BMJ
Confidential: For Review Only
Impact o f searching clinical trial registries in systematic reviews of pharmaceutical treatments
Journal: BMJ
Manuscript ID BMJ.2016.034967.R1
Article Type: Research methods and reporting
BMJ Journal: BMJ
Date Submitted by the Author: 02-Nov-2016
Complete List of Authors: Gauthier, Marie; INSERM, Yavchitz, Amélie; INSERM U1153, Centre de recherche Epidémiologies et Biostatistique; Cochrane France Ravaud, Philippe; Inserm U1153; Cochrane France Perrodeau, Elodie; CHU Hôtel Dieu, Centre d’épidémiologie clinique Boutron, Isabelle ; Center of Research in Epidemilogy and Statistics Sorbone Paris, CRESS UMR 1153, METHODS team; Cochrane France
systematic review, registration, meta-analysis, publication bias, clinical Keywords: trial
https://mc.manuscriptcentral.com/bmj Page 1 of 70 BMJ
1 2 3 1 4 5 2 .Impact of searching clinical trial registries in systematic 6 7 8 3 Confidential:reviews of pharmaceutical For treatments. Review Only 9 10 11 4 12 13 5 Marie Baudard (1,2), Amélie Yavchitz (1,2,3,4), Philippe Ravaud (1,2,3,4), Elodie Perrodeau 14 15 6 (1,2), Isabelle Boutron (1,2,3,4) 16 17 7 1 Paris Descartes University, Paris, France 18 19 2 20 8 INSERM, UMR 1153 Epidemiology and Biostatistics Sorbonne Paris Cité Center (CRESS), 21 9 METHODS team, Paris, France 22 23 10 3 Centre d’Épidémiologie Clinique, Hôpital Hôtel Dieu, Assistance Publique des Hôpitaux de 24 11 Paris, Paris, France 25 26 12 4 Cochrane France, Paris, France 27 28 29 13 30 31 14 32 15 33 34 16 Corresponding author : 35 36 17 Amélie Yavchitz 37 38 18 Centre d’Epidémiologie Clinique, Hôpital Hôtel Dieu 39 40 19 1, Place du Parvis Notre Dame, 41 20 75181 PARIS Cedex 4 42 43 21 Tel 33 (0) 1 42 34 78 33 44 45 22 Fax 33 (0) 1 42 34 87 90 46 23 [email protected] 47 48 24 49 25 Word count abstract 406 50 26 Word count full text 3672 51 27 52 53 54 55 56 57 58 59 60 1 https://mc.manuscriptcentral.com/bmj BMJ Page 2 of 70
1 2 3 1 ABSTRACT 4 5 6 2 Objective To evaluate the impact of searching clinical trial registries on including the results 7 8 3 Confidential:of additional randomized controlled trials For (RCTs) i n Reviewsystematic reviews (ie, eligibleOnly RCTs not 9 10 4 originally included in the systematic review classified as completed or terminated in the 11 12 13 5 registry). 14 15 6 Design 1) We identified systematic reviews of RCTs assessing pharmaceutical treatments 16 17 7 published between June 2014 and January 2015. 2) For all systematic reviews that did not 18 19 8 report a trial registry search but reported the information to perform it, we searched the World 20 21 9 Health Organization International Trials Registry Platform (WHO ICTRP Search Portal) for 22 23 24 10 completed or terminated RCTs not originally included in the systematic review. 3) We 25 26 11 searched the results for all completed or terminated RCTs identified and 4) performed meta 27 28 12 analyses when additional data were retrieved. 29 30 13 Data source MEDLINE and WHO ICTRP Search Portal 31 32 33 14 Data extraction For each systematic review, two researchers independently extracted the 34 35 15 outcomes analyzed, the number of patients included and the treatment effect estimated. For 36 37 16 each RCT identified, two researchers independently determined whether the results were 38 39 17 available (ie, posted, published or available on the sponsor website) and extracted the data. 40 41 18 Results Among 223 selected systematic reviews, 116 (52%) did not report a search of trial 42 43 44 19 registries; 21 of these did not report the information to perform the search (key words, search 45 46 20 date). We performed the search for 95 systematic reviews; for 54/95 (57%), we found no 47 48 21 additional RCTs and for 41/95 (43%) we identified 122 additional RCTs. The search allowed 49 50 22 for increasing the number of patients by more than 10% in 19 systematic reviews, 20% in 10, 51 52 53 23 30% in 7, and 50% in 4. Moreover, 63 RCTs had results available; the results for 45 could be 54 55 24 included in a meta analysis. We reanalyzed 14 systematic reviews including 45 RCTs. The 56 57 25 weight of the additional RCTs in the re calculated meta analyses ranged from 0% to 58% and 58 59 60 2 https://mc.manuscriptcentral.com/bmj Page 3 of 70 BMJ
1 2 3 1 was greater than 10% in 5 of 14 systematic reviews, 20% in 3, and 50% in 1. The change in 4 5 2 summary statistics ranged from 0% to 29% and was greater than 10% for 5 of 14 systematic 6 7 3 reviews and greater than 20% for 2. However, none of the changes to summary effect 8 Confidential: For Review Only 9 10 4 estimates led to a qualitative change in the interpretation of the results once the new trials 11 12 5 were added. 13 14 15 6 Conclusions Trial registries are an important source for identifying additional RCTs. The 16 17 7 additional number of RCTs and patients included if a search were performed varied across 18 19 8 systematic reviews. 20 21 9 22 23 24 10 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 3 https://mc.manuscriptcentral.com/bmj BMJ Page 4 of 70
1 2 3 1 INTRODUCTION 4 5 2 Systematic reviews are considered to provide the highest level of evidence[1,2]. They are 6 7 3 widely used by clinical practice guideline developers, granting health agencies and journal 8 Confidential: For Review Only 9 10 4 editors [3–6] . A major challenge of systematic reviews is to identify all relevant randomized 11 12 5 controlled trials (RCTs), whatever their publication status[7–13]. Indeed, results for half of 13 14 6 RCTs are never published and the publication status is affected by the nature and direction of 15 16 7 results, which may bias the results of the systematic review[14]. In some cases, the 17 18 8 importance of unpublished trials can be considerable; for example, the addition of 19 20 21 9 unpublished data in the updated Cochrane review assessing the efficacy of neuraminidase 22 23 10 inhibitor for influenza modified the conclusion[15,16]. Initiatives aimed at reducing 24 25 11 publication bias include the trial registration policy initiated by the International Committee 26 27 12 of Medical Journal Editors (ICMJE) in 2005[17]. In 2007, the US Food and Drug 28 29 30 13 Administration Amendments Act (FDAAA) required the posting of clinical trial results at 31 32 14 ClinicalTrials.gov no later than one year after the date of final collection of data for the pre 33 34 15 specified primary outcome, for all phase II to IV trials of drugs, biologic treatments and 35 36 16 devices having at least one site in the United States[18,19]. The research community has 37 38 17 embraced this policy, and there was a marked increase in trial registration around the time of 39 40 41 18 implementation of the ICMJE policy[20]. In April 2016, about 90,000 completed 42 43 19 experimental studies were registered at ClinicalTrials.gov (the largest registry), and 16,500 44 45 20 have results posted. 46 47 21 When performing systematic reviews, the search of trial registries is now considered an 48 49 50 22 essential tool [3,21–23]. Previous studies showed that clinical trial registry search is not 51 52 23 systematically reported by authors of systematic reviews [24–26], but to our knowledge, none 53 54 24 had systematically performed a trial registry search to quantify the impact of searching trial 55 56 25 registries. The objectives of this study were to 1) describe whether and how clinical trial 57 58 59 60 4 https://mc.manuscriptcentral.com/bmj Page 5 of 70 BMJ
1 2 3 1 registries were searched in published systematic reviews of pharmaceutical treatments and 2) 4 5 2 evaluate the impact of searching trial registries on the identification of additional RCTs (ie, 6 7 3 eligible completed or terminated RCTs not included in the systematic review). For this 8 Confidential: For Review Only 9 10 4 purpose, we identified a sample of systematic reviews of RCTs assessing pharmaceutical 11 12 5 treatments indexed in PubMed and recorded whether a search of clinical trial registries was 13 14 6 performed. Then, for all systematic reviews not reporting a search in clinical trial registries 15 16 7 but reporting the information to perform it, we systematically performed the search. 17 18 8 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 5 https://mc.manuscriptcentral.com/bmj BMJ Page 6 of 70
1 2 3 1 METHODS 4 5 6 2 Identification of systematic reviews 7 8 Confidential: For Review Only 9 3 Search strategy 10 11 4 We systematically searched MEDLINE via PubMed for all systematic reviews of RCTs 12 13 5 assessing pharmaceutical treatments that were published in English between June 1, 2014 and 14 15 16 6 January 31, 2015 by searching for “Meta Analysis[ptyp] AND ("2014/06/01"[PDAT]: 17 18 7 "2015/01/31"[PDAT]) AND English[lang] appearing in the title, abstract or keywords (date 19 20 8 search: March 16, 2015) 21 22 9 23 24 10 Inclusion and exclusion criteria 25 26 27 11 One researcher screened all titles and abstracts of citations retrieved to identify all reports of 28 29 12 systematic reviews of RCTs with at least one meta analysis including at least two RCTs and 30 31 13 assessing pharmaceutical treatment (ie, drug, health related biological product or biologic 32 33 14 supplementation). We excluded updates of previously published systematic reviews and 34 35 36 15 systematic reviews of diagnostic test accuracy, prognosis, economics evaluations, genetics, 37 38 16 non RCT studies, network meta analyses, indirect comparison meta analysis as well as 39 40 17 individual patient data meta analyses. The full text of potentially relevant citations was 41 42 18 obtained. As a quality control procedure, another researcher independently screened 20% of 43 44 19 the citations and confirmed the eligibility of all systematic reviews included. Discrepancies 45 46 47 20 were discussed to reach consensus. 48 49 21 50 51 22 Data extraction 52 53 54 23 From the published reports and supplementary appendices when available, two researchers 55 56 24 independently recorded the following: 57 58 59 60 6 https://mc.manuscriptcentral.com/bmj Page 7 of 70 BMJ
1 2 3 1 1) the general characteristics of the systematic review (ie, the type of journal: general 4 5 2 medical journal, specialty journal or Cochrane review), the funding source (none 6 7 3 profit, for profit, not reported or unclear), and the number of RCTs and participants 8 Confidential: For Review Only 9 10 4 included in the systematic review. 11 12 5 2) the reporting of the clinical trial registry search (ie, whether a search in a clinical trial 13 14 6 registry was reported, the name and type of registries searched, and whether the results 15 16 7 of the search were reported (the number and identification of RCTs identified from the 17 18 8 clinical trial registry search). 19 20 21 9 Any disagreement was resolved by discussion and consensus. 22 23 10 24 25 11 Impact of searching clinical trial registries 26 27 12 For each systematic review that did not report a search in clinical trial registries, we 28 29 30 13 systematically performed a search reproducing the conditions of the original search reported 31 32 14 in the systematic review, particularly taking into account the date of the search and the 33 34 15 inclusion criteria of the systematic review. 35 36 37 16 We screened the retrieved records and identified all eligible RCTs classified in the registries 38 39 17 as completed (i.e., RCTs that ended normally) or terminated (i.e., RCTs that stopped 40 41 18 recruiting or enrolling participants early and would not start again) that were not initially 42 43 44 19 included in the systematic reviews 45 46 47 20 Search strategy 48 49 21 Our search strategy followed the same search and selection process described by the authors 50 51 22 of the published systematic reviews. 52 53 23 1) From the selected full text articles and all available supplementary materials, we 54 55 56 24 systematically recorded the search terms related to the condition and interventions 57 58 59 60 7 https://mc.manuscriptcentral.com/bmj BMJ Page 8 of 70
1 2 3 1 used by authors and the date of last electronic search. Systematic reviews that did not 4 5 2 provide search terms or the date of search were excluded from this analysis. 6 7 3 2) We searched the World Health Organization International Trials Registry Platform 8 Confidential: For Review Only 9 10 4 (WHO ICTRP Search Portal). We chose this portal (i.e., a portal provides access to a 11 12 5 central database containing the trial registration data sets provided by several 13 14 6 registries) because it includes 16 national and international primary registries 15 16 7 including ClinicalTrials.gov. In the advanced search window of the WHO ICTRP 17 18 8 Search Portal ( http://apps.who.int/trialsearch/ ), we entered the search terms recorded 19 20 21 9 in the “condition” and “intervention” fields with Boolean operators. We chose “all” in 22 23 10 the “recruitment status” field and “Search for clinical trials in children” when 24 25 11 appropriate. Details of the search strategies and keywords for each systematic review 26 27 12 are available in Appendix 1. 28 29 30 13 31 32 14 Identification of completed or terminated RCTs 33 34 15 For each search, we downloaded all the citations retrieved and identified all studies with a 35 36 16 recruitment status recorded as “completed” or “terminated” registered before the date of the 37 38 17 last search reported in the systematic review and 39 40 41 18 For each systematic review, two researchers independently screened the records retrieved and 42 43 44 19 selected all completed or terminated RCTs not already included in the systematic review that 45 46 20 fulfilled the systematic review eligibility criteria in terms of participants, interventions, and 47 48 21 comparator. We systematically verified in the history or archives of the registry that the 49 50 22 recruitment status was recorded as “completed” or “terminated” before the date of the search 51 52 53 23 (Appendix 2). Any disagreements were resolved by consensus. A third researcher screened all 54 55 24 selected records to confirm their inclusion. 56 57 25 58 59 60 8 https://mc.manuscriptcentral.com/bmj Page 9 of 70 BMJ
1 2 3 1 Availability of RCT results 4 5 2 For each selected RCT, two researchers independently determined whether the trial results 6 7 3 were available (ie, posted, published or available on the sponsor website). We searched for 1) 8 Confidential: For Review Only 9 10 4 results posted on clinical trial registries and 2) publications referenced on the trial registry and 11 12 5 3) performed an electronic search of PubMed and Google and searched the sponsor website. 13 14 6 All trials with results available were screened, and we selected only trials for which the results 15 16 7 became available before the last electronic search of the systematic review. 17 18 8 19 20 21 9 Inclusion of the RCT results in meta analyses 22 23 10 We recorded the number of meta analyses reported in the systematic review, the number of 24 25 11 meta analyses that could include the additional RCTs, and the number of meta analyses for 26 27 12 which all the RCTs identified had available results and could be included in the meta analysis. 28 29 30 13 Finally, we determined the impact of including the RCTs on treatment effect estimates. For 31 32 33 14 this purpose, we used an algorithm to select one meta analysis in which at least one RCT with 34 35 15 results available could be included. 36 37 38 16 We proceeded as follows: 39 40 41 17 1. For each systematic review, we recorded all the outcomes of the meta analysis 42 43 18 reported in the systematic review report. 44 45 46 19 2. For each eligible RCT with results available, we determined whether the RCT could be 47 48 20 included in the meta analyses previously recorded, i.e., the RCT reports included the 49 50 51 21 following: 52 53 54 22 a. For continuous outcomes: sample size, mean and one measure of dispersion 55 56 23 (standard deviation, standard error or confidence interval) by group. Standard 57 58 24 errors and confidence intervals were converted in standard deviations to 59 60 9 https://mc.manuscriptcentral.com/bmj BMJ Page 10 of 70
