Magnetic resonance imaging of the hand and wrist in a randomized, double-blind, multicenter, placebo-controlled trial of infliximab for rheumatoid arthritis: Comparison of dynamic contrast enhanced assessments with semi-quantitative scoring

The objective of this study was to compare the scope and the discriminative power of Dynamic Contrast Enhanced Magnetic Resonance Imaging (DCE-MRI) to those of semi-quantitative MRI scoring for evaluating treatments for rheumatoid arthritis (RA) in multicenter randomized clinical trials (RCTs). Sixty-one patients with active RA participated in a double-blind, parallel group, randomized, multicenter methodology study receiving infliximab or placebo through 14 weeks. The most symptomatic wrist and metacarpophalangeal joints (MCPs) were imaged using MRI. In addition to clinical assessments with DAS28(CRP), the severity of inflammation was measured as synovial leak of gadolinium based contrast agent (GBCA) using DCE-MRI (Ktrans, primary endpoint) at weeks 0, 2, 4, and 14. Two radiologists independently scored synovitis, osteitis and erosion using RA MRI Score (RAMRIS) and cartilage loss using a 9-point MRI scale (CARLOS). Infliximab showed greater decrease from baseline in DAS28(CRP), DCE-MRI Ktrans of wrist and MCP synovium, and RAMRIS synovitis and osteitis at all visits compared with placebo (p<0.001). Treatment effect sizes of infliximab therapy were similar for DAS28(CRP) (1.08; 90% CI (0.63–1.53)) and MRI inflammation endpoints: wrist Ktrans (1.00 (0.55–1.45)), RAMRIS synovitis (0.85 (0.38–1.28)) and RAMRIS osteitis (0.99 (0.52–1.43)). Damage measures of bone erosion (RAMRIS) and cartilage loss (CARLOS) were reduced with infliximab compared to with placebo at 14 weeks (p≤0.025). DCE-MRI and RAMRIS were equally sensitive and responsive to the anti-inflammatory effects of infliximab. RAMRIS and CARLOS showed suppression of erosion and cartilage loss, respectively, at 14 weeks. (ClinicalTrials.gov registration: NCT01313520)


C O N F I D E N T I
To compare the effect of infliximab on synovial inflammation as measured with Dynamic Contrast Enhanced (DCE)-MRI of one wrist with placebo over 14 weeks of treatment. The primary hypothesis is that the change in K trans in total enhancing synovium with infliximab is larger than placebo.

Secondary Objectives:
! To compare the effect of infliximab with placebo on various MRI measures of disease activity including: o Other DCE-MRI endpoints including K trans of total enhancing tissue of wrist; IAUC90, Early Enhancement Rate (EER) and maximal enhancement (ME) of total enhancing tissue and total enhancing synovium of wrist, and enhancing synovial volume of wrist and/or metacarpophalangeal (MCP) o RA MRI Scoring (RAMRIS) of osteitis and synovitis of wrist and MCP. ! To compare the effect of infliximab with placebo over 14 weeks of treatment on the following clinical measures of disease activity: ! Clinical disease activity (DAS28 (CRP)) ! American College of Rheumatology criteria for 20% improvement (ACR20) and for 50% improvement (ACR50). ! American College of Rheumatology criteria for x % of improvement over the range of x=5 to 45% (ACRx). (The purpose of this objective is to assess for the possibility of a failed clinical study and to benchmark imaging biomarkers against standard clinical biomarkers.) ! To compare infliximab against placebo over 14 weeks with two composite measures of RA disease activity that incorporate DAS28(CRP) with K trans or RAMRIS.
! To correlate changes in clinical disease activity (DAS28(CRP)) and changes in MRI biomarkers of disease activity: o RAMRIS synovitis and osteitis o DCE-MRI parameters Exploratory Objectives:

!
To assess the sensitivity and specificity of blood mRNA signatures and DNA polymorphisms that predict clinical response to infliximab. !
To evaluate the responsiveness of endpoints available in Dynamika RA software to infliximab compared with placebo !
To compare the effect of infliximab with placebo on synovial inflammation as measured by K trans derived from DCE-MRI of all segmentable synovium in wrist and all MCP in the field of view (FOV). !
To compare the effect of infliximab with placebo on RAMRIS scoring of bone erosion in wrist and MCP. !
To compare the effects of infliximab with placebo on a measure of inflammation that sums RAMRIS osteitis and synovitis scores, and a measure of damage that sums RAMRIS bone erosion and non-RAMRIS cartilage damage. !
To compare the effect of infliximab with placebo on the volume of osteitis TNF gene promoter, known as the 308 G/A variant (TNF-308) (rs1800629) The effect of the following gene transcript signatures at baseline on clinical response to infliximab will be investigated: •  Week -2 to 0 -1 to 0 0 2 4 6 10 14 0 to 14 16 Clinic and MRI visit must occur within this time frame (+/-calendar days) N/A N/A -7 +/-4 +/-4 +/-4 +/-4 -7 N/A N/A o X X X X X Treatment Administration (IV) p X X X X Record AEs and Concomitant Meds X X X X X X X X X a. These include the main study consent, informed consent for pharmacogenetic sample, and future use consent. b: On treatment days, blood samples should be drawn predose. c: For women of childbearing potential only. d: For post-menopausal women only. e: Performed at investigator discretion. f: At prestudy/screening, subjects must satisfy all TB criteria including a negative skin test with PPD or a QuantiFERON-TB Gold test (Appendix 2). At Weeks 0, 2, 4, 6, 10, and 14, subjects must show no signs or symptoms suggestive of active TB upon medical history and/or physical examination, and have had no recent close contact with a person with active TB or, if there has been such contact, will be referred to a physician specializing in TB to undergo additional evaluation. g: Oral temperature measurement is preferred, but axillary temperature is also acceptable. h. Refer to Appendix 3. i. Results from these assessments will be used by the Sponsor to calculate DAS28(CRP) and ACR20, ACR50 and ACRx. At Week 2, 4, and 14, the Tender and Swollen Joint Count, CRP, GADP, and other Investigator and Patient assessments may be performed before or after the MRI as long as they are performed within the permissible time frame. These assessments must be performed before the administration of study medication. j. Tender and Swollen Joint Counts (28 joint set) will be performed by a masked joint examiner with no other role in the clinical care of the patient. The same masked joint examiner should be used for all assessments of an individual patient. Baseline Tender and Swollen Joint Counts must occur before administration of study medication at Week 0 (Visit 2). PROTOCOL  Week -2 to 0 -1 to 0 0 2 4 6 10 14 0 to 14 16 Clinic and MRI visit must occur within this time frame (+/-calendar days) N/A N/A -7 +/-4 +/-4 +/-4 +/-4 -7 N/A N/A k. At Screening, an ESR by Westergren method may be used to satisfy the inclusion criteria in Section 7.3.1.
l. Informed consent for pharmacogenomic samples must be obtained before the DNA, RNA, Plasma, and serum future use samples are collected. DNA sample for analysis should be obtained on Day 1 on randomized subjects only, or at a later date as soon as the informed consent is obtained. m. Future use consent must be signed before these samples are collected. n. High dimensionality cytometry assays are contingent upon operational availability as described in Section 7.6.
o. Screening MRI must be obtained within 7 days prior to Baseline and approved by Sponsor or Sponsor's agent before initiating study treatment at Week 0. If subject is eligible, the screening MRI will be used as the baseline MRI at Week 0. An MRI is required at the time of study discontinuation unless an MRI was obtained within 4 weeks prior. At the scheduled timepoints, the MRI must be obtained before the administration of study medication. p. At all visits, study drug is administered after performance of Tender and Swollen Joint Count, serum for CRP is obtained, and after the GADP(VAS) and other investigator and patient questionnaires are completed AND after the MRI is collected. Study medication is prepared and hung on an IV pole with obscuring drape by an unmasked trained person with no other role in study conduct. PROTOCOL

Background
Evaluation of disease activity in rheumatoid arthritis (RA) is complex, with no single marker reflecting all aspects. In the past 20 years, composite disease activity instruments (e.g. DAS28) have improved the ability to evaluate the course of RA. DAS28 combines measures of swollen and tender joint counts in a 28 joint set, serum C-reactive protein (CRP), and global health questionnaire into a more responsive metric. However, DAS28 is a surrogate marker of the underlying pathological process -synovitis of the RA joint. This synovitis leads to bone and cartilage destruction, causes pain and ultimately leads to joint destruction and physical impairment.
The efficiency of proof of concept (POC) trials in RA may be improved by the incorporation of sensitive imaging biomarkers of joint synovitis to complement clinical change using validated instruments like DAS28. This is a methodology study designed to assess two MRI biomarkers to improve clinical decision-making for new drug development. Imaging endpoints that satisfy criterion validity as a measure of disease activity, that are objective, and that are at least as discriminatory as clinical endpoints to treatment effects will improve the assurance of development decisions.
A placebo-controlled, randomized clinical trial to benchmark MRI biomarkers against DAS28 at time points less than 12 weeks in established RA has never been published yet potentially offers several benefits for novel drug development. First, the addition of another objective endpoint to DAS28 provides greater assurance in development decisions. Currently objective endpoints such as C-Reactive Protein (CRP) are heavily weighted in efficacy assessments, yet typically do not offer sufficient power in small, early clinical studies. Second, it may be useful to combine imaging biomarkers with DAS28 to create a composite endpoint to support smaller sample sizes or greater power with a fixed sample size. This was done in MK-0000 P-88 in prespecified fashion. Since the correlation between imaging biomarkers and DAS28 is only moderate, t h e introduction of new information from imaging increased the treatment difference substantially from effect size~1 for either DAS28 or imaging biomarkers alone to about 1.6 -2 for the composite, with the understanding that imaging composite measures with DAS28 are not fully validated. Third, it may be possible to use imaging to study treatment effects in populations with less severe disease activity, which may be important as efficacious treatments are increasingly utilized. This approach was successfully explored by identifying treatment effects using imaging biomarkers in patients enrolled in MK-0000 P88 with baseline DAS28 values at and below the median values for the enrolled population. PROTOCOL NO. P08136 PROTOCOL 05 JAN 2011 INITIAL PROTOCOL

Class or Type of Drug Being Studied/Description of Drug
Infliximab, at the dose recommended for initial therapy in RA will be used in this study at a dose of 3 mg/kg with an induction regimen of 0, 2, 6 wks, and an additional infusion at 14 weeks. Infliximab is a chimeric mouse-human IgG1 monoclonal antibody to human Tumor Necrosis Factor (TNF) ( 1 ).

Study Conduct Rationale
Information produced from this methodology study will be applied to the large potential pipeline for RA, a high priority disease area. The precise way it will be utilized depends on the prior level of information regarding the mechanism. Programs are expected to take advantage of a positive imaging biomarker study by either utilizing the information to reduce sample size or to improve the certainty of decisions, incorporated into Go-No-Go criteria, as long as the costs and complexity of MRI do not outweigh larger sample sizes. Beyond NO-GO decisions, imaging biomarkers may be useful to establish the clinically active dose-range. Establishing dose-response in RA trials is difficult using DAS28 and ACR endpoints. Group sizes of 40-70 are typically used, but are still not powered to statistically distinguish amongst doses. If a convincing trend is not seen, then several doses may need to be confirmed in phase 3, or additional work might be required to establish the minimally non-efficacious dose ( 2,3,4 ). Very few imaging studies have investigated dose response in RA. MK-0000 Protocol 088 investigated 2 doses of prednisone on clinical and power Doppler ultrasound in an adaptive design. The higher dose of prednisone had similar treatment effect on imaging biomarkers and DAS28. The lower dose of prednisone did not show statistically significant effects on either DAS28 or the prespecified imaging endpoint. However, a prespecified combination endpoint of DAS28 with ultrasound measures did demonstrate a treatment effect at 14 and 7 days, as did 3 of 8 secondary ultrasound endpoints at 14 days and 7 of 8 secondary ultrasound endpoints at 7 days, some with effect sizes of~2, indicating substantial pharmacological activity of the 7.5 mg dose. Neither the DAS nor any individual ultrasound endpoint demonstrated statistical separation between doses, but the higher dose had numerically higher effects than the lower dose on 6 of 8 imaging endpoints and DAS28 at 14 days. In the phase 3 GO-FORWARD and GO-BEFORE studies, 50 and 100 mg of golimumab were investigated in active RA using MRI. There were no clear differences between these doses on MRI endpoints, but these doses are likely at the top of the dose-response curve for the mechanism. Finally, Kalden-Nemeth studied two doses of infliximab (1 and 10 mg/kg) and showed numerical differences between doses on a number of DCE-MRI endpoints in a PROTOCOL NO. P08136 PROTOCOL 05 JAN 2011 INITIAL PROTOCOL handful of patients after 4 weeks ( 5 ). In this study we will not investigate doseresponse.
MRI is capable of identifying subclinical synovitis, implying that MRI may be uniquely suited to measure changes in disease activity in patients with lower levels of disease activity ( 6 ). This finding is pertinent to drug development since the activity of RA patients appears lower than in the pre-biological era ( 7 ). The greater availability of potent disease modifying agents will continue to challenge the ability to identify biological naïve subjects with high degrees of disease activity, the preferred population for assessment of disease activity in early clinical drug development. These trends are likely to continue; thus, identifying an imaging biomarker with sufficient dynamic range to be useful in patients lacking high degrees of symptomatic arthritis may enhance the future conduct of RA proof-of-concept studies. In support of this notion MK-0000 Protocol 088 identified treatment effects using power Doppler ultrasound in subjects with baseline DAS below the median enrolled in that small study. A similar approach can be taken in this study, though there is no plan to intentionally study patients with less active disease.

