British Journal of Dermatology 2004; 151: 1123–1132. DOI: 10.1111/j.1365-2133.2004.06323.

Guidelines
Guidelines for prescribing azathioprine in dermatology
A.V.ANSTEY, S.WAKELIN* AND N.J.REYNOLDS†
Department of Dermatology, Royal Gwent Hospital, Cardiff Road, Newport, Gwent NP20 2UB, U.K.
*Department of Dermatology, St Mary’s Hospital, Praed Street, London W2 1NY, U.K.
†Department of Dermatology, School of Clinical and Laboratory Sciences, University of Newcastle upon Tyne, Framlington Place,
Newcastle upon Tyne NE2 4HH, U.K.

Accepted for publication 14 May 2004

Summary Azathioprine has been available as an immunosuppressive agent for over 40 years, and current
routine usage in dermatology is not restricted to licensed indications. Advances in understanding the
metabolic fate of azathioprine have led to significant changes in prescribing practice and toxicity
monitoring by U.K. dermatologists. The current state of knowledge concerning the use of
azathioprine in dermatology is summarized, with identification of strength of evidence. Clinical
indications and contraindications for azathioprine usage in dermatology are identified. Evidence-
based recommendations are made for routine safety monitoring of patients treated with azathioprine,
including pretreatment assessment of red blood cell thiopurine methyltransferase activity.

of autoimmune and immune-mediated dermatological

Disclaimer
conditions. Licensed indications include dermatomyo-
These guidelines have been prepared for dermatologists sitis, systemic lupus erythematosus and pemphigus
on behalf of the British Association of Dermatologists and vulgaris. It is used for these conditions either alone or,
reflect the best data available at the time the report was more commonly, in combination with corticosteroids.
prepared. Caution should be exercised in interpreting the Therapeutic effect is typically delayed for weeks or even
data; the results of future studies may require alteration 2–3 months, and includes a steroid-sparing effect that
of the conclusions or recommendations in this report. It reduces long-term toxicity of corticosteroids. Azathio-
may be necessary to depart from the guidelines in the prine is also used quite frequently as monotherapy in
interests of specific patients and in special circumstances. nonlicensed indications including atopic eczema.
Just as adherence to guidelines may not constitute Adverse drug reactions with azathioprine occur in
defence against a claim of negligence, so deviation from 15–28% of patients1–3 and include myelosuppression,
them should not necessarily be deemed negligent. nausea and vomiting, rash, pancreatitis and hypersen-
sitivity. Polymorphism in the thiopurine methyltrans-
ferase (TPMT) gene predicts haematological adverse
Introduction
drug reactions in 5–10% of patients treated with
Azathioprine is an immune-modulating drug that was thiopurine drugs.1 The remaining adverse drug reac-
originally developed for the control of graft rejection in tions are unexplained, and may be mediated by
transplant surgery. Since it became available in the early immune mechanisms or by other variables affecting
1960s, azathioprine has also been prescribed for a range the metabolic fate of the drug. It is therefore essential to
continue monitoring blood counts throughout treat-
ment with azathioprine.4
These guidelines were commissioned by the British Association of
Dermatologists Therapy, Guidelines and Audit Subcommittee. Mem-
bers of the committee are A.D.Ormerod (Chairman), A.S.Highet, Methods
D.Mehta, R.H.Meyrick Thomas, C.H.Smith, J.C.Sterling and A.Anstey.
Conflicts of interest: No additional funding was received by the The aim of the search strategy was to identify recent
authors of these guidelines to support this work. None of the authors
and past publications relating to the clinical pharma-
has a conflict of interest to declare.
Correspondence: Alex Anstey. cology of azathioprine that were relevant to its current
E-mail: [email protected] usage within dermatology. The evidence gathered

2004 British Association of Dermatologists 1123

1124 A . V . A N S T E Y et al.

includes some derived from disciplines other than

Pharmacology
dermatology, such as gastroenterology, where aza-
thioprine usage is high and experience with TPMT- Azathioprine is a prodrug with no inherent immuno-
guided prescribing is established. suppressive activity. Azathioprine is an imidazole
derivative of 6-mercaptopurine (6-MP) and is therefore
classed as a purine analogue. It was developed in an
Types of studies
attempt to produce a drug with the same immuno-
Randomized, double-blind, placebo-controlled trials; suppressive activity as 6-MP but with significantly
well-designed controlled trials without randomization; slower metabolism. Following oral administration,
well-designed cohort or case–control analytical studies; azathioprine is rapidly and almost completely
evidence from multiple time series with or without absorbed from the gut. No azathioprine crosses the
intervention; opinions of respected authorities based on blood–brain barrier, but there is otherwise an even
clinical experience, descriptive studies or reports of distribution of the drug throughout the body. The
expert committees. plasma half-life is just 3 h owing to rapid nonenzy-
matic conversion to 6-MP, the active compound, and
imidazole derivatives. 6-MP has a long half-life and is
Search strategy for identification of studies
metabolized by three competing pathways (Fig. 1).
A computer-assisted search of the online bibliographic Active 6-thioguanine nucleotide metabolites are gen-
databases Medline, PubMed and Embase was carried erated by the action of hypoxanthine guanine phos-
out to identify potentially relevant papers published phoribosyl transferase (HPRT). Xanthine oxidase,
between 1966 and 2003. Other databases searched which is inhibited by allopurinol, metabolizes 6-MP
included the Royal College of Physicians Guidelines to inactive 6-thiouric acid. The third of the competing
database, CINAHL, the Cochrane library, DARE, AMED pathways involves methylation of 6-MP to 6-methyl
and HMIC. The following search terms were used: mercaptopurine, an inactive compound, which is
azathioprine; 6-mercaptopurine; dermatology; adverse catalysed by TPMT.
drug reactions; clinical monitoring; TPMT; thiopurine
methyltransferase. Citations were limited to those in
Azathioprine mode of action
English, French, Spanish, Italian and German. Manual
searches of the reference lists from the relevant papers 6-MP readily crosses cell membranes, and is converted
were performed in order to identify additional studies by the enzyme HPRT into a number of active purine
that may have been missed by the computer-assisted thionucleotide metabolites. The exact mode of action of
strategy. These guidelines include evidence derived azathioprine at a cellular level remains unclear, but
from the key papers identified by the search strategy. thionucleotide metabolites of 6-MP are believed to
compete with their endogenous counterparts in many
biochemical pathways. Nucleotides have a number of
Risks and benefits of azathioprine therapy in
important roles in all cells: they are precursors of DNA
dermatology
and RNA, they are essential carriers of energy (e.g.
Particular emphasis has been placed on assessing the adenosine triphosphate and guanosine triphosphate),
risks and benefits of azathioprine therapy for patients and they function as cellular second messengers. These
with dermatological disease. These guidelines attempt nucleotide-dependent processes endow azathioprine
to establish an explicit link between evidence and with both immunosuppressive and cytotoxic properties.
recommendations for clinical usage. This is sometimes There is also evidence that imidazole derivatives and
difficult, as decisions in clinical medicine occur in the the thiopurine intermediates have independent immu-
context of single patients and do not always relate to the nosuppressive properties.5
context from which a guideline recommendation has
been made. Finally, these guidelines have been subjec-
The significance of thiopurine
ted to expert review by nondermatologists (acknow-
methyltransferase in the clinical
ledged) with recognized expertise in the prescribing of
pharmacology of azathioprine
azathioprine. In line with other British Association of
Dermatologists (BAD) guidelines it is the intention for There is marked interpatient variability in the genera-
these guidelines to be reviewed and updated in 5 years. tion of immunosuppressant metabolites, resulting from