1 2 3 1 perform the meta analyses. When results were given for separate subgroups, 4 5 2 we pooled the results, the pooled sample size being the sum of the subgroup 6 7 3 sample sizes, the pooled mean being the weighted mean of the subgroups and 8 Confidential: For Review Only 9 10 4 the pooled standard deviation combined[3]. 11 12 13 5 b. For binary outcomes: sample size and number of events by group. 14 15 16 6 c. For time to event outcomes: hazard ratio and 95% confidence interval or 17 18 7 median survival times and confidence intervals by group[27]. 19 20 21 8 3. When RCT(s) could be included in several meta analyses, we selected only one meta 22 23 9 analysis according to the following order of outcomes analyzed: 1) the primary 24 25 10 efficacy outcome of the systematic review, 2) the primary safety outcome, 3) the 26 27 28 11 patient important outcome such as mortality, quality of life or morbidity outcome. If 29 30 12 several of these outcomes could be used to include new RCT(s), we selected the first 31 32 13 meta analysis reported. If none of these outcomes could be used to include a new 33 34 14 RCT, we selected the first meta analysis reported. 35 36 37 15 For each meta analysis selected, we extracted from the RCTs identified the outcome data (ie, 38 39 16 number of events and number of patients in each group, means, standard deviations, etc). 40 41 42 17 When the outcome data were available in several sources, we considered in priority the data 43 44 18 reported 1) in the registry, 2) in a published report and 3) on the sponsor website. 45 46 47 19 48 49 20 Data analysis 50 51 21 Statistical analyses involved use of R v3.1.0 ( http://www.R project.org , the R foundation for 52 53 54 22 statistical Computing, Vienna, Austria). Qualitative variables are represented by percentages. 55 56 23 Quantitative variables are represented by medians (quartile 1–quartile 3 [Q1–Q3]). In a post 57 58 24 hoc analysis, we used a chi square test to compare the proportion of reviews reporting a trial 59 60 10 https://mc.manuscriptcentral.com/bmj Page 11 of 70 BMJ
1 2 3 1 registry search according to the type of systematic review (Cochrane vs non Cochrane) and 4 5 2 funding source (not for profit funding or not funded vs for profit funding, funding not 6 7 3 reported or unclear). 8 Confidential: For Review Only 9 10 4 For the meta analysis selected for recalculation (one per selected systematic review), we 11 12 13 5 calculated summary statistics (risk ratios, odds ratios, hazard ratios, mean differences or 14 2 15 6 standardized mean differences) and the I statistic (measure of heterogeneity) with and 16 17 7 without trials retrieved by a trial registry search. We reported the magnitude of the change in 18 19 8 the result of the meta analysis as a percentage change in the summary statistic after including 20 21 9 the RCTs retrieved. We re analysed the published meta analyses by using the same statistical 22 23 24 10 method (Peto, Mantel Haenszel, inverse variance), analysis model (fixed v random effects), 25 26 11 and measure of effect (risk ratio, odds ratio, weighted mean difference) used by the original 27 28 12 authors. For all meta analyses, we assessed heterogeneity by calculating the I² statistic and τ² 29 30 13 (DerSimonian Laird estimate). 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 11 https://mc.manuscriptcentral.com/bmj BMJ Page 12 of 70
1 2 3 1 RESULTS 4 5 6 2 Identification and characteristics of reports 7 8 Confidential: For Review Only 9 3 Among the 2,249 citations retrieved, we included 223 reports of systematic reviews with 10 11 4 meta analyses (fig 1). The characteristics of the included systematic reviews are in table 1. 12 13 14 5 One third (35%) were Cochrane reviews; the median [Q1 Q3] number of RCTs included in 15 16 6 systematic reviews was 10 [6 18] and the median [Q1 Q3] number of patients was 1,594 17 18 7 [614 5027]. 19 20 21 22 8 Reporting of clinical trial registry search in systematic reviews 23 24 9 Among the 223 systematic review reports included, 107 (48%) reported searching at least one 25 26 10 clinical trial registry: 48 of these (45%) reported searching only individual registries, 11 27 28 29 11 (10%) only portals and 44 (41%) a combination of individual registries and portals. The portal 30 31 12 and individual register most frequently searched were the WHO ICTRP Search Portal 32 33 13 (n=53/107, 50%) and ClinicalTrials.gov (n=89/107, 83%), and for 40 studies (37%), both 34 35 14 were searched. In only 47 of the 107 (21%) reports were the results of the clinical trial 36 37 38 15 registry search clearly described (i.e., with a description of the number and identification of 39 40 16 RCTs found from the search) (fig 1, table 1): 16 of these 47 reviews (34%) did not retrieve 41 42 17 any eligible RCTs, 11 (23%) retrieved only ongoing studies, 13 (28%) retrieved at least one 43 44 18 completed or terminated RCT without results available and 7 (15%) retrieved at least one 45 46 19 completed or terminated RCT with results identified. Of these last 7, 3 included RCTs in at 47 48 49 20 least one meta analysis. 50 51 52 21 A search of a trial registry was more frequent in Cochrane than non Cochrane reviews [65/77 53 54 22 (84%) vs 42/146 (29%), p<0.001] and not for profit funding or no funding than for profit 55 56 23 funding, funding not reported or unclear [79/139 (57%) vs 28/84 (33%), p<0.001]. 57 58 59 60 12 https://mc.manuscriptcentral.com/bmj Page 13 of 70 BMJ
1 2 3 1 Impact of searching clinical trial registries 4 5 2 Identification of completed or terminated RCTs 6 7 3 Among the 116 systematic reviews not reporting a search in trial registries, for 21 (18%), we 8 Confidential: For Review Only 9 10 4 were not able to perform the clinical trial registry search because the search date or the 11 12 5 keywords were not reported. Therefore, we performed the search for 95 systematic reviews. 13 14 6 Among the 15,282 records screened (median [Q1 Q3] records screened for each systematic 15 16 7 review = 23 [6 150]), we identified 122 eligible RCTs terminated or completed (involving 17 18 8 52,743 patients) not originally included in the systematic review. Among the 122 RCTS, 104 19 20 21 9 (85%) were classified as completed and 18 (15%) as terminated. Among the 18 RCTs 22 23 10 classified terminated, 3 had results available and were included in meta analyses, 2 were 24 25 11 stopped early because of adverse events and 1 was stopped early because of futility. The 26 27 12 remaining 15 RCTs had no results available and no information on the reason for stopping 28 29 30 13 early. 31 32 33 14 Availability of RCT results 34 35 15 Overall, the trial registry searches identified at least one eligible RCT for 41 of 95 (43%) 36 37 16 systematic reviews, with a median [Q1 Q3] of 9% [4 18] additional patients per systematic 38 39 17 review (fig 2, table 2, Appendix 3). Among these 41 systematic reviews with additional RCTs 40 41 18 identified, the number of patients included was increased by 10% in 19, 20% in 10, 30% in 7, 42 43 44 19 and 50% in 4. 45 46 20 We identified results for 63 of 122 RCTs (52%) involving 42,202 patients, and 45 of 122 47 48 21 (37%) involving 21,358 patients could be included in the quantitative analyses (i.e., reported 49 50 22 sufficient data to be included in at least one meta analysis of the systematic review). The 18 51 52 53 23 remaining RCTs with results could not contribute to the quantitative analysis because of 54 55 24 differences in definition or metrics used between the outcome reported in the RCT and the 56 57 25 outcome of the systematic review or outcome reporting bias. 58 59 60 13 https://mc.manuscriptcentral.com/bmj BMJ Page 14 of 70
1 2 3 1 The results of the RCTs identified were 1) posted (n=41/63, 65%); 2) published as identified 4 5 2 by a reference reported on the registry (n=21/63, 33%) or from a complementary search 6 7 3 (n=10/63, 16%); or 3) were available on the company’s Web site (n=31/63, 49%). The results 8 Confidential: For Review Only 9 10 4 were available in one (n= 29/63, 46%), two (n=27/63, 43%) or three sources (n= 7/63, 11%). 11 12 5 For 14 systematic reviews, the trial registry searches allowed for identifying RCTs with 13 14 6 results (n= 45) that could contribute to the quantitative analysis. Among the 73 meta analyses 15 16 7 reported in these 14 systematic reviews; the search in trial registries retrieved additional 17 18 8 results that could be included in 59 meta analyses. Overall, 31 of 59 (53%) meta analyses 19 20 21 9 were considered complete (ie, all the RCTs identified had available results and could be 22 23 10 included in the meta analysis). 24 25 11 26 27 12 Inclusion of the RCT results in meta analyses 28 29 30 13 Finally, we recalculated the effect estimates for the selected meta analyses from the 14 31 32 14 systematic reviews including RCTs that could contribute to the quantitative analysis. Among 33 34 15 the 14 meta analyses selected, 6 involved safety outcomes and 8 efficacy outcomes. In the 35 36 16 meta analysis without additional RCTs, results for 12 of 14 outcomes significantly favoured 37 38 17 the experimental treatment and results for 2 did not differ from the comparator. 39 40 41 18 The weight of the eligible RCTs included ranged from 0% to 58% and was greater than 10% 42 43 44 19 for 5 of 14 systematic reviews, 20% for 3, and 50% for 1. The change in summary statistics 45 46 20 ranged from 0% to 29% and was greater than 10% for 5 of 14 systematic reviews and greater 47 48 21 than 20% for 2. For example, in the meta analysis with a 29% change in summary effect, the 49 50 22 mean difference changed from 0.35 [ 0.51; 0.19] to 0.45 [ 0.55; 0.36], for a larger effect 51 52 53 23 after inclusion of the new RCTs. However, including the RCTs identified by a trial registry 54 55 24 search did not change the statistical significance or direction of the results. Detailed 56 57 25 descriptions of the 14 meta analyses are provided in table 2. 58 59 26 60 14 https://mc.manuscriptcentral.com/bmj Page 15 of 70 BMJ
1 2 3 1 DISCUSSION 4 5 6 2 Summary of findings 7 8 Confidential: For Review Only 9 3 Despite recommendations [23], about half of the published systematic reviews performed a 10 11 4 trial registry search and only one fifth reported the results of the search. When we performed 12 13 14 5 the search, we identified additional studies for 43% of the systematic reviews. However, 15 16 6 because of the lack of data availability, data for half of the eligible RCTs retrieved could not 17 18 7 be included in systematic reviews. We re analyzed 14 meta analyses to include data from 19 20 8 RCTs retrieved by the trial registry search. The weight of the eligible RCTs included ranged 21 22 9 from 0% to 58% and the change in summary statistics from 0% to 29%. The addition of data 23 24 25 10 from registries mainly adds to the precision of summary estimates, but none of the changes 26 27 11 led to a qualitative change in the interpretation of the results once the new trials were added. 28 29 30 12 Comparison with other studies 31 32 33 13 Our results are consistent with other studies showing that the search for unpublished trial data 34 35 14 is still often lacking in systematic reviews[24–26,28,29] as in a random sample of 300 recent 36 37 38 15 systematic reviews indexed in MEDLINE in February 2014 of which 19% reported searching 39 40 16 trial registries[30]. A previous study by Hart in 2012 aimed to re analyze meta analyses by 41 42 17 adding unpublished trial outcome data obtained from the US Food and Drug Administration 43 44 18 (US FDA) to published meta analyses[31]. The study documented that the addition of 45 46 19 unpublished trial data obtained from the US FDA could change the magnitude of the effect 47 48 49 20 size or in a few cases the statistical significance of meta analyses. Moreover, in a systematic 50 51 21 review by Golder in 2016, aiming at quantifying the impact of the underreporting of adverse 52 53 22 events in systematic reviews showed that the inclusion of unpublished data may reduce the 54 55 23 imprecision of pooled effect estimates in meta analysis of adverse events[32]. However, to 56 57 58 59 60 15 https://mc.manuscriptcentral.com/bmj BMJ Page 16 of 70
1 2 3 1 our knowledge, the impact of searching trial registries in terms of identifying trials and their 4 5 2 inclusion in the analysis when results are available has never been evaluated. 6 7 3 8 Confidential: For Review Only 9 10 11 4 Limitations 12 13 14 5 Our study has some limitations. First, we searched only the WHO ICTRP Search Portal using 15 16 6 the keywords reported by authors for their electronic search. Consequently, we cannot claim 17 18 7 that we identified all RCTs. However, this portal brings together 16 national and international 19 20 8 primary registries including ClinicalTrials.gov. Furthermore, in a previous study, the overlap 21 22 9 between ICTRP and ClinicalTrials.gov was good, because all records identified in 23 24 25 10 ClinicalTrials.gov were also identified in ICTRP[33]. Second, we did not account for 26 27 11 eligibility criteria related to trial quality. The quality assessment of data recorded from trial 28 29 12 registries is difficult and some trials could secondarily be excluded because of insufficient 30 31 13 quality. Third, we did not attempt to contact investigators of the unpublished trials to obtain 32 33 34 14 results. In fact, we aimed to reproduce the condition the authors encountered and it would not 35 36 15 be appropriate to ask authors for results after such a delay. Further, we did not search for 37 38 16 additional data presented in conference abstracts or searched US FDA and EMA websites. 39 40 17 Therefore, the number of systematic reviews with trials identified by a search of clinical trial 41 42 18 registries and the amount of data from RCTs retrieved from clinical trial registries may be 43 44 45 19 underestimated. Fourth, we choose to include only one meta analysis per systematic review to 46 47 20 make the workload manageable. Finally, we focused on only systematic reviews of 48 49 21 pharmaceutical treatment and cannot extrapolate to non pharmaceutical treatments because 50 51 22 the regulation for trial registration and posting of results is less stringent with these 52 53 54 23 treatments. 55 56 57 58 59 60 16 https://mc.manuscriptcentral.com/bmj Page 17 of 70 BMJ
1 2 3 1 Implications for clinicians and policy makers 4 5 6 2 Clinical trial registries have been developed and their use enforced by editors and policy 7 8 3 Confidential:makers to reduce waste in research andFor publication Review bias. They have been Only considered an 9 10 4 important step toward more transparency and increasing research value. 11 12 13 5 Searching clinical trial registries is recommended when performing systematic reviews. In our 14 15 6 study, the addition of new RCTs in meta analyses affected treatment effect estimates but did 16 17 18 7 not change the statistical significance of the results or the direction of the treatment effect, 19 20 8 although it increased precision. 21 22 23 9 Nevertheless, searching clinical trials registers remains an essential recommendation for the 24 25 10 conduct of systematic reviews and should be enforced. In fact, the objective of systematic 26 27 11 reviews is to collate all empirical evidence [3]. However, overall, results for only about half 28 29 12 of clinical trials are published, and searching only electronic bibliographic databases gives 30 31 32 13 access to only the “tip of the iceberg” [24–26]. In our study, searching clinical trials registries 33 34 14 allowed for finding new evidence for almost half of the systematic reviews (41/95), and this 35 36 15 new evidence was usable in at least one meta analysis in one third of these systematic reviews 37 38 16 (14/41). Finally, searching trial registries in general represented a low burden. The median 39 40 41 17 (Q1 Q3) number of records to screen by systematic review was low (23 [6 150]). The results 42 43 18 for 41 of 63 trials were posted at ClinicalTrials.gov and therefore immediately available. 44 45 19 Furthermore, a previous study showed that the reporting of results was more complete at 46 47 20 ClinicalTrials.gov than in published reports[34]. Of course, one important limitation of this 48 49 21 search is the lack of availability of the results for completed trials and the low level of details 50 51 52 22 on the methodologic quality recorded in the registries. Some initiatives to facilitate the access 53 54 23 to clinical trial results, such as the 2007 FDAAA, which requires the posting of clinical trial 55 56 24 results[18] or pharmaceutical company policies[35], have been implemented. Some 57 58 59 60 17 https://mc.manuscriptcentral.com/bmj BMJ Page 18 of 70
