ARTHRITIS ADVISORY COMMITTEE MEETING

Transcription

1 ARTHRITIS ADVISORY COMMITTEE MEETING July 23, 2013 sbla /323: adalimumab for the treatment of Active non-radiographic axial spondyloarthritis in adults with objective signs of inflammation by elevated c-reactive protein (CRP) or magnetic resonance imaging (MRI), who have had an inadequate response to, or are intolerant to, a nonsteroidal anti-inflammatory drug

2 Disclaimer Statement The attached package contains background information prepared by the Food and Drug Administration (FDA) for the panel members of the advisory committee. The FDA background package often contains assessments and/or conclusions and recommendations written by individual FDA reviewers. Such conclusions and recommendations do not necessarily represent the final position of the individual reviewers, nor do they necessarily represent the final position of the Review Division or Office. The background package may not include all issues relevant to the final regulatory recommendation and instead is intended to focus on issues identified by the Agency for discussion by the advisory committee. The FDA will not issue a final determination on the issues at hand until input from the advisory committee process has been considered and all reviews have been finalized. The final determination may be affected by issues not discussed at the advisory committee meeting.

3 FDA Briefing Package Table of Contents I. Division Memorandum II. Clinical Briefing Document III. Statistical Briefing Document Page 3 of 14

4 Division Memorandum Date: June 24, 2013 From: To: Subject: Sarah Yim, MD Associate Director, Division of Pulmonary, Allergy, and Rheumatology Products, CDER, FDA Members, Arthritis Advisory Committee Overview of the FDA background materials for supplemental biologic license application (sbla) /323 Humira (adalimumab) for active non-radiographic axial spondyloarthritis in adults with objective signs of inflammation by elevated c-reactive protein (CRP) or magnetic resonance imaging (MRI), who have had an inadequate response to, or are intolerant to, a nonsteroidal anti-inflammatory drug Introduction Thank you for your participation in the Arthritis Advisory Committee (AAC) meeting to be held on July 23, As members of the AAC, you provide important expert scientific advice and recommendations to the US Food and Drug Administration (the Agency) on the regulatory decision-making process related to the approval of a drug or biologic product for marketing in the United States. The upcoming meeting is to discuss supplemental biologic license application (sbla) /323 Humira (adalimumab) for active non-radiographic axial spondyloarthritis in adults with objective signs of inflammation by elevated c-reactive protein (CRP) or magnetic resonance imaging (MRI), who have had an inadequate response to, or are intolerant to, a nonsteroidal antiinflammatory drug. The content of this document and the materials prepared by the Agency reflect the preliminary findings and opinions based on reviews of the information submitted by the applicant, AbbVie, Inc. These materials do not represent the final position of the Agency. The opinions and insights provided by you at this AAC meeting will be an important factor in our decision on this application. The clinical and statistical issues related to the adalimumab clinical trial results in nonradiographic axial spondyloarthritis (nr-axspa) are the primary focus of this AAC meeting. In determining approvability of additional indications for a product, there may be factors, other than clinical data, that the Agency may take into consideration in the regulatory decision-making process. These additional factors will not be the focus of this AAC meeting. Attached are the background materials for this meeting. In addition to this memorandum, the FDA background materials include the clinical and statistical briefing documents. Page 4 of 14

5 FDA Division Summary Memo s, Inc. Humira (adalimumab) for non-radiographic axial spondyloarthritis (nr-axspa) Background The spondyloarthritides (also known as spondyloarthropathies ) are a group of inflammatory rheumatic disorders distinguished from rheumatoid arthritis (RA) by absence of rheumatoid factor, axial skeleton involvement, enthesitis, dactylitis, and extraarticular features such as uveitis and skin rash. Also, in contrast to RA, peripheral arthritis, when present, tends to be asymmetrical. The spectrum of diagnoses includes ankylosing spondylitis (AS) the prototypical axial spondyloarthritis along with psoriatic arthritis, inflammatory bowel disease-related arthritis, reactive arthritis, and other undifferentiated spondyloarthritides. Since 2003, four Tumor Necrosis Factor (TNF) inhibitors have been approved for the treatment of ankylosing spondylitis: Enbrel (etanercept), Remicade (infliximab), Humira (adalimumab), and Simponi (golimumab). TNF inhibitor treatment has demonstrated efficacy for multiple aspects of clinical disease activity in AS, but it not yet known whether treatment has a beneficial effect on structural damage progression. The patient population studied in the respective clinical development programs for the AS indication were patients with established AS, as defined by the modified New York Criteria, which includes a radiologic criterion of sacroiliitis grade >2 bilaterally or grade 3-4 unilaterally. This is a population that is well characterized, with a chronic and progressive form of disease. However, it may take years from the onset of inflammatory back pain symptoms until the appearance of radiographic sacroiliitis, and patients disease activity and severity is not contingent on the presence or absence of radiographic sacroiliitis. Thus the Assessment of Spondyloarthritis international Society (ASAS) developed criteria for the classification of axial spondyloarthritis with the intent of defining a broader population of patients with inflammatory back pain but who may not have radiographic changes, to allow for earlier diagnosis and treatment. The ASAS criteria require patients to have back pain for at least three months and age of onset less than 45 years. In addition, patients should have either sacroiliitis on imaging (MRI or x-ray) along with at least 1 SpA feature or be HLA-B27 positive and have at least 2 other SpA features. SpA features include inflammatory back pain, arthritis, enthesitis (heel), uveitis, dactylitis, psoriasis, Crohn s disease or ulcerative colitis, good response to NSAIDs, family history of SpA, HLA-B27 positive, and elevated C-reactive protein (CRP). Adalimumab has been approved for AS since August In sbla /323, AbbVie is seeking to expand the indication to include non-radiographic axial spondyloarthritis (nr-axspa), on the basis of study M10-791, a 12-week double-blind placebo-controlled study with a 144-week open-label treatment period in 192 patients who met the ASAS classification criteria but did not fulfill modified New York Criteria for AS, and who had inadequate response or intolerance to a non-steroidal antiinflammatory drug (NSAID). Notably, the population described by AbbVie s proposed indication is a subgroup of the total nr-axspa patients evaluated in the program; specifically those who had objective signs of inflammation by elevated CRP or MRI (identified as the Adalimumab Target Population, ATP). This subgroup in turn Page 5 of 14

6 FDA Division Summary Memo s, Inc. Humira (adalimumab) for non-radiographic axial spondyloarthritis (nr-axspa) required additional subgrouping, as differences in the interpretation of screening anteroposterior (AP) pelvis x-rays used to exclude patients meeting radiographic sacroiliitis by the modified New York Criteria resulted in 38 of 102 patients being reclassified as fulfilling the modified New York Criteria for AS, when x-rays were reread centrally (only patients with x-rays available at screening and week 104 were reread). Of the remaining 64 nr-axspa patients, 45 had evidence of inflammation on MRI or had elevated CRP. FDA is seeking the Committee s input on whether these data are adequate to determine the risk-benefit profile of adalimumab-treatment in the population sought, and whether that risk-benefit profile is favorable to support approval of adalimumab for the proposed expanded indication. Relevant Regulatory History An End-of-Phase 2 (EOP2) meeting was held with adalimumab s sponsor in August 2008 regarding a broader spondyloarthropathy indication. At that time, the ASAS classification criteria had not yet been published, but FDA was open to the use of updated classification criteria, provided that the criteria would have been validated to identify a discrete population of patients. If the population selected was adequate, the sponsor s plan to perform a 12-week double-blind, placebo-controlled trial followed by a one-year open-label extension was deemed reasonable at that time. A pre-sbla meeting was held in June At this meeting, FDA expressed concerns regarding a lack of clarity in the axial spondyloarthritis indication, what entities would be encompassed, and whether patients could evolve into diagnoses for which the risk-benefit profile of treatment would be unfavorable. Further discussion occurred in a Type C meeting held in July The sponsor provided additional information regarding available data on axial spondyloarthritis and nr-axspa prevalence and burden of disease. FDA reiterated previously expressed concerns and indicated that discussion of the proposed expanded indication in the setting of an advisory committee would likely be needed. Product Information Adalimumab is a recombinant human IgG1 monoclonal antibody specific for human tumor necrosis factor (TNF)-alpha. It is marketed as sterile, preservative-free solution for subcutaneous administration, available as a 40 mg/0.8 ml single-use prefilled pen, 40 mg/0.8 ml single-use prefilled glass syringe, 20 mg/0.4 ml single-use prefilled glass syringe, and 40 mg/0.8 ml single-use glass vial for institutional use only. Humira (adalimumab) was first approved on December 31, 2002 for the treatment of rheumatoid arthritis. Subsequently, it has been approved for psoriatic arthritis (October 2005), ankylosing spondylitis (August 2006), Crohn s Disease (February 2007), plaque psoriasis (January 2008), polyarticular juvenile idiopathic arthritis (February 2008), and Page 6 of 14

