A multicentre randomiSed controlled TRial of IntraVEnous immunoglobulin compared with standard therapy for the treatment of transverse myelitis in adults and children (STRIVE)

Article history

The research reported in this issue of the journal was funded by the HTA programme as project number 11/129/148. The contractual start date was in June 2014. The draft report began editorial review in October 2016 and was accepted for publication in February 2017. The authors have been wholly responsible for all data collection, analysis and interpretation, and for writing up their work. The HTA editors and publisher have tried to ensure the accuracy of the authors’ report and would like to thank the reviewers for their constructive comments on the draft document. However, they do not accept liability for damages or losses arising from material published in this report.

Declared competing interests of authors

Intravenous immunoglobulin was provided by Biotest AG, Germany, and, should any commercial opportunity arise, the industrial partner has an option for an exclusive licence from the sponsor (Guy’s and St Thomas’ NHS Foundation Trust) and the potential for a revenue-sharing arrangement. Michael Absoud serves on the data safety monitoring board for a study sponsored by Neurim Pharmaceuticals, has received consultation fees from Novartis and is on the editorial advisory board for the International Journal of Language & Communication Disorders. Peter Brex has received fees for speaking and consulting from Biogen Idec, Roche, Sanofi Genzyme, Teva Pharmaceuticals Industries Ltd and Merck Serono. Olga Ciccarelli serves as consultant for Novartis, Biogen Inc. and GE Healthcare and is an Associate Editor of Neurology. Gavin Giovannoni has received consultation and speaking fees from Biogen Idec, GlaxoSmithKline, Merck Serono, Novartis, Sanofi Genzyme and Synthon BV and is on the steering committee for studies sponsored by AbbVie, Biogen Idec, Novartis, Teva Pharmaceuticals Industries Ltd and Roche. Jackie Palace serves on the scientific advisory board for the Charcot Foundation, has performed advisory work for Biogen Idec, Merck Serono, Bayer Schering Pharma, Novartis, Teva Pharmaceuticals Industries Ltd, Gilenya, Ono Pharmaceutical Co., Primary i-Research, Alexion Pharmaceuticals Inc. and Chugai Pharma Europe, receives research support from Merck Serono, Bayer Schering Pharma, Biogen Idec and Teva Pharmaceuticals Industries Ltd and has received conference expenses from Novartis, Merck Serono and Biogen Idec. Michael Pike has received a meeting support grant from EUROIMMUN. Anu Jacob is supported by the NHS National Specialised Commissioning Group for Neuromyelitis Optica, has been a consultant for Shire, Alexion Pharmaceuticals Inc. and Chugai Pharmaceutical Co. Ltd and has received research funding from Biogen Inc. and Alexion Pharmaceuticals Inc. Ming Lim has received consultation fees from CSL Behring and travel grants from Merck Serono and has been awarded educational grants to organise meetings by Novartis, Biogen Idec, Merck Serono and Bayer.

Copyright statement

© Queen’s Printer and Controller of HMSO 2017. This work was produced by Absoud et al. under the terms of a commissioning contract issued by the Secretary of State for Health. This issue may be freely reproduced for the purposes of private research and study and extracts (or indeed, the full report) may be included in professional journals provided that suitable acknowledgement is made and the reproduction is not associated with any form of advertising. Applications for commercial reproduction should be addressed to: NIHR Journals Library, National Institute for Health Research, Evaluation, Trials and Studies Coordinating Centre, Alpha House, University of Southampton Science Park, Southampton SO16 7NS, UK.