1 2 3 1 researchers have developed an intervention to improve posting, such as emailing a reminder 4 5 2 of the FDAAA 801 requirement to responsible parties[36]; other interventions are necessary. 6 7 3 Recently, the new rules of trial registration at ClinicalTrials.gov requires submission of a full 8 Confidential: For Review Only 9 10 4 protocol and statistical analysis plan at the same time as submission of results [37]. Registries 11 12 5 could be an even more important source of results in the future.” 13 14 15 6 CONCLUSION 16 17 18 7 Searching clinical trial registries is essential for identifying additional trials that could 19 20 8 increase the value of systematic reviews. However, the lack of availability of RCT results 21 22 9 limits the value of the search. Trial registry searching should be promoted and enforced, as 23 24 25 10 should the posting of trial results. 26 27 28 11 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 18 https://mc.manuscriptcentral.com/bmj Page 19 of 70 BMJ
1 2 3 1 Acknowledgements 4 2 We thank Laura Smales (BioMedEditing, Toronto, Canada) for English language 5 3 proofreading. We thank Carolina Riveros for data extraction. 6 4 7 5 8 Confidential: For Review Only 9 6 Contributions 10 7 Conceived, designed and experiments: MB, AY, IB, PR, Wrote the first draft: 11 8 MB, IB, AY Contributed to the writing of the manuscript: AY, IB, EP, PR 12 9 Data analyses MB, AY, EP and IB. 13 10 14 11 15 16 12 Transparency declaration 17 13 AY affirms that this manuscript is an honest, accurate, and transparent account of the study 18 14 being reported; that no important aspects of the study have been omitted; and that any 19 15 discrepancies from the study as planned have been explained. 20 21 16 AY and IB had access to all of the data in the study and take responsibility for the integrity of 22 23 17 the data and the accuracy of the data analysis. 24 25 18 26 27 19 Data sharing 28 20 All data from this study—including literature searches, additional explanatory material, and 29 30 21 data extraction forms—are available on request. 31 32 22 33 34 23 Conflicts of interest 35 24 None of the authors have conflicts of interest to declare. 36 37 38 25 39 26 Financial support 40 27 This study received no funding. 41 28 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 19 https://mc.manuscriptcentral.com/bmj BMJ Page 20 of 70
1 2 3 4 1 TABLES 5 6 7 2 Table 1 Characteristics of included systematic reviews and registry searches 8 Confidential: For Review Only 9 10 3 Table 2 Effect of adding randomized controlled trials retrieved from clinical trial registries on 11 12 4 meta analyses 13 14 15 16 5 17 6 FIGURES 18 19 20 7 Fig 1 Study flow diagram 21 22 8 Fig 2 Identification of trials by searching trials registries 23 24 25 9 26 27 28 10 APPENDIX 29 30 31 32 11 Appendix 1 Verification of the recruitment status according to the registry 33 34 35 12 Appendix 2 Keywords, date of search and detection in the WHO ICTRP Search Portal 36 37 38 13 Appendix 3 Systematic reviews for which data could not be added to the meta analysis 39 40 41 14 Appendix 4 Impact of trial registry searches on summary statistics 42 43 15 44 16 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 20 https://mc.manuscriptcentral.com/bmj Page 21 of 70 BMJ
1 2 3 4 5 1 REFERENCES 6 2 7 8 3 Confidential:1 Cook DJ, Mulrow CD, Haynes RB.For Systematic Review Reviews: Synthesis ofOnly Best Evidence 9 10 4 for Clinical Decisions. Ann Intern Med 1997; 126 :376–80. doi:10.7326/0003 4819 126 11 5 5 199703010 00006 12 13 6 2 Bastian H, Glasziou P, Chalmers I. Seventy five trials and eleven systematic reviews a 14 7 day: how will we ever keep up? PLoS Med 2010; 7:e1000326. 15 8 doi:10.1371/journal.pmed.1000326 16 17 9 3 Higgins JPT GS. Cochrane Handbook for Systematic Reviews of Interventions. 18 10 Cochrane Collab 2011; version 5. 19 20 11 4 Higgins JP, Altman DG, Gotzsche PC, et al. The Cochrane Collaboration’s tool for 21 12 assessing risk of bias in randomised trials. BMJ 2011; 343 :d5928. 22 23 13 5 Moher D, Tetzlaff J, Tricco AC, et al. Epidemiology and reporting characteristics of 24 14 systematic reviews. PLoS Med 2007; 4:e78. doi:10.1371/journal.pmed.0040078 25 26 15 6 Ebrahim S, Bance S, Athale A, et al. Meta analyses with industry involvement are 27 16 massively published and report no caveats for antidepressants. J Clin Epidemiol 28 17 2015; 70 :155–63. doi:10.1016/j.jclinepi.2015.08.021 29 30 18 7 Chalmers I, Glasziou P. Avoidable waste in the production and reporting of research 31 19 evidence. Lancet 2009; 374 :86–9. doi:10.1016/s0140 6736(09)60329 9 32 33 20 8 Chan AW, Song F, Vickers A, et al. Increasing value and reducing waste: addressing 34 21 inaccessible research. Lancet 2014; 383 :257–66. doi:10.1016/s0140 6736(13)62296 5 35 36 22 9 Easterbrook PJ, Berlin JA, Gopalan R, et al. Publication bias in clinical research. 37 23 Lancet 1991; 337 :867–72. doi:0140 6736(91)90201 Y [pii] 38 39 24 10 Song F, Parekh S, Hooper L, et al. Dissemination and publication of research findings: 40 25 An updated review of related biases. Health Technol Assess (Rockv) 2010; 14 :1–220. 41 26 doi:10.3310/hta14080 42 43 27 11 Hopewell S, Loudon K, Clarke MJ, et al. Publication bias in clinical trials due to 44 28 statistical significance or direction of trial results. Cochrane Database Syst Rev 45 29 2009;:MR000006. doi:10.1002/14651858.MR000006.pub3 46 47 30 12 Rising K, Bacchetti P, Bero L. Reporting bias in drug trials submitted to the Food and 48 31 Drug Administration: review of publication and presentation. PLoS Med 2008; 5:e217; 49 50 32 discussion e217. doi:08 PLME RA 1177 [pii]10.1371/journal.pmed.0050217 [doi] 51 33 13 Dwan K, Gamble C, Williamson PR, et al. Systematic review of the empirical evidence 52 53 34 of study publication bias and outcome reporting bias an updated review. PLoS One 54 35 2013; 8:e66844. doi:10.1371/journal.pone.0066844 55 56 36 14 Schmucker C, Schell LK, Portalupi S, et al. Extent of non publication in cohorts of 57 37 studies approved by research ethics committees or included in trial registries. PLoS 58 38 One 2014; 9:e114023. doi:10.1371/journal.pone.0114023 59 60 21 https://mc.manuscriptcentral.com/bmj BMJ Page 22 of 70
1 2 3 1 15 Jefferson T, Jones MA, Doshi P, et al. Neuraminidase inhibitors for preventing and 4 2 treating influenza in healthy adults and children. Cochrane database Syst Rev 5 3 2014;:CD008965. doi:10.1002/14651858.CD008965.pub4 6 7 4 16 Jefferson T, Jones M, Doshi P, et al. Neuraminidase inhibitors for preventing and 8 5 Confidential:treating influenza in healthy adults:For systematic Review review and meta analysis. Only BMJ 9 6 2009; 339 :b5106.http://www.ncbi.nlm.nih.gov/pubmed/19995812 (accessed 10 10 7 Oct2016). 11 12 8 17 De Angelis C, Drazen JM, Frizelle FA, et al. Clinical trial registration: a statement 13 9 from the International Committee of Medical Journal Editors. Ann Intern Med 14 15 10 2004; 141 :477. 16 11 18 United States Congress (2007) Food and Drug Administration Amendments Act 17 18 12 (FDAAA) of 2007: public law no. 110–85. Available: 19 13 http://www.gpo.gov/fdsys/pkg/PLAW-110publ85/pdf/PLAW-110publ85.pdf 20 21 14 19 Groves T. Mandatory disclosure of trial results for drugs and devices. BMJ 22 15 2008; 336 :170. doi:10.1136/bmj.39469.465139.80 23 24 16 20 Laine C, Horton R, DeAngelis CD, et al. Clinical trial registration: looking back and 25 17 moving ahead. JAMA 2007; 298 :93–4. doi:10.1001/jama.298.1.jed70037 26 27 18 21 Shea BJ, Grimshaw JM, Wells GA, et al. Development of AMSTAR: a measurement 28 19 tool to assess the methodological quality of systematic reviews. BMC Med Res 29 20 Methodol 2007; 7:10. doi:10.1186/1471 2288 7 10 30 31 21 22 Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for reporting 32 22 systematic reviews and meta analyses of studies that evaluate health care interventions: 33 23 explanation and elaboration. J Clin Epidemiol 2009; 62 :e1 34. 34 24 doi:10.1016/j.jclinepi.2009.06.006 35 36 25 23 Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting items for systematic reviews 37 26 and meta analyses: the PRISMA statement. BMJ 2009; 339 :b2535–b2535. 38 27 doi:10.1136/bmj.b2535 39 40 28 24 Keil LG, Platts Mills TF, Jones CW. Systematic Reviews Published in Emergency 41 29 Medicine Journals Do Not Routinely Search Clinical Trials Registries: A Cross 42 30 sectional Analysis. Ann Emerg Med Published Online First: 2014. 43 44 31 doi:10.1016/j.annemergmed.2014.10.001 45 32 25 van Enst WA, Scholten RJ, Hooft L. Identification of additional trials in prospective 46 47 33 trial registers for Cochrane systematic reviews. PLoS One 2012; 7:e42812. 48 34 doi:10.1371/journal.pone.0042812 49 50 35 26 Jones CW, Keil LG, Weaver MA, et al. Clinical trials registries are under utilized in 51 36 the conduct of systematic reviews: a cross sectional analysis. Syst Rev 2014; 3:126. 52 37 doi:10.1186/2046 4053 3 126 53 54 38 27 Cortés J, González JA, Campbell MJ, et al. A hazard ratio was estimated by a ratio of 55 39 median survival times, but with considerable uncertainty. J Clin Epidemiol 56 40 2014; 67 :1172–7. doi:10.1016/j.jclinepi.2014.05.017 57 58 41 28 Dechartres A, Boutron I, Trinquart L, et al. Single center trials show larger treatment 59 60 22 https://mc.manuscriptcentral.com/bmj Page 23 of 70 BMJ
1 2 3 1 effects than multicenter trials: evidence from a meta epidemiologic study. Ann Intern 4 2 Med 2011; 155 :39–51. doi:10.1059/0003 4819 155 1 201107050 00006 5 6 3 29 Potthast R, Vervolgyi V, McGauran N, et al. Impact of inclusion of industry trial 7 4 results registries as an information source for systematic reviews. PLoS One 8 5 Confidential:2014; 9:e92067. doi:10.1371/journal.pone.0092067 For Review Only 9 10 6 30 Page MJ, Shamseer L, Altman DG, et al. Epidemiology and reporting characteristics of 11 7 systematic reviews of biomedical research: a cross sectional study. PLoS Med 12 8 2016; 13 :e1002028. doi:10.1371/journal.pmed.1002028 13 14 9 31 Hart B, Lundh A, Bero L. Effect of reporting bias on meta analyses of drug trials: 15 10 reanalysis of meta analyses. Bmj 2012; 344 :d7202–d7202. doi:10.1136/bmj.d7202 16 17 11 32 Golder S, Loke YK, Wright K, et al. Reporting of Adverse Events in Published and 18 12 Unpublished Studies of Health Care Interventions: A Systematic Review. PLoS Med 19 13 2016; 13 :e1002127. doi:10.1371/journal.pmed.1002127 20 21 14 33 Glanville JM, Duffy S, McCool R, et al. Searching ClinicalTrials.gov and the 22 15 International Clinical Trials Registry Platform to inform systematic reviews: what are 23 24 16 the optimal search approaches? J Med Libr Assoc 2014; 102 :177–83. doi:10.3163/1536 25 17 5050.102.3.007 26 27 18 34 Riveros C, Dechartres A, Perrodeau E, et al. Timing and Completeness of Trial Results 28 19 Posted at ClinicalTrials.gov and Published in Journals. PLoS Med 2013; 10 :e1001566. 29 20 doi:10.1371/journal.pmed.1001566 30 31 21 35 Strom BL, Buyse M, Hughes J, et al. Data Sharing, Year 1 — Access to Data from 32 22 Industry Sponsored Clinical Trials. N Engl J Med 2014; 371 :2052–4. 33 23 doi:10.1056/NEJMp1411794 34 35 24 36 Maruani A, Boutron I, Baron G, et al. Impact of sending email reminders of the legal 36 25 requirement for posting results on ClinicalTrials.gov: cohort embedded pragmatic 37 26 randomized controlled trial. BMJ 2014; 349 :g5579. doi:10.1136/bmj.g5579 38 39 27 37 Zarin DA, Tse T, Williams RJ, et al. Trial Reporting in ClinicalTrials.gov — The Final 40 28 Rule. N Engl J Med 2016;:NEJMsr1611785. doi:10.1056/NEJMsr1611785 41 42 29 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 23 https://mc.manuscriptcentral.com/bmj BMJ Page 24 of 70
1 2 3 Figure 1: Study flow diagram 4 5 Citations identified at PubMed 6 1 807 excluded based on titles and abstract 7 n= 2 249 8 Confidential: For Review Only Non-pharmaceutical treatment: n=1255 9 Updated meta-analyses: n=264 10 Including non-randomized trials: n=204 11 12 Individual patient data: n=42 13 Network meta-analyses: n=29 14 Duplicates: n=4 15 Not meta-analysis: n=1 16 Animal studies: n=1 17 Full-text not retrieved: n=7 18 19 Full-text articles 20 21 n=442 22 23 219 excluded after reading the full text
24 Including non-randomized trials: n=84 25 26 Non-pharmaceutical treatment: n=76 27 Updated meta-analyses: n=26 28 Not meta-analysis: n=20 29 Individual patient data: n=9 30 Network meta-analyses: n=3 31 Animal studies: n=1 32 33 223 systematic reviews 34 35 included 36 37 38 39 40 41 107 (48%) systematic 42 116 (52%) systematic 43 reviews reporting a search of reviews not reporting a 44 at least one trial registry search of a trial registry 45 46 47 48 60 s ystematic reviews 21 systematic reviews 49 not reporting results of not reporting information to perform 50 a trial registry search 51 a trial registry search 52 53 54 47 systematic reviews 95 systematic reviews 55 reporting the results of reporting information 56 a trial registry search to perform a trial 57 registry search 58 59 60 https://mc.manuscriptcentral.com/bmj Page 25 of 70 BMJ Figure 2: Identification of trials by searching trials registries (WHO ICTRP Search Portal)
1 2 95 systematic reviews 3 Search in trials registries to identify new eligible 4 5 Confidential: Forcompleted Review or terminated RCTsOnly 6 15,282 records screened 7 8 9 10 11 54 systematic reviews 12 13 No new eligible RCT identified in 14 the registries 15 16 17 41 systematic reviews 18 with at least 1 new eligible completed/terminated RCT 19 identified 20 21 122 new eligible RCTs involving 52,743 patients identified 22 23 24 25 2659 RCTs without results identified 27 28 29 30 22 systematic reviews 31 32 Identification of at least 1 new eligible RCT with results 33 available 34 35 63 RCTs involving 42,202 patients with results identified 36 37 38 39 40 18 RCTs with results identified 41 but that could not be included in 42 an MA 43 44 45 46 14 systematic reviews 47 Identification of at least 1 new eligible RCT that could 48 49 contribute to an MA of the SR 50 45 RCTs involving 21,358 patients could contribute to 51 52 quantitative analyses 53 54 55 https://mc.manuscriptcentral.com/bmj 56 MA= meta-analysis, IQR= Interquartile range, Med= median, RCT = randomized controlled trial 57* The RCTs could not contribute to the quantitative analyses because outcomes of interest were not reported 58 59 60 BMJ Page 26 of 70
1 2 3 4 Table 1: Characteristics of included systematic reviews and registry searches 5 6 Systematic reviews 7 Characteristics of the systematic reviews 8 Confidential: For Reviewn=223 Only (%) 9 Type of review - Cochrane reviews 77 (35) 10 11 - Non-Cochrane reviews 146 (65) 12 Funding - Not-for-profit 106 (47.5) 13 - For-profit 14 3 (1.3) 15 - No funding 33 (14.8) 16 - Not reported or unclear 81 (36.3) 17 18 Number of RCTs included in the Median [Q1 -Q3] 10.0 [6.0-18.0] systematic reviews 19 Min-max 2-158 20 21 Number of patients included in the Median [Q1 -Q3] 1,594.0 [614.0-5,027.0] 22 systematic reviews * Min-max 47-102,607 23 24 Clinical trial registry search (yes) 107 (48.0) 25 26 Characteristics of registry search n=107 (%) 27 28 Search portal (at least one portal searched) 57 (53.3) 29 - WHO ICTRP 53 (49.5) 30 - MetaRegister of Current Controlled Trials 15 (14.0) 31 - International Federation of Pharmaceutical 32 1 (0.9) 33 Manufacturers and Associations 34 Individual clinical trial registries approved by the WHO or 93 (86.9) 35 ICMJE (at least one searched) 36 - ClinicalTrials.gov 89 (83.2) 37 - International Standard Randomised Controlled Trial 38 22 (20.6) 39 Number Register - EU Clinical Trials Register 5 (4.7) 40 41 - Australian New Zealand Clinical Trials Registry 5 (4.7) 42 - Japan Primary Registries Network 3 (2.8) 43 44 - Chinese Clinical Trial Registry 1 (0.9) 45 Non-approved or unclear individual clinical trial registries 11 (10.3) 46 RCTs: Randomized controlled trials 47 [Q1-Q3] : Interquartil range 48 WHO ICTRP: World Health Organization International Trials Registry Platform 49 * Number of patients included was unclear or missing in 9 non-Cochrane systematic reviews 50 51 52 53 54 55 56 57 58 59 60 https://mc.manuscriptcentral.com/bmj Page 27 of 70 BMJ
1 2 3 4 5 Table 2Confidential: Effect of adding randomized controlled tria ls Forretrieved from Review clinical trial registries on Only meta-analyses 6 7 Number of Number of 8 Number of RCTs (and Weight of the RCTs (and 9 RCTs2 (and patients) Description of the new RCTs Change in Direction of patients) Summary statistic of the Summary statistic of the Change in 10 patients) retrieved with selected 5 outcomes included in summary change in 1 retrieved selected meta-analysis in selected meta-analysis statistical 11ID included in results that (type of outcome: the selected statistic summary from trial the original SR with new RCTs included significance 12 the original could PO, SO or U 6) meta- (%) statistic 13 3 registry SR contribute to at analysis (%) 14 search least one MA 4 15 16