2004 British Association of Dermatologists, British Journal of Dermatology, 151, 1123–1132

 GUIDELINES FOR PRESCRIBING AZATHIOPRINE IN DERMATOLOGY 1125

Metabolism of azathioprine

Azathioprine

TPMT XO

6-methyl
6-MP 6-thiouric acid
Figure 1. Metabolism of azathioprine. The
mercaptopurine
first metabolite is 6-mercaptopurine (6-MP). HPRT
This may be metabolized in three ways:
(i) methylation to inactive 6-methyl mercap- 6-thioinosine 5-
topurine (6-MMP) catalysed by thiopurine monophosphate
methyl transferase (TPMT); (ii) oxidation to 6-MMP
inactive 6-thiouric acid catalysed by xanthine
6-thioguanine
ribonucleotides
oxidase (XO); and (iii) hypoxanthine guanine
phosphoribosyl transferase (HPRT) catalyses
the conversion of 6-MP to 6-thioinosine
5-monophosphate which is further converted
to 6-thioguanine. 6-Thioguanine has an Purine synthesis DNA synthesis
inhibitory effect on DNA synthesis.

a common genetic polymorphism in TPMT, one of Table 1 Licensed and unlicensed indications for azathioprine in the
treatment of dermatological disorders
three competing pathways for 6-MP metabolism
(Fig. 1). Eleven per cent of the population has low Licensed indications Unlicensed indications
TPMT activity and is vulnerable to myelosuppression Systemic lupus erythematosus Atopic dermatitis
with azathioprine treatment.6 Furthermore, one in 300 Dermatomyositis Psoriasis
Pemphigus vulgaris Bullous pemphigoid
individuals has undetectable TPMT activity,6 and is Chronic actinic dermatitis
susceptible to rapid-onset, prolonged, life-threatening Pyoderma gangrenosum
pancytopenia if treated with conventional doses of Pityriasis rubra pilaris
azathioprine.7 A recent large clinical study on patients Wegener’s granulomatosis
Cutaneous vasculitis
being treated with azathioprine showed the rate of
undetectable TPMT activity to be higher, at one in 200
patients;4 however, the cases in this series were not is used as a steroid-sparing agent. Accumulating
randomly selected and there may have been some bias evidence also suggests a role for azathioprine as a
towards a higher rate of
problem cases. Inhibition of single agent in the treatment of severe, recalcitrant
xanthine oxidase, the third enzyme in this system atopic dermatitis15–17 (Grade A; level I). Double-blind,
(Fig. 1), produces the same effect on azathioprine placebo-controlled trials have shown azathioprine to be
metabolism as deficiency in TPMT: decreased metabo- of benefit in chronic actinic dermatitis18 (Grade A; level
lism of 6-MP to inactive metabolites, and increased I) and Behçet’s disease19 (Grade A; level I). It may also
generation of immunosuppressant 6-thioguanine nu- be effective as monotherapy in the treatment of severe,
cleotides. This explains the potentially serious drug recalcitrant psoriasis20 (Grade C; level IV). Azathio-
interaction between azathioprine and allopurinol.8,9 prine is sometimes used in the treatment of other rare
dermatological conditions including Wegener’s gra-
nulomatosis,21 pyoderma gangrenosum,22 pityriasis
Azathioprine indications in dermatology
rubra pilaris,23 lupus erythematosus24 and lichen
Indications are listed in Table 1. Azathioprine is a planus,25 but evidence to support its usage in these
popular drug with dermatologists, as it is perceived as conditions is anecdotal26 (Grade C; level IV).
an immunosuppresive drug in which the potential
benefits outweigh the potential risks. The main area of
Azathioprine contraindications
use in dermatology is in the treatment of autoimmune
dermatoses, in particular bullous pemphigoid10–12 Azathioprine is contraindicated in patients with known
(Grade B; level IV; see Appendix 1), and pemphigus hypersensitivity to the drug27–29 (Grade A; level III).
vulgaris13,14 (Grade B; level II-iii) where azathioprine Evidence of teratogenicity with azathioprine in humans

