk of lymphoma potentially directly
stemming from the use of TNF inhibitors are mixed.
Evidence for an increased risk is provided by three main
reports. A meta-analysis of nine RCTs [18] identified
10 lymphoma cases in anti-TNF-treated patients compared
with no lymphomas in DMARD-treated patients. A
retrospective cohort study from Sweden [123] reported an
SIR for lymphoma of 11.5 (95% CI 3.7, 26.9) compared
with that of 1.3 (95% CI 0.2, 4.5) for controls (DMARD
users). The 5-fold increase of lymphoma incidence
among RA patients exposed to anti-TNF agents may, in
part, have been explained by confounding by indication
<strong>and</strong> a low rate of lymphoma in the control population. An
observational study from the USA [131] also reported a
increased risk with an SIR for lymphoma in patients treated
with biologics of 2.9 (95% CI 1.7, 4.9). However, no
adjustment for baseline differences in disease duration
<strong>and</strong> severity was made in this study.
Three main studies have reported no increase in lymphoma
risk with anti-TNF therapies. A retrospective cohort
study [122] reported a pooled HR for haematological
malignancy, when compared with MTX users, of 1.37
(95% CI 0.71, 2.65). An observational study from
Sweden [132] reported SIRs for lymphoma of 1.9 in the
RA cohort vs 2.9 in the biologics user cohort. There was
no significant increase in risk when compared with the
control group (RR of 1.1; 95% CI 0.6, 2.1). A further observational
study from the USA [135] showed no increase
in risk [OR of 1.0 when compared with controls (95% CI
0.6, 1.8)].
Solid malignancy
There are also conflicting data on the risks of solid malignancy
associated with anti-TNF therapy. An increased risk
was reported from the meta-analysis of Bongartz et al.
[18] with 15 solid malignancies identified among 29 malignancies
in the TNF inhibitor group. In contrast, only
www.rheumatology.oxfordjournals.org 9

Tina Ding et al.
one solid malignancy was observed among the three
malignancies seen in the comparison group. However,
no increase in risk was found from the Swedish Early
Arthritis Registry [132], where rates of solid malignancy
among patients treated with anti-TNF therapy was no
higher than expected for the underlying disease (SIR
0.9; 95% CI 0.7, 1.2).
Non-melanoma <strong>and</strong> melanoma skin cancer
There is evidence emerging for an increased risk of
non-melanotic skin cancer (NMSC) <strong>and</strong> also for malignant
melanomas (MMs) among patients treated with anti-TNF.
Wolfe <strong>and</strong> Michaud [126] reported increased rates of
NMSC with biologic therapy (OR 1.5; 95% CI 1.2, 1.8)
<strong>and</strong> a possible, but not statistically significant, higher
risk for MM (OR 2.3; 95% CI 0.9, 5.4). Chakravarty et al.
[136] have also reported an increased risk for developing
NMSC from a large cohort study of RA patients treated
with anti-TNF agents (HR 1.19; P = 0.042) <strong>and</strong> the Swedish
Biologic Register has also reported a non-significant
increased occurrence of NMSC in their RA cohort treated
with anti-TNF therapy (RR 1.55; 95% CI 0.76, 3.15; n = 10)
[124]. Many of the malignancies reported from the
meta-analysis of Bongartz et al. [18] were NMSC (9 out
of 35).
More recently, Amari et al. [137] reported an increased
incidence of both NMSC <strong>and</strong> MM from a large retrospective
cohort of RA patients (16 829 RA patients, of whom
3096 were on anti-TNF treatment). The incidence of
NMSC was 25.9 per 1000 pyrs in patients on TNF antagonists
<strong>and</strong> 19.6 per 1000 pyrs in those on non-biologic
DMARDs. Patients on anti-TNF agents had a higher risk
of developing NMSC than those on non-biologic
DMARDs with an HR of 1.34 (95% CI 1.15, 1.58;
P < 0.0001). Factors that increased the risk of NMSC
included older age, male gender, glucocorticoid use,
a history of prior malignancies <strong>and</strong> duration of anti-TNF
therapy. The incidence of MM was also increased at 3.7
per 1000 pyrs in patients on TNF antagonists, compared
with 2.6 per 1000 pyrs in those on non-biologic DMARDs.
Patients on anti-TNF agents had a higher risk of MM with
an HR of 1.5 (95% CI 1.01, 2.24; P < 0.05).
The <strong>BSR</strong>BR also published its data on the risk of NMSC
in 2009 [138]. Among 11 757 consecutive anti-TNF-treated
patients with RA, compared with 3515 biologic-naïve subjects
with active RA receiving traditional DMARD, there
was a non-significant but 70% increased risk of NMSC.
Infliximab was associated with the highest risk of NMSC in
this group (an almost 3-fold increase in risk). This data has
resulted in recommendation that vigilance for NMSC
should be maintained in all patients with RA, but especially
in those treated with anti-TNF therapy.
Previous history of malignancy
Little data are available on the risks of anti-TNF therapies
in patients with a past history of malignancy but caution
has generally been advised on the basis of theoretical risk.
Watson et al. [125] (<strong>BSR</strong>BR) compared a group of 154 patients
with RA <strong>and</strong> previous cancer [65 (42%) of these
patients had had their malignancy within 10 years of starting
anti-TNF therapy] with a group of 9844 patients
with RA but no previous malignancy. Six patients (4%)
who had previous malignancy developed a new cancer,
compared with 158 patients (1.6%) without previous
malignancy [IRR 2.5 (95% CI 1.2, 5.8)]. Three of these
cancers (50%) occurred in patients with a previous malignancy
that had developed

RA <strong>guidelines</strong> on safety of anti-TNF therapies
group receiving only DMARD therapy. Of these, 40
anti-TNF-treated patients have developed a new lupus
event, compared with only one of the DMARD-treated patients
(adjusted IRR 3.17; 95% CI 0.38, 26.26). Although
the number of events was small, there was a trend towards
an increased incidence of lupus events in those
receiving anti-TNF therapies. The most common lupus
symptom was skin rash; lupus nephritis or neuropsychiatric
symptoms were not reported [146].
Ramos-Casals et al. [147] assessed all cases of autoimmune
diseases induced by anti-TNF therapies reported
in the literature through to December 2006 (233 cases).
One hundred <strong>and</strong> thirteen cases of vasculitis were identified
in association with anti-TNF therapy (59 cases with
etanercept, 47 with infliximab, 5 with adalimumab).
Ninety-eight (87%) of the patients with vasculitis had cutaneous
involvement including palpable purpura [55 patients
(57%)], erythematous papules/macules/punctate
lesions [11 patients (11%)], ulcers [9 patients (9%)]
<strong>and</strong> nodules [9 patients (9%)]. Visceral involvement of
the peripheral nerves <strong>and</strong> kidneys were reported in
18 (16%) <strong>and</strong> 15 (13%) patients, respectively. Eightythree
of the 113 vasculitis cases were confirmed histopathologically.
Among these, leucocytoclastic vasculitis
was reported in 52 patients (63%), necrotizing vasculitis
in 14 (17%), lymphocytic vasculitis in 5 (6%) <strong>and</strong> other
findings in 12 (14%). Ninety-two of the cases of vasculitis
resolved following the discontinuation of anti-TNF therapy.
Among the 16 patients re-challenged with a TNF inhibitor,
vasculitis recurred in 12 (75%). Ninety-two cases
of lupus following anti-TNF therapy (40 cases with infliximab,
37 with etanercept, 15 with adalimumab) were also
reported. Nearly half the cases fulfilled four or more ACR
classification criteria for SLE; this proportion fell to
one-third after discarding cases with pre-existing
lupus-like features. The symptoms <strong>and</strong> signs of SLE
resolved following the discontinuation of the TNF inhibitor
in 71 of the 72 patients on whom information regarding
disease outcome was available.
In summary, <strong>and</strong> as highlighted from these reports, the
clinical features of anti-TNF-induced lupus appear distinct
from classical drug-induced lupus (DIL) with a phenotype
more similar to idiopathic SLE [145], albeit a mild form. It is
associated with more cerebral <strong>and</strong> renal involvement than
classical DIL. Withdrawal of anti-TNF therapy usually
leads to resolution of symptoms. Steroids <strong>and</strong>/or immunosuppressive
therapy may be required in severe cases.
Demyelination
Recommendation 30: anti-TNF therapy should not be
given when there is a clear history of multiple sclerosis
<strong>and</strong> should be used with caution with other demyelinating
diseases. (Level III evidence, Grade of recommendation
B.)
Recommendation 31: anti-TNF therapy should be withdrawn
if demyelination occurs <strong>and</strong> the patient should be
referred for specialist investigation. (Level III evidence,
Grade of recommendation B.)
Stubgen et al. [148] have recently reviewed ongoing
case histories <strong>and</strong> series reports <strong>and</strong> highlighted the association
between anti-TNF treatment <strong>and</strong> various disorders
of peripheral nerves such as Guillain–Barré
syndrome, Miller Fisher syndrome, chronic inflammatory
demyelinating polyneuropathy, multifocal motor neuropathy
with conduction block, mononeuropathy multiplex
<strong>and</strong> axonal sensorimotor polyneuropathies. Cases were
reported for all three first-generation anti-TNF agents
with symptoms developing over a wide range of time intervals
from starting anti-TNF treatment (8 h to 2 years).
Withdrawal of anti-TNF resulted in slow resolution of
symptoms in many cases; however, additional treatment
with i.v. steroid, with or without IVIG or cyclophosphamide,
or plasmapheresis was required for others. With
these interventions symptoms resolved in the vast majority
of cases over time periods ranging from 3 weeks to
35 months. A small minority of patients, however, never
achieved symptom resolution. There were some reports of
successful re-challenge with anti-TNF without recurrence
of neurological symptoms, but there are also reports of
worsening symptoms or recurrence on re-challenge, <strong>and</strong>
re-challenge is therefore not recommended but may be
considered on an individual case basis. Additional case
reports reinforce the risks of neurological complications
from anti-TNF therapies.
CNS demyelination
There are further reports implicating all three firstgeneration
anti-TNF agents in new cases of demyelination
with the highest rate reported with etanercept [149–152],
followed by infliximab [153, 154] <strong>and</strong> then adalimumab
[155].
Optic neuritis
Simsek et al. [156] have reported 15 patients who developed
optic neuritis while on anti-TNF therapy—8 of
these patients had received infliximab, 5 had received
etanercept <strong>and</strong> 2 patients had received adalimumab. On
stopping anti-TNF treatment, nine patients experienced
complete resolution <strong>and</strong> two patients had partial resolution
of their symptoms while four patients continued
to have symptoms.
Demyelinating peripheral nervous system disease
There are also further reports implicating all three firstgeneration
anti-TNF agents in the development of peripheral
nerve demyelination [157]. Shin et al. [158] report
one incident case from the authors themselves <strong>and</strong>
15 patients identified from the FDA database, in whom
Guillain–Barre syndrome developed following anti-TNF
therapy. Guillain–Barre syndrome developed following
infliximab therapy in 10 patients, following etanercept
therapy in 5 patients <strong>and</strong> following adalimumab therapy
in 1 patient. Among the 13 patients for whom follow-up
data were available, 1 patient experienced no resolution,
9 patients had partial resolution <strong>and</strong> 3 patients had complete
resolution of Guillain–Barre syndrome following
therapy.
www.rheumatology.oxfordjournals.org 11

Tina Ding et al.
