Available online at www.sciencedirect.com

Joint Bone Spine 75 (2008) 315e318

http://france.elsevier.com/direct/BONSOI/

Original article

Onset or exacerbation of cutaneous psoriasis during TNFaantagonist therapy

Daniel Wendling a,*, Jean-Charles Balblanc b, Daniel Briancon c, Alain Brousse d,Anne Lohse b, Philippe Deprez e, Philippe Humbert f, Francois Aubin f

a Rheumatology Department, CHU Jean Minjoz, Minjoz Teaching Hospital, Franche-Comte University, Boulevard Fleming, 25030 Besancon Cedex, Franceb Rheumatology Department, Hospital Center, Belfort, France

c Rheumatology Department, Reine Hortense Hospital Center, Aix-les-Bains, Franced Rheumatology Department, Hospital Center, Dole, France

e Dermatology Office, Dole, Francef Dermatology Department, St Jacques Teaching Hospital, Franche-Comte University, Besancon, France

Accepted 22 June 2007

Available online 11 February 2008

Abstract

The widespread use of TNFa antagonists in recent years has led to the recognition of paradoxical adverse effects, defined as the onset orexacerbation of disorders that are usually improved by TNFa antagonists. Cutaneous psoriasis is an example, of which several cases havebeen reported.Objective: To identify cases of psoriasis onset or exacerbation during TNFa antagonist therapy and to look for potential predictive factors.Methods: We retrospectively reviewed cases of psoriasis onset or exacerbation during TNFa antagonist therapy. For each case we recorded thefollowing data: age, sex, underlying disease, nature of the TNFa antagonist, effectiveness in improving the underlying disease, history ofpsoriasis in the patient or family, time to psoriasis development, type of psoriasis (confirmed by an experienced dermatologist), concomitanttreatments, whether the TNFa antagonist was stopped or continued, and the outcome of the psoriasis. These data were compared to those inthe literature.Results: We identified 12 patients, six men and six women, with a mean age of 45.5 years. The TNFa antagonist was adalimumab in fourpatients, etanercept in six patients, and infliximab in two patients. The underlying disease was ankylosing spondylitis in six cases, rheumatoidarthritis in four, and psoriatic arthritis in two. Mean time from treatment initiation to psoriasis was 4.1 months (range, 1e15 months). A previoushistory of psoriasis in the patient was noted in six cases, including four of the six patients taking etanercept. TNFa antagonist therapy waseffective on the underlying disease in 11 of the 12 patients. The drug was discontinued in five patients, of whom four experienced resolutionof their psoriasis. In the remaining seven patients, the drug was continued and the skin lesions remained unchanged. Most of the patients hadpsoriasis vulgaris (plaque psoriasis); palmoplantar pustulosis was a feature in five patients.Discussion: Over 40 cases of psoriasis onset or exacerbation during TNFa antagonist therapy have been reported in the literature. The prevalenceof this adverse effect has been estimated at 1.5e5% of patients taking TNFa antagonists. The findings from our case series are consistent withdata in the literature. Psoriasis is a class effect that has been reported with all the currently available TNFa antagonists. The skin lesions developwithin the first few months of therapy. Patients with a wide range of underlying diseases can be affected. Palmoplantar pustulosis is a commonfeature. A previous history of psoriasis seems more common in patients who experience psoriasis onset or exacerbation during etanercepttherapy (four of six patients in our study and 55% in the literature); thus, previous psoriasis may be a risk factor for psoriasis exacerbation duringetanercept therapy.� 2008 Elsevier Masson SAS. All rights reserved.

Keywords: TNFa antagonists; Psoriasis; Adverse drug effects

* Corresponding author. Tel.: þ33 3 81 66 82 41; fax: þ33 3 81 66 86 86.

E-mail address: dwendling@chu-besancon.fr (D. Wendling).

1297-319X/$ - see front matter � 2008 Elsevier Masson SAS. All rights reserved.

doi:10.1016/j.jbspin.2007.06.011

316 D. Wendling et al. / Joint Bone Spine 75 (2008) 315e318

1. Introduction

TNFa antagonists have been found effective in variousrheumatic and nonrheumatic diseases. The increasing use ofthese agents has led to the recognition of several paradoxicaladverse effects. One example is the new onset or exacerbationof cutaneous psoriasis. TNFa antagonists are effective in treat-ing psoriasis and are licensed for use in patients with severeforms of the disease. We retrospectively identified 12 casesof psoriasis onset or exacerbation during TNFa antagonisttherapy.