7 FDA Division Summary Memo s, Inc. Humira (adalimumab) for non-radiographic axial spondyloarthritis (nr-axspa) ulcerative colitis (September 2012). The dose for rheumatoid arthritis, psoriatic arthritis, and ankylosing spondylitis is 40 mg subcutaneously every other week. The proposed dose for nr-axspa is also 40 mg SC every other week. Clinical and Statistical Overview of the clinical program As previously noted, the data for the proposed nr-axspa indication are derived from study M This study enrolled 192 patients and had a 12-week double-blind, placebo-controlled period where patients were randomized 1:1 to receive either adalimumab 40 mg SC every other week or a matching placebo. After the double-blind treatment period, all patients received open-label treatment with adalimumab 40 mg SC every other week for up to 144 weeks. The applicant submitted data through week 68. Patients could continue on stable doses of methotrexate (MTX), sulfasalazine (SSZ), hydroxychloroquine (HCQ), azathioprine, prednisone, and/or NSAIDS during the study. Patients were required to have had an inadequate response or intolerance to at least 1 NSAID and were required to meet ASAS axspa classification criteria. However, patients fulfilling a diagnosis of AS, as defined by the modified New York criteria, were to be excluded. AP pelvis x-rays were obtained within 180 days prior to screening and were initially read locally for purposes of patient enrollment. At participating sites, pelvis x-rays were repeated at Week 104 to evaluate for the presence of sacroiliitis that fulfilled the radiographic criterion of the modified New York Criteria for AS. Screening and Week 104 images were sent to a central imaging laboratory for interpretation. All patients were to have MRI of the cervical, thoracic, and lumbar spine as well as SI joints using the Spondyloarthritis Research Consortium of Canada (SPARCC) method. These were performed at screening through baseline, Week 12, Week 52, and Week 104. Images were sent to the central imaging laboratory for interpretation. Brief Description of Efficacy Endpoints The primary efficacy endpoint for Study M was the proportion of patients having an Assessment in Ankylosing Spondylitis (ASAS) 40 response at Week 12. ASAS40 response is defined as an improvement of at least 40% and an absolute improvement of at least 20 units (on a scale of 0 to 100) in at least 3 of 4 domains, and no worsening in the remaining domain: patient global assessment (visual analog scale (VAS), 0 to 100) total back pain (VAS 0 to 100) function as assessed by the Bath AS Functional Index (BASFI) inflammation, as assessed using the last two stiffness assessments in the Bath AS Disease Activity Index (BASDAI) Page 7 of 14

8 FDA Division Summary Memo s, Inc. Humira (adalimumab) for non-radiographic axial spondyloarthritis (nr-axspa) The BASFI is a functional index based on the patient s assessment of his/her ability to perform 10 selected activities during the past week on a VAS from 0 to 100, with 0 being easy and 100 being impossible. The overall BASFI score is the average of the 10 items. A numeric rating scale from 0 to 10 may be used instead, but the 0 to 100 scale was used for Study M The BASDAI is a 6-item disease activity index that includes patient ratings of fatigue, spinal pain, peripheral arthritis, enthesitis, intensity of morning stiffness, and duration of morning stiffness. Five of these items are rated from 0 to 10, with 0 being no activity and 10 being very severe. The duration scale is rated from 0 to 10, with 0 being 0 hours and 10 being 2 or more hours. The overall BASDAI score is calculated as the sum of the values of questions 1 to 4, added to the average of questions 5 and 6 (intensity and duration of morning stiffness), and the total then divided by 5. Multiple other endpoints were pre-specified in the study and are described in detail in the clinical and statistical briefing documents. Study Population Study M enrolled adult patients meeting ASAS axial SpA classification criteria who have had an inadequate response or are intolerant to at least 1 NSAID, but excluded the subset of patients who fulfilled modified New York criteria for definite AS. Otherwise, the only baseline disease activity requirement was a total back pain VAS score of at least 40 mm (on a 0 to 100 scale) and a BASDAI score of at least 4 at both the screening and baseline visits. However, subsequently, the applicant has proposed narrowing the indicated population of nr-axspa patients to the subgroup of patients with objective signs of inflammation by CRP or MRI (annotated as the adalimumab target population or ATP). Inflammation for the ATP was defined as a baseline SPARCC MRI score of at least 2 for either the sacroiliac joints or spine, or an elevated baseline CRP. This subgroup comprised 73 of the 94 (78%) placebo group patients and 69 of the 91 (76%) adalimumab group patients in the full analysis set (FAS). The demographic characteristics and baseline disease characteristics of the ATP were similar to the FAS. However, the demographic characteristics of the ATP and FAS were different than the demographic characteristics expected in AS populations. In contrast to AS, the study population in M included more females (~55%, compared to ~45% males). With respect to the defining characteristics of the ATP, CRP values and MRI scores were only slightly higher in the ATP, as shown in Table 1, below. Page 8 of 14

9 FDA Division Summary Memo s, Inc. Humira (adalimumab) for non-radiographic axial spondyloarthritis (nr-axspa) Table 1: Baseline CRP and MRI in FAS and ATP FAS ATP FAS Placebo Adalimumab Placebo Adalimumab N Hs-CRP (mg/l) Median (range) 3 (0-43) 2 (0-77) 5 (0-43) 3 (0-77) Mean (SD) 7 (10) 7 (13) 9 (11) 9 (14) SPARCC MRI scores, mean (SD) Spine (0 to 108 range) 5 (6) 4 (5) 6 (7) 5 (6) Sacroiliac joints (0 to 72 range) 5 (10) 5 (10) 6 (11) 5 (6) Source: Excerpted from Table 13 of the FDA Clinical Briefing Document Adalimumab is already approved for AS, so patients fulfilling modified New York Criteria were intended to be excluded. Given the challenges of reading plain films for sacroiliitis, there was some discrepancy between local and central interpretation. While the local interpretation would likely be consistent with the scenario in clinical practice, the presence of a subgroup of AS patients in the study raises the question of whether the apparent treatment effect is being driven by this subgroup. Figure 1 below illustrates the various patient subgroups by analysis population and centrally-read x-ray status. Figure 1: Patient Subgroups by Analysis Population and Centrally-Read X-Ray Status Randomized, n=192 FAS, n=185 (ADA 91, PBO 94) ATP with no centralread x-rays; n=61 (ADA 30, PBO 31) ATP, n=142 (ADA 69, PBO 73) ATP with central-read AS n=36 (ADA 14, PBO 22) ATP with central-read nr-axspa n=45 (ADA 25, PBO 20) Central-read AS, n=38 (ADA 15, PBO 23) Central-read nr-axspa, Total n=64 (ADA 33, PBO 31) FAS = Full Analysis Set; ATP = Adalimumab Target Population; AS = ankylosing spondylitis; nr-axspa = non-radiographic axial spondyloarthritis; ADA = adalimumab; PBO = placebo Page 9 of 14

10 FDA Division Summary Memo s, Inc. Humira (adalimumab) for non-radiographic axial spondyloarthritis (nr-axspa) The largest circle represents the randomized patient population (n=192). The next circle represents the FAS population (n=185). The ATP is the inner green circle (n=142). Patients with centrally read x-rays consistent with AS are represented by the smaller, upper rectangle (n=38). These patients were intended to be excluded from the study, as they fulfilled modified New York Criteria for AS. Thirty-six of 38 patients were also in the ATP. The larger lower rectangle represents patients with centrally read x-rays that were negative (n=64), and thus is consistent with the nr-axspa population as defined by the applicant for this study. Forty-five of 64 patients were also in the ATP. This small subgroup represents the patients who are definitely included by the applicant s proposed indication. An additional 61 patients in the ATP did not have x-rays that were centrally read, but may also be included by the applicant s proposed indication. Although one cannot draw definitive conclusions from post-hoc subgroup analyses, this is a proposed indication and application where subgroups take center stage nonetheless. Efficacy findings The primary endpoint for study M was the proportion of ASAS40 responders at Week 12. Table 2 below summarizes the results in the overall population (FAS) and results in the subgroup of patients proposed for the indication (ATP), who were defined by having objective evidence of inflammation on baseline MRI or elevated CRP. Results for the FAS and ATP are consistent in showing a higher proportion of ASAS40 responders with adalimumab treatment. However the treatment difference was slightly higher in the ATP group due to a higher proportion of adalimumab group responders. Differences in both the FAS and ATP were statistically significant. Results for the ranked secondary endpoints showed a similar pattern and are discussed in detail in the FDA clinical and statistical briefing documents. Table 2: Primary endpoint ASAS40 response at Week 12 in overall nonradiographic axial SpA group (FAS, pre-specified analysis) and adalimumab target population (ATP, post-hoc analysis) FAS ATP FAS ASAS40 PBO N=94 n (%) ADA N=91 n (%) Diff % p-value PBO N=73 n (%) ADA N=69 n (%) Diff % p-value Responders 14 (15) 33 (36) 21 < (14) 28 (41) 27 <0.001 FAS=Full Analysis Set; ATP=Adalimumab Target Population; PBO=placebo; ADA=adalimumab; Diff=difference between adalimumab and placebo Source: Table 20 from FDA Clinical Briefing Document To explore the potential impact of including AS patients in the study, FDA reviewers evaluated treatment effect by centrally-read x-ray status, keeping in mind the limitations of this analysis and the fact that not all x-rays were re-read. Table 3 below summarizes the results of this analysis for the primary endpoint of ASAS40 responders. Notably, in the subgroup of patients who fulfilled the modified New York Criteria for AS, the effect of adalimumab treatment was the largest. In the subgroup of patients with confirmed nrax-spa by centrally-read x-rays, the proportion of ASAS40 responders was similar Page 10 of 14