TABLE 2 New York Heart Association (NYHA) grading of
heart failure
Cardiac failure <strong>and</strong> ischaemic heart
disease
Recommendation 32: anti-TNF therapy should not be
initiated in patients with New York Heart Association
(NYHA) Grade 3 or 4 cardiac failure (CF) (Table 2) <strong>and</strong>
should be used with caution in patients with mild (NYHA
Grade 1 or 2) CF. (Level Ib evidence, Grade of recommendation
A.)
Recommendation 33: anti-TNF therapy should be discontinued
if CF develops or worsens while on treatment.
(Level IV evidence, Grade of recommendation C.)
CF
Class
Class I (mild)
Class II (mild)
Class III
(moderate)
Class IV (severe)
Patient symptoms
No limitation of physical activity.
Ordinary physical activity does not
cause undue fatigue, palpitation or
dyspnoea (shortness of breath)
Slight limitation of physical activity.
Comfortable at rest, but ordinary
physical activity results in fatigue,
palpitation or dyspnoea
Marked limitation of physical activity.
Comfortable at rest, but less than
ordinary activity causes fatigue, palpitation
or dyspnoea
Unable to carry out any physical activity
without discomfort. Symptoms
of cardiac insufficiency at rest. If any
physical activity is undertaken, discomfort
is increased
Concern about the possible adverse effect of CF associated
with anti-TNF therapy stems from r<strong>and</strong>omized clinical
trials of TNF inhibitors when used as a potential
therapy for CF {R<strong>and</strong>omized Etanercept North American
Strategy to Study Antagonism of Cytokines
(RENAISSANCE) trial [159], Research into Etanercept
Cytokine Antagonism in Ventricular Dysfunction
(RECOVER) trial [160], Anti-TNF Therapy Against Congestive
Heart Failure (ATTACH) trial [161]}. Because of
the adverse experiences reported with etanercept <strong>and</strong>
infliximab with CF, no treatment trials of adalimumab in
CF have ever been attempted. Data from trials specifically
assessing the risk of CF with the use of anti-TNF therapy
at the licensed RA doses are inconclusive [162]. Interpretation
of the results from these trials is hampered by issues
relating to potential confounding by indication or contraindication
<strong>and</strong> by lack of control for confounding conditions
or severity of RA. Most studies <strong>and</strong> case reports,
however, have been generally reassuring with regard to
the risks of CF as a complication of anti-TNF therapy.
Post-marketing surveillance data gathered by the FDA
identified 47 cases of CF associated with etanercept or
infliximab reported to the AERS through to January 2002
[163]. Thirty-eight of these patients developed new-onset
CF <strong>and</strong> nine experienced an exacerbation of their CF.
Bernatsky et al. [164] reported a beneficial effect for
anti-TNF therapies on the rate of hospitalization for CF
(RR 0.5; 95% CI 0.2, 0.9). As the study only considered
hospitalized patients, no information is available for milder
degrees of CF. Cole et al. [165], however, found no difference
in the number of admissions for CF between veterans
with RA treated with TNF inhibitors (n = 103), RA
patients not treated with TNF inhibitors (n = 100) <strong>and</strong>
non-RA patients (n = 100). Listing et al. [166] also reported
no significant difference in the HR for developing
new-onset CF or worsening CF in patients treated with
anti-TNF therapy (n = 2757) when compared with those
treated with conventional DMARDs (n = 1491). Curtis
et al. [167] reported a non-significant increased risk of
CF in patients

RA <strong>guidelines</strong> on safety of anti-TNF therapies
exposed to TNF inhibitors. The rate of first CV event (not
clearly defined) was lower in patients treated with TNF
inhibitors (14.0/1000 pyrs at risk; 95% CI 5.7, 22.4) compared
with those not exposed (35.4/1000 pyrs at risk;
95% CI 16.5, 54.4).
Solomon et al. [172] undertook a nested case–control
study comparing 946 elderly patients (mean age 82 years)
with RA <strong>and</strong> a history of hospitalization with MI or stroke
with age- <strong>and</strong> gender-matched controls without a CV
event. MTX monotherapy was used as the reference
group. There was no beneficial or detrimental affect of
biologic use (with <strong>and</strong> without MTX) compared with MTX
monotherapy. The use of biologics was associated with a
decreased risk of MI or stroke when compared with the
use of other immunosuppressant agents (AZA, ciclosporin
<strong>and</strong> LEF) <strong>and</strong> glucocorticoids without concomitant use
of MTX.
Carmona et al. [20] reported on two cohorts of patients;
the first treated with TNF inhibitors (n = 4459) <strong>and</strong> the
second not treated with TNF inhibitors (n = 789). There
were significant reductions in the risk of ischaemic heart
disease, CF <strong>and</strong> stroke in those treated with TNF inhibitors.
The st<strong>and</strong>ardized mortality ratios for CV disease
were not significantly different between the groups
(mortality rate ratio 0.58; 95% CI 0.24, 1.41).
Haematological complications
Recommendation 34: for all patients on anti-TNF therapy
including those not on concomitant DMARDs, full blood
counts should be monitored regularly. (Level III evidence,
Grade of recommendation B.)
Cases of haematological abnormality have continued to
be reported in patients treated with each of the three
first-generation anti-TNF agents but in many cases there
have been other co-prescribed medications that may
have been responsible for these changes. One case
series reports neutropenia (

Tina Ding et al.
7 therapeutic terminations <strong>and</strong> 1 miscarriage (with malformation
secondary to trisomy 18). There were 4 live births,
3 miscarriages, 1 therapeutic termination <strong>and</strong> 1 malformation
(intestinal malrotation) among 20 cases exposed to
infliximab <strong>and</strong> 1 therapeutic termination with no miscarriages
among 5 patients exposed to adalimumab.
Outcome data from a post-marketing safety database
were reported by Katz et al. [180] for 96 of 131 patients
with Crohn’s disease or RA who had exposure to infliximab
during pregnancy. Sixty-seven per cent of the pregnancies
resulted in live births, 15% in spontaneous
abortions <strong>and</strong> 19% in terminations. Among the live
births there were five cases of varying fetal complications.
A recent abstract from the <strong>BSR</strong>BR reported on 99 pregnancies
<strong>and</strong> suggested a possible increase in miscarriage
rate in pregnancies conceived while taking anti-TNF.
However, other confounders (other DMARDs/unplanned
pregnancy factors) may also have played a role in these
miscarriages [181]. TNF is important during early pregnancy,
hence the possible explanation for the effect of
anti-TNF therapy in increasing the risk of miscarriage
[182, 183].
In addition to these reports of pregnancy complications
<strong>and</strong> congenital abnormalities associated with anti-TNF
therapy, there are, however, reports of successful pregnancy
outcomes despite treatment with anti-TNF. Rosner
et al. in 2007 [184] reported on four women with severe
intractable <strong>arthritis</strong> who successfully conceived <strong>and</strong> maintained
their pregnancies to full term while under continuous
anti-TNF therapy. In all cases, the newborn babies
were entirely well <strong>and</strong> remained so after 6 months to
2 years of follow-up. A recent publication by Berthelot
et al. [185] is also reassuring, with a good reported outcome
for 15 pregnancies in women treated with anti-TNF
drugs. This publication also included a literature review
[185].
Most other reports on anti-TNF agents <strong>and</strong> pregnancy
to date have focused on inflammatory bowel disease but
at least two ongoing registries for rheumatic disease patients
also suggest their apparent safety [186, 187]. The
number of reported pregnancies associated with anti-TNF
treatment now exceeds 300, but only 29 women were
treated during their whole pregnancy. The rate of congenital
malformations observed so far for women exposed to
anti-TNF only at the time of conception appear reassuring
when comparison is made with the general population.
However, there are too few reports of exposure during
pregnancy to allow any firm conclusion about the safety
of TNF blockers in pregnancy. Adalimumab <strong>and</strong> infliximab
are IgG antibodies <strong>and</strong> as such cannot cross the placenta
in the first trimester [188]. Therefore, it would appear to be
entirely safe to administer them to the mother during this
time [189]. Long-term follow-up of children born to parents
exposed to anti-TNF at conception or through pregnancy
to identify any potential minor forms of VACTERL
association, overlooked at birth, would be desirable.
Accordingly, anti-TNF drugs are still classified by the
FDA as ‘pregnancy risk category B’, as human studies
are still insufficient to guarantee their safety for the
fetus. No adverse effects have been observed in animal
pregnancies thus far.
Fathering children
There remains no data to provide guidance on the safety
of fathering children while taking anti-TNF therapy or on
how long before conception patients should be recommended
to discontinue anti-TNF therapy. Viktil et al.
[190] suggests that there were no obvious problems for
the 28 fathers taking a biologic drug.
Lactation
There remains insufficient data to address the issue of the
safety of each of the anti-TNF drugs for infants during
lactation. Ostensen <strong>and</strong> Eigenmann [191] report an etanercept
concentration in breast milk of 75 ng/ml in a lactating
mother; a level that was 4% of the maternal serum
concentration (2058 ng/ml). However, it is highly unlikely
that oral administration of anti-TNF therapy via breast milk
is likely to have any effect on the child [192].
Interstitial lung disease
Recommendation 38: patients with pre-existing interstitial
lung disease (ILD) should have monitoring of their lung
function if treated with anti-TNF therapies, <strong>and</strong> consideration
should be given to stopping anti-TNF therapy in patients
with worsening, or new features of ILD. (Level III
evidence, Grade of recommendation B.)
The 2005 <strong>BSR</strong> <strong>guidelines</strong> made no recommendations
on anti-TNF therapy <strong>and</strong> pre-existing lung disease. Data
have since emerged on ILD in particular, but also on bronchiectasis
(as reported under ‘Infection’ section).
There are reported cases of ILD associated with each of
the first-generation anti-TNF therapies. The majority of
cases have been reported in patients with RA but such
patients are known to have an increased risk of ILD, <strong>and</strong>
confounding by indication may therefore be a factor in
these cases.
Ramos-Casals et al. [147] analysed autoimmune diseases
reported to have been induced by anti-TNF therapies
(233 cases in 226 patients). Twenty-four cases of ILD
associated with anti-TNF therapy were reported within
this group <strong>and</strong> further details on the cases highlighted
the poor prognosis of ILD in these patients, despite cessation
of anti-TNF therapy. There was a reported outcome
for only 19 cases but over half showed no resolution of
their ILD after stopping anti-TNF therapy <strong>and</strong> one-third of
the patients died; most of these patients had known pulmonary
disease prior to commencing anti-TNF treatment.
The possible adjuvant role of concomitant MTX therapy,
however, may also have been a factor in these cases as,
overall, half the patients who developed ILD had also
received MTX.