Background

Transverse myelitis (TM) is an immune-mediated disorder of the spinal cord affecting children and adults and is usually characterised by a rapid onset of paraplegia or tetraplegia, loss of sensation and sphincter disturbance. 1 Histologically, TM is characterised by spinal cord immune cellular infiltration, with pathogenesis mediated by a variety of immunological mechanisms. 1 Attacks usually develop over 24 hours and, in some cases, can progress rapidly to a potentially devastating and life-threatening condition. The severity of symptoms depends on the spinal cord level affected, with patients with high cervical lesions often requiring intensive care support to maintain respiratory function. Patients can recover fully from TM but a large number are left with significant disabilities. Recovery occurs within weeks of the onset of symptoms and is most rapid during the first 3–6 months, although further improvement may be seen for up to 2–4 years. 2

Diagnostic criteria for TM were established by the Transverse Myelitis Consortium Working Group in 2002. 3 A proportion of patients initially diagnosed with TM will subsequently relapse, often with the involvement of other parts of the central nervous system (CNS), and may then be diagnosed with either multiple sclerosis (MS) or neuromyelitis optica (NMO). NMO is a relapsing subset of TM, caused by antibodies to aquaporin 4 (AQP4). 4 Clinically, patients have predominantly recurrent episodes of myelitis and optic neuritis. Neurodisability accrues with progressive relapses. Initial presentation may be with myelitis alone, making it clinically and radiologically indistinguishable from TM, and patients are thus subjected to the same acute therapeutic strategies.

The precise number of patients who make a full recovery from TM remains unclear. Studies carried out before the development of the Transverse Myelitis Consortium Working Group criteria may have included patients with a wider range of myelopathies, such as spinal cord infarction,5 or may reflect the greater severity of cases seen at tertiary referral centres such as the Johns Hopkins Transverse Myelitis Centre,6 where up to 20% of patients are reported to make a good recovery. Currently, the only report to reliably inform on the outcome of adult-onset TM is a retrospective French multicentre study applying Transverse Myelitis Consortium Working Group criteria, with 36% of patients with TM having a poor prognosis, defined by death or non-ambulating. 7 Approximately half of children with TM make a good recovery. 8 Hence, the majority of adults and children presenting with TM have either a fair outcome (functional and ambulatory but with varying degrees of spasticity, urgency and/or constipation and some sensory signs) or a poor outcome (remaining completely or largely unable to walk, having at best partial sphincter control and being left with severe sensory deficits). 2 These outcomes have a huge burden on patients and their carers. With conservative estimates of the incidence of TM in the UK being 350 cases per year (based on an incidence of three to seven cases per million),9,10 this clearly imposes a significant cumulative demand on health resources in the UK. Moreover, many patients are affected when they are in the prime of their working life, thus resulting in a loss of productivity and imposing a further financial impact on the country.

Importantly, strategies to reduce disability in patients are urgently required, but there have been no robust randomised controlled trials (RCTs) carried out in children or adults to inform on the optimal treatment of TM. The current clinical consensus is derived from data that are extrapolated mainly from class IV evidence from case series or clinical trials on the treatment of exacerbations of adult MS. 9–12 In adults, this evidence suggests that treatment of relapses with intravenous methylprednisolone (IVMP) shortens relapse duration and speeds up recovery. It is on this evidence that the current standard therapy for TM has been based, with both children and adults treated with high-dose intravenous (IV) steroids for 3–7 days to reduce inflammation, hasten recovery and restore neurological function. 13

Although IV steroids are now the most common treatment for TM, there are other interventions that have proved effective in aiding recovery, but which are not routinely applied. In a retrospective analysis of 122 adults with TM, acute therapies given at one centre between 2001 and 2005 were evaluated, with the finding that some patients benefited from the addition of plasma exchange (PLEX) to IVMP. 11 The efficacy of PLEX was also demonstrated in a small RCT in adults with acute CNS demyelination (including four patients with TM), in whom steroids had failed to induce a remission of symptoms. 14 However, administering PLEX is technically difficult and costly, making it challenging to deliver within the NHS and resulting in it not being universally available.

Treatment with intravenous immunoglobulin (IVIG) is also used increasingly in the management of a range of neurological conditions and its efficacy has been clearly established in RCTs for a handful of these conditions. 15 In adults and children with CNS demyelination who do not respond to steroids, IVIG is often used, although supporting data are limited to small case series and single case reports. 16,17 The most relevant actions of IVIG in the therapy of neurological diseases include (1) the inhibition of complement binding, (2) the neutralisation of pathogenic cytokines, (3) the down-regulation of antibody production and (4) the modulation of Fc-receptor-mediated phagocytosis. Additional actions include the modulation of T-cell function and the enhancement of remyelination. 18 The majority of these factors are common across inflammatory disorders of the CNS, including TM,19 providing a strong rationale for the use of IVIG in the management of TM. In addition, IVIG is cost-effective compared with PLEX and more readily accessible.