Study Design Rationale
This study will examine the discriminative power of several MRI measures of RA inflammation and to benchmark these imaging biomarkers to clinical activity measured using DAS28. We concentrate on inflammation because bone and joint destruction are less critical in early drug development, given the shorter periods of clinical studies during early development. A large body of work indicates synovitis and osteitis are the underlying processes that explain subsequent bone erosions in RA ( 8,9 ). The imaging biomarkers incorporated in this study are the Rheumatoid Arthritis MRI Score (RAMRIS) system, and parameters regarding the vascularization of synovium obtained during a time course of contrast injection, a technique called dynamic contrast enhanced magnetic resonance imaging (DCE-MRI). Moreover, this study will also demonstrate the feasibility of using DCE-MRI at multiple centers capable of enrolling rheumatoid arthritis patients to drive program decisions. Only RAMRIS scoring has been successfully employed in placebo controlled multicenter RA trials, but the shortest timepoints reported in such a rigorous study design are at 12 weeks. Therefore there is a need for methodology studies incorporating more sensitive imaging biomarkers to define the reproducibility and the treatment effect size at short time points (< 12 weeks) and benchmark these endpoints against clinical disease activity measures. PROTOCOL

Rationale for Study Design, Population, Treatment and Effect Size
Clinical disease activity at enrollment is required for any of the imaging methods described herein to show a treatment effect, and this is also the case for improvement in clinical disease activity. RA patients have a fluctuating disease course, and enrollment of patients at an apogee of disease activity probably explains up to a third of the total DAS28 improvement in response to TNF inhibitors ( 10 ). Consequently, randomized, placebo controlled studies are the only study design from which definitive determinations of efficacy can be reached. The relative performance of imaging biomarkers could certainly be explored in open label treatment studies. However, when trying to compare effect sizes of imaging biomarkers against clinical disease measures (e.g. DAS28), it is essential to have a reference control arm and patient and physician masking to treatment assignment. The available evidence suggests all the MRI imaging biomarkers mentioned here are adequately sensitive to treatment effects, and for avoidance of doubt, a placebo control arm will allow a clear decision to be made whether the cost and complexity of imaging endpoints is worthwhile in RA study paradigms. Because patients may improve in many ways with infliximab therapy, this study will use a joint examiner masked to treatment assignment to record the number of tender and swollen joints using the 28 joint set for the calculation of DAS28(CRP), but will have no other involvement in patient management, including administration of study drug or recording of AE.
Current treatment paradigms in RA emphasize prompt, thorough treatment of active disease, ideally to clinical remission, and increasingly introduced as soon as a diagnosis can be reached. True placebo controlled trials cannot be conducted. Instead patients with persistent disease activity on background treatment with DMARDs such as methotrexate (MTX) are randomized to placebo or active therapy for a short period of time. Anti-TNF inhibitors form the most commonly prescribed, highly active initial biological therapy in such patients. Treatment effects are prompt, with CRP response by 2 weeks, swollen joint counts by 6 weeks, and a plateau in clinical response using ACR20 criteria by 14 weeks ( 11 ). In this study we plan to use infliximab will be administered according to a standard approved dosing regimen of 3 mg / kg at 0, 2, and 6 weeks (induction), with one additional infusion 8 weeks later at 14 weeks, which is the primary time point for the study. Infliximab is indicated in addition to methotrexate therapy. At the end of the double-blind portion of this study, the study will close. The Sponsor will make available a limited amount open-label infliximab therapy for the use of patients with their treating physician. This study will enroll RA patients with established, active disease despite DMARD treatment, typical of those that participate in therapeutic trials of new therapeutics. Patients will be on a stable dose of methotrexate, as required for infliximab therapy. Patients will have sufficient disease activity using tender and swollen joint count criteria. Because this is a MRI study, patients will be required to have at least one wrist swollen at baseline and have evidence of synovitis, as determined by a central reader, prior to randomization.
As described below, the best validated MRI scoring system, RAMRIS, is at least as responsive to therapy as DAS28, and we hypothesize that some MRI measures are even more responsive. It is highly desirable to compare the PROTOCOL NO. P08136 PROTOCOL 05 JAN 2011 INITIAL PROTOCOL   responsiveness of DCE-MRI biomarkers to DAS28, so we select the sample size  based on DAS28. Reasonable estimates for treatment effects for infliximab  treatment are derived from more recently approved TNF inhibitors or other potent  biological agents in active RA patients with inadequate response to DMARDs, which are summarized in the Statistical Analysis Plan. Based on these considerations we choose to power this study based on an estimate treatment effect size of 0.7 to 0.8.

Rationale for Methods and Endpoints
The major consideration in choosing MRI endpoints to support the need of early drug development are a certain degree of criterion validation (correlation with gold-standard disease activity metrics, histological correlation, or alignment with other valid imaging modalities), and, critically, evidence that the endpoints are precise and highly discriminative of treatment effects. The MRI endpoints discussed here appear to meet these needs. Moreover, they demonstrate change with several efficacious agents, suggesting they will be general purpose tools in RA drug development.
As described below, the RAMRIS imaging biomarker system is validated and can be implemented in multicenter trials. There is abundant data in established RA at 12 weeks, but little data useful for our purposes from earlier time points. Studies in patients within 6 months of diagnosis suggest MRI may be responsive at shorter timepoints, as reported by Quinn at 4 weeks for bone marrow edema ( 12 ), but this population (i.e., patients with early poor-prognosis RA) is not available for study for investigational agents in early drug development. The chief deficit of RAMRIS is its ordinal scoring system, which may limit the ability to observe true, rapid improvements in synovitis. DCE-MRI shows much promise, with several reports of significant change after 1 to 4 weeks. There is a single randomized clinical trial using DCE-MRI and it is not possible to estimate a treatment effect size. It is possible that publication bias against negative studies precludes a balanced assessment of the utility of DCE-MRI in RA. If this study achieves its primary objectives there is the possibility of dramatically smaller, shorter clinical POC studies. If RAMRIS scoring shows, as expected, similar effect size as DAS28 with earliest effect detected difference at~12 weeks, then we will accrue operational experience with MRI in RA and will have a dataset on which the franchise can optimize decision-making rules using imaging endpoints or composite clinicalimaging endpoints PROTOCOL NO. P08136 PROTOCOL 05 JAN 2011 INITIAL PROTOCOL

RAMRIS scoring of osteitis and synovitis
The RAMRIS (Rheumatoid Arthritis Magnetic Resonance Imaging Scoring) system is the most mature semi-quantitative system for RA MRI, a well-defined, reproducible measurement system of one wrist and MCP joints suitable for multicenter use, utilize scoring of bone erosions, bone marrow edema (osteitis) and synovial volume. Cartilage loss or joint destruction or tenosynovitis is not yet a valid component of RAMRIS. The RAMRIS sequences utilize a set of T1-weighted images before and after intravenous GBCA (typically 0.1 mmol/kg) to identify enhancing synovitis.
Bone marrow edema is assessed on pre-contrast fat suppressed T2-weighted coronal images.
Synovitis is scored 0-3 in distal radioulnar, radiocarpal, intercarpal-carpometacarpal and 1-5 MCP joints. Osteitis is scored 0-3 in the fraction of bone involved within 1 cm of joint line of carpal bones, distal radial, distal ulnar and metacarpalbases. Expert readers examine image sets simultaneously from multiple timepoints blinded to treatment and acquisition order. Scoring can take an hour per patient-timepoint ( 13 ). Maximum total scores for erosion, osteitis and synovitis are 250, 75 and 24, respectively.
The reliability of RAMRIS components shows high intra-reader reliability (ICC >0.9) and in good hands, substantial inter-reader reliability ( 14,15 ). Comparisons in early and late RA show high correlations between MR synovial membrane thickening and enhancement post-gadolinium, and histological inflammatory changes ( 16,17,18,19 ). Osteitis correlates with pain, CRP and histological osteitis ( 20 ). Bone erosions scores using RAMRIS have been validated against radiographic skeletal erosions and are detected more sensitively in a shorter period of time. MRI using RAMRIS scoring systems has been successfully employed in numerous multicenter studies of RA therapeutics. Correlation of change with RAMRIS score with clinical status is low, suggesting that RAMRIS provides additional unique information on disease status. In the golimumab MRI dataset the correlation between change in DAS28 and change in RAMRIS synovitis score was very modest, about 0.2 (data on file). PROTOCOL  In summary, RAMRIS is capable of identifying treatment effects of potent biological drugs with effect sizes at least comparable to clinical change scores and it is straightforward to apply in multicenter clinical trials. There is no need to further validate the RAMRIS endpoint, which is included in this trial as a benchmark imaging endpoint. Adequate data with RAMRIS at~12 weeks exist in established RA in order to estimate the additional utility of including MRI in treatment studies. There is a paucity of data at shorter timepoints in established RA.
In general it appears that RAMRIS scores in patients with established disease tend to remain steady in the control group. This behavior differs from clinical improvement which occurs within the control group. Inflammatory change appears to temporally track with improvement in clinical endpoints. Few studies have looked at time-points shorter than 3 months. Limitations of RAMRIS are the relative insensitivity to change, perhaps due to the ordinal scoring system.