11 FDA Division Summary Memo s, Inc. Humira (adalimumab) for non-radiographic axial spondyloarthritis (nr-axspa) ASAS40 between placebo and adalimumab-treated patients. In the subgroup of patients whose x- rays were not centrally-read, very few placebo-treated patients had an ASAS40 response, and the proportion of adalimumab-treated patients experiencing a response was similar to the confirmed nr-axspa subgroup. Combining this group with the confirmed nr-axspa subgroup would primarily reduce the placebo response rate and increase the treatment difference to approximately 15%. A similar pattern was observed for the ranked secondary endpoints, which are discussed in detail in the FDA clinical and statistical briefing documents. Thus, although we cannot draw definitive conclusions, having AS patients in the study does appear to increase the apparent treatment effect of adalimumab in the overall study population. Table 3: Primary endpoint results at Week 12 in study M10-791, by centrally-read x-ray status (FAS) Centrally-read x-rays Nr-axSpA AS X-rays not read PBO N=31 % ADA N=33 % Diff % 95% CI PBO N=23 % ADA N=15 % Diff % 95% CI PBO N=40 % ADA N=43 % Diff % 95% CI Responders (-18,27) (29,83) (8,38) FAS=Full Analysis Set; PBO=placebo; ADA=adalimumab; Diff=difference between adalimumab and placebo; Non-responder imputation was used for missing data Source: Excerpted from Table 9 of the FDA Statistical Briefing Document As the ATP is the population proposed for the indication, similar subgroup analyses were done for the ATP patients by centrally-read x-ray status, and results are summarized in Table 4 below. Compared to the overall population, ATP patients experienced slightly higher response rates with adalimumab treatment, although the patterns of response to placebo vs. adalimumab are similar in the x-ray subgroups. Also consistent with the overall findings, a markedly higher proportion of ATP patients fulfilling modified New York Criteria for AS responded to adalimumab treatment compared to placebo. Notably, in the confirmed nr-axspa subgroup, the use of the ATP-defining criteria for inflammation appeared to be associated with only a modest numerical increase in the treatment difference between adalimumab and placebo. Thus it is not clear from these data whether using the ATP criteria would enhance the risk-benefit profile of treatment in the nr-axspa population. Table 4: Primary endpoint results at Week 12 in study M10-791: ATP Population by centrally-read x-ray status ATP Centrally-read negative (nr-axspa subgroup) ATP Centrally-read positive (AS subgroup) ATP x-rays not centrally-read ASAS40 PBO ADA Diff 95% CI PBO ADA Diff 95% CI PBO ADA Diff 95% CI N=20 % N=36 % % N=22 % N=14 % % N=31 % N=30 % % Responders (-16,38) (39,91) (2,38) PBO=placebo; ADA=adalimumab; Diff=difference Source: Tables 3 and 10 from the FDA Statistical Briefing Document Page 11 of 14

12 FDA Division Summary Memo s, Inc. Humira (adalimumab) for non-radiographic axial spondyloarthritis (nr-axspa) In summary, on the basis of post-hoc subgroup analyses, the efficacy and risk-benefit profile of adalimumab for the treatment of nr-axspa patients overall and of nr-axspa patients with objective signs of inflammation, as defined for the ATP in this study, is not well characterized. At best, these data would be considered suggestive and hypothesisgenerating. Typically, such findings would be confirmed in the setting of at least one additional controlled trial where the population is prospectively defined. Safety findings Adalimumab has a Boxed Warning for serious infections leading to hospitalization or death, including tuberculosis (TB), bacterial sepsis, invasive fungal infections, and infections due to opportunistic pathogens. The Boxed Warning also includes malignancy; especially lymphoma and hepatosplenic T-cell lymphoma. Adalimumab has Warnings and Precautions related to serious infections, malignancy, hypersensitivity reactions, Hepatitis B Virus reactivation, central and peripheral demyelinating disease, cytopenias, worsening of congestive heart failure, lupus-like syndrome, and against use with anakinra or abatacept. The safety data in the sbla are derived from study M through week 68 of the open-label extension. In study M10-791, 190 patients received at least one dose of adalimumab during the study. The mean duration of exposure to adalimumab in the study was approximately 372 days (range 14 to 488 days; median duration 393 days). Thus the safety database for the nr-axspa population was limited, and the bulk of safety information related to adalimumab treatment is from data in other indications (including AS) and the post-marketing experience since its approval in In study M10-791, there were no deaths during the 12-week controlled period and two deaths during the open-label extension through Week 68. The first patient committed suicide on Day 223 (40 days after the last dose of adalimumab), and the second patient was found unresponsive on Day 649 (later determined to be death related to opiate toxicity). FDA reviewers considered neither case to be likely related to adalimumab. There were few serious adverse events (SAEs) observed during the controlled period of study M Three of 95 adalimumab-treated patients experienced an SAE compared to 1/97 placebo-treated patients. The three adalimumab-treated patients experienced an SAE of acute hepatitis, breast dysplasia, and induced abortion, respectively. No serious infections were reported during the 12-week controlled period and three serious infections were reported during the open-label period, including one case of disseminated tuberculosis. No malignancies were reported in the study. Overall, the safety profile of adalimumab in the data submitted appeared to be consistent with the known safety profile of adalimumab, however the data were limited. Details of the safety evaluation are available in the FDA clinical briefing document. Long-term data were not provided that would enable a determination of the safety profile of adalimumab over time in the different axial spondyloarthritis entities captured by the nraxspa classification. Page 12 of 14

13 FDA Division Summary Memo s, Inc. Humira (adalimumab) for non-radiographic axial spondyloarthritis (nr-axspa) Benefit-risk assessment Based on the information provided, the Advisory Committee will be asked to consider whether the benefit-risk profile of adalimumab is adequate for the proposed indication of active non-radiographic axial spondyloarthritis in adults with objective signs of inflammation by elevated CRP or MRI, who have had inadequate response or are intolerant to an NSAID. There are a number of points to consider: Study M was intended to enroll non-radiographic axial spondyloarthritis (nraxspa) patients but in actuality enrolled a subgroup of patients who fulfilled modified New York Criteria for ankylosing spondylitis (AS). A markedly higher proportion of patients in the AS subgroup responded to adalimumab-treatment compared to placebo-treatment and skewed results for the overall group higher. In contrast, the proportion of ASAS40 responders in the nr-axspa subgroup was much lower, with a relatively modest treatment difference between adalimumab and placebo-treated patients. The proportion of ASAS40 responders in the nr-axspa subgroup was slightly higher if the population was selected for objective signs of inflammation as defined for the adalimumab target population (ATP), but the difference is small. Thus it is not clear whether use of objective signs of inflammation in the indication would be warranted by the data. There is still uncertainty regarding what disease entities are being captured by the proposed nr-axspa indication. Study M suggests nr-axspa patients are different than AS patients in terms of demographics and response to treatment. Data have not been provided regarding the natural history of nr-axspa patients that would enable a determination of whether the risk-benefit profile of treatment would be expected to remain the same over time (e.g., chronic immunosuppressive therapy may not be warranted if patients might be expected to remit). Summary The purpose of this AAC meeting is to discuss the adequacy of the data submitted by AbbVie to support approval of adalimumab for the proposed indication of active nonradiographic axial spondyloarthritis in adults with objective signs of inflammation by elevated CRP or MRI, who have had inadequate response or are intolerant to an NSAID. The Committee s input will be invaluable in this determination. In this regard, we ask the Committee to keep in mind the following discussion topics. Draft Topics for Discussion 1. Discuss the efficacy data for adalimumab. a) Discuss the treatment effect of adalimumab for AS patients versus nr-axspa patients. b) Discuss whether the data for the subpopulation with objective signs of inflammation by elevated CRP or MRI supports specifying this population in the proposed indication. Page 13 of 14

14 FDA Division Summary Memo s, Inc. Humira (adalimumab) for non-radiographic axial spondyloarthritis (nr-axspa) 2. Discuss the safety data for adalimumab. a) Discuss whether nr-axspa may include disease entities that are not chronic and progressive. In light of your position, discuss whether the currently available safety data are adequate or whether additional data are needed. If the latter, please explain what additional data should be required. 3. Overall, do the data provide substantial evidence that adalimumab provides a clinically meaningful beneficial effect in the treatment of active non-radiographic axial spondyloarthritis in adults with objective signs of inflammation by elevated CRP or MRI, who have had inadequate response or are intolerant to an NSAID? 4. Is the safety profile of adalimumab adequate to support approval of adalimumab for the treatment of active non-radiographic axial spondyloarthritis in adults with objective signs of inflammation by elevated CRP or MRI, who have had inadequate response or are intolerant to an NSAID? 5. Does the Committee recommend approval of adalimumab for the proposed indication of active non-radiographic axial spondyloarthritis in adults with objective signs of inflammation by elevated CRP or MRI, who have had inadequate response or are intolerant to an NSAID. Page 14 of 14

15 Clinical Review for the Arthritis Advisory Committee Meeting July 23, 2013 Humira (Adalimumab) BLA /323 Dose: 40mg subcutaneous every other week Proposed indication: Active non-radiographic axial spondyloarthritis in adults with objective signs of inflammation by elevated c-reactive protein (CRP) or magnetic resonance imaging (MRI), who have had an inadequate response to, or are intolerant to, a nonsteroidal antiinflammatory drug Department of Health & Human Services Food & Drug Administration Center for Drug Evaluation & Research Division of Pulmonary, Allergy and Rheumatology Products Silver Spring, MD

16 Table of Contents 1 EXECUTIVE SUMMARY Brief Overview of Clinical Program Summary of Efficacy Summary of Safety INTRODUCTION AND REGULATORY BACKGROUND Product Information Tables of Currently Available Treatments for Proposed Indications Availability of Proposed Active Ingredient in the United States Important Safety Issues with Consideration to Related Drugs Summary of Pre-submission Regulatory Activity Related to Submission SOURCES OF CLINICAL DATA Tables of Studies/Clinical Trials Review Strategy Discussion of Individual Studies/Clinical Trials Study M REVIEW OF EFFICACY Efficacy Summary for Non-Radiographic Axial Spondyloarthritis Indication Methods Demographics Subject Disposition Analysis of Primary Endpoint(s) Analysis of Secondary Endpoints(s) Other Endpoints Subpopulations Analysis of Clinical Information Relevant to Dosing Recommendations Discussion of Persistence of Efficacy and/or Tolerance Effects Additional Efficacy Issues/Analyses REVIEW OF SAFETY Safety Summary Methods Studies/Clinical Trials Used to Evaluate Safety Categorization of Adverse Events Pooling of Data Across Studies/Clinical Trials to Estimate and Compare Incidence Adequacy of Safety Assessments