2004 British Association of Dermatologists, British Journal of Dermatology, 151, 1123–1132

1126 A . V . A N S T E Y et al.

is equivocal,30 but adequate contraceptive precautions significance of interpatient variability in TPMT activity
are advised when either partner is taking azathioprine. that coincided with availability of a National Health
Azathioprine is also contraindicated in pregnancy Service laboratory service for TPMT assay (Guy’s
(except where benefit may outweigh risk such as in Hospital, Purine Research Laboratory). The justifica-
allograft recipients)30–32 (Grade A; level II-ii). 6-MP has tion for pretreatment TPMT measurement is the
been identified in colostrum and in the breast milk of presumed improvements in safety and efficacy that
women receiving azathioprine treatment. Women on follow from knowing its activity. In acute lymphoblas-
azathioprine should therefore be advised to bottle feed tic leukaemia (ALL) treated with 6-MP, lower TPMT
their babies. It is strongly recommended that azathio- activity tended to be associated with a better out-
prine should not be used in patients whose TPMT come,35 although ALL patients with lower TPMT
status is unknown4,15 (Grade A; level II-ii). Very low or activity appeared to be at greater risk of developing
absent TPMT activity is a contraindication to the second malignancies.35 Although there are currently
usage of azathioprine because of the high risk of life- no published prospective studies for dermatological
threatening pancytopenia24 (Grade A; level II-ii). conditions which demonstrate improved prognosis,
Concurrent treatment with allopurinol results in an TPMT screening prior to azathioprine treatment is
important drug interaction which may cause considered by some clinicians to be essential (Grade A;
significant myelosuppression, and should therefore be level II-ii).4,10
avoided8,9 (Grade A; level III). Knowledge of TPMT prevents those with very low or
There are concerns that azathioprine treatment undetectable TPMT from receiving azathioprine, and
increases the risk of developing a malignancy. Accu- may prevent fatal myelosuppression. This is the most
mulating evidence suggests that this risk is smaller important aspect of measuring TPMT. Prior TPMT
than was originally feared.33,34 Nevertheless, patients assessment also identifies those with intermediate
should be advised of this risk, and it is recommended activity who would be predicted to suffer toxicity on
that azathioprine treatment should not usually be standard doses of azathioprine but who might tolerate
initiated or continued in patients with known malig- and respond to tailored doses of azathioprine. Other
nancy (Grade A; level III). components of the azathioprine metabolic pathways
may rarely be deficient or exert an immunosuppressive
Contradictions to azathioprine: effect, including deficiency in purine-5¢-nucleotidase
and xanthine oxidase, and lethal immunosuppressive
• It is strongly recommended that azathioprine should
effects mediated by imidazole derivatives.5 Dermatolo-
not be used in patients whose TPMT status is unknown
gists should therefore be aware that knowledge of TPMT
• Known hypersensitivity to azathioprine (or 6-MP)
status does not preclude the necessity for monitoring for
• Azathioprine is contraindicated in patients who may
azathioprine toxicity by regular blood tests. Genotype
be pregnant or hope to become pregnant in the near
tests for the commonest mutations in the TPMT gene
future (except where benefit may outweigh risk)
are of interest to clinicians, but do not provide the
• Women taking azathioprine should not breast feed
functional result which measurement of the enzyme
their babies
activity provides.36 Furthermore, the TPMT genotyping
• Very low or absent TPMT activity
tests are not yet available in the U.K. as a routine
• Concurrent allopurinol treatment
laboratory test. Thus, dermatologists are advised to
• Concurrent malignant disease where azathioprine
continue to use the TPMT enzymatic assay.
treatment may increase the risk of disease progression
• Renal or hepatic insufficiency (relative contraindi- • Pretreatment TPMT measurement should be per-
cation) formed in all patients prescribed azathioprine for
treatment of dermatological conditions
Pretreatment thiopurine methyltransferase
assessment Azathioprine dosage
Azathioprine prescribing practice among U.K. derma- The recommended dosage of azathioprine for dermato-
tologists has changed in recent years, with increased logical indications is 1–3 mg kg)1 daily, adjusted
demand for pretreatment assessment of TPMT activ- within these limits according to response. If no
ity.4 This is explained by greater awareness of the improvement occurs in the patient’s condition within

