Abstract Objectives Minimally important changes (MICs) for SPondyloArthritis Research Consortium of Canada (SPARCC) MRI scores are ⩾2.5 for SI joint and ⩾5 for spine. This post hoc analysis assessed achievement of MIC in SPARCC scores in biologic-naïve patients with AS treated with tofacitinib or placebo, and correlation with clinical responses. Methods Adult AS patients in a 12-week phase 2 study (n = 207) were randomized 1: 1: 1: 1 to tofacitinib 2, 5 or 10 mg twice daily (BID) or placebo. MIC in SPARCC SI joint and spine scores were assessed for patients with available MRI data (N = 164; 79%). Clinical endpoints at week 12, including Assessment of SpondyloArthritis international Society 20% improvement (ASAS20), were compared between patients achieving/not achieving MIC. Results A greater proportion of patients achieved MIC with tofacitinib 2, 5 and 10 mg BID vs placebo for SI joint (28.6, 38.6, 29.6 vs 11.8%) and spine scores (29.3, 36.4, 40.9 vs 11.8%). Generally, a greater proportion of patients treated with tofacitinib 2, 5 and 10 mg BID or placebo, respectively, who achieved MIC for SI joint and spine scores achieved ASAS20 (SI joint: 75.0, 88.2, 69.2, 75.0%; spine: 91.7, 85.7, 72.2, 75.0%) vs patients who did not achieve MIC (SI joint: 51.7, 84.0, 58.1, 48.3%; spine: 46.4, 85.7, 53.8, 48.3%). Numerically greater responses were seen in those patients achieving vs not achieving MIC across a range of other efficacy assessments. Conclusion Approximately one-third of tofacitinib-treated AS patients experienced clinically meaningful reductions in spinal MRI inflammation at week 12. Patients achieving MIC for MRI inflammation had greater clinical response.