Dixon et al. [193] have presented data from the <strong>BSR</strong>BR
showing that baseline ILD is a strong predictor of
increased all-cause mortality in RA patients (mortality
OR 2.12; 95% CI 1.52, 2.95), but that this risk is irrespective
of the use or not of anti-TNF treatment.
14 www.rheumatology.oxfordjournals.org

RA <strong>guidelines</strong> on safety of anti-TNF therapies
It is important to note that data thus far indicate that
ILD can be severe <strong>and</strong> life threatening in RA patients, regardless
of the treatment received. Some data suggest
that anti-TNF therapy may worsen both all-cause <strong>and</strong>
ILD-specific mortality in RA patients with baseline ILD but
more data are required before any firm conclusions can be
drawn on the risks specifically linked to anti-TNF therapies.
Psoriasis
Recommendation 39: if psoriasis develops on anti-TNF
therapy patients should receive conventional psoriasis
treatment <strong>and</strong> consideration should be given to stopping
anti-TNF therapy if the skin lesions persist despite
specialist dermatology treatment <strong>and</strong> advice or are particularly
severe. (Level III evidence, Grade of recommendation
B.)
A significant number of reports of psoriasis developing
in patients treated with all three first-generation anti-TNF
agents have emerged in recent years, despite NICE now
having approved the use of anti-TNF therapies for the
treatment of psoriasis [194–196].
Harrison et al. [197] reported data from the <strong>BSR</strong>BR suggesting
the incidence of psoriasis is increased in patients
treated with anti-TNF therapy. A total of 9826 anti-TNFtreated
<strong>and</strong> 2880 DMARD-treated RA patients with
new-onset psoriasis as an adverse event were recorded
on the <strong>BSR</strong>BR. Twenty-five incident cases were reported
in patients receiving anti-TNF therapy <strong>and</strong> none in the
comparison cohort. The crude incidence rate of psoriasis
in those treated with anti-TNF therapy was elevated at
1.04 (95% CI 0.67, 1.54) per 1000 pyrs compared with
the rate of 0 (upper 97.5% CI 0.71) per 1000 pyrs in the
DMARD-treated patients. Adalimumab-treated patients
had a significantly higher rate of incident psoriasis compared
with etanercept-treated (IRR 4.6; 95% CI 1.7, 12.1)
<strong>and</strong> infliximab-treated (IRR 3.5; 95% CI 1.3, 9.3) patients.
Thirteen patients developed psoriasis within the first
6 months of anti-TNF therapy; eight of these patients
were receiving adalimumab, three had had infliximab
<strong>and</strong> two had had etanercept. Of these 13 patients,
4 stopped anti-TNF therapy due to their psoriasis <strong>and</strong> reported
improvement in their psoriasis after stopping treatment.
No information on the course of skin disease in
patients who did not stop treatment due to psoriasis
was available.
In a literature review, Collamer et al. [198] reported on
the incidence of new or worsening psoriasis in patients
treated with TNF inhibitors. Nearly 50% of patients were
treated for RA <strong>and</strong> similar rates of occurrence of psoriasis
were reported for each of the anti-TNF therapies (infliximab
37%, etanercept 31%, adalimumab 32%). Newonset
psoriasis was found to occur at any time after
initiation of TNF antagonist therapy, to often present
with an uncommon morphology, <strong>and</strong> to usually respond
to psoriasis treatments. Seventy-nine per cent of patients
were able to continue with anti-TNF therapy (the original
agent or an alternative).
Ko et al. [199] undertook a further Ovid MEDLINE
<strong>and</strong> PubMed literature search for cases of psoriasis in
patients receiving TNF-blocking agents published between
January 1990 <strong>and</strong> September 2007. Several reports
of new-onset <strong>and</strong> worsening psoriasis in association with
TNF blockade therapy were found. A series of 127 cases
(from 1990 to 2007; 70 on infliximab, 35 on etanercept <strong>and</strong>
22 on adalimumab) showed a prevalence of psoriasis in
anti-TNF-treated patients ranging from 0.6 to 5.3%, with a
range of time to onset after starting treatment of a few
days to 4 years. Among these 127 reports there was a
disproportionately higher representation of the primary
conditions of AS (in 16.1% of the cases) <strong>and</strong> Behçet’s
disease (in 1.5%). While psoriasis classically presents as
thick, erythematous plaques with an adherent silvery
scale on the extensor surfaces of extremities (with
>80% of psoriasis patients having plaque type), psoriasis
occurring during TNF blockade has been mostly reported
as the pustular type occurring on the palms <strong>and</strong> soles
(pustular, palmer/plantar type in >40% of the cases, compared
with plaque-type psoriasis in just 33% of the
cases). Various treatment approaches were employed
for the psoriasis with mixed results. Topical corticosteroids
were used in 55% of cases but when used alone, led
to resolution of the psoriasis in only 25%. Thirteen patients
had their TNF inhibitor switched to an alternative
but this led to resolution of the psoriasis in just 15% of
the cases. Stopping TNF blockade alone improved psoriasis
in 50% of patients. Stopping TNF inhibitors <strong>and</strong> starting
systemic therapy was the most effective treatment
approach from this review, resulting in resolution of the
psoriasis in >64% of the cases.
Uveitis
Recommendation 40: if patients develop uveitis while
on anti-TNF, a trial of an alternative anti-TNF agent
could be considered. (Level III evidence, Grade of recommendation
B.)
Recommendation 41: anti-TNF should be used with
caution in patients with a history of previous uveitis <strong>and</strong>
the RRs of the available anti-TNF agents should be reviewed
prior to selecting which treatment to use. (Level
III evidence, Grade of recommendation B.)
There are now several case reports of uveitis developing
in patients treated with anti-TNF therapy despite the
fact that anti-TNF therapy has also been reported to successfully
treat patients with resistant uveitis. Most cases
of anti-TNF therapy-associated uveitis have been reported
with etanercept [200, 201]. Most reports of successful
treatment of resistant uveitis with anti-TNF
therapy have been with infliximab [202–205].
Lim et al. [206] reported on a registry-based study that
collected data on spontaneous reports of uveitis during
the period 1 January 1998 through 1 January 2006.
Forty-three cases of uveitis were recorded in patients
who were receiving anti-TNF therapy. Seventeen patients
were known to have diseases that are associated with an
increased risk of uveitis (i.e. AS, Crohn’s disease or JIA).
RA patients were included in the remaining 26 cases in
which no known condition predisposing to uveitis was
identified; the number taking etanercept, infliximab or
www.rheumatology.oxfordjournals.org 15

Tina Ding et al.
adalimumab in this group was 20, 4 <strong>and</strong> 2, respectively.
The authors highlighted that etanercept therapy was
associated with a significantly greater number of reported
uveitis cases in comparison with infliximab (P < 0.001) but
also conceded that there were several limitations to this
study; in particular, there was limited information on the
clinical diagnoses available to the authors [206]. More
data are therefore required on uveitis associated with
each of the available anti-TNF agents before further conclusions
can be drawn on the RR associated with individual
agents.
Summary <strong>and</strong> conclusions
Inhibitors of TNF-a represent important treatment
advances in RA. The use of anti-TNF drugs in RA has
been increasing significantly over the past few years <strong>and</strong>
information on the safety of these agents continues to be
collected. Data are particularly scarce for the secondgeneration
anti-TNF agents but there are no data thus far
to suggest that safety guidance for these agents should
differ from those provided for the first-generation agents.
Data from post-marketing surveillance, observational
studies, individual r<strong>and</strong>omized trials <strong>and</strong> from national
registries (in a number of different countries) have been
used as sources of evidence for these <strong>guidelines</strong> on
anti-TNF safety in RA. Many questions remain unanswered,
however, <strong>and</strong> ongoing studies <strong>and</strong> data collection
will be essential to further clarify the answers to these
questions.
Acknowledgements
Disclosure statement: J.L. has no personal conflicts of
interest. Her department has received support from all
the companies for educational meetings <strong>and</strong> specifically
from Wyeth for its patient education programme <strong>and</strong> from
Abbott for anti-CCP testing for its patients <strong>and</strong> for the
purchase of an ultrasound machine. A.B.’s organization
has received educational grants from MSD (Schering-
Plough), Abbott, Pfizer (Wyeth) <strong>and</strong> UCB. R.A. has previously
sat on an advisory board for Roche <strong>and</strong> received
honoraria for talks at symposia sponsored by Wyeth <strong>and</strong>
Abbott. She has received personal support from Wyeth
<strong>and</strong> Abbott to attend international educational meetings
<strong>and</strong> the Department of Rheumatology at St Helens
Hospital has received sponsorship from Wyeth, Abbott,
Roche, Bristol-Myers Squibb <strong>and</strong> Schering-Plough
Pharmaceuticals for support of clinical meetings. M.B.
has attended meetings organized by Schering Plough
<strong>and</strong> Wyeth, has been sponsored to attend international
meetings by Pfizer, Merck, Roche, Wyeth <strong>and</strong> Abbott,
<strong>and</strong> has accepted honoraria for educational meetings
from Merck, Wyeth, Abbott, Roche, UCB Celltech,
Mennarini <strong>and</strong> Pfizer. He has also departmental sponsorship
for software from Wyeth, Abbott <strong>and</strong> Roche. R.L. has
received financial support from Wyeth, Roche <strong>and</strong> Abbott
to attend ACR <strong>and</strong> EULAR meetings. S.W. has received
an unconditional education grant from Abbott as part of
the 3E project. A.O. has received support from (including
attendance at conferences), undertakes clinical trials <strong>and</strong>
acts as a consultant to Roche, Chugai, Schering-Plough/
MSD, Abbott, Wyeth, BMS, GSK, MerckSorono <strong>and</strong> UCB.
P.K. has received departmental support for service <strong>and</strong>
research from Sanofi-Aventis, Schering-Plough <strong>and</strong>
Wyeth <strong>and</strong> has received advisory fees, speaker fees <strong>and</strong>
unrestricted educational grants from Abbot, Bristol Myers
Squibb, Novartis, Roche, Schering-Plough, UCB <strong>and</strong>
Wyeth. K.G. sits on advisory boards for Schering-
Plough, UCB <strong>and</strong> Roche <strong>and</strong> has received honoraria for
talks at symposia sponsored by Wyeth, Abbott <strong>and</strong>
Roche. C.D. has received honoraria for talks at symposia
sponsored by Wyeth <strong>and</strong> Abbott. The Department of
Rheumatology at Derbyshire Royal Infirmary has received
sponsorship from Wyeth, Abbott <strong>and</strong> Schering-Plough
Pharmaceuticals for support of clinical meetings, <strong>and</strong> unrestricted
grants to support an US machine, an anti-TNF
audit clerk <strong>and</strong> research nurse. All other authors have
declared no conflicts of interest.
References
1 British Society for Rheumatology. Guidelines for
prescribing TNF-a blockers in adults with <strong>rheumatoid</strong>
<strong>arthritis</strong>. 2001, http://www.rheumatology.org.uk/
<strong>guidelines</strong>/clinical<strong>guidelines</strong> (20 January 2010, date last
accessed).
