PATIENTS AND METHODS

RESULTS

Table 1⇓ shows the demographic and descriptive characteristics of the 21 patients who were enrolled in the pilot study.²⁰ Three patients were classified as having AS with only axial involvement, seven with AS and peripheral arthritis, nine with PsA, and two with uSpA. Of the nine patients with PsA, one also fulfilled the modified New York criteria for AS.

Nineteen patients completed the one year follow up for efficacy. Two patients (one patient with PsA, one with uSpA) changed to another dosing regimen with infliximab owing to partial lack of efficacy. One patient with uSpA had a disease flare at week 12 (patient global assessment of disease activity on a 100 mm visual analogue scale (VAS) 73 mm, four swollen joints, C reactive protein (CRP) 217.6 mg/l). After re-treatment with infliximab (5 mg/kg) at week 16, the patient's disease remained stable, but the swollen joint count and the CRP did not improve. Because there was a new relapse at week 20, it was decided to give infliximab monthly (weeks 22, 26, 30, 34, 38, 42, 46, 50). Moreover as this patient had had a partial response to methotrexate, it was decided to add methotrexate to the infliximab regimen. With this regimen the disease symptoms were adequately controlled (at week 52: VAS global patient score 2 mm, four swollen joints, CRP 5.1 mg/l).

The other patient, with polyarticular PsA, had a disease flare at week 16 (VAS global patient score 88 mm, six swollen joints, CRP 49.3 mg/l). After re-treatment with infliximab (5 mg/kg) at week 18 the patient's disease remained stable, but the swollen joint count and CRP did not improve. Because there was a new relapse at week 30, it was decided to give infliximab monthly (weeks 30, 34, 38, 42, 46, 50), which controlled the disease symptoms adequately (at week 52: VAS global patient score 3 mm, three swollen joints, CRP 20.8 mg/l). However, these two patients remained in the follow up for safety evaluation.

Efficacy evaluation

Global (n=19) and peripheral (n=16) assessments were performed at weeks 20, 22, 28, 34, 36, 42, 48, and 50, and were compared with the baseline value. Axial assessments (n=11) were carried out at weeks 20, 34, 48, and 50. Table 2⇓ summarises the results (data for weeks 28 and 42 are not shown). At week 50 a significant decrease (p≤0.05) for all variables was observed, comparable with the effect seen at week 12. When the global assessments were analysed at different times, this significant decrease was maintained for all variables, except for erythrocyte sedimentation rate (ESR) and CRP at week 48, immediately before the third re-treatment, reflecting a trend towards relapse in some patients. A similar trend was noted for morning stiffness of the spine, spinal pain assessed by the patient on a VAS, and the Dougados functional index.

View this table:

• View inline
• View popup

Table 2

Global, peripheral, and axial assessments at baseline, and at weeks 12, 20, 22, 34, 36, 48, and 50

Figure 1⇓ shows the evolution over this one year period for patient global assessment (100 mm VAS), CRP (mg/l), swollen joint count, and spinal pain assessment (100 mm VAS).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Evolution of patient assessments over time. The box plots show the median value (horizontal line) and range (first to third quartiles in boxes) of the chosen parameter (y axis) over time (weeks). Significance (p) was calculated by the Wilcoxon signed ranks test: *p≤0.05, **p≤0.01, ***p≤0.001, NS, not significant. (A) Patient global assessment (100 mm VAS); (B) C reactive protein (mg/l); (C) swollen joint count; (D) spinal pain assessment (100 mm VAS).

When patients were asked about recurrence of symptoms, reflecting a possible disease relapse, 3/19 (16%) patients reported recurrence of symptoms at week 20, 13/19 (68%) reported a relapse at week 34, and 15/19 (79%) reported a relapse at week 48. However, two weeks after each re-treatment, at weeks 22, 36, and 50, none of the 19 patients reported such disease manifestations. The moment of recurrence of symptoms varied, though for most of these patients (3/3 at week 20, 12/13 at week 34, and 12/15 at week 48) symptoms occurred between weeks 10 and 14 after re-treatment. Nevertheless, none of them reported a full relapse comparable with baseline symptoms.

All variables (global, peripheral, and axial) were evaluated in the different subgroups—namely, AS, AS with peripheral arthritis, PsA, and uSpA (data not shown). All the subgroups responded in the same way to the maintenance treatment with infliximab.

Severity of psoriatic skin involvement was initially evaluated in eight patients with PsA, using the PASI index. The marked improvement in the PASI score noted throughout the three month study was not maintained over the one year follow up, even though there was a trend at weeks 20 and 34. The evolution of the median score (range) for baseline, weeks 20, 34, 48, and 50, respectively, was as follows: 0.72 (0.12–15.8), 0.10 (0.04–3.62), 0.11 (0.02–2.56), 1.79 (0.02–4.80), and 0.92 (0.00–2.4).

