CASE REPORT

Treatment of a 51 year old woman with severe seropositive erosive RA with methotrexate, sulfasalazine, Myocrisin (sodium aurothiomalate), d-penicillamine, hydroxychloroquine, and leflunomide had previously failed and she continued to require repeated courses of systemic corticosteroids to control her disease. She started etanercept, receiving twice weekly doses of 25 mg subcutaneously, and achieved a good clinical response, with a significant reduction in disease activity scores. Other drugs taken were meloxicam and co-codamol. The etanercept was withheld for a total of four weeks during an intercurrent chest infection, but no other side effects were noted until after seven months of etanercept treatment.

At this stage the patient developed a photosensitive “burning” rash on the exposed areas of her face, neck, and arms. She felt generally lethargic, and described Raynaud’s phenomenon. She had clinically active synovitis. There was no evidence of sepsis. An antinuclear antibody test, which had previously been negative, was now strongly positive (titre 1 in 320 by indirect immunofluorescence on rat liver sections), and dsDNA antibodies were present (9.2 mg/l by the Farr assay) (normal range 0–8). Antinucleosome antibodies were strongly positive by enzyme linked immunosorbent assay (ELISA; Organtec) at 180 U/ml (normal range 0–20 U/ml). Total white cell count had fallen to 1.62×10⁹/l (neutrophils 0.58, lymphocytes 0.88). Haemoglobin and platelet counts were normal. Vasculitic screen was negative, while rheumatoid factor remained positive. There was a polyclonal rise in immunoglobulins, and erythrocyte sedimentation rate was raised (46 mm/1st h). Renal and liver function tests were normal.

A diagnosis of SLE was made and her etanercept was discontinued. She received three intramuscular injections of triamcinolone 40 mg over three weeks, and treatment with prednisolone 10 mg/day and alendronate 70 mg/week was started. Three weeks after stopping etanercept and starting corticosteroids the rash had completely disappeared, the patient was feeling generally better. The synovitis had also improved. After three months the total white cell count had increased to 3.63×10⁹/l (neutrophils 2.21, lymphocytes 1.07). A yellow card reporting this probable drug reaction was completed.

DISCUSSION

We have described the new onset of SLE in a patient receiving etanercept for RA. Before receiving etanercept the patient had, on a number of occasions, had a total white cell count between 2.5 and 4×10⁹/l (normal range 4–11). An ultrasound scan of the abdomen had shown a mild degree of splenomegally of normal consistency in keeping with Felty’s syndrome. There were no other prior features of SLE. In particular, antinuclear antibody and anti-double stranded DNA antibody tests were negative (on two separate occasions), there was no history of rash or photosensitivity, and no history of renal or neurological disease. The improvement in symptoms, signs, and white cell count on stopping the etanercept would be in keeping with a case of drug-induced SLE. The alternative explanation is that the pre-existing leucopenia was a manifestation of a pre-existing lupus-like disease which was unmasked by the etanercept. This would be supported by occasional pre-etanercept episodes of lymphopenia, which is less common in Felty’s syndrome.

Why should TNFα blocking therapy cause or unmask SLE? One possible explanation is that TNFα may up regulate the cellular expression of the adhesion molecule CD44, which has a role in the clearance of apoptotic neutrophils by phagocytes.^7,8 Impaired clearance of apoptotic cells, and reduced leucocyte CD44 expression, have been described in SLE.^9,10 This is, however, a complex area which requires clarification by further study. There is no doubt that etanercept and infliximab represent great advances in the treatment of RA. This case serves to illustrate that patients receiving these drugs should be closely monitored, and caution exercised in any patient with lupus-like or leucopenic disease features.