Dynamic Contrast Enhanced MRI
DCE-MRI involves acquisition of sequential images every few seconds during and after the standardized intravenous injection of GBCA, typically with a power injector (0.1 mmol/kg). In t h e setting of inflammation t h e contrast material extravasates from the vascular space to interstitial space but does not penetrate viable cells. The resultant increase in T1-weighted signal intensity strongly relates to local areas of highly metabolically active synovial tissue -areas with associated extreme hypoxia, metabolic activity, angiogenesis, endothelial activation and upregulated cytokine release ( 24,25 ). Thus one would expect relationships between vascular leak and local tissue destruction in RA DCE-MRI directly assesses these processes by measuring the dynamics of contrast change.
In order to capture the rate of synovial enhancement, MRI protocols are chosen with short repetition times, typically a spoiled gradient echo sequence, allowing an entire image set to be acquired every few seconds to capture the dynamic enhancement phase. To maximize signal-to-noise, a small field of view is selected. In this study we will study one wrist as it is commonly involved in RA. In order to follow the same tissues longitudinally during treatment, it is recommended to obtain 3D images where signal from a slab of tissue is simultaneously obtained. The analysis of DCE-MRI is complex and typically involves segmentation of synovium from the bone.
DCE-MRI measures have been quantified by extracting such features as the rate of contrast change, often called the Early Enhancement Rate (EER in % / sec), the maximal Relative Enhancement (RE or ME) (compared with precontrast signal), and the extent of contrast washout. These are the common endpoints reported in the RA literature and are graphically illustrated: PROTOCOL  In an attempt to reduce variability in the relationship between observed signal and contrast agent concentrations between patients, centers and time, more rigorous approaches were developed, such as a pharmacokinetic compartment model, for instance the Toft model ( 27 ), that describes the exchange between the plasma space into which the Gd contrast is injected and the tissue extracellular space (the synovium). Assumptions that are required are that the relative signal enhancement curve scales linearly with contrast agent tissue concentration. The rate of exchange depends on capillary blood flow, capillary permeability and area product, volume of distribution of the contrast, and blood contrast concentration over time. The model outputs are K trans (min -1 ), the volume transfer rate from the blood plasma to the enhancing synovium; the fractional volume of enhancing synovium; and k ep (min -1 ), the volume transfer rate from the enhancing synovium to the blood plasma ( 28 ). In this study we will use K trans in enhancing synovium as the primary endpoint. In the RA field it is not known whether the derivation of a PK compartment model will provide better sensitivity to treatment effects at the group level, but the oncology literature suggests improved precision when using individual patient arterial input functions rather than an assumed general function ( 29 ). One challenge of using a compartment model approach is that an arterial input function (AIF) has to be assumed or measured, for example using an automated data-driven method ( 29  Cross-sectional studies provide criteria validity for t h e ra t e of synovial enhancement on MRI. Several studies have correlated DCE-MRI with histological features. Ostergaard and colleagues found vascularity was the histology parameter with the closest correlation with the EER on DCE-MRI ( 18 ). The rate of maximal enhancement had very high correlation with inflammation on biopsy, while static MRI measures (e.g synovial volume) correlated more weakly with inflammation ( 30 ). The rate of synovial enhancement was related to fibrin exudation, cellular infiltrate, villous hypertrophy and blood vessel density but not fibrosis in RA knees. Clinical correlates (swelling tenderness or disease indices) were weak or absent ( 31 ). Cimmino and colleagues used a 0.2T extremity MRI to show in 36 patients with RA and 5 healthy controls that EER and maximal RE were significantly higher in patients with active RA than in those with inactive RA and controls ( 32  The published work is all from single-centers, suggesting that the technical protocol will need to consider issues associated with different site equipment. DCE-MRI has been developed for multicenter applications in the oncology field, and the use of phantoms has been suggested to improve the comparability of parameters between centers. We are aware of a study conducted by Amgen in 2 centers to evaluate etanercept on the effects of DCE-MRI in an open label trial (OLT) of 14 patients with RA (NCT00361634). The study, by report, was unable to identify a significant treatment effect. Questions have been raised about the magnitude of signals observed. A randomized, multicenter double blind study of tocilizumab in RA with inadequate response to anti-TNF is recruiting and uses DCE-MRI as one imaging endpoint (NCT01034397). The developers of Dynamika RA software described below are involved in multicenter studies. Therefore, it is likely that the technology can be applied across multiple centers.
Considered together, the available data suggest DCE-MRI has reasonable criterion validity and can identify changes in synovium and bone marrow edema (osteitis) due to treatment. In a number of studies, detectable changes occurred very quickly after treatment, suggesting large treatment effects. In one study, a trend towards dose-discrimination for EER suggests additional utility. The analysis methods are complicated, and improvements in measurement may also improve the discriminant capacity. The relative sensitivity of DCE-MRI endpoints to clinical endpoints will require additional experience. In this study we hypothesize that the DCE-MRI parameter, K trans , will discriminate infliximab treatment effects more efficiently than DAS28 or RAMRIS synovitis. PROTOCOL

Exploratory Endpoints
Exploratory endpoints include evaluation of the volume of synovium and DCE-MRI endpoints calculated using the Dynamika RA software. A blood RNA profile that may predict the clinical response to TNF antagonists will also be validated in this study. Synovial volume is a continuous measure of the same feature quantified with ordinal scores in RAMRIS. The synovial membrane volume has been shown to be correlated with local clinical signs of inflammation in cross-sectional studies ( 39 ). The major impediment to utilizing synovial volume as a clinical trial endpoint is the analytical complexity of obtaining its measure, but semi-automated computerized methods can speed this process. The volume of enhancing synovium can be taken from MRI sequences used for RAMRIS scoring. Statistically significant reductions in synovial volume have been seen in double blind trials with time points as short as 3 months ( 36,40 ).
Dynamika RA software is a computer-aided semi-automated method for quantitative analysis of DCE-MRI and endpoints available in this software are included for exploratory analyses. There are four major features of Dynamika RA software. First, a motion correction algorithm is applied to remove small motions that are common in RA patients and when analysis of small structures is desired. The 3D motion correction algorithm substantially raises the number of overlapping pixels compared with 2D or un-corrected frame sets. Artifactual enhancements from tissues adjacent to high contrast voxels are reduced ( 41 ). Second, Dynamika applies a modeling method to each voxel with signal intensity vs. time. Each voxel is categorized in one of 4 contrast categories-no signal, continuous uptake, presence of plateau, or presence of wash-out, and assigns a color to each enhancement condition for graphical display. Third, the model on which Dynamika is based uses a noise-based approach, avoiding the use of the threshold conditions to identify enhancing synovium, and contributes to the objectivity of the results. Fourth, the parametric graphic display of the modeled contrast vs time intensity curves allows the analyst to quickly quantify the distribution and volume of inflammatory activity using the number of voxels of each contrast category. Parameters of Early Enhancement Rate (called EER) and Relative Enhancement can be computed from the models rather than average raw signal intensity vs. time curves within handdrawn ROI. This permits the assessment of the average value in the whole wrist and metacarpophalangeal joints, largely obviating the entire issue of ROI choice. If required, the interface allows for placement of ROI for analysis and the exclusion of vessels. In addition, the number of voxels with contrast uptake, contrast plateau, and contrast washout, alone and in combination can be easily reported.
The ability to predict clinical response to infliximab is of great interest. An exploratory endpoint of this study is to assess the sensitivity and specificity of severa; blood gene expression signatures to predict, using only the baseline samples, the response at 14 weeks to infliximab as measured by DAS28(CRP). As an example, an 8-gene signature has been reported to have an 85% prediction PROTOCOL NO. P08136 PROTOCOL 05 JAN 2011 INITIAL PROTOCOL accuracy ( 42 ). Such advances require replication in other datasets. See Section 6.3 for a list of particular gene signatures.

Dose Rationale
Infliximab, at the dose recommended for initial therapy in RA will be used in this study at a dose of 3 mg/kg with an induction regimen of 0, 2, and 6 weeks, and an additional infusion at 14 weeks. Infliximab is administered intravenously in 250 ml of 0.9% sodium chloride solution (normal saline) with an in-line sterile filter over a 2hour period. This facilitates compliance and blinding of study treatment. The safety and efficacy of infliximab is well-described (SmPC). The major side effect associated with infliximab are hypersensitivity reactions and infections, including serious infections.
Infusion reactions are common and anaphylaxis may occur. Management of infusion reactions is not mandated in this protocol. However, patients who receive glucocorticoids to manage infusion reactions or for premedication to prevent infusion reactions will be discontinued from this study, as this may confound studying the effects of infliximab on clinical and MRI endpoints. Treatment of infusion reactions should include stopping or reducing the rate of the infusion, and optional treatment with antihistamine, acetaminophen/paracetamol, epinephrine and fluids. Several experience studies have published infusion reaction management schemes allowing almost all patients to tolerate infliximab therapy without glucocorticoids except in the case of treatment failure or angioedema, wheezing, hypotension or anaphylaxis ( 1,43 ).

Primary Objective
To compare the effect of infliximab on synovial inflammation as measured with Dynamic Contrast Enhanced (DCE)-MRI of one wrist with placebo over 14 weeks of treatment.

Exploratory Objectives
! To assess the sensitivity and specificity of each of the following mRNA signatures that predict clinical response to infliximab at 14 weeks using DAS28(CRP ! To assess t h e sensitivity and specificity of each of the following genetic polymorphisms to predict, using only the baseline samples, clinical response to infliximab at 14 weeks using DAS28(CRP).
! v-maf musculoaponeurotic fibrosarcoma oncogene homolog B (MAFB) on chromosome 20 (rs6028945 & rs6071980) ! type I interferon gene (IFNk) on chromosome 9; (rs7046653) ! paraoxonase I (PON1) gene on chromosome 7 (rs854555, rs854548, rs854547) ! TNF gene promoter, known as the 308 G/A variant (TNF-308) (rs1800629) ! To evaluate the responsiveness of endpoints available in Dynamika RA software to infliximab compared with placebo. These endpoints include EER and ME and the number of voxels showing contrast uptake, contrast plateau, and contrast washout, alone and in combination. ! To compare the effect of infliximab with placebo on synovial inflammation as measured by K trans derived from DCE-MRI of all segmentable synovium in wrist and all MCP in the field of view (FOV). ! To compare the effect of infliximab with placebo on RAMRIS scoring of bone erosions in wrist and MCP. ! To compare the effects of infliximab with placebo on a measure of inflammation that sums RAMRIS osteitis and synovitis scores, and a measure of damage that sums RAMRIS bone erosion and non-RAMRIS cartilage damage. ! To compare the effect of infliximab with placebo on the volume of osteitis.

Design of the Study/Methodology
Patients with active RA on a stable dose of methotrexate and with evidence of wrist synovitis on MRI that meet study eligibility criteria will participate. Patients will be randomized to receive intravenous infusion of either infliximab 3 mg/kg or placebo for 14 weeks, with infusions administered at Weeks 0, 2, 6, and 14. A MRI with intravenous GBCA taken of one wrist and MCP joints will be obtained within 7 days before the patient undergoes treatment with either infliximab or placebo and will be assessed for eligibility. Patients will be assessed for RA activity using DAS28 (an assessment of tender and swollen joints, the blood inflammation marker C-reactive protein (CRP) and patient's global assessment of disease status using a visual PROTOCOL NO. P08136 PROTOCOL 05 JAN 2011 INITIAL PROTOCOL analog scale (VAS)), patient assessment of pain, patient's and physician's global assessment, health assessment questionnaire (HAQ) and for adverse events before each infusion or MRI. MRI with GBCA will be obtained at Week 2, 4 and 14. The 2 week MRI may be removed from t h e study procedures based on Sponsor assessment of feasibility. The study will conclude after completion of procedures at Week 14. Adverse experiences will be collected though out the study, and for an additional 2 weeks afterwards. At the completion of week 14 procedures, receipt of a technically adequate MRI scan and successful resolution of safety data for each patient, the treatment assignment will be unblinded to the Investigator. The Sponsor of the study will cover the cost of open-label 3 mg / kg infliximab for the optional use of all patients completing the study in consultation with a treating rheumatologist for up to 16 weeks, as described in the infliximab dosing instructions. Patients previously assigned to placebo may receive a standard induction regimen at 0, 2, 6 and 14 weeks while patients previously assigned to infliximab may receive two additional treatments every 8 weeks. The use of infliximab during this period is optional, and the decision to treat with infliximab will be made by the patient and treating rheumatologist according to the infliximab SmPC.

Planned Flexibility in this Study
This protocol is written with some flexibility to accommodate for the inherent dynamic nature of Experimental Medicine clinical trials. The disease specific inclusion criteria, imaging endpoints, and the timing of the MRI procedure currently outlined in the protocol may be modified during the study based on newly available data. The flexibility required for these aspects of the study along with the rationale for such adaptive design are further described below. may be reordered to accommodate either data with technically satisfactory collection or to accommodate shorter MRI acquisitions. This decision will be communicated by a protocol clarification letter, and changes to the analysis plan will be finalized by dated memoranda prior to breaking of the treatment blind.
Finally, the 2 week MRI may be removed from the protocol procedures based on Sponsor assessment of feasibility. Any changes made to this protocol will be detailed in a protocol clarification letter and sent to the investigators for retention.
The investigators must forward a copy of this memo to the IRB/ERC. These changes will not increase the number of study procedures for a given subject during his/her participation in the entire study. PROTOCOL NO. P08136 PROTOCOL 05 JAN 2011 INITIAL PROTOCOL

Screening
Within 2 weeks prior to treatment, the investigator or qualified designee shall discuss with each subject the nature of the study, its requirements, and its restrictions. Written informed consent will be obtained from each subject prior to any study-related procedures being performed, and a signed copy will be given to the volunteer. The inclusion/exclusion criteria will be reviewed, and a complete medical history, physical examination, laboratory evaluations and other assessments, delineated in Section 2.2 and Section 7.6 will be performed. A screening number will be assigned to each subject on completion of the study-specific informed consent.
The hand and wrist MRI with intravenous GBCA will be the last screening procedure performed. The wrist with the greatest clinical involvement at baseline will be studied throughout the trial by MRI. The screening MRI must be graded as RAMRIS synovitis 1 in either radiocarpal or intercarpal joint regions by the central reader. MRI with intravenous GBCA without adequate image quality may be repeated once, after GBCA washout. For eligible subjects, the screening MRI will serve as the baseline scan.

