Patients were aged ≥18 years old with active RA (determined according to ACR 1987 criteria)13 of over 3 months duration, with 28-joint Disease Activity Score (DAS28) (erythrocyte sedimentation rate [ESR]) >4.5 and C-reactive protein (CRP) >1.0mg/dL at baseline. Additionally, all patients had previously been treated with synthetic disease modifying anti-rheumatic drugs (DMARDs). Patients either had no other prior anti-TNF treatment (naïve patient) or CZP was administered after failure of the first anti-TNF treatment (first-switch patient). Patients enrolled onto the study were actively receiving CZP treatment and were not required to have a wash-out period prior to enrollment. The safety set (SS) consisted of all enrolled patients who took at least 1 dose of CZP. The full analysis set (FAS) consisted of all patients in the SS who had at least 1 valid baseline and post-baseline effectiveness assessment for any effectiveness variable and who had no important protocol deviations. Patients identified as having important protocol deviations were those not initiating CZP treatment, or not fulfilling the study inclusion criteria outlined above.

SONAR (NCT01526434) was a 12-week, open-label, prospective, observational, multicenter, post-marketing study. Both naïve and first switch patients were treated with CZP (400mg at Weeks 0, 2 and 4 followed by 200mg every 2 weeks). CZP was administered subcutaneously, either by patients or according to the standard clinical setting for the prescribing physician and as defined by the Summary of Product Characteristics (SmPC). Permitted concomitant treatments included methotrexate, corticosteroids and analgesics including nonsteroidal anti-inflammatory drugs (NSAIDs). Assessments were carried out at weeks 0, and 12, with a 16-week safety follow-up visit. All patients provided written consent to participate in this study, which was carried out in accordance with local regulations, the Declaration of Helsinki and the local laws of Spain.

The primary effectiveness endpoint was the mean change from baseline in Health Assessment Questionnaire-Disability Index (HAQ-DI) for RA,14 at Week 12 in the FAS population; this population was also grouped by prior anti-TNF experience to compare HAQ-DI scores between anti-TNF naïve and first switch patients from baseline to Week 12. Other assessments included DAS28 (ESR),15 Patient's Assessment of Arthritis Pain (PtAAP) – Visual Analog Scale (VAS)16 and Short Form 36-item Health Survey (SF-36) physical component summary (PCS) and mental component summary (MCS).17 For effectiveness assessments, the data were assigned to a visit based on the actual date of the routine clinical visit. Day was calculated relative to the baseline visit (recorded at Week 0). Any visit between Day ≥63 and ≤98 was assigned as Week 12 for the purposes of the analysis.

PDUS assessment involved the detection and grading (from 0 to 3) of gray-scale synovitis and Power Doppler (PD) synovial signal of 24-joints (12-joint US assessment for RA18 and the bilateral 4 and 5 metacarpophalangeal [MCP] joints and 2 to 5 metatarsophalangeal joints). Synovial hypertrophy and effusion were graded, semi-quantitatively, from 0 to 3, where 0=absence; 1=mild; 2=moderate; and 3=marked. PD signal was graded according to the following scale: 0=absence, no synovial flow; 1=mild, ≤3 isolated signals; 2=moderate, >3 isolated signals or confluent signal in less than half of the synovial area; 3=marked, signals in more than half of the synovial area.19 Assessments were made according to the Outcome Measures in Rheumatology definitions and published scoring systems.20 An index for effusion, synovial hypertrophy and synovial PD signal was calculated by summing the corresponding scores from each assessed joint. PDUS outcomes measured included change from baseline to Week 12 in global US 24-joint indices (including synovial hypertrophy index, effusion index, and PD signal index). Intra-reader reliability was monitored by evaluating the first two patients at each center twice (2–4 days between evaluations) before the treatment period and was tested on representative images of joints included in the study. Intra-class correlation coefficients are reported for synovial hypertrophy index, effusion index, and PD signal index and were based on the repeated assessments at baseline for these patients.

Adverse drug reactions (ADRs), were recorded and treatment-emergent adverse drug reactions (TEADRs) were defined as ADRs if they started on or after the date of first study medication and up to 70 days after the last (most recent) dose of CZP. Adverse events were classified as ADRs when a causal relationship between the product and the occurrence was suspected by the reviewing healthcare professional. TEADRs were coded using the Medical Dictionary for Regulatory Activities (MedDRA) Version 16.1.