2 Ledingham J, Deighton C. Update on the British Society
for Rheumatology <strong>guidelines</strong> for prescribing TNF-a
blockers in adults with <strong>rheumatoid</strong> <strong>arthritis</strong> (update of
previous <strong>guidelines</strong> of April 2001). Rheumatology 2005;44:
157–63.
3 Furst DE, Keystone EC, Kirkham B et al. Updated consensus
statement on biological agents for the treatment of
rheumatic diseases, 2008. Ann Rheum Dis 2008;67(Suppl.
3):iii2–25.
4 Saag KG, Teng GG, Patkar NM et al. American College of
Rheumatology 2008 recommendations for the use of
nonbiologic <strong>and</strong> biologic disease-modifying antirheumatic
drugs in <strong>rheumatoid</strong> <strong>arthritis</strong>. Arthritis Rheum 2008;59:
762–84.
5 Doran MF, Crowson CS, Pond GR, O’Fallon WM,
Gabriel SE. Frequency of infection in patients with
<strong>rheumatoid</strong> <strong>arthritis</strong> compared with controls: a populationbased
study. Arthritis Rheum 2002;46:2287–93.
6 Gomez-Reino JJ, Carmona L, Valverde VR, Mola EM,
Montero MD. Treatment of <strong>rheumatoid</strong> <strong>arthritis</strong> with tumor
necrosis factor inhibitors may predispose to significant
increase in tuberculosis risk: a multicenter activesurveillance
report. Arthritis Rheum 2003;48:2122–7.
7 Wallis RS, Broder MS, Wong JY, Hanson ME,
Beenhouwer DO. Granulomatous infectious diseases
associated with tumor necrosis factor antagonists. Clin
Infect Dis 2004;38:1261–5.
8 St Clair EW, van der Heijde DM, Smolen JS et al.
Combination of infliximab <strong>and</strong> methotrexate therapy for
early <strong>rheumatoid</strong> <strong>arthritis</strong>: a r<strong>and</strong>omized, controlled trial.
Arthritis Rheum 2004;50:3432–43.
9 Breedveld FC, Weisman MH, Kavanaugh AF et al. The
PREMIER study: a multicenter, r<strong>and</strong>omized, double-blind
16 www.rheumatology.oxfordjournals.org

RA <strong>guidelines</strong> on safety of anti-TNF therapies
clinical trial of combination therapy with adalimumab plus
methotrexate versus methotrexate alone or adalimumab
alone in patients with early, aggressive <strong>rheumatoid</strong>
<strong>arthritis</strong> who had not had previous methotrexate
treatment. Arthritis Rheum 2006;54:26–37.
10 Curtis JR, Patkar N, Xie A et al. Risk of serious bacterial
infections among <strong>rheumatoid</strong> <strong>arthritis</strong> patients exposed to
tumor necrosis factor alpha antagonists. Arthritis Rheum
2007;56:1125–33.
11 Bernatsky S, Hudson M, Suissa S. Anti-rheumatic drug
use <strong>and</strong> risk of serious infections in <strong>rheumatoid</strong> <strong>arthritis</strong>.
Rheumatology 2007;46:1157–60.
12 Salliot C, Gossec L, Ruyssen-Witr<strong>and</strong> A et al. Infections
during tumour necrosis factor-alpha blocker therapy
for rheumatic diseases in daily practice: a systematic
retrospective study of 709 patients. Rheumatology 2007;
46:327–34.
13 Westhovens R, Yocum D, Han J et al. The safety of
infliximab, combined with background treatments, among
patients with <strong>rheumatoid</strong> <strong>arthritis</strong> <strong>and</strong> various comorbidities:
a large, r<strong>and</strong>omized, placebo-controlled trial. Arthritis
Rheum 2006;54:1075–86.
14 Weisman MH, Paulus HE, Burch FX et al. A
placebo-controlled, r<strong>and</strong>omized, double-blinded study
evaluating the safety of etanercept in patients with
<strong>rheumatoid</strong> <strong>arthritis</strong> <strong>and</strong> concomitant comorbid diseases.
Rheumatology 2007;46:1122–5.
15 Klareskog L, van der Heijde D, de Jager JP et al.
Therapeutic effect of the combination of etanercept <strong>and</strong>
methotrexate compared with each treatment alone in
patients with <strong>rheumatoid</strong> <strong>arthritis</strong>: double-blind
r<strong>and</strong>omised controlled trial. Lancet 2004;363:675–81.
16 Wolfe F, Caplan L, Michaud K. Treatment for <strong>rheumatoid</strong>
<strong>arthritis</strong> <strong>and</strong> the risk of hospitalization for pneumonia:
associations with prednisone, disease-modifying
antirheumatic drugs, <strong>and</strong> anti-tumor necrosis factor
therapy. Arthritis Rheum 2006;54:628–34.
17 Schneeweiss S, Setoguchi S, Weinblatt ME et al.
Anti-tumor necrosis factor alpha therapy <strong>and</strong> the risk
of serious bacterial infections in elderly patients with
<strong>rheumatoid</strong> <strong>arthritis</strong>. Arthritis Rheum 2007;56:1754–64.
18 Bongartz T, Sutton AJ, Sweeting MJ, Buchan I,
Matteson EL, Montori V. Anti-TNF antibody therapy in
<strong>rheumatoid</strong> <strong>arthritis</strong> <strong>and</strong> the risk of serious infections <strong>and</strong>
malignancies: systematic review <strong>and</strong> meta-analysis of rare
harmful effects in r<strong>and</strong>omized controlled trials. JAMA
2006;295:2275–85.
19 Listing J, Strangfeld A, Kary S et al. Infections in patients
with <strong>rheumatoid</strong> <strong>arthritis</strong> treated with biologic agents.
Arthritis Rheum 2005;52:3403–12.
20 Carmona L, Descalzo MA, Perez-Pampin E et al. All-cause
<strong>and</strong> cause-specific mortality in <strong>rheumatoid</strong> <strong>arthritis</strong> are not
greater than expected when treated with tumour necrosis
factor antagonists. Ann Rheum Dis 2007;66:880–5.
21 Dixon WG, Watson K, Lunt M, Hyrich KL, Silman AJ,
Symmons DP. Rates of serious infection, including
site-specific <strong>and</strong> bacterial intracellular infection, in
<strong>rheumatoid</strong> <strong>arthritis</strong> patients receiving anti-tumor
necrosis factor therapy: results from the British Society
for Rheumatology Biologics Register. Arthritis Rheum
2006;54:2368–76.
22 Dixon WG, Symmons DP, Lunt M, Watson KD, Hyrich KL,
Silman AJ. Serious infection following anti-tumor necrosis
factor alpha therapy in patients with <strong>rheumatoid</strong> <strong>arthritis</strong>:
lessons from interpreting data from observational studies.
Arthritis Rheum 2007;56:2896–904.
23 Galloway JB, Hyrich KL, Mercer LK et al. Risk of septic
<strong>arthritis</strong> in patients with <strong>rheumatoid</strong> <strong>arthritis</strong> treated with
anti-TNF therapy: results from the <strong>BSR</strong> Biologics Register
(<strong>BSR</strong>BR). Arthritis Rheum 2009;60(10 Suppl.):S771.
24 Ding T, Deighton C. Complications of anti-TNF therapies.
Future Rheumatol 2007;2:587–97.
25 Kling A, Mjorndal T, Rantapaa-Dahlqvist S. Sepsis as a
possible adverse drug reaction in patients with <strong>rheumatoid</strong>
<strong>arthritis</strong> treated with TNF-alpha antagonists. J Clin
Rheumatol 2004;10:119–22.
26 Lieberman-Maran L, Orzano IM, Passero MA, Lally EV.
Bronchiectasis in <strong>rheumatoid</strong> <strong>arthritis</strong>: report of four cases
<strong>and</strong> a review of the literature–implications for management
with biologic response modifiers. Semin Arthritis Rheum
2006;35:379–87.
27 Seong SS, Choi CB, Woo JH et al. Incidence of tuberculosis
in Korean patients with <strong>rheumatoid</strong> <strong>arthritis</strong> (RA):
effects of RA itself <strong>and</strong> of tumor necrosis factor blockers.
J Rheumatol 2007;34:706–11.
28 Watanabe A. Infliximab <strong>and</strong> tuberculosis. Bunshi
Ryumachi [Molecular Rheumatology] 2005;2:272–7.
29 Fonseca JE, Canhao H, Silva C et al. [Tuberculosis in
rheumatic patients treated with tumour necrosis factor
alpha antagonists: the Portuguese experience]. Acta
Reumatol Port 2006;31:247–53.
30 Wolfe F, Michaud K, Anderson J, Urbansky K.
Tuberculosis infection in patients with <strong>rheumatoid</strong> <strong>arthritis</strong>
<strong>and</strong> the effect of infliximab therapy. Arthritis Rheum 2004;
50:372–9.
31 Askling J, Fored CM, Br<strong>and</strong>t L et al. Risk <strong>and</strong> case
characteristics of tuberculosis in <strong>rheumatoid</strong> <strong>arthritis</strong>
associated with tumor necrosis factor antagonists in
Sweden. Arthritis Rheum 2005;52:1986–92.
32 Hamilton CD. Tuberculosis in the cytokine era: what
rheumatologists need to know. Arthritis Rheum 2003;48:
2085–91.
33 Tubach F, Salmon D, Ravaud P et al. Risk of tuberculosis
is higher with anti-tumor necrosis factor monoclonal
antibody therapy than with soluble tumor necrosis factor
receptor therapy: the three-year prospective French
Research Axed on Tolerance of Biotherapies registry.
Arthritis Rheum 2009;60:1884–94.
34 Dixon WG, Hyrich KL, Watson KD et al. Drug-specific risk
of tuberculosis in patients with <strong>rheumatoid</strong> <strong>arthritis</strong> treated
with anti-TNF therapy: results from the British Society for
Rheumatology Biologics Register (<strong>BSR</strong>BR). Ann Rheum
Dis 2009;69:522–8.
35 Keystone E, Mason D, Combe B. The anti-TNF
certolizumab pegol in combination with methotrexate is
significantly more effective than methotrexate alone in the
treatment of patients with active <strong>rheumatoid</strong> <strong>arthritis</strong>:
preliminary results from the RAPID 1 study. Ann Rheum
Dis 2007;66(Suppl. 2):55.
36 Smolen J, Brezezicki J, Mason D, Kavanaugh A. Efficacy
<strong>and</strong> safety of certolizumab pegol in combination with
methotrexate (MTX) in patients with active <strong>rheumatoid</strong>
<strong>arthritis</strong> despite MTX therapy: results from the RAPID
2 study. Ann Rheum Dis 2007;66(Suppl. 2):187.
37 Gardam MA, Keystone EC, Menzies R et al. Anti-tumour
necrosis factor agents <strong>and</strong> tuberculosis risk: mechanisms
www.rheumatology.oxfordjournals.org 17

Tina Ding et al.
of action <strong>and</strong> clinical management. Lancet Infect Dis 2003;
3:148–55.