17 5.2.1 Overall Exposure at Appropriate Doses/Durations and Demographics of Target Populations Explorations for Dose Response Routine Clinical Testing Evaluation for Potential Adverse Events for Similar Drugs in Drug Class Major Safety Results Deaths Nonfatal Serious Adverse Events Dropouts and/or Discontinuations Significant Adverse Events Submission Specific Primary Safety Concerns Supportive Safety Results Common Adverse Events Laboratory Findings Vital Signs Electrocardiograms (ECGs) Other Safety Explorations Dose Dependency for Adverse Events Time Dependency for Adverse Events Drug-Demographic Interactions Drug-Disease Interactions POSTMARKET EXPERIENCE APPENDICES Table of Tables Table 1: Modified New York criteria for ankylosing spondylitis... 7 Table 2: Definitions and terms used in this briefing document... 9 Table 3: Approved products for the treatment of AS in the United States Table 4: Approved biologic TNF-inhibitors in the United States as of May 30, Table 5: Overview of safety concerns with TNF-inhibitors Table 6: Overview of regulatory interactions for the non-radiographic axial spondyloarthritis development program Table 7: Overview of the randomized, controlled study in non-radiographic axial spondyloarthritis patients Table 8: Overview of Protocol Amendments to study M Table 9: Overview of monitoring in study M Table 10: The Bath AS Function Index (BASFI), Bath AS Disease Activity Index (BASDAI), the Bath AS Metrology Index (BASMI), and Health Assessment Questionnaire (HAQ) Table 11: BASMI Linear Definitions Table 12: Baseline demographics in study M (FAS and ATP FAS) Table 13: Baseline disease characteristics in study M (FAS and ATP FAS) Table 14: Axial SpA-related medical history in study M (FAS and ATP FAS)

18 Table 15: Proportion of patients who reported prior DMARD use at baseline study M (FAS and ATP FAS) Table 16: Proportion of patients using concomitant DMARDs in study M (FAS and ATP FAS) Table 17: Patient disposition through Week 68 in study M (FAS and ATP FAS) 40 Table 18: Comparison of investigator s and central reader s assessment of pelvis x-rays Table 19: The Bath AS Function Index (BASFI) and Bath AS Disease Activity Index (BASDAI) Table 20: Analysis of primary endpoint ASAS40 response at Week 12 in overall nonradiographic axial SpA group (pre-specified analysis) and adalimumab target population (post-hoc analysis) Table 21: Analysis of ASAS response component endpoints at Week 12 in overall nonradiographic axial SpA group (FAS) Table 22: Ranked dichotomous secondary efficacy results at Week 12 in study M (Statistical reviewer s NRI analyses in FAS and ATP-FAS) Table 23: Ranked continuous secondary efficacy results at Week 12 in study M (Statistical reviewer s BOCF analyses FAS and ATP-FAS) Table 24: Ranked dichotomous primary and secondary efficacy results at Week 12 in study M based on central x-ray interpretation (Statistical reviewer s NRI analyses)...46 Table 25: Ranked continuous secondary efficacy results at Week 12 in study M based on central x-ray interpretation (Statistical reviewer s LOCF analyses FAS and ATP-FAS) Table 26: Primary efficacy results at Week 12 in study M in overall population (FAS) and subgroups of patients with centrally-read negative screening pelvis x-rays (FAS and ATP subgroups) Table 27: Efficacy results at Week 12 in study M in ATP centrally-read negative screening pelvis x-rays Table 28: Ranked dichotomous efficacy results at Week 12 in study M in ATP centrally-read positive (AS) and x-rays not read subgroups Table 29: Efficacy (ASAS40 responders at week 12) of the adalimumab group versus placebo group by demographic and disease characteristics in study M Table 30: Patient disposition through Week 68 in study M (FAS and ATP FAS) 51 Table 31: Extent of exposure to adalimumab up to week 12 (FAS and ATP FAS) Table 32: Extent of exposure to adalimumab up to week 68 (any adalimumab safety set and ATP any adalimumab safety set) Table 33: Comparison of number of patients studied in different clinical development programs for AS and nr-axspa development programs Table 34: Clinical testing to elicit AEs, vital signs, and laboratory parameters in study M Table 35: Serious adverse events by system organ class and MedDRA preferred term in study M through Week 12 (safety analysis set; ATP safety analysis set) Table 36: Discontinuations secondary to adverse events by system organ class and MedDRA preferred term in study M through Week 12 (safety analysis set)

19 Table 37: Adverse events related to infections experienced by >2 patients in any treatment group during the 12 week, double-blind treatment period (safety analysis set; ATP safety analysis set) Table 38: Listing of serious infections by randomized treatment group (safety analysis set) Table 39: Maximum post-baseline ALT and AST measurements during the 12-week double-blind treatment period in study M stratified by whether baseline AST and ALT values were normal or elevated (Safety Analysis Set) Table 40: Injection site reaction-related adverse events during the 12 week, doubleblind treatment period (Safety analysis set; ATP safety analysis set) Table 41: Shifts in hematology parameters from baseline to the low final value of the 12- week, double-blind treatment period (Safety analysis set) Table 42: Adverse events related to infections experienced by 3% of patients in any treatment group during the 12 week, double-blind treatment period (safety analysis set)

20 Table of Figures Figure 1: Study M Schematic Figure 2: Patient flow diagram for centrally-read x-rays

21 1 Executive Summary 1.1 Brief Overview of Clinical Program Background The focus of discussion for the morning of July 23, 2013 Arthritis Advisory Committee (AAC) meeting is adalimumab (Humira ) for the proposed indication of non-radiographic axial spondyloarthritis (nr-axspa). The Sponsor proposes use of adalimumab for the treatment of adult patients with active nr-axspa with objective signs of inflammation by elevated c-reactive protein (CRP) or magnetic resonance imaging (MRI), who have had an inadequate response to, or are intolerant to, a nonsteroidal anti-inflammatory drug (NSAID). In the United States, this product is currently approved for the treatment of rheumatoid arthritis (RA), juvenile idiopathic arthritis (JIA), psoriatic arthritis (PsA), ankylosing spondylitis (AS), Crohn s disease (CD), ulcerative colitis (UC), and plaque psoriasis (Ps). The contents of this document reflect preliminary findings and opinions based on reviews of the information submitted by the Sponsor in the supplemental biologic license application (sbla) /323. This document does not represent the final position of the Agency. The opinions and insights provided at this AAC meeting will be important factors in the Agency s decision on this application. Background on the Proposed Indication: Non-radiographic Axial Spondyloarthritis The Sponsor s proposed indication is non-radiographic axial spondyloarthritis (nraxspa). Nr-axSpA is a subgroup of the broader group of inflammatory diseases frequently referred to as axial spondyloarthritis (axial SpA). Axial SpA encompasses a spectrum of disease severity that spans from self-limited inflammation to bony destruction of the spine. AS is a well-characterized, chronic and progressive form of axial SpA. The majority of research performed over the last two decades has used the modified New York Critieria 1 to identify patients with AS (Table 1). Further, these criteria were used in the clinical trials performed to support product registration in AS. Table 1: Modified New York criteria for ankylosing spondylitis Clinical Criteria Low Back pain and stiffness for longer than 3 months, which improve with exercise, but are not relieved by rest Restriction of motion of the lumbar spine in both the sagittal and frontal planes Restriction of chest expansion relative to normal values correlated for age and sex Radiologic criterion Sacroiliitis grade 2 bilaterally, or grade 3-4 unilaterally Definitive ankylosing spondylitis is present if the radiologic criterion is associated with at least one clinical criterion Source: van der Linden S. Arthritis Rheum 1984;27(4):

22 The modified New York Criteria 1 have potential limitations in clinical practice as they are designed to identify patients with established AS and do not identify all patients in the larger spectrum of inflammatory back pain. Although the long-term benefit of early treatment in AS, with respect to disease progression, remains unclear, the Assessment of Spondyloarthritis international Society (ASAS) developed criteria for axial SpA with the goal of identifying more patients in the spectrum of inflammatory back pain, including patients with early AS 2, 3. By design, these criteria were meant to be inclusive, as not to miss patients with the potential for developing progressive disease. As a result, the ASAS criteria identify a heterogeneous group of patients as described below. The ASAS criteria for axial SpA require patients to have back pain for at least three months and age of onset less than 45 years. Subsequently, patients must meet clinical or imaging criteria (Figure 1). The imaging criteria include evidence of sacroiliitis on magnetic resonance imaging (MRI) or x-ray and at least one additional clinical feature. The clinical criteria include the presence of HLA-B27 and at least two other SpA features. The clinical SpA features include inflammatory back pain, arthritis, enthesitis, uveitis, dactylitis, psoriasis, Crohn s disease or ulcerative colitis, good response to NSAIDs, family history for SpA, HLA-B27, and elevated CRP. As a consequence of being more inclusive, the criteria increase the heterogeneity of the resulting disease group and may reduce the utility of these criteria for drug development. Figure 1: ASAS Classification Criteria for Axial SpA Source: Rudwaleit. M Ann Rheum Dis 2009;68(6):777. We acknowledge that there is debate within the AS research community regarding the nomenclature for the spectrum of inflammatory diseases encompassed by the term 8

