Article Text

Download PDFPDF

Time to next relapse as a primary endpoint in neuromyelitis optica clinical trials
  1. Joanna Kitley1,
  2. M Isabel Leite1,
  3. Liene Elsone2,
  4. Anu Jacob2,
  5. Jackie Palace1
  1. 1Nuffield Department of Clinical Neurosciences, John Radcliffe Hospital, University of Oxford, Oxford, UK
  2. 2The Walton Centre for Neurology and Neurosurgery, Liverpool, UK
  1. Correspondence to Jackie Palace, Nuffield Department of Clinical Neurosciences, John Radcliffe Hospital, University of Oxford, Oxford OX3 9DU, UK; Jacqueline.palace{at}

Statistics from

Neuromyelitis optica (NMO) and its spectrum disorder (NMOSD) is a severe autoimmune inflammatory disorder of the central nervous system, with recurrent attacks mainly focused on the optic nerves and spinal cord. Untreated, it has a high morbidity and mortality. Serum aquaporin-4 (AQP4) antibodies are present in the majority of patients, are recognised as pathogenic and are a predictor of further relapses.1

Most specialists treat AQP4 antibody-positive patients early and aggressively with immunosuppression to prevent further attacks. There have been no randomised trials in NMO, but prednisolone, azathioprine, mycophenolate mofetil, methotrexate and rituximab are among the therapies currently used and appear to have comparable efficacy.2 ,3 Because of the potentially devastating nature of NMO relapses, many clinicians would consider a single relapse on treatment to represent treatment failure and an indication to alter therapy, particularly if the relapse is severe and the therapy has had adequate time to work. This alteration can take the form of an increase in dosage of current treatment or a switch to an alternative agent.

There is an urgent need to identify the most effective treatment regimens, particularly because there are many novel therapeutic candidates being developed, including those which act on the antibody-producing or complement pathways, and an AQP4 monoclonal antibody blocker.4 Current NMO study designs under discussion vary, with some regulatory authorities requesting placebo comparator arms but others accepting standard treatments (azathioprine or alternative plus low dose oral steroids)3 as comparators either as add-on or stand-alone. Time to relapse appears to be an acceptable outcome measure. Whereas placebo studies assessing new treatments will not tell us if the new drug is superior to standard treatment, comparative studies can assess treatments head to head and would need to demonstrate superiority in one or more of efficacy, risk and cost. Add-on studies can only assess the extra efficacy of the new treatment and analyse cost–benefit of combined treatment but the side-effect profile may be worse.

We have previously reported the outcomes over time of AQP4 antibody-positive patients5 and here have re-analysed our updated UK cohort data to assess the time to relapse off and on treatment in an observational retrospective study. These data are informative for planning randomised trials.

Treatment and relapse data were collated from 97 AQP4 antibody-positive NMO/NMOSD patients seen at one of two UK specialist centres for NMO. The demographic data of the cohort were in line with previous publications, with a median onset age of 39.2 years (range 3–75 years) and strong female preponderance (85%). The majority of patients (72%) were Caucasian and the median disease duration was 70.2 months (range 5–417 months). Forty-four patients (45%) had the full NMO phenotype and 53 had limited disease. Most patients (85%) had relapsing disease and those without relapses had a shorter follow-up time (median 18.7 vs 80.0 months) with the majority being treated at onset (80%). The median time to treatment from disease onset was 13 months (range 0–276 months). Twenty-five patients received early treatment (within 3 months of disease onset) and, as expected, the median time to first relapse from disease onset in these patients was significantly longer than in those for whom immunosuppressive treatment was delayed (median 30 vs 12 months; p=0.009).

All patients ultimately received treatment with immunosuppressive agents. The majority (91%) received therapy with azathioprine (n=62), mycophenolate mofetil (n=5), methotrexate (n=9) or rituximab (n=2) with or without steroids or with prednisolone alone (n=10) and only a minority (n=9) received non-conventional treatment, including ciclosporin (n=1), mitoxantrone (n=1) and cyclophosphamide (n=7). Only one patient discontinued maintenance immunosuppressive treatment over the follow-up period.