The sample size was calculated based on the precision of the 95% confidence interval (CI) around the expected HAQ-DI score mean change from Baseline to Week 12. A difference of at least ±0.22 was assumed to be sufficient to recognize a minimal clinically important difference (MCID) in HAQ-DI score, and the proportion of patients achieving MCID in HAQ-DI was calculated post hoc.21 A sample size of 95 patients was estimated to provide at least 90% of statistical power to establish a difference of 0.22 as significant with a SD of 0.6 in a two-sided one-sample t-test, with a significance level of 0.05. A HAQ-DI difference greater than 0.22 could be confirmed with fewer patients.

The mean changes in HAQ-DI scores from Baseline to Week 12 were analyzed by comparing the values at Baseline and Week 12 using a paired t-test. The primary analysis was run based on last observation carried forward (LOCF) imputation of missing data from dropouts. A paired t-test was also used to compare Baseline and Week 12 values for SF-36, PtAAP-VAS and DAS28(ESR). Apart from the p-value related to the primary endpoint, all p-values are nominal.

Statistical analysis and generation of tables, figures and patient data listings were performed using Statistical Analysis System (SAS®) Version 9.4. P-values are only provided for the mean change from baseline, to Week 12, in HAQ-DI (the primary endpoint), DAS28(ESR), PtAAP and SF36 PCS and MCS, all other variables are represented by CIs. Nominal p values were not calculated for comparisons between the patients split by prior anti-TNF exposure, due to the small sample number, therefore data are shown as the mean and 95% CI for HAQ-DI and PDUS outcomes for each subgroup.

A total of 77/80 enrolled patients received at least one dose of CZP during the trial and were included in the SS (Table 1), of which 65 patients provided at least one valid baseline and post-baseline effectiveness assessment, without any important protocol deviations during the study (FAS); the remaining 12 recruited patients did not meet these criteria for inclusion within the FAS. The reasons for excluding these patients from the FAS were: 3 patients did not have active RA (defined as DAS28(ESR)>4.5 and CRP>1.0mg/dL) at baseline, prior anti-TNF use was unknown for 1 patient, and 9 patients did not have a valid baseline and post-baseline effectiveness assessment. At baseline, 57/77 (74%) patients within the SS and 48/65 (73.8%) patients in the FAS were anti-TNF naïve. Of patients in the FAS, 17/65 (26.2%) patients were switched to CZP due to lack of effectiveness or adverse events. A large proportion of patients in the FAS completed CZP treatment to Week 12 of the study (60/65, 92.3%), 1 patient (1.5%) permanently withdrew CZP treatment before Week 12 due to an ADR, and 4 patients (6.2%) withdrew due to “other” reasons. “Other” reasons included the patient not attending the visit, patient decision, hospitalization, and temporary interruption of CZP due to vertigo. Among the 61 patients with a Week 16 visit, 56 patients (91.8%) opted to continue CZP treatment, and 1 patient (1.6%) withdrew due to an ADR, 1 patient (1.6%) withdrew due to ineffectiveness and 3 patients (4.9%) withdrew for “other” reasons. The mean duration of RA since diagnosis (SD) for enrolled patients was 7.8 (9.8) years. The mean age of patients was similar between the two study populations, 53.8 years (SS) and 53.7 years (FAS) (Table 1). The gender of patients within the study population was predominantly female (SS: 74.0% and FAS: 73.8%) (Table 1). Disease activity and patient-reported outcomes (PROs) were similar between the SS and the FAS (Table 1). The SS had a greater median CRP (1.83mg/dL) than the FAS (1.65mg/dL) but a lower median ESR (SS: 31.0mm/h, FAS: 34.0mm/h) (Table 1). US outcomes were similar between patients in the SS and the FAS (Table 1) at baseline. At baseline, 57/65 (87.7%) patients in the FAS (69/77 [89.6%] patients in the SS) were taking concomitant DMARDs (Table 1). At Week 12, 10 (15.4%) patients in the FAS were taking concomitant DMARDs, with 4 (6.2%) patients receiving a higher DMARD dose compared to baseline.