Background Tofacitinib is an oral JAK inhibitor that has been investigated for the treatment of ankylosing spondylitis (AS). Objectives To investigate for the first time the effects of tofacitinib in adult patients (pts) with active AS. Methods This was a 16-week (wk), Phase 2, multicentre, randomised, double-blind, placebo (PBO)-controlled, dose-ranging study (NCT01786668) to investigate efficacy, safety and dose-response of tofacitinib in pts with AS. Pts fulfilling the modified New York criteria (central read) were randomised 1:1:1:1 to PBO or tofacitinib 2, 5 or 10 mg BID for 12 wks plus 4 wks follow-up. Primary efficacy endpoint was ASAS20 response rate at Wk 12 using a 3-parameter Bayesian Emax model by assuming a monotonic response. Secondary endpoints included ASAS40 response rate, AS disease activity score using C-reactive protein (ASDAS), Bath AS disease activity index 50% (BASDAI50) response rate, Bath AS functional index, Bath AS metrology index (linear method), Spondyloarthritis Research Consortium of Canada (SPARCC) score of sacroiliac (SI) joints and spine. Safety endpoints included adverse events (AEs) and laboratory outcomes. Results 208 pts were randomised and 207 treated; 196 pts completed the study. 51 pts (PBO group) and 52 (each tofacitinib group) were included in analyses. Baseline demographics and disease characteristics were balanced between groups and typical of AS populations (87.4% HLA-B27 positive; 81.2% white; 69.1% male; mean age 42 years; mean disease duration 6.3 years; mean BASDAI 6.7). The table summarises safety and efficacy results. The ASAS20 Emax model showed tofacitinib 10 mg BID had a high response rate and its confidence bounds met a pre-specified efficacy decision rule. ASAS20 response rates by NRI were significantly greater with tofacitinib 5 mg BID vs PBO, compared with 2 or 10 mg BID. All tofacitinib groups had ASAS40, ASDAS and BASDAI50 improvements of similar magnitude vs PBO. Tofacitinib 2 mg BID did not differ vs PBO in remaining clinical efficacy measures or SPARCC scores. Tofacitinib 5 and 10 mg BID had greater clinical efficacy vs PBO, with minimal differences between doses, and significantly improved SPARCC SI joint and spine scores vs PBO. No tofacitinib-related safety issues unique to the AS population or new safety concerns were identified. Two treatment-related herpes zoster cases were reported (1 each with tofacitinib 2 and 10 mg BID). No cases of tuberculosis, malignancy, gastrointestinal perforation or death were reported. Dose-dependent changes in laboratory outcomes commonly reported in other tofacitinib studies were observed and returned to approximately baseline values by Wk 16. Conclusions Tofacitinib 5 and 10 mg BID demonstrated greater clinical and imaging efficacy vs PBO in reducing the signs and symptoms of AS in adults with active AS. Safety was similar to that reported for tofacitinib studies in other indications. Acknowledgement Previously presented (van der Heijde D et al. Arthritis Rheumatol 2015; 67(S10): Abstr 5L) and reproduced with permission from Arthritis Rheumatol. This study was funded by Pfizer Inc. Editorial support was provided by S Johnson of CMC and funded by Pfizer Inc. Disclosure of Interest D. van der Heijde Consultant for: AbbVie, Amgen, AstraZeneca, Augurex, Bristol-Myers Squibb, Boehringer Ingelheim, Celgene, Centocor, Chugai, Covagen, Daiichi, Eli Lilly, Galapagos, GlaxoSmithKline, Janssen Biologics, Merck, Novartis, Novo Nordisk, Otsuka, Pfizer Inc, Roche, Sanofi-Aventis, UCB, Vertex, A. Deodhar Grant/research support from: AbbVie, Amgen, Eli Lilly, Janssen, Novartis, Pfizer Inc, and UCB, Consultant for: AbbVie, Amgen, Boehringer Ingelheim, Janssen, Novartis, Pfizer Inc, and UCB, J. Wei Grant/research support from: AbbVie, Bristol-Myers Squibb, Celgene, Chugai, Eisai, Janssen, Novartis, Pfizer Inc, Sanofi-Aventis, TSH Taiwan, and UCB, Consultant for: AbbVie, Bristol-Myers Squibb, Celgene, Chugai, Eisai, Janssen, Novartis, Pfizer Inc, Sanofi-Aventis, TSH Taiwan, and UCB, E. Drescher Grant/research support from: AbbVie, Bristol-Myers Squibb, Celgene, Pfizer Inc, Novartis, UCB, Amgen, Lilly, and Sanofi-Aventis, D. Fleishaker Shareholder of: Pfizer Inc, Employee of: Pfizer Inc, T. Hendrikx Shareholder of: Pfizer Inc, Employee of: Pfizer Inc, D. Li Shareholder of: Pfizer Inc, Employee of: Pfizer Inc, S. Menon Shareholder of: Pfizer Inc, Employee of: Pfizer Inc, K. Kanik Shareholder of: Pfizer Inc, Employee of: Pfizer Inc

Background Our objective was to evaluate the efficacy (clinical and biomarker) and safety of intravenous bapineuzumab in patients with mild to moderate Alzheimer’s disease (AD). Methods Two of four phase 3, multicenter, randomized, double-blind, placebo-controlled, 18-month trials were conducted globally: one in apolipoprotein E ε4 carriers and another in noncarriers. Patients received bapineuzumab 0.5 mg/kg (both trials) or 1.0 mg/kg (noncarrier trial) or placebo every 13 weeks. Coprimary endpoints were change from baseline to week 78 on the 11-item Alzheimer’s Disease Assessment Scale–Cognitive subscale and the Disability Assessment for Dementia. Results A total of 683 and 329 patients completed the current carrier and noncarrier trials, respectively, which were terminated prematurely owing to lack of efficacy in the two other phase 3 trials of bapineuzumab in AD. The current trials showed no significant difference between bapineuzumab and placebo for the coprimary endpoints and no effect of bapineuzumab on amyloid load or cerebrospinal fluid phosphorylated tau. (Both measures were stable over time in the placebo group.) Amyloid-related imaging abnormalities with edema or effusion were confirmed as the most notable adverse event. Conclusions These phase 3 global trials confirmed lack of efficacy of bapineuzumab at tested doses on clinical endpoints in patients with mild to moderate AD. Some differences in the biomarker results were seen compared with the other phase 3 bapineuzumab trials. No unexpected adverse events were observed. Trial registration Noncarriers (3000) ClinicalTrials.gov identifier NCT00667810 ; registered 24 Apr 2008. Carriers (3001) ClinicalTrials.gov identifier NCT00676143 ; registered 2 May 2008.