Baseline and Treatment Visits
At the Baseline visit, the investigator or designee will review the inclusion/exclusion criteria and record adverse events (AEs) and medications taken within the last 14 days. Laboratory evaluations, vital signs and other assessments, delineated in Section 2.2 and Section 7.6 will be performed and/or recorded. If not previously assigned, a screening number will be assigned. When a subject fulfills the entry criteria, the subject will be assigned a randomization number that will be used to determine the treatment according to a randomization code (see Section 7.4.1.3).

Study Completion/ Follow-Up
On the last day of their s t u d y participation (including premature discontinuation), subjects will be discharged from the study following the collection of the last blood sample/assessment delineated in Section 2.2 and Section 7.6.
Approximately 2 weeks after t h e last s t u d y visit (including premature discontinuation), subjects will be contacted by phone to assess for potential adverse events. Subjects who completed the study at Week 14 will be offered the option to receive open label infliximab treatment following the regimen described in the infliximab product leaflet. The cost of this additional treatment is covered by the Sponsor for up to 16 weeks.

Participation in and Completion of the Study
Each subject is considered to be enrolled in the trial when the subject (or the subject's legal representative) has provided written informed consent. All subjects will have their information entered into the screening log maintained by the study A subject is considered to have discontinued participation after he/she has withdrawn consent or has been discontinued under the conditions specified in A subject is considered to have been lost to follow-up if the investigator is unable to contact the subject. The end of participation for a subject lost to follow-up is the last known contact (eg, visit or telephone contact).
The trial begins when the first subject is enrolled (i.e., signs the informed consent form). The trial ends when t h e last remaining subject has ended participation in the trial, by completing the trial, being discontinued from the trial, or being lost to follow-up.
Follow-up procedures related to pregnancy or (S)AEs may continue beyond the end of the clinical trial.
For patients completing the study assessments at Week 14, the Sponsor of the study will cover the cost of a limited supply of open-label infliximab at 3 mg/kg for optional use after the trial concludes. A patient and treating rheumatologist will elect whether or not to initiate or continue infliximab treatments. It is recommended that infliximab dosing instructions are followed in accordance with the SMPC (See also Appendix 5) Unused infliximab may not be used for other patients. No additional provision for infliximab is available through the Sponsor.

Study Population
Approximately 60 adult subjects with a diagnosis of Rheumatoid Arthritis will be selected for the study. Approximately 20 subjects will be enrolled at each site.
There is no quota for enrollment at any given site. Subjects must meet all the inclusion criteria and none of the exclusion criteria to receive treatment assignment. PROTOCOL NO. P08136 PROTOCOL 05 JAN 2011 INITIAL PROTOCOL

Subject Inclusion Criteria
The subject must meet ALL the criteria listed below for entry: 1. Subjects must be willing to give written informed consent for the trial and able to adhere to dose and visit schedules.

2.
Subjects must be willing to give written informed consent for pharmacogenomic testing, and able to adhere to applicable visit schedules.

Note:
Subjects who are unwilling to sign the informed consent for pharmacogenomic testing may be included into t h e trial, however, pharmacogenomic samples must not be obtained.

3.
Subjects of either gender and of any race 18 years of age.

4
Subjects must have a clinical diagnosis of rheumatoid arthritis based on the 1987 ACR clinical criteria for at least 6 months (Appendix 3).
Note: Newly diagnosed subjects must have had clinical symptoms consistent with rheumatoid arthritis for at least 6 months as assessed by medical history and must fulfill the clinical criteria specified in Appendix 3.
(Note: this is not an inclusion criterion but is the criterion for wrist selection).
b. At the prestudy/screening visit:

8.
Women of childbearing potential and all men must agree to use a medically accepted method of contraception prior to entering the study (beginning at least 2 weeks prior to administration of the first dose of study drug, throughout the trial and for 6 weeks after the completion of the trial (at Week 14)). Acceptable methods of contraception include condoms (male or female) with or without spermicide, diaphragm or cervical cap with spermicide, medically prescribed intrauterine device (IUD), oral or injectable hormonal contraceptive, and surgical sterilization (e.g. hysterectomy, tubal ligation for women, and vasectomy for men). Women of non-childbearing potential (menopausal) do not have to use any contraceptive methods (menopause is defined as the time when there have been no menstrual periods for 12 consecutive months, with an FSH value in the postmenopausal range at the prestudy/screening visit, and no other biological or physiological cause can be identified).

9.
Women of childbearing potential must demonstrate a serum -hCG level consistent with the non-gravid state at the prestudy/screening visit. f. Have a chest radiograph (both posterior-anterior and lateral views), taken within 3 months prior to the first administration of study agent and read by a qualified radiologist, with no evidence of current active TB or old inactive TB.

11.
Subjects have received methotrexate therapy for 3 months prior to the Baseline visit. The dose of methotrexate must be stable for 8 weeks prior to Baseline. In addition, subjects on methotrexate must also be taking a stable dose of folate for 4 weeks prior to Baseline. Subjects receiving 7.5 mg methotrexate (alone or in combination with other DMARDS) must have received methotrexate therapy for 8 weeks and at a stable dose for at least 8 weeks prior to baseline.

12.
Subjects taking the following DMARDs in combination with methotrexate must be on a stable dose for the duration outlined below: a.
For subjects taking D-penicillamine, hydroxychloroquine, chloroquine, oral or parenteral gold, sulfasalazine, mycophenolate mofetil, the dose must be stable for 6 weeks prior to the Baseline visit b.
For subjects taking leflunamide, the dose must be stable for 8 weeks prior to the Baseline visit.
c. For subjects taking Cyclosporine A and/or azathioprine, the dose must be stable for 12 weeks prior to the Baseline visit.
Note: Any changes in stable doses of DMARDs during the course of the trial will result in discontinuation of the subject from t h e trial at the discretion of the SPONSOR medical monitor.

13.
Subjects taking oral corticosteroids must be on a stable dose equivalent to 10 mg of prednisone (or prednisolone) per day for 2 weeks prior to the Baseline visit.

14.
Subjects taking daily NSAID must be on a stable dose for 2 weeks prior to the Baseline visit. The dose of NSAID must remain stable throughout the trial and until completion of trial at Week 14.

15.
Subjects taking NSAID for pain on an as-needed basis must agree to discontinue NSAID use for at least 3 days and use only paracetamol for breakthrough pain for 3 days before each MRI and clinic visit (at Baseline, Weeks 2, 4, and 14). Further, these subjects must withhold taking any pain medication (including paracetamol) for at least 12 hours prior to each MRI and clinic visit.

16.
Subjects who have received biological therapies (infliximab, etanercept, adalimumab, or pegsunercept) may participate, as long as the last dose of these drugs was received 3 months prior to the Baseline visit AND the reason for discontinuations was not for safety considerations OR lack of efficacy.

17.
Subjects agree to avoid unaccustomed strenuous physical activity (e.g., unaccustomed weight lifting, initiation of new physical therapy program) for at least 3 days before each MRI or clinic visit (at Baseline, Weeks 2, 4, and 14).

18.
Subjects are able to understand and complete study questionnaires (including the use of a visual analog scale [VAS] to report responses).

Subject Exclusion Criteria
The subject will be excluded from entry if ANY of the criteria listed below are met: 1. Female subjects who are pregnant, intend to become pregnant, or are breastfeeding.

4.
Subjects with moderate or severe congestive heart failure.

5.
Subjects with a history of or current signs and/or symptoms of severe,

20.
Subjects who have received any treatment listed in Table 2 more recently than the indicated washout period prior to the Baseline visit, which, in the opinion of the investigator and sponsor, interferes with their ability to participate in the trial.

Subject Discontinuation Criteria
Subject participation may be terminated during the study for any of the ! Ü A diagnosis of active TB is made or suspected.
It is the right and the duty of the investigator to interrupt treatment of any subject if he/she feels that study discontinuation is necessary to protect the subject, or that there are unmanageable factors, that may interfere significantly with the study procedures and/or the interpretation of results.
If a subject prematurely discontinues, or is discontinued from the study, the primary reason for the discontinuation will be obtained and recorded on the eCRF.

Replacement of Subjects
After consultation between the sponsor and t h e principal investigator, enrollment may be extended to replace subject(s) discontinued during the study.

Treatments
Subjects randomized to infliximab will receive infliximab 3 mg/kg by intravenous (IV) infusion over a 2-hour period at Week 0, Week 2, Week 6, Week 14. Subjects randomized to placebo will receive 0.9% sodium chloride (normal saline) by intravenous (IV) infusion over a 2-hour period at Week 0, Week 2, Week 6, Week 14.
Subjects will continue to receive their standard dose and regimen of DMARDS (e.g. methotrexate and folate), NSAID or COXibs, and/or glucocorticoid (e.g. prednisone as long as dose is 10 mg per day). However, patients who receive glucocorticoids to manage infusion reactions or for premedication to prevent infusion reactions will be discontinued from the study. The bag is hung on an IV pole and covered by a concealing drape. The IV tubing will also be covered by a concealing drape. Infliximab or placebo is administered intravenously via a filter. Infusions of infliximab should be administered over at least 2 hours. The administration of study drug is performed by study personnel blinded to treatment assignment.
No other drugs should be administered simultaneously in the same intravenous line as infliximab.

Infusion Reactions
Infusion reactions are common and anaphylaxis may occur. However, patients who receive glucocorticoids to manage infusion reactions or for premedication to prevent infusion reactions will be discontinued from this study. Treatment of infusion reactions should include stopping or reducing the rate of t h e infusion, and optional treatment with antihistamine, acetaminophen/paracetamol, epinephrine and fluids. Several experience studies have published infusion reaction management schemes allowing almost all patients to tolerate infliximab therapy without glucocorticoids except in the case of treatment failure or angioedema, wheezing, hypotension or anaphylaxis (LeQuerre et al., 2006) (SmPC).

Timing of Dose for Each Subject
Subjects will receive study medication in the clinic. At the Baseline visit, it is critical that study medication be administered after the collection of a technically adequate hand and wrist MRI; samples for laboratory evaluation and clinical assessments for the computation of DAS28(CRP) at baseline should also be performed prior to initiating study treatment. See Section 2.2 and Section 7.6.

Method of Treatment Assignment, Randomization, and/or Stratification
The study will be stratified by center. Stratified randomization with permuted block block allocation within each stratum will be used and the allocation schedule will be PROTOCOL NO.  1 and 7.3.2 will be assigned a randomization number by the site's unblinded personnel. Subjects will then receive the study treatment corresponding to their assigned randomization number.
Randomization Nos. 001 to 060 will receive Treatment A or Treatment B.
The randomization number will be recorded in the study case record (eg, eCRF).
The unblinded trained person will be provided with the randomization number to be assigned to each replacement that corresponds to the same treatment as that received by the subject being replaced.

Management of Blinding of Study Treatments
Blinding will be preserved for the infliximab and placebo infusion treatments.
Infliximab and placebo infusion treatments will be blinded to the investigator, study staff, and subject. The infusions will be prepared (according to the randomization schedule provided by the sponsor) by an unblinded trained person at each study center in accordance with Section 7.

Investigational Product
The investigator shall take responsibility for and shall take all steps to maintain appropriate records and ensure appropriate supply, handling, storage, distribution, and usage of these materials in accordance with the protocol and any applicable laws and regulations.
The following Investigational product(s) will be used in the study and will be supplied in sufficient quantities for subjects and any replacement subjects.

Identity of Investigational Product(s)
Refer to the SmPC and Appendix 5 for a description of the investigational drug product(s).

Source
Infliximab will be purchased by the site from the same batch/lot if possible.
The site will be responsible for recording the lot number, manufacturer, and expiry date of any locally purchased drugs and for forwarding this information along with the package insert to the sponsor.