We aimed to conduct a multicentre, single-blind, parallel-group RCT to generate evidence to inform clinical and health economic decisions with regard to IVIG use in adults and children with TM.

Risks and benefits

Study aim

The randomiSed controlled TRial of IntraVEnous immunoglobulin compared with standard therapy for the treatment of transverse myelitis in adults and children (STRIVE) study was a multicentre, single-blind, parallel-group RCT aimed at generating evidence to inform clinical and health economic decisions regarding IVIG use in adults and children with TM.

The primary objective was to evaluate whether or not additional, and early, treatment with IVIG is of extra benefit in TM compared with current standard therapy with IV steroids.

The secondary objectives were to investigate the following potential benefits:

• The clinical and para-clinical data collected from patients will provide a robust resource and platform for other clinical studies, including the identification of early predictors of poor outcome.

• Biobanked samples from patients recruited to the study will be collected and used for carefully designed biological studies by a consortium of established basic science researchers in the field.

Trial design

This was a UK, multicentre, single-blind, parallel-group RCT.

Patients randomised to the control arm were prescribed IVMP in line with local clinical practice. Recommended dosages were as follows:

• paediatric patients: 30 mg/kg or 500 mg/m2 capped to a maximum dose of 1 g/day for 5 days

• adult patients: 1 g/day for 5 days.

Patients in the intervention arm received the above standard therapy plus additional IVIG:

• adults: 2 g/kg administered in five divided doses

• children > 41.2 kg: 2 g/kg administered as for adults

• children ≤ 41.2 kg: 2 g/kg administered in two divided doses.

Although IVIG dosing does not need to be administered over consecutive days, it must be administered in accordance with the dosing schedule (see Appendix 1).

Patients could be recruited and randomised up to 5 days from the date of first commencing steroid therapy or up to 21 days from the onset of symptoms (if definitely known).

In patients who did not respond to standard IVMP treatment or adjunctive treatment with IVIG, rescue therapy, such as PLEX, would be instituted.

If PLEX was administered it might attenuate the treatment effect of IVIG and, indeed, might have a treatment effect of its own. Guidance parameters were set out to define and standardise the PLEX regime as follows:

• treatment failure would be considered if no improvement was seen or if deterioration occurred after 14 days from presentation or 5 days after completion of either treatment arm

• complete PLEX treatment would consist of at least five cycles, with at least 75% of plasma volume exchanged in each cycle, with a 24- to 48-hour interval between each cycle

• an extra course of IVMP may be given by physicians, often during the lag phase from the decision to proceed with rescue therapy to its initiation (usually 5–7 days).

End-point measures

Tertiary end-point measures

• International SCI-QOL, Pain, Bladder and Bowel Function Basic Data Sets27,28 at 6 and 12 months post randomisation for patients aged ≥ 13 years (at presentation).

• Paediatric Quality of Life Inventory (PedsQL™) Parent Report for Toddlers29 at 6 and 12 months for patients aged 2–4 years (at presentation).

• PedsQL (Parent Report for Young Children30) at 6 and 12 months for patients aged 5–7 years of age (at presentation).

• International SCI-QOL Pain Basic Data Set31 at 6 and 12 months for patients aged ≥ 13 years (at presentation).

The overall flow of study participants from admission through to randomisation and the final visit is summarised in Figure 1.

FIGURE 1.

Participant flow chart. A&E, accident and emergency; CRF, case report form; GP, general practitioner; MRI, magnetic resonance imaging; PIS, patient information sheet.

Study subjects

Participants were individuals who met the eligibility criteria and diagnostic algorithm (see Appendix 2) and who lived in the catchment area of participating tertiary neurology centres (although neurologists could also recruit patients at district general hospitals or from rapid general practitioner referrals). Eligibility for the study was determined using the following criteria.

