Daily low-dose prednisolone to prevent relapse of steroid-sensitive nephrotic syndrome in children with an upper respiratory tract infection: PREDNOS2 RCT

Article history

The research reported in this issue of the journal was funded by the HTA programme as project number 11/129/261. The contractual start date was in October 2012. The draft report began editorial review in January 2020 and was accepted for publication in June 2021. 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.

Copyright statement

Copyright © 2022 Christian et al. This work was produced by Christian et al. under the terms of a commissioning contract issued by the Secretary of State for Health and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, distribution, reproduction and adaption in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.

2022 Christian et al.

Trial rationale/introduction

Minimal change nephrotic syndrome is the most common glomerular disease of childhood. 2 The presenting episode is treated with high-dose oral prednisolone and it is expected that > 90% of children will have a complete response, with responders receiving the diagnostic label of steroid-sensitive nephrotic syndrome (SSNS). 2 The optimum duration of prednisone/prednisolone therapy at presentation was recommended by the International Study of Kidney Disease in Children as 60 mg/m² daily for 4 weeks, followed by 40 mg/m² on alternate days for 4 weeks. 3 Subsequent trials showed benefits with longer durations of corticosteroid;4 however, in the last decade, four well-designed randomised controlled trials have demonstrated no clinical benefit to an extended course of prednisolone beyond the accepted 8- to 12-week course. 5–8 The most recent of these was the PREDNOS (PREDnisolone in NephrOtic Syndrome) trial, which was undertaken by this trial group. Although the PREDNOS trial demonstrated no clinical benefit to an extended corticosteroid course, it did show evidence of cost-effectiveness. 9

Following successful initial treatment, at least 80% of children develop disease relapses necessitating further courses of high-dose prednisolone. The PREDNOS trial found that 80% of children relapse within 12 months of initial presentation8 and around 50% develop frequently relapsing disease. 10 Long-term low-dose maintenance prednisolone therapy is the most commonly prescribed therapy to reduce relapse frequency, although a number of children will require additional immunosuppressive agents, including levamisole, cyclophosphamide, ciclosporin, tacrolimus, mycophenolate mofetil (MMF) and rituximab.

Nephrotic syndrome relapses are associated with a risk of significant complications, including sepsis, thrombosis, dyslipidaemia and malnutrition. 11 The treatment of relapses with high-dose prednisolone is associated with major short-12 and long-term13 adverse effects, including hip avascular necrosis, hypertension, diabetes and behavioural problems. This places financial pressure on the health-care system and leads to reduced quality of life. Furthermore, children are kept off school during relapses, resulting in impaired education performance and parental absence from work.

It is well recognised that the majority of relapses are precipitated by viral upper respiratory tract infection (URTI). Alwadhi et al. 14 followed 68 Indian children with 76 initial presentations or relapses of nephrotic syndrome over 12 months. Of 68 episodes of nephrosis that occurred while not taking corticosteroid, there was an infectious trigger in 57 (84%). The range of infections included URTI (28%), lower respiratory tract infection (a further 19%) and other causes, including urinary tract infection (23%), peritonitis (16%) and diarrhoeal illness (11%). Arun et al. 15 carried out a study designed to evaluate the effectiveness of supplemental zinc in reducing relapse rate in Indian children with nephrotic syndrome. They reported subgroup data for children with frequently relapsing disease. Within that subgroup, 52 out of 86 relapses (60%) were preceded by infections. MacDonald et al. 16 followed 32 Canadian children with nephrotic syndrome over two successive winters. There were 61 URTIs over that period and 41 exacerbations (29 full relapses) of nephrotic syndrome. Seventy-one per cent of exacerbations (69% of full relapses) were associated with URTIs within the prior 10 days.

Furthermore, in children with frequently relapsing steroid-sensitive nephrotic syndrome (FRSSNS), the development of an URTI frequently precipitates a relapse. Moorani et al. 17 documented infections in 62 Pakistani children with nephrotic syndrome over 12 months. A total of 74 infections were observed. Acute respiratory infections were the most common infection (29%). Nephrotic syndrome relapse or initial presentation occurred with 78% of infections or with 80% of acute respiratory infections. In their Canadian cohort, MacDonald et al. 16 demonstrated 47.5% of URTIs associated with disease exacerbation or 32.8% associated with overt relapse.

Given these strong links between viral URTI and relapse, and the morbidity and cost associated with relapse and its treatment, it is logical that attempts are made to ameliorate the URTI-driven disease modification.

Summary of previous studies investigating the use of daily prednisolone therapy at the time of upper respiratory tract infection

Current practice in the majority of UK centres has been for no change to be made to immunosuppressive therapy at the time of development of an URTI. Between 2000 and 2011, three studies18–20 assessed whether or not the use of daily prednisolone at the time of URTI reduced the subsequent risk of relapse of nephrotic syndrome.

Mattoo and Mahmoud18 published the first of these trials in 2000. Mattoo and Mahmoud18 studied 36 Saudi children with relapsing SSNS who were receiving a long-term maintenance dose of alternate-day prednisone of approximately 0.5 mg/kg per day. Starting on the day of onset of URTI (defined by onset of cough and/or cold with or without fever), children were alternately assigned in an unblinded manner to receive either 5 days of daily prednisone at the same dose or to remain on alternate-day prednisone. Patients who did not relapse and patients who did not achieve remission with cyclophosphamide were excluded. The number of disease relapses [defined as Albustix (Siemens Healthcare Diagnostics Ltd, Frimley, UK) 3+ positivity on morning urinalysis for 3 days] was documented in each group. Patients were followed for 2 years and the main reported outcome was the mean number of relapses per patient over that period for each arm.

The arms were well matched for age, sex, use of prior cyclophosphamide and histology (where performed). In the 18 children assigned to daily prednisone at the time of URTI, the rate of relapse was lower than in the 18 children who continued on alternate-day prednisone [mean ± standard deviation (SD) relapse rate 2.2 ± 0.87 vs. 5.5 ± 1.33; p = 0.04]. No data on the individual risk of relapse following URTI were reported. The intervention arm received a median of seven courses of daily prednisone over the 2-year follow-up period. No difference was noted in the frequency of hospitalisations or the length of treatment for relapses between the two treatment arms.

Abeyagunawardena and Trompeter19 recruited 48 Sri Lankan children to a randomised double-blind crossover study. All were receiving long-term low-dose alternate-day prednisolone (mean 0.36 mg/kg, range 0.1–0.6 mg/kg). Children were studied over two consecutive URTIs (defined as cough, runny nose, sore throat, lethargy, body aches and fever). Throat swabs were taken and children with bacterial infection were excluded. Children were randomised, using sealed envelopes, either to receive prednisolone at their usual maintenance dose for 7 days given daily instead of on alternate days or to continue on alternate-day prednisolone. This was achieved by dispensing investigational medicinal product (IMP) in one of two containers, one of which contained prednisolone and the other placebo. Parents were asked to administer the study drug on the child’s usual non-treatment day and to continue so that a total of 7 days of daily treatment were given. In this way, children received three or four doses of study drug.

In the crossover design, those who received daily prednisolone for the first URTI received alternate-day prednisolone plus placebo for the second URTI, and vice versa. Recruitment continued until 40 children had experienced two URTIs. Children were reviewed on days 3 and 7 to assess for evidence of disease relapse (defined as Albustix 3+ proteinuria for 3 consecutive days). Those who developed prednisolone-related adverse events (AEs), those who required steroid-sparing therapy for frequent relapses, those in whom prednisolone was discontinued because of sustained remission and those who did not have two viral infections were all excluded from the study (8/48 of the recruited number). The main reported outcome was the difference in infection-associated relapse (IAR).