Rapid improvements in clinical and PROs were observed between baseline and Week 12 (Fig. 1). An overall significant mean (SD) reduction of −0.6 (0.6) in HAQ-DI from baseline to Week 12 was achieved (p<0.001, Fig. 1A) and 33/65 (50.8%) patients achieved MCID in HAQ-DI at Week 12. A similar significant improvement was observed for DAS28 (ESR), with an overall mean reduction of −2.2 (1.5) from baseline to Week 12 (p<0.001, Fig. 1B). PROs also showed a significant improvement between baseline and Week 12, with a substantial reduction in PtAAP (VAS Scale 0–100) from 66.8 to 27.6 (p<0.001, Fig. 1C). Further to this an overall improvement in mean SF-36 PCS and SF-36 MCS was seen to rise from 32.2 at baseline to 38.4 by Week 12 (Fig. 1D) and from 42.4 at baseline to 50.1 at Week 12 (Fig. 1E) (both p<0.001), respectively.

Fig. 1.

Clinical and patient-reported outcomes at baseline and Week 12. Data are presented for the full analysis set. Values are means and error bars represent the 95% confidence intervals. ***Statistically significant difference between baseline and Week 12 mean values; p<0.001. Patient's assessment of arthritis pain (PtAAP) was recorded on a visual analog scale (VAS) ranging from 0mm (no pain) to 100mm (maximum pain). DAS28: 28-joint count Disease Activity Score, ESR: erythrocyte sedimentation rate, HAQ-DI: health assessment questionnaire-disability index, MCS: mental component summary, PCS: physical component summary, PtAAP: patient's assessment of arthritis associated pain, VAS: visual analog scale.

(0.19MB).

A comparison between anti-TNF naïve and anti-TNF first switch patient's HAQ-DI score indicated that functional improvements were seen in both anti-TNF naïve and first switch patients, although the patient numbers in these sub-group analyses were low (Fig. 2). The improvement seen in the naïve group (34 patients) was −0.67 (95% CI at Week 12: −0.86, −0.47) whereas the improvement in the first switch patients (13 patients) was numerically smaller, −0.28 (95% CI at Week 12: −0.65, 0.09). At Week 12, 27/34 anti-TNF naïve patients achieved MCID in HAQ-DI, compared to 6/13 first switch patients.

Fig. 2.

Mean HAQ-DI in anti-TNF naïve and first switch patients at Week 12 versus baseline (observed data). Data are presented for the full analysis set (n=65) and patients are grouped by prior anti-TNF treatment (naïve or first-switch). Values are means and the error bars represent the 95% confidence intervals. For one patient within the first switch group the HAQ-DI score was not available at baseline. The Week 12 data are reported for the number of patients with HAQ-DI scores available at this time point.

(0.05MB).

Reductions from baseline to Week 12 were seen in synovial hypertrophy index, from 19.3 to 12.3, effusion index, from 14.3 to 7.7, and PD signal index, from 8.9 to 4.6 (Table 2). These reductions were observed in both anti-TNF naïve and first switch patients (Fig. 3). The mean [95% CI] effusion index for the naïve (13.5 [10.7, 16.3]) and first switch patients (16.8 [9.4, 24.1]) differed at baseline (Fig. 3D).

Fig. 3.

Power Doppler ultrasound scores at baseline and Week 12 for anti-TNF naïve and first-switch patients for (A) synovial hypertrophy index, (B) effusion index, and (C) Power Doppler signal index. Data are reported for the full analysis set. Values are means and the error bars represent the 95% confidence intervals.

(0.19MB).

The safety profile of CZP was consistent with other reports of CZP in RA; no new safety signals were identified (Table 3). A total of 19 TEADRs were reported by 13/77 (16.9%) patients, 9 of which led to the discontinuation of 8/77 (10.4%) patients from the study. A total of 4 serious TEADRs were experienced by 3/77 (3.9%) patients. The serious TEADRs were lymphadenopathy mediastinal, tuberculosis, skin reaction and non-Hodgkin's lymphoma. The case of active tuberculosis was reported by the investigators to be related to CZP and the patient permanently discontinued CZP treatment. As the patient recovered after treatment with anti-tuberculosis therapy, with the disappearance of pulmonary nodules, the case was diagnosed clinically as tuberculosis, despite the absence of microbiological confirmation. The skin reaction was considered to be a medically important reaction and the patient permanently discontinued CZP treatment; the patient recovered from this event. Non-Hodgkin's lymphoma was reported 69 days after initiation of CZP in one patient, and was considered severe in intensity. The patient permanently discontinued CZP treatment due to non-Hodgkin's lymphoma. The event resulted in the death of the patient 64 days after onset. Because the study investigator stated that they were unable to confirm the relationship, UCB Pharmacovigilance Department, per protocol, classified the relationship of this adverse drug reaction conservatively as ‘related to study drug’.