38 Carmona L, Gomez-Reino JJ, Rodriguez-Valverde V et al.
Effectiveness of recommendations to prevent reactivation
of latent tuberculosis infection in patients treated with
tumor necrosis factor antagonists. Arthritis Rheum 2005;
52:1766–72.
39 Gomez-Reino JJ, Carmona L, Angel Descalzo M. Risk of
tuberculosis in patients treated with tumor necrosis factor
antagonists due to incomplete prevention of reactivation
of latent infection. Arthritis Rheum 2007;57:756–61.
40 Lalvani A, Richeldi L, Kunst H. Interferon gamma assays
for tuberculosis. Lancet Infect Dis 2005;5:322–4, author
reply 325–7.
41 Winthrop KL, Chang E, Yamashita S, Iademarco MF,
LoBue PA. Nontuberculous mycobacteria infections <strong>and</strong>
anti-tumor necrosis factor-alpha therapy. Emerg Infect Dis
2009;15:1556–61.
42 Salvana EM, Cooper GS, Salata RA. Mycobacterium other
than tuberculosis (MOTT) infection: an emerging disease
in infliximab-treated patients. J Infect 2007;55:484–7.
43 Mufti AH, Toye BW, McKendry RR, Angel JB.
Mycobacterium abscessus infection after use of tumor
necrosis factor alpha inhibitor therapy: case report <strong>and</strong>
review of infectious complications associated with tumor
necrosis factor alpha inhibitor use. Diagn Microbiol Infect
Dis 2005;53:233–8.
44 Thomas JE, Taoka CR, Gibbs BT, Fraser SL. Fatal pulmonary
Mycobacterium abscessus infection in a patient
using etanercept. Hawaii Med J 2006;65:12–5.
45 Okubo H, Iwamoto M, Yoshio T et al. Rapidly aggravated
Mycobacterium avium infection in a patient with <strong>rheumatoid</strong>
<strong>arthritis</strong> treated with infliximab. Mod Rheumatol 2005;
15:62–4.
46 Phillips K, Husni ME, Karlson EW, Coblyn JS. Experience
with etanercept in an academic medical center: are
infection rates increased? Arthritis Rheum 2002;47:17–21.
47 Diaz F, Urkijo JC, Mendoza F et al. Mycobacterium
chelonae infection associated with adalimumab therapy.
Sc<strong>and</strong> J Rheumatol 2008;37:159–60.
48 Adenis-Lamarre E, Kostrzewa E, Texier-Maugein J,
Doutre MS. [Cutaneous infection due to Mycobacterium
chelonae during anti-TNF therapy.]. Ann Dermatol
Venereol 2009;136:811–4.
49 Boulman N, Rozenbaum M, Slobodin G, Rosner I.
Mycobacterium fortuitum infection complicating infliximab
therapy in <strong>rheumatoid</strong> <strong>arthritis</strong>. Clin Exp Rheumatol 2006;
24:723.
50 Shehan JM, Sarma DP. Mycobacterium mucogenicum:
report of a skin infection associated with etanercept.
Dermatol Online J 2008;14:5.
51 van Ingen J, Boeree M, Janssen M et al. Pulmonary
Mycobacterium szulgai infection <strong>and</strong> treatment in a patient
receiving anti-tumor necrosis factor therapy. Nat Clin
Pract Rheumatol 2007;3:414–9.
52 Maimon N, Brunton J, Chan AK, Marras TK. Fatal
pulmonary Mycobacterium xenopi in a patient with
<strong>rheumatoid</strong> <strong>arthritis</strong> receiving etanercept. Thorax 2007;62:
739–40.
53 Arthritis Research Campaign. http://cks.library.nhs.
uk/patient_information_leaflet/etanercept_arc#
(20 January 2010, date last accessed).
54 Arthritis Research Campaign. http://cks.library.nhs.
uk/patient_information_leaflet/infliximab_arc# (20 January
2010, date last accessed).
55 Arthritis Research Campaign. http://cks.library.nhs.
uk/patient_information_leaflet/adalimumab_arc#.
(20 January 2010, date last accessed).
56 Slifman NR, Gershon SK, Lee JH, Edwards ET, Braun MM.
Listeria monocytogenes infection as a complication of
treatment with tumor necrosis factor alpha-neutralizing
agents. Arthritis Rheum 2003;48:319–24.
57 Makkuni D, Kent R, Watts R, Clunie G. Two cases of
serious food-borne infection in patients treated with
anti-TNF-alpha Are we doing enough to reduce the risk?
Rheumatology 2006;45:237–8.
58 Pagliano P, Attanasio V, Fusco U, Mohamed DA, Rossi M,
Faella FS. Does etanercept monotherapy enhance the risk
of Listeria monocytogenes meningitis? Ann Rheum Dis
2004;63:462–3.
59 Bowie VL, Snella KA, Gopalachar AS, Bharadwaj P.
Listeria meningitis associated with infliximab. Ann
Pharmacother 2004;38:58–61.
60 Katsarolis I, Tsiodras S, Panagopoulos P et al. Septic
<strong>arthritis</strong> due to Salmonella enteritidis associated with
infliximab use. Sc<strong>and</strong> J Infect Dis 2005;37:304–5.
61 Rijkeboer A, Voskuyl A, Van Agtmael M. Fatal Salmonella
enteritidis septicaemia in a <strong>rheumatoid</strong> <strong>arthritis</strong> patient
treated with a TNF-alpha antagonist. Sc<strong>and</strong> J Infect Dis
2007;39:80–3.
62 Tsiodras S, Samonis G, Boumpas DT, Kontoyiannis DP.
Fungal infections complicating tumor necrosis factor
alpha blockade therapy. Mayo Clin Proc 2008;83:181–94.
63 Kaur N, Mahl TC. Pneumocystis jiroveci (carinii) pneumonia
after infliximab therapy: a review of 84 cases. Dig Dis
Sci 2007;52:1481–4.
64 Tai TL, O’Rourke KP, McWeeney M, Burke CM,
Sheehan K, Barry M. Pneumocystis carinii pneumonia
following a second infusion of infliximab. Rheumatology
2002;41:951–2.
65 Velayos FS, S<strong>and</strong>born WJ. Pneumocystis carinii
pneumonia during maintenance anti-tumor necrosis
factor-alpha therapy with infliximab for Crohn’s disease.
Inflamm Bowel Dis 2004;10:657–60.
66 Mori S, Tomita Y, Horikawa T, Cho I, Sugimoto M. Delayed
spinal infection after laminectomy in a patient with <strong>rheumatoid</strong>
<strong>arthritis</strong> interruptedly exposed to anti-tumor necrosis
factor alpha agents. Clin Rheumatol 2008;27:937–9.
67 Eguchi K. [Efficacy <strong>and</strong> adverse events of etanercept in
patients with <strong>rheumatoid</strong> <strong>arthritis</strong>: reports of postmarketing
surveillance in Japan]. Nippon Rinsho 2007;65:1259–66.
68 Kalyoncu U, Karadag O, Akdogan A et al. Pneumocystis
carinii pneumonia in a <strong>rheumatoid</strong> <strong>arthritis</strong> patient treated
with adalimumab. Sc<strong>and</strong> J Infect Dis 2007;39:475–8.
69 Rodriguez M, Fishman JA. Prevention of infection due
to Pneumocystis spp. in human immunodeficiency
virus-negative immunocompromised patients. Clin
Microbiol Rev 2004;17:770–82.
70 British Society for Rheumatology. Guidelines for
vaccinations in the immunocompromised person. 2002,
http://www.rheumatology.org.uk/includes/documents/
cm_docs/2009/v/vaccinations_in_the_immunocom
promised_person.pdf (20 January 2010, date last
accessed).
18 www.rheumatology.oxfordjournals.org

RA <strong>guidelines</strong> on safety of anti-TNF therapies
71 Choi HJ, Kim MY, Kim HO, Park YM. An atypical varicella
exanthem associated with the use of infliximab. Int J
Dermatol 2006;45:999–1000.
72 Vonkeman H, ten Napel C, Rasker H, van de Laar M.
Disseminated primary varicella infection during infliximab
treatment. J Rheumatol 2004;31:2517–8.
73 Tresch S, Trueb RM, Kamarachev J, French LE,
Hofbauer GF. Disseminated herpes zoster mimicking
<strong>rheumatoid</strong> vasculitis in a <strong>rheumatoid</strong> <strong>arthritis</strong> patient on
etanercept. Dermatology 2009;219:347–9.
74 Lee DH, Kim HS, Song YW, Cho KH. Development of
varicella during adalimumab therapy. J Eur Acad Dermatol
Venereol 2007;21:687–8.
75 Strangfeld A, Listing J, Herzer P et al. Risk of herpes
zoster in patients with <strong>rheumatoid</strong> <strong>arthritis</strong> treated with
anti-TNF-alpha agents. JAMA 2009;301:737–44.
76 Smitten AL, Choi HK, Hochberg MC et al. The risk of
herpes zoster in patients with <strong>rheumatoid</strong> <strong>arthritis</strong> in the
United States <strong>and</strong> the United Kingdom. Arthritis Rheum
2007;57:1431–8.
77 McDonald JR, Zeringue AL, Caplan L et al. Herpes zoster
risk factors in a national cohort of veterans with <strong>rheumatoid</strong>
<strong>arthritis</strong>. Clin Infect Dis 2009;48:1364–71.
78 Wolfe F, Michaud K, Chakravarty EF. Rates <strong>and</strong> predictors
of herpes zoster in patients with <strong>rheumatoid</strong> <strong>arthritis</strong> <strong>and</strong>
non-inflammatory musculoskeletal disorders.
Rheumatology 2006;45:1370–5.
79 Zaia JA, Levin MJ, Preblud SR et al. Evaluation of
varicella-zoster immune globulin: protection of immunosuppressed
children after household exposure to varicella.
J Infect Dis 1983;147:737–43.
80 Calabrese LH, Zein N, Vassilopoulos D. Safety of
antitumour necrosis factor (anti-TNF) therapy in patients
with chronic viral infections: hepatitis C, hepatitis B, <strong>and</strong>
HIV infection. Ann Rheum Dis 2004;63(Suppl. 2):ii18–24.
81 Oniankitan O, Duvoux C, Challine D et al. Infliximab therapy
for rheumatic diseases in patients with chronic
hepatitis B or C. J Rheumatol 2004;31:107–9.
82 Roux CH, Brocq O, Breuil V, Albert C, Euller-Ziegler L.
Safety of anti-TNF-alpha therapy in <strong>rheumatoid</strong> <strong>arthritis</strong>
<strong>and</strong> spondylarthropathies with concurrent B or C chronic
hepatitis. Rheumatology 2006;45:1294–7.
83 Kaur PP, Chan VC, Berney SN. Histological evaluation of
liver in two <strong>rheumatoid</strong> <strong>arthritis</strong> patients with chronic
hepatitis B <strong>and</strong> C treated with TNF-alpha blockade: case
reports. Clin Rheumatol 2008;27:1069–71.