Study data

The study data were managed as previously described in the published trial protocol. 33

Investigational medicinal product

The investigational medicinal product (IMP) was provided as human normal immunoglobulin (Intratect^®) [100 g/l solution for infusion in a single 5-g (50-ml) or 10-g (100-ml) glass vial] by Biotest AG, Dreieich, Germany, the marketing authorisation holder of the commercially available Intratect.

An Annex 13 clinical trial labelling exemption37 was in place and was approved by the Medicines and Healthcare products Regulatory Agency (MHRA) for this trial. A standard pharmacy dispensing label was applied to the IMP at the point of dispensing by the pharmacy at each investigator site.

The IMP was stored in a secure area with limited access. Site pharmacies were responsible for the safe and appropriate storage of the IMP at their site in accordance with the manufacturer’s instructions:

• do not store above 25°C

• do not freeze

• keep the vial in the outer carton, to protect from light.

Trial sites and study duration

The study was conducted across 15 clinical sites within the UK, with an additional site acting as a participant information centre. The first of the 15 sites was opened for recruitment on 4 March 2015, whereas the final site received its approval to commence recruiting on 8 February 2016. Recruitment was terminated early across all 15 recruiting sites and the single participant identification centre on 11 March 2016.

Trial procedure

Table 1 provides a summary of the trial visit schedule and the associated assessments to be carried out. Some of the questionnaires were specific to particular age groups, with age on presentation used as the criterion.

Analysis

A comprehensive statistical analysis plan was developed and a descriptive analysis (e.g. summary statistics and plots) was to be performed to investigate the distribution of the primary outcome, that is, the ASIA impairment score, across participants.

Ethics and approvals

This study is registered as EudraCT 2014–002335–34, ClinicalTrials.gov NCT02398994 and Current Controlled Trials ISRCTN12127581. Research Ethics Committee approval was obtained (South Central – Berkshire B; reference number 14/SC/1329) alongside MHRA notification.

Baseline and 3-month data summary

Descriptive summarisation of both participants’ data was carried out following a lock of the study data; the trial statistical team remained masked to treatment allocation up to this point. One patient was a white female child aged between 10 and 15 years (student), with an unknown body mass index (BMI). This participant had an ASIA impairment score of C prior to randomisation, meaning that motor function was preserved below the neurological level and more than half of the key muscles below this level had a muscle grade of < 3. At 3 months post randomisation this participant had improved to an ASIA impairment score of D, meaning that at least half of the key muscle functions below the neurological level had a muscle grade of ≥ 3. The other participant was a white adult male aged between 60 and 65 years in full-time employment and with a BMI of 30 kg/m². This participant had an ASIA impairment score of A prior to randomisation, meaning that no motor or sensory function was preserved in sacral segments S4–S5. At 3 months post randomisation, this participant had not improved. Neither participant suffered a relapse during the trial. The adult participant suffered from two AEs during the trial, a chest infection and pressure sores, which were not related to the trial; the paediatric participant suffered no AEs. Because of the very small sample size it is not possible to comment on whether there is any relationship between the trial arm and any change in ASIA impairment scores.

Withdrawal

Both randomised patients remained in the trial until their second follow-up visit (6 months post randomisation), at which point they were withdrawn from the study by the study team.

The rationale behind the termination and subsequent withdrawal was explained to both participants, both of whom had recourse to further information.

Health economics analysis

As the trial did not recruit a sufficient number of participants, it was not possible to conduct the health economic analysis.

Investigator (opinion leader) feedback

Following the termination of the study, the study team developed a survey that sought to both gain feedback on the study and develop a better understanding of the reasons, in the opinion of the investigators, the study did not recruit to target. The survey questions were sent to 13 of the 15 recruiting sites, as the remaining two sites were those of the chief investigator and the co-investigator/project co-ordinator, both of whom were involved in the design of the survey. Of the 13 investigators contacted, 11 responded. A summary of these responses is provided in Table 2.