17Efficacy outcomes 18 19 Atrial fibrillation (PO) Decrease 201 18 (9952) 1 (73) 1 (73) OR 7 0.51 [0.36 ; 0.70] OR 0.53 [0.38 ; 0.73] 1.9 6 No 21 efficacy 22 23 Decrease 242 9(11390) 2 (355) 1 (322) PASI 75 8 (PO) RR 9 18.28 [12.76 ; 26.17] RR 14.20 [10.72 ; 18.81] 37.6 9 no efficacy 25 26 27 Overall survival (U) Decrease No 3 20 (8225) 8 (1806) 2 (1400) HR 10 0.87 [0.82 ; 0.91] HR 0.88 [0.84 ; 0.93] 14.8 8 28 efficacy 29 30 Overall survival (U) Decrease No 314 6 (2264) 1 (1029) 1 (1029) HR 0.89 [0.80 ; 0.99] HR 0.90 [0.83 ; 0.98] 34.8 10 32 efficacy 33 34 5 12 (6297) 2 (340) 1 (102) Overall survival (U) HR 0.99 [0.90 ; 1.09] HR 0.99 [0.90 ; 1.08] 3.5 0 No change No 35 36 37 Neuropsychiatric No Decrease 386 32 (6812) 8 (3831) 5 (2942) inventory total score SMD 11 -0.21 [-0.29 ; -0.12] SMD -0.19 [-0.28 ; -0.11] 9.0 10 efficacy 39 (U) 40 41 42 43 44 45 46 https://mc.manuscriptcentral.com/bmj 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Page 28 of 70
1 2 3 4 5 Confidential:12 For Review Only 6 UPDRS scale (U) Decrease No 7 9 (2857) 1 (514) 1 (514) MD 13 -1.77 [-2.13 ; -1.41] MD -1.66 [-1.99 ; -1.32] 16.1 6 7 efficacy 8 9 HbA1c 14 (U) Increase No 108 23 (18980) 28 (14733) 21 (11298) MD -0.35 [-0.51 ; -0.19] MD -0.45 [-0.55 ; -0.36] 58.3 29 efficacy 11 12 13 Safety outcomes 14 15 169 14 (42602) 1 (166) 1 (166) Major bleeding (U) OR 0.88 [0.79 ; 0.99] OR 0.88 [0.79 ; 0.98] 0.2 0 No change No 17 18 Opportunistic infection No 1910 70 (32054) 4 (2039) 4 (2039) OR 1.79 [1.17 ; 2.74] OR 1.52 [1.04 ; 2.23] 18.7 28 Less harm 20 (PO) 21 22 Withdrawal due No 11 9 (11007) 4 (810) 2 (550) RR 0.83 [0.74 ; 0.93] RR 0.85 [0.76 ; 0.94] 0.2 13 More harm 23 adverse event (SO) 24 25 2612 16 (33958) 1 (129) 1(129) Major bleeding (PO) RR 0.79 [0.52 ; 1.19] RR 0.80 [0.54 ; 1.20] 1.5 5 More harm No 27 28 29 Treatment No 3013 19 (101801) 2 (317) 2 (317) discontinuation due to RR 1.40 [1.08 ; 1.82] RR 1.37 [1.06 ; 1.75] 8.6 6 Less harm 31 all cause (PO) 32 3314 43 (16011) 7 (943) 2 (477) Fatal adverse event (U) RR 1.63 [1.32 ; 2.01] RR 1.62 [1.32 ; 2.99] 1.2 1 Less harm No 34 35 36 37 1 ID = Identification of the systematic review, corresponding to appendix 4 38 2 RCT = randomised controlled trial 39 3 SR = systematic review 40 4 MA = meta-analysis 41 42 43 44 45 46 https://mc.manuscriptcentral.com/bmj 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 29 of 70 BMJ
1 2 3 4 5 5 Confidential: For Review Only 6 For each systematic review, we selected one meta-analysis in which at least one RCT with results available could be included according a predefined hierarchical order of 7 outcomes analyzed as follows: 1) the primary efficacy outcome, 2) the primary safety outcome, and 3) the most clinically relevant outcome. If none of these meta-analyses 8 could include an RCT, we selected the meta-analysis that could include at least one RCT that was reported first. 9 6 PO = the outcome was defined as primary in the SR, SO = the outcome was defined as secondary in the SR, U = primary and secondary outcome were not pre specified in 10 the SR 11 7 OR = odds ratio 12 8 PASI 75 = 75% reduction in the Psoriasis Area Severity Index 13 9 RR = Risk ratio 14 10 HR = hazard ratio 15 11 SMD = standardized mean difference 16 12 UPDRS scale = Unified Parkinson Disease Rating Scale 17 13 MD = mean difference 18 14 HbA1c = glycated hemoglobin 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 https://mc.manuscriptcentral.com/bmj 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Page 30 of 70
1 2 3 4 5 APPENDIX 6 7 8 Confidential:Appendix 1: Verification of the recruitment For status Review according to the registry Only 9 10 11 ClinicalTrials.gov 12 13 We systematically verified the Last Verified date recorded in the registry (ie, the most recent 14 15 date on which all of a clinical study's information on ClinicalTrials.gov was confirmed as 16 accurate and current). 17 18 If the Last Verified date was before the date of the search, the trial was included. 19 20 If the Last Verified date was after the date of the search, we verified in the archives of the 21 registry Web site when the status was modified and we excluded trials that were recorded as 22 23 “completed” or “terminated” after the date of search. 24 25 26 27 28 UMIN registry: 29 30 We systematically verified the “Date of last update”. If this date was before the date of the 31 search, the trial could be included. 32 33 If the” Date of last update” was after the date of search, we verified that the “date trial data 34 35 considered complete” and the “date analysis concluded” was before the date of the search and 36 we verified in the history of the registry that these dates were recorded before the date of 37 38 search; if not, the trial was excluded. 39 40 41 ISRCTN: 42 43 We systematically verified the “Last edited” date. If this date was before the date of the 44 45 search, the trial could be included. 46 If the” Last edited” date was after the date of the search, we verified the “Recruitment end 47 48 date” and the “Overall trial end date.” 49 50 Because this registry did not give access to archives, if additional identifiers with a 51 ClinicalTrials.gov number was provided, we searched this registry. 52 53 54 55 56 57 58 59 60 https://mc.manuscriptcentral.com/bmj Page 31 of 70 BMJ
1 2 3 ANZCTR: 4 5 We systematically verified the timing of the registration status in the history. 6 7 The history reported the timing of the modification with the reason for the modification. 8 Confidential: For Review Only 9 10 11 Eudract: 12 13 We downloaded the full trial details; we checked the trials status and the date on which this 14 15 record was first entered in the EudraCT database 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 https://mc.manuscriptcentral.com/bmj BMJ Page 32 of 70
1 2 3 4 Appendix 2: Keywords, date of search and finding in the WHO ICTRP Search Portal 5 First Last 6 Key words used in Citation ID Key words used in "Conditions" field limit of electronic 7 "Interventions" field findings 8 Confidential: For Reviewsearch Onlysearch 1 stomach neoplasms OR gastric cancer 9 paclitaxel OR S 1 OR fluorouracil 30/11/2013 350 10 OR carcinosis 2 tenofovir OR gs4331 OR gs 4331 11 HIV OR antiretroviral naïve 31/10/2013 344 12 OR gs 4331 3 13 Sleep Bruxism 31/08/2014 23
14 4 constipation OR fecal impaction polyethylene glycol OR laxative 10/02/2014 48 15 16 5 perphenazine 31/10/2013 14 17 18 6 probiotics OR bifidobacterium OR food hypersensitivity OR food allergy 30/09/2013 10 19 lactobacillus 20 statin OR atorvastatin OR 21 rosuvastatin OR cerivastatin OR 22 simvastatin OR pravastatin OR 23 coronary angiography lovastatin OR 31/01/2014 321
24 Hydroxymethylglutaryl CoA 25 7 reductase inhibitors OR HMG CoA 26 reductase inhibitors 27 Myelodysplastic Syndromes OR 28 refractory anemia OR Preleukemia OR refractory cytopenia OR Refractory Romiplostim OR eltrombopag 28/02/2014 64 29 8 anemia excess blasts OR 30 Thrombocytopenia 31 abitesartan OR azilsartan OR 32 candesartan OR elisartan OR 33 embusartan /// eprosartan OR 34 forasartan OR irbesartan OR 35 losartan OR milfasartan OR 36 hypertension OR blood pressure olmesartan OR saprisartan OR 15/01/2014 909
37 tasosartan OR telmisartan OR 38 valsartan OR zolasartan OR KT3 39 671 OR atacand OR teveten OR 40 9 avapro OR cozaar OR benicar OR 41 micardis OR diovan 42 new oral anticoagulants OR direct coagulation OR Xa inhibitor OR IIa 43 thrombosis OR embolism OR inhibitor OR thrombin inhibitor OR 28/02/2014 106 thromboembolism 44 10 rivaroxaban OR dabigatran OR 45 apixaban OR edoxaban 46 antioxidan t OR ascorbic acid OR 47 bilirubin OR butylated 48 hydroxyanisole OR butylated 49 hydroxytoluene OR canthaxanthin 50 OR carotenoids OR catalase OR ergothioneine //// grape seed extract 51 alcoholic pancreatitis OR chronic OR melatonin OR 31/03/2010 6 52 pancreatitis 53 nordihydroguaiaretic acid OR 54 probucol OR propyl gallate OR 55 pyrogallol OR quercetin OR 56 selenium OR silymarin OR thioctic 11 acid OR tocopherols /// tocotrienols 57 OR uric acid OR vitamin OR alpha 58 59 60 https://mc.manuscriptcentral.com/bmj Page 33 of 70 BMJ
1 2 First Last 3 Key words used in Citation ID Key words used in "Conditions" field limit of electronic 4 "Interventions" field findings 5 search search 6 tocopherol OR beta tocopherol OR 7 gamma tocopherol OR zeta 8 Confidential:carotene For OR beta carotene Review OR Only curcumin OR methionine OR 9 allopurinol OR oxidizing agent 10 11 12 13 14 15 12 chronic kidney disease AND 16 15/11/2012 11 hyperuricemia 17 acute coronary syndromes OR ST bivalirudin OR angiomax OR 18 elevation myocardial infarction OR hirulog OR stent OR percutaneous 09/04/2014 71 19 13 non ST elevation myocardial infarction coronary 20 OR unstable angina 21 14 atrial fibrillation atorvastatin 30/04/2014 12 22 23 arterial compliance OR pulse wave OR antioxydants OR ascorbic acid OR 24 vascular siffnesss OR applanation vitamin E OR vitamin A OR 31/12/2013 5 25 15 tonometry OR arterial stiffness OR vitamin C OR tocopherol OR 26 pulse carotene OR dietary supplements 16 27 fluphenazine 01/05/2010 6 28 17 Nalbuphine OR en2234a OR en 29 31/07/2013 10 30 2234a OR nubain 31 Biological agent OR Biological 32 therapy OR VEGF A OR VEGFA colon OR rectum OR colorectal OR EGF receptor OR bevacizumab 5/31/2013 684 33 18 OR cetuximab OR panitumumab 34 OR aflibercept OR regorafenib 35 new oral anticoagulant OR oral 36 thrombin inhibitor OR factor Xa 01/01/20 23/03/2014 217 37 19 inhibitor OR dabigatran OR 01 38 rivaroxaban OR apixaban 39 appetite stimulants OR 40 cyproheptadine OR prednisolone 41 OR progestational agents OR 42 progestins OR anabolic agents OR 43 megesterol OR megace OR 44 mirtazapine OR antidepressive 45 agents OR antidepressants OR cannaboids OR 46 cystic fibrosis OR CF OR tetrahydrocannabinol /// 08/04/2014 4 mucovicidosis 47 antihistamines OR histamine 48 antagonists OR corticosteroids OR 49 prednisone OR steroids OR 50 hormone therapy OR growth 51 hormone OR hormones OR 52 dronabinol OR pizotyline OR 53 20 pizotifen OR risperidone OR 54 olanzapine 21 colorectal cancer OR colon cancer OR 55 panitumumab OR vectibix 31/03/2014 122 56 rectal cancer 57 58 59 60 https://mc.manuscriptcentral.com/bmj BMJ Page 34 of 70
1 2 First Last 3 Key words used in Citation ID Key words used in "Conditions" field limit of electronic 4 "Interventions" field findings 5 search search 6 gonadotropin releasing hormone OR 7 GnRH analogue OR GnRH agonist breast cancer OR triptrorelin OR goserelin OR 31/03/2014 84 8 Confidential: For Review Only 22 leuprolide OR busselin OR 9 nafarenlin 10 23 haloperidol 01/05/2010 82 11 12 24 metformin AND (repaglinide OR 30/11/2013 11 13 novonorm) 14 25 trifluoperazine 01/05/2010 3 15 16 contrast induced acute kidney injury statin OR 3 hydroxy 3 17 OR CIN OR contrast induced methylglutaryl coenzyme A 18 nephropathy OR contrast nephropathy 10/02/2014 1 reductase inhibitor OR HMG CoA 19 26 OR AKI OR acute kidney injury OR OR CI AKI OR CI AKI OR 20 ARF OR acute renal failure 21 rivaroxaban OR dabigatran OR apixaban OR new oral anticoagulant 01/01/20 22 15/09/2013 180 23 27 OR oral thrombin inhibitors OR oral 01 24 factor Xa inhibitors 28 25 tonsillectomy OR adenotonsillectomy ketamine OR analgesics OR opioid 01/02/2013 1
26 gastrointestinal cancer OR gastric 27 29 cancer OR colorectal cancer OR colon S 1 OR 5 fluorouracil 31/12/2013 631
28 cancer OR rcetal cancer 29 probiotics OR prebiotics OR 30 dermatitis OR eczema OR atopy OR synbiotics OR lactobacillus OR 31/12/2013 42 31 30 atopic lactobacilli bifidobacteria OR 32 bifidobacterium 33 31 heart failure AND congestive adrenergic beta antagonists 31/12/2013 6 34 35 32 36 33 37 agitation OR delirium sevaflurane OR dexmedetomidine 15/03/2014 49 38 39 recombinant human thyroid 34 thyroid cancer hormone stimulating hormone OR 31/08/2013 2 40 41 thyroid hormone withdrawal 35 42 hypertension portal propranolol AND carvedilol 31/03/2013 0
43 36 cancer OR tumour OR carcinoma OR vitamin D OR cholecalciferol OR 4/30/2014 44 197 neoplasm ergocalciferol 45 inflammation OR high sensitivity C 46 37 reactive protein OR high sensitive C vitamin D OR cholecalciferol 28/02/2014 18
47 reactive protein OR hs CRP 48 Embolism OR Thrombosis OR Anticoagulants OR heparin OR 49 Postoperative Complications OR UFH OR LMWH OR warfarin OR 50 Intraoperative Complications OR deep coumadin OR vitamin K antagonist 51 venous thrombosis OR DVT OR OR VKA OR aspirin OR ASA OR 52 pulmonary embolism OR thrombosis 6/30/2013 289 factor Xa inhibitor OR OR thrombotic OR emboli OR 53 fondaparinux OR rivaroxaban OR thromboemboli OR thromboprophyla 54 apixaban OR thrombin inhibitor OR 38 OR bleed OR hemorrhag OR 55 dabigatran 56 complication 39 57 dabigatran OR BIBR 1048 08/12/2013 76 58 59 60 https://mc.manuscriptcentral.com/bmj Page 35 of 70 BMJ
1 2 First Last 3 Key words used in Citation ID Key words used in "Conditions" field limit of electronic 4 "Interventions" field findings 5 search search 40 operable advanced breast cancer OR neoadjuvant OR trastuzumab OR 6 31/03/2014 20 7 locally advanced breast cancer lapinib OR pertuzumab contrast medium OR contrast OR 8 Confidential:hydroxyl For methylglutaryl Review coenzyme Only radiography OR angiocardiography OR 9 A reductase inhibitor OR HMG angiography OR heart catheterization 10 CoA reductase inhibitor OR statins OR cardiac catheterization OR kidney OR atorvastatin OR rosuvastatin 1/1/1950 1/31/2014 65 11 diseases OR kidney failure OR nephritis OR simvastatin OR pravastatin OR 12 OR kidney disease OR nephrotoxicity cerivastatin OR fluindostatin OR 13 41 OR nephrotoxic OR contrast fluvastatin 14 nephropathy 15 42 gastric cancer OR stomach cancer S 1 OR fluouracil 20/02/2014 201 16 17 peri operative period OR 18 postoperativeperiod OR surgery OR melatonin 30/09/2013 9 19 43 surgical OR operation OR surgical 20 procedures OR operative procedures 21 “tiotropium” AND “fluticasone “chronic obstructive pulmonary propionate/salmeterol” et 22 31/12/2013 3 23 44 disease” OR “COPD” “tiotropium” AND “fluticasone– 24 salmeterol” “bevacizumab” OR “avastin” OR 25 “aflibercept” OR “VEGFR TKIs” 26 OR “sorafenib” OR “nexavae” OR 27 “sunitinib” / “sutent” OR “SU1248” 28 OR “vandetanib” OR “caprelsa” OR 01/01/20 cancer 28/02/2014 2680 29 “ZD6474” OR “axitinib” OR 04 30 “pazopanib” OR “votrient” OR 31 “GW786034” OR “regorafenib” OR 32 45 “apatinib” OR “ramucirumab” OR 33 “angiogenesis inhibitors” 46 “Kashin Beck disease” or “KBD” or “hyaluronic acid” or “hyaluronan” 34 30/11/2013 1 35 “Urov” or “hyaluronate” or “HA” extended release pramipexole OR 36 Parkinson’s disease OR Parkinson’s OR 47 ropinirole prolonged released OR 10/02/2013 3 37 PD 38 rotigotine transdermal patch 48 non small cell lung cancer OR EGFR epidermal growth factor receptor 39 31/07/2013 362 wild type OR EGFR mutation negative inhibitors OR erlotinib OR gefitinib 40 Gestational diabetes OR gestational Metformin OR hypoglycemic drugs 41 49 diabetes mellitus OR diabetes OR Hypoglycemic Agents OR 31/12/2012 664
42 pregnancy Antidiabetic 43 50 schizophrenia chlorpromazin 30/06/2013 6 44 45 ‘‘herpes simplex virus thymidine (‘‘malignant glioma’’ or ‘‘high grade 46 51 kinase’’ or ‘‘HSV tk’’ or ‘‘gene 30/11/2013 1 glioma’’ or ‘‘GBM’’ or ‘‘HGG’’ 47 therapy’’ or ‘‘genetic therapy’’ 48 nonalcoholic fatty liver disease OR 49 52 NAFLD OR nonalcoholic pentoxifylline 31/01/2013 2
50 steatohepatitis OR NASH add back OR HRT OR GnRHa OR 51 01/01/19 53 GnRH a OR GnRH agonist OR GnRH 28/02/2013 12 52 98 53 analogues 54 Clonidin OR Catapres OR 54 06/02/2013 12 55 Dexmedetomidine 56 57 58 59 60 https://mc.manuscriptcentral.com/bmj BMJ Page 36 of 70
1 2 First Last 3 Key words used in Citation ID Key words used in "Conditions" field limit of electronic 4 "Interventions" field findings 5 search search 6 “alfusosin” OR “doxazosin” OR 7 “tamsulosin” OR “terazosin” OR 8 Confidential:“silodosin” OR “fiansteride” OR For Review Only “dutasteride” OR “sildenafil” OR 9 “tadalafil” OR “vardenafil” OR 10 “oxybutynin” OR “tolterodine” OR 11 “trospium chloride” OR “darifenacin” 12 OR “solifenacin” / “fesoterodine” OR 31/01/2013 104 13 “mirabegron” / “serenoa” OR 14 “Adrenergic alpha Antagonists” OR “5 15 alpha reductase inhibitors” OR 16 “phosphodiesterase 5 inhibitors” OR 17 “cholinergic antagonists” OR “2 (2 18 aminothiazol 4 yl) 4′ (2 ((2 hydroxy 2 19 55 phenylethyl)amino)ethyl)acetanilide” 20 OR “serenoa” 56 local analgesia OR "intra articular 21 31/08/2013 1 22 analgesia 57 23 chemotherapy OR per formance status 31/07/2013 274
24 58 systematic chemotherapy OR 01/01/20 ovarian cancer 31/01/2013 41 25 pegylated liposomal doxorubicin 00 26 infliximab OR etanercept OR 27 adalimumab OR certolizumab OR 28 rheumatoid AND arthritis golimumab OR anakinra OR 6/24/2013 581