Survival analysis calculated the median time to relapse from treatment initiation for all patients to be 23 months, or 36 months when excluding patients who relapsed within 3 months of treatment initiation (figure 1), with 25% reaching the endpoint at 7 months and 14 months, respectively. There was no significant difference in time to next relapse from treatment initiation between those treated early (maintenance immunosuppression commenced within 3 months of disease onset) (median 30 months; 25% relapsing by 15 months) and those with delayed treatment (immunosuppression commenced ≥3 months after disease onset) (median 21 months, 25% relapsing by 6 months; p=0.452 log rank), which would suggest that it is reasonable to include in trials all patients starting treatment regardless of disease duration. Additionally, a cross-sectional analysis of time to next relapse of all patients on treatment from 1 March 2009 (selected as a time point giving long-enough follow-up to perform meaningful analysis) revealed a comparable median time to next relapse of 19 months, with 25% relapsing by 7 months.

Figure 1

Kaplan–Meier analysis of time to next relapse in neuromyelitis optica/neuromyelitis optica spectrum disorder patients. The median time to next relapse after initiation of immunosuppressive treatment for all patients was 36 months, with 25% patients relapsing within 14 months (blue line). Patients who relapsed within 3 months of treatment initiation were excluded from the analysis. A cross-sectional analysis of time to next relapse in all patients on treatment on 1 March 2009 revealed a median time to next relapse of 19 months, with 25% of patients relapsing within 7 months (green line).

In contrast to multiple sclerosis trials, where annualised relapse rate (ARR) can be used as a primary endpoint, the substantial relapse-related disability in NMO means that it is not always ethical to continue patients on a stable treatment regime they have already relapsed on. Thus, ARR may not be an acceptable primary endpoint for NMO treatment trials and time to next attack may provide a better primary endpoint. This endpoint would leave clinicians free to alter treatment regimens following treatment failure, without compromising trial data.

We have calculated ‘time to relapse’ on current conventional therapies; these data will be useful for performing power calculations when planning clinical trials in NMO. Our analysis is encouraging because it suggests that studies allowing recruitment of ‘all patients starting treatment’ (not only newly diagnosed patients) and even ‘all patients already on treatment’ (not only those with a recent relapse) are feasible options. Although we previously found5 that patient factors such as age, gender and ethnicity influenced outcome, they had no effect on time to first relapse; however, as the ethnic groups assessed were limited these factors should be adjusted for when planning multinational studies.


View Abstract


  • Contributors Conception and design of study: Drs Palace and Kitley; Data collection: Drs Kitley, Leite and Elsone; Data analysis: Drs Kitley and Palace; Drafting of manuscript: Drs Kitley and Palace; Critical appraisal and revision of manuscript: all authors.

  • Competing interests JK is supported by the NHS National Specialised Commissioning Group for Neuromyelitis Optica and has received travel grants from Biogen Idec, Novartis and Teva and speaker honoraria from Novartis. MIL is involved in AQP4 and MOG antibody testing, is supported by the NHS National Specialised Commissioning Group for Neuromyelitis Optica and by the NIHR Oxford Biomedical Research Centre and has received speaking honoraria from Biogen Idec and travel grants from Novartis. LE is supported by the NHS National Specialised Commissioning Group for Neuromyelitis Optica and has received travel grants from Novartis and Teva. AJ is supported by the NHS National Specialised Commissioning Group for Neuromyelitis Optica. JP has received unrestricted grants and support for scientific meetings and scientific advisory honorariums from Merk Serono, TEVA, Biogen, Bayer Schering and Novartis, has held MS society grants, is a clinical lead for the UK DOH RSS and is supported by the NHS National Specialised Commissioning Group for Neuromyelitis Optica.

  • Ethics approval Oxfordshire REC.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.