- Top of page
- PATIENTS AND METHODS
Psoriatic arthritis (PsA) has long been considered a relatively mild nondeforming arthropathy. Its treatment consists of nonsteroidal antiinflammatory drugs (NSAIDs) and local corticosteroid injections, with second line drugs being reserved for NSAID-resistant or progressively destructive forms. However, this view has been challenged over the last decade. One of the largest series of PsA patients revealed the development of erosive and deforming arthritis in 40% of cases, a rate similar to that observed in rheumatoid arthritis (RA) (1–3).
Methotrexate (MTX), sulfasalazine (SSZ), and cyclosporin A (CSA) are the most widely used symptom modifying antirheumatic drugs (SMARDs), also in combination, in the treatment of PsA (4–7). Tumor necrosis factor α (TNFα) inhibition has been shown to be effective in spondylarthropathies, including PsA (8–15). As in the case of RA (16, 17), patients with aggressive PsA who have failed therapy with SMARDs may be treated with the addition of anti-TNFα agents (11).
We treated 16 patients with resistant PsA with the addition of infliximab to previous MTX therapy.
PATIENTS AND METHODS
- Top of page
- PATIENTS AND METHODS
This was a 30-week, multicenter, prospective, open pilot study evaluating the efficacy and tolerability of infliximab administered at a dosage of 3 mg/kg at 0, 2, 6, 14, 22, and 30 weeks while continuing MTX therapy.
The study protocol was reviewed and approved by the ethical committees of the participating centers. Before entering the trial, each potential study participant was informed of the nature, duration, and purpose of the study, and all the potential benefits, inconveniences, and hazards that could reasonably be expected.
Sixteen consecutive PsA patients, aged 16–65 years, with active peripheral arthritis (≥5 swollen and ≥5 tender joints, pain on a visual analog scale [VAS] >25 mm) and an erythrocyte sedimentation rate (ESR) >25 mm/hour despite at least 6 months of MTX therapy at a stable dosage of 10–15 mg per week were enrolled. Table 1 shows the baseline demographic, clinical, and laboratory characteristics of the patients. Clinical and radiographic evidence of active axial involvement was present in 3 patients, who had inflammatory back pain according to the criteria of Calin et al (18) and bilateral sacroiliitis grade 2–3.
Table 1. Demographic, clinical, and laboratory features of the patients at study onset*
| ||Median (range)|
|Age, years||51.5 (32–69)|
|Age at onset of arthritis, years||41.0 (22–61)|
|Duration of arthritis, months||142.0 (12–245)|
|Number of painful joints||10.0 (5–29)|
|Number of swollen joints||6.5 (5–14)|
|Pain, VAS, mm||71.0 (50–87)|
|Morning stiffness, minutes||140.0 (15–240)|
|Dougados articular index||17.5 (1–26)|
|Patient's general evaluation||3.0 (1–4)|
|Physician's general evaluation||3.0 (1–4)|
|ESR (mm/first hour)||44.0 (27–80)|
|CRP (mg/dl)||1.79 (0.3–8.4)|
|Daily dosage of prednisone, mg||4.0 (0–7.5)|
|Concomitant therapy at baseline|
| NSAIDs, %||100|
| Corticosteroids, % (n)||62.5 (11)|
The diagnosis of psoriasis was confirmed by an expert dermatologist.
All 16 patients had been previously treated during the disease course with one or more SMARD administered alone or in association: 14 patients received MTX alone (median 15 mg/week, range 10–15 mg/week), 9 received CSA (median 3.5 mg/day, range 3–4 mg/day), 7 received SSZ (800 mg/day), 2 received gold sodium thiomalate (50 mg/week), and 3 received MTX (10 mg/week) plus CSA (4 mg/day).
All 16 patients at the time of the study entry were undergoing MTX therapy; infliximab was added to this. All patients were being treated with NSAIDs and 11 also were taking prednisone (≤7.5 mg/day). The patients taking prednisone had to be on a stable dose for at least 1 month before entry.