2004 British Association of Dermatologists, British Journal of Dermatology, 151, 1123–1132

 GUIDELINES FOR PRESCRIBING AZATHIOPRINE IN DERMATOLOGY 1127

3 months, consideration should be given to withdraw- • In patients with low TPMT activity, azathioprine
ing azathioprine. Care should be taken when prescri- should either not be prescribed or, if used, the dose
bing azathioprine in the elderly: it is recommended that should be low (0Æ5–1 mg kg)1 daily) with careful
the dosage used is at the lower end of the range. There monitoring for myelosuppression
are currently no data to support prescribing azathio- • In patients with normal or high TPMT activity,
prine in doses outside the above range. However, azathioprine dosage should commence at the top of
modified dosage regimens based on TPMT activity have the 1–3 mg kg)1 daily dosage range. In patients
been published for both adults16,37 and children38 who fail to respond, and in whom no adverse effects
(Grade C; level III), and are the logical progression of occur, dosage above the 1–3 mg kg)1 daily range
this pharmacogenetic assessment. might be considered for a trial period
Azathioprine should not be used in patients with very
low ⁄ absent TPMT activity (deficient), as the danger of
Monitoring for azathioprine-induced toxicity
severe and prolonged myelosuppression is signifi-
cant20,39 (Grade A; level II-ii). Patients with inflamma- The data sheet for azathioprine recommends weekly
tory bowel disease and low TPMT activity have been monitoring of full blood count (FBC) for the first
shown to be at increased risk of azathioprine toxicity.39 8 weeks of treatment. However, the British National
Thus, for patients with low TPMT activity, alternative Formulary states that the
evidence of the practical
systemic therapies should be considered (Grade A; level value (of weekly FBC for 8 weeks) is unsatisfactory,
II-ii). If a trial of azathioprine is deemed appropriate in and recommends weekly FBCs for the first 4 weeks,
this situation, a low-dosage regimen should be used followed by reduced frequency of monitoring to a
(0Æ5–1 mg kg)1 daily) and extra care taken with haem- minimum of once every 3 months. Routine monitoring
atological surveillance (Grade B; level III). for azathioprine toxicity should also include liver
In patients with high TPMT activity (see Appendix 2 function tests (LFTs) as hepatoxicity is a recognized
for laboratory ranges), the azathioprine dose should be complication of azathioprine therapy.10,11,40,41 Thus, it
at the higher end of the range of 1–3 mg kg)1 daily. It is advised that dermatologists carry out weekly blood
is probably safe to treat these patients from the outset tests (FBCs and LFTs) until maintenance dose is
with dosages of azathioprine towards the top end of this achieved, followed by regular monitoring reducing to
dosage range provided the usual measures are taken to a minimum of once every 3 months for the duration of
monitor for myelosuppression. However, azathioprine therapy42 (Grade A; level I).
intolerance unrelated to TPMT activity is not uncom- For higher dosages and for patients with hepatic or
mon, and a lower initial dose of azathioprine is renal impairment, initial blood count monitoring more
advocated by some authors for the first month of frequently than once weekly is advised. Return to weekly
therapy, even in patients with high TPMT activity.10 In FBCs and LFTs should also follow an increase in dosage in
patients with inflammatory bowel disease, high TPMT azathioprine in patients already established on this
activity predicts treatment failure with azathioprine.3 treatment. It is also advised that patients on azathioprine
Thus, in dermatology patients with high TPMT activ- be instructed to report immediately any evidence of
ity, azathioprine dosage should be at the top of the infection, unexpected bruising or bleeding or jaundice.
recommended dose range of 1–3 mg kg)1 daily. In In patients with low TPMT activity (3–8 nmol h)1 mL)1
patients who fail to respond to 3 months of this red blood cells), monitoring for FBC and LFTs should be
dosage regimen, and in whom no adverse effects occur, more frequent than outlined above due to the increased
dosage above the 1–3 mg kg)1 daily range might be risk of toxicity. Acute pancreatitis is a rare but well-
considered for a trial period (Grade C; level III). recognized side-effect of azathioprine treatment.43 In
However, if this approach is adopted, care should be azathioprine-treated patients with acute abdominal pain
taken in monitoring for myelosuppression and possible and ⁄ or severe vomiting, acute pancreatitis should be
hepatotoxicity. considered and serum amylase measured.
• If no therapeutic response is observed within Monitoring for azathioprine toxicity should include:
3 months of starting azathioprine, treatment should • Weekly monitoring of FBC and LFTs for the first
usually be withdrawn 4 weeks of therapy, or until the maintenance dose is
• In patients with very low ⁄ absent TPMT activity achieved; reducing to a minimum of once every
(deficient), azathioprine is contraindicated 3 months for the duration of therapy

2004 British Association of Dermatologists, British Journal of Dermatology, 151, 1123–1132

1128 A . V . A N S T E Y et al.

• More frequent monitoring of FBC and LFTs is advised which may be fatal in the elderly. Dermatologists
in patients with hepatic or renal impairment, in the are advised to use the minimum necessary doses of
elderly and in those treated with high doses of immunosuppressive therapies to control immuno-
azathioprine bullous diseases in the elderly10 (Grade A; level
• Increase in dosage of azathioprine should be accom- II-ii)
panied by return to weekly FBC and LFTs for • Live vaccines are contraindicated for patients receiv-
4 weeks, reducing to a minimum of once monthly ing azathioprine (Grade A; level III)
or every 2 months for the duration of therapy • Killed vaccines may elicit a diminished immune
response in patients receiving azathioprine (Grade B;
level II-ii)
Azathioprine-induced susceptibility to
infection
Azathioprine-related malignancy
In transplant recipients, the immunosuppressive activ-
ity of azathioprine in combination with corticosteroids The inhibitory effect of azathioprine on the immune
can lead to an increased susceptibility to viral, surveillance system could, on theoretical grounds,
bacterial and fungal infections which manifest in the lead to an increased rate of malignancy with long-
skin and in other body organs (azathioprine data term therapy. The increased incidence of neoplasms in
sheet). In dermatology, azathioprine is most com- azathioprine-taking immunosuppressed renal trans-
monly prescribed for management of immunobullous plant recipients46 cannot easily be compared with
disorders, where combination with systemic steroids is patients receiving azathioprine for other diseases due
the norm. Evidence of infection in such patients is to the strong antigenic stimulation by the graft and
inconsistent, and in the absence of reliable informa- the taking of two or more immunosuppressive
tion on this topic dermatologists should assume an drugs.47 Importantly, there was no increase in the
increased rate of infection as for transplant recipients. rate of malignancies in a large number of azathio-
Infection in elderly patients with bullous pemphigoid prine-treated, nontransplant patients compared with
treated with azathioprine and prednisolone has been placebo-treated controls.48 Long-term treatment with
identified as a significant cause for mortality, partic- azathioprine for inflammatory bowel disease was not
ularly when compared with rates of such fatal associated with a significant increase in malignancy
infections in patients treated with prednisolone compared with matched controls.34 However, long-
alone.44 term treatment with azathioprine for rheumatoid
The use of azathioprine monotherapy does not arthritis showed an increased rate of lymphoma
appear to give rise to a marked increase in susceptibility which was estimated at one case of lymphoma per
to such infections (azathioprine data sheet). However, 1000 patient years of azathioprine treatment.33 The
the immunosuppressive activity of azathioprine could evidence concerning mutagenicity in relation to
result in an atypical and potentially deleterious azathioprine is unclear, and there is also no clear
response to live vaccines. Thus, administration of live evidence that azathioprine per se is oncogenic in
vaccines to patients receiving azathioprine is contrain- humans. However, the significantly increased rate of
dicated on theoretical grounds. A diminished response skin malignancies in transplant recipients compared
to killed vaccines may occur, and has been observed with the general population is undisputed, and may in
with hepatitis B vaccine in patients treated with a part be due to the immunosuppressive effects of
combination of azathioprine and corticosteroids. In azathioprine. Despite this, reports of skin malignancy
view of the potential severity of primary varicella zoster in patients receiving long-term azathioprine mono-
in the immunosuppressed, patients who have not therapy are rare, suggesting that the risk, if it exists, is
previously had chickenpox should be identified at the probably small.
outset of azathioprine prescribing and advised to seek
• Dermatologists should make patients aware of the
immediate attention if they subsequently come into
possible increased risk of malignancy related
contact with someone who has chickenpox or shin-
to long-term azathioprine therapy (Grade B; level
gles.45
IV)
• Azathioprine in combination with prednisolone is • Skin photoprotection should be advised when rele-
associated with an increased risk of infection, vant (Grade B; level IV)