Labeling, Storage and Dispensing
Study drug supplies must be stored in a secure, limited-access location under the storage conditions specified on the drug supply label.
Receipt and dispensing of study medication must be recorded by an authorized person at the investigator's site. Clinical supplies may not be used for any purpose other than that stated in the protocol.
An adequate quantity of study drug will be provided to replace units damaged in shipment/handling, to replace subjects meeting the entry criteria.

Investigational Product Accountability
Accurate and current accounting of t h e dispensing and return of investigational product(s) will be maintained on an ongoing basis by a member of the trial site staff: ! Investigational medicinal product(s) dispensed to each site will be recorded in the trial-specific Site Investigational Medicinal Product (IMP) Accountability Log (or equivalent document approved by the sponsor); ! Investigational medicinal product(s) dispensed to each subject will be recorded in the trial-specific Subject IMP Accountability Log (or equivalent document approved by the sponsor).
The Site IMP Accountability Log and Subject IMP Accountability Log will be verified by the sponsor's trial monitor. The original Site IMP Accountability Log and Subject IMP Accountability Log will be approved by the investigator and retained at the trial site and a copy supplied to the sponsor when the trial is complete.
As infliximab is an intravenous treatment, there are no test articles that the subject needs to return to the sites.
The following test articles may be discarded by the site immediately after use: to government inspection at any time.

Modification of Dose and/or Administration of Investigational Product for a Subject
The dose and administration of the IMP to any subject may not be modified.
If necessary a subject must be discontinued for t h e reasons described in Section 7.3.3.

Non-Study Treatments
Patients will continue to receive their standard dose and regimen of DMARDS (e.g. methotrexate and folate), NSAID or COXibs, and/or glucocorticoid (e.g. prednisone as long as dose is 10 mg per day). The Sponsor will not provide these medications.
Subjects taking daily NSAID must be on a stable dose for 2 weeks prior to the Baseline visit. The dose of NSAID must remain stable throughout the trial and until completion of trial at Week 14.
Subjects taking NSAID for pain on an as-needed basis must agree to discontinue NSAID use for at least 3 days and use only paracetamol for breakthrough pain for 3 days before each MRI and clinic visit (at Baseline, Weeks 2, 4, and 14). Further, these subjects must withhold taking any pain medication PROTOCOL NO. P08136 PROTOCOL 05 JAN 2011 INITIAL PROTOCOL (including paracetamol) for at least 12 hours prior to each MRI and clinic visit (at the timepoints listed above).

Prior and Concomitant Medications
All prior medication taken by t h e subject 14 days prior to treatment intervention and all concomitant therapy taken by the subject during the study are to be recorded on the eCRF. The identity of the therapy, the dose, route, and regimen, the dates started and stopped (or notation of "continuing" if that is the case), and the reason for use must be recorded. The use of any concomitant medication must relate to an AE or the subject's medical history.

Medications, Supplements, and Other Substances from Baseline and During the Study
The medications prohibited prior to Baseline and during the study are listed in Section 7.3.2 with the subject exclusion criteria.
During non-confined study period, subjects should refrain from the use of nutraceuticals, vitamins at doses that exceed daily replacement requirements, and antacids.

Medications Allowed During the Study
The

! Explain Study and Obtain Written Informed Consent
The investigator or qualified designee will explain the study and all study requirements to the subject, answer all of his/her questions, and obtain written informed consent before performing any study-related procedure, including pharmacogenomic sampling and the collection of any samples for future use. A copy of the informed consents will be given to the subject.

! Review Inclusion/Exclusion Criteria Including Concomitant Medications
The inclusion and exclusion criteria will be reviewed by the investigator or qualified designee to ensure that the subject qualifies for the study. All appropriate treatment and washout times will be discussed with the subject. All medications used during the 14 days prior to first drug administration or treatment intervention, will be recorded on the eCRF.

! Demographic Profile
The demographic profile required by the study eCRF and permitted by local regulations should be entered. This information may include the subject's age (at baseline), ethnicity, gender and race. PROTOCOL

! Medical History
A detailed medical history will be obtained by the investigator or qualified designee. Subject history should include information on clinically significant family and personal history, smoking history, and history of diseases (including past history of hepatitis, allergies or reactions to medications and food, seizures, previous surgery, trauma, syncope, and arrhythmias). For rheumatoid arthritis, year of onset, duration, complications, surgeries and treatments should be recorded. Any clinically relevant changes found at any time during the study will be recorded as adverse events on the eCRF.

! Physical Examination
The principal investigator or designee will perform a physical examination of the following organ systems at the times specified in Section 2.2, Study Flow Chart: Eyes, oropharynx, thyroid, respiratory system, cardiovascular system, abdomen, skin, extremities and reflexes.
Clinically significant changes from the Screening physical examinations will be recorded as adverse events.

! Body Height (cm) and Weight (kg) Measurements (cm) Without Shoes
Body height and weight measurements should be obtained and recorded to the nearest kg and cm on the eCRF.

! Laboratory Tests (CBC, Chemistry, Urinalysis)
The following will be performed according to standard laboratory procedures. All samples should be collected prior to study drug administration. PROTOCOL  If during the trial any laboratory result is outside the reference range and is considered to be clinically significant by the investigator, the test should be repeated at appropriate time intervals until it returns to baseline or becomes a clinically insignificant finding. Any adverse clinically significant change will be recorded on the eCRF as an Adverse Event. All laboratory tests, including urinalysis results, will be recorded on the eCRF;

! Pregnancy Tests and FSH
A serum pregnancy test must be performed at the Screening visit and a urine pregnancy test must be performed at the Baseline visit for females of childbearing/conceiving potential (including females with a partial hysterectomy or tubal ligation). The urine pregnancy test is also performed periodically throughout the trial as outlined in Section 2.2. Results must be available prior to dosing. Subjects with a positive pregnancy test at screening or prior to the first dose will not be allowed to enter into the study (screen failure). Subjects with a positive pregnancy test anytime after first dose will be discontinued from the study. Positive results will be recorded on the Adverse Event section of the eCRF. The pregnancy will be reported and monitored according to The FSH test is performed on post-menopausal women only.at the Screening visit.

! Test for HIV Antibodies, Hepatitis B Surface Antigen and HCV
Tests will be performed according to standard local procedures and are performed at the discretion of the Investigator. A plan must be in place at the site(s) for the management of a positive test result according to local requirements. Results will not be recorded on the eCRF.

! Screen for Drugs With a High Potential for Abuse
Urine tests will be conducted according to standard procedures and are performed at the discretion of the Investigator. In addition to commonly abused drugs, any drugs abused locally with a high prevalence, should also be tested. Positive results will be recorded on the Comment section of the eCRF, and will be reported to the sponsor.

! Tuberculosis Screening and Evaluation
Tuberculosis screening includes a review of subjects' medical history, a tuberculin skin test or QuanitferonTB Gold test (Appendix 2) and chest xray. Local guidelines specifically directed at high risk and immunocompromised subjects should be used for screening and interpretation of positive test results. Tuberculin skin testing must be performed and read by experienced, trained, and licensed personnel according to published local guidelines, either at the participating trial site, a local public health clinic, or a primary care physician's office. If the test was performed at any site other than the principle investigator's facility, the results must be made available in written form as source documentation.
QuantiFERON-TB Gold testing is an acceptable alternative to tuberculin skin testing. In the case of a positive TB-test (either positive skin test or QuantiFERON-TB Gold test or findings on chest x-ray suggestive of latent TB such as pleural scarring), a subject can participate in this trial only if the subject agrees to receive TB prophylaxis with isoniazid 300 mg once daily starting from 2 weeks prior to Baseline.

!
Following completion of the study-specific informed consent by the subject, a screening number will be assigned to the subject by the site. Screening numbers will be recorded on the eCRF. PROTOCOL

! Randomization Number Assignment
Prior to the administration of study treatment, a randomization number will be assigned to each volunteer by the site using the randomization numbers provided in Section 7.4.1.3.

! Issue or Collect Subject Identification Card
The investigator or qualified designee will provide the subject with a Subject Identification Card when the subject is no longer under the direct observation of the investigator/site. The investigator or qualified designee will retrieve the card from the subject at the last contact (see Section 9.1.3 for further description of the Subject Identification Card).

! Clinical Assessment of Electrocardiograms
The principal investigator/designee will record Screening ECG data and any unscheduled ECG data in full on the eCRF.

Local ECGs
The 12-lead ECG (12-lead at 25 mm/sec reporting rhythm, ventricular rate, PR, QRS, and QT intervals) will be performed by the principal investigator or designee after the subject has been in the supine position for at least 5 minutes.
A commercial 12-lead electrocardiograph capable of recording, storing, and printing on full-size paper, high resolution 12-lead data, should be used. On the electrocardiogram should be recorded the date, time, subject demographics (eg, age, and sex), subject number, and studyrelated information (eg, protocol number and nominal time).

! Vital Signs
Vital signs will be obtained by the principal investigator or designee. Subjects will be in a semi-recumbant position for at least 3 minutes. Systolic and diastolic blood pressure (mm Hg), pulse rate (bpm) and oral body temperature ( C) and the actual clock time (24:00) will be recorded on the eCRF. Any clinically significant change from baseline for vital sign measurement (not associated with another adverse event) will be recorded on the adverse event eCRF form. PROTOCOL NO. P08136 PROTOCOL 05 JAN 2011 INITIAL PROTOCOL All blood pressure measurements should be made by consistently using the same arm and type of equipment such as a mercury sphygmomanometer (preferred), a recently calibrated aneroid manometer, or a validated automated electronic device, with an appropriate cuff size. Auscultatory or oscillometric technique should be used with the mercury sphygmomanometer or aneroid manometer. Pulse rate should be obtained manually by radial pulse palpation over a 30 second count or by a validated automated electronic device.
Body temperature should be measured consistently using the same approach (oral or axillary). Since drinking hot or cold water (or other beverages) has a significant impact on recorded oral temperature, hot or cold beverages should not be ingested within 15 minutes of the oral temperature measurement.
If the scheduled time for vital sign measurements coincides with a blood collection, the vital signs should be performed prior to the blood collection, or at least 5 minutes afterwards. This same timing for obtaining the vital signs (before or after the blood collection) should be used for all vital sign measurements.

! ACR Criteria for Diagnosis of RA
Subjects will be evaluated against t h e ACR Criteria through an assessment of medical history, physical examination, laboratory and radiographic findings. See Appendix 3 for the ACR criteria.

! Swollen and Tender Joint Count
The 28 joint count for pain and swelling will be used. The following 28 joints will be assessed: Finger Proximal Interphalangeal Joints (8), thumb interphalangeal joint (2), metacarpophalangeal (MCP) (10), wrists (2) (includes carpometacarpal, intercarpal, and radiocarpal), elbows (2), shoulders (2), and knees (2). Tender and Swollen Joint Counts will be performed by a masked joint examiner with no other role in the clinical care of the patient. The same masked joint examiner should be used for all assessments of an individual patient. Tender and Swollen Joint Counts must occur before initiating study treatment at the relevant visits.

! Patient's Global Assessment of Disease Activity (GADP) Visual Analog Scale (VAS)
Subjects will be asked to make an overall global assessment of their disease activity on a VAS (scale of 0 to 100 mm, with 0 being very well to 100 being very poor).

! Patient's Global Assessment of Pain Visual Analog Scale (VAS)
Subjects will be asked to make an overall global assessment of pain due to their disease during the past week on a VAS (scale of 0 to 100 mm, with 0 being no pain to 100 being extreme pain).

! Physician's Global Assessment of Disease Activity Visual Analog Scale (VAS)
Physician's global assessment of disease activity will be recorded on a VAS (scale of 0 to 100 mm, with 0 being very well to 100 being very poor).

! CRP, ESR, Rheumatoid Factor
An ESR by Westergren method or the CRP maybe used to satisfy the inclusion criteria outlined in Section 7.3.1. The CRP is also used to compute the DAS28(CRP). Laboratory tests for CRP and rheumatoid factor will be performed by the central laboratory.

! Samples for Responder Analyses of Genetic Polymorphisms and Gene Expression Signatures
Blood samples are required from all randomized subjects for analyses of DNA polymorphisms and blood mRNA signatures. An 8.5 mL blood sample (for DNA) and a 2.5 mL blood sample (for RNA) are collected at baseline.