Patients were seen for clinical evaluation by the same investigator at baseline and weeks 2, 6, 14, 22, and 30 for the study variables. The patient self assessment measures included severity of pain (current global level of pain) using a 100-mm VAS; duration of morning stiffness; global disease assessment graded on a 5-point scale (0 = none, 1 = mild, 2 = moderate, 3 = severe, 4 = very severe); and arthritis-related functional disability measured by the Health Assessment Questionnaire (HAQ) and the Dougados functional index (19).
The clinical assessments included the number of tender and swollen joints (57 and 54 sites respectively); the number of fingers showing dactylitis, defined as the presence of tenderness and swelling of entire digits; and a physician global disease assessment graded on a 5-point scale (0 = none, 1 = mild, 2 = moderate, 3 = severe, 4 = very severe). We evaluated the following swollen joints (54 sites): right (R) and left (L) temporomandibular, R and L shoulder, R and L sternoclavicular, R and L elbow, R and L wrist, R and L metacarpophalangeal, R and L hand proximal interphalangeal, R and L hand distal interphalangeal, R and L knee, R and L ankle, R and L metatarsophalangeal, R and L first toe interphalangeal. We evaluated 57 tender joints: the same sites evaluated for swollen joints plus R and L hip joints and cervical spine.
In the patients with axial involvement, spinal pain (100-mm VAS) and spinal morning stiffness were also assessed. In these patients, mobility impairment was evaluated by chest expansion, a modified Schober's test, finger-to-floor distance, and cervical spine flexion/extension distances (20).
A dermatologist examined the extension and severity of cutaneous psoriasis at each visit. The evaluation was performed using the Psoriasis Area Severity Index (PASI), which summarizes the degree of erythema, desquamation, infiltration, and the percentage of body surface area involved (21).
Laboratory evaluations were made at every visit and included biochemical surveys, routine hematologic variables, urinalysis, ESR according to Westergren's method, and C-reactive protein (CRP) levels.
The patients were screened for adverse events at every visit. A 5-mg reduction in the dose of MTX was allowed for patients whose serum aminotransferase concentrations increased to at least 2.0 times the upper limit of the normal range. All patients also received 1 mg of folic acid per day. The dose of infliximab was not reduced.
Each patient was considered a treatment responder or nonresponder on the basis of the American College of Rheumatology (ACR) improvement criteria for RA with 20%, 50%, and 70% improvement (22, 23) using ESR or CRP.
The changes in outcome measures from baseline to the second and 30th weeks were expressed as mean values with their 95% confidence intervals. Significance of the change from baseline was measured by the Wilcoxon signed rank test (P < 0.05). Fisher's exact test was used when necessary.
- Top of page
- PATIENTS AND METHODS
Efficacy evaluations at weeks 2 and 30 are shown in Table 2. Significant reductions in the swollen and tender joint count, pain score, HAQ, Dougados functional index, and patient's and physician's global assessment were registered at the 2-week visit and persisted throughout the study.
Table 2. Mean variation of clinical and laboratory variables during followup*
| ||Week 2 versus time 0 (95% CI)||P (Wilcoxon)||Week 14 versus time 0 (95% CI)||P (Wilcoxon)||Week 30 versus time 0 (95% CI)||P (Wilcoxon)|
|Δ number of tender joints||−71% (−32, −100)||0.002||−76% (−43, −100)||0.002||−65.5% (−18, −100)||0.02|
|Δ number of swollen joints||−72% (−43, −100)||0.004||−77% (−43, −100)||0.001||−81% (−40, −100)||0.003|
|Δ pain, VAS||−26% (−24, 38)||0.003||−66% (−45, −87)||0.001||−33% (−24, 38)||0.004|
|Δ HAQ||−39% (−17, −97)||0.03||−89% (−31, −100)||0.03||−85% (−23, −100)||0.05|
|Δ Dougados functional index||−49% (−11, −87)||0.007||−64% (−25, −100)||0.006||−54% (−6, −100)||0.03|
|Δ patient global evaluation||−26% (−10, −43)||0.01||−58% (−29, −87)||0.002||−65% (−26, −100)||0.008|
|Δ physician global evaluation||−38% (−23, −52)||0.004||−58% (−34, −84)||0.001||−78% (−47, −100)||0.001|
|Δ ESR||−56% (−33, −79)||0.001||−57% (−31, −83)||0.001||−47% (−12, −83)||0.01|
|Δ CRP||−70% (−28, −100)||0.001||−68% (−24, −100)||0.003||−48% (−18, −100)||0.03|
No patient developed dactylitis during the study period.