Overview of the trial evolution

The first trial recruiting site received approval to start recruiting on 4 March 2015, whereas the final site to receive its approval did so on 8 February 2016. At the point at which the study was suspended (11 March 2016), 15 sites were open to recruitment, with two sites each recruiting one subject. The TSC met on 11 March 2106, when strategies to further improve recruitment into the study were extensively discussed. Following this meeting, the TSC and Trial Management Group (TMG) jointly decided that further recruitment into the study should be suspended with immediate effect. The TSC and TMG came to this decision because the study, on its current trajectory, was unlikely to recruit to sufficient numbers to allow scientifically valid deductions to be made. The study co-ordinating team consulted the National Institute for Health Research (NIHR) monitoring team and, following further discussion with the respective programme directors, early termination of the study was agreed.

The need for better understanding of transverse myelitis epidemiology

A lower than expected frequency of patient encounters is believed to be the main factor that led to the low levels of recruitment observed in this study. The challenges associated with recruitment to rare disease trials are not restricted to this study alone. A large-scale comparison of interventional trials (24,088 trials), of which 2759 (11.5%) were classified as rare disease trials and 21,329 (88.5%) as non-rare disease trials, found that rare disease trials are more likely to be terminated early (13.7%) than non-rare disease trials (6.3%). 41 The same study found that, on average, fewer patients were recruited into rare disease trials (70.1%) than into trials of non-rare diseases (81.6%) than initially estimated. It is perhaps intuitively understandable that, because of the uncommon nature of the disorders under study, one of the most frequent problems faced during the conduct of rare disease trials is recruitment of the requisite number of study subjects. 41,42

The trial team were aware of the challenges associated with recruitment into rare disease trials and a prior feasibility assessment suggested a higher frequency of patient encounters than was eventually observed. With the yearly incidence of TM being approximately 350 cases (109 cases proportionately across the 15 open sites, based on a study team projection that 24 sites would capture 50% of cases of TM in the UK), the relatively low rate of patient encounters reported by all sites was unexpected. Furthermore, based on our estimated 35% recruitment rate, approximately 38 subjects could theoretically have been recruited by the open sites per year, rather than the two actually recruited. Interestingly, although numerous evidence (screening logs, informal discussions with recruiting site personnel and the investigator survey responses) pointed towards a low rate of patient encounters, anonymised data provided by the TM patient group, the Transverse Myelitis Society (TMS), showed that 61 registered members of the society reported having been diagnosed with TM in 2015 (of whom 57 had a confirmed TM diagnosis) (Barbara Babcock, TMS chairperson, May 2016, personal communication) (see Appendix 5). These data suggest that the incidence of the disease may not have been overestimated by the study team, especially as the TMS data are likely to be an underestimate of the true incidence of TM (reflecting only the individuals who actually took time to register with the society). However, an analysis of the geographical distribution of these subjects indicated that more than half resided > 25 miles away from a recruiting site (see Appendix 5). This observation points towards a level of geographical disconnect between the potential patients and the trial sites. Additionally, although local TM cases were often referred to, or had their treatment discussed with, the investigators involved in the trial, it is likely that some potentially eligible subjects were seen and treated exclusively by their local physician. The study team did take steps to ensure wide dissemination of the study among the neurology network by inclusion of the study details in the Association of British Neurologists newsletter and conferences, with an e-mail sent to members, presentation of the study at the TMS annual general meeting and conference, distribution of ward leaflets and admission unit posters at participating sites and the provision of study flyers for use in district general hospitals.

The low frequency of patient encounters was further complicated by the fact that a significant proportion of subjects who were considered for the trial were not eligible for inclusion into the study. One reason for this may be that the investigators may not have wholly appreciated the strictness of the eligibility criteria during the feasibility assessment. Indeed, ‘strict inclusion criteria’ was cited by 82% (n = 9/11) of respondents in the post-trial survey as a key factor that they believed had inhibited recruitment into the study. Notably, these inclusion criteria were mandated by the primary end point of the study, with recruited patients needing to have at least an ASIA impairment score of C to observe at least a two-grade improvement in this scale (see Considerations of the design of the study). This highlights the absolute necessity for further studies to establish both the incidence and, equally as important, the severity of TM cases that are observed.