2. Methods

We conducted a retrospective observational study ofpatients with new onset or exacerbation of cutaneous psoriasisduring treatment with TNFa antagonists for a variety ofdiseases. All the patients were evaluated at our hospitaldepartments.

For each patient, we recorded the following data: age, sex,underlying disease with its duration, time from TNFa antago-nist initiation to psoriasis onset or exacerbation, type anddosage of the TNFa antagonist, concomitant treatments,previous history of psoriasis in the patient or family, type ofpsoriasis, extent and severity of the skin lesions, effectivenessof TNFa antagonist therapy on the underlying disease, theopinion of the dermatologist, whether the drug was stoppedor continued, and the outcome of the skin lesions.

3. Results

We identified 12 patients, whose main characteristics areshown in Table 1. There were six women and six men, witha mean age of 45.5 years. Of the 12 patients, six were takingetanercept, four adalimumab, and two infliximab. The drugswere used in standard dosages. The underlying disease wasankylosing spondylitis or another spondyloarthropathy in sixpatients, rheumatoid arthritis in four patients, and psoriaticarthritis in two patients. A previous history of psoriasis wasnoted in six patients, including four of the six taking etaner-cept. One patient reported psoriasis in a family member.Concomitant medications were noted in seven patients andconsisted of nonsteroidal antiinflammatory drugs in five.

Mean time from TNFa antagonist initiation to onset orexacerbation of psoriasis was 4.1 months (range, 1e15months). TNFa antagonist therapy was effective in improvingthe signs and symptoms of the underlying disease in 11 of the12 patients. Most of the patients had psoriasis vulgaris (plaquepsoriasis). Palmoplantar pustulosis was noted in five patients,psoriasis of the scalp in one patient, and hyperkeratosis in onepatient. The diagnosis was confirmed by an experienceddermatologist in all 12 patients. The psoriasis was consideredmoderate in six patients and severe in six patients.

The TNFa antagonist was continued in seven patients,among whom one responded well to topical treatment andsix experienced no change in their skin lesions. Of the five pa-tients who stopped TNFa antagonist therapy, four experienced

resolution of their skin lesions. Re-challenge testing was notperformed.

4. Discussion

We report on 12 cases of psoriasis onset or exacerbationduring TNFa antagonist therapy for a variety of diseases.Three different TNFa antagonists were involved.

This side effect may seem paradoxical, given that TNFaantagonists are used to treat severe forms of psoriasis. About40 similar cases have been reported in the medical literature[1e10]. Psoriasis occurred during six of 400 TNFa antagonisttreatments at a rheumatology center [2] and in five of 100patients treated by Sfikakis et al. [10].

Psoriasis onset or exacerbation has been reported withvarious TNFa antagonists. Thus, of 30 cases in a literature re-view [11], 14 occurred with infliximab, eight with etanercept,and eight with adalimumab. Of these 30 patients, 18 hadrheumatoid arthritis, eight ankylosing spondylitis, and twoinflammatory bowel disease [11]. Psoriasis has been reportedin patients treated for Behcet’s disease [10]. Finally, exacerba-tion of palmoplantar pustulosis has been described in patientswith SAPHO syndrome treated with TNFa antagonists [12].

Among the 30 patients in the above-mentioned literaturereview [11], six had a history of psoriasis and two reportedpsoriasis in their family. The proportion of patients with a his-tory of psoriasis varied across TNFa antagonists, from 55%with etanercept to 14% with infliximab and 10% withadalimumab [9]. Similarly, in our study four of the six patientson etanercept had a previous history of psoriasis.

The time from TNFa antagonist initiation to psoriasis onsetor exacerbation ranged from a few days to 48 months [9].However, most cases occurred between 1 and 3 months aftertreatment initiation. Most of the patients had a good responseto TNFa antagonist therapy at the time of the skin reaction.

Several forms of psoriasis can be induced by TNFa antag-onist therapy. However, palmoplantar pustulosis seemscommon, with five of 12 cases in our study. In the previousliterature review [11], 19 of 30 patients had pustular psoriasisand 10 had psoriasis vulgaris. Nail involvement was noted inthree of five patients [11].