In the 3 patients with axial disease, spinal pain and stiffness almost completely resolved at day 14 and the improvement persisted throughout the study period (Table 3). A significant decrease of ESR and CRP was similarly observed at day 14 and was maintained during the study period (Table 2).
Table 3. Evolution of the axial signs and symptoms during the study period in the 3 patients with psoriatic spondylitis*
| ||Patient 1||Patient 2||Patient 3|
|Time 0||Week 2||Week 30||Time 0||Week 2||Week 30||Time 0||Week 2||Week 30|
|Spinal pain, VAS, mm||70||16||3.5||65||0||0||50||10||0|
|Morning stiffness, minutes||90||0||15||160||15||0||150||0||30|
|Dougados articular index||20||7||5||7||3||0||4||1||1|
|Disease duration, months|| ||165|| || ||124|| || ||110|| |
|Age at study onset, years|| ||64|| || ||33|| || ||36|| |
|Sex|| ||Male|| || ||Male|| ||Male|| |
At day 14, the percentages of the patients satisfying ACR 20% and ACR 50% response rates were 67% (10 of 15 patients) and 40% (6 of 15 patients). No patients satisfied the ACR 70% response criteria. At week 14, the percentages of the patients satisfying ACR 20%, ACR 50%, and ACR 70% were 60% (9 of 15 patients), 47% (7 of 15 patients), and 47% (7 of 15 patients), respectively. At the end of study period, 9 of 14 patients (64%) satisfied ACR 20% and 8 of 14 patients (57%) satisfied both ACR 50% and ACR 70% response rates.
The 14 patients who completed the study period were divided in 2 groups based on baseline CRP values. The median value was selected as the cutoff point (2 mg/dl). At the end of the study period, we used this cutoff point to compare the frequencies of patients with at least 70% improvement in clinical variables considered in Table 2. Patients with baseline CRP value ≥2 mg/dl had a statistically significant more frequent improvement in the number of tender joints (100% versus 33.3%; P = 0.02), pain score (100% versus 28.6%; P = 0.02), Dougados functional index (100% versus 16.7%; P = 0.02), patient global evaluation (100% versus 14.3%; P = 0.005), and physician global evaluation (100% versus 28.6%; P = 0.02) compared with patients with baseline CRP value <2 mg/dl.
Psoriasis, measured by PASI index, improved by 37% at week 2 and 86% at week 30. There were no changes in NSAID or corticosteroid use during the study period.
No patient dropped out due to treatment failure. Side effects were observed in 4 of 16 patients, 2 of whom suspended the therapy for a severe infusion reaction (dyspnea, broncospasm, and hypotension): 1 during the first infusion, and the second during the fifth (14th week). In these 2 patients, symptoms were not controlled by slowing down the infusion rate. In the other 2 patients, infusion reaction was mild (headache and nausea) and symptoms were controlled by slowing down the infusion rate. Prophylactic antihistamines were not used. No patient had infections requiring antibiotic treatment. Minor, uncomplicated infections of the upper respiratory tract occurred in 3 patients. No patient reduced the weekly MTX dosage for aminotransferase elevation.
- Top of page
- PATIENTS AND METHODS
TNFα has been shown to be increased in synovial fluid and synovial membranes in patients with PsA (24, 25). Therefore, TNFα blockade, shown to be effective in treating RA and ankylosing spondylitis (8, 9, 12–17), may be useful in the treatment of PsA as well.
Four studies suggesting that TNFα blockade may be an effective and safe therapy in PsA have been published (9–12). Two of them were open pilot studies on a limited number of patients (9, 10). They used TNFα blockade as monotherapy. The first was a short-term study that evaluated the efficacy of a loading dose regimen of 3 intravenous infusions (5 mg/kg) with infliximab in 21 patients with active spondylarthropathy (9). Only 8 of these patients had PsA. A significant improvement of axial and peripheral articular manifestations was found by day 3 and it was maintained 6 weeks after the third infusion. The second study enrolled 10 patients with severe peripheral polyarticular PsA refractory to previous SMARDs (10). In this study, 25 mg of etanercept was administered subcutaneously twice weekly. Eight of the 10 patients were still taking etanercept with continued good response at month 12. However, this study was limited by the unclear design that included a retrospective part and by the poorly defined inclusion criteria and outcomes.