2004 British Association of Dermatologists, British Journal of Dermatology, 151, 1123–1132

 GUIDELINES FOR PRESCRIBING AZATHIOPRINE IN DERMATOLOGY 1129

dosage reduction or withdrawal of treatment. Prac-

Azathioprine-induced hypersensitivity
tical suggestions on how to manage gastrointestinal
reactions
upset include splitting the dose of azathioprine,
Idiosyncratic hypersensitivity reactions with azathio- starting at a lower dose and then increasing to the
prine are recognized, but are rare.49 Manifestations required dose, and taking azathioprine with, or
include nausea, diarrhoea and vomiting, malaise, shortly after food.
dizziness, fever, rigors, rashes (urticarial, maculopapular
and vasculitic) and even circulatory collapse. The drug
Patient information and informed consent
should be discontinued immediately and circulatory
support initiated if necessary. The aetiology of the Because of the potential for serious adverse effects
reaction is unknown. Rechallenge should be avoided, associated with azathioprine, patients should be care-
as the reaction is generally more severe and occasionally fully counselled before starting treatment. Patients
life-threatening. who are unable to comply with close monitoring are
unsuitable for treatment with this drug. It is recom-
mended that the following points are discussed, an
Management of azathioprine-induced
azathioprine patient information sheet is provided,
complications
and written documentation of this made in the case
Clinical manifestations of chronic azathioprine over- notes.
dosage are those of bone marrow suppression: unex-
1 Azathioprine has a slow onset of action and
plained infection, ulceration of the throat, bruising and
benefit may not be apparent until 2–3 months
bleeding. The most obvious aspect of management of
after starting treatment. The importance of regular
this complication is to take the necessary measures to
blood tests, which are particularly frequent at the
prevent it happening in the first place. This should
start of treatment, should be reinforced. The drug
include measurement of TPMT status, in combination
should be taken once or twice daily, with or after
with monitoring of the FBC as described above. Isolated
food.
lymphopenia is not uncommon with azathioprine
2 The prescribing physician should explain whether
therapy and may be due to lymphocytotoxicity induced
use of azathioprine is for a licensed indication or
by azathioprine-derived imidazole derivatives.5 Dose
not.
reduction is recommended if the lymphocyte count falls
3 Patients should be advised to seek urgent medical
below 0Æ5 · 109 L)1.
attention if they develop signs and symptoms of bone
If bone marrow suppression occurs, the earliest
marrow impairment or liver impairment, such as
feature is leucopenia followed by reduction in platelet
unexplained bruising, sore throat, high fever or
count. Early changes in these parameters with a
jaundice. It is important to explain that symptoms of
downward trend but absolute levels within normal
azathioprine hypersensitivity may initially be mista-
ranges should alert the clinician to the need for
ken for
flu.
continued vigilance in haematological assessment and
4 Patients taking azathioprine who have not had
should raise the option of reduced dosage or with-
chickenpox should seek immediate attention if they
drawal of treatment. More significant myelosuppres-
come into contact with someone who has chicken-
sion with blood indices below the normal ranges should
pox or shingles for consideration of zoster immune
be managed by immediate withdrawal of azathioprine.
globulin. Patients taking azathioprine should not be
Regular monitoring of the FBC will indicate whether
given live vaccines, and due to the risk of orofaecal
the blood count suppression is significant. Platelet
transmission, other members of their household
count below 50 · 10)9 L)1 and neutrophil count
should be given inactive (rather than live) polio
below 1Æ0 · 109 L)1 should be managed jointly with
vaccine.
a haematologist.
5 There may be a small increase in the risk of
Azathioprine-induced hepatotoxicity should be
malignancy with long-term treatment with aza-
diagnosed early (by carrying out frequent monitoring
thioprine. The drug should be continued only when
of LFTs as recommended in these guidelines) and
benefits outweigh the risks.
managed by dosage reduction or withdrawal of
6 Patients should be warned that the following treat-
azathioprine. Gastrointestinal upset is common with
ments may interact with azathioprine:
azathioprine treatment and may occasionally require