Immunotherapy with monoclonal antibodies that target amyloid beta has been under investigation as a treatment for patients with Alzheimer’s disease (AD). The 3000 and 3001 phase 3 clinical studies of intravenous bapineuzumab assessed safety and efficacy in patients with mild to moderate AD recruited in over 26 countries. This article describes the long-term safety and tolerability of bapineuzumab in the extension studies for these two protocols.

Diffusion weighted imaging (DWI) is an MRI technique measuring the apparent diffusion coefficient (ADC) of water molecules. As Alzheimer's disease (AD) progresses, neuronal degeneration and cerebral atrophy may manifest as an increase in ADC values owing to increased diffusivity of water molecules. This analysis compares DWI-ADC histogram values to clinical characteristics in subjects with mild to moderate AD. DWI was performed at Screening as part of a standard MRI protocol to assess subject eligibility for two clinical trials (see http://clinicaltrials.gov/ct2/show/NCT00996918 and http://clinicaltrials.gov/ct2/show/NCT00998764). ADC values were computed for each brain voxel, resulting in normalized histograms with a bin width of 0.1x10–4 mm 2/s, before and after thresholding (18x10–4 mm 2/s) to account for the increased water signal resulting from cerebral atrophy. Histograms were characterized by mean, peak and height. For the first 1000 subjects screened, Pearson correlation was computed between these histogram values and age, time since AD diagnosis, MMSE, ADAS-COG 11 and CDR-SOB. Sub-groups were analyzed based on MRI scanner field strength (1T, 1.5T and 3T). Mean, peak and thresholded mean over the whole brain showed a moderate positive linear correlation with age (n = 996, r = 0.34, 0.31 and 0.35 respectively, with p-values<0.0001). Mean was also moderately correlated with ADAS-COG and CDR (n = 643, r = 0.31 and 0.32, P <0.0001), while correlations with peak and thresholded mean were lower. A weak correlation was found between mean and MMSE (r = -.18, P <0.0001) and time since diagnosis (r = 0.15, P <0.0001). Subgroup analyses by field strength (n = 41 for 1T, 787 for 1.5T and 168 for 3T) demonstrated correlations in a similar range (for mean: r = 0.31, 0.31 and 0.28 for ADAS, r = 0.19, 0.32 and 0.28 for CDR), with some variation due to sample sizes. In subjects with mild to moderate AD, increased DWI ADC values over the whole brain correlate moderately with older subjects and more impaired ADAS-COG 11 and CDR-SOB. These results are independent of MRI scanner field strength. Future studies will focus on longitudinal analyses in AD subjects.

Diffusion weighted imaging (DWI) is an MRI technique measuring the apparent diffusion coefficient (ADC) of water molecules. As Alzheimer's disease (AD) progresses, neuronal degeneration and cerebral atrophy may manifest as an increase in ADC values owing to increased diffusivity of water molecules. This analysis compares DWI-ADC histogram values to clinical characteristics in subjects with mild to moderate AD. DWI was performed at Screening as part of a standard MRI protocol to assess subject eligibility for two clinical trials (see http://clinicaltrials.gov/ct2/show/NCT00996918 and http://clinicaltrials.gov/ct2/show/NCT00998764). ADC values were computed for each brain voxel, resulting in normalized histograms with a bin width of 0.1x10 -4 mm 2/s, before and after thresholding (18x10 -4 mm 2/s) to account for the increased water signal resulting from cerebral atrophy. Histograms were characterized by mean, peak and height. For the first 1000 subjects screened, Pearson correlation was computed between these histogram values and age, time since AD diagnosis, MMSE, ADAS-COG 11 and CDR-SOB. Sub-groups were analyzed based on MRI scanner field strength (1T, 1.5T and 3T). Mean, peak and thresholded mean over the whole brain showed a moderate positive linear correlation with age (n = 996, r = 0.34, 0.31 and 0.35 respectively, with p-values<0.0001). Mean was also moderately correlated with ADAS-COG and CDR (n = 643, r = 0.31 and 0.32, P <0.0001), while correlations with peak and thresholded mean were lower. A weak correlation was found between mean and MMSE (r = -0.18, P <0.0001) and time since diagnosis (r = 0.15, P <0.0001).Subgroup analyses by field strength (n = 41 for 1T, 787 for 1.5T and 168 for 3T) demonstrated correlations in a similar range (for mean: r = 0.31, 0.31 and 0.28 for ADAS, r = 0.19, 0.32 and 0.28 for CDR), with some variation due to sample sizes. In subjects with mild to moderate AD, increased DWI ADC values over the whole brain correlate moderately with older subjects and more impaired ADAS-COG 11 and CDR-SOB. These results are independent of MRI scanner field strength. Future studies will focus on longitudinal analyses in AD subjects.