The third study was a 12-week randomized, double-blind, placebo-controlled trial that assessed the efficacy of etanercept or placebo in 60 patients with PsA (11). The selected patients had peripheral active disease and an inadequate response to NSAIDs; 47% were also receiving MTX therapy. The proportion of patients who met ACR 20%, which constituted the main endpoint, was 73% in the etanercept-treated arm and 13% in the placebo-treated arm.
The fourth study was a placebo-controlled, double-blind study on infliximab in 40 patients with active spondylarthropathy, which also included 13 patients with PsA (12). They were assigned to receive a loading dose of 5 mg/kg infliximab (weeks 0, 2, and 6) or placebo. Both primary endpoints, patient and physician assessments of global disease activity, improved significantly in the infliximab group compared with the baseline value, with no improvement in the placebo group. A significant improvement was present as early as week 2 that persisted up to week 12.
In all 4 studies, TNF blockade controlled psoriasis (as shown by the improvement in PASI) and was well tolerated without serious side effects, except in 1 patient receiving infliximab therapy who developed disseminated tubercolosis (9–12).
Our study suggests the efficacy of infliximab (3 mg/kg every 8 weeks) in combination with MTX in controlling disease activity in the patients with severe PsA resistant to previous second line therapy. The treatment response was rapid and did not diminish over time. A statistically significant improvement of clinical variables, functional status, and acute phase reactants was already present at day 14 and persisted for the entire study period. At week 30, the percentage of the patients satisfying ACR 20%, ACR 50%, and ACR 70% response rates were 64%, 57%, and 57%, respectively.
Given the open, nonrandomized nature of our study, the results need to be confirmed by a double-blind, placebo-controlled, randomized study. However, placebo effect of intravenous infusions cannot explain the high percentage of response (more than 60% of the patients). Furthermore, the evidence of efficacy of the anti-TNFα was strengthened by the fact that all patients had long-term, severe PsA resistant to most second line drugs normally used in PsA and that the serum levels of ESR and CRP decreased significantly during the treatment.
Our preliminary data suggest a correlation between responsiveness to TNFα treatment and baseline CRP values. The patients with higher baseline levels seem to have a better response. However, this observation needs to be confirmed by studies with larger numbers of patients.
Although limited by the low number of patients with axial disease included, our study confirms the efficacy of TNF blockade on spinal symptoms observed in some studies on ankylosing spondylitis (8, 9, 12–15). Our 3 patients had long-term, continuous, inflammatory spinal pain with marked aching and morning stiffness that strongly limited the quality of their lives. Previous SMARD therapy was not very effective on axial symptoms and only NSAIDs gave some relief. TNF blockade dramatically improved spinal pain and morning stiffness, which were almost absent at week 2. SSZ and other second line drugs, while effective on peripheral arthritis, are rarely effective on axial disease of PsA (4–7, 26). These data reinforce the usefulness of TNF blockade in a heterogeneous disease, such as PsA, characterized by a different degree of axial and peripheral involvement.
Major side effects that caused patients to withdraw from the study were observed in 2 patients who had serious infusion reactions at the first infusion and at week 14, respectively. In our study, the percentage of patients (12.5%) who stopped infliximab due to an infusion reaction was higher than those reported in RA studies. The lack of prophylactic use of antihistamine agents could partially explain our higher rate. Brockbank et al reported a high frequency of infections (47%) in a series of 15 PsA patients treated with infliximab (27). Unlike these authors, we had minor, uncomplicated infections of the upper respiratory tract in only 3 patients; they did not require antibiotic therapy. As observed in studies on RA patients, infliximab was well tolerated by PsA patients as well.
As previously reported, anti-TNFα was also effective on psoriasis as shown by the improvement of PASI. As did Mease et al (11), we observed an important improvement of PASI of 37% at week 2 and 86% at week 30.
Our findings in this pilot study strongly indicate that infliximab combined with MTX is effective in improving signs and symptoms of inflammation and physical function in patients with active PsA who are resistant to the most commonly used second line drugs.