23 axial SpA. Some authors have argued that the term AS should be retained because axial SpA cases have far greater clinical heterogeneity than AS and have broader etiologies 5. For the purposes of this briefing document we will use the definitions and terms listed in Table 2. Table 2: Definitions and terms used in this briefing document Term Definition of this term as used in this document Axial spondyloarthritis (axial Patients fulfilling the ASAS criteria 2-4 for axial SpA) spondyloarthritis. Ankylosing spondylitis (AS) Patients fulfilling the modified New York criteria for AS 1. Thus, patients with pelvis x-ray changes consistent with Non-radiographic axial spondyloarthritis (nr-axspa) Source: FDA generated AS. Patients fulfilling the ASAS criteria 2-4 for axial spondyloarthritis, but without pelvis x-ray changes consistent with AS. Of note, these patients may have MRI changes suggestive of sacroiliitis. Data suggest that the entities nr-axspa and AS might describe different patient populations due to differences in demographics, genetics, and response to treatment Further, limited data are available on the prevalence and natural history of nr-axspa. To our knowledge, no prospective evaluations have been conducted to describe the natural history of nr-axspa as defined by the ASAS classification criteria 2, 3. While the natural history of nr-axspa is unknown, patients with nr-axspa can have significant pain and functional limitations secondary to their illness. Thus, clinical studies have evaluated treatment options, such as TNF-inhibitors, for nr-axspa. A topic of this advisory committee meeting will be whether the risk/benefit profile of treating patients with nr-axspa is different than the risk/benefit profile of treating patients with AS. Overview of Clinical Program This document provides a clinical review of sbla /323, a supplemental application proposing to expand the indication of adalimumab to include the treatment of active nr-axspa with objective signs of inflammation by elevated CRP or MRI. Adalimumab is a monoclonal antibody to tumor necrosis factor (TNF) alpha. Elevated TNF is thought to play an important role in inflammation and joint destruction that occurs in a variety of forms of inflammatory arthritis, including RA, JIA, PsA, and AS. The efficacy and safety of adalimumab has been established in these forms of inflammatory arthritis. The Sponsor submitted the results from one study with a 12-week placebo controlled portion as the primary basis for the efficacy and safety of adalimumab for the treatment of signs and symptoms of nr-axspa. This study was: Study M10-791: Randomized, placebo-controlled, 12-week double-blind period followed by a 144-week open-label extension in patients with nr-axspa. During the double-blind period, 192 patients were randomized (1:1) to adalimumab or placebo. 9

24 The dose of adalimumab was 40mg subcutaneous (SC) every other week and was based on the approved dose of adalimumab for AS. Patients were included who met the ASAS classification criteria for axial SpA, but did not meet the radiographic modified New York criterion for AS. These patients were referred to as nr-axspa. Importantly, exclusion of patients with AS occurred at screening when local investigators evaluated anteroposterior (AP) pelvis x-rays. In post-hoc analyses performed by the Sponsor, these pelvis x-rays were re-read by central readers and differences between central and local x-ray evaluations were found. Of the 102 patients with pelvis x-rays available for central evaluation, 38 patients were reclassified as fulfilling the modified New York criterion for AS. Thus, there was a difference in the number of patients with AS in the study depending on whether the x-rays were read locally or centrally. Patients included in study M had an inadequate response to NSAIDs, intolerance to at least 1 NSAID, or a contraindication for NSAIDs as defined by the investigator. In addition, patients needed to have active baseline disease, which was defined by a total back pain visual analogue score 40 (on a scale of 0 to 100) and Bath AS Disease Activity Index (BASDAI) 4 (on a scale of 0 to 10) at both the screening and baseline visits. The patient population is the major issue for discussion at this meeting. While the Sponsor utilized the ASAS criteria for entry, patients had to have active disease as defined by the sponsor. In addition, there are post-hoc analyses with different patient subgroups that had an impact on the response to adalimumab. All the variations in defining the patient population raise questions regarding whether axial SpA is understood well enough to support an indication. 1.2 Summary of Efficacy Study M was submitted as the primary source of efficacy data for adalimumab in the treatment of nr-axspa. The trial consisted of a 12-week double-blind period followed by a 144-week open label extension. The Sponsor submitted data to Week 68. All primary and secondary efficacy analyses were performed at Week 12. The primary endpoint was the ASessment in Ankylosing Spondylitis (ASAS) 40 response. In order to achieve an ASAS40 response, a patient had to demonstrate improvement of at least 40% relative to baseline and absolute improvement of at least two units (on a scale of 0 to 10) from baseline in at least three of the following four domains, with no deterioration in the potential remaining domain: patient global disease activity, total back pain, function (measured by the Bath Ankylosing Spondylitis Function Index [BASFI]), and inflammation (measured by two questions from the Bath Ankylosing Spondylitis Disease Activity Index [BASDAI]). This trial was well-controlled and had endpoints that are considered acceptable for efficacy evaluations in AS. Whether the trial design and endpoints are reasonable for patients with nr-axspa will be a key discussion issue at this meeting. In the pre-specified primary analyses, adalimumab met the primary and key secondary 10

25 endpoints. The Sponsor performed additional post-hoc analyses in patients with objective baseline evidence of inflammation in CRP or MRI. The magnitude of effect was greater in this subgroup than in the overall population. After 104 weeks of study participation, screening pelvis x-rays were re-read centrally. The central x-ray readings revealed that the trial included patients with x-ray changes consistent with AS at screening. Patients with AS should have been excluded from the study population. Post-hoc analyses evaluated whether the magnitude of the treatment effect varied based on the presence or absence of radiographic changes consistent with AS. While there are limitations to these post-hoc analyses, they suggest that the presence or absence of radiographic changes consistent with AS correlated with patients response to adalimumab. Specifically, patients with AS at screening had larger treatment differences than patients without AS at screening. Patients without x- ray changes based on central reading had small, numerical differences between adalimumab and placebo for all of the primary and ranked secondary endpoints. Whether or not these small treatment differences are clinically meaningful will be an important discussion point at this meeting. Further, we will discuss the implications of potential differences in treatment response based on the presence or absence of radiographic changes consistent with AS. Generally, post-hoc analyses are considered hypothesis generating and require further substantiation in at least one additional clinical trial. 1.3 Summary of Safety The safety information for adalimumab in nr-axspa is obtained from one study during which 190 patients received at least one dose of adalimumab. During the 12 week double-blind period, the median treatment duration was 83 days (range 14 to 93 days). Following the 12-week double-blind period, there was a 144-week open label extension. The Sponsor submitted data to 68 weeks, with overall median treatment duration of 393 days (range 14 to 488 days). There were no deaths during the 12-week, double-blind treatment period. There were two deaths during the open label extension when all patients were receiving adalimumab. The first patient died secondary to suicide 40 days after the last dose of adalimumab. The second patient was found dead on day 649. The cause of death was unclear. Based on FDA review of the data provided, it was felt unlikely that either case was related to study drug. Serious adverse events (SAEs) were uncommon during the 12-week, double-blind treatment period. A slightly higher proportion of adalimumab-treated patients had SAEs compared to placebo-treated patients. This slight imbalance did not appear to be secondary to a particular class of adverse events (AEs). The most common SAEs in the adalimumab-treatment group were acute hepatitis (1%), breast dysplasia (1%), and induced abortion (1%). There were no notable imbalances between individual SAEs. 11

26 The most common adverse events (AEs) leading to discontinuation were nausea, vomiting, acute hepatitis, and decreased weight. There were not significant imbalances between groups. The most commonly reported adverse events in the adalimumabtreatment group were nasopharyngitis (12%), nausea (7%), headache (6%), diarrhea (4%), and injection site reaction (4%). All of these AEs were more common in the adalimumab-treatment group than the placebo-treatment group, except for nausea. However, there were not marked differences in the proportion of these AEs in the adalimumab versus placebo-treatment group. Due to specific safety concerns with TNF-inhibitors, analyses were conducted related to adverse events of special interest, including infections, malignancies, hepatotoxicity, cardiovascular events, demyelinating disorders, injection site reactions, autoimmunity, and hematologic cytopenias. These analyses did not reveal new safety concerns with the use of adalimumab in nr-axspa. Of note, there was a case of disseminated tuberculosis during the open label extension in a patient without known risk factors for tuberculosis. Overall, no new safety signals were identified in study M10-791and adalimumab had a similar safety profile in nr-axspa to that reflected in the product label. While no new safety signals were identified in study M10-791, the safety profile of adalimumab is wellcharacterized for the approved rheumatologic indications and the known risks of adalimumab should be considered in the risk/benefit consideration. 2 Introduction and Regulatory Background 2.1 Product Information Proposed Trade Name (established name): Humira (adalimumab) Proposed Indication: The proposed indication in the submitted label is reducing signs and symptoms in adult patients with active non-radiographic axial spondyloarthritis with objective signs of inflammation by elevated CRP or MRI, who have had an inadequate response to, or are intolerant to, a nonsteroidal anti-inflammatory drug. Proposed Age Group: Adult patients Proposed Dose Regimen: 40mg SC every other week Pharmacological Class: Monoclonal antibody to TNF-alpha Description: Adalimumab is a recombinant human IgG1 monoclonal antibody to human tumor necrosis factor (TNF)-alpha. Adalimumab is produced by recombinant DNA technology in a mammalian cell expression system and is purified by a process that includes specific viral inactivation and removal steps. It consists of 1330 amino acids and has a molecular weight of approximately 148 kilodaltons. 12

27 How supplied: Adalimumab is supplied as a sterile, preservative-free solution of adalimumab for subcutaneous administration. The drug product is supplied as either a single-use, prefilled pen (HUMIRA Pen) or as a single-use, 1 ml prefilled glass syringe. Enclosed within the pen is a single-use, 1mL prefilled glass syringe. The solution of HUMIRA is clear and colorless, with a ph of about Tables of Currently Available Treatments for Proposed Indications There are no approved products in the United States to treat non-radiographic axial spondyloarthritis (nr-axspa). Patients with nr-axspa are in the subset of the larger category of inflammatory disease, frequently called axial spondyloarthritis (axial SpA). An additional subgroup of axial SpA is ankylosing spondylitis (AS). Patients with AS have radiographic evidence of disease in the sacroiliac joints. There are four biologic TNF-inhibitors that are approved in the United States for the treatment of AS (Table 3). Table 3: Approved products for the treatment of AS in the United States Product BLA Date of approval (Sponsor) for AS Characteristic ROA 1 Etanercept /24/03 Fusion protein SC (Enbrel ) (Immunex) (TNF-inhibitor) 2 Infliximab (Remicade ) (Centocor) 12/17/04 Monoclonal antibody (TNF- IV 3 Adalimumab (Humira ) 4 Golimumab (Simponi ) (Abbott) (Centocor) inhibitor) 8/28/06 Monoclonal antibody (TNFinhibitor) 4/24/09 Monoclonal antibody (TNFinhibitor) NSAIDs (e.g., celecoxib, diclofenac, indomethacin, naproxen, sulindac) and steroids (e.g., betamethasone, cortisone, dexamethasone, hydrocortisone, methylprednisolone, prednisone, prednisolone, and triamcinolone) are also approved for the treatment of AS Etanercept was originally approved in 1998 for RA, infliximab was originally approved in 1998 for Crohn s Disease, adalimumab was originally approved in 2002 for RA, and golimumab was approved for RA, PsA, and AS concurrently Abbreviations: BLA=Biologics License Applications; ROA=route of administration; SC=subcutaneous; IV=intravenous; RA=rheumatoid arthritis; PsA=psoriatic arthritis; AS=ankylosing spondylitis Source: FDA generated from prescribing information for infliximab, etanercept, adalimumab, golimumab (accessed May 20, 2013) 2.3 Availability of Proposed Active Ingredient in the United States Adalimumab is commercially available in the United States. Adalimumab received FDA approval on December 31, 2002 for the treatment of RA. Adalimumab was subsequently approved for multiple indications at various dates, including the treatment of adults with signs and symptoms of active AS on August 28, Important Safety Issues with Consideration to Related Drugs There are five biologic TNF-inhibitors that are approved in the United States (see Table 4) for multiple inflammatory diseases in adults (RA, psoriatic arthritis, AS, PsA, CD, and SC SC 13