2004 British Association of Dermatologists, British Journal of Dermatology, 151, 1123–1132

1130 A . V . A N S T E Y et al.

Allopurinol inhibits the enzyme xanthine oxidase and

Audit of azathioprine prescribing
prolongs the action of azathioprine with potential for
increased toxicity. Concomitant administration This paper includes highlighted recommendations for
should be avoided. good practice in relation to the prescribing of azathio-
Sulfasalazine inhibits TPMT activity and may potentiate prine. Any one of these could be used to establish
azathioprine toxicity. agreed local standards of care against which audit
Warfarin. The anticoagulant effect may be impaired by could be performed. For example,
indications for usage
azathioprine. of azathioprine might be one area, taking into account
Myelosuppressive drugs such as penicillamine and the licensed indications for azathioprine and the
co-trimoxazole should be avoided due to the possi- conditions for which there is good evidence to support
bility of inducing serious haematological toxicity. its use from randomized controlled trials. There is now
Angiotensin-converting enzyme inhibitors have been a broad consensus among U.K. dermatologists that
reported to induce severe leucopenia in patients pretreatment TPMT measurement is necessary for
taking azathioprine. improved safety and more effective dosage selection
Live vaccines are contraindicated on theoretical grounds for patients treated with azathioprine. Thus, audit on
in patients taking azathioprine. usage of TPMT screening could follow the development
7 Pregnancy should be avoided during treatment with of agreed local standards of care concerning this
azathioprine, and women taking this drug must ensure investigation. Routine monitoring of azathioprine tox-
that they use adequate contraceptive precautions. icity is another area where recommendations are
8 Patients should be warned that sudden onset of made, and locally agreed standards could precede an
abdominal pain, with or without severe vomiting, audit to assess compliance with that standard. Perhaps
may be due to pancreatitis related to the azathio- the most novel audit project concerns what we tell our
prine treatment. Patients should seek urgent medical patients. The recommendations for patient information
attention. Serum amylase should be checked in this document exceed what most dermatologists
urgently in this situation. currently do. A patient information sheet has also been
developed by the BAD (available on the patient section
• Before azathioprine is prescribed, the clinician
of the BAD website: http://www.bad.org.uk/patient) to
should provide the patient with an azathioprine
be used in combination with detailed patient-focused
patient information sheet, and discuss the anticipa-
discussion of the merits and hazards of this drug.
ted benefits and possible side-effects

Conclusions Acknowledgments
The authors wish to acknowledge the following con-
Better understanding of azathioprine therapeutics has
tributors: Dr Alex Holme, Dr Simon Meggit, Dr Jeremy
resulted from its continued use for more than 40 years.
Sanderson, Dr Tony Marinaki, Dr Catherine Davies and
Insight into the genetic basis of drug metabolism has
Dr Jonathan Berg. Mrs Joanna Dundon is acknow-
resulted in changes in routine usage of azathioprine
ledged for assistance with the search strategy.
aimed at improving drug efficacy and safety. Measure-
ment of TPMT activity before prescribing azathioprine
combined with alterations in the routine prescribing of References
azathioprine is the best example to date of the evolving
1 Lennard L. TPMT in the treatment of Crohn’s disease with aza-
field of pharmacogenetics.50,51 This is due to the thioprine. Gut 2002; 51: 143–6.
unusual occurrence of a single allelic variant (i.e. that 2 Sandborn W, Sutherland L, Pearson D et al. Azathioprine or
encoding low ⁄ absent TPMT activity) resulting in a 6-mercaptopurine for induction of remission of Crohn’s disease
(Cochrane Review). In: The Cochrane Library, Issue 2. Oxford:
significant clinical effect. Thus, the case for pre-
Update Software, 2003.
azathioprine screening is clear-cut and is increasingly 3 Ansari A, Hassan C, Duley JA et al. Thiopurine methyltransferase
endorsed by clinician demand for this service. There is activity and the use of azathioprine in inflammatory bowel dis-
now the need to reassess azathioprine usage in the face ease. Aliment Pharmacol Ther 2002; 16: 1743–50.
4 Holme SA, Duley JA, Sanderson J et al. Erythrocyte thiopurine
of the change in the way it is prescribed in clinical
methyl transferase assessment prior to azathioprine use in the
practice, by further epidemiological research on ad- UK. Q J Med 2002; 95: 439–44.
verse and beneficial drug effects.52