The outcome was variable, although the skin lesionsusually persisted unchanged when that TNFa antagonist wascontinued (23/30) [11]. A few patients experienced resolutionof the skin lesions despite treatment continuation [3]. In somecases, the skin lesions prompted discontinuation of TNFa an-tagonist therapy [10]; this occurred in five of our 12 patients.

Psoriasis onset or exacerbation seems to be a class effect, ascases have occurred with all the available TNFa antagonists[7]. In several cases, the same patient experienced psoriasiswith two different TNFa antagonists [3,9,10]. In two patients[9], however, skin lesions occurred with a monoclonal anti-body but not the soluble receptor or vice versa. Similarly,two of our patients (#4 and #10) experienced psoriasis duringetanercept therapy after 4 and 3 years, respectively, of inflixi-mab therapy without psoriasis.

Table 1

Recapitulation of the study cases

Patient

1 2 3 4 5 6 7 8 9 10 11 12

Age 22 54 46 49 43 44 39 46 62 49 49 43

Sex F M M M F F F F M M M F

Underlying disease &

duration (years)

SpA (1) AS (20) AS RCH (10) AS (20) RA (10) PsA (12) RA (30) RA (8) PsA (15) AS (20) AS (30) RA (5)

Time from anti-TNF

initiation to

psoriasis (months)

1 1 2 3 1.5 10 3 3 15 3 6 2

Anti-TNF Adalimumab Adalimumab Adalimumab Etanercept Etanercept Etanercept Adalimumab Etanercept Infliximab Etanercept Infliximab Etanercept

Dosage 40 mg/15 d 40 mg/15 d 40 mg/15 d 50 mg/wk 50 mg/wk 50 mg/wk 40 mg/15 d 50 mg/wk 5 mg/kg 50 mg/wk 5 mg/kg 50 mg/wk

Concomitant

medications

0 0 0 NSAID MTX Leflunomide MTX 0 0 NSAID NSAID MTX

NSAID NSAID Glucocorticoids

Glucocorticoids

Previous history

of psoriasis

Yes No No No Yesþ family Yes No Yes Yes No No Yes

Type of psoriasis PPP Scalp Plaque Plaque Palmoplantar Limbs Plaque Plaque Plaque Plaque Hyperkeratotic Palmoplantar

Pustular

Psoriasis

Pustular

psoriasis

Lower limbs Soles

of feet

Palmoplantar

Extent/severity Moderate Extensive Moderate Extensive Mild Severe

Anti-TNF

effective on

underlying

disease

Yes Yes Yes Yes Yes Yes Yes Somewhat

effective

Yes Yes Yes Yes

Evaluated by

a dermatologist

Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes

Outcome Stable Favorable Stable Stable Favorable Favorable Persistence Stable Flares Stable Improvement Improvement

Anti-TNF stopped No No No No Yes Yes Yes No No No Yes Yes

F, female; M, male; SpA, spondyloarthropathy; AS, ankylosing spondylitis; UC, ulcerative colitis; RA, rheumatoid arthritis; PsA, psoriatic arthritis; NSAIDs, nonsteroidal antiinflammatory drugs; MTX, meth-

otrexate; PPP, palmoplantar pustulosis.

31

7D

.W

endlinget

al./

JointB

oneSpine

75(2008)

315e318

318 D. Wendling et al. / Joint Bone Spine 75 (2008) 315e318

A causal link between TNFa antagonists and the onset orexacerbation of psoriasis seems likely. Arguments that supportcausality include the timing of the skin reaction relative totreatment initiation, the number of reported cases, the resolu-tion of the skin lesions after treatment discontinuation [9], andreports of positive re-challenge tests [5,10].

The development of psoriasis-like lesions in a patient tak-ing TNFa antagonist therapy may suggest a number of diagno-ses [13] including infection, cutaneous lupus, a malignancy, orvasculitis [14]. In some patients, a skin biopsy may be requiredto establish the diagnosis.

The mechanism underlying psoriasis onset or exacerbationduring TNFa antagonist therapy is not agreed on. TNFaantagonist promotes infection, which may trigger the skindisorder [11], in keeping with the high rate of palmoplantarpustulosis. Alternatively, a treatment-induced cytokineimbalance may be involved: inhibition of TNFa can induceoverexpression of cutaneous interferon alpha, which in turnpredisposes to psoriasis [15].