28 UC) and two pediatric diseases (pediatric CD and pjia). Of the five approved biologic TNF-inhibitors, four, including adalimumab, are approved for AS. Table 4: Approved biologic TNF-inhibitors in the United States as of May 30, 2013 Product 1 Etanercept (Enbrel ) 2 Infliximab (Remicade ) 3 Golimumab (Simponi ) 4 Certolizumab (Cimzia ) 5 Adalimumab (Humira ) BLA (Sponsor) (Immunex) (Centocor) (Centocor) (UCB) (Abbott) Year of initial approval Characteristic ROA Approved indications 1998 Fusion protein (TNF-inhibitor) 1998 Monoclonal antibody (TNFinhibitor) 2009 Monoclonal antibody (TNFinhibitor) 2008 Fab fragment (TNF-inhibitor) 2002 Monoclonal antibody (TNFinhibitor) SC RA, PsA, AS, Ps, pjia N/A IV RA, PsA, AS, Ps, N/A CD, pediatric CD, UC, pediatric UC SC RA, PsA, AS N/A Proposed indications SC RA, CD Axial SpA SC RA, PsA, AS, Ps, CD,UC, pjia Nr-axSpA supplemental BLA under review Abbreviations: BLA=Biologics License Applications; ROA=route of administration; SC=subcutaneous; IV=intravenous; RA=rheumatoid arthritis; PsA=psoriatic arthritis; AS=ankylosing spondylitis; Ps=psoriasis; CD=Crohn s disease; pjia=polyarticular juvenile idiopathic arthritis; UC=ulcerative colitis; nr-axspa=non-radiographic axial spondyloarthritis Source: FDA generated from prescribing information for infliximab, etanercept, adalimumab, golimumab, certolizumab (accessed May 20, 2013) 14

29 Table 5 displays adverse reactions that have been associated with the use of TNFinhibitors and appear in the Boxed Warnings and Warnings and Precautions sections of at least one biologic TNF-inhibitor product label. See Section (Submission Specific Primary Safety Concerns) for an evaluation of the association of adalimumab with these known TNF-inhibitor adverse events. Table 5: Overview of safety concerns with TNF-inhibitors Location in label Safety concerns Boxed Warnings 1. Serious infections, including bacterial sepsis, tuberculosis, invasive fungal infections, histoplasmosis, and other opportunistic infections 2. Malignancies, including hepatosplenic T-cell lymphoma, lymphoma, and other malignancies in children and adolescents Warnings/Precautions 1. Infections 2. Hepatitis B reactivation 3. Invasive fungal infections 4. Malignancies 5. Hepatoxicity 6. Hypersensitivity disorders 7. Demyelinating disorders (e.g., multiple sclerosis) 8. Adverse outcomes in patients with heart failure 9. Pancytopenia, leukopenia, neutropenia, thrombocytopenia 10. Autoimmune disorders (e.g., lupus-like syndrome) 11. Infusion-related reactions/injection-site reactions All TNF-inhibitor labels do not contain all of these safety concerns Source: FDA generated from prescribing information for infliximab, etanercept, adalimumab, golimumab, certolizumab(accessed May 20, 2013) 2.5 Summary of Pre-submission Regulatory Activity Related to Submission An overview of the important regulatory interactions pertaining to the current submission is shown in Table 6. In 2008, prior to publication of the ASAS classification criteria, the Sponsor proposed studies in patients with axial and peripheral SpA, excluding patients who fulfilled the modified New York criteria or Classification Criteria for Psoriatic Arthritis (CASPAR) for AS or PsA, respectively. The Sponsor proposed use of the ASAS classification criteria to categorize patients as axial or peripheral SpA. At this point, the ASAS criteria were not published or finalized. This clinical review will not address considerations with regards to peripheral SpA. Rather, this clinical review will focus on the discussions in regards to axial SpA. In regards to axial SpA, the FDA noted that it was unclear if the Sponsor would be able to identify a truly distinct patient population. After discussion, it was noted that in principle, the plan to use updated SpA classification criteria was reasonable, but a final determination would depend upon the Sponsor s ability to establish that the criteria were valid and enabled one to select a discrete population of patients. If it was determined that the population selected was adequate, the Sponsor s plan to perform a 12-week, double-blind, placebo-controlled trial, followed by a one year open label extension, was reasonable. On June 13, 2011, the Sponsor had a pre-sbla meeting with the FDA. This meeting occurred after publication of the ASAS classification criteria. FDA raised concerns regarding use of these criteria to define a patient population that would be the basis of a 15

30 new indication. Specific concerns included the lack of clarity of the entities the criteria would encompass and whether patients would evolve into other diagnoses for which treatment may have an unfavorable or unknown risk/benefit profile. The Sponsor had an additional Type C meeting with the FDA on July 20, 2012 to discuss these concerns. At this meeting, the Sponsor provided additional data to support that axial SpA is not a new disease entity, but represents a spectrum of inflammatory involvement of the axial skeleton. In addition, the Sponsor provided data suggesting unmet medical need in nraxspa patients and data suggesting that the burden of disease in nr-axspa and AS patients is similar. FDA recognized that inflammation can be present in the sacroiliac joints despite the absence of sacroiliitis on radiographs, but remained concerned about the proposed indication, nr-axspa. FDA s specific concerns were that the indication is overly broad, the clinical course of patients is unclear, and the trial design that could establish an appropriate risk/benefit assessment is unknown. FDA noted that discussion of the proposed indication would likely occur in a public forum, such as an advisory committee meeting. 16

31 Table 6: Overview of regulatory interactions for the non-radiographic axial spondyloarthritis development program Type of FDA recommendations and key discussion topics Meeting (date) EOP2 Meeting 1. Sponsor proposed to study patients with SpA with axial and peripheral (8/7/08) disease (who do not fulfill a diagnosis of AS or PsA) using ASAS criteria. At this point, the ASAS criteria were not published or finalized. 2. FDA noted that is was unclear if the Sponsor would be able to exclude patients with AS and identify a truly distinct population 3. FDA stated the Sponsor would need to provide rationale for the diagnostic criteria selected. In principle, it was felt that the plan to use updated SpA classification criteria was reasonable, but a final determination would depend upon the Sponsor s ability to establish that the criteria are valid and enable the one to select a discrete population of patients 4. If it were determined that the population selected was adequate, the Sponsor s plan to perform a 12-week, double-blind, placebo controlled Pre-sBLA meeting (6/13/11) Type C meeting (7/20/12) trial, followed by a one-year open label extension appeared reasonable 1. FDA raised regulatory concerns regarding the use of the new ASAS axial SpA classification criteria to define a new indication. Specifically, there was concerns regarding the lack of clarity of the entities the criteria would encompass and whether patients could evolve into other diagnoses for which treatment may have an unfavorable or unknown risk/benefit profile. 2. FDA recognized that early treatment of spondyloarthropathy with the intent of preventing or reducing structural damage is desirable, but noted the difficulty in labeling given the disease state may be transitory 3. FDA advised that it is impractical to try to define the disease and obtain approval at the same time. FDA recommended that the Sponsor pursue collaboration with the academic community and other industry personnel interested in spondyloarthropathy to validate and better characterize these entities and their evolution over time, particularly in the United States population 4. From a scientific perspective, the FDA agreed with the overall study design and statistical methodology, but noted that the primary concern was the proposed clinical indication 1. Sponsor provided data to support that axial SpA is not a new disease entity, but represents a spectrum of inflammatory involvement of the axial skeleton, evidence demonstrating unmet medical need in nr-axial SpA patients, and data suggesting that the burden of disease in nraxspa and AS patients is similar 2. Sponsor provided data regarding prevalence of axial SpA in the United States 3. FDA recognized that inflammation can be present in the sacroiliac joints despite the absence of sacroiliitis on plain radiographs, but remained concerned about the proposed indication: nr-axspa 4. FDA's specific concerns were that the indication is overly broad, the clinical course of patients in unclear, and the trial design that could establish an appropriate risk/benefit assessment is unknown 5. Discussion of the proposed indication would likely occur in a public forum, such as an advisory committee meeting Abbreviations: SpA=spondyloarthritis; FDA=Food and Drug Administration; ASAS=Assessment of Spondyloarthritis international Society; EOP2=end of phase 2; sbla=supplemental biologic license application; AS=ankylosing spondylitis; PsA=psoriatic arthritis Source: FDA generated 17