2004 British Association of Dermatologists, British Journal of Dermatology, 151, 1123–1132

 GUIDELINES FOR PRESCRIBING AZATHIOPRINE IN DERMATOLOGY 1131

5 Sauer H, Hantke U, Wilmanns W. Azathioprine lymphocytotox- 28 Jones JJ, Ashworth J. Azathioprine-induced shock in dermatology
icity. Potentially lethal damage by its imidazole derivatives. Arz- patients. J Am Acad Dermatol 1993; 29: 795–6.
neim-Forsch ⁄ Drug Res 1988; 6: 820–4. 29 Hinrichs R, Schneider LA, Ozdemir C et al. Azathioprine hyper-
6 Weinshilboum RM, Sladek SL. Mercaptopurine pharmacogenet- sensitivity in a patient with peripheral demyelinating polyneuro-
ics: monogenic inheritance of erythrocyte thiopurine methyl- pathy. Br J Dermatol 2003; 148: 1076–7.
transferase activity. Am J Hum Genet 1980; 32: 651–62. 30 Bar J, Stahl B, Hod M et al. Is immunosuppression therapy in
7 Anstey A, Lennard L, Mayou SC, Kirby JD. Pancytopenia related renal allograft recipients teratogenic? A single-center experience.
to azathioprine—an enzyme deficiency caused by a common Am J Med Genet 2003; 116: 31–6.
genetic polymorphism: a review. J R Soc Med 1992; 85: 752–6. 31 The Registration Committee of the European Dialysis and
8 Jeurissen MEC, Boerbooms AMT, van de Putte LBA. Pancytopenia Transplantation Association. Successful pregnancies in women
related to azathioprine in rheumatoid arthritis. Ann Rheum Dis treated by dialysis and kidney transplantation. Br J Obstet
1988; 47: 503–5. Gynaecol 1980; 87: 839–45.
9 Bacon BR, Treuhaft WH, Goodman AM. Azathioprine-induced 32 Davison JM, Dellagrammatikas H, Parkin JM. Maternal
pancytopenia. Occurrence in two patients with connective-tissue azathioprine therapy and depressed haemopoiesis in the babies
disease. Arch Intern Med 1981; 141: 223–6. of renal allograft patients. Br J Obstet Gynaecol 1985; 92:
10 Wojnarowska F, Kirtschig G, Highet AS et al. Guidelines for the 233–9.
management of bullous pemphigoid. Br J Dermatol 2002; 147: 33 Silman AJ, Petrie J, Hazleman B, Evans SJW. Lymphoproliferative
214–21. cancer and other malignancy in patients with rheumatoid arth-
11 Khumalo N, Murrell DF, Wojnarowska F, Kirtschig G. A sys- ritis treated with azathioprine: a 20 year follow-up study. Ann
tematic review of treatments for bullous pemphigoid. Arch Der- Rheum Dis 1988; 47: 988–92.
matol 2002; 138: 385–9. 34 Connell WR, Kamm MA, Dickson M et al. Long-term neoplasia
12 Burton JL, Harman RMM, Peachey RDG, Warin RP. A controlled risk after azathioprine treatment in inflammatory bowel disease.
trial of azathioprine in the treatment of pemphigoid. Br J Dermatol Lancet 1994; 343: 1249–52.
1978; 99 (Suppl. 16): 14. 35 McLeod HL, Krynetski EY, Relling MV, Evans WE. Genetic poly-
13 Harman KE, Albert S, Black MM. Guidelines for the management morphism of thiopurine methyltransferase and its clinical rele-
of pemphigus vulgaris. Br J Dermatol 2003; 149: 926–37. vance for childhood acute lymphoblastic leukemia. Leukemia
14 Aberer W, Wolff-Schreiner EC, Stingl G, Wolff K. Azathioprine in 2000; 14: 567–72.
the treatment of pemphigus vulgaris. J Am Acad Dermatol 1987; 36 Coulthard SA, Rabello C, Robson J et al. A comparison of
16: 527–33. molecular and enzyme-based assays for the detection of thiopu-
15 Lear JT, English JSC, Jones P, Smith AG. Retrospective review of rine methyltransferase mutations. Br J Haematol 2000; 110:
the use of azathioprine in severe atopic dermatitis. J Am Acad 599–604.
Dermatol 1996; 35: 642–3. 37 Snow JL, Gibson LE. The role of genetic variation in thiopurine-
16 Meggitt SJ, Reynolds NJ. Azathioprine for atopic dermatitis. Clin methyltransferase activity and the efficacy and ⁄ or side effects of
Exp Dermatol 2001; 26: 369–75. azathioprine in dermatologic patients. Arch Dermatol 1995; 131:
17 Berth-Jones J, Takwale A, Tan E et al. Azathioprine in severe 193–7.
adult atopic dermatitis: a double-blind, placebo-controlled, 38 Murphy L-A, Atherton D. A retrospective evaluation of
crossover trial. Br J Dermatol 2002; 147: 324–30. azathioprine in severe childhood atopic eczema, using thiopurine
18 Murphy GM, Maurice PM, Norris PG et al. Azathioprine in the methyltransferase levels to exclude patients at high risk of
treatment of chronic actinic dermatitis: a double-blind controlled myelosuppression. Br J Dermatol 2002; 147: 308–15.
trial with monitoring of exposure to ultraviolet radiation. Br J 39 Colombel J-F, Ferrari N, Debuysere H et al. Genotypic analysis of
Dermatol 1989; 121: 639–46. thiopurine S-methyltransferase in patients with Crohn’s disease
19 Yazici H, Pazarli H, Barnes CG et al. A controlled trial of aza- and severe myelosuppression during azathioprine therapy. Gas-
thioprine in Behçet’s syndrome. N Engl J Med 1990; 322: 281–5. troenterology 2000; 118: 1025–30.
20 du Vivier A, Munro DD, Verbov J. Treatment of psoriasis with 40 Dubinsky MC, Yang H, Hassard PV et al. 6-MP metabolite profiles
azathioprine. Br Med J 1974; i: 49–51. provide a biochemical explanation for 6-MP resistance in patients
21 Wisehart JM. Wegener’s granulomatosis—controlled by aza- with inflammatory bowel disease. Gastroenterology 2002; 122:
thioprine and corticosteroids. Br J Dermatol 1975; 92: 461–7. 904–15.
22 Chow RK, Ho VC. Treatment of pyoderma gangrenosum. J Am 41 Guillaume J-C, Vaillant L, Bernard P et al. Controlled trial of
Acad Dermatol 1996; 34: 1047–60. azathioprine and plasma exchange in addition to prednisolone in
23 Hunter GA, Forbes IJ. Treatment of pityriasis rubra pilaris with the treatment of bullous pemphigoid. Arch Dermatol 1993; 129:
azathioprine. Br J Dermatol 1972; 87: 42–5. 49–53.
24 Callen JP, Spencer LV, Burruss JB et al. Azathioprine: an effective, 42 Nielsen OH, Vainer B, Rask-Madsen J. Review article: the treat-
corticosteroid-sparing therapy for patients with recalcitrant ment of inflammatory bowel disease with 6-mercaptopurine or
cutaneous lupus erythematosus or with recalcitrant cutaneous azathioprine. Aliment Pharmacol Ther 2001; 15: 1699–708.
leucocytoclastic vasculitis. Arch Dermatol 1991; 127: 515–22. 43 Sturdevant RAL, Singleton JW, Deren JJ et al. Azathioprine-rela-
25 Lear JT, English JSC. Erosive and generalised lichen planus ted pancreatitis in patients with Crohn’s disease. Gastroenterology
responsive to azathioprine. Clin Exp Dermatol 1996; 21: 56–7. 1979; 77: 883–6.
26 Anstey A. Azathioprine in dermatology; a review in the light of 44 Venning VA, Wojnarowska F. Lack of predictive factors for the
advances in the understanding of methylation pharmacogenetics. clinical course of bullous pemphigoid. J Am Acad Dermatol 1992;
J R Soc Med 1995; 88: 155–60. 26: 585–9.
27 Saway PA, Heck LW, Bonner JR et al. Azathioprine hypersensi- 45 Spencer ES, Anderson HK. Viral infections in renal allograft
tivity. Case report and review of the literature. Am J Med 1988; recipients treated with long-term immunosuppression. Br Med J
84: 960–4. 1979; ii: 829–30.