Overall, there were seven (18%) IARs following 40 URTIs treated with daily prednisolone and 19 (48%) IARs following 40 URTIs where alternate-day prednisolone and placebo were administered (p = 0.014; two-sided Fisher’s exact test). Response to the initial URTI was a relapse in 4 of 18 (22%) children in the intervention arm and 10 of 22 (45%) children in the placebo arm (no significance reported). No significant adverse effects were encountered.

The third and largest of these three initial studies was performed in 100 Indian children recently diagnosed with FRSSNS who were on long-term alternate-day prednisolone with or without levamisole. 20 Children were recruited in stable remission, having received alternate-day 1.5 mg/kg of prednisolone for 4 weeks and then tapered by 0.25 mg/kg every 2 weeks until a dose of 0.5–0.75 mg/kg on alternate days was reached. If a dose of prednisolone > 1 mg/kg on alternate days was required, then levamisole was added. Children were randomised, stratified according to whether they received levamisole (n = 32) or not (n = 68), to either receive daily prednisolone for 7 days or remain on alternate-day prednisolone. Prednisolone was given at the same dose for non-treatment-days in an unblinded manner at the time of development of intercurrent infection [defined as fever (i.e. axillary-measured temperature of > 38 °C on two occasions and more than 1 hour apart), rhinorrhoea or cough for more than 1 day or diarrhoea (i.e. three or more semiformed stools per day for more than 2 days)]. Children were reviewed every 2 months for a total of 12 months.

The primary end point was the incidence of IAR, with secondary end points of overall relapse rate, infection frequency and type and cumulative dose of prednisolone. Patients exited the study if there were two or more relapses in any 6-month period. Daily prednisolone therapy at the time of an URTI resulted in a reduction in the incidence of IAR [0.7 ± 0.3 vs. 1.4 ± 0.5; rate difference 0.7, 95% confidence interval (CI) 0.3 to 1.1; p < 0.01] and the overall relapse rate (0.9 ± 0.4 vs. 1.8 ± 0.5; rate difference 0.9, 95% CI 0.4 to 1.4; p < 0.0001). Although not powered to do so, a subanalysis showed that this difference was lost in those receiving levamisole. Nineteen children in the daily prednisolone group, compared with seven in the alternate-day group, remained relapse free over the entire 12-month study period (p = 0.03). There was no difference in cumulative prednisolone dose between the two treatment arms. More infections occurred in the daily prednisolone group (226 vs. 161; p = 0.04), although no difference was detected in height SD score, cushingoid features, cataract or serious infection. Six children (two in the daily steroid group) exited the study because of treatment failure, necessitating treatment with cyclophosphamide or calcineurin inhibitors.

Critique of previous studies investigating the use of daily prednisolone therapy at the time of upper respiratory tract infection and summary of findings

Subsequent research

Since commencement of the PREDNOS2 trial, a second trial from Abeyagunawardena et al. 22 has been published. In contrast to the group’s previous study19 and the other two studies18,20 prior to PREDNOS2, the investigators studied children with SSNS on no maintenance therapy. In a double-blind placebo-controlled crossover study, 48 children were randomised to receive 5 days of daily prednisolone (0.5 mg/kg) or placebo at the start of an URTI. Children on second-line treatments were excluded and children also had to have discontinued corticosteroid treatment at least 3 months prior to recruitment. A sealed envelope method was used to randomise patients in the same way as the group’s previous study. 19 Both investigators and patients/parents were blinded to the contents of the pot until the end of the study. Viral infections were defined more loosely than for the previous study, with the presence of two or more of the following criteria: fever > 38 °C; runny nose; cough; body aches, lethargy or loss of appetite; or sore throat. Study treatment was not given or was subsequently stopped if microbiological evidence of a bacterial infection was found. Patients were followed for 12 months, with any subsequent URTIs treated with the same study drug. Thereafter, a crossover took place, with group 1 taking placebo and group 2 taking prednisolone at the time of an URTI for the next 12 months.

A sample size matching their previous study appears to have been chosen, but no power calculation was provided. Investigators recruited 27 children who were randomised to group 1 and 21 children randomised to group 2. Of the 48 children recruited, only 33 (69%) completed the study (19 children in group 1 and 14 children in group 2). Of those exclusions, 12 children were because of non-compliance and three children needed treatment escalation in the form of maintenance immunosuppressive therapy.

There were 11 relapses following 115 episodes of URTI (9.5%) in the treatment group and 25 relapses following 101 episodes of URTI (24.8%) in the control group. Despite no significant differences in the numbers of URTIs between groups, children in the treatment group had significantly fewer relapses than children in the control group (p = 0.014). Within the treatment group, 22 of 33 children did not relapse compared with 14 of 33 children within the placebo group (p = 0.049).

The IAR in this study, at 25%, was half of that reported in earlier studies. 19,20 The authors explain that this is likely because the study was conducted in children with more stable disease, who do not require long-term prophylactic treatments. Interestingly, the only relapses reported within the study period were those following URTIs.

In common with their previous study,19 the sample size is small and the study was vulnerable to inadvertent unblinding. The high dropout rate weakens the findings and the decision to exclude children who commenced maintenance treatment introduces a potential bias. In addition, along with the studies prior to PREDNOS2, generalisability of the findings to a different georacial group is problematic.

Summary

In the 2015 update of the Cochrane review of corticosteroid treatment for nephrotic syndrome in children,4 the authors noted that the combined weight of the three studies described above18–20 increased the power of the analysis so that low-dose daily prednisolone may be considered at the time of viral infections in children on maintenance alternate-day prednisolone. Abeyagunawardena et al. ’s22 subsequent study, which was available in abstract form only at the time of the Cochrane review, has shown that the same strategy may benefit children on no maintenance treatment. In the 2020 Cochrane update,4 the authors concluded that the limited data provided by a single study of 48 children meant that it remains unclear whether or not children not already on alternate day corticosteroid should restart daily corticosteroid for around a week at the onset of viral infections.

The review authors also stated that there were some concerns about the possibility of bias4 (Table 2). Mattoo and Mahmoud’s study18 was rated as having a high risk of selection bias through inadequate random sequence generation and allocation concealment. Only the two studies by Abeyagunawardena et al. 19,22 were blinded studies. The study by Mattoo and Mahmoud18 reported that fewer than 10% of participants were lost to follow-up or excluded from the analysis. Three of the studies18,19,22 were rated as having a high risk of reporting bias due to a combination of outcome data not including one or more outcomes of FRSSNS, relapse rate and AEs, providing data in a format that could not be entered into the meta-analyses or from inability to separate first and second parts in crossover studies.

Therefore, although providing proof of concept that increased corticosteroid dosing may reduce the risk of relapses in children with SSNS, the methodological issues and other limitations of all these studies have not been able to address the following questions:

• Do children from developed temperate countries where the pattern of childhood URTI is significantly different benefit from the same intervention?

• Does this effect occur in children receiving long-term maintenance therapy with other immunosuppressive therapies (e.g. levamisole, ciclosporin, tacrolimus and MMF) in conjunction with prednisolone or without prednisolone or, indeed, does this effect occur in children on no long-term immunosuppressive therapy?

• Is there an effect on the cumulative dose of prednisolone or in the incidence of corticosteroid-related adverse effects, including behavioural issues associated with the use of this intervention?