29 59 abatacept OR tocilizumab OR 30 rituximab 31 60 01/01/20 ranibizumab OR bevacizumab 31/03/2013 215 32 04 33 axitinib OR cabozantinib OR 34 erlotinib OR gefitinib OR lapatinib OR pazopanib OR regorafenib OR 3/31/2013 3576 35 36 61 sorafenib OR sunitinib OR 37 vandetanib 62 38 statin 31/07/2013 0
39 alpha blockers OR doxazosin OE 40 Erectile dysfunction OR Lower urinary alfuzosin OR tamsulosin OR PDE5 tract symptoms OR Benign prostatic 30/11/2013 52 41 63 OR sildenafil OR tadalafil OR hyperplasia OR ED OR LUTS OR BPH 42 vardenafil OR udenafil 43 palonosetron AND (antineoplastic 44 malignant OR neoplasms OR cancer agents OR neoplastic OR 30/06/2013 9 45 64 OR oncology chemotherapy OR 46 chemoradiotherapy) 47 65 hypotonic AND isotonic 31/01/2013 3 48 49 body fat OR body weight OR fat free mass OR fat mass OR adiposity OR fat 50 vitamin D OR vitamin D 01/01/19 distribution OR body fat regulation OR 31/03/2013 51 51 supplementation 95 52 66 BMI OR weight loss OR body 53 composition. 54 steroid OR corticosteroid glucocorticoid OR dexamethasone 55 cardiac surgery OR cardiopulmonary OR prednisolone OR prednisone 1996 30/04/2013 9 bypass OR heart surgery 56 67 OR methylprednisolone OR 57 hydrocortisone 58 59 60 https://mc.manuscriptcentral.com/bmj Page 37 of 70 BMJ
1 2 First Last 3 Key words used in Citation ID Key words used in "Conditions" field limit of electronic 4 "Interventions" field findings 5 search search 6 cardiovascular disease OR coronary OR 7 myocardial ischemia OR stenosis OR restenosis OR revascularization OR folic acid OR folate OR 8 Confidential: For Review1966 Only 30/09/2013 109 coronary OR coronary intervention OR multivitamin 9 68 cerebrovascular OR percutaneous 10 disease OR stroke 11 Serotonin Uptake Inhibitors OR 12 SSRI OR fluoxetine OR citalopram 13 OR dapoxetine OR escitalopram Multiple Sclerosis 20/03/2013 8 14 OR fluvoxamine OR indalpine OR 15 69 paroxetine OR sertraline OR 16 vilazodone OR zimeldine 17 70 postoperative pain OR postoperative nicotine 31/07/2012 2 18 nausea vomiting 19 chronic obstructive pulmonary dise ase 71 OR chronic bronchitis OR pulmonary NAC OR acetylcysteine 01/08/2013 12 20 21 emphysema OR COPD 72 22 hyperglycemia OR stroke intravenous insulin 1966 15/02/2013 3 23 24 gonadotropin releasing hormone agonist OR luteinizing hormone 01/01/19 25 30/08/2013 258 73 releasing hormone agonist OR 92 26 triptorelin OR goserelin 27 bacterial vaginitides OR bacterial Probiotics OR lactobacillus OR 28 74 vaginoses OR bacterial vaginitis OR bifidobacterium OR lactobacilli OR 31/05/2013 18
29 bacterial vaginosis. lactic acid bacteria. 30 Mirodenafil OR 5 ethyl 2 (5 (4 (2 31 hydroxyethyl)piperazine 1 32 Erectile Dysfunction OR Impotence sulfonyl) 2 propoxyphenyl) 7 1966 31/03/2013 6 33 75 propyl 3,5 dihydro 4H pyrrolo(3,2 34 d)pyrimidin 4 one OR SK3530 35 Lidocaine AND (opioid OR 76 fentanyl OR remifentanil OR 31/03/2013 36 36 37 sufentanil OR alfentanil) 38 myocardial infarction OR percutaneous 77 coronary intervention OR acute cangrelor 30/04/2013 7 39 coronary syndrome 40 78 41 arthroscopic OR postoperative pain bupivacaine 30/04/2013 76
42 79 anaesth OR anaesth OR nerve block dexamethasone 16/05/2014 3 43 44 80 miralax and gatorade OR 31/01/2014 3 45 46 81 rhinoplasty 28/02/2014 12 47 48 atrial fibrillation OR atrial tachycardia catheter ablation OR radiofrequency 49 82 OR atrial tachyarrhythmia OR AT OR 14/03/2014 107 ablation 50 atrial flutter 51 52 53 54 55 56 57 58 59 60 https://mc.manuscriptcentral.com/bmj BMJ Page 38 of 70
1 2 First Last 3 Key words used in Citation ID Key words used in "Conditions" field limit of electronic 4 "Interventions" field findings 5 search search 6 cholinesterase inhibitors OR 7 donepezil OR galantamine OR 8 Confidential:rivastigmine For OR metrifonateReview OR Only tacrine OR antipsychotics OR 9 haloperidol OR thioridazine OR 10 thiothixene OR chlorpromazine OR 11 acetophenazine OR clozapine OR 12 olanzapine //// risperidone OR Alzheimer disease OR AD 31/12/2013 227 13 quetiapine OR aripiprazole OR 14 antidepressants OR setraline OR 15 fluoxetine OR citalopram OR 16 trazodone OR mood stabilizers OR 17 valproate OR carbamazepine OR 18 lithium OR anticonvulsants OR 19 83 benzodiazepines OR memantine OR 20 psychotropic drugs 21 cardiac surgery OR valve surgery OR glucocorticoid OR steroid OR 84 coronary surgery OR cardiopulmonary hydrocortisone OR dexamethasone 31/08/2013 7 22 bypass OR extracorporeal circulation OR methylprednisolone” 23 85 CPR OR cardio pulmonary vasopressine OR epinephrine OR 20/08/2013 0 24 resuscitation OR cardio arrest adrenaline 25 DPP IV inhibitors OR vildagliptin 26 OR sitagliptin OR saxagliptin OR 27 alogliptin OR linagliptin OR 1/31/2013 1661
28 86 dutogliptin OR metformin OR 29 sulfonylureas 30 carotenoids and visual function OR 31 visual performance OR visual acuity OR vision OR contrast sensitivity OR 32 lutein OR zeaxanthin OR glare sensitivity OR AMD OR age 30/04/2014 27 33 xanthophyll 34 related maculopathy OR choroidal 35 87 neovascularization OR geographic 36 atrophy ustekinumab OR CNTO 1275 OR 37 psoriasis OR pustulosis of palms OR 88 interleukin 12/23 monoclonal 01/08/2013 26 pustulosis of soles 38 antibody OR sterala 39 89 lapatinib 28/02/2014 325 40 41 hormone therapy OR intermittent 42 androgen OR androgen antagonists 43 /// hormone blockade OR androgen 44 deprivation OR continuous 45 androgen OR hormone deprivation prostat 4/30/2013 303 46 OR LHRH OR luteinising hormone 47 releasing hormone OR flutamide 48 OR bicalutamide OR cyproterone 49 90 OR buserelin OR goserelin OR 50 leupro OR triptorelin OR nilutamide 51 Premature OR infant OR newborn OR low birth weight OR neonate OR 52 premature OR neurodevelopment OR Erythropoietin OR epo OR epogen 30/11/2012 11 53 neuroprotection OR neurobehavioral OR epoetin OR rhuepo 54 91 development OR neurological 55 development OR neural development 56 57 58 59 60 https://mc.manuscriptcentral.com/bmj Page 39 of 70 BMJ
1 2 First Last 3 Key words used in Citation ID Key words used in "Conditions" field limit of electronic 4 "Interventions" field findings 5 search search 6 spastic colon OR irritable colon OR irritable bowel OR functional bowel OR peppermint oil OR mintoil OR 7 28/01/2013 2 8 92Confidential: colonic disease OR colonic diseases OR Forcolpermin Review Only IBS OR gastrointestinal sydrome 9 93 10 tramadol AND ondansetron 18/08/2014 1
11 94 colorectal OR neoplasms cetuximab 16/02/2014 665 12 13 95 nicergoline 16/08/2013 3 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 https://mc.manuscriptcentral.com/bmj BMJ Page 40 of 70
1 2 3 Appendix 3: Systematic reviews for which data could not be added to the meta-analysis 4 5 Number of RCTs (and 6 Number of RCT is (and Number of RCTs (and patients) retrieved with 7 patients) included in the patients) retrieved from 8 Confidential:ii For Reviewresults that Only could contribute original SR WHO ICTRP search iii 9 to at least one MA 10 1 21 (12242) 2 (1587) 0 11 2 10 (1052) 4 (274) 0 12 3 7 (27024) 1 (60) 0 13 4 10 iv 7 (15613) 0 14 5 5 (4155) 1 (9) 0 15 6 5 (613) 1(400) 0 16 7 25 (1599) 3 (132) 0 17 8 9 (2812) 4 (745) 0 18 9 10 (924) 2 (162) 0 19 20 10 24 (1794) 1 (100) 0 21 11 6 (1268) 1 (50) 0 22 12 7 (2340) 1 (8) 0 23 13 6 (1420) 1 (217) 0 24 14 128 v 1 (66) 0 25 15 23 (24370) 5 (3291) 0 26 16 12 (1268) 2 (49 0) 0 27 17 8 (4855) 1 (501) 0 28 18 18 (2305) 2 (80) 0 29 19 3 (130) 1 (20) 0 30 20 9 (662) 1 (80) 0 31 21 11 (2587) 1 (240) 0 32 33 22 9 (765) 2 (430) 0 34 23 12 (1304) 1 (70) 0 35 24 11 (1481) 2 (142) 0 36 25 8 (1176) 2 (181) 0 37 26 15 (8332) 1 (688) 0 38 27 7 (523) 1 (22) 0 39 1RCT = Randomized controlled trial; 2 SR = Systematic review; 3 MA = Meta-analysis; 4 5 Number of 40 patients included was unclear or missing in two SRs 41 42 43
44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 https://mc.manuscriptcentral.com/bmj Page 41 of 70 BMJ
1 2 3 4 5 6 7 8 9 Confidential: For Review Only 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Appendix 4: Results of the trials registries searches and their 26 27 impact on summary statistics 28 29 30 31 32 33 October 19, 2016 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
https://mc.manuscriptcentral.com/bmj BMJ Page 42 of 70
1 2 3 4 5 6 Contents 7 1 Systematic review 1 3 8 1.1 Title of the systematic review ...... 3 9 1.2Confidential: Inclusion criteria ...... For ...... Review ...... Only ...... 3 10 1.3 Comparison assessed ...... 3 11 1.4 Results ...... 4 12 13 2 Systematic review 2 5 14 2.1 Title of the systematic review ...... 5 15 2.2 Inclusion criteria ...... 5 16 2.3 Comparison assessed ...... 5 17 2.4 Results ...... 5 18 19 3 Systematic review 3 6 20 3.1 Title of the systematic review ...... 6 21 3.2 Inclusion criteria ...... 6 22 3.3 Comparison assessed ...... 6 23 3.4 Results ...... 6 24 25 4 Systematic review 4 7 26 4.1 Title of the systematic review ...... 7 27 4.2 Inclusion criteria ...... 7 28 4.3 Comparison assessed ...... 7 29 4.4 Results ...... 8 30 31 5 Systematic review 5 9 32 5.1 Title of the systematic review ...... 9 33 5.2 Inclusion criteria ...... 9 34 5.3 Comparison assessed ...... 9 35 5.4 Results ...... 9 36 37 6 Systematic review 6 10 38 6.1 Title of the systematic review ...... 10 39 6.2 Inclusion criteria ...... 10 40 6.3 Comparison assessed ...... 10 41 6.4 Results ...... 11 42 43 7 Systematic review 7 12 44 7.1 Title of the systematic review ...... 12 45 7.2 Inclusion criteria ...... 12 46 7.3 Comparison assessed ...... 12 47 7.4 Results ...... 12 48 49 8 Systematic review 8 13 50 8.1 Title of the systematic review ...... 13 51 8.2 Inclusion criteria ...... 13 52 8.3 Comparison assessed ...... 13 53 8.4 Results ...... 14 54 55 9 Systematic review 9 15 56 9.1 Title of the systematic review ...... 15 57 9.2 Inclusion criteria ...... 15 58 9.3 Comparison assessed ...... 15 59 9.4 Results ...... 15 60
October 19, 2016 EP – CEC – PhR 1
https://mc.manuscriptcentral.com/bmj Page 43 of 70 BMJ
1 2 3 4 5 10 Systematic review 10 16 6 7 10.1 Title of the systematic review ...... 16 8 10.2 Inclusion criteria ...... 16 9 10.3Confidential: Comparison assessed ...... For ...... Review ...... Only ...... 16 10 10.4 Results ...... 17 11 11 Systematic review 11 18 12 11.1 Title of the systematic review ...... 18 13 14 11.2 Inclusion criteria ...... 18 15 11.3 Comparison assessed ...... 18 16 11.4 Results ...... 19 17 12 Systematic review 12 20 18 12.1 Title of the systematic review ...... 20 19 12.2 Inclusion criteria ...... 20 20 21 12.3 Comparison assessed ...... 20 22 12.4 Results ...... 20 23 13 Systematic review 13 21 24 13.1 Title of the systematic review ...... 21 25 13.2 Inclusion criteria ...... 21 26 13.3 Comparison assessed ...... 21 27 28 13.4 Results ...... 21 29 14 Systematic review 14 22 30 14.1 Title of the systematic review ...... 22 31 14.2 Inclusion criteria ...... 22 32 14.3 Comparison assessed ...... 22 33 14.4 Results ...... 23 34 35 15 Summary 24 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
October 19, 2016 EP – CEC – PhR 2
https://mc.manuscriptcentral.com/bmj BMJ Page 44 of 70
1 2 3 4 5 6 1 Systematic review 1 7 8 1.1 Title of the systematic review 9 TheConfidential: preventive effect of atorvastatin on atrial For fibrillation: Review a meta-analysis of randomized Only controlled trials 10 11 12 1.2 Inclusion criteria 13 Studies that met the following specified criteria: 14 15 • comparison of atorvastatin with placebo or control treatment, regardless of the background therapy; 16 • randomized controlled human trials; 17 18 • new-onset AF or recurrent AF in each group as an outcome. 19 20 21 1.3 Comparison assessed 22 23 Experimental Control 24 NCT00756886 Atorvastatin Placebo 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
October 19, 2016 EP – CEC – PhR 3
https://mc.manuscriptcentral.com/bmj Page 45 of 70 BMJ
1 2 3 4 5 6 1.4 Results 7 8 9 Confidential: For Review Only 10 Atorvastatin Control Atrial fibrillation 11 Study Events Total Events Total OR 95%−CI W(random) 12 13 New study 14 NCT00756886 6 30 1 25 6.00 [0.67; 53.68] 1.9% 15 Random effects model 6 30 1 25 6.00 [0.67; 53.68] 1.9% 16 Heterogeneity: not applicable for a single study 17 Original study 18 MIRACL 2004 93 1539 96 1548 0.97 [0.72; 1.31] 9.4% 19 Dernellis 2005 14 40 36 40 0.06 [0.02; 0.20] 4.2% 20 Chello 2006 2 20 5 20 0.33 [0.06; 1.97] 2.6% 21 Ozaydin 2006 3 24 11 24 0.17 [0.04; 0.72] 3.4% 22 ARMYDA−3 2006 35 101 56 99 0.41 [0.23; 0.72] 7.8% 23 Tsai 2008 3 52 10 54 0.27 [0.07; 1.04] 3.7% Song 2008 8 62 17 62 0.39 [0.15; 0.99] 5.6% 24 Almroth 2009 54 111 64 111 0.70 [0.41; 1.18] 8.0% 25 Melina 2009 94 315 106 317 0.85 [0.61; 1.18] 9.2% 26 Ji 2009 10 71 23 69 0.33 [0.14; 0.76] 6.1% 27 Spadaccio 2010 2 25 4 25 0.46 [0.08; 2.75] 2.5% 28 Sun 2011 9 49 21 51 0.32 [0.13; 0.80] 5.7% 29 SToP AF 2011 22 33 26 31 0.38 [0.12; 1.28] 4.3% SPARCL 2011 139 2365 122 2366 1.15 [0.89; 1.47] 9.6% 30 Demir 2011 6 22 3 22 2.38 [0.51; 11.05] 3.2% 31 Baran 2012 1 30 7 30 0.11 [0.01; 0.99] 1.9% 32 Suleiman 2012 9 62 8 63 1.17 [0.42; 3.25] 5.1% 33 Jiang 2013 10 50 20 49 0.36 [0.15; 0.89] 5.8% 34 Random effects model 514 4971 635 4981 0.51 [0.36; 0.70] 98.1% 35 Heterogeneity: I−squared=72.3%, tau−squared=0.2675, p<0.0001 36 Random effects model 520 5001 636 5006 0.53 [0.38; 0.73] 100% 37 Heterogeneity: I−squared=72.2%, tau−squared=0.2798, p<0.0001 38 39 0.1 0.5 1 2 10 40 Favors Atorvastatin Favors Control 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
October 19, 2016 EP – CEC – PhR 4
https://mc.manuscriptcentral.com/bmj BMJ Page 46 of 70
1 2 3 4 5 6 2 Systematic review 2 7 8 2.1 Title of the systematic review 9 SystematicConfidential: review and meta-analysis of ustekinumab For for moderateReview to severe psoriasis Only 10 11 12 2.2 Inclusion criteria 13 Firstly, the RCTs had to include patients with a proven diagnosis of plaque psoriasis for at least 6 months. 14 Exclusion criteria for patients included known malignancy (except treated basal cell skin cancer or squa- 15 mous cell skin cancer of at least 5 years duration) or recent serious systemic or local infection. Exclusion 16 criteria for controls included systemic use of corticosteroids, immunosuppressants or agents specifically 17 targeting IL-12 or IL-23 with a withdrawal time of < 2 weeks. Thirdly, articles lacking original data for 18 meta-analysis and review articles were excluded. 19 20 21 2.3 Comparison assessed 22 23 Experimental Control 24 NCT01008995 Ustekinumab Placebo 25 26 27 28 2.4 Results 29 30 31 32 33 Ustekinumab 45 mg Placebo PASI75 34 Study Events Total Events Total RR 95%−CI W(fixed) 35 New study 36 NCT01008995 132 160 18 162 7.42 [ 4.78; 11.54] 37.6% 37 Fixed effect model 132 160 18 162 7.42 [ 4.78; 11.54] 37.6% 38 Heterogeneity: not applicable for a single study 39 40 Original study 41 Igarashi 2012 38 64 2 31 9.20 [ 2.37; 35.70] 5.7% 42 Krueger 2007 33 64 1 64 33.00 [ 4.65; 234.05] 2.1% Leonardi 2008 171 255 8 255 21.38 [10.75; 42.50] 16.8% 43 Papp 2008 273 409 15 410 18.24 [11.05; 30.12] 31.5% 44 Tsai 2011 41 61 3 60 13.44 [ 4.40; 41.07] 6.4% 45 Fixed effect model 556 853 29 820 18.28 [12.76; 26.17] 62.4% 46 Heterogeneity: I−squared=0%, tau−squared=0, p=0.7682 47 48 Fixed effect model 688 1013 47 982 14.20 [10.72; 18.81] 100% Heterogeneity: I−squared=57.3%, tau−squared=0.1919, p=0.0388 49 50 0.01 0.1 1 10 100 51 52 Favors Placebo Favors Ustekinumab 53 54 55 56 57 58 59 60
October 19, 2016 EP – CEC – PhR 5
https://mc.manuscriptcentral.com/bmj Page 47 of 70 BMJ
1 2 3 4 5 6 3 Systematic review 3 7 8 3.1 Title of the systematic review 9 TheConfidential: role of biological therapy in metastatic colorectalFor cancer Review (mCRC) after first-line Only treatment : a meta- 10 analysis of randomised trials 11 12 13 3.2 Inclusion criteria 14 Studies included were registered RCTs evaluating second- or third line (or beyond) therapy for mCRC, 15 which reported at least one of the following: OS, PFS, ORR and toxicity. 16 17 3.3 Comparison assessed 18 19 20 Experimental Control 21 NCT00063141 Cetuximab+Irinotecan Irinotecan 22 NCT00061815 Cetuximab+FOLFOX4 FOLFOX4 23 24 25 3.4 Results 26 27 28 29 30 Study TE seTE OS HR 95%−CI W(fixed) 31 32 New study 33 NCT00063141 −0.03 0.0676 0.98 [0.85; 1.11] 13.7% NCT00061815 0.09 0.2410 1.10 [0.68; 1.76] 1.1% 34 Fixed effect model 0.98 [0.87; 1.12] 14.8% 35 Heterogeneity: I−squared=0%, tau−squared=0, p=0.6362 36 37 Original study 38 2010 Peeters Study 181 −0.16 0.0991 0.85 [0.70; 1.03] 6.4% 39 2013 Seymour PICCOLO 0.01 0.1001 1.01 [0.83; 1.23] 6.2% 40 2008 Amado −0.01 0.1417 0.99 [0.75; 1.31] 3.1% 41 2008 Karapetis CO17 −0.60 0.1499 0.55 [0.41; 0.74] 2.8% 42 2007 Giantonio E3200 −0.29 0.0890 0.75 [0.63; 0.89] 7.9% 43 2012 Arnold TML −0.22 0.0789 0.81 [0.69; 0.94] 10.0% 44 2012 Van Cutsem VELOUR −0.20 0.0695 0.82 [0.71; 0.94] 12.9% 2013 Masi BEBYP −0.28 0.1721 0.76 [0.54; 1.06] 2.1% 45 2011 Van Cutsem CONFIRM2 0.00 0.0735 1.00 [0.87; 1.15] 11.6% 46 2012 Grothey CORRECT −0.26 0.0943 0.77 [0.64; 0.93] 7.0% 47 2012 Siu CO20 −0.13 0.0884 0.88 [0.74; 1.05] 8.0% 48 2011 Watkins 10 mg dalo 0.35 0.1945 1.42 [0.97; 2.08] 1.7% 49 2011 Watkins 7.5 mg Dalo 0.14 0.1916 1.15 [0.79; 1.67] 1.7% 50 2012 Cohn conatumumab −0.12 0.2549 0.89 [0.54; 1.47] 1.0% 51 2012 Cohn conatumumab b 0.24 0.2620 1.27 [0.76; 2.12] 0.9% 52 2013 Eng tivantinib −0.37 0.2606 0.69 [0.42; 1.16] 0.9% 53 2013 Eloehler sorafenib 0.45 0.2563 1.57 [0.95; 2.59] 1.0% 54 Fixed effect model 0.87 [0.82; 0.91] 85.2% 55 Heterogeneity: I−squared=59.6%, tau−squared=0.0191, p=0.0009 56 Fixed effect model 0.88 [0.84; 0.93] 100% 57 Heterogeneity: I−squared=58.1%, tau−squared=0.0171, p=0.0008 58 59 0.5 1 2 60 Favors Biologic Favors Control