Depression is a common comorbidity in patients with AD. In two recent clinical trials for treatment of mild to moderate AD, patients with a medical history of depression were permitted to participate except those with current major depressive disorder. The prevalence of depression and clinical characteristics of these patients were investigated. A total of 2490 patients with AD who reported their medical history at a screening visit of two trials between December 2007 and October 2011 were included (ClinicalTrials.gov identifiers NCT00667810 and NCT00676143). The prevalence of a past medical history of depression was described for the overall patient sample and for patient subgroups defined by race, gender, and age at AD diagnosis. Onset of depression in relation to date of AD diagnosis was described. A statistical-modeling approach was used to identify the AD patient subgroup with the highest prevalence of depression. Approximately 27% of patients reported a past medical history of depression at screening. Prevalence of depression was 33% in subjects with a diagnosis of AD at or before age 65 and 24% in those with a diagnosis of AD after age 65, with 31% in women, 22% in men, 33% in whites, and 15% in other races. The prevalence did not vary with ApoE e4 carrier status or MMSE scores (<21 vs ≥21). The statistical model estimated that white women with a diagnosis of AD would have the highest estimated prevalence of depression (45%). Approximately 89% of patients who had medical history of depression reported concurrent depression at screening. In comparing the time of concurrent depressive onset and diagnosis of AD in these patients, we found that depression occurred prior to AD diagnosis in approximately 44% of these patients. The percentage of women (49%) whose depressive onset preceded the diagnosis of AD was higher than that for men (34%). Depression was common among AD patients who were screened for these two clinical trials. Prevalence of depression was found to be highest among patients with diagnosis of AD at or before age 65, women, and whites in our AD patient sample.

Cerebral microhemorrhages (MHs) are common in persons with Alzheimer disease (AD), but their exact prevalence is not well established, nor are the epidemiologic factors with which they are associated. Between December 2007 and December 2010, 2010 subjects underwent a brain MRI (including GRE/T2*) as part of screening for participation in two global clinical trials investigating the efficacy and safety of an immunotherapeutic agent for the treatment of mild to moderate AD (ClinicalTrials.gov identifiers NCT00667810 and NCT00676143). Eligible subjects were between age 50 and 88 and met NINCDS-ADRDA criteria for probable AD. Subjects with a history of stroke, recent myocardial infarction, or known MRI abnormality were excluded from the trial. Screening MRIs were centrally read by a group of neuroradiologists. The number of MHs at baseline was quantified. The recursive partitioning and regression tree (rpart) approach was used to identify risk factors for baseline MHs. Logistic regression was employed to build a prediction model for the likelihood of identifying MHs. 24% of subjects had one or more MH at screening. Prevalence of baseline MHs increased with age (age 50-60 = 15%, 61-70 = 20%, 71-80 = 26%, 81-88= 36%). MHs were more common in male than female subjects (27% vs 22%), in Asian than white subjects (26% vs 23%), and in Apolipoprotein E e4 (ApoE4) carriers than non-carriers (26% vs 21%). Among ApoE4 carriers, the rate of MHs was higher in homozygotes than in heterozygotes (32% vs 25%). There was no notable increase in MH prevalence in subjects with a history of cardiovascular disorders, but the rate appeared higher in subjects taking aspirin (27% treated vs 21% non-treated). A multivariate rpart analysis showed that age was a major risk factor for MHs. In older subjects, ApoE4 status and gender were predictive of MHs. There was a trend suggesting that higher field strength scanners (2-3T) were more likely to detect MHs than 1-1.5T MRIs. The presentation will include an updated data cut. Subjects screening for participation in AD clinical trials have a high prevalence of baseline MHs. MH prevalence varies with age, ApoE4 status, and gender.