! Pharmacogenomic (PGx) Samples
Informed consent specific for PGx sampling must be obtained prior to collection. To obtain sufficient DNA, RNA, Serum, Plasma for pharmacogenomic studies, a single 8.5 mL blood sample, 6 x 2.5 mL RNA samples, 6 x 6 mL Plasma samples, 6 x 10 mL Serum samples will be drawn at the specified time point indicated in Section 2.2, Study Flow Chart, into the appropriate tube provided by the sponsor (see

Serum Samples for Determination of Pharmacodynamic Responses (Future Use)
Serum samples are collected for archive at the specified time points indicated in Section 2.2, Study Flow Chart. Informed consent for future use must be obtained before these samples can be collected.

! Samples for High Dimensionality Cytometry Cell Stimulation and High-Dimensionality Flow Cytometry
This high dimensionality cytometry assay involves stimulation of peripheral blood mononuclear cells with certain stimulants (e.g. cytokines, TLR ligands, etc.) followed by high-dimensionality flow cytometry on t h e stimulated cells.
The method emphasizes measurement of activation states of proteins thought to be nodes in the signaling networks of various gated cell populations.
While the method is labor intensive and technically complex, published results ( 50 ) and preliminary results (based on internal discussions with and Nodality) suggest the method may have unique promise as a predictive platform for subtle immunologic phenomena. Furthermore, Nodality has validated SOPs for sample collection, shipping, processing, and assay; they also have a CLIA-certified laboratory.
The site based procedure involves drawing blood from the patient. Further, the site must be able to process blood to generate and store peripheral blood mononuclear cells (PBMC). These assays may not be incorporated into the study since they are contingent on factors unable to be resolved at the time of protocol authorship.

! Dynamic Contrast Enhanced Magnetic Resonance Imaging
One wrist and MCP joints will be imaged with MRI using GBCA at various timepoints in this trial. The first MRI will be performed during Screening, as part of the review of inclusion criteria. This MRI must show evidence of synovitis in the wrist, as determined by the central reader. For eligible subjects, the screening MRI will also serve as the baseline scan. The same wrist and MCP joints will be followed and imaged at the subsequent visits. The MRI should be performed prior to study drug administration at the relevant visits.

DCE-MRI:
! K trans of the total enhancing tissue and total enhancing synovium of the wrist.
K trans reflects flow and permeability surface area of enhancing synovium.
! IAUC 90 the total enhancing tissue or synovium reflects flow and permeability surface area and the contrast agent distribution volume in the first 90 seconds after injection.
! The number of voxels showing contrast uptake, contrast plateau, and contrast washout, alone and in combination using Dynamika RA software.

High Dimensionality Cytometry
This assay may be performed. See Section 7.6. PROTOCOL NO. P08136 PROTOCOL 05 JAN 2011 INITIAL PROTOCOL

Pharmacogenomics Analysis
The effect of the following genetic polymorphisms on clinical response to infliximab will be investigated: •

PD/PGx, PK/PGx and Safety/PGx Analysis
Pharmacogenomics inter-relationships may be explored.

Other Assessments
Other assessments include the Health Assessment Questionnaire ,the Patient and Physician's Global Assessment Visual Analog Scale, and the Patient's Assessment of Pain Visual Analog Scale as described in Section 7.6.

Safety Monitoring and Assessments
Safety variables to be assessed include: vital signs, ECGs, reporting of adverse events, hematology and blood chemistry. Life-threatening in the definition of serious adverse event, refers to an event in which the subject was at risk of death at the time of the event; it does not refer to an event which hypothetically might have caused death if it were more severe.
Medical judgment should be exercised in deciding whether an adverse event/reaction is serious in other situations. Important adverse events/reactions that are not immediately life-threatening, or do not result in death or hospitalization, but may jeopardize the subject or may require intervention to prevent one of the outcomes listed in the definition above, should also be considered serious.

Closely Monitored Event
A "closely monitored event" is a non-serious adverse event or occurrence that is designated to be of special interest and must be reported to the sponsor as though it were a serious adverse event -as described in Section 7.7.2.5.1.
No adverse events are considered closely monitored for this protocol.

Potential Medication Error
A potential medication error is an individual case safety report of information or complaint about product name, labeling, or packaging similarities that does not involve a subject or patient (eg, If a subject reports that one of the investigational products looks like a different product, the report would be considered a potential medication error).

Incidents
Product complaints which led to or might have lead to death or serious deterioration of health/serious injury/serious illness for the user or any other person.

Monitoring Adverse Events
Subjects will be monitored for the occurrence of AEs -including SAEsimmediately after the subject has signed the informed consent form. For all AEs that require the subject to be discontinued from the trial and SAEs, relevant clinical assessments and laboratory tests will be repeated as clinically appropriate, until final resolution or stabilization of the event(s).

Monitoring Laboratory Assessments
The clinical laboratory values will reviewed by the investigator for significance and consideration as an AE.

Assessment of Severity
Where t h e determination of adverse event severity rests on medical judgment, t h e determination of severity must be made with t h e appropriate involvement of a medically-qualified investigator.
The severity of AEs will be graded according to the following definitions: Mild: awareness of sign, symptom, or event, but easily tolerated; Moderate: discomfort enough to cause interference with usual activity and may warrant intervention; Severe: incapacitating with inability to do normal daily living activities, or significantly affects clinical status, and warrants intervention; PROTOCOL NO. P08136 PROTOCOL 05 JAN 2011 INITIAL PROTOCOL

Assessment of Causality
A medically-qualified investigator must assess the relationship of any AE (including SAEs) to the use of the investigational product, as unlikely related, possibly related, or probably related, based on available information, using the guidelines listed below: Unlikely related: no temporal association, or the cause of the event has been identified, or the drug, biological, or device cannot be implicated based on available information; Possibly related: temporal association, but other etiologies are likely to be the cause; however, involvement of the drug, biological, or device cannot be excluded based on available information; Probably related: temporal association, other etiologies are possible, but unlikely based on available information.

Reference Safety Information (RSI) for the Assessment of Expectedness of Adverse Events
The Reference Safety Information (RSI) for assessing the expectedness of an adverse event for infliximab in this current trial is to be the most recent SmPC.

Known Potential Toxicities of Study Drug
The following AEs were reported during clinical trials and/or during postmarketing experience with infliximab: 5.
Incidents associated with the device;
Any occurrence of a product quality complaint by a subject in the trial must be reported expeditiously by the investigator or qualified designee to the sponsor or designee using t h e Investigational Medicinal Product Quality Complaint Form provided by the sponsor/designee within 1 working day of becoming aware of the event.
Any occurrence of the following events or outcomes in a subject in the trial must be reported expeditiously by the investigator or qualified designee to the sponsor or designee using the Safety Data Reporting Form 1727 within 5 working days of becoming aware of the event.
1. Pregnancy, exposure during pregnancy or lactation NOT associated with an SAE-including the pregnancy of a male subject's female partner who has provided written informed consent to provide information regarding pregnancy; Any observation reported to the sponsor or designee via the Safety Data Reporting Form 1727 that is also an AE, is to be recorded in the CRF (Section 9.2), as well as in the subject's source documentation along with any actions taken as a result of AE and follow-up results.
If an autopsy is performed, the de-identified autopsy report must be provided to the sponsor within 1 working day of the results being available.
The Safety Data Reporting Form 1727 requires that the investigator assess causality of the event relative to the investigational product administered in the trial (Causality is described in Section 7.7.

2.3.2).
Investigators are to follow the requirements for reporting to the Sponsor information regarding AEs, SAEs, closely monitored AEs, pregnancy exposure, lactation exposure, overdoses, medication errors (including potential), preplanned hospitalizations, investigational medicinal product quality complaints, and subject deaths.

Expedited Reporting by the Sponsor to a Regulatory Health Authority
GPV will monitor data for safety. The Sponsor will manage the expedited reporting of relevant safety information to concerned health authorities, competent authorities, and IRBs/IECs in accordance with local laws and regulations. Should a subject be discontinued from the trial, complete the visit activities as specified for discontinuation in the Trial Flow Chart in Section 2.2.

Temporary Interruption of Treatment for a Subject
A Subject may not temporarily interrupt and then restart treatment. The investigator is to discontinue a subject as necessary according to the criteria provided in Section 7.3.3.

Criteria for Early Termination of the Trial
The clinical trial may be stopped if the extent (incidence and/or severity) of emerging effects/clinical endpoints is such that the risk/benefit ratio to the trial population as a whole is unacceptable. The clinical trial may also be stopped for administrative reasons.
In addition, further recruitment in the trial or at (a) particular site(s) may be stopped due to insufficient compliance with t h e protocol, GCP and/or other applicable regulatory requirements, procedure-related problems, or the number of discontinuations for administrative reasons is too high. PROTOCOL NO. P08136 PROTOCOL 05 JAN 2011 INITIAL PROTOCOL

All-Subjects-Evaluable group
The All-Subjects-Evaluable group will consist of all subjects from the All-Subjects-Treated group for whom at least one pharmacodynamic parameter can be calculated according to the protocol and who did not have any protocol violation interfering with pharmacodynamics.

Demographic and Other Baseline Characteristics
Demographic variables (eg, sex, race, age, weight, elbow breadth) will be listed and summarized using descriptive statistics for the entire study population.

METHODS:
The primary metric for assessing treatment effect at Weeks 14, 4, and 2 is change from baseline. Currently, K trans is considered a primary metric, provided that technically adequate MRI scans are obtained from the first 6 to 10 patients enrolled in this study. If a patient-specific Arterial Input Function (AIF) cannot be measured, then a population AIF may be substituted for patient-specific AIF. Primary endpoint will be compared among the treatment groups via a constrained Longitudinal Data Analysis (cLDA) according to Liang and Zeger ( 51   Statistical significance of the dichotomous endpoints (ACRx) will be determined using Fisher exact test comparing placebo and treatment groups.
P-values and 90% CI's will be computed. These analyses will be carried out for change from baseline at each of weeks 14 and 6. Interpretation of p-value testing for each endpoint will be made in step-down fashion (using closed stepwise procedure) with week 14 interpreted first. Only if significant treatment difference is observed for week 14, then treatment difference testing of week 4 and then week 2 will be examined. Significance test will be carried out at alpha=0.05, 1-sided for the primary and other study endpoints.
Effect sizes and associated 90% CI's will be computed to assess precision of each endpoint. PROTOCOL NO. P08136 In addition, the gene signatures will be tested for their mean difference of expression values between responder and non-responder groups using a combination of a fixed sequence step-down procedure and Holm's step-down procedure to control the overall alpha level at 0.05. Among the seven gene signatures, the Julia 8-gene Signature will be tested first at alpha=0.05, two-sided. If the test result is nonsignificant, we will stop and the remaining six signatures remain non-significant for the purpose of conclusions. If the "Julia" signature is shown to be significant, then the other six signatures will be tested by Holm's step-down procedure to control the overall alpha=0.05. More specifically, the p-values from the six signatures will be ordered from the smallest to the largest, p(1) p(2) p(3) p(4) p(5) p(6), where p(i) is the i th largest p-value for 1 i k. Here k=6. If p(1) < alpha / k= 0.0083, the corresponding signature is significant. We will move on to compare p(2) with alpha / (k-1)= 0.01. If the corresponding signature is significant, we will continue to compare p(3) with alpha / (k-2)= 0.0125. If the corresponding signature is significant, we will continue to compare p(4) with alpha / (k-3)= 0.016. If the test result is significant, we will compare p(5) with alpha / (k-4) = 0.025. And if again that is significant, we will test the last signature with the largest p-value, i.e. to compare p(6) with alpha=0.05.
If p(6) < 0.05, we can make a conclusion that all signatures are statistically significant. In this Holm's procedure, whenever a non-significant test occurs, the procedure stops for conclusion purposes, and the remaining tests remain non- In addition to the classification models for the prediction of classes of responder nonresponders obtained by dichotomization of the clinical scores, regression model based on random forest will be also developed to predict DAS28(CRP) scores on the continuous scale. The model will be validated using external (nested) crossvalidation. Cross-validated (out-of-bag) MSE and R 2 and variable importance for each gene in a particular signature will be reported for this analysis.