October 19, 2016 EP – CEC – PhR 6
https://mc.manuscriptcentral.com/bmj BMJ Page 48 of 70
1 2 3 4 5 6 4 Systematic review 4 7 8 4.1 Title of the systematic review 9 S-1-basedConfidential: versus 5-FU-based chemotherapy asFor first-line treatmentReview in advanced gastricOnly cancer: a meta- 10 analysis of randomized controlled trials 11 12 13 4.2 Inclusion criteria 14 Studies meeting the following inclusion criteria were included: 15 16 • patients suffering from histological confirmed, inoperable, advanced, or recurrent adenocarcinoma of 17 the stomach or gastroesophageal junction at baseline; 18 19 • phase II or phase III RCT; 20 • trials comparing S-1-based with 5-FU-based regimens given as first-line palliative chemotherapy and 21 not confounded by additional agents or interventions; 22 23 • if there were multiple articles based on similar patients, only the largest or the most recently article 24 was included. 25 26 Exclusion criteria included the following: 27 • letters, reviews, case reports, editorials, and expert opinion; 28 29 • non-prospective trials. 30 31 32 4.3 Comparison assessed 33 34 S-1-based regimen 5-FU-based regimen 35 NCT00400179 S-1/Cisplatin 5-FU/cisplatin 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
October 19, 2016 EP – CEC – PhR 7
https://mc.manuscriptcentral.com/bmj Page 49 of 70 BMJ
1 2 3 4 5 6 4.4 Results 7 8 9 Confidential: For Review Only 10 11 Study TE seTE OS HR 95%−CI W(fixed) 12 13 New study 14 NCT00400179 −0.08 0.0713 0.92 [0.80; 1.06] 34.8% Fixed effect model 0.92 [0.80; 1.06] 34.8% 15 Heterogeneity: not applicable for a single study 16 17 Original study 18 Jin 2008 −0.65 0.2900 0.52 [0.30; 0.92] 2.1% 19 Boku 2009 −0.19 0.1000 0.83 [0.68; 1.01] 17.7% 20 Ajani 2010 −0.08 0.0700 0.92 [0.80; 1.06] 36.1% 21 Nishikawa 2012 −0.04 0.1900 0.96 [0.66; 1.39] 4.9% Xu 2013 0.05 0.2000 1.05 [0.71; 1.56] 4.4% 22 Fixed effect model 0.89 [0.80; 0.99] 65.2% 23 Heterogeneity: I−squared=20.8%, tau−squared=0.0047, p=0.282 24 25 Fixed effect model 0.90 [0.83; 0.98] 100% 26 Heterogeneity: I−squared=3.7%, tau−squared=0.0005, p=0.393 27 28 0.5 1 2 29 30 Favors S−1 regimen Favors 5−Fu regimen 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
October 19, 2016 EP – CEC – PhR 8
https://mc.manuscriptcentral.com/bmj BMJ Page 50 of 70
1 2 3 4 5 6 5 Systematic review 5 7 8 5.1 Title of the systematic review 9 EfficacyConfidential: and toxicity of adding cetuximab toFor chemotherapy Review in the treatment ofOnly metastatic colorectal 10 cancer: a meta-analysis from 12 randomized controlled trials 11 12 5.2 Inclusion criteria 13 14 Studies that met the following criteria were considered for inclusion: 15 16 • randomized controlled trials; 17 • the study population of patients aged ≥18 years; 18 19 • eligible patients with histologically or cytologically confirmed mCRC; 20 • randomized allocation to cetuximab plus chemotherapy group or chemotherapy group; 21 22 • results reported data on efficacy and safety. 23 24 Reports were excluded from the final analysis if they described studies with a single-arm design or ran- 25 domized controlled trials that assigned cetuximab into the two treatment arms. 26 27 5.3 Comparison assessed 28 29 Experimental Control 30 NCT00061815 Cetuximab+FOLFOX4 FOLFOX4 31 32 33 34 5.4 Results 35 36 37 38 OS 39 Study TE seTE HR 95%−CI W(random) 40 New study 41 NCT00061815 0.09 0.2410 1.10 [0.68; 1.76] 3.5% 42 Random effects model 1.10 [0.68; 1.76] 3.5% 43 Heterogeneity: not applicable for a single study 44 45 Original study 46 Borner −0.27 0.3703 0.76 [0.37; 1.57] 1.5% Stintzing −0.15 0.2291 0.86 [0.55; 1.35] 3.8% 47 Tol 0.14 0.1071 1.15 [0.93; 1.42] 13.6% 48 Saltz 0.25 0.1740 1.29 [0.92; 1.81] 6.2% 49 Tveit 0.06 0.1241 1.06 [0.83; 1.35] 10.9% 50 Dewdney −0.63 0.3680 0.53 [0.26; 1.09] 1.6% 51 Bokemeyer 0.01 0.1271 1.01 [0.79; 1.30] 10.5% 52 Van Cutsem −0.13 0.0667 0.88 [0.77; 1.00] 24.4% 53 Sobrero −0.02 0.0681 0.98 [0.86; 1.12] 23.9% Random effects model 0.99 [0.90; 1.09] 96.5% 54 Heterogeneity: I−squared=29.6%, tau−squared=0.006, p=0.1821 55 56 Random effects model 0.99 [0.90; 1.08] 100% 57 Heterogeneity: I−squared=22.4%, tau−squared=0.0044, p=0.2373 58 59 0.5 1 2 60 Favors Cetux + Chemo Favors Chemo
October 19, 2016 EP – CEC – PhR 9
https://mc.manuscriptcentral.com/bmj Page 51 of 70 BMJ
1 2 3 4 5 6 6 Systematic review 6 7 8 6.1 Title of the systematic review 9 PharmacologicalConfidential: treatment of neuropsychiatric For symptoms Review in Alzheimer’s disease: aOnly systematic review and 10 meta-analysis 11 12 13 6.2 Inclusion criteria 14 Trials were selected for inclusion if they met all of the following criteria: 15 16 • double-blind, placebo controlled, randomised controlled trials (RCTs); 17 18 • the design of the trial was either parallel or crossover; for a crossover trial, it had a washout period 19 greater than 1 week; 20 • patients enrolled were diagnosed as probable or possible AD according to the Diagnostic and Sta- 21 tistical Manual of Mental Disorders?Fourth Edition or the criteria of the National Institute of Neu- 22 rological and Communicative Disorders and Stroke/Alzheimer?s Disease and Related Disorders As- 23 sociation; 24 25 • studies compared any medicine at any dose with placebo, with any treatment durations; 26 27 • neuropsychiatric outcomes were measured with the most common neuropsychiatric scales Neuropsy- 28 chiatric Inventory (NPI) (NPI-10 or NPI-12) or Neuropsychiatric Inventory-Nursing Home version 29 (NPI-NH). 30 31 6.3 Comparison assessed 32 33 34 Medicine Placebo 35 NCT01438060 Aripiprazole Placebo 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
October 19, 2016 EP – CEC – PhR 10
https://mc.manuscriptcentral.com/bmj BMJ Page 52 of 70
1 2 3 4 5 6 6.4 Results 7 8 9 Confidential: For Review Only 10 Atypical antipsychotic Placebo NPI Total Score 11 Study Total Mean SD Total Mean SD SMD 95%−CI W(random) 12 New study 13 NCT01438060 103 −11.20 23.84 100 −9.75 23.70 −0.06 [−0.34; 0.21] 9.0% 14 Random effects model 103 100 −0.06 [−0.34; 0.21] 9.0% 15 Heterogeneity: not applicable for a single study 16 Original study 17 De Dyen 2004 520 −16.13 15.94 129 −13.70 20.30 −0.14 [−0.34; 0.05] 18.3% De Dyen 2005 106 −11.20 18.81 102 −9.75 18.81 −0.08 [−0.35; 0.20] 9.2% 18 Mintzer 2007 366 −15.90 18.18 121 −13.00 16.40 −0.16 [−0.37; 0.04] 16.1% 19 Street 2000 159 −6.23 7.24 47 −3.70 10.30 −0.31 [−0.64; 0.01] 6.4% 20 Streim 2008 131 −16.43 17.32 125 −10.01 18.83 −0.35 [−0.60; −0.11] 11.2% Sultzer 2008 85 −11.60 15.40 142 −4.20 20.00 −0.40 [−0.67; −0.13] 9.3% 21 Sultzer 2008 94 −7.30 20.20 142 −4.20 20.00 −0.15 [−0.41; 0.11] 10.0% 22 Sultzer 2008 100 −7.00 18.10 142 −4.20 20.00 −0.15 [−0.40; 0.11] 10.4% Random effects model 1561 950 −0.21 [−0.29; −0.12] 91.0% 23 Heterogeneity: I−squared=0%, tau−squared=0, p=0.5898 24 Random effects model 1664 1050 −0.19 [−0.28; −0.11] 100% 25 Heterogeneity: I−squared=0%, tau−squared=0, p=0.5862 26 27 −0.6 −0.4 −0.2 0 0.2 0.4 0.6 28 Favors Medicine Favors Placebo 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
October 19, 2016 EP – CEC – PhR 11
https://mc.manuscriptcentral.com/bmj Page 53 of 70 BMJ
1 2 3 4 5 6 7 Systematic review 7 7 8 7.1 Title of the systematic review 9 Meta-analysisConfidential: of the efficacy and safety of long-acting For non-ergot Review dopamine agonists Only in Parkinson’s disease 10 11 12 7.2 Inclusion criteria 13 Trials were included in the study if they met all of the following criteria: 14 15 • RCT, 16 • study participants were required to have a clinical diagnosis of PD, 17 18 • intervention therapies consisting of long-acting NEDA versus placebo, 19 20 • assessment of the efficacy data in the form of Unified Parkinson’s Disease Rating Scale (UPDRS) 21 scores, "off" time and/or "on" time without troublesome dyskinesia measured by patient diaries, 22 tolerability data in the form of withdrawals, and safety data in the form of adverse events. 23 24 7.3 Comparison assessed 25 26 27 Long-acting NEDA Control 28 NCT00522379 Rotigotine Placebo 29 30 31 32 7.4 Results 33 34 35 36 37 Long acting NEDA Placebo UPDRS ALD 38 Study Total Mean SD Total Mean SD MD 95%−CI W(fixed) 39 New study 40 NCT00522379 392 −1.97 4.37 105 −0.90 3.70 −1.07 [−1.90; −0.24] 16.1% 41 Fixed effect model 392 105 −1.07 [−1.90; −0.24] 16.1% 42 Heterogeneity: not applicable for a single study 43 44 Original study 45 Jankovic 2007 177 −0.39 3.46 96 0.92 3.43 −1.31 [−2.16; −0.46] 15.2% Poewe 2011 213 −2.10 3.35 103 −0.20 3.37 −1.90 [−2.69; −1.11] 17.7% 46 LeWitt 2007 168 −3.15 5.26 92 −0.50 5.27 −2.65 [−3.99; −1.31] 6.2% 47 Poewe 2007 204 −4.20 4.50 101 −2.00 4.30 −2.20 [−3.24; −1.16] 10.2% 48 Pahwa 2007 197 −3.50 5.47 184 −0.90 5.36 −2.60 [−3.69; −1.51] 9.4% 49 Schapira 2011 161 −3.80 4.67 174 −2.60 4.67 −1.20 [−2.20; −0.20] 11.1% 50 Trenkwalder 2011 178 −2.60 3.60 89 −1.30 3.40 −1.30 [−2.18; −0.42] 14.2% 51 Fixed effect model 1298 839 −1.77 [−2.13; −1.41] 83.9% 52 Heterogeneity: I−squared=26%, tau−squared=0.0856, p=0.2306 53 Fixed effect model 1690 944 −1.66 [−1.99; −1.32] 100% 54 Heterogeneity: I−squared=32.7%, tau−squared=0.1132, p=0.1673 55 56 −2 0 2 57 58 Favors Long acting NEDA Favors Placebo 59 60
October 19, 2016 EP – CEC – PhR 12
https://mc.manuscriptcentral.com/bmj BMJ Page 54 of 70
1 2 3 4 5 6 8 Systematic review 8 7 8 8.1 Title of the systematic review 9 TheConfidential: long-term efficacy and safety of DPP-IV inhibitorsFor monotherapy Review and in combination Only with metformin 10 in 18 980 patients with type-2 diabetes mellitus a meta-analysis 11 12 13 8.2 Inclusion criteria 14 Studies were deemed eligible for inclusion in pooled analysis if they met the following criteria: 15 16 • were phase 3 or later, prospected and randomized controlled trials of ≥ 24 weeks’ duration, 17 18 • enrolled adult patients with T2DM, 19 • were comparing DPP-IV inhibitors with placebo, DPP-IV inhibitors +metformin with metformin 20 and DPP-IV inhibitors + metformin with sulphonylureas + metformin and 21 22 • have at least 50 subjects in every arm of the studies. 23 24 25 8.3 Comparison assessed 26 27 Experimental Control 28 NCT00087516 Sitagliptin Placebo 29 NCT00099892 Vildagliptin Placebo 30 NCT00099905 Vildagliptin Placebo 31 NCT00101712 Vildagliptin Placebo 32 NCT00121641 Saxagliptin Placebo 33 NCT00305604 Sitagliptin Placebo 34 NCT00316082 Saxagliptin Placebo 35 NCT00363948 Sitagliptin Placebo 36 NCT00396357 Vildagliptin Placebo 37 NCT00646542 Vildagliptin Placebo 38 NCT00698932 Saxagliptin Placebo 39 NCT00728351 Vildagliptin Placebo 40 NCT00813995 Sitagliptin Placebo 41 NCT00860288 Vildagliptin or Sitagliptin Placebo 42 NCT00918879 Saxagliptin Placebo 43 NCT01023581 Alogliptin alone or in combination with metformin Metformin or Placebo 44 NCT01076088 Sitagliptin alone or in combination with metformin Metformin or Placebo 45 46 NCT01128153 Saxagliptin Placebo 47 NCT01194830 Linagliptin Placebo 48 NCT01214239 Linagliptin Placebo 49 NCT01215097 Linagliptin Placebo 50 51 52 53 54 55 56 57 58 59 60
October 19, 2016 EP – CEC – PhR 13
https://mc.manuscriptcentral.com/bmj Page 55 of 70 BMJ
1 2 3 4 5 6 8.4 Results 7 8 9 Confidential: For Review Only 10 Study TE seTE HbA1c MD 95%−CI W(random) 11 New study 12 NCT00087516 −0.87 0.0758 −0.87 [−1.02; −0.72] 3.0% 13 NCT00099892 −0.90 0.1229 −0.90 [−1.14; −0.66] 2.7% 14 NCT00099905 −0.52 0.1489 −0.52 [−0.81; −0.23] 2.5% 15 NCT00101712 −0.28 0.0848 −0.28 [−0.45; −0.11] 3.0% NCT00121641 −0.67 0.1151 −0.67 [−0.90; −0.44] 2.8% 16 NCT00305604 −0.70 0.1641 −0.70 [−1.02; −0.38] 2.4% 17 NCT00316082 −0.39 0.1149 −0.39 [−0.62; −0.16] 2.8% 18 NCT00363948 −0.70 0.1444 −0.70 [−0.98; −0.42] 2.5% NCT00396357 −0.14 0.0425 −0.14 [−0.22; −0.06] 3.2% 19 NCT00646542 −0.50 0.1000 −0.50 [−0.70; −0.30] 2.9% 20 NCT00698932 −0.50 0.0937 −0.50 [−0.68; −0.32] 2.9% 21 NCT00728351 −0.33 0.1062 −0.33 [−0.54; −0.12] 2.8% 22 NCT00813995 −0.88 0.1045 −0.88 [−1.08; −0.68] 2.8% NCT00860288 −0.49 0.0347 −0.49 [−0.56; −0.42] 3.2% 23 NCT00918879 −0.46 0.1383 −0.46 [−0.73; −0.19] 2.6% 24 NCT01023581 −0.32 0.0775 −0.32 [−0.47; −0.17] 3.0% 25 NCT01076088 −0.35 0.1066 −0.35 [−0.56; −0.14] 2.8% 26 NCT01128153 −0.66 0.1061 −0.66 [−0.87; −0.45] 2.8% NCT01194830 −0.59 0.2196 −0.59 [−1.02; −0.16] 1.9% 27 NCT01214239 −0.50 0.1221 −0.50 [−0.74; −0.26] 2.7% 28 NCT01215097 −0.52 0.0860 −0.52 [−0.69; −0.35] 3.0% 29 Random effects model −0.53 [−0.63; −0.43] 58.3% 30 Heterogeneity: I−squared=85.7%, tau−squared=0.0448, p<0.0001 31 Original study 32 Del Prato 2011 −0.69 0.0857 −0.69 [−0.86; −0.52] 3.0% 33 Haak 2012 −0.60 0.0151 −0.60 [−0.63; −0.57] 3.3% Scherbaum 2008 −0.30 0.1409 −0.30 [−0.58; −0.02] 2.6% 34 Pfützner 2011 −0.52 0.0991 −0.52 [−0.71; −0.33] 2.9% 35 Pan 2012 −0.38 0.1114 −0.38 [−0.60; −0.16] 2.8% 36 Bosi 2009 −0.40 0.0846 −0.40 [−0.57; −0.23] 3.0% 37 Nauck 2009 −0.50 0.1415 −0.50 [−0.78; −0.22] 2.5% Taskinen 2011 −0.64 0.0720 −0.64 [−0.78; −0.50] 3.1% 38 DeFronzo 2009 −0.82 0.0990 −0.82 [−1.01; −0.63] 2.9% 39 Haak 2012 −0.50 0.0120 −0.50 [−0.52; −0.48] 3.3% 40 Göke 2010 0.06 0.0537 0.06 [−0.05; 0.17] 3.1% Gallwitz 2012 0.20 0.0423 0.20 [ 0.12; 0.28] 3.2% 41 Filozof 2010 0.04 0.0849 0.04 [−0.13; 0.21] 3.0% 42 Ferrannini 2009 0.09 0.0302 0.09 [ 0.03; 0.15] 3.2% 43 Random effects model −0.35 [−0.51; −0.19] 41.7% 44 Heterogeneity: I−squared=98.4%, tau−squared=0.0856, p<0.0001 45 Random effects model −0.45 [−0.55; −0.36] 100% 46 Heterogeneity: I−squared=96.5%, tau−squared=0.0699, p<0.0001 47 48 −1 −0.5 0 0.5 1 49 Favors DPP−IV Favors Control 50 51 52 53 54 55 56 57 58 59 60
October 19, 2016 EP – CEC – PhR 14
https://mc.manuscriptcentral.com/bmj BMJ Page 56 of 70
1 2 3 4 5 6 9 Systematic review 9 7 8 9.1 Title of the systematic review 9 DabigatranConfidential: Etexilate and Risk of Myocardial For Infarction, Review Other Cardiovascular Events,Only Major Bleeding, 10 and All-Cause Mortality: A Systematic Review and Meta-analysis of Randomized Controlled Trials 11 12 13 9.2 Inclusion criteria 14 To be included in the meta-analysis, clinical trials should present the following criteria: (1) it should be an 15 RCT and (2) the follow-up should have been the same between the different groups. In addition, (3) the 16 control groups should receive a placebo or the reference treatment when applicable. This meant (3a) war- 17 farin was the reference treatment in patients with NVAF and in the treatment of venous thromboembolism 18 (VTE) or pulmonary embolism; (3b) enoxaparin was the reference treatment for the prevention of VTE 19 events in patients undergoing total hip or knee surgery; and (3c) placebo was used for the prevention of 20 recurrence of coronary events in patients receiving antiplatelet therapy or for the prevention of recurrence 21 of VTE events in patients who had completed a first period of anticoagulant therapy. 22 23 24 9.3 Comparison assessed 25 26 Experimental Control 27 NCT01136408 Dabigatran Etexilate Warfarin 28 29 30 31 9.4 Results 32 33 34 35 Study TE seTE Major bleeding OR 95%−CI W(fixed) 36 37 New study 38 NCT01136408 −1.23 1.2523 0.29 [0.02; 3.39] 0.2% Fixed effect model 0.29 [0.02; 3.39] 0.2% 39 Heterogeneity: not applicable for a single study 40 41 Original study RE−NOVATE 0.05 0.2854 1.05 [0.60; 1.85] 3.8% 42 RE−MODEL 0.06 0.4031 1.06 [0.48; 2.34] 1.9% 43 RE−NOVATE II 0.43 0.4189 1.54 [0.68; 3.50] 1.8% 44 RE−MOBILIZE −0.96 0.4531 0.38 [0.16; 0.93] 1.5% 45 BISTRO II 0.41 0.3323 1.50 [0.78; 2.88] 2.8% RE−DEEM 0.56 0.5925 1.75 [0.55; 5.58] 0.9% 46 RE−SONATE 1.97 1.4143 7.20 [0.45; 115.07] 0.2% 47 Fuji 0.55 0.8869 1.73 [0.30; 9.83] 0.4% 48 RE−LY −0.14 0.0639 0.87 [0.76; 0.98] 76.5% 49 PETRO 1.16 1.4632 3.20 [0.18; 56.37] 0.1% RE−COVER −0.19 0.3042 0.83 [0.46; 1.50] 3.4% 50 RE−MEDY −0.64 0.3272 0.53 [0.28; 1.00] 2.9% 51 RE−COVER II −0.38 0.3311 0.69 [0.36; 1.31] 2.9% 52 RE−ALIGN 0.52 0.7184 1.68 [0.41; 6.85] 0.6% Fixed effect model 0.88 [0.79; 0.99] 99.8% 53 Heterogeneity: I−squared=24.3%, tau−squared=0.0323, p=0.1914 54 55 Fixed effect model 0.88 [0.79; 0.98] 100% 56 Heterogeneity: I−squared=22%, tau−squared=0.0304, p=0.2086 57 0.01 0.1 1 10 100 58 59 Favors Dabigatran Favors Control 60