DISCUSSION

In this study we report the safety and efficacy of a maintenance regimen of 5 mg/kg infliximab every 14 weeks in patients with active SpA. The patients included in this extension protocol were initially treated with three infusions of infliximab in a 12 week open label pilot study²⁰: after this loading dose regimen, a rapid and significant improvement in global, peripheral, and axial disease manifestations was seen, without major side effects.

From this one year follow up study the following conclusions can be reached. Firstly, a statistically significant decrease of global, peripheral, and axial disease manifestations was seen after one year of treatment in the 19 evaluated patients. Two patients changed to another dosing regimen owing to partial lack of efficacy of the drug. Improvements in BASMI, chest expansion, fingers to floor, and occiput to wall distance in the group of patients with AS, paralleled global improvement of the disease activity in these patients. It is unclear if this improvement in spinal metrology is an indication of an effect on the progressive axial ankylosis which is typical for AS. This study was designed to evaluate the clinical efficacy of infliximab in SpA, and not to evaluate the effect on ankylosis in AS: therefore one cannot draw conclusions from this small subgroup. In order to detect such a structural improvement, homogeneous patient groups, adequate sample size, and sensitive outcome measurements, including radiological assessments evaluated over longer periods, are mandatory.²⁹ During this one year follow up study, general wellbeing was furthermore shown by the fact that a statistically significant increase was found for body weight, as well as for body mass index (data not shown). In addition, in this patient cohort, 42% had a normocytic normochromic anaemia at baseline, probably reflecting severe chronic inflammatory disease, which can be deduced from the patient demographics. Patients had longstanding disease with high ESR, CRP, and marked inflammatory symptoms. After one year's maintenance treatment with infliximab, anaemia was no longer present. A comparable effect, of a dose dependent increase in haemoglobin levels after only four weeks' treatment, has been reported in patients with RA.³⁰ Whether the beneficial effect of infliximab on haemoglobin in our patients reflects the reduction in their disease activity or a possible direct effect of infliximab on erythropoiesis, has yet to be determined.³¹

Secondly, recurrence of symptoms, such as axial night pain, morning stiffness, or synovitis, was reported by three patients (16%) at week 20, 13 patients (68%) at week 34, and 15 patients (79%) at week 48, with the moment of recurrence between 10 and 14 weeks after re-treatment. This relapse was furthermore documented by the fact that ESR, CRP, axial morning stiffness, and axial pain assessed by the patient, were no longer statistically significant when compared with baseline, immediately before the third re-treatment, at week 48. Together with observations made in other open label studies,^32,33 these results indicate that a maintenance regimen of 5 mg/kg infliximab every 14 weeks could not control the inflammatory disease activity continuously. Adjustment of the maintenance regimen seems warranted; however, it is not clear whether this should be achieved by increasing the dose or shortening the interval between doses.

Thirdly, re-treatment of patients whose symptoms recurred induced an improvement comparable with the response seen after the induction scheme reported in the 12 week pilot study. All variables of disease activity were again significantly improved two weeks after every re-treatment.

Fourthly, the treatment with infliximab in this patient group was shown to be safe and the infusion procedure was generally well tolerated. Only one mild self limiting infusion reaction was seen, and no major infectious episodes or malignancies occurred. Previous studies indicate that treatment with TNFα inhibitors, such as infliximab, can induce the development of ANA in patients with RA.³⁴ So far no data on the occurrence of such antibodies in patients with SpA treated with infliximab have been reported: here, up to 12/21 (57%) patients developed ANA; in four of them (19%) anti-dsDNA antibodies were also detected. However, no significant laboratory changes in peripheral blood count or complement occurred in these patients; no patients developed proteinuria, nor where there any other lupus-like symptoms. The overall frequency of autoantibodies is comparable with the data reported in RA.³⁴ However, we have to remember that the background for developing such antibodies—in RA versus SpA—is different. Occurrence of ANA in RA is not considered unusual and a prevalence up to 60% has been seen³⁵; in SpA and AS, however, only scarce data are available, showing a prevalence ranging from 11.4% to 19.1%.^36,37 In these studies a bias attributed to the use of sulfasalazine is likely. A retrospective analysis of a patient group with SpA disclosed a prevalence of 8% ANA positive patients before sulfalsalazine treatment with no anti-dsDNA antibodies detected. After two to five years of sulfasalazine treatment, the incidence rose to 24% and 4%, respectively for ANA and anti-dsDNA antibodies.³⁸

In conclusion, this one year open pilot study suggests that infliximab is a safe and effective drug in the treatment of SpA. Regardless of the fact that these data need further confirmation in ongoing double blind, placebo controlled trials, the extent of the clinical improvements, their consistency throughout the study group, the parallel changes in laboratory indices of inflammation, and the maintenance of this therapeutic effect over one year, are encouraging. Further accurate follow up of this patient group is mandatory for the optimisation of the maintenance regimen and for the evaluation of longer term safety with regard to infections, malignancies, and the occurrence of ANA.