Genetic Polymorphisms (Single Nucleotid Polymorphisms SNPs) -Responder Analysis
First, the responder (good response) and non-responder (moderate or bad response) status for each patient will be inferred by DAS28-ESR EULAR criteria on the change in disease activity measured by DAS28-ESR EULAR response at 14 weeks as carried out for the gene signatures before.
Then a logistic regression model on the responder status will be fitted for each of the SNP with term SNP genotype and any other relevant covariates. An additive inheritance model will be used as the primary genetic model. Dominant, recessive and general (categorical) model will also be considered for sensitivity analysis. For each candidate SNP, the odds ratio of the risk allele being 1 in responder vs. nonresponder group will be tested. For each significant SNP, a prediction rule will be developed using the univariate analysis. Multivariate classifier using the SNP signature with more than 1 SNP will also be fitted with the SNPs as the independent variables. The sensitivity and specificity will be evaluated on those prediction rules.
Receiver operating characteristic (ROC) curves will be drawn for those genetic polymorphisms. The performance of the SNP signature will be quantified by the area under the ROC curve. Similarly to the analysis of gene signatures, external crossvalidation will be applied to correct for the over-optimism as an internal validation.
There are four genes (seven SNPs) to be tested for its odds ratio of risk allele in responder versus non-responder groups. Multiplicity will be adjusted by Holm correction to control the overall type I error rate at level 0.05. PROTOCOL NO. P08136 PROTOCOL 05 JAN 2011 INITIAL PROTOCOL In addition to the classification models for the prediction of classes of responders vs non-responders, obtained by dichotomization of clinical scores, a regression model based on random forest will be also developed to predict DAS28(CRP) scores on the continuous scale. The model will be validated using external (nested) crossvalidation. Cross-valided (out-of-bag) MSE and R 2 and variable importance for each SNP will be reported for this analysis.

Pharmacodynamic-Pharmacodynamic Analysis
Exploratory pharmacodynamic/pharmacodynamic relationships and analyses will be conducted if deemed appropriate.

Determination of Sample Size/Power/Level of Significance
Reasonable estimates for treatment effects for infliximab treatment are were enrolled at a large number of clinical sites, where it is difficult to standardize clinical assessment. We believe in a smaller study we should be able to standardize examination practices to achieve slightly higher treatment effect sizes. Using these references as a guide we choose to power this study based on an estimate treatment effect size of 0.7 to 0.8.
POWER: It is assumed that infliximab effect sizes for K trans will be at least as large as those for DAS28. The sample size of 26 patients per group has 80% power to yield a statistically significant (alpha=0.05, 1-tailed) difference between treatments if the true underlying effect size is 0.7.

Interim Analysis
No formal interim analysis is planned.

Adverse Events
All AEs noted during the study will be listed. The number of subjects reporting each AE (by treatment group) and severity will also be presented.
Treatment-emergent and treatment-related AEs will be tabulated by body system/organ class.

Clinical Laboratory Tests
The results of hematology and blood chemistry will be listed for each subject.
Laboratory values outside the normal ranges will be flagged.

Vital Signs
Systolic and diastolic blood pressures, heart rate, and body temperature will be listed for each subject.

Physical Examination
The results of the physical examinations at Screening will be listed in the medical history listing. Post-baseline findings of the physical examinations that meet the criteria of an AE (Section 7.7.2.2) will be listed in the relevant AE listings.

Electrocardiogram
The results of the ECG will be listed for each subject. If applicable, summary statistics of ECG results will be provided.

Other Safety
Not applicable.

Other Analyses
Data collected from the Health Assessment Questionnaire and the Physician's Global Assessment Visual Analog Scale will be reviewed. In the event that the IRB/IEC requires changes in the protocol, the sponsor shall be advised and must approve the changes prior to implementation. The investigator shall not modify the trial described in the protocol once finalized and after approval by the IRB/IEC without the prior written approval of sponsor.
In countries where the investigator submits the trial protocol and statement of informed consent to the IRB/IEC, the investigator or qualified designee will forward the approvals to the sponsor.

Subject Information and Consent
The details of the protocol must be provided in written format and discussed with each potential subject, and written informed consent must be obtained for all subjects before any trial-related procedure is performed. In obtaining informed consent, the information must be provided in language and terms understandable to

Subject Identification Card
A Subject Identification Card will be provided to each subject to carry on his or her person (eg, in a wallet) at all times while the subject is participating in the trial, if required in section 2.2. The card is to be shown to caregivers in the event of an emergency.
At a minimum, the card must contain the following information: 1. Protocol number; 2. The subject's protocol identification number;

3.
A statement identifying the card-carrier as a participant in a clinical trial (eg, "This person is participating in a clinical research trial.");

4.
A statement indicating the person might be taking an investigational drug (eg, "This person is taking an experimental drug which could have interactions with other medications, or placebo"); and

5.
Contact information in the event of an emergency or hospitalization. The contact information on the card is to be the investigator or a designated site contact, rather an contact from within the sponsor; The cards may also include other trial-specific information to assist with treatment decisions in the event of an emergency, such as types of concomitant therapies that may, or may not be, permitted as part of emergency treatment. As with any other information provided to subjects, the Subject Identification Card must be approved by the IRB/IEC. Monitors will request that Investigators provide Subject Identification Cards to each subject. Investigators will be asked to request that subjects carry the cards with them while they are participating in the trial.

Registration of the Trial
The trial will be registered by the sponsor on appropriate free public web sites such as www.clinicaltrials.gov, which is a service of the United States National Institutes of Health, and EudraCT, https://eudract.ema.europa.eu/.

9.2
Reporting Trial Data to the Sponsor

Data Collection Forms
The Sponsor will provide the site with data collection forms, be they Case Report Forms (CRF), either in paper format or electronic Case Report Forms (eCRF); diaries; Electronic Data Capture (EDC) screens; or other appropriate data collection forms as the trial requires. The investigator is to provide subject data according to the Sponsor's instructions, in the designated data collection form, compliant with GCP practices. The Sponsor will also provide the site with instructions for assisting other parties -such as a central laboratory -to collect data.
As instructed by the Sponsor, a designated central laboratory may collect data in a database and provide the completed database to sponsor. All data collection forms and the databases from the trial are the exclusive property of sponsor. The investigator will ensure that there are sufficient time, staff, and facilities available for the duration of the trial to conduct and record the trial as described in the protocol and according to all applicable guidances, laws, and regulations.
All data collection forms such as CRFs, diaries; EDC screens; electronic database entries, should be completed as soon as possible after the evaluation has occurred. All dates appearing on the sponsor's subject data collection forms for laboratory tests, cultures, and other data collected, must be the dates on which the specimens were obtained, or the procedures performed.

Preparing Case Report Forms for All Subjects
The Sponsor must not collect subject names, initials, or other personal information that is beyond the scope of the trial from any subject. Subjects are not to be identified by name or initials on the CRF or any trial documents. The only acceptable identification for a subject that may appear on a CRF or trial document is the unique subject identification number. The investigator must maintain contact information for each participant so that all can be quickly contacted by the investigator, if necessary.
All entries into CRFs are the responsibility of the investigator and must be completed by the investigator or a qualified designee. The investigator will attest in writing at the beginning of the trial that his/her electronic signature is the legally binding equivalent of a written signature and will acknowledge by entering his/her electronic signature that he/she has verified the accuracy of the recorded data. 2. the accuracy of the information contained in the publication; and

Preparing Case Report Forms for Subjects Who Fail Screening
3. to ensure that the presentation is fairly balanced and in compliance with FDA regulations.
If the parties disagree concerning the appropriateness of the data analysis and presentation, and/or confidentiality of the sponsor's confidential information, investigator agrees to meet with the sponsor's representatives at the clinical trial site or as otherwise agreed, prior to submission for publication, for the purpose of making good faith efforts to discuss and resolve any such issues or disagreement.

Use of Proprietary or Confidential Information in a Publication
No publication or manuscript shall contain any trade secret information of the sponsor or any proprietary or confidential information of the sponsor and shall be confined to new discoveries and interpretations of scientific fact. If the sponsor believes there is patentable subject matter contained in any publication or manuscript submitted for review, the sponsor shall promptly identify such subject matter to investigator. If sponsor requests and at sponsor's expense, investigator shall use its best efforts to assist sponsor to file a patent application covering such subject matter with the USA Patent and Trademark Office or through the Patent Cooperation Treaty prior to any publication.

Use of Trial Information in a Publication
Investigator is granted the right subject to the provisions of this protocol to use the results of all work provided by investigator under this protocol, including but not limited to, the results of tests and any raw data and statistical data generated for PROTOCOL NO. P08136 PROTOCOL 05 JAN 2011 INITIAL PROTOCOL investigator's own teaching, research, and publication purposes only.
Investigator/Institution agrees, on behalf of itself and its employees, officers, trustees, and agents, not to cause said results to be knowingly used for any commercial purpose whatsoever except as authorized by the sponsor in writing.

Authorship of Publications
Authors of publications must meet the International Committee of Medical Journal Editors (ICMJE) guidelines for authorship and must satisfy the 3 criteria that follow: 1.
Authors must make substantial contributions to the conception and design of the trial, acquisition of data, or analysis of data and interpretation of results; 2. Authors must draft the publication or, during draft review, provide contributions (data analysis, interpretation, or other important intellectual content) leading to significant revision of the manuscript with agreement by the other authors;

3.
Authors must provide written approval of the final draft version of the publication prior to submission.
All contributors who do not meet the 3 criteria for authorship should be listed in an acknowledgments section within the publication, if allowed by the journal, per the ICMJE guidelines for acknowledgment.

Trial Documents and Records Retention
During the trial and after termination of the trial -including after early termination of the trial -the investigator must maintain copies of all documents and records relating to the conduct of the trial. Subject files and other source data must be kept for the maximum period of time permitted by the hospital, institution or private practice, or as specified below. The sponsor must be consulted if the investigator wishes to assign the files to someone else, remove them to another location, or is unable to retain them for the specified period.
The investigator must retain trial records for the amount of time specified by applicable laws and regulations. At a minimum, trial records must be retained for the amount of time specified by ICH Guidelines or the EU Good Clinical Practices Directive, or applicable local laws, whichever is longer: 1.
The ICH Guidelines specify that records must be retained for a minimum of 2 years after a marketing application for the indication is approved (or not approved) or 2 years after notifying the appropriate regulatory agency that an investigation is discontinued.

Sponsor
The sponsor of this trial is indicated in Section 1, Title Page.

Selecting Investigators
Only investigators qualified by training and experience to perform this clinical trial are selected. The sponsor will contact and select all investigators (ie, the legally responsible party[ies] at each trial site), who, in turn, will select their staff.

Financial Disclosure Requirement
In connection with the clinical trial described in the protocol, the investigator certifies that, if asked, t h e investigator will read and answer the Must be willing and capable of completing the necessary reviews and providing approval of the CSR in writing.

Central Organizations
Central organizations may be contracted to perform site qualification, study monitoring, imaging data analyses, and other study related tasks. A central laboratory will be used for the measurement of CRP.

Summary of Procedures for Pharmacogenomics
a. Subjects for Enrollment: All subjects enrolled in the current clinical trials will be considered for enrollment.

b. Consent
Informed consent for biosamples (ie, DNA, RNA, protein, etc) will be obtained during screening for protocol enrollment from all subjects or legal guardians, at an outpatient visit, or during an inpatient stay by the investigator or his or her designate.
Subjects are not required to participate in the pharmacogenomic substudy in order to participate in the main trial.

Scope of Pharmacogenomic Study
The DNA, RNA, serum, and plasma samples collected in the current trial will be used to study various genetic causes for how subjects may respond to a drug. The DNA, RNA, serum, and plasma, samples will be stored to provide a resource for future studies conducted by Schering-Plough focused on the study of genes responsible for how a drug enters and is removed by the body, how a drug works, other pathways a drug may interact with, or other aspects of disease. All samples will be used by Schering-Plough or designees and research will be monitored and reviewed by a committee of our scientists and clinicians.

Techniques to Collect Samples
Blood, plasma, serum samples will generally be obtained for all study participants. Blood samples for both DNA and RNA isolation will usually be obtained at a time when the subject is having blood drawn for other study purposes.