Navigating operational barriers in the set-up and running of a rare disease trial

Although the low frequency of patient encounters primarily led to the study under-recruiting, other barriers played an important role in complicating the study set-up and the running of the study and ultimately the recruitment of subjects into the study. Hurdles in setting up the trial within the NHS sites, including contract and cost negotiations, local governance and a lack of capacity of research staff (research nurses’ and/or investigators’ competing priorities, i.e. clinical and trial commitments) led to the staggered opening of target sites. The overall impact of this was that, at the time of the study closure, only 62.5% (n = 15/24) of target sites were opened and recruiting. As recruitment logs were not available for sites that failed to open (n = 9), we are not able to say what the frequency of patient encounters would have been. The impact of this is best understood in light of the fact that the 24 originally targeted sites were assumed to represent the neurology services that catered for approximately half of the UK population. With only 15 of these target sites open, approximately 31% of new TM/NMO cases were geographically covered.

Lack of research capacity, described as the unavailability of site research staff or co-investigators or competing clinical priorities and commitments to other trials, played a significant role in five of the 24 target sites failing to open. Unsurprisingly, in the survey of investigators who took part in this study, 63.6% (n = 7/11) found it difficult to run all kinds of clinical trials at the present time because of a lack of time; 27.3% (n = 3/11) found it neither difficult nor easy; and one investigator (9.1%) did not currently find it difficult to run clinical trials because of a lack of time (see Appendix 6 for details).

The protracted contract negotiations and cost allocations highlight a wider issue related to differences in health service provision within different regions, in particular the interpretation of standard of care across different (neurology) services. For instance, although the premise of the study (which informed the funding structure) was that most of the study activities were found within standard patient care, some of these activities, such as the ASIA assessment, were evidently not routinely used within some neurological services. These differences impacted not only on the ability of target sites to be opened, but also on how and when sites could be authorised to commence patient recruitment. As site activation was predicated on at least one researcher within a site being ASIA assessment trained, sites that did not routinely use this assessment tool required a member of the research team (preferentially the principal investigator) to undertake the training and gain competence in ASIA assessment. Moreover, competency in ASIA assessment required online training that could take up to 6 hours. At least one site remained unopened at the point of the study closure despite having all other approvals in place and having already had their site initiation visit, as the primary investigator was unable to complete the ASIA assessment training.

It is not immediately apparent how best to overcome the barriers associated with research capacity. Investigators, especially research-active ones, are likely to routinely balance clinical activities with existing and new clinical trials, whereas research and development (R&D) departments balance the low cost-effectiveness associated with initiating rare disease trials with the cost-effectiveness of high throughput or commercial studies. Ultimately, fiscal remunerations for adopting very rare disease trials are likely to be required to facilitate the running of such trials with their associated barriers.

Considerations of the design of the study

Although the paucity of patient encounters and other operational barriers played a significant role in the study under-recruiting, the contributory effect of the study design (in terms of eligibility criteria) to the under-recruitment of the study cannot be overlooked. An analysis of site screening logs and feedback from investigators indicated that the exclusion of ‘less severe’ patients (i.e. patients with an ASIA impairment score of D) had some impact on the recruitment of patients into the study. Approximately 48% (n = 11/23) of screened patients who did not meet the eligibility criteria did not do so as a result of their symptoms being ‘mild’ or being assessed as having an ASIA impairment score of D. The study team recognised this trend while the study was still active and considered amending the protocol to allow the inclusion of patients assessed as having an ASIA impairment score of D or those whose symptoms were considered to be ‘mild’ at the time of presentation. However, the primary end point of the study required patients to have at least an ASIA impairment score of C to enable a two-grade improvement in this scale to be observed.

The short window for recruitment (5 days from the date of first commencement of steroid therapy) was the second factor that contributed to the low recruitment rate. Although it would have been easy to remove this restriction, the early treatment paradigm was key to the study question and so was retained.