2004 British Association of Dermatologists, British Journal of Dermatology, 151, 1123–1132

1132 A . V . A N S T E Y et al.

46 Hoover R, Fraumeni JF. Risk of cancer in transplant recipients. Appendix 2. Measurement of erythrocyte thiopurine
Lancet 1973; ii: 55–7. methyltransferase (TPMT) activity
47 Gleichmann E, Gleichmann H. Immunosuppression and neopla-
sia. Klin Wochenschr 1973; 51: 260–5. • TPMT activity should be checked before starting therapy, as aza-
48 McEwan A, Petty LG. Oncogenicity of immunosuppressive drugs. thioprine may induce TPMT enzyme activity
Lancet 1972; i: 326–7. • In cases of doubt, clinicians should discuss the result with the
49 Knowles SR, Gupta AK, Shear NH et al. Azathioprine hypersen- testing laboratory. Genotyping may be helpful in selected cases
sitivity-like reactions—a case report and review of the literature.
• Whole blood is required for analysis (4 mL in ethylenediamine
Clin Exp Dermatol 1995; 20: 353–6.
tetraacetic acid)
50 Ameen M, Smith CH, Barker JNWN. Pharmacogenetics in clinical
• Turnaround time for results varies according to laboratory and
dermatology. Br J Dermatol 2002; 146: 2–6.
demand, but is usually within 2 weeks
51 Krynetski EY, Evans WE. Pharmacogenetics as a molecular basis
for individualized drug therapy: the thiopurine S-methyltransf- • The blood sample should be appropriately packaged and labelled
erase paradigm. Pharm Res 1999; 16: 342–9. and sent by first-class post at room temperature
52 Jick H, Rodriguez LAG, Perez-Gutthann S. Principles of epidemi- • Patients who have recently received blood products can have
ological research on adverse and beneficial drug effects. Lancet misleading TPMT results
1998; 352: 1767–70. • The following two U.K. National Health Service laboratories pro-
53 Griffiths CEM. The British Association of Dermatologists guide- vide TPMT testing. Costs of tests are available on demand from
lines for the management of skin disease. Br J Dermatol 1999; each laboratory:
141: 396–7. The Purine Research Laboratory, 5th Floor, Thomas Guy House,
54 Cox NH, Williams HC. The British Association of Dermatologists Guy’s Hospital, London SE1 9RT, U.K.
therapeutic guidelines: can we AGREE? Br J Dermatol 2003; 148:
Department of Clinical Biochemistry, City Hospital, Dudley Road,
621–5.
Birmingham B18 7QH, U.K.
The details of TPMT testing by the two laboratories are as follows
Appendix 1. Levels of evidence on which the guideline is
based Birmingham City
The consultation process and background details for the British Guy’s Hospital Hospital
Association of Dermatologists (BAD) guidelines have been published Method Tandem mass High-performance
previously.53,54 The patient information sheet to accompany this spectrometry liquid chromato-
guideline is available on the BAD website: http://www.bad.org/ graphy
patient Units pmol h)1 mg)1 nmol h)1 g)1
haemoglobin haemoglobin
Level Type of evidence Very low ⁄ absent < 10 Deficient < 5
Low 10– < 25 Low 6–24
I Evidence obtained from at least one properly designed, Normal 25–50 Normal 25–55
randomized controlled trial High > 50 High > 55
II-i Evidence obtained from well-designed controlled trials
without randomization
II-ii Evidence obtained from well-designed cohort or
case–control analytical studies, preferably from more than The methodology for TPMT testing used by the laboratory at Guy’s
one centre or research group Hospital measures the formation of 6-methyl mercaptopurine from
II-iii Evidence obtained from multiple time series with or the substrates 6-mercaptopurine and S-adenosylmethionine. The 6-
without the intervention. Dramatic results in uncontrolled methyl mercaptopurine product is quantified by tandem mass
experiments (such as the results of the introduction of spectrometry relative to a deuterated 6-methyl mercaptopurine
penicillin treatment in the 1940s) could also be regarded internal standard.
as this type of evidence
The methodology for TPMT testing at Birmingham City Hospital
III Opinions of respected authorities based on clinical
uses 6-thioguanine as substrate and measures the product 6-methyl-
experience, descriptive studies or reports of
thioguanine by high-performance liquid chromatography with
expert committees
fluorescence detection.
IV Evidence inadequate due to problems of methodology
(e.g. sample size, or length of follow-up, or
conflicts of interest)
Grade of
recommendation
A There is good evidence to support the use of the procedure
B There is fair evidence to support the use of the procedure
C There is poor evidence to support the use of the procedure
D There is fair evidence to support the rejection of the use of
the procedure
E There is good evidence to support the rejection of the use of
the procedure

2004 British Association of Dermatologists, British Journal of Dermatology, 151, 1123–1132