We read with great interest the article by Feletar et al.¹ We were surprised by their findings of only a modest response to infliximab treatment in patients with treatment refractory psoriatic arthritis (PsA), as this seems to be in contrast with our observations.

Since 2001 nine patients with refractory PsA according to American College of Rheumatology criteria have been treated with infliximab at our division. All patients had active joint disease with a tender joint count (TJC) and a swollen joint count (SJC) of at least 6, with the exception of one patient (patient 3), whose cervical spine was affected. Quantitative assessment of skin involvement was not available. Table 1 shows further demographic data.

Seven patients received infliximab for more than 52 weeks, and five of those for more than 78 weeks. Clinical assessment, including TJC and SJC, performed by experienced rheumatologists, and laboratory tests (haematology, biochemistry, antinuclear antibodies (ANA), including subsets) was made routinely at each visit (fig 1). During the treatment period toxicity occurred on three occasions, including one labial herpes infection (patient 4), one leucopenia of 2.6×10⁹/l (patient 3), which led to the discontinuation of methotrexate, and one allergic reaction (patient 9). Patient 9 had to stop infliximab treatment permanently despite an excellent clinical response, whereas the other two patients received infliximab again after the adverse events resolved. Six of the nine patients became positive for ANA—titre >1/80 on at least two successive visits. No serum conversion of ANA subsets occurred and no lupus-like symptoms were seen.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

 Changes in (A) TJC, (B) SJC, and (C) C reactive protein levels (normal value <9 mg/l) of each patient during the respective treatment period. Infliximab treatment was stopped because of insufficient response (1) or allergic reaction (2). Disease exacerbation occurred when infliximab was stopped owing to herpes labialis infection (3). Shortening of the infliximab application interval to 4 weeks (4).

Eight of the nine patients showed rapid improvement and a clinical response. Only 1 patient (patient 8) was unresponsive to the treatment regimen, which was stopped. At 52 weeks five patients met the PsA response criteria,² four patients were in complete remission (patients 1, 3, 5, 6) of joint tenderness and swelling, one patient (patient 2) had an incomplete response. Patient 4 stopped infliximab treatment in week 49 for 10 weeks because of a labial herpes infection that was treated with acyclovir. Thereafter tumour necrosis factor blocking therapy was successfully restarted.

At 78 weeks 4/5 patients (patients 1, 3, 4, 5) were in complete remission. One patient (patient 2) who had an incomplete response with a 6 weeks’ treatment interval showed complete remission after the interval was shortened to 4 weeks. Patient 1 who received his first infliximab dose at our division 160 weeks ago had been responding so well to the treatment that we stopped giving infliximab after his 16th infusion at the end of week 120. Because of disease exacerbation, treatment had to be restarted 26 weeks later and within a few days (data not shown) the patient regained complete remission.

These data, although only retrospectively obtained, suggest that infliximab is a valuable therapeutic tool for treating refractory PsA. So far, permanent discontinuation of treatment owing to toxicity or inefficiency has only been necessary for one of the nine patients.

One possible reason for the different outcome in the effect of infliximab on PsA in our patients and those of Feletar et al might be the severity of the joint disease in Feletar’s patients. Fifteen out of 16 of their patients had mutilating arthritis, whereas this form of joint destruction was not present in our group. To clarify this possibility, further studies with larger groups and well defined joint disease are necessary.