84 Ostuni P, Botsios C, Punzi L, Sfriso P, Todesco S.
Hepatitis B reactivation in a chronic hepatitis B surface
antigen carrier with <strong>rheumatoid</strong> <strong>arthritis</strong> treated with
infliximab <strong>and</strong> low dose methotrexate. Ann Rheum Dis
2003;62:686–7.
85 Wendling D, Auge B, Bettinger D et al. Reactivation of a
latent precore mutant hepatitis B virus related chronic
hepatitis during infliximab treatment for severe spondyloarthropathy.
Ann Rheum Dis 2005;64:788–9.
86 Sakellariou GT, Chatzigiannis I. Long-term anti-TNFalpha
therapy for ankylosing spondylitis in two patients with
chronic HBV infection. Clin Rheumatol 2007;26:950–2.
87 Chung SJ, Kim JK, Park MC, Park YB, Lee SK.
Reactivation of hepatitis B viral infection in inactive HBsAg
carriers following anti-tumor necrosis factor-alpha therapy.
J Rheumatol 2009;36:2416–20.
88 Li S, Kaur PP, Chan V, Berney S. Use of tumor necrosis
factor-alpha (TNF-alpha) antagonists infliximab, etanercept,
<strong>and</strong> adalimumab in patients with concurrent
<strong>rheumatoid</strong> <strong>arthritis</strong> <strong>and</strong> hepatitis B or hepatitis C: a
retrospective record review of 11 patients. Clin
Rheumatol 2009;28:787–91.
89 Robinson H, Walker-Bone K. Anti-TNF-alpha therapy for
<strong>rheumatoid</strong> <strong>arthritis</strong> among patients with chronic
hepatitis B infection. Rheumatology 2009;48:448–50.
90 Charpin C, Guis S, Colson P et al. Safety of TNF
blocking agents in rheumatic patients with serology
suggesting past-hepatitis B state: results from a cohort
of 21 patients. Arthritis Res Ther 2009;11:R179.
91 Raftery G, Griffiths B, Kay L, Kane D. Chronic viral
hepatitis <strong>and</strong> TNF-alpha blockade. Rheumatology 2007;
46:1381; author reply 1381–2.
92 Peterson JR, Hsu FC, Simkin PA, Wener MH. Effect of
tumour necrosis factor alpha antagonists on serum
transaminases <strong>and</strong> viraemia in patients with <strong>rheumatoid</strong>
<strong>arthritis</strong> <strong>and</strong> chronic hepatitis C infection. Ann Rheum
Dis 2003;62:1078–82.
93 Zein NN. Etanercept as an adjuvant to interferon <strong>and</strong>
ribavirin in treatment-naive patients with chronic hepatitis
C virus infection: a phase 2 r<strong>and</strong>omized,
double-blind, placebo-controlled study. J Hepatol 2005;
42:315–22.
94 Linardaki G, Katsarou O, Ioannidou P, Karafoulidou A,
Boki K. Effective etanercept treatment for psoriatic
<strong>arthritis</strong> complicating concomitant human immunodeficiency
virus <strong>and</strong> hepatitis C virus infection.
J Rheumatol 2007;34:1353–5.
95 Pitarch G, Sanchez-Carazo JL, Mahiques L,
Perez-Ferriols MA, Fortea JM. Treatment of psoriasis
with adalimumab. Clin Exp Dermatol 2007;32:18–22.
96 Bellisai F, Giannitti C, Donvito A, Galeazzi M.
Combination therapy with cyclosporine A <strong>and</strong>
anti-TNF-alpha agents in the treatment of <strong>rheumatoid</strong>
<strong>arthritis</strong> <strong>and</strong> concomitant hepatitis C virus infection. Clin
Rheumatol 2007;26:1127–9.
97 Aslanidis S, Vassiliadis T, Pyrpasopoulou A,
Douloumpakas I, Zamboulis C. Inhibition of TNFalpha
does not induce viral reactivation in patients with chronic
hepatitis C infection: two cases. Clin Rheumatol 2007;
26:261–4.
98 Parke FA, Reveille JD. Anti-tumor necrosis factor agents
for <strong>rheumatoid</strong> <strong>arthritis</strong> in the setting of chronic hepatitis
C infection. Arthritis Rheum 2004;51:800–4.
99 Ferri C, Ferraccioli G, Ferrari D et al. Safety of anti-tumor
necrosis factor-alpha therapy in patients with <strong>rheumatoid</strong>
<strong>arthritis</strong> <strong>and</strong> chronic hepatitis C virus infection.
J Rheumatol 2008;35:1944–9.
100 Pritchard C. Etanercept <strong>and</strong> hepatitis C. J Clin
Rheumatol 1999;5:179.
101 Cepeda EJ, Williams FM, Ishimori ML, Weisman MH,
Reveille JD. The use of anti-tumour necrosis factor
therapy in HIV-positive individuals with rheumatic
disease. Ann Rheum Dis 2008;67:710–2.
102 Gaylis N. Infliximab in the treatment of an HIV positive
patient with Reiter’s syndrome. J Rheumatol 2003;30:
407–11.
103 Bartke U, Venten I, Kreuter A, Gubbay S, Altmeyer P,
Brockmeyer NH. Human immunodeficiency
www.rheumatology.oxfordjournals.org 19

Tina Ding et al.
virus-associated psoriasis <strong>and</strong> psoriatic <strong>arthritis</strong> treated
with infliximab. Br J Dermatol 2004;150:784–6.
104 Sellam J, Bouvard B, Masson C et al. Use of infliximab to
treat psoriatic <strong>arthritis</strong> in HIV-positive patients. Joint
Bone Spine 2007;74:197–200.
105 Aboulafia DM, Bundow D, Wilske K, Ochs UI. Etanercept
for the treatment of human immunodeficiency
virus-associated psoriatic <strong>arthritis</strong>. Mayo Clin Proc 2000;
75:1093–8.
106 den Broeder AA, Creemers MC, Fransen J et al. Risk
factors for surgical site infections <strong>and</strong> other complications
in elective surgery in patients with <strong>rheumatoid</strong>
<strong>arthritis</strong> with special attention for anti-tumor necrosis
factor: a large retrospective study. J Rheumatol 2007;34:
689–95.
107 Bibbo C, Goldberg JW. Infectious <strong>and</strong> healing complications
after elective orthopaedic foot <strong>and</strong> ankle surgery
during tumor necrosis factor-alpha inhibition therapy.
Foot Ankle Int 2004;25:331–5.
108 Giles JT, Bartlett SJ, Gelber AC et al. Tumor necrosis
factor inhibitor therapy <strong>and</strong> risk of serious postoperative
orthopedic infection in <strong>rheumatoid</strong> <strong>arthritis</strong>. Arthritis
Rheum 2006;55:333–7.
109 Ruyssen-Witr<strong>and</strong> A, Gossec L, Salliot C et al.
Complication rates of 127 surgical procedures performed
in rheumatic patients receiving tumor necrosis
factor alpha blockers. Clin Exp Rheumatol 2007;25:
430–6.
110 Dixon WG, Lunt M, Watson KD, Hyrich KL,
Symmons DP. Anti-TNF therapy <strong>and</strong> the risk of
serious post-operative infection: results from the <strong>BSR</strong>
Biologics register (<strong>BSR</strong>BR). Ann Rheum Dis 2007;
66(Suppl. II):118.
111 Pappas DA, Giles JT. Do antitumor necrosis factor
agents increase the risk of postoperative orthopedic
infections? Curr Opin Rheumatol 2008;20:450–6.
112 Abu-Shakra M, Press J, Varsano N et al. Specific
antibody response after influenza immunization in
systemic lupus erythematosus. J Rheumatol 2002;29:
2555–7.
113 Kaine JL, Kivitz AJ, Birbara C, Luo AY. Immune
responses following administration of influenza <strong>and</strong>
pneumococcal vaccines to patients with <strong>rheumatoid</strong>
<strong>arthritis</strong> receiving adalimumab. J Rheumatol 2007;34:
272–9.
114 Fomin I, Caspi D, Levy V et al. Vaccination against
influenza in <strong>rheumatoid</strong> <strong>arthritis</strong>: the effect of disease
modifying drugs, including TNF alpha blockers. Ann
Rheum Dis 2006;65:191–4.
115 Kapetanovic MC, Saxne T, Nilsson JA, Geborek P.
Influenza vaccination as model for testing immune
modulation induced by anti-TNF <strong>and</strong> methotrexate
therapy in <strong>rheumatoid</strong> <strong>arthritis</strong> patients. Rheumatology
2007;46:608–11.
116 Kapetanovic MC, Saxne T, Sjoholm A, Truedsson L,
Jonsson G, Geborek P. Influence of methotrexate, TNF
blockers <strong>and</strong> prednisolone on antibody responses to
pneumococcal polysaccharide vaccine in patients with
<strong>rheumatoid</strong> <strong>arthritis</strong>. Rheumatology 2006;45:106–11.
117 Mease PJ, Ritchlin CT, Martin RW et al. Pneumococcal
vaccine response in psoriatic <strong>arthritis</strong> patients during
treatment with etanercept. J Rheumatol 2004;31:
1356–61.
118 O’Dell J, Gilg J, Palmer W. Pneumococcal vaccination:
decreased antibody response in <strong>rheumatoid</strong> <strong>arthritis</strong>
patients on methotrexate. Arthritis Rheum 1992;9:1971.
119 Visvanathan S, Keenan GF, Baker DG, Levinson AI,
Wagner CL. Response to pneumococcal vaccine in patients
with early <strong>rheumatoid</strong> <strong>arthritis</strong> receiving infliximab
plus methotrexate or methotrexate alone. J Rheumatol
2007;34:952–7.
120 Ravikumar R, Owen T, Barnard J. Anti-TNF therapy in RA
patients alters hepatitis B vaccine responses. Arthritis
Rheum 2006;56:S366.
121 Avery RK. Vaccination of the immunosuppressed adult
patient with rheumatologic disease. Rheum Dis Clin
North Am 1999;25:567–84, viii.
122 Setoguchi S, Solomon DH, Weinblatt ME et al. Tumor
necrosis factor alpha antagonist use <strong>and</strong> cancer in
patients with <strong>rheumatoid</strong> <strong>arthritis</strong>. Arthritis Rheum 2006;
54:2757–64.
123 Geborek P, Bladstrom A, Turesson C et al. Tumour
necrosis factor blockers do not increase overall tumour
risk in patients with <strong>rheumatoid</strong> <strong>arthritis</strong>, but may be
associated with an increased risk of lymphomas. Ann
Rheum Dis 2005;64:699–703.
124 Askling J. No increase in overall risk for cancer in
<strong>rheumatoid</strong> <strong>arthritis</strong> treated with TNF-antagonists, but
risks for certain cancer types may be elevated.
Ann Rheum Dis 2007;66(Suppl. 2):S54.