Achievements of the trial

The barriers described individually in the previous section and summarised in the timeline in Figure 3 collectively contributed to a formidable barrier that prevented the study from maintaining steady momentum and recruiting to target. Nevertheless, despite the early termination of the study, the efforts involved in trying to run this trial have resulted in two major hidden rewards. First, the study team has established a network of investigators, including both adult and paediatric neurologists, who are ready to collaborate in future studies and clinical trials in TM and other neuroinflammatory conditions. Second, set-up of the study resulted in the provision of important training to local investigators with regard to evaluating patients with TM. As such, we now have 15 paediatric and adult neurology centres that are more equipped to manage adults and children with TM.

FIGURE 3.

A summary of some of the barriers encountered during the conduct of the trial. CRF, case report form.

Conclusion and recommendations

With 91% (n = 10/11) of the trial investigators (opinion leaders) surveyed agreeing that the study question is important and should be a priority and 82% (n = 9/11) noting that the outcome of the study would have changed how they treat TM patients, it is clear that our study question remains pertinent.

However, a future TM intervention trial would benefit from having robust, up-to-date information on the incidence of TM and the distribution of the severity of patients presenting with TM. Multisource incidence studies in both adults and children within the intended intervention study population would negate the need to extrapolate data. Such a trial must also accommodate the evidence accrued on the spectrum of the severity of TM within patients and thus an alternative outcome measure is required beyond the ASIA Impairment Scale. Here, prospective studies recruiting patients to carefully evaluate the utility of other available measures, such as the ASIA motor score, will be required. Importantly, the lessons learned from this study are easily applicable to other very rare disorders.

Patient and public involvement

Patient and public representatives were actively involved in the design of the research, the management of the research and the development of the participant information resources.

The TMS, Multiple Sclerosis Society and Guthy-Jackson Charitable Foundation were involved in the design of the research. Two members from the TMS served on the TSC. Their involvement was crucial both in finalising the research protocol and patient information sheet and in the overall direction that the study took. Additionally, they provided a lay perspective on the probable public perceptions of the study and offered key advice on how to maximise recruitment.

Collaborators

Contributions of authors

Michael Absoud (Consultant and Research Fellow, Paediatric Neurodisability) contributed to the study and to the protocol conceptualisation and design.

Peter Brex (Consultant Neurologist, Adult Clinical Neurology) provided methodological input and clinical expertise.

Olga Ciccarelli (Reader in Neurology, Neuroimaging) provided methodological input and clinical expertise.

Onyinye Diribe (Clinical Trial Manager) contributed to study management and drafting of the final report.

Gavin Giovannoni (Professor of Neurology, Immunobiology) provided methodological input and clinical trial expertise.

Jennifer Hellier (Statistician) was the primary statistician and provided methodological input.

Rosemary Howe (Clinical Trial Manager) contributed to study management and to initial drafts of aspects of the final report.

Rachel Holland (Statistician) was the primary statistician and provided methodological input.

Joanna Kelly (Data Management Strategic Lead, Trial Data Management) contributed to the trial methodology.

Paul McCrone (Professor of Health Economics) provided methodological input.

Caroline Murphy (Operational Director of KCTU) provided methodological input.

Jackie Palace (Consultant Neurologist, Adult Clinical Neurology) provided methodological input, clinical input and clinical trial expertise.

Andrew Pickles (Professor of Biostatistics) contributed to the statistical analysis.

Michael Pike (Consultant Neurologist, Paediatric Neurology) provided methodological input and clinical expertise.

Neil Robertson (Professor of Neurology, Adult Clinical Neurology) provided methodological input and clinical expertise.

Anu Jacob (Consultant Neurologist, Adult Neurology) contributed to the study and to the protocol conceptualisation and design.

Ming Lim (Consultant and Reader in Neurology, Paediatric Neurologist) contributed to the study and protocol conceptualisation and design and to the drafting of the final report.

All authors reviewed and approved the final version of the report.