32 3 Sources of Clinical Data 3.1 Tables of Studies/Clinical Trials The Sponsor submitted the following data from one completed study of adalimumab in their sbla to support the approval of adalimumab for the treatment of non-radiographic axial spondyloarthritis (nr-axspa): 1. One randomized, 12-week double-blind, parallel-group placebo controlled study, followed by a 144-week open label extension (M10-791, see Table 7). The Sponsor submitted data to Week 68. Table 7: Overview of the randomized, controlled study in non-radiographic axial spondyloarthritis patients Study Design/ Sites M R, DB, PC/ 37 sites Patient population Active nonradiographic axial spondylitis with an inadequate response to at least one NSAID or had a contraindication to NSAIDs Treatment and duration 12-week controlled period: Adalimumab 40mg SC eow Placebo 144-week open label extension: Adalimumab 40mg SC eow N 94 adalimumab; 91 placebo Primary efficacy variable ASAS40 at 12 weeks Ranked secondary efficacy variables ASAS20, BASDAI50, SF-36 PCS, ASAS partial remission, ASAS5/6 response, HAQ-S total score, hs- CRP, SPARCC MRI score for SI joints, SPARCC MRI score for the spine =Number of patients in Full Analysis Set (FAS) Abbreviations: R=randomized, DB=double-blind, PC=placebo controlled, eow=every other week; ASAS=ASessemnt in Ankylosing Spondylitis; BASDAI50=At least 50% improvement in Bath Ankylosing Spondylitis Disease Activity Score; SF-36 PCS=short form-36 physical component score; ASAS partial remission=absolute score of<20 units for each of the 4 domains identified in the ASAS40; ASAS5/6=20% improvement in 5 out of the following 6 domains: BASFI, total back pain, patient global disease activity, inflammation from questions 5 and 6 of BASDAI, lateral lumbar flexion from BASMI (Bath AS Metrology Index), and acute phase reactants; HAQ- S=Health Assessment Questionnaire for Spondyloarthropathies; CRP=C-reactive protein; SPARCC MRI=Spondyloarthritis Research Consortium of Canada Magnetic Resonance Imaging Index; SI=sacroiliac joints 3.2 Review Strategy Efficacy: Study M served as the one phase 3 trial for the evaluation of the efficacy of adalimumab in the treatment of signs and symptoms of nr-axspa. This trial was well-controlled and had endpoints that are considered acceptable for efficacy evaluation in AS. Whether the trial design and endpoints are reasonable for patients with nr-axspa will be a key discussion issue at this meeting. The trial selected for patients who did not meet modified New York criteria for AS. However, post-hoc analyses revealed that the trial included patients with and without x-ray changes consistent with AS based on central x-ray interpretation. Potential effects of this on the overall interpretation of the efficacy results will be further discussed in Section (Other Endpoints). The Sponsor s proposes approval of adalimumab for a subgroup of the patients originally enrolled in their clinical trial. To support efficacy in this subgroup, the Sponsor performed post-hoc analyses in patients with evidence of inflammation at 18

33 baseline based on a positive MRI (e.g. SPARCC MRI score at baseline of at least 2 for either the sacroiliac (SI) joints or the spine) or an elevated CRP at baseline. The Sponsor defines this subgroup as the adalimumab target population (ATP). Safety: The major safety evaluation of adalimumab for the treatment of signs and symptoms of nr-axspa was study M The primary safety analyses focused on data through Week 12 the double-blind, controlled period. For the Week 12 analyses, the treatment groups included the treatment groups assigned at randomization. There was no escape option during the 12 week study period. Additional safety analyses were conducted through Week 68 to help evaluate the possibility of time-dependency for adverse reactions. However, limited conclusions are possible after 12 weeks due to the lack of a placebo control group 3.3 Discussion of Individual Studies/Clinical Trials Study M The following description of the protocol for study M is based on amendment 4 dated August 1, 2011 and the statistical analysis plan (SAP) dated November 7, See Table 8 for the dates and description of all amendments to the protocol for study M The final protocol amendment, but not the final SAP occurred prior to the 24-week database lock. Two interim database locks were planned: when all subjects completed at least week 24 and when all subjects had completed up to week 68 of the study. Prior to the first database lock and unblinding, the SAP was modified to remove 7 subjects from efficacy analyses from the site of Investigator due to investigator noncompliance with the protocol requirements. The Full Analysis Set (FAS) was defined in the SAPs as the subset of the ITT Analysis Set excluding subjects from the site of Investigator Additional minor modifications to the data analysis methods were made prior to unblinding of the data. Additional changes were made after the Week 24 SAP was finalized, but prior to the Week 68 interim SAP. These changes included modifications to the statistical analysis methods for demographic and baseline characteristics and specification of imputation methods for missing data. Additional modifications were made to the SAP after the Week 68 database lock. The most relevant change was regarding additional analyses in the adalimumab target population (ATP). The ATP was a subpopulation and consisted of all patients with objective evidence of inflammation at baseline based on a positive MRI (e.g. SPARCC MRI score at baseline of at least 2 for either of the SI joints or the spine) or an elevated CRP at baseline. 19

34 Table 8: Overview of Protocol Amendments to study M Amendment Date Major modifications Original Protocol Amendment 1 November 16, 2009 Modifications made to the inclusion/exclusion criteria for clarity Amendment 2 November 19, 2009 Correction to the SAE process Amendment 3 February 10, 2010 Clarified time frames for evaluations at screening of the ASAS criteria Correction and clarification to concomitant mediation acceptability Clarification that the Week 12 MRI should be completed prior to Week 12 OL dose Amendment 4 August 10, 2011 Extended OL treatment to 144 weeks Included addition of confirmatory HLA-B27 test if the result was initially reported as equivocal Added an AP pelvis x-ray at Week 104 for monitoring progression of disease Abbreviations: SAE=serious adverse event; ASAS=Assessment of SpondyloArthritis international Society; MRI=magnetic resonance imaging; OL=open label; AP=anteroposterior Source: FDA generated Title: Study M is entitled, A multicenter study of the efficacy and safety of anti- TNF monoclonal antibody adalimumab in subjects with axial spondyloarthritis. Study Dates: August 11, 2009 through study visit week 68 for all subjects as of October 19, Sites: US, Canada, Europe, and Australia Objectives of study M10-791: The primary objective of study M was to evaluate the efficacy and safety of adalimumab 40mg given every other week SC compared to placebo for 12 weeks followed by open label safety and efficacy assessments in patients with active axial SpA based on the ASAS criteria, but not fulfilling the modified New York criteria for AS who had an inadequate response to or intolerance to at least one NSAID, or had a contraindication for NSAIDs. Overall Design of study M10-791: Study M was a phase 3, placebo-controlled, double-blind, randomized study with an open label phase conducted in the United States, Canada, Europe, and Australia in patients with active axial SpA not fulfilling the modified New York criteria for AS who had an inadequate response to or intolerance to at least one NSAID, or had a contraindication for NSAIDs. Approximately 194 patients were planned to be enrolled. The study included a 30-day screening period, a 12-week double-blind placebo-controlled treatment period, a 144-week open label treatment period, and a follow-up telephone call 70 days after the last dose of study drug. Patients were randomized in a 1:1 ratio to receive either adalimumab 40mg SC every other week or matching placebo for the 12 week double-blind treatment period. After the double-blind treatment period, all patients received open label treatment with adalimumab 40mg SC every other week. The study schematic is shown in Figure 1. 20

35 The Sponsor submitted data to Week 68. Figure 1: Study M Schematic Source: Study M Complete Study Report (Module ), received 11/29/12, page 215 Patient Selection for study M10-791: Study M randomized 192 patients who met the following inclusion and exclusion criteria: Inclusion criteria: 1. Patient was 18 years of age 2. Patient must have had an inadequate response to NSAIDs, intolerance to 1 NSAID, or had a contraindication for NSAIDs as defined by the investigator 3. Patient must have had chronic back pain (of at least 3 months duration) with onset at age < 45 years 4. MRI evidence of active inflammatory lesions of sacroiliac (SI) joints (past or present) with definite bone marrow edema/osteitis, suggestive of sacroiliitis associated with SpA plus 1 of the clinical criteria listed below: Or Positive human leukocyte antigen-b27 (HLA-B27) plus 2 of the clinical criteria listed below other than HLA-B27 positivity: Inflammatory back pain defined as the presence at screening of at least 4 out of the following 5 parameters: 1) age at onset < 40 years, 2) insidious onset, 3) improvement with exercise, 4) no improvement with rest, 5) night pain with 21

36 improvement upon getting up Arthritis (past or present) Heel enthesitis (past or present) Anterior uveitis confirmed by an ophthalmologist (past or present) Dactylitis (past or present) Crohn's Disease (CD) or ulcerative colitis (UC) (past or present) Good prior response to an NSAID back pain was not present anymore or much better 24 to 48 hours after a full dose of an NSAID Family history of SpA Positive HLA-B27 Elevated CRP 5. Patient must have had baseline disease activity as defined by having a Total Back Pain VAS score 40 mm and Bath AS Disease Activity Index (BASDAI) 4 at both the screening and baseline visits 6. If female, patient was either not of childbearing potential, defined as postmenopausal for at least 1 year or surgically sterile (bilateral tubal ligation, bilateral oophorectomy or hysterectomy) or was of childbearing potential and was practicing an approved method of birth control throughout the study and for 150 days after last dose of study drug. The results of the serum pregnancy test performed during the screening period and urine pregnancy test performed at the baseline visit had to be negative. Examples of approved methods of birth control included the following: Condoms, sponge, foams, jellies, diaphragm, or intrauterine device (IUD); Oral, parenteral, or intravaginal contraceptives for 90 days prior to study drug administration; A vasectomized partner. 7. Patient was judged to be in good health as determined by the PI based upon the results of medical history, laboratory profile, physical examination, chest x-ray (CXR), and a 12-lead electrocardiogram (ECG) performed at screening; Patient had a negative purified protein derivative (PPD) test (or equivalent) and CXR (posterior-anterior [PA] and lateral view) at screening. If the subject had a positive PPD test (or equivalent) and/or CXR (PA and/or lateral view) consistent with prior TB exposure, the subject was to initiate, be currently receiving, or have documented completion of a course of anti-tb therapy. 8. This inclusion criterion was deleted because it was incorporated as the bullet under Inclusion Criterion 7 in Protocol Amendment 1 9. Patient was able and willing to provide written informed consent and comply with the requirements of this study protocol. 10. Patient was able and willing to self-administer SC injections or have a qualified person available to administer SC injections. Exclusion Criteria: 22