No risk factors for psoriasis during TNFa antagonisttherapy have been identified to date. Nevertheless, data fromthe literature suggest a higher risk of psoriasis exacerbationwith etanercept and a higher risk of new-onset psoriasis withmonoclonal antibodies to TNFa [9].

Other paradoxical side effects of TNFa antagonist therapy,defined as the occurrence of disorders that usually improvewith TNFa antagonists, include vasculitides [16] and inflam-matory bowel disease [17]. The development of a paradoxicalside effect does not always require treatment discontinuation,and neither does it correlate with lack of efficacy.

References

[1] Aslanidis S, Pyrpasopoulou A, Leontsini M, et al. Anti-TNF-alpha-

induced psoriasis: case report of an unusual adverse event. Int J Dermatol

2006;45:982e3.

[2] Cassano N, Coviello C, Loconsole F, et al. Psoriasis exacerbation after

a flu-like syndrome during anti-TNF-alpha therapy. Eur J Dermatol

2006;16:316e7.

[3] Cohen JD, Bournerias I, Buffard V, et al. Psoriasis induced by tumor

necrosis factor-alpha antagonist therapy: a case series. J Rheumatol

2007;34:380e5.

[4] Goiriz R, Dauden E, Perez-Gala S, et al. Flare and change of psoriasis

morphology during the course of treatment with tumour necrosis factor

blockers. Clin Exp Dermatol 2007;32:176e9.

[5] Kary S, Worm M, Audring H, et al. New onset or exacerbation of psoriatic

skin lesions in patients with definite rheumatoid arthritis receiving tumour

necrosis factor alpha antagonists. Ann Rheum Dis 2006;65:405e7.

[6] Pirard D, Arco D, Debrouckere V, et al. Anti-tumor necrosis factor alpha-

induced psoriasiform eruptions: three further cases and current overview.

Dermatology 2006;213:182e6.

[7] Richette P, Viguier M, Bachelez H, et al. Psoriasis induced by anti-tumor

necrosis factor therapy: a class effect? J Rheumatol 2007;34:438e9.

[8] Roux CH, Brocq O, Leccia N, et al. New-onset psoriatic palmoplantaris

pustulosis following infliximab therapy: a class effect? J Rheumatol

2007;34:434e7.

[9] Sari I, Akar S, Birlik M, et al. Anti-tumor necrosis factor-alpha-induced

psoriasis. J Rheumatol 2006;33:1411e4.

[10] Sfikakis PP, Iliopoulos A, Elezoglou A, et al. Psoriasis induced by anti-

tumor necrosis factor therapy: a paradoxical adverse reaction. Arthritis

Rheum 2005;52(8):2513e8.

[11] Ritchlin C, Tausk F. A medical conundrum: onset of psoriasis in patients re-

ceiving anti-tumour necrosis factor agents. Ann Rheum Dis 2006;65:1541e4.

[12] Massara A, Cavazzini PL, Trotta F. In SAPHO syndrome anti-TNF-alpha

therapy may induce persistent amelioration of osteoarticular complaints,

but may exacerbate cutaneous manifestations. Rheumatology (Oxford)

2006;45:730e3.

[13] Flendrie M, Vissers WH, Creemers MC, et al. Dermatological conditions

during TNF-alpha-blocking therapy in patients with rheumatoid arthritis:

a prospective study. Arthritis Res Ther 2005;7:R666e76.

[14] Wendling D, Streit G, Lehuede G, et al. Cutaneous lymphocytic vasculi-

tis during TNFalpha antagonist therapy for polyarthritis. Joint Bone

Spine 2006;73:215e6.

[15] Dutz JP. Tumor necrosis factor-a inhibition and palmoplantar pustulosis:

Janus-faced therapy? J Rheumatol 2007;34:247e9.

[16] Saint Marcoux B, De Bandt M. Vasculitides induced by TNFalpha antag-

onists: a study in 39 patients in France. Joint Bone Spine 2006;73:710e3.

[17] Oh J, Arkfeld DG, Horwitz DA. Development of Crohn’s disease in a pa-

tient taking etanercept. J Rheumatol 2005;32:752e3.