Publications

Absoud M, Gadian J, Hellier J, Brex PA, Ciccarelli O, Giovannoni G, et al. Protocol for a multicentre randomiSed controlled TRial of IntraVEnous immunoglobulin versus standard therapy for the treatment of transverse myelitis in adults and children (STRIVE). BMJ Open 2015;5:e008312. http://dx.doi.org/10.1136/bmjopen-2015-008312

Data sharing statement

All key data are included in appendices to the report and any further data desired but not included here can be obtained from the corresponding author.

Disclaimers

This report presents independent research funded by the National Institute for Health Research (NIHR). The views and opinions expressed by authors in this publication are those of the authors and do not necessarily reflect those of the NHS, the NIHR, NETSCC, the HTA programme or the Department of Health. If there are verbatim quotations included in this publication the views and opinions expressed by the interviewees are those of the interviewees and do not necessarily reflect those of the authors, those of the NHS, the NIHR, NETSCC, the HTA programme or the Department of Health.

Proposal for an interim futility analysis

Investigator

1. I think addressing the study question is highly important and should be a priority (11 responses) (1 = strongly disagree, 2 = disagree, 3 = neither agree nor disagree, 4 = agree, 5 = strongly agree).

2. The outcome of this study would have changed how I treat my TM patients (11 responses) (1 = strongly disagree, 2 = disagree, 3 = neither agree nor disagree, 4 = agree, 5 = strongly agree).

3. Within my daily workload, I believe I had enough time to allocate to activities that the study required (both clinical and administrative) (11 responses) (1 = strongly disagree, 2 = disagree, 3 = neither agree nor disagree, 4 = agree, 5 = strongly agree).

4. I generally find it difficult to participate in all kinds of clinical trials nowadays because of a lack of time (11 responses) (1 = strongly disagree, 2 = disagree, 3 = neither agree nor disagree, 4 = agree, 5 = strongly agree).

5. A study visit by a participant had (or would have had) an undesirable impact on my clinic (e.g. significantly increasing the clinic time or time for administration of the clinic) (11 responses) (1 = strongly disagree, 2 = disagree, 3 = neither agree nor disagree, 4 = agree, 5 = strongly agree).

6. I generally find it difficult to find time for all study visits nowadays (11 responses) (1 = strongly disagree, 2 = disagree, 3 = neither agree nor disagree, 4 = agree, 5 = strongly agree).

7. What in your opinion was the single most important reason this trial failed (important, please answer) (11 responses)

8. What would you have changed about the study (important, please answer) (11 responses)

Study set-up and recruitment

1. I was actively involved in the site feasibility discussion before the study opened at my site (10 responses) (1 = strongly disagree, 2 = disagree, 3 = neither agree nor disagree, 4 = agree, 5 = strongly agree).

2. Set-up of the study felt complicated at times (10 responses) (1 = strongly disagree, 2 = disagree, 3 = neither agree nor disagree, 4 = agree, 5 = strongly agree).

3. All things considered, I think funding for sites for this study was not really adequate (11 responses) (1 = strongly disagree, 2 = disagree, 3 = neither agree nor disagree, 4 = agree, 5 = strongly agree).

4. I found my interaction with our R&D department useful during the set-up and administration of this study (if applicable) (11 responses) (1 = strongly disagree, 2 = disagree, 3 = neither agree nor disagree, 4 = agree, 5 = strongly agree).

5. I was unable to recruit because (chose as many as applicable) (11 responses)

Other

1. I received as much assistance as needed from the study co-ordinating team (trial manager, sponsors, etc.) (11 responses) (1 = strongly disagree, 2 = disagree, 3 = neither agree nor disagree, 4 = agree, 5 = strongly agree).

2. The study co-ordinating team could have done more to increase awareness of the study among other clinicians or the potential patient population (11 responses) (1 = strongly disagree, 2 = disagree, 3 = neither agree nor disagree, 4 = agree, 5 = strongly agree).

3. In future, I would be interested in collaborating with this team in designing and running another study (11 responses) (1 = strongly disagree, 2 = disagree, 3 = neither agree nor disagree, 4 = agree, 5 = strongly agree).

4. Any other comments? Please feel free to be critical, as we would like to learn from any mistakes made in this study (five responses).