Understanding both the placebo response and the natural disease progression in Alzheimer's disease clinical trials is crucial to designing appropriate long term trials. It is sometimes difficult to differentiate the placebo effect and underlying disease progression in longitudinal studies, sometimes resulting in misinterpretation of the treatment effect. The magnitude of placebo effect may also differ by several factors, such as baseline severity, background therapy, or other study-specific factors. In order to evaluate these factors, the placebo response in the Alzheimer's Disease Assessment Scale cognitive subscale (ADAS-cog) was quantitatively evaluated across studies. In the previous analysis, we conducted systematic search in public data sources from 1990 to 2008 for all available AChE inhibitors studies and have developed a model to describe the longitudinal response in ADAS-cog. The disease progression in mild to moderate AD patients was estimated as 5.5 points per year. In this analysis, placebo response component from this model was used to estimate the historical placebo response, which provided a robust understanding of the placebo response. The placebo response from double blind studies (study level) in mild to moderate AD patients was compared by covariates of interests. The patient-level control arm data (approximately 3500 patients information) from clinical trials in AD from the Coalition Against Major Disease (CAMD) database (http://www.c-path.org) were also used, and longitudinal placebo response was compared. The magnitude of placebo response was similar between patients who were and were not on stable background AD therapy. The preliminary results also indicate that the placebo response was similar between Phase 2 and Phase 3 studies, and were within the range of the historical placebo response derived from the disease progression model. Recent failed Phase 3 studies in Alzheimer's disease may not due to the lack of enough cognitive decline/disease progression in the placebo group. Understanding placebo effect is critical to predict and evaluate the outcome of the clinical trial.

Slope analysis has been recommended by certain regulatory agencies to support characterization of disease-modifying effect of an experimental drug. Two issues are to be resolved to conduct a valid slope analysis: 1) linearity assumption of responses, and 2) missing data due to dropouts. A new statistical slope analysis (divergence of effect analysis) will be considered. This approach is to confirm increasing separation of response differences over time between treated and placebo group. No linearity assumption of response in each group is needed. Missing data issue can also be addressed via sensitivity analyses. Clinical trials will be simulated to evaluate the performance of this approach and other approaches, in characterizing the disease-modifying effect in AD trials. The pros and cons of different approaches will be summarized. It is confirmed that the new statistical approach is free of assumption that the response in each group is linear, and is robust to missing data mechanism in general. This new statistical approach of slope analysis (divergence of effect) should be recommended to use in AD trials.

It has been suggested that the ADAS-Cog is not uniformly sensitive to measuring cognitive decline in Alzheimer disease across mild and moderate patients. As a result, the NTB (neuropsychological test battery) was introduced as an alternative which measures cognitive decline in both mild and moderate patients equally well (Harrison et al, Arch Neurol 64:1323-9, 2007). The psychometric properties of NTB are examined in two randomized, double-blind trials; one 12-week placebo- and active-controlled (~ 350 patients) and the other 24-week placebo-controlled (~ 330). The test-retest reliability, ratio of within subject variability to between subject variability, and internal consistency of NTB in addition to correlation of NTB with other cognitive and functional measures are evaluated using data from the placebo patients. The sensitivities of NTB and ADAS-Cog are assessed as a function of baseline MMSE (Mini-Mental State Examination) scores using data from the active and the placebo arms. It is confirmed that the NTB has good psychometric properties. Although the NTB and ADAS-Cog are equally sensitive in detecting change in the overall population, differences are observed when stratified by baseline MMSE scores. While the NTB is more sensitive than ADAS-Cog among the mild patients, the reverse is observed among the moderate patients. The NTB has good psychometric properties. Based on two studies of relatively short duration, both the NTB and ADAS-Cog are shown to be good instruments for detecting drug effects on cognitive decline among AD patients. The sensitivity of these instruments needs to be evaluated in validation studies of longer duration.