October 19, 2016 EP – CEC – PhR 15
https://mc.manuscriptcentral.com/bmj Page 57 of 70 BMJ
1 2 3 4 5 6 10 Systematic review 10 7 8 10.1 Title of the systematic review 9 BiologicConfidential: Therapies in Rheumatoid Arthritis andFor the Risk Review of Opportunistic Infections: Only A Meta-analysis 10 11 12 10.2 Inclusion criteria 13 A randomized trial of a biologic agent was considered eligible if it fulfilled all of the following conditions: 14 15 • randomized patients with RA, 16 17 • randomized Food and Drug Administration?approved biologic agents for treatment of RA, 18 • compare the effect of a biologic agent with that of a control drug, and, 19 20 • provided safety data to calculate ?1 outcome of interest. 21 22 The control arm included either placebo or disease-modifying antirheumatic drugs/conventional ther- 23 apy. Low-dose corticosteroids (<10 mg equivalent to prednisolone) were permitted in all arms. 24 A study was considered ineligible if it included 25 26 • no data on OIs; 27 28 • compared different dosing, schemes, or routes of the same biologic agent; 29 • randomized 2 biologic agents; 30 31 • or included agents not approved for RA. 32 33 10.3 Comparison assessed 34 35 36 Biologic Control 37 NCT00048581 Abatacept Placebo 38 NCT00048932 Abatacept Placebo 39 NCT00420199 Abatacept + MTX Placebo + MTX 40 NCT00533897 Abatacept Placebo 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
October 19, 2016 EP – CEC – PhR 16
https://mc.manuscriptcentral.com/bmj BMJ Page 58 of 70
1 2 3 4 5 6 10.4 Results 7 8 9 Confidential: For Review Only 10 11 12 13 Biologic Control OIs 14 Study Events Total Events Total OR 95%−CI W(fixed) 15 New study 16 NCT00048581 4 258 3 133 0.67 [0.14; 3.24] 5.9% 17 NCT00048932 2 959 2 482 0.47 [0.06; 3.80] 3.4% 18 NCT00420199 1 27 0 23 6.37 [0.12; 325.18] 0.9% NCT00533897 3 40 8 80 0.74 [0.20; 2.75] 8.5% 19 Fixed effect model 10 1284 13 718 0.74 [0.31; 1.79] 18.7% 20 Heterogeneity: I−squared=0%, tau−squared=0, p=0.7193 21 22 Original study Abe 2006 1 100 0 47 4.35 [0.07; 290.79] 0.8% 23 Durez 2004 1 12 0 15 9.49 [0.18; 489.97] 0.9% 24 Maini 1999 2 340 0 88 3.53 [0.11; 109.40] 1.2% 25 Schiff 2008 5 165 0 110 5.43 [0.89; 32.91] 4.5% 26 St Clair 2004 4 722 0 282 4.03 [0.45; 35.82] 3.1% Westhovens 2006 4 721 0 363 4.52 [0.56; 36.13] 3.4% 27 Zhang 2006 1 87 0 86 7.30 [0.14; 368.15] 1.0% 28 Combe 2009 1 204 0 50 3.47 [0.03; 480.27] 0.6% 29 Emery 2008 0 265 1 263 0.13 [0.00; 6.77] 1.0% Kim 2012 3 197 0 103 4.63 [0.42; 50.62] 2.6% 30 Lan 2004 1 29 0 29 7.39 [0.15; 372.38] 1.0% 31 O'Dell 2013 2 175 0 178 7.56 [0.47; 121.36] 1.9% 32 Smoen 2013 2 404 5 200 0.17 [0.04; 0.85] 5.8% 33 Bejarano 2008 0 75 2 73 0.13 [0.01; 2.10] 1.9% Breedveld 2006 1 542 0 257 4.37 [0.07; 290.06] 0.8% 34 Chen 1008 4 35 0 12 4.21 [0.41; 42.99] 2.7% 35 Furst 2003 1 318 0 318 7.39 [0.15; 372.38] 1.0% 36 Kavanaugh 2013 2 515 3 517 0.67 [0.12; 3.89] 4.7% 37 Keystone 2004 3 419 0 200 4.40 [0.39; 49.69] 2.5% Kim 2007 1 65 0 63 7.17 [0.14; 361.27] 1.0% 38 Takeuchi 2013 0 171 1 163 0.13 [0.00; 6.50] 1.0% 39 Choy 2012 4 126 2 121 1.90 [0.38; 9.55] 5.6% 40 Keystone 2008 5 783 0 199 3.52 [0.40; 31.32] 3.1% Smolen 2009 5 492 0 127 3.55 [0.40; 31.31] 3.1% 41 Weinblatt 2012 2 851 0 212 3.49 [0.11; 112.21] 1.2% 42 Emery 2009 1 477 0 160 3.80 [0.04; 348.87] 0.7% 43 Smolen 2012 2 306 0 155 4.53 [0.24; 85.34] 1.7% 44 Bresnihan 1998 1 351 0 121 3.84 [0.04; 341.61] 0.7% Kremer 2005 0 220 1 119 0.06 [0.00; 3.52] 0.9% 45 Kremer 2006 2 433 1 219 1.01 [0.09; 11.15] 2.5% 46 Weinblatt 2006 0 856 2 418 0.05 [0.00; 0.91] 1.7% 47 Genovese 2008 4 803 0 413 4.56 [0.57; 36.23] 3.4% Jones 2010 5 286 4 284 1.24 [0.33; 4.64] 8.4% 48 Kremer 2011 1 797 0 393 4.45 [0.07; 287.30] 0.8% 49 Nishimoto 2004 1 109 0 53 4.42 [0.07; 288.21] 0.8% 50 Smolen 2008 1 418 0 204 4.43 [0.07; 287.96] 0.8% 51 Tak 2012 2 499 1 249 1.00 [0.09; 11.05] 2.5% Fixed effect model 75 13368 23 6864 1.79 [1.17; 2.74] 81.3% 52 Heterogeneity: I−squared=2.9%, tau−squared=0.0527, p=0.4191 53 54 Fixed effect model 85 14652 36 7582 1.52 [1.04; 2.23] 100% 55 Heterogeneity: I−squared=3.8%, tau−squared=0.062, p=0.4025 56 0 0.1 1 10 1000 57 58 59 Favors Biologic Favors Control 60
October 19, 2016 EP – CEC – PhR 17
https://mc.manuscriptcentral.com/bmj Page 59 of 70 BMJ
1 2 3 4 5 6 11 Systematic review 11 7 8 11.1 Title of the systematic review 9 AngiotensinConfidential: converting enzyme (ACE) inhibitors For versus angiotensinReview receptor blockers Only for primary hyper- 10 tension 11 12 13 11.2 Inclusion criteria 14 Studies: 15 16 • directly compared an ACE inhibitor and an ARB; 17 18 • randomized participants to the ACE inhibitor group or the ARB group; 19 • had the same protocol regarding continuation or discontinuation of pre-study blood pressure lowering 20 therapy in both arms; 21 22 • had the same protocol for adding background blood pressure lowering therapy during the trial in 23 both arms; 24 25 • had a prespecified duration of at least one year; 26 27 • were double blinded when included for WDAE. 28 29 11.3 Comparison assessed 30 31 ARBs ACE inhibitors 32 33 NCT00433836 Valstartan Enalapril 34 NCT00446511 Valstartan Enalapril 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
October 19, 2016 EP – CEC – PhR 18
https://mc.manuscriptcentral.com/bmj BMJ Page 60 of 70
1 2 3 4 5 6 11.4 Results 7 8 9 Confidential: For Review Only 10 ARBs ACE inhib. Withdrawals due to AE 11 Study Events Total Events Total RR 95%−CI W(fixed) 12 13 New study NCT00433836 7 151 0 148 14.70 [0.85; 255.14] 0.1% 14 NCT00446511 3 103 1 109 3.17 [0.34; 30.03] 0.2% 15 Fixed effect model 10 254 1 257 7.12 [1.28; 39.57] 0.2% 16 Heterogeneity: I−squared=0%, tau−squared=0, p=0.3883 17 18 Original study 19 Fogari 2011 3 132 7 130 0.42 [0.11; 1.60] 1.1% ONTARGET 2008 465 4711 535 4687 0.86 [0.77; 0.97] 86.1% 20 Bremner 1997 37 334 30 167 0.62 [0.40; 0.96] 6.4% 21 DETAIL 2004 16 100 24 102 0.68 [0.38; 1.20] 3.8% 22 Fogari 2008 1 122 5 124 0.20 [0.02; 1.71] 0.8% 23 Fogari 2012 3 102 7 103 0.43 [0.12; 1.63] 1.1% 24 Lacourciere 2000 2 52 1 51 1.96 [0.18; 20.97] 0.2% 25 Spoelstra−de 2006 3 24 1 22 2.75 [0.31; 24.52] 0.2% 26 Fixed effect model 530 5577 610 5386 0.83 [0.74; 0.93] 99.8% Heterogeneity: I−squared=11.5%, tau−squared=0.013, p=0.341 27 28 Fixed effect model 540 5831 611 5643 0.85 [0.76; 0.94] 100% 29 Heterogeneity: I−squared=31.7%, tau−squared=0.058, p=0.1549 30 31 0.01 0.1 1 10 100 32 Favors ARBs Favors ACE inhib. 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
October 19, 2016 EP – CEC – PhR 19
https://mc.manuscriptcentral.com/bmj Page 61 of 70 BMJ
1 2 3 4 5 6 12 Systematic review 12 7 8 12.1 Title of the systematic review 9 BivalirudinConfidential: versus heparin in patients planned For for percutaneous Review coronary intervention: Only a meta-analysis of 10 randomised controlled trials 11 12 13 12.2 Inclusion criteria 14 Trials were included if they enrolled individuals with planned PCI and randomly assigned patients to 15 treatment with bivalirudin (using the approved dosing regimen) or heparin (mostly unfractionated heparin 16 [UFH], but also low-molecularweight heparin) with or without a GPI. Trials that did not report clinical 17 outcomes, involved fi brinolytics, were done before coronary stenting was available, or compared bivalirudin 18 with anticoagulant regimens other than heparin or low-molecular-weight heparin were excluded from the 19 analysis. 20 21 22 12.3 Comparison assessed 23 24 Experimental Control 25 NCT00464087 Bivalirudin Heparin 26 27 28 29 12.4 Results 30 31 32 33 34 Bivalirudin Heparin Major bleeding 35 Study Events Total Events Total RR 95%−CI W(random) 36 New study 37 NCT00464087 1 51 0 49 2.88 [0.12; 69.11] 1.5% 38 Random effects model 1 51 0 49 2.88 [0.12; 69.11] 1.5% 39 Heterogeneity: not applicable for a single study 40 41 Original study 42 ISAR−REACT 3 70 2289 104 2281 0.67 [0.50; 0.90] 38.6% 43 ARMYDA−7 BIVALVE2 2 198 6 203 0.34 [0.07; 1.67] 5.5% BRIGHT (heparin alone) 4 729 11 725 0.36 [0.12; 1.13] 9.7% 44 HEAT PPCI 32 905 28 907 1.15 [0.70; 1.89] 27.7% 45 NAPLES III 14 418 11 419 1.28 [0.59; 2.78] 17.0% 46 Random effects model 122 4539 160 4535 0.79 [0.52; 1.19] 98.5% 47 Heterogeneity: I−squared=47.2%, tau−squared=0.093, p=0.1082 48 49 Random effects model 123 4590 160 4584 0.80 [0.54; 1.20] 100% Heterogeneity: I−squared=39.3%, tau−squared=0.083, p=0.1435 50 51 0.1 0.5 1 2 10 52 53 Favors Bivalirudin Favors Heparin 54 55 56 57 58 59 60
October 19, 2016 EP – CEC – PhR 20
https://mc.manuscriptcentral.com/bmj BMJ Page 62 of 70
1 2 3 4 5 6 13 Systematic review 13 7 8 13.1 Title of the systematic review 9 TreatmentConfidential: Discontinuations With New Oral AgentsFor for Long-term Review Anticoagulation: Only Insights from a Meta- 10 Analysis of 18 Randomized Trials Including 101 801 patients. 11 12 13 13.2 Inclusion criteria 14 Studies included compared NOACs with conventional anticoagulants or placebo for the treatment of 15 VTE/pulmonary embolism (PE), ACS, and stroke prevention in patients with AF. The included studies 16 had to have at least 12 weeks of follow-up. Studies of orthopedic operations were not included. Both 17 double-blind and open-label trials were eligible for inclusion. 18 19 20 13.3 Comparison assessed 21 22 NOACs Control 23 NCT00852397 Apixaban Placebo 24 25 26 27 13.4 Results 28 29 30 31 32 NOACs Control Treatment D/C due to all cause 33 Study Events Total Events Total RR 95%−CI W(random) 34 35 New study NCT00852397 22 99 11 52 1.05 [0.55; 1.99] 8.6% 36 Random effects model 22 99 11 52 1.05 [0.55; 1.99] 8.6% 37 Heterogeneity: not applicable for a single study 38 Original study 39 APPRAISE 2009 480 1104 84 611 3.16 [2.56; 3.90] 17.5% 40 APPRAISE−2 2011 863 3705 748 3687 1.15 [1.05; 1.25] 19.6% 41 ATLAS ACS−TIMI 46 2009 347 1823 149 907 1.16 [0.97; 1.38] 18.2% 42 ATLAS ACS 2−TIMI 51 2012 2914 10350 1355 5176 1.08 [1.02; 1.14] 19.8% RE−DEEM 2011 270 1505 53 373 1.26 [0.96; 1.66] 16.2% 43 Random effects model 4874 18487 2389 10754 1.40 [1.08; 1.82] 91.4% 44 Heterogeneity: I−squared=95.9%, tau−squared=0.0829, p<0.0001 45 46 Random effects model 4896 18586 2400 10806 1.37 [1.06; 1.75] 100% Heterogeneity: I−squared=94.8%, tau−squared=0.0811, p<0.0001 47 48 0.5 1 2 49 50 Favors NOACs Favors Control 51 52 53 54 55 56 57 58 59 60
October 19, 2016 EP – CEC – PhR 21
https://mc.manuscriptcentral.com/bmj Page 63 of 70 BMJ
1 2 3 4 5 6 14 Systematic review 14 7 8 14.1 Title of the systematic review 9 SafetyConfidential: and efficacy of addition of VEGFR and For EGFR-family Review oral small-molecule tyrosineOnly kinase inhibitors 10 to cytotoxic chemotherapy in solid cancers: A systematic review and meta-analysis of randomized con- 11 trolled trials 12 13 14 14.2 Inclusion criteria 15 Clinical trials that met the following criteria were included: 16 17 • phase II and III trials in patients with solid cancers; 18 19 • random assignment of participants to treatment with chemotherapy plus VEGFR or EGFR-targeted 20 TKI or chemotherapy alone; 21 • reporting data for at least one of the safety or efficacy outcomes. 22 23 24 14.3 Comparison assessed 25 26 TKI Control 27 NCT00447057 Pemetrexed + Erlotinib Pemetrexed 28 NCT00486954 Paclitaxel + Lapatinib Paclitaxel 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
October 19, 2016 EP – CEC – PhR 22
https://mc.manuscriptcentral.com/bmj BMJ Page 64 of 70
1 2 3 4 5 6 14.4 Results 7 8 9 Confidential: For Review Only 10 11 12 13 14 15 TKI Control Fatal adverse event 16 Study Events Total Events Total RR 95%−CI W(random) 17 New study 18 NCT00447057 2 102 1 102 2.00 [0.18; 21.71] 0.8% 19 NCT00486954 0 131 1 129 0.33 [0.01; 7.98] 0.4% 20 Random effects model 2 233 2 231 1.05 [0.15; 7.07] 1.2% 21 Heterogeneity: I−squared=0%, tau−squared=0, p=0.3731 22 Original study 23 Spano 2008 1 68 0 31 1.38 [0.06; 32.93] 0.4% 24 Kindler 2011 5 305 2 308 2.52 [0.49; 12.91] 1.6% Rugo 2011 1 111 0 56 1.52 [0.06; 36.72] 0.4% 25 Hauschild 2009 4 134 0 134 9.00 [0.49; 165.53] 0.5% 26 Scagliotti 2010 13 463 4 459 3.22 [1.06; 9.81] 3.5% 27 Abou−Alfa 2010 2 47 2 49 1.04 [0.15; 7.10] 1.2% 28 Molina 2011 3 49 3 51 1.04 [0.22; 4.91] 1.8% Baselga 2012 0 112 2 112 0.20 [0.01; 4.12] 0.5% 29 Paz−Ares 2012 5 385 2 384 2.49 [0.49; 12.77] 1.6% 30 Goncalves 2012 12 50 6 52 2.08 [0.85; 5.12] 5.3% 31 Flaherty 2012 9 393 10 397 0.91 [0.37; 2.21] 5.4% Schwartzberg 2012 1 79 0 77 2.92 [0.12; 70.70] 0.4% 32 Gradishar 2012 9 115 3 118 3.08 [0.85; 11.08] 2.6% 33 Crown 2010 14 217 9 215 1.54 [0.68; 3.48] 6.5% 34 Bergh 2012 2 295 0 293 4.97 [0.24; 103.00] 0.5% 35 Heist 2012 4 41 0 41 9.00 [0.50; 161.92] 0.5% Carrato 2013 12 384 4 379 2.96 [0.96; 9.10] 3.4% 36 Heymach 2007 4 86 2 41 0.95 [0.18; 5.00] 1.6% 37 Heymach 2008 4 56 0 52 8.36 [0.46; 151.61] 0.5% 38 Herbst 2010 42 689 38 690 1.11 [0.72; 1.69] 23.7% Boer 2010 0 33 1 29 0.29 [0.01; 6.93] 0.4% 39 de Boer 2011 14 260 12 273 1.22 [0.58; 2.60] 7.6% 40 Choueiri 2011 1 70 0 72 3.09 [0.13; 74.47] 0.4% 41 Herbst 2005 33 209 15 208 2.19 [1.23; 3.91] 12.8% 42 Gatzemeier 2007 8 580 1 579 7.99 [1.00; 63.65] 1.0% Moore 2007 6 282 0 280 12.91 [0.73; 228.04] 0.5% 43 Mok 2009 1 74 2 79 0.53 [0.05; 5.76] 0.8% 44 Pawel 2011 2 76 0 83 5.46 [0.27; 111.88] 0.5% 45 Stinchcombe 2011 2 51 0 44 4.32 [0.21; 87.63] 0.5% 46 Mok 2012 11 226 6 222 1.80 [0.68; 4.79] 4.5% Herbst 2004 4 684 2 341 1.00 [0.18; 5.42] 1.5% 47 Argiris 2013 6 124 2 129 3.12 [0.64; 15.17] 1.7% 48 Di Leo 2009 8 293 2 286 3.90 [0.84; 18.23] 1.8% 49 Cameron 2010 4 207 6 191 0.62 [0.18; 2.15] 2.8% Random effects model 247 7248 136 6755 1.63 [1.32; 2.01] 98.8% 50 Heterogeneity: I−squared=0%, tau−squared=0, p=0.6001 51 52 Random effects model 249 7481 138 6986 1.62 [1.32; 1.99] 100% 53 Heterogeneity: I−squared=0%, tau−squared=0, p=0.6467 54 0.01 0.1 1 10 100 55 56 57 Favors TKI Favors Control 58 59 60
October 19, 2016 EP – CEC – PhR 23
https://mc.manuscriptcentral.com/bmj Page 65 of 70 BMJ
1 2 3 4 5 6 15 Summary 7 8 Systematic Proportion Summary statistic Change in Direction of 9 ReviewConfidential:of patients For Review summaryOnlychange 10 added in statistic 11 meta- (%) 12 analysis 13 (%) 14 Without new studies With new studies 15 16 1 1 OR 0.51 [0.36 ; 0.70] OR 0.53 [0.38 ; 0.73] 6 Decrease 17 2 19 RR 18.28 [12.76 ; 26.17] RR 14.20 [10.72 ; 18.81] 9 Decrease 18 3 19 HR 0.87 [0.82 ; 0.91] HR 0.88 [0.84 ; 0.93] 8 Decrease 19 4 51 HR 0.89 [0.80 ; 0.99] HR 0.90 [0.83 ; 0.98] 10 Decrease 20 5 2 HR 0.99 [0.90 ; 1.09] HR 0.99 [0.90 ; 1.08] 0 No change 21 6 8 SMD -0.21 [-0.29 ; -0.12] SMD -0.19 [-0.28 ; -0.11] 10 Decrease 22 7 23 MD -1.77 [-2.13 ; -1.41] MD -1.66 [-1.99 ; -1.32] 6 Decrease 23 8 112 MD -0.35 [-0.51 ; -0.19] MD -0.45 [-0.55 ; -0.36] 29 Increase 24 9 0 OR 0.88 [0.79 ; 0.99] OR 0.88 [0.79 ; 0.98] 0 No change 25 10 10 OR 1.79 [1.17 ; 2.74] OR 1.52 [1.04 ; 2.23] 28 Less harm 26 11 5 RR 0.83 [0.74 ; 0.93] RR 0.85 [0.76 ; 0.94] 13 More harm 27 12 1 RR 0.79 [0.52 ; 1.19] RR 0.80 [0.54 ; 1.20] 5 More harm 28 13 1 RR 1.40 [1.08 ; 1.82] RR 1.37 [1.06 ; 1.75] 6 Less harm 29 14 3 RR 1.63 [1.32 ; 2.01] RR 1.62 [1.32 ; 2.99] 1 Less harm 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
October 19, 2016 EP – CEC – PhR 24
https://mc.manuscriptcentral.com/bmj BMJ Page 66 of 70
1 2 3 BMJ 4 Licence to BMJ Publishing Group Limited (“BMJ Group”) for Publication 5 To be agreed to by the corresponding author or guarantor on behalf of all authors ("Corresponding 6 Author"). All authors collectively are referred to as the “Contributors” 7 8 Confidential:In consideration of the BMJ Group (“the For Publishers” ) consideringReview to publish the articleOnly contained 9 within the original manuscript which includes without limitation any diagrams, photographs, other illustrative material, video, film or any other material howsoever submitted by the Contributor(s) at any 10 time and related to the Contribution (“the Contribution”) in the BMJ (“the Journal”), certain rights are 11 required to be granted by each different category of author(s), which are as follows: 12 13 1. For employees of the UK Crown acting in the course of their employment - a non exclusive 14 Licence, as set out below. All provisions of this document apply. The non exclusivity relates to the 15 original submitted manuscript video, films, images, photographs, diagrams and/or illustrative 16 material only). 17 18 2. For employees of the US Federal Government employees acting in the course of their 19 employment , no copyright exists and the Contribution is in the public domain so no licence is 20 required to be granted. The Author Warranties below apply (excluding 1.iii). 21 22 3. For all other authors, an exclusive Licence, as set out below. All provisions of this document apply. 23