Confidential Subject Information for Pharmacogenomic Analysis
Samples will be collected and sent to the laboratory designated for the trial where they will be processed (ie, DNA or RNA extraction, etc) following the Schering-Plough approved policies and procedures for sample handling and preparation.
When samples are collected for a specific genotype or expression analysis, this analysis will be detailed in the main body of the clinical protocol. These samples will be processed, analyzed, and the remainder of the sample will be destroyed. The results of these analyses will be reported along with the other study results. A separate sample will be obtained from subjects in these protocols for storage in the biorepository for future analyses.
To maintain privacy of information collected from samples obtained for storage and future analysis, Schering-Plough has developed secure policies and procedures to maintain subject privacy. At the clinical site, a unique Code will be placed on the blood sample for transfer to the storage facility. The Code is a random number used only to identify the biosample of each subject. No other personal identifiers will appear on the sample tube. The first Code will be replaced with a Sample Code (eg, Genetic Sample Code for DNA sample, Serum Sample code for serum sample) at the Central Laboratory or at the Schering-Plough designated facility. This sample is now a single coded sample. The Sample Code is stored separately from all previous sample identifiers. A secure code, hereinafter referred to as a "first coding key", will be utilized to match the Sample Code to the original blood code and subject number to allow clinical information collected during the course of the study to be associated with the biosample. This "first coding key" will be transferred by the central laboratory or Schering-Plough designated facility under secure procedures to the Schering-Plough group designated as the entrusted keyholder to maintain confidentiality of the biosamples. The Sample Code will be logged into the primary biorepository database, and in this database this identifier will not have identifying demographic data or identifying clinical information (ie, race, sex, age, diagnosis, lab values) associated with it. The sample will be stored in a designated repository site with secure policies and procedures for sample storage and usage. This storage code will be stored separately from all previous sample identifiers. The second secure key referred to as a "second coding key" file will be transferred by the Schering-Plough designated facility under secure procedures to the Schering-Plough entrusted keyholder. Samples with the second code are sometimes referred to as de-identified samples. The use of the second code provides additional confidentiality and privacy protection for subjects over the use of a single code. Access to both coding keys is needed to link any data or samples back to a subject identifier.
The "keys" could be utilized to reconstruct t h e link between genetic information and identifiable clinical information, at the time of analysis. This linkage would not be possible for the investigator conducting the analysis, but may only be done by the Schering-Plough entrusted keyholder under strict security policies and procedures. The Schering-Plough entrusted keyholder will link the information, conduct the analysis, then issue an anonymized data summary on the initially single or double coded samples to the investigator conducting the genomic analysis. The only circumstance by which genomic information would be linked to clinical information would be those situations mandated by health authorities (eg, EMEA, FDA), whereby this information would be directly transferred to the health authority. Once the link between subject's identifiers and the unique codes is deleted, it is no longer possible to trace the data and samples back to individual subjects through the coding keys. Anonymization is intended to prevent subject re-identification.

Biorepository Sample Usage
Samples obtained for the Schering-Plough Biorepository will be used for analyses using good scientific practices. Exploratory analyses will not be conducted under highly validated conditions. The scope of research performed on these samples is limited to the investigation of the variability in inherited biomarkers that may correlate with a clinical phenotype in subjects.
Genomic analysis utilizing the DNA,RNA, plasma, serum samples may be performed by the sponsor, or an additional third party (eg, a university investigator) designated by the sponsor. The investigator conducting the analysis will be provided with a double (single) coded sample. Reassociation of analysis results with corresponding clinical data will only be conducted by the Schering-Plough entrusted keyholder. Any contracted third party genomic analysis will conform to the specific genomic analysis outlined in the clinical protocol. DNA, RNA, plasma, serum sample remaining with the third party vendor after genomic analysis will be returned to the sponsor or destroyed and documentation of destruction will be reported to Schering-Plough. PROTOCOL NO. P08136 PROTOCOL 05 JAN 2011 INITIAL PROTOCOL Consent form signed by the subject will be kept under secure storage for regulatory reasons. Information contained on the consent form alone cannot be traced to any samples, test results, or medical information once the specimens have been rendered de-identified. Laboratory personnel performing the genomic testing will not have access to the informed consent document, nor will they be able to identify subjects from the double (single) coded specimens. Specimens will be identified to the laboratory only by the Sample double (single) code. Subjects who decline to sign the informed consent document for the sub-study will not have the sample collected or stored, nor will they be discontinued from the main study unless the pharmacogenomics sample is specifically required for study enrollment.
A template of each site's informed consent will be stored in the Sponsor's clinical document repository. Each consent will be assessed for appropriate sample permissions. The tracking number on this document will be used to assign sample permissions for each sample in the entrusted keyholder's Sample Database.

Withdrawal From the Biorepository and Pharmacogenomic Database
Subjects may withdraw their consent to store the blood, plasma, and serum sample or the DNA or RNA derived from it. Subjects can also request that their sample be destroyed at any time. If samples can be identified in any way (ie, are not anonymized samples), subjects may withdraw consent for banking samples at any time by contacting the investigator responsible for administering their initial informed consent. At that time, subject samples will be removed from the biorepository. Any DNA, RNA, or other biologic samples will be destroyed, destruction will be documented, and sample database information deleted. However, any analyses performed or data obtained from the samples prior to the subject withdrawing consent will not be deleted.

Retention of Data and Biosamples
It is anticipated that data generated from processed samples collected during the course this study will be retained for an indefinite period. DNA specimens will be maintained for potential analysis for 20 years from the acquisition. Samples will be destroyed according to Schering-Plough policies and procedures and this destruction will be documented in t h e repository database. PROTOCOL NO. P08136 PROTOCOL 05 JAN 2011 INITIAL PROTOCOL

Data Security
Pharmacogenomic and other research databases are accessible only to authorized sponsor and study administrator research personnel and/or designated collaborators and are only stored and accessible as anonymized data. Database user authentication is highly secure, and is accomplished using network security policies and practices based in international standards (eg, ISO17799) to protect against unauthorized access. The Schering Corporation entrusted key holder maintains control over access to all sample data. These data are collected for pharmacogenomic research purposes only as specified in the clinical protocol and will not be used for any other purpose without explicit consent from the research subject.

Reporting of Data to Subjects
There is no definitive requirement in either authoritative ethical guidelines or in relevant laws/regulations globally that research results have to be, in all circumstances, returned to study participant. Some guidelines advocate a proactive return of data in certain instances.
No information obtained from exploratory laboratory studies will be reported to the subject or family, and this information will not be entered into the clinical database maintained by Schering-Plough on subjects. Principle reasons not to inform or return results to the subject include: lack of relevance of data, limitations of predictive capability of research data, concerns of misinterpretation of data, absence of good clinical practices standards in exploratory research.
If any exploratory results are definitively associated with clinical significance for subjects while t h e Schering-Plough clinical trial is still ongoing, investigators will be contacted with information as to how to offer genomic testing (paid for by Schering-Plough) to subjects enrolled and will be advised that genomic counseling should be made available for all who choose to participate.
If any exploratory results are definitively associated with clinical significance after completion of a clinical trial, Schering-Plough will publish the results without revealing specific subject information, inform all sites who participated in the Schering-Plough clinical trial, and post the anonymized results on our website or other accredited website(s) that allow for public access (eg, Disease-societies who have primary interest in the results) in order that physicians and patients may pursue genomic testing if they wish to do so. PROTOCOL NO. P08136 PROTOCOL 05 JAN 2011 INITIAL PROTOCOL

Gender, Ethnicity, and Minorities
Although many diagnoses differ in terms of frequency by ethnic population and gender, every effort will be made to recruit all subjects diagnosed and treated on Schering-Plough clinical trials for pharmacogenomic sampling.
When studies with samples are conducted and subjects identified to serve as controls, every effort will be made to group samples from subjects and controls to represent the ethnic and gender population representative of the disease under current investigation.

Risks Versus Benefits of Pharmacogenomic Testing
For pharmacogenomic testing, risks to the subject have been minimized. Risks include those associated with venipuncture to obtain the whole blood sample. This sample will be obtained at the time of routine blood samples drawn for clinical reasons.
Data privacy concerns of the subject have been strictly protected against with Schering-Plough security, policies and procedures. Data privacy risks are largely limited to rare situations involving possible breach of confidentiality. In this highly unlikely situation there is risk that the information, like all medical information, may be misused.
It is necessary for subject-related data (ie, ethnicity, diagnosis, drug therapy and dosage, age, toxicities, etc) to be reassociated to double (single) coded samples at the time of data analysis. These subject data will be kept in a separate, secure Schering-Plough database, and all samples will be stripped of subject identifiers. No information concerning results obtained from genotyping or biomarker studies conducted with samples from the biorepository will be entered into clinical records, nor will it be released to outside persons or agencies, in any way that could be tied to an individual subject.

Self-Reported Ethnicity
Subjects who participate in pharmacogenomic study will be asked to provide self-reported ethnicity. Subjects who do not wish to provide this data may still participate in the pharmacogenomic study.

Questions
Any questions related to the genomic informed consent, genomic sampling, genomic sample handling, or genomic sample storage should be e-mailed directly to PPD PROTOCOL NO. The area of induration (palpable raised hardened area) around the site of injection is the reaction to tuberculin. For standardization, the diameter of the induration should be measured transversely (perpendicular) to the long axis of the forearm. Erythema (redness) should not be measured. All reactions should be recorded in millimeters, even those classified as negative.

Interpreting the Tuberculin Skin Test Results
In the US and many other countries, the most conservative definition of positivity for the tuberculin skin test is reserved for immunocompromised patients, and this definition is to be applied here, even though the subjects entering this study may or may not be immunocompromised at Baseline. The purpose of using this conservative definition of positivity is to maximize the likelihood of detecting latent TB.
In the US and Canada, an induration of 5 mm or greater in response to the intradermal tuberculin skin test is considered to be a positive result and evidence for either latent or active TB. Although the performance of the new IFN--based blood tests for active or latent M.
tuberculosis infection have not been well-validated in t h e immunosuppressed population, experts believe these new tests will be at least as sensitive, if not more, and definitely more specific than the tuberculin skin test (Barnes, 2004).

Performing the QuantiFERON-TB Gold Test
The QuantiFERON-TB Gold test is available in 2 formats, a standard method and an In-Tube method. The 2 formats differ logistically in the manner in which they are performed. In addition, the In-Tube format contains 1 additional M. tuberculosisspecific antigen, TB7.7(p4), which is thought to increase the specificity of the test.
Results of the QuantiFERON-TB Gold test using either format are acceptable for subjects to enter this trial; however, only the In-Tube format will be provided for this study.
Published which this trial will be conducted. The QuantiFERON TB Gold test with the standard format was recently approved by the FDA for use in the US. The In-Tube format will be submitted to the FDA for review in the next few years. To perform the test using the In-Tube format, blood is drawn through standard venipuncture into supplied tubes that already contain the M. tuberculosis-specific antigens. Approximately 3 tubes will be needed per patient, each requiring 1 mL of blood. One tube contains the M.Tuberculosis-specific antigens, while the remaining tubes contain positive and negative control reagents. Thorough mixing of the blood with the antigens is done by shaking the tubes. The blood is then incubated for 16 to 24 hours at 37°C, after which tubes are centrifuged for approximately 5 to 10 minutes at 1500 to 2200 g.
Following centrifugation, plasma may be harvested from each tube and placed in aliquot tubes, or the centrifuged tubes themselves may be shipped to the central laboratory at 2°C to 8°C. The central laboratory will perform an enzyme-linked immunosorbent assay (ELISA) to quantify the amount of IFN-present in the plasma using spectrophotometry and computer software analysis.

Interpreting the QuantiFERON-TB Gold Test Results
The central laboratory will analyze and report results for each subject, and sites will be informed of the results prior to randomization. Subjects who have a positive

ACR Classification Criteria for Diagnosis of Rheumatoid Arthritis
Using history, physical examination, laboratory and radiographic findings, 4 of the following must be present with 1 through 4 present a minimum of 6 weeks: ! Morning Stiffness 1 hour ! Arthritis of 3 or more of the following joints: right or left PIP, MCP, wrist, elbow, knee, ankle, and MTP joints ! Arthritis of wrist, MCP, or PIP joint ! Symmetric involvement of joints ! Rheumatoid nodules over bony prominences, or extensor surfaces, or in juxta-articular regions ! Positive serum rheumatoid factor ! Radiographic changes including erosions or bony decalcifications localized in or adjacent to the involved joints.