37 1. Patient fulfilling a diagnosis of AS (as defined by the modified New York criteria) at or prior to the screening visit 2. Patient with a past or present diagnosis of psoriasis (Ps) or PsA 3. Prior exposure to any biologic therapy with a potential therapeutic impact on SpA, including anti-tnf therapy 4. Patient had received any live (attenuated) vaccines within 52 weeks prior to the baseline visit (This exclusion criterion was removed in Protocol Amendment 1) 5. If entering the study on concomitant disease-modifying anti-rheumatic drugs (DMARDs), subject was not on stable dose of methotrexate (MTX) ( 25 mg per week) and/or sulfasalazine (SSZ) ( 3 g per day), and/or hydroxychloroquine ( 400 mg per day) for 28 days prior to the baseline visit 6. If entered the study on concomitant oral corticosteroids, patient was not on stable dose of prednisone ( 10 mg per/day), or oral corticosteroid equivalents, for at least 14 days prior to the baseline visit 7. Patient had received cyclosporine or other second line anti-rheumatic therapy (except MTX, SSZ, hydroxychloroquine, or azathioprine) within 28 days prior to the baseline visit 8. Patient had been treated with intra-articular joint injection(s) or spinal/paraspinal injection(s) of corticosteroids in the preceding 28 days prior to the baseline visit 9. Patient had been treated with any investigational drug of chemical or biologic nature within a minimum of 30 days or 5 half-lives (whichever was longer) of the drug prior to the baseline visit 10. Infection(s) requiring treatment with intravenous (IV) antibiotics, IV antivirals, or IV antifungals within 30 days prior to the baseline visit or oral antibiotics, oral antivirals, or oral antifungals within 14 days prior to baseline visit 11. Patient with extra-articular manifestations (e.g., inflammatory bowel disease, uveitis, etc.) that were not clinically stable for at least 30 days prior to study entry 12. Patient had a history of inflammatory arthritis of a different etiology other than axial SpA (e.g., rheumatoid arthritis, gout, systemic lupus erythematosus, polyarticular, or systemic juvenile idiopathic arthritis) 13. Known hypersensitivity to the excipients of adalimumab as stated in the label 14. History of central nervous system (CNS) demyelinating disease or neurologic symptoms suggestive of CNS demyelinating disease 15. History of listeriosis, histoplasmosis, chronic or active hepatitis B infection, human immunodeficiency virus infection, immunodeficiency syndrome, chronic recurring infections, or active tuberculosis (TB) 16. History of moderate to severe congestive heart failure (New York Association [NYHA] class III or IV), recent cerebrovascular accident (CVA) and any other condition which, in the opinion of the investigator, would put the subject at risk by participation in the protocol 17. Evidence of dysplasia or history of malignancy (including lymphoma and leukemia) other than a successfully treated non-metastatic cutaneous squamous cell, basal cell carcinoma, or localized carcinoma in situ of the cervix 18. Female patients who are pregnant or breastfeeding or considering becoming pregnant during the study 23

38 19. History of clinically significant drug or alcohol abuse in the last 12 months 20. Clinically significant abnormal screening laboratory results as evaluated by the investigator 21. Patient was considered by the investigator, for any reason, to be an unsuitable candidate for the study 22. If entering the study on concomitant azathioprine, subject not on stable dose ( 150 mg/day) for 28 days prior to the baseline visit or on azathioprine and another concomitant immunosuppressive drug at study entry 23. If entering the study on concomitant NSAIDs and/or analgesics, subject on opioid analgesics (other than tramadol) within 14 days prior to baseline visit or subject not on stable doses of NSAIDs and/or analgesics for 14 days prior to the baseline visit 24. Patient had undergone spinal surgery within 2 months prior to baseline Treatments in study M10-791: During the 12 week double-blind stage, study drug was provided as a sterile subcutaneous (SC) injection solution in 1-mL pre-filled syringes containing either adalimumab 40mg/0.8mL or matching placebo for adalimumab. Study drug was to be self-administered SC every other week at approximately the same time of day. At baseline, Week 2, and Week 12, patients were to self-administer study drug under guidance of study site personal following all specified study procedures. Selection of doses in study M10-791: The dose in the study was selected based on the approved dose of adalimumab for AS. Efficacy and safety of adalimumab in AS were evaluated in Study M (ATLAS) and Study M Concomitant medications in study M10-791: Each vaccine and all medications, except study drug, administered to a subject during the study were to be listed as concomitant medications. The patients were not to initiate any therapies for axial SpA treatment during the study other than non-drug therapy, such as physiotherapy. Patients could continue on stable doses of MTX, SSZ, hydroxychloroquine, azathioprine, prednisone, and/or NSAIDs provided they were at stable doses (see exclusion criteria). All current second-line anti-rheumatic therapies, corticosteroids, NSAIDs, and analgesics specifically administered for axial SpA were to be captured on the concomitant medication electronic case report form (ecrf). Doses of these concomitant medications were to remain stable for the first 24 weeks of participation (except as medically required due to an AE). Dose adjustments or induction of treatment with these agents were permitted after Week 24. Patients on stable doses of analgesic(s) were allowed to continue during the study. However, opioid analgesics (except for tramadol) were prohibited from baseline to Week 24. The dose of analgesic(s) being used for axial SpA pain was not to be changed through Week 24. For patients who were taking analgesics on an as needed basis, analgesics were to be discontinued for at least 24 hours prior to a study visit. 24

39 Analgesic(s) could be initiated during the study after Week 24. Only 1 intra-articular corticosteroid injection for a peripheral joint was to be allowed during the first 24 weeks of the study. After Week 24, intra-articular corticosteroid injections were to be allowed at the investigator's discretion. Once a joint was injected it was to be considered not evaluable/assessable ("NA") during the 28 days following injection. No spinal, para-spinal, or SI joint injections were to be allowed during the first 24 weeks of the study. Prohibited therapies in study M The following were prohibited medications during the study: All biologic therapy with a potential therapeutic impact on SpA including but not limited to the following: Enbrel (etanercept) Remicade (infliximab) Orencia (abatacept) Kineret (anakinra) Rituxan (rituximab) Tysabri (natalizumab) Actemra (tocilizumab) Raptiva (efalizumab) Simponi (golimumab) Cimzia (certolizumab) Live vaccines (during the study and for 70 days after the last dose of study drug) Rifampin/pyrazinamide combination All other second-line anti-rheumatic therapies other than MTX, SSZ, azathioprine, or hydroxychloroquine Opioid analgesics (other than tramadol) until Week 24 High potency opiates such as methadone, hydromorphone, and morphine Any investigational drug of chemical or biologic nature Patients were to be discontinued from the study if any of the prohibited medications were used during the study. Patient stopping rules in study M Voluntary withdrawal of the patient 2. The investigator could discontinue the patient for any reason, including AE, safety concern, or failure to comply with the protocol 3. Patients were to be withdrawn from the study immediately if any 1 of the following occurred: Clinically significant abnormal laboratory result(s) or AE(s), which rule out continuation of the study drug, as determined by the investigator an the Abbott Medical Monitor The investigator believed it was in the best interest of the patient An inclusion or exclusion criteria violation(s) was noted after the patient 25

40 started the study drug and continuation of the study drug placed the patient at risk as determined by the Abbott Medical Monitor Introduction of prohibited medications or dosages when continuation of the study drug would place the patient at risk as determined by the Abbott Medical Monitor Patient non-compliant with tuberculosis prophylaxis Patient became pregnant during the study Patient known to have dysplasia of the gastrointestinal tract or malignancy, except for localized non-melanoma skin cancer Patient diagnosed with lupus like syndrome, multiple sclerosis, or demyelinating disease Patient significantly non-compliant with study procedures which would put the patient at risk for continued participation in the study, in consultation with the Abbott Medical Monitor and the investigator Study monitoring and evaluations for study M10-791: Study monitoring and evaluations for study M are presented in Table 9. Table 9: Overview of monitoring in study M

41 27

42 28

43 Source: Study Report Body (Module ), submitted 11/29/12, Table 2, pages AP Pelvis x-ray Anteroposterior (AP) pelvis x-rays to evaluate the SI joints were obtained at screening. The AP pelvis x-ray was not required if the patient had a previous AP pelvis x-ray film 180 days prior to screening. A qualified radiologist or rheumatologist was to read the screening visit AP pelvis x-ray. A repeat AP pelvis x-ray was to be performed at Week 104 visit to evaluate for the presence of sacroiliitis that fulfilled the radiologic component of the modified New York criteria for AS. A qualified radiologist or rheumatologist was to read the Week 104 visit AP pelvic x-ray film. The x-ray was only performed at sites where the local IEC/IRB approved it. Screening and Week 104 images were sent to the central imaging laboratory designated by the Sponsor. Differences in local and central interpretations of screening x-rays are discussed further in Section (Subject Disposition). MRI of the Spine and SI joints (Spondyloarthritis Research Consortium of Canada [SPARCC] Method) All patients were to have an MRI evaluation of the cervical, thoracic, and lumbar regions of the spine as well as the SI joints at the designated study visits with in the following visit windows: Screening: screening period until baseline Week 12: up to 7 days prior to Week 12 visit Week 52 and 104: ±7 days from forecasted date of study visit Images were sent to the central imaging laboratory designated by the Sponsor. Efficacy Endpoints 29