24 NB where a Contribution is a multi authored work, each author’s element of the Contribution will be 25 dealt with in accordance with 1, 2 or 3 above, as applicable. 26 27 The Licence 28 The Licence granted in accordance with 1 or 3 above is: 29 A worldwide, licence, to the Publishers and its licensees in perpetuity (subject to the Reversion of 30 Rights set out below), in all forms, formats and media (whether known now or created in the future), to 31 i) publish, reproduce, distribute, display and store the Contribution, ii) translate the Contribution into 32 other languages, create adaptations, reprints, include within collections and create summaries, extracts and/or, abstracts of the Contribution and convert or allow conversion into any format including without 33 limitation audio, iii) create any other derivative work(s) based in whole or part on the Contribution, iv) 34 to exploit all subsidiary rights that currently exist or as may exist in the future in the Contribution, v) 35 the inclusion of electronic links from the Contribution to third party material where-ever it may be 36 located; and, vi) licence any third party to do any or all of the above. 37 38 If you and/or any co-author’s employer owns the copyright to your contribution you must obtain in 39 writing the relevant employers’ consent to grant the licence and agree to all obligations herein. The 40 author(s) hereby agree that in the event that the BMJ Group sell, the whole or part of its journal 41 business to any third party, the benefit and the burden of the Licence contained herein shall be assigned 42 to that third party. 43 44 Additional Rights and Obligations 45 The author(s) (and their employers as applicable), hereby authorise the Publishers to take such steps as they consider necessary at their own expense in the copyright owners name and on their behalf, if they 46 believe that a third party is infringing or is likely to infringe copyright or the rights granted to the 47 Publishers herein in the Contribution without further recourse to the copyright owner(s). 48 49 For Original Research articles and Open Access Funded Articles (as both defined below), the 50 Publishers expressly agree to place the published Contribution for display on PubMed Central 51 (including its international mirror sites) promptly without charge to the authors or their employers 52 (provided Pubmed Central does not charge the Publishers), which will include any Publishers’ supplied 53 amendments or retractions . 54 55
56
57 58 59 60 24349793_File000007_539274517.doc https://mc.manuscriptcentral.com/bmj Page 67 of 70 BMJ
1 2 3 “Original Research ” means an article reporting a research study, with a research structured abstract 4 and normally appearing in the Research section of the BMJ. “Open Access Funded Articles ” means 5 articles funded in whole or part by a research grant from a government and/or charitable organisation(s) 6 that requires open access deposit in PubMed Central. Such articles are identified by the reference to a 7 Creative Commons licence.
8 Confidential:The author(s) acknowledge and accept that ForBMJ Group may Review make additional changes toOnly the 9 contribution as considered necessary in accordance with standard editorial processes whether before or 10 after publication. The Corresponding Author will usually see proofs for their Contribution s and every 11 effort will be made to consult with the Corresponding Author if substantial alterations are made. The 12 BMJ Group may also retract or publish a correction or other notice when it considers this appropriate 13 for legal or editorial reasons and this shall be at its absolute discretion which shall be exercised 14 reasonably. 15 16 Reversion of Rights 17 If the Contribution is not published in either the print or electronic versions of the Journal or any other 18 Publisher(s) products , within 12 months of final acceptance by the BMJ Group, (or as otherwise agreed), any Licence granted herein shall automatically terminate and all rights shall revert to the 19 copyright owner. The Publishers may keep a copy of the Contribution as a record (including via any 20 contractor). 21 22 Rights Granted to Owners of the Contribution 23 Ownership of copyright remains with the author(s) or their employers. All rights not expressly granted 24 are, subject to the Licence terms, reserved by the Publishers. In return for the grant of the Licence 25 herein, the copyright owner(s) shall have the following rights for non -Commercial Use (unless 26 otherwise stated) of the Contribution: 27 1.The right to reproduce a reasonable number (no more than 100) print copies of the final 28 Contribution, by copying or downloading from the BMJ Group website, for personal use and to send 29 copies to colleagues in print or electronic form provided no fee is charged and this is not done on a 30 systematic basis ( which includes via mass e-mailings). 31 32 2. The right to include the Contribution in a compilation for classroom use (course packs) to be 33 distributed free of charge (other than for direct photocopying cost) to students at the Contributor(s)’s 34 institution or to be stored in digital format in data rooms for access by students as part of their course 35 work and for in house training programmes of the Contributor(s)’s employer or at seminars or 36 conferences subject to a limit of 100 copies per conference or seminar . 37 38 3. For all articles (excluding articles commissioned by the Publishers), the right to post a version of 39 the final published version of the Contribution, or any abstract of the final published Contribution on 40 the Contributor(s)’s own and/or his/her institution’s website after the Publisher’s publication. 41 4. For all Publisher commissioned articles, the right to post a version of the final published version of 42 the Contribution, or any abstract of the final published Contribution on the Contributor(s)’s own and/or 43 his/her institution’s website 12 months after publication. 44 45 5. The following statement must accompany the articles posted on the Contributor(s)’s and/or his/her 46 institution’s website: 47 48 “This article has been published in the BMJ [insert full citation reference] and can also be viewed on the journal’s website at www.bmj.com“ 49
50 51 6. In addition, for Original Research articles and Open Access Funded Articles copyright owners (and 52 the Publishers) may and may allow third parties to use the Contribution in accordance one of the 53 following Creative Commons licences depending on the source of the research funding as per below: 54 a) where the Original Research article and/or Open Access Funded Articles is not funded by the 55 Wellcome Trust or UK Research Council, the artciles may be re-used under the terms of the Creative 56 Commons Attribution-Non Commercial 3.0 Unported (CC BY-NC 3.0) see: 57 http://creativecommons.org/licenses/by-nc/3.0/ 58 59 60 24349793_File000007_539274517.doc https://mc.manuscriptcentral.com/bmj BMJ Page 68 of 70
1 2 3 and 4 http://creativecommons.org/licenses/by-nc/3.0/legalcode 5 6 or any updated versions as determined by the Publisher from time to time. 7 or 8 Confidential: For Review Only b) where the Original Research article and/or Open Access Funded Articles is funded by the Wellcome 9 Trust or UK Research Council, the Contribution may be re-used under the terms of the Creative 10 Commons Attribution 3.0 Unported Licence (CC BY 3.0) see: 11 12 http://creativecommons.org/licenses/by/3.0/ 13 and 14 15 http://creativecommons.org/licenses/by/3.0/legalcode 16 or any updated versions as determined by the Publisher from time to time 17 18 subject to ensuring the Publishers and the Journal are referenced (including a full citation) as set out 19 above; all third party rights within all images, diagrams, photograph, other illustrative material or 20 films, not owned by the authors or BMJ Group are cleared independently and appropriately; and all the 21 Publishers trademarks are removed from any derivative works and ensuring any translations, for which 22 a prior translation agreement with BMJ Group has not been established, must prominently display the statement: “This is an unofficial translation of an article that appeared in a BMJ Group publication. 23 BMJ Group has not endorsed this translation”. 24 25 7. The right to publish with the necessary acknowledgement of the Publishers and the Journal, all or 26 part of the material from the published Contribution in a book, essay, position paper, or other non peer 27 reviewed publication authored or edited by the Contributor(s)’s (which may be a Commercial Use). 28 This does not apply to multiple Contributions in the same journal, for which permission from the 29 Publishers must be sought. 30 31 8. The right to use selected figures and tables and (of which the author or his employer owns or has 32 licensed) and selected text (up to 300 words) from the Contribution for incorporation within another work published in print or digital format by a third party, so long as full credit is given to the Publishers 33 and use of the parts of the Contribution is non Commercial Use. 34 35 9. Subject to it not being contrary to English law to do so (such as for example where the UK has 36 trading or other bans with the country of the Corresponding Authors origin or certain groups of people 37 within), the right to receive a royalty for up to 5 years from publication of 10% of any net receipts less 38 sales commission on single orders in excess of £2000 received by the Publisher for any single 39 Contribution reprint or translation sales to a single third party, subject however to any fee being 40 determined (if charged) at the absolute discretion of the Publishers as may be altered from time to time. 41 If the Publishers receive such an order for reprint sales of the Contribution, they will contact the 42 Corresponding Author at the address given on the published Contribution to find out to whom payment 43 should be made. Corresponding Authors have the responsibility to ensure that all authors have agreed 44 what should be done with any such royalty payment and to keep the Publisher updated with current contact details. 45
46 For permission to use materials that are beyond permitted here, visit http://www.bmj.com/about- 47 bmj/resources-readers/permissions 48 49 “Commercial Use” includes: 50 - copying or downloading of documents, or linking to such postings, for further 51 redistribution, sale or licensing, for a fee; 52 - copying, downloading or posting by a site or service that incorporates advertising 53 with such content; 54 - the inclusion or incorporation of document content in other works or services (other 55 than for legally permitted quotations with an appropriate citation) that is then 56 available for sale or licensing, for a fee. 57 - use of documents or document content (other than for legally permitted quotations with appropriate citations) by organisations for any promotional or advertising 58 59 60 24349793_File000007_539274517.doc https://mc.manuscriptcentral.com/bmj Page 69 of 70 BMJ
1 2 3 purposes whether direct or indirect, whether for a fee or otherwise. Distribution by 4 or on behalf of pharmaceutical organisations is considered in all cases as Commercial 5 Use; 6 - use for the purposes of monetary reward by means of sale, resale, license, loan, hire 7 transfer or other form of commercial exploitation.
8 Confidential: For Review Only Author warranties 9 1. The author(s) warrant that: i) they are the sole author(s) of the Contribution which is an original 10 work; ii) the whole or a substantial part of the Contribution has not previously been published; iii) they 11 or their employers are the copyright owners of the Contribution; iv) to the best of their knowledge that 12 the Contribution does not contain anything which is libellous, illegal or infringes any third party’s 13 copyright or other rights; v) that they have obtained all necessary written consents for any patient 14 information which is supplied with the Contribution and vi) that they have declared or will accurately 15 declare all competing interests to the Publisher. 16 17 Anti Bribery 18 As a service provider to the BMJ Group, you agree that you shall: (a)comply with all applicable laws, 19 statutes, regulations and codes relating to anti-bribery and anti-corruption including but not limited to 20 the Bribery Act 2010 (Relevant Requirements); b) not engage in any activity, practice or conduct which would constitute an offence under sections 1, 2 or 6 of the Bribery Act 2010 (as amended) if 21 such activity, practice or conduct had been carried out in the UK; (c) comply with any Publisher Ethics 22 and Anti-bribery Policy supplied to you from time to time including as contained as follows (Relevant 23 Policies): 24 http://group.bmj.com/group/about/corporate/Anti-Bribery%20and%20Corruption%20Policy%20- 25 August%202012.pdf; 26 (d) promptly report to the Chief Executive Officer or Chairman of the Publisher any request or demand 27 for any undue financial or other advantage of any kind received by you in connection with the 28 performance of this Agreement; Breach of this Clause shall be deemed a material breach of this 29 Agreement. 30 31 Law and Jurisdiction 32 To the fullest extent permitted by law, this Agreement will be governed by the laws of England and shall be governed and construed in accordance with the laws of England whose courts shall have 33 exclusive jurisdiction, unless as at the date of formation of this Agreement either i) an English 34 judgement could not be enforced in the Corresponding Author’s stated country location; or ii) it would 35 take six months or more for the BMJ Group to enforce an English judgement in the Corresponding 36 Author’s stated country location, then it is hereby agreed that this Agreement shall be governed by the 37 laws of the Corresponding Author’s stated country (or state if applicable) and their courts shall have 38 jurisdiction. Notwithstanding any of the above, this clause is governed by the laws of England. 39 40 The following statement must be included in your manuscript, together with the relevant tick box 41 line below: 42 “I, Amélie Yavchitz, The Corresponding Author of this article contained within the original manuscript 43 which includes any diagrams & photographs within and any related or stand alone film submitted (the 44 Contribution”) has the right to grant on behalf 45 of all authors and does grant on behalf of all authors, a licence to the BMJ Publishing Group Ltd and its licencees, to permit this Contribution (if accepted) to be published in the BMJ and any other BMJ 46 Group products and to exploit all subsidiary rights, as set out in our licence set out at: 47 http://www.bmj.com/about-bmj/resources-authors/forms-policies-and-checklists/copyright-open- 48 access-and-permission-reuse.” 49 50 IF YOU ARE A NATIONAL INSTITUTE OF HEALTH (“NIH”) EMPLOYEE, CONTRACTOR OR 51 TRAINEE the following cover sheet will be accepted by the BMJ Group and NIH and incorporated into 52 the above Licence. 53 54 Please tick one or more boxes as appropriate: 55 I am the sole author of the Contribution. 56 X I am one author signing on behalf of all co-owners of the Contribution. 57 The Contribution has been made in the course of my employment and I am signing as authorised by my employer. 58 59 60 24349793_File000007_539274517.doc https://mc.manuscriptcentral.com/bmj BMJ Page 70 of 70
1 2 3 I am a US Federal Government employee acting in the course of my employment. 4 I am not a US Federal Government employee, but some or all of my co-authors are. 5 I am an employee of the UK Crown* acting in the course of my employment 6 I am a US Federal Government employee acting in the course of my employment. 7 I am not a US Federal Government employee, but some or all of my co-authors are. I am an employee of the UK Crown acting in the course of my employment 8 Confidential: For Review Only I am not an employee of the UK Crown acting in the course of my employment but some/all 9 of my co-authors are.* 10 11 *Such authors should consult guidance and if necessary return any completed form. 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 24349793_File000007_539274517.doc https://mc.manuscriptcentral.com/bmj