125 Watson KD, Dixon WG, Hyrich KL, Lunt M,
Symmons DP, Silman AJ. Influence of anti-TNF therapy
<strong>and</strong> previous malignancy on cancer incidence in patients
with <strong>rheumatoid</strong> <strong>arthritis</strong> (RA): results from the <strong>BSR</strong>
biologics register. Ann Rheum Dis 2006;65(Suppl. 2):
S512.
126 Wolfe F, Michaud K. Biologic treatment of <strong>rheumatoid</strong>
<strong>arthritis</strong> <strong>and</strong> the risk of malignancy: analyses from a large
US observational study. Arthritis Rheum 2007;56:
2886–95.
127 Isomaki HA, Hakulinen T, Joutsenlahti U. Excess risk
of lymphomas, leukemia <strong>and</strong> myeloma in patients with
<strong>rheumatoid</strong> <strong>arthritis</strong>. J Chronic Dis 1978;31:691–6.
128 Gridley G, McLaughlin JK, Ekbom A et al. Incidence
of cancer among patients with <strong>rheumatoid</strong> <strong>arthritis</strong>.
J Natl Cancer Inst 1993;85:307–11.
129 Mellemkjaer L, Linet MS, Gridley G, Frisch M, Moller H,
Olsen JH. Rheumatoid <strong>arthritis</strong> <strong>and</strong> cancer risk. Eur J
Cancer 1996;32A:1753–7.
130 Ekstrom K, Hjalgrim H, Br<strong>and</strong>t L et al. Risk of malignant
lymphomas in patients with <strong>rheumatoid</strong> <strong>arthritis</strong> <strong>and</strong> in
their first-degree relatives. Arthritis Rheum 2003;48:
963–70.
131 Wolfe F, Michaud K. Lymphoma in <strong>rheumatoid</strong> <strong>arthritis</strong>:
the effect of methotrexate <strong>and</strong> anti-tumor necrosis factor
therapy in 18,572 patients. Arthritis Rheum 2004;50:
1740–51.
132 Askling J, Fored CM, Baecklund E et al. Haematopoietic
malignancies in <strong>rheumatoid</strong> <strong>arthritis</strong>: lymphoma risk <strong>and</strong>
characteristics after exposure to tumour necrosis factor
antagonists. Ann Rheum Dis 2005;64:1414–20.
133 Franklin J, Lunt M, Bunn D, Symmons D, Silman A.
Incidence of lymphoma in a large primary care derived
cohort of cases of inflammatory poly<strong>arthritis</strong>. Ann Rheum
Dis 2006;65:617–22.
20 www.rheumatology.oxfordjournals.org

RA <strong>guidelines</strong> on safety of anti-TNF therapies
134 Brown SL, Greene MH, Gershon SK, Edwards ET,
Braun MM. Tumor necrosis factor antagonist therapy
<strong>and</strong> lymphoma development: twenty-six cases reported
to the Food <strong>and</strong> Drug Administration. Arthritis Rheum
2002;46:3151–8.
135 Wolfe F, Michaud K. The effect of methotrexate <strong>and</strong>
anti-tumor necrosis factor therapy on the risk of
lymphoma in <strong>rheumatoid</strong> <strong>arthritis</strong> in 19,562 patients
during 89,710 person-years of observation. Arthritis
Rheum 2007;56:1433–9.
136 Chakravarty EF, Michaud K, Wolfe F. Skin cancer,
<strong>rheumatoid</strong> <strong>arthritis</strong>, <strong>and</strong> tumor necrosis factor inhibitors.
J Rheumatol 2005;32:2130–5.
137 Amari W, Zeringue AL, McDonald JR et al.
Non-melanoma <strong>and</strong> melanoma skin cancer risk in a
National Cohort of Veterans with Rheumatoid Arthritis.
Arthritis Rheum 2009;60(10 Suppl.):S516.
138 Mercer LK, Galloway JB, Lunt M et al. The Influence of
anti-TNF therapy upon incidence of non-melanoma skin
cancer (NMSC) in patients with <strong>rheumatoid</strong> <strong>arthritis</strong> (RA):
results from the <strong>BSR</strong> Biologics Register (<strong>BSR</strong>BR).
Arthritis Rheum 2009;60(10 Suppl.):S772.
139 Dixon WG, Watson K, Lunt M, Hyrich KL,
Symmons DPM. The influence of anti-TNF therapy upon
cancer incidence in patients with RA who have had prior
malignancy: results from the <strong>BSR</strong>BR. Arthritis Rheum
2008;58:S638–9.
140 Rennard SI, Fogarty C, Kelsen S et al. The safety <strong>and</strong>
efficacy of infliximab in moderate to severe chronic
obstructive pulmonary disease. Am J Respir Crit Care
Med 2007;175:926–34.
141 Morel<strong>and</strong> LW, Schiff MH, Baumgartner SW et al.
Etanercept therapy in <strong>rheumatoid</strong> <strong>arthritis</strong>.
A r<strong>and</strong>omized, controlled trial. Ann Intern Med 1999;130:
478–86.
142 Weinblatt ME, Kremer JM, Bankhurst AD et al. A trial of
etanercept, a recombinant tumor necrosis factor
receptor:Fc fusion protein, in patients with <strong>rheumatoid</strong>
<strong>arthritis</strong> receiving methotrexate. N Engl J Med 1999;340:
253–9.
143 Maini R, St Clair EW, Breedveld F et al. Infliximab
(chimeric anti-tumour necrosis factor alpha monoclonal
antibody) versus placebo in <strong>rheumatoid</strong> <strong>arthritis</strong> patients
receiving concomitant methotrexate: a r<strong>and</strong>omised
phase III trial. ATTRACT Study Group. Lancet 1999;354:
1932–9.
144 Benucci M, Saviola G, Baiardi P, Cammelli E,
Manfredi M. Anti-nucleosome antibodies as prediction
factor of development of autoantibodies during
therapy with three different TNFalpha blocking
agents in <strong>rheumatoid</strong> <strong>arthritis</strong>. Clin Rheumatol 2008;
27:91–5.
145 Williams EL, Gadola S, Edwards CJ. Anti-TNF-induced
lupus. Rheumatology 2009;48:716–20.
146 Thornhill J, Watson KD, Lord PA, Symmons DP,
Hyrich KL. Drug-induced lupus in patients with inflammatory
<strong>arthritis</strong> treated with TNF blocking agents: results
from the <strong>BSR</strong> Biologics Register. Rheumatology 2009;
47(Suppl. 2):ii15.
147 Ramos-Casals M, Brito-Zeron P, Munoz S et al.
Autoimmune diseases induced by TNF-targeted
therapies: analysis of 233 cases. Medicine 2007;86:
242–51.
148 Stubgen JP. Tumor necrosis factor-alpha antagonists
<strong>and</strong> neuropathy. Muscle Nerve 2008;37:281–92.
149 Kur-Zalewska J, Swarowska-Knap J, Tlustochowicz W.
Neurological disorders with demyelinating brain white
matter lesions in a patient with <strong>rheumatoid</strong> <strong>arthritis</strong>
treated with etanercept. Pol Arch Med Wewn 2008;118:
234–7.
150 Kameda T, Dobashi H, Kittaka K et al. A case of
<strong>rheumatoid</strong> <strong>arthritis</strong> complicated by demyelination in
both cerebral cortex <strong>and</strong> spinal cord during etanercept
therapy. Mod Rheumatol 2008;18:399–402.
151 Clemente RE. Demyelinating disease <strong>and</strong> cutaneous
lymphocitic vasculitis after etanercept therapy in a
patient with <strong>rheumatoid</strong> <strong>arthritis</strong>. Sc<strong>and</strong> J Rheumatol
2007;36:244–5, author reply 245.
152 Yamamoto M, Takahashi H, Wakasugi H et al.
Leukoencephalopathy during administration of etanercept
for refractory <strong>rheumatoid</strong> <strong>arthritis</strong>. Mod Rheumatol
2007;17:72–4.
153 Jar<strong>and</strong> J, Zochodne DW, Martin LO, Voll C. Neurological
complications of infliximab. J Rheumatol 2006;33:
1018–20.
154 Tanno M, Nakamura I, Kobayashi S, Kurihara K, Ito K.
New-onset demyelination induced by infliximab therapy
in two <strong>rheumatoid</strong> <strong>arthritis</strong> patients. Clin Rheumatol
2006;25:929–33.
155 Bensouda-Grimaldi L, Mulleman D, Valat JP,
Autret-Leca E. Adalimumab-associated multiple sclerosis.
J Rheumatol 2007;34:239–40; Discussion 240.
156 Simsek I, Erdem H, Pay S, Sobaci G, Dinc A. Optic
neuritis occurring with anti-tumour necrosis factor a
therapy. Ann Rheum Dis 2007;66:1255–8.
157 Lozeron P, Denier C, Lacroix C, Adams D. Long-term
course of demyelinating neuropathies occurring during
tumor necrosis factor-alpha-blocker therapy. Arch
Neurol 2009;66:490–7.
158 Shin IS, Baer AN, Kwon HJ, Papadopoulos EJ,
Siegel JN. Guillain-Barre <strong>and</strong> Miller Fisher syndromes
occurring with tumor necrosis factor alpha antagonist
therapy. Arthritis Rheum 2006;54:1429–34.
159 Mann DL, McMurray JJ, Packer M et al. Targeted
anticytokine therapy in patients with chronic heart
failure: results of the R<strong>and</strong>omized Etanercept Worldwide
Evaluation (RENEWAL). Circulation 2004;109:1594–602.
160 Anker SD, Coats AJ. How to RECOVER from
RENAISSANCE? The significance of the results of
RECOVER, RENAISSANCE, RENEWAL <strong>and</strong> ATTACH. Int
J Cardiol 2002;86:123–30.
161 Chung ES, Packer M, Lo KH, Fasanmade AA,
Willerson JT. R<strong>and</strong>omized, 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;107:3133–40.
162 Gabriel SE. Tumor necrosis factor inhibition: a part of the
solution or a part of the problem of heart failure in
<strong>rheumatoid</strong> <strong>arthritis</strong>? Arthritis Rheum 2008;58:637–40.
163 Kwon HJ, Cote TR, Cuffe MS, Kramer JM, Braun MM.
Case reports of heart failure after therapy with a tumor
necrosis factor antagonist. Ann Intern Med 2003;138:
807–11.
www.rheumatology.oxfordjournals.org 21

Tina Ding et al.
164 Bernatsky S, Hudson M, Suissa S. Anti-rheumatic
drug use <strong>and</strong> risk of hospitalization for congestive heart
failure in <strong>rheumatoid</strong> <strong>arthritis</strong>. Rheumatology 2005;44:
677–80.
165 Cole J, Busti A, Kazi S. The incidence of new onset
congestive heart failure <strong>and</strong> heart failure exacerbation
in Veteran’s Affairs patients receiving tumor necrosis
factor alpha antagonists. Rheumatol Int 2007;27:369–73.
166 Listing J, Strangfeld A, Kekow J et al. Does tumor
necrosis factor alpha inhibition promote or prevent heart
failure in patients with <strong>rheumatoid</strong> <strong>arthritis</strong>? Arthritis
Rheum 2008;58:667–77.