• What are the quality-of-life implications of this strategy?

• What would be the cost-effectiveness of such an approach?

Research question

A meeting was convened in January 2012 to discuss this trial proposal, to which the members of the National Institute for Health Research (NIHR) Medicines for Children Research Network Nephrology Clinical Studies Group, which represents each of the 13 tertiary paediatric nephrology centres in the UK, were invited to attend. A number of consumer representatives, including the chairperson of the UK Nephrotic Syndrome Trust (NeST) (Somerset, UK), and other parents of children with nephrotic syndrome also attended.

It was clear from this meeting that interest lay not only in further generalisability of the use of daily low-dose prednisolone at the time of an URTI in children on alternate-day prednisolone and children on no maintenance treatment, but also in those receiving other immunosuppressant therapies (with or without prednisolone) for their nephrotic syndrome. Therefore, a single pragmatic trial was proposed, which would include all patients with relapsing SSNS, regardless of background therapy, to answer the overarching question of whether or not treatment with a short course of daily prednisolone at the time of URTI in children with relapsing SSNS reduced the subsequent development of nephrotic syndrome relapse.

A reduction in the nephrotic syndrome relapse rate would reduce relapse and treatment-associated morbidity, hospitalisation rates and parental time absent from work.

The research question agreed was ‘Does a 6-day course of daily prednisolone given early in the course of URTI in children with relapsing SSNS effectively and safely reduce the incidence of subsequent URTI-related relapse?’.

A 6-day course was chosen as an even number of days of administration of trial drug would be required to avoid the bias from children already taking alternate-day prednisolone who might receive different dosing of trial drug depending on whether or not it was commenced on a day on which they usually took prednisolone.

Aim

The aim was to evaluate the effectiveness of a 6-day course of daily prednisolone therapy at the time of URTI in reducing the development of subsequent nephrotic syndrome relapse in children with relapsing SSNS.

Objectives

The specific trial objectives were to determine whether or not a 6-day course of oral prednisolone given at the time of an URTI:

• reduces the incidence of first upper respiratory tract infection-related relapse (URR) in children with relapsing SSNS

• reduces the overall rate of URR in children with relapsing SSNS

• reduces the overall rate of relapse in children with relapsing SSNS

• reduces the cumulative dose of prednisolone over the 12-month study period

• reduces the incidence and prevalence of adverse effects of prednisolone, including behavioural abnormalities

• reduces the number of children undergoing escalation of background immunosuppressive therapy

• increases the number of children undergoing reduction of background immunosuppressive therapy

• is more cost-effective than standard therapy

• improves quality of life as measured using the Child Health Utility 9D (CHU-9D), EuroQol-5 Dimensions (EQ-5D) and the Pediatric Quality of Life Inventory (PedsQL).

Trial design

The PREDNOS2 trial was a Phase III randomised parallel-arm placebo-controlled double-blind trial that compared a 6-day course of daily prednisolone with no change in therapy using a matching placebo. Children with relapsing nephrotic syndrome who met eligibility criteria were randomised in a 1 : 1 ratio (with minimisation according to background treatment for nephrotic syndrome) to receive either 6 days of prednisolone or 6 days of placebo at the time of an URTI. The participant, clinician and trial teams were masked to treatment allocation.

The trial protocol for PREDNOS2 was published in open access format in 2014. 1

The trial schema is shown in Figure 1.

FIGURE 1.

Trial schema.

Participants

Outcome measures

Sample size

In children with FRSSNS, the development of an URTI results in relapse in around 50% of instances. 19 In the Abeyagunawardena and Trompeter19 study, 40 URTIs treated with placebo were followed by 19 relapses (48%), compared with seven relapses (18%) in the prednisolone-treated group. This corresponds to an absolute difference of 30% (a 62.5% proportional reduction). In the first treatment period, there were 10 relapses (45%) in 22 placebo-treated children, compared with four relapses (22%) in 18 prednisolone-treated children (i.e. a 23% absolute difference and 51% proportional reduction). 19 This was a large treatment effect based on a small study of children on long-term maintenance alternate-day prednisolone therapy in a developing country. Therefore, to detect a more conservative difference of 17.5% (i.e. 35% proportional reduction) in URR rate (i.e. from 50% to 32.5%), with 80% power, a two-sided test and an alpha of 0.05, required 250 children in total (comparison of two proportions23). An allowance was made for between 10% and 20% dropout (e.g. subject withdrawal, lost to follow-up or subject not having an URTI during the 12-month follow-up period), which required recruitment of between 280 and 320 children. Therefore, it was proposed to recruit 300 children, 150 to each arm. However, with a treatment effect more in line with the 50% reduction observed in the first treatment period of the Abeyagunawardena and Trompeter19 study, the PREDNOS2 trial would have sufficient power (> 95%) to detect this difference [i.e. to detect a 50% proportional reduction (i.e. from 50% to 25%) with 90% power and an alpha of 0.05, required 160 children, increasing to 200 with allowance for 20% dropout].

During the trial, it became apparent that a larger number of children than expected were reaching the end of the 12-month follow-up period without experiencing an URTI and, therefore, were unable to contribute to the analysis. The original sample size was increased by between 10% and 20% to account for this, but, in late 2015, the actual number of children who had completed 12 months’ follow-up without an URTI was 28%. Therefore, the TSC recommended revising the sample size using a 30% attrition rate, which increased the sample size to 360 patients.

Recruitment

Children with relapsing SSNS under the care of a paediatric nephrologist and/or a general paediatrician were recruited from throughout the UK. Potentially eligible children were identified from clinic lists and departmental databases. Information sheets outlining the trial were mailed to the parents or guardians of potentially eligible children (and the child, where age appropriate) 1–2 weeks prior to their next clinical appointment. Following confirmation of eligibility with regard to the inclusion and exclusion criteria, and a further full discussion of the trial, informed consent was sought from the parents (or guardians) and children (informed consent or assent, according to age) at the time of this appointment.

Randomisation

Following the informed consent process and completion of the baseline assessments, eligible children were randomised at the level of the individual in a 1 : 1 ratio to either the intervention arm or the control arm. To ensure concealment of treatment allocation, randomisation was provided using a central secure 24-hour internet-based randomisation service, or by a telephone call to the University of Birmingham Clinical Trials Unit (BCTU) (Birmingham, UK). The randomisation process used minimisation by the background treatment regimen that the child was receiving at randomisation (i.e. no treatment, long-term maintenance with prednisolone therapy only, long-term maintenance prednisolone therapy with other immunosuppressant therapy, and other immunosuppressant therapy only).

Randomisation took place at the time of the initial trial visit, prior to the development of the first URTI, to ensure that the trial drug reached the parents/child by the time the first URTI developed. The investigator completed the randomisation notepad to prepare for randomisation. The randomisation notepad was signed by the investigator to indicate that all of the eligibility criteria had been checked. It was also noted in the medical records that the investigator had checked all of the eligibility criteria and that the child met all of the inclusion criteria and none of the exclusion criteria. The signed randomisation notepad was kept in the ISF and a copy sent to the trials office.

After randomisation, each child was assigned a unique trial identification number to be used on all trial-related material for the child. Confirmation of the randomisation and trial identification number was forwarded to the trial manager, the local investigator and the Birmingham Children’s Hospital Pharmacy Department by e-mail immediately on randomisation. Once a PREDNOS2 trial number had been obtained, the investigator faxed or e-mailed a signed copy of the clinical trial prescription form and the consent/assent form(s) to the pharmacy at the Birmingham Children’s Hospital to order the PREDNOS2 trial drug, which was then sent directly to the parents/child (see Investigational medicinal product). Only delegated staff at the Birmingham Children’s Hospital NHS Trust Pharmacy were able to view the treatment allocation to dispatch the pots of trial drug. This was carried out using a secure login link to the randomisation program once the child had been randomised. This ensured that investigators and the trial office remained blind to the treatment allocation.

Blinding

The PREDNOS2 trial was a double-blind placebo-controlled trial. The secure internet-based central randomisation service provided by BCTU ensured concealment of treatment allocation. Essential Nutrition Ltd (Brough, UK), a Medicines and Healthcare products Regulatory Agency (MHRA)-licensed good manufacturing practice manufacturer of solid oral dosage form products, manufactured both the active and placebo tablets for the PREDNOS2 trial. To ensure that the tablets were identical in size, shape, taste and smell, Essential Nutrition Ltd conducted batch testing throughout the trial. To maintain blinding, children randomised to the control arm received the same number of study tablets as they would have received in the intervention arm through supplemental placebo tablets. Therefore, outcome assessments by families and local investigators were carried out masked to treatment allocation. Only staff at the Birmingham Children’s Hospital Pharmacy Department were aware of the individual allocations of children. The trial statistician was unblinded to the intervention code for all of the interim analyses and presented unblinded analyses to the DMC.

Planned interventions

Current practice in the UK has been for no change to be made to immunosuppressive therapy at the time of development of an URTI. The intervention being assessed within the PREDNOS2 trial was a 6-day course of daily prednisolone therapy at the time of onset of URTI.

Those randomised to the intervention arm commenced a 6-day course of daily prednisolone each time they developed an URTI during the 12-month follow-up period (see Definition of upper respiratory tract infection for strict definition). Those randomised to the control arm received an identical number of placebo tablets, therefore allowing double blinding of the trial. If the child was receiving background long-term immunosuppressive therapy (e.g. levamisole or ciclosporin), then this continued unchanged.

Trial procedures

Statistical methods

The primary comparison trial arms were those who were randomised to 6 days of daily prednisolone and those randomised to placebo. Analyses were based on the intention-to-treat (ITT) principle. The analysis population did not include all of the randomised population, as analyses were based on a modified ITT population, which included only those participants who had an URTI over the 12-month follow up period. Excluding those participants who did not have an URTI resulted in no bias, as taking the study medication was conditional on the child experiencing an URTI and this should, by chance, have been balanced between the two arms. Participants were analysed in the intervention arm to which they were randomised and participants were included whether or not they received the allocated intervention following an URTI.

Estimates of treatment effects are presented with two-sided 95% CIs. p-values are reported from two-sided tests at the 5% significance level. Analyses were adjusted for the minimisation variable (i.e. background therapy at randomisation) and baseline scores (where appropriate). The placebo arm is the reference group. No corrections for multiple tests were made. All analyses were carried out using SAS^®, version 9.4 (SAS Institute Inc., Cary, NC, USA), or Stata^®, version 16 (StataCorp LP, College Station, TX, USA).

The primary outcome was the incidence of first URR of nephrotic syndrome following any URTI during the 12-month follow-up period. This outcome is a binary outcome (i.e. yes/no). The number and percentage of children reporting an URR was reported. An adjusted risk difference and relative risk along with the respective 95% CI were estimated using binomial models with the identity and log-links, respectively. The statistical significance of the treatment arm parameter was determined from the p-value generated by the model. The number needed to treat (NNT) was also calculated. Three sensitivity analyses to assess the impact of missing data were undertaken for the primary outcome: (1) no URR (best)/URR (worse), (2) URR (worse)/no URR (best) for the prednisolone and placebo arms, respectively, and (3) an analysis that included the eight participants who had URTIs without relapse, but who did not complete 12 months’ follow-up and so it was not possible to assume that they did not have an URR during the 12-month trial follow-up period. In this third analysis, the data available were used and so they were included as ‘no’ for the primary outcome.

The number of children experiencing any relapse (i.e. both URRs and non-URRs), the number of children having an escalation of background immunosuppressant therapy and the number of children having a reduction in background immunosuppressant therapy were analysed as per the primary outcome measure. The rate of URRs and rate of any relapse were determined in the first instance as numbers of children who had a relapse with the respective percentage and in the second instance as a median and interquartile range (IQR). An adjusted incidence rate ratio with 95% CI was estimated using a Poisson regression model. An offset for the length of time the child was in the trial was included in the model.

The cumulative dose of prednisolone received over the 12-month period was initially planned to be reported as means with SDs, with an adjusted mean difference and 95% CI estimated from a linear regression model. These data were, however, skewed and so the median and IQR were also reported, along with the unadjusted median difference between arms and 95% CI estimated using bootstrapping methods, and the arms were instead compared using a Wilcoxon rank-sum test.

The PedsQL quality-of-life measure and ACBC measures were collected at multiple time points. The scores at each time point were summarised using means and SDs. Data were analysed using mixed-effect repeated measures models, with the minimisation variable and the baseline score included in the model as covariates. Time was included as a continuous variable in the model. In the initial model, a treatment-by-time cross-term was included in the model. If this was not significant, it was assumed that the treatment effect was constant over time and models without the treatment-by-time cross-term were fitted. Adjusted mean differences between arms are presented, alongside the respective 95% CIs.

The prednisolone AE data were collected at each time point (i.e. 3, 6, 9 and 12 months). To assess whether or not each AE occurred over the whole trial period, the 3-, 6-, 9- and 12-month data were also assessed collectively, with statistical significance determined by a chi-squared or Fisher’s exact test (as appropriate). The SAE data were summarised descriptively. The number of children experiencing any SAEs and suspected unexpected serious adverse reactions (SUSARs) was presented by arm. Statistical significance was determined by a chi-squared or Fisher’s exact test (as appropriate).

Subgroup analyses were limited to the primary outcome and the following defined categories based on the child’s background treatment regimen at randomisation:

• no background immunosuppressive therapy compared with long-term maintenance prednisolone therapy, long-term maintenance prednisolone therapy plus other immunosuppressive therapies and other immunosuppressive therapies alone

• on prednisolone compared with not on prednisolone

• on other immunosuppressant compared with not on other immunosuppressant.

The effects of these subgroups were examined by including a treatment arm by subgroup interaction parameter in the final binomial model.

Ethics approval, regulations and trial registration

Patient, parent and public involvement

The PREDNOS2 trial design was discussed extensively with a number of patient and public involvement representatives, including representatives from NeST. This provided valuable input regarding the development of essential and patient-facing documentation for the trial and the acceptability of study visit frequency, as well as insight into the issues surrounding self-medication and addressing the complexity of calculating study drug dosage at home.

Patient and public involvement representatives worked with members of the trial team to develop advertising materials for social media to encourage patient recruitment through avenues such as the NeST Facebook page (URL: www.facebook.com, Facebook, Inc., Menlo Park, CA, USA). Representatives also disseminated information about the trial among their nephrotic syndrome contacts, encouraging those families/patients who may have benefited from participation in the PREDNOS2 trial to seek out more information and ask their local clinician about participation. These actions undoubtedly aided the trial’s recruitment.

Patient and public involvement representatives reviewed any changes to essential or patient-facing documentation throughout the trial as amendments were made. Representatives also had input throughout the trial through their membership of the TMG and TSC, and regularly attended meetings for both oversight groups.

Recruitment

The PREDNOS2 trial opened to recruitment in February 2013 and the first patient was randomised on 19 March 2013. The trial closed to recruitment, after 6 years, on 31 January 2019, with a total of 365 patients randomised from 91 of the 122 approved centres. Five patients more than the 360 patient recruitment target were included, as these patients had all received written information on the trial and verbally indicated their willingness to participate in the trial. The Trial Ethics Committee agreed that these patients should be offered randomisation to the trial.

The average rate of recruitment was 5.1 patients per month. After 2 years where the average recruitment rate had been 6.4 patients per month, the rate slowed to 4.6 patients per month for the remaining 47 months. Quarterly recruitment by randomised treatment is shown in Figure 3. The first period is 4 months (March–June 2013) and the final period is a single month (January 2019).

FIGURE 3.

Quarterly recruitment by randomised treatment.

Patients were recruited from 91 paediatric units of 122 approved units (74.6%) across the whole of the UK. The number of patients recruited per centre varied from 1 to 35. Individual site recruitment data are shown in Appendix 1, Table 31. The 13 UK tertiary paediatric renal units recruited 133 (36%, range 3–35) patients of the total trial population. As was seen with the PREDNOS trial,9 this highlights the important contribution of district general paediatric units, in which individual recruitment ranged from 1 to 17 participants.

Participants had completed 12 months’ follow-up by 31 January 2020.

Participant flow

Of the 365 children randomised into the PREDNOS2 trial, 182 were randomised to the intervention arm and 183 to the placebo arm. Thirty-two children (8.8%) did not complete the 12 months’ follow-up (13 children from the intervention arm and 19 children from the placebo arm). Of these withdrawals, 14 patients were withdrawn before having an URTI (six patients from the intervention arm and eight patients from the placebo arm). Eighteen patients were withdrawn after having an URTI (seven patients from the intervention arm and 11 patients from the placebo arm). Eighty children completed the 12 months’ follow-up without experiencing an URTI (42 children in the prednisolone arm and 38 children in the placebo arm). Therefore, in total, 48 children in the prednisolone arm and 46 children in the placebo arm were excluded from the trial analyses, as they did not have an URTI during the trial. This gave a modified ITT analysis population of 271 patients (i.e. 365 randomised patients – 94 patients with no URTI).

The CONSORT (Consolidated Standards of Reporting Trials) flow diagram36 for all randomised patients is shown in Figure 4. Patient flow for the modified ITT population is shown in Figure 5.

FIGURE 4.

A CONSORT flow diagram for all randomised patients.

FIGURE 5.

A CONSORT flow diagram for the modified ITT population. Reproduced with permission from Christian et al. 37 This is an Open Access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. The figure includes minor additions and formatting changes to the original figure.

Reproduced with permission from Christian et al. 37 This is an Open Access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. The figure includes minor additions and formatting changes to the original figure.

Overall reasons for withdrawal are detailed in Table 5.

Withdrawals and analysis population

Of the 365 children randomised:

• 32 children were withdrawn or lost to follow-up from the trial prior to completing the 12 months assessment –

  • 18 of these children had an URTI prior to withdrawal or being lost to follow-up

  • 14 of these children had no URTI prior to withdrawal or being lost to follow-up

• 333 children completed the 12-month follow-up assessment –

  • 253 of these children had an URTI

  • 80 of these children had no URTI at any point in the trial.

This gives a modified ITT analysis population of 271 patients (i.e. 253 + 18).

Details of the children who were excluded are shown in Table 5 for the whole randomised population and in Table 6 for the modified ITT population.

Completeness of data

Data completion was very high. For the modified ITT population at all four data collection time points, a total of 1040 forms out of a possible 1055 were received (98.6%; 99.0% from the intervention arm and 98.1% from the placebo arm).

Baseline data

Randomisation was minimised according to baseline treatment. Details of baseline demographics for the whole randomised population and for the modified ITT population are shown in Tables 7 and 8, respectively. For the modified ITT analysis population, the mean age at recruitment was 7.6 (SD 3.6) years, 64% of participants were male, 64% of participants were white and the duration of SSNS at trial entry was 3.2 years.

Adherence

Adherence to the trial medication was summarised at each follow-up visit according to the following:

• Did the parent commence their child on trial medication at the time of the URTI?

  • If yes, what was the number of days to starting medication from the start of the URTI?

• Was the whole 6-day course of trial mediation taken?

  • If no, how many days were missed?

• Was the 6-day course prematurely discontinued?

Overall adherence for the 12-month trial period is shown in Table 9. In total, trial medication was commenced at the time of an URTI for 691 of 791 URTIs (87.3%; 85.4% of URTIs in the intervention arm and 89.2% of URTIs in the placebo arm). There were no differences in rates of adherence for each 3-month period throughout the trial. Reasons for not starting the trial medication at the time of URTI included the parent/guardian forgetting, medication had not arrived in time and child already relapsed and/or on treatment for relapse.

Primary outcome

All outcome analyses were carried out on the modified ITT population of 271 children and excluded the 80 children who completed 12 months of follow-up without having had an URTI and the 14 children who were withdrawn without having had an URTI.

The median time from randomisation to first URTI was 61 (IQR 21–126) days in the intervention arm and 54 (IQR 23–98) days in the placebo arm. There were 384 URTIs and 82 URRs in the intervention arm and 407 URTIs and 82 URRs in the placebo arm. One patient in the placebo arm had an URTI, but no information was provided on whether or not they had a URR and so their data were classed as missing. Three patients in the intervention arm and five patients in the placebo arm had URTIs but withdrew before the 12-month follow-up and did not report any URR for any time points for which they provided data. These patients were excluded from the primary analysis, but were included in a sensitivity analysis (Table 10).

There was no difference in the proportion of children experiencing an URR between the two arms of the trial. In the intervention arm, 56 children (42.7%) experienced an URR, compared with 58 children (44.3%) in the placebo arm. The adjusted risk difference was –0.02 (95% CI –0.14 to 0.09; p = 0.70) and the adjusted risk ratio was 0.96 (95% CI 0.74 to 1.26). The NNT with prednisolone to avoid one URR was 42.

There were three predefined subgroup analyses according to background treatment. There was no evidence that the treatment effect differed when the data were analysed by these subgroups (Table 11).

Secondary outcomes

Introduction

The economic evaluation was undertaken alongside the PREDNOS2 trial to estimate the cost-effectiveness of a short course of daily prednisolone therapy at the time of URTI compared with usual care (use of placebo) for treating children with SSNS. The primary evaluation was a model-based cost–utility analysis, with the outcome measured in terms of QALYs.

Aim

The aim was to estimate the cost-effectiveness of a 6-day course of daily prednisolone given early in the course of an URTI compared with standard care in treating children with relapsing SSNS.

Methods

A model-based economic evaluation in the form of a cost–utility analysis based on the outcome of cost per QALY was performed alongside the PREDNOS2 trial. The analysis assessed the difference in costs and difference in QALYs from a NHS perspective. An ITT analysis was adopted, as the economic evaluation was conducted alongside the clinical trial. Two different time horizons were considered: (1) the primary analysis assessed the cost-effectiveness at 1 year and (2) the secondary analysis extrapolated the results over 16 years (when the model cohort reached 18 years). Although a societal perspective was considered, this was not feasible within the resource constraints of the trial.

Results

The results for the base-case and sensitivity analyses are presented separately for the primary and secondary analyses.

Summary of findings

In, to the best of our knowledge, the largest ever clinical trial of an IMP in children with nephrotic syndrome, the PREDNOS2 trial has shown that giving 6 days of low-dose prednisolone to children with relapsing SSNS at the time of an URTI does not reduce the risk of a subsequent URR. The findings apply across the four subgroups: (1) children on no background treatment, (2) children on long-term maintenance prednisolone only, (3) children on non-corticosteroid maintenance immunosuppression only and (4) children on both long-term maintenance prednisolone and non-corticosteroid maintenance immunosuppression.

There were also no differences in secondary outcomes between the two treatment arms, including the overall incidence of relapses (whether URTI related or not), escalations and reductions in background therapy, cumulative dose of prednisolone, adverse effects of prednisolone, ACBC scores or quality of life (measured using the PedsQL).

However, the economic analysis showed that giving daily prednisolone at the time of an URTI, compared with standard care, led to reduced overall health-care and background therapy costs, increased QALYs and was, therefore, the ‘dominant’ treatment option. This result was maintained over a 16-year time horizon and was robust to all sensitivity analyses.

Comparison of demographics with other studies

The findings do not support the four previously published trials,18–20,22 which all showed that low-dose daily prednisolone for 5–7 days given at the time of an intercurrent infection was effective in preventing a relapse, although, with p-values of 0.04, 0.01, 0.01 and 0.049, the results were not overwhelmingly in favour of daily prednisolone. Rather, the findings suggested that it may work. Furthermore, the small populations and the methodological issues with these studies limited their impact, leading the latest Cochrane review to conclude that ‘clinicians are unlikely to use this regimen without additional data to confirm its efficacy and safety’. 4

Table 27 shows how the PREDNOS2 trial compares with the four previous trials of daily corticosteroid at the time of an intercurrent illness. The PREDNOS2 trial sample size was larger than those of all four previous studies combined. In terms of demographics, the average age at recruitment for the PREDNOS2 trial population (7.6 years) was in the middle of the age ranges for previous trials, with Abeyagunawardena and Trompeter19 having the youngest population (median age 5.3 years) and Abeyagunawardena et al. 22 having the oldest population (mean age 12.3 years). There was a considerable difference in the duration of nephrotic syndrome prior to recruitment in the two trials that reported this. Gulati et al. ’s20 subjects had been diagnosed only 9.8 and 10.5 months prior to recruitment for intervention and control arms, respectively, whereas children in the trial by Abeyagunawardena et al. 22 were older, with a much longer nephrotic syndrome course (diagnosed 90 and 76.8 months prior to recruitment for intervention and control arms, respectively). The length of time that the PREDNOS2 trial ITT population had SSNS was between these values (i.e. 40 months for those taking prednisolone and 37 months for the placebo arm).

No comment about ethnicity is made in any of the previous studies and it is assumed that the populations comprised a single ethnicity that corresponded to the location of the trial. Nephrotic syndrome is much more common in individuals of South Asian ethnicity47 and three of these previous studies19,20,22 were undertaken in ethnic groups that are considered to have a higher incidence of SSNS. In the mixed-ethnic PREDNOS2 modified ITT population, 21.4% of the population reported South Asian ethnicity (i.e. Pakistani, Indian, Bangladeshi, Sri Lankan, other Asian or mixed white Asian). Thirty of 134 (22.3%) patients in the prednisolone arm and 28 of 137 (20.4%) patients in the placebo arm were of South Asian ethnicity. We undertook a post hoc subgroup analysis that assessed the primary outcome in those of South Asian ethnicity (risk ratio 0.66, 95% CI 0.4 to 1.10), compared with those of other ethnicity (risk ratio 1.11, 95% CI 0.81 to 1.54), which suggested that there may be evidence that daily prednisolone is effective in those of South Asian ethnicity (test for interaction p = 0.09). However, as this was not a planned subgroup analysis, the numbers are small and this needs confirmation in a larger study.

In comparison with previous studies, the PREDNOS2 trial included children on a range of background treatments. Three of the previous studies18–20 included children on maintenance prednisolone only (with levamisole in addition for some patients in one study), with the fourth study22 comprising children on no background therapy only. In Gulati et al. ’s study,20 it is noteworthy that the lower relapse rate seen in children taking background levamisole as well as prednisolone was non-significant.

Abeyagunawardena et al. ’s22 second study replicated their previous design in children on no background immunosuppression and is the only previous study examining the effect of daily low-dose prednisolone on preventing the risk of URR. The results mirrored the group’s earlier findings of a benefit to the intervention. However, compared with all other studies, including the PREDNOS2 trial, this population was an outlier in terms of the longer time for which children had nephrotic syndrome and the lower relapse frequency during the trial, suggesting that their nephrotic syndrome was progressing towards long-term remission. Only the PREDNOS2 trial has considered a real-world scenario of children with relapsing SSNS taking a range of background treatments or none.

The PREDNOS2 trial was also the only trial to include patients by the number of relapses experienced prior to recruitment. The three earlier studies18–20 inferred frequently relapsing or steroid-dependent disease through the need for maintenance corticosteroid. Abeyagunawardena et al. ’s second study22 included children with previous steroid-dependent nephrotic syndrome (SDNS) but who had been off maintenance prednisolone for at least 3 months (it is noteworthy that they had the lowest frequency of relapses during the study period for any trial). The only trial that reported rate of relapses prior to recruitment was Gulati et al. ,20 with a mean of 4.1 relapses per year. That metric is noteworthy because Gulati et al. 20 reported a lower frequency of relapse during the trial itself for both arms [mean (± SD) 0.9 (± 0.4) and 1.8 (± 0.5) for intervention and control arms, respectively]. Reasons for this reduction are not discussed. Given that it was a population of children early in their nephrotic syndrome history, it is most likely that maintenance prednisolone was commenced at some point in the year prior to the trial as a response to frequent relapses. Whatever the explanation, the background reduction in frequency mitigates the extent of the treatment effect they report.

During the trial periods, there were differences in the overall rate of relapses, varying between 1.93 relapses per patient per year for Mattoo and Mahmoud’s population18 and 0.55 relapses per patient per year for Abeyagunawardena et al. ’s second trial. 22 The rate of relapse for the PREDNOS2 trial was 1.67 relapses per patient per year. This difference in relapse frequency may reflect the fact that Abeyagunarwardena et al. ’s second trial comprised a population of children who may have been outgrowing the disease, as discussed above. Interestingly, however, in all of the studies, including the PREDNOS2 trial, the average relapse rate of the control groups did not meet the definition of FRSSNS (i.e. four or more relapses per year) over the study period.

The rate of infections, where reported, was similar across the previous trials, varying between 3.3 and 3.8 infections per patient per year. The overall rate of infections in the PREDNOS2 trial was slightly lower, at 2.95 URTIs per patient per year. The PREDNOS2 trial-recruited population included 80 children (24% of those who completed 12 months of follow-up) who did not have any URTI. Analysis of a modified ITT population of those who did have an URTI was part of the PREDNOS2 trial design, but was not considered by previous studies. Three of these studies18,20,22 did not mention children who did not experience an intercurrent infection during the trial period and the inference is that all children had at least one infection. The study by Abeyagunawardena and Trompeter19 was the only other study to have excluded children for this reason, but with a much lower percentage (6.3%) than that of the PREDNOS2 trial. There was a small difference in age at recruitment between the PREDNOS2 trial and the three earlier studies18–20 and, in general, viral URTIs are more common in younger children. 48 When the modified ITT PREDNOS2 population was split by age into < 4 years (pre-school) and ≥ 4 years (school age), the breakdown of URTI confirmed a greater incidence of URTIs in younger children (Table 28).

A lower rate of intercurrent infection but similar rate of relapses means that either there is a higher incidence of non-URRs in the PREDNOS2 trial population or the children have had more URTIs than have been recorded in the trial. Differences in the URTI-to-relapse ratio may be explained partly by the epidemiology of viral URTIs, suggesting that the types of infections encountered in the UK are less likely to trigger infections than those in more tropical countries. One justification for the PREDNOS2 trial was the need to determine if findings were generalisable to a European temperate climate. The difference in definitions of viral infections, as presented in Table 1, shows that fever was an essential component for three of the four previous studies (although not with Mattoo and Mahmoud’s population18), whereas it could be one or two out of six symptoms to define URTI in the PREDNOS2 trial. In two of those trials,20,22 a temperature of > 38 °C was required, implying a more virulent infection than a typical URTI in the UK. The lower URTI-to-relapse ratio in the PREDNOS2 trial may be explained by a milder viral illness that was less likely to trigger a relapse. Overall, there were more relapses in the placebo arm (237 relapses in 137 patients in the placebo arm vs. 216 relapses in 134 patients in the prednisolone arm), but this small increase in number was non-significant (p = 0.33).

The risk of any individual URTI triggering a relapse, what was described as the event rate in planning the sample size for the trial design, is a different measure to the relapse-to-URTI ratio, as it eliminates the possibility of other relapses being caused by undiagnosed infections. It can be extracted from published data for most of the trials where the number of relapses following URTIs is documented. The appropriate comparator between trials of the property of the infections within the trial population is the measurement of the event rate within the control arm. In Abeyagunawardena and Trompeter’s first (and younger) population,19 a viral infection was likely to trigger a relapse 48% of the time, but only 25% of the time in the second (older) population. 22 In the PREDNOS2 trial, the likelihood was even lower, with just 82 of 407 URTIs (20.1%) within the placebo arm resulting in a relapse. Again, this might be a property of the epidemiology of viral URTIs, but interpretation is problematic because of the large differences in event rates seen between the two trials of Abeyagunawardena et al. 19,22 and it suggests differences in how relapses are triggered in individuals with SSNS. In Abeyagunawardena and Trompeter’s first trial,19 the population comprised children early on in their history of nephrotic syndrome who were taking alternate-day maintenance corticosteroid, whereas in the population of Abeyagunawardena et al. ’s second trial22 children had experienced SSNS for a longer duration and had stopped taking background corticosteroid for at least 3 months before recruitment. The natural history of SSNS is for the disease to enter permanent remission in the majority of patients and this mostly occurs in teenage years. 11 It is self-evident then that as children grow older the risk of URTIs triggering relapses must reduce.

It is also possible, however, that children on maintenance alternate-day prednisolone may be more vulnerable to relapses triggered by intercurrent infection. Evidence to support this hypothesis comes from the historical finding that adrenal suppression is common in children with SSNS and may increase the risk of relapse49 and the more recent finding that low-dose maintenance corticosteroid is more effective in preventing relapses if the same total 48-hour dose is divided into daily doses rather than given in a standard alternate-day fashion. 50

Gulati et al. ’s study20 was the only study to calculate NNT. The authors presented mean (± SD) numbers of relapses per infection in the intervention and control groups of 0.13 (± 0.1) and 0.35 (± 0.2), respectively, giving a mean difference of 0.22 (95% CI 0.16 to 0.28). Using adjusted Poisson regression, Gulati et al. 20 calculated a 59% independent reduction in the risk of relapse following an URTI and estimated that the intervention would result in a reduction in the frequency of relapses to fewer than three per year for one out of six patients with frequent relapses. Note that the NNT calculated is one to reduce relapses below the FRSSNS definition, rather than one required to abolish relapses completely. Given the a priori reduction in relapse frequency seen in the control group from the year prior to recruitment to the 12 months of the trial, this NNT may well be an underestimate. The NNT increases as the URTI rate and the event rate reduce. Therefore, the optimistic NNT calculated by Gulati et al. 20 would be even greater if applied to the lower event rate of Abeyagunawardena et al. ’s second trial,22 implying that the intervention effect may have been considerably overestimated prior to the PREDNOS2 trial.

Comparison of trial design with previous research

Many similar demographics and trial design issues allow for careful comparison of results (Table 29). The average background dose of prednisolone/prednisone was similar for all trials that included children on long-term maintenance prednisolone, including the relevant subgroups in the PREDNOS2 trial. The dose of prednisolone used as trial medication was also similar between all trials. In the PREDNOS2 trial, the dose of prednisolone for children not already taking it as part of long-term maintenance treatment was 15 mg/m², which is equivalent to Abeyagunawardena et al. ’s dose of 0.5 mg/kg. 22,51

The PREDNOS2 trial’s large sample size, double-blinded randomised controlled design and low dropout rate make its data more reliable than those of previous trials. Lack of blinding, exclusion of children whose background therapy either escalated or reduced and exclusion of children where concerns about adherence became apparent may all have introduced bias in previous trials. One other trial design issue that may also have unwittingly introduced bias into previous studies is the duration of trial medication, which ranged from 5 to 7 days. We chose 6 days in the trial design of the PREDNOS2 trial so that an even number would remove any inadvertent bias caused if there were an imbalance in the actual number of doses of trial medication between arms for those children already taking long-term maintenance prednisolone. As this factor was not addressed in previous studies, results may have been inadvertently biased.

Adherence

To the best of our knowledge, the PREDNOS2 trial is the only study of daily corticosteroid at the time of an URTI in children with SSNS that has systematically reported adherence data. Abeyagunawardena et al. 22 reported that 12 patients who failed to report with viral infections, did not maintain daily urine protein records or failed to comply with trial protocol at the time of acute infections were excluded from the final assessment of the study. Other previous studies made no assessment of adherence.

In the PREDNOS2 trial, adherence was assessed by verbal recall of how trial medication had been given during any URTI prior to the trial visit. Counting trial tablets at trial visits, as was performed with the PREDNOS trial,9 was felt to document the number of trial tablets remaining only and not necessarily to reflect how it had been given. Three questions were asked of parents/carers:

1. Was trial medication started at the time of an URTI? If yes, after what length of time?

2. Was the whole 6-day course of trial mediation taken? If no, how many days were missed?

3. Was the 6-day course prematurely discontinued?

Answers to questions 2 and 3 were very similar, but not identical. Question 2 was felt to be a more robust question, covering the eventuality that mid-course doses were missed without the 6-day course finishing prematurely. The percentage of URTIs where medication was commenced at the time of URTI symptoms was 85.4% in the prednisolone arm and 89.2% in the placebo arm. Of those URTIs where medication was commenced, the 6-day course was completed in 82.9% of URTIs in the prednisolone arm and 86.2% of URTIs in the placebo arm. Overall, medication was commenced and a 6-day course completed for 70.8% of URTIs in the prednisolone arm and 76.9% of URTIs in the placebo arm (see Table 29).

Table 30 shows adherence data for individual URTIs in each arm of the trial according to whether or not an URR resulted. Adherence appeared to be better when a relapse was not triggered compared with when there was an URR. In the prednisolone arm, trial medication was commenced and 6 days’ treatment completed in 253 of 298 (84.9%) URTIs where an URR did not occur, compared with 19 of 82 (23.2%) URTIs where an URR did occur. However, this difference in adherence was observed for both the prednisolone arm and placebo arm and, on careful inspection of the data, the most common reason for ‘non-adherence’ was the need to stop trial medication to begin treatment for a relapse.

It is still possible that higher adherence within the prednisolone arm with those URTIs that did not result in URRs may also have meant that beneficial effects of the intervention treatment have been missed because the treatment was not administered as intended, either through medication not being given at all or through an incomplete course of prednisolone. An exploratory per-protocol analysis of those URTIs where medication was started within 2 days of URTI gives an adjusted risk difference of –0.05 (95% CI –0.17 to 0.08).

Secondary outcomes

There was a small difference in the number of relapses between the two trial arms (216 relapses were reported in 91 patients the prednisolone arm vs. 237 relapses in 98 patients in the placebo arm). The difference in the number of patients experiencing a relapse was non-significant (p = 0.33), but the difference in the number of relapses between trial arms was sufficient for the health economic analysis to conclude a benefit in the trial intervention. If a small increase in relapses in the placebo arm was genuine, the NNT calculation that 42 URTIs are required to be treated with 6 days of daily low-dose prednisolone to prevent one URR would be considered unacceptable by most clinicians.

The cumulative dose of prednisolone, which was calculated including the prednisolone used as the trial drug in the intervention arm, was not different between the two arms (median 2060 mg, IQR 1127.5–3355 mg in the prednisolone arm; median 1880 mg, IQR 1115–3295 mg in the placebo arm; p = 0.72 from a Wilcoxon rank-sum test). Therefore, there was no benefit or detriment from occasional use of this additional 6-day course of low-dose prednisolone.

Similar to findings of the PREDNOS study,8 the incidence of most specific known corticosteroid adverse effects was as expected and not different between treatment arms. There was a tendency for increased reporting of poor behaviour in the placebo arm (54% vs. 45%). This was also supported by ACBC scores, which were, on average, lower (better) in the prednisolone arm (although the differences were small). The scores are comparable to those found in the PREDNOS trial;9 however, these ACBC mean total problem T-scores are low compared with those found in a study by Mishra et al. 52 Mishra et al. 52 looked at ACBC individual and total problem T-scores for children with SSNS at different points of the disease course and matched controls. Total problems T-scores for children in both arms of the PREDNOS2 trial fall between the newly diagnosed pre-corticosteroid treatment and the control group in Mishra et al. ’s study (59.8 and 44.7, respectively). In addition, the differences between groups in that study were much greater than the differences seen between treatment arms in the PREDNOS2 trial. Given that the PREDNOS2 trial involved giving low doses of prednisolone and the average relapse rate was relatively low, meaning that relatively few courses of high-dose prednisolone were required, it is perhaps not surprising to see these low ACBC scores.

An expected number of SAEs was seen, with no significant difference between treatment arms. Absence of SUSARs and deaths is consistent with known safety data on prednisolone.

Escalation and reduction of background therapy were considered important secondary outcomes. They were not included in the design of previous trials, in which children who escalated or reduced background therapy were often excluded from the analysis, therefore introducing potential bias. Should the trial intervention be effective, it was hypothesised that patients in the treatment arm should see a reduction in the level of background therapy. Similarly, patients in the placebo arm continuing to experience frequent relapses would be more likely to require escalation in background therapy. There was no difference in escalation of background therapy between treatment arms, with an adjusted risk ratio close to 1 (p = 0.96). In terms of reduction of background therapy, more patients in the placebo than in the prednisolone arm (48.1% vs 43%) moved to a lower level of maintenance treatment, although the difference was non-significant (p = 0.42).

Health economic findings

The economic evaluation combined the costs associated with giving daily low-dose prednisolone at the time of an URTI with the HRQoL benefits and found it to increase QALYs and decrease costs when compared with standard care. This finding was robust to sensitivity analysis and was maintained over the long term, until all patients had reached adulthood (i.e. 18 years).

There are two main reasons for this finding. First, the main driver of the cost difference was the difference in costs of background therapy plus the hospitalisation costs associated with relapse. After 1 year, proportionally more patients in the prednisolone arm had discontinued background therapy and moved to group 1 than in the standard care arm. In addition, there were fewer cases of hospitalisation after relapse (see Table 23). These differences led to an incremental cost saving that favoured the daily low-dose prednisolone arm that was sustained and accumulated over 16 years. Second, when the utility decrement associated with having a relapse was accounted for, this led to low-dose prednisolone gaining more QALYs than standard care.

The model used the trial data over 12 months’ follow-up and projected the findings over 16 years (i.e. until all patients reached 18 years). The robustness of the findings was tested in a PSA by plotting 10,000 paired cost and QALY estimates on the cost-effectiveness plane and CEAC. According to published recommendations, CEACs are considered an effective tool for dealing with uncertainty in economic evaluations and remove reliance from statistics, such as the 95% CIs, which are not always defined for cost-effectiveness ratios. 53 Although the difference in overall relapse rate between the two trial arms did not reach statistical significance in the clinical analysis, the health economic model provided insight into the trade-off between the costs and effects when considered simultaneously. It provided an economic case for administering low-dose prednisolone at the time of an URTI to avoid the high-cost and HRQoL impacts associated with relapse events, and to take account of the costs associated with background therapy, therefore leading to daily prednisolone at the time of an URTI being the recommended option.

These economic findings are particularly interesting as the clinical effectiveness results concluded that using prednisolone at the time of an URTI does not lead to a statistically significant reduction in SSNS relapses and, therefore, should not be routinely recommended as a strategy. Therefore, it appears that the clinical effectiveness and cost-effectiveness results from the PREDNOS2 trial are producing different recommendations. The explanation for this stems from the underlying differences in the theoretical approaches. Unlike the clinical effectiveness analysis, the economic analysis is focused on the ratio of costs and outcomes and assessing value, based on society’s WTP for a unit gain in outcome (QALYs). With this in mind, although the PREDNOS2 trial showed that the difference in SSNS relapse between the treatment arms did not reach statistical significance, the direction of effect favoured the prednisolone arm. As these relapse episodes are frequently associated with high hospital costs and a corresponding reduction in HRQoL, and when this is combined with the relatively cheap costs of the intervention (prednisolone), this small difference in relapse rate, albeit not statistically significant for the clinical effectiveness, combined with the difference in costs associated with background therapy led to the low-dose prednisolone strategy being the ‘dominant’ treatment option. This apparent difference in findings between the clinical effectiveness and cost-effectiveness of this trial emphasises the importance of the morbidity and economic consequences of SSNS relapses and on the underlying costs associated with background therapy.

Strengths of the trial

The strengths of the PREDNOS2 trial are its size and methodology. With 365 patients recruited and 271 patients forming the modified ITT population, it, to the best of our knowledge, constitutes the largest clinical trial of an IMP in childhood nephrotic syndrome ever undertaken. A trial of this scale could be undertaken as a multicentre study only and the PREDNOS2 trial was built on the successful research network developed with its predecessor study (i.e. the PREDNOS trial9). The extremely high case report form return rate (98.6% data completion) (see Figure 5) demonstrates the effectiveness of this research network and adds to the robustness of the data.

The PREDNOS2 trial is unique among other trials evaluating the use of daily corticosteroid to prevent relapses in SSNS in its comprehensive trial design and methodology. In contrast to other trials, the PREDNOS2 trial was double blinded and placebo controlled. Care was taken to avoid bias from excluding groups such as those experiencing corticosteroid adverse effects or escalating