167 Curtis JR, Kramer JM, Martin C et al. Heart failure among
younger <strong>rheumatoid</strong> <strong>arthritis</strong> <strong>and</strong> Crohn’s patients
exposed to TNF-alpha antagonists. Rheumatology 2007;
46:1688–93.
168 Kent JD, Pangan AL, Fitzpatrick SB. Analysis of the US
postmarketing safety of adalimumab (HUMIRA Õ )in
patients with <strong>rheumatoid</strong> <strong>arthritis</strong> during the first 2 years
after approval. Arthritis Rheum 2005;52:1455.
169 Suissa S, Bernatsky S, Hudson M. Antirheumatic drug
use <strong>and</strong> the risk of acute myocardial infarction. Arthritis
Rheum 2006;55:531–6.
170 Dixon WG, Watson KD, Lunt M, Hyrich KL, Silman AJ,
Symmons DP. Reduction in the incidence of myocardial
infarction in patients with <strong>rheumatoid</strong> <strong>arthritis</strong> who
respond to anti-tumor necrosis factor alpha therapy:
results from the British Society for Rheumatology
Biologics Register. Arthritis Rheum 2007;56:2905–12.
171 Jacobsson LT, Turesson C, Gulfe A et al. Treatment with
tumor necrosis factor blockers is associated with a lower
incidence of first cardiovascular events in patients with
<strong>rheumatoid</strong> <strong>arthritis</strong>. J Rheumatol 2005;32:1213–8.
172 Solomon DH, Avorn J, Katz JN et al. Immunosuppressive
medications <strong>and</strong> hospitalization for cardiovascular
events in patients with <strong>rheumatoid</strong> <strong>arthritis</strong>. Arthritis
Rheum 2006;54:3790–8.
173 Rajakulendran S, Gadsby K, Allen D, O’Reilly S,
Deighton C. Neutropenia while receiving anti-tumour
necrosis factor treatment for <strong>rheumatoid</strong> <strong>arthritis</strong>.
Ann Rheum Dis 2006;65:1678–9.
174 Yazdani R, Simpson H, Kaushik V. Incidence of
cytopoenias in anti-TNF-alpha therapy. Rheumatology
2007;46(Suppl. 1):i33.
175 Litwic AE, Tann M, Ledingham JM. Neutropenia with
anti-TNF therapy for <strong>rheumatoid</strong> <strong>arthritis</strong>. Rheumatology
2009;48(Suppl. 1):i68.
176 Menon Y, Cucurull E, Espinoza LR. Pancytopenia in a
patient with scleroderma treated with infliximab.
Rheumatology 2003;42:1273–4, author reply 1274.
177 Stern A, Riley R, Buckley L. Worsening of macrophage
activation syndrome in a patient with adult onset Still’s
disease after initiation of etanercept therapy. J Clin
Rheumatol 2001;7:252–6.
178 Carter JD, Ladhani A, Ricca L, Valeriano J, Vasey FB.
A safety assessment of TNF antagonists during
pregnancy: a review of the FDA Database. Arthritis
Rheum 2007;56:S667.
179 Roux CH, Brocq O, Breuil V, Albert C, Euller-Ziegler L.
Pregnancy in rheumatology patients exposed to
anti-tumour necrosis factor (TNF)-alpha therapy.
Rheumatology 2007;46:695–8.
180 Katz JA, Antoni C, Keenan GF, Smith DE, Jacobs SJ,
Lichtenstein GR. Outcome of pregnancy in women
receiving infliximab for the treatment of Crohn’s disease
<strong>and</strong> <strong>rheumatoid</strong> <strong>arthritis</strong>. Am J Gastroenterol 2004;99:
2385–92.
181 King YE, Watson KD, Symmons DPM, Hyrich KL.
Pregnancy outcome in women exposed to anti-TNF
agents: an update from the British Society for
Rheumatology Biologics Register (<strong>BSR</strong>BR). Arthritis
Rheum 2008;58:S542.
182 Hunt JS, Chen HL, Miller L. Tumor necrosis factors:
pivotal components of pregnancy? Biol Reprod 1996;54:
554–62.
183 Haider S, Knofler M. Human tumour necrosis factor:
physiological <strong>and</strong> pathological roles in placenta <strong>and</strong>
endometrium. Placenta 2009;30:111–23.
184 Rosner I, Haddad A, Boulman N et al. Pregnancy in
rheumatology patients exposed to anti-tumour necrosis
factor (TNF)-alpha therapy. Rheumatology 2007;46:
1508.
185 Berthelot JM, De B<strong>and</strong>t M, Goupille P et al. Exposition to
anti-TNF drugs during pregnancy: outcome of 15 cases
<strong>and</strong> review of the literature. Joint Bone Spine 2009;76:
28–34.
186 Hyrich KL, Watson KD, Dixon WG, Silman AJ,
Symmons DPM. Pregnancy experience in women with
rheumatic diseases exposed to biologic agents: results
from the <strong>BSR</strong> biologic register. Ann Rheum Dis 2006;
65(Suppl. 2):321.
187 Chambers CD, Johnson DL, Jones KL. Pregnancy
outcome in women exposed to anti-TNF alpha
medications: the OTIS <strong>rheumatoid</strong> <strong>arthritis</strong> in Pregnancy
Study. Arthritis Rheum 2004;50:S479.
188 Pentsuk N, van der Laan JW. An interspecies
comparison of placental antibody transfer: new insights
into developmental toxicity testing of monoclonal
antibodies. Birth Defects Res B Dev Reprod Toxicol
2009;86:328–44.
189 Gisbert JP. Safety of immunomodulators <strong>and</strong> biologics
for the treatment of inflammatory bowel disease during
pregnancy <strong>and</strong> breast-feeding. Inflamm Bowel Dis 2009;
16:881–95.
190 Viktil KK, Engel<strong>and</strong> A, Furu K. Use of antirheumatic
drugs in mothers <strong>and</strong> fathers before <strong>and</strong> during
pregnancy—a population-based cohort study.
Pharmacoepidemiol Drug Saf 2009;18:737–42.
191 Ostensen M, Eigenmann GO. Etanercept in breast milk.
J Rheumatol 2004;31:1017–8.
192 Murashima A, Watanabe N, Ozawa N, Saito H,
Yamaguchi K. Etanercept during pregnancy <strong>and</strong> lactation
in a patient with <strong>rheumatoid</strong> <strong>arthritis</strong>: drug levels in
maternal serum, cord blood, breast milk <strong>and</strong> the infant’s
serum. Ann Rheum Dis 2009;68:1793–4.
193 Dixon WG, Watson KD, Lunt M, Hyrich KL,
Symmons DP. Rheumatoid <strong>arthritis</strong>, interstitial lung disease,
mortality <strong>and</strong> anti-TNF therapy: results from the
<strong>BSR</strong> biologics register (<strong>BSR</strong>BR). Ann Rheum Dis 2007;
66(Suppl. II):55.
194 National Institute of Health <strong>and</strong> Clinical Excellence.
Infliximab for the treatment of psoriasis. 2008,
http://www.nice.org.uk/Guidance/TA134 (20 January
2010, date last accessed).
22 www.rheumatology.oxfordjournals.org

RA <strong>guidelines</strong> on safety of anti-TNF therapies
195 National Institute of Health <strong>and</strong> Clinical Excellence.
Etanercept <strong>and</strong> efalizumab for the treatment of adults
with psoriasis. 2006, http://www.nice.org.uk/
Guidance/TA103 (20 January 2010, date last accessed).
196 National Institute of Health <strong>and</strong> Clinical Excellence.
Adalimumab for the treatment of psoriasis. 2008,
http://www.nice.org.uk/Guidance/TA146 (20 January
2010, date last accessed).
197 Harrison MJ, Dixon WG, Watson KD et al. Rates of
new-onset psoriasis in patients with <strong>rheumatoid</strong> <strong>arthritis</strong>
receiving anti-TNFa therapy. Results from the British
Society for Rheumatology Biologics Register. Ann
Rheum Dis 2008;68:209–15.
198 Collamer AN, Guerrero KT, Henning JS, Battafarano DF.
Psoriatic skin lesions induced by tumor necrosis factor
antagonist therapy: a literature review <strong>and</strong> potential
mechanisms of action. Arthritis Rheum 2008;59:
996–1001.
199 Ko JM, Gottlieb AB, Kerbleski JF. Induction <strong>and</strong>
exacerbation of psoriasis with TNF-blockade therapy: a
review <strong>and</strong> analysis of 127 cases. J Dermatolog Treat
2009;20:100–8.
200 Taban M, Dupps WJ, M<strong>and</strong>ell B, Perez VL. Etanercept
(enbrel)-associated inflammatory eye disease: case
report <strong>and</strong> review of the literature. Ocul Immunol Inflamm
2006;14:145–50.
201 Wang F, Wang NS. Etanercept therapy-associated acute
uveitis: a case report <strong>and</strong> literature review. Clin Exp
Rheumatol 2009;27:838–9.
202 Saurenmann RK, Levin AV, Rose JB et al. Tumour
necrosis factor alpha inhibitors in the treatment of
childhood uveitis. Rheumatology 2006;45:982–9.
203 Tynjala P, Lindahl P, Honkanen V, Lahdenne P,
Kotaniemi K. Infliximab <strong>and</strong> etanercept in the treatment
of chronic uveitis associated with refractory juvenile
idiopathic <strong>arthritis</strong>. Ann Rheum Dis 2007;66:548–50.
204 Braun J, Baraliakos X, Listing J, Sieper J. Decreased
incidence of anterior uveitis in patients with ankylosing
spondylitis treated with the anti-tumor necrosis factor
agents infliximab <strong>and</strong> etanercept. Arthritis Rheum 2005;
52:2447–51.
205 Galor A, Perez VL, Hammel JP, Lowder CY. Differential
effectiveness of etanercept <strong>and</strong> infliximab in the treatment
of ocular inflammation. Ophthalmology 2006;113:
2317–23.
206 Lim LL, Fraunfelder FW, Rosenbaum JT. Do tumor
necrosis factor inhibitors cause uveitis? A registry-based
study. Arthritis Rheum 2007;56:3248–52.
Appendix I
Rating the strength of evidence for recommendations
For each final recommendation, the strength of evidence
was assigned using the grading suggested by the Royal
College of Physicians of London (National Clinical
Guidelines for Stroke by the Intercollegiate Stroke
Working Party, June 2004 http://www.rcplondon.ac
.uk/pubs/books/stroke/stroke_<strong>guidelines</strong>_2ed.pdf, table
1.1; 22 June 2010, date last accessed).
The data supporting recommendations on safety aspects
of anti-TNF therapy were derived primarily from observational
studies, <strong>and</strong> to a lesser degree from evidence
from RCTs. Many of these studies might have addressed
the general topic under discussion but did not specifically
address the questions of relevance. Therefore, the recommendations
could not be derived directly from the evidence
but required synthesis of the data from studies
plus extrapolation to the specific clinical scenarios under
consideration. As a result, many of the recommendations
were graded as Level C.
www.rheumatology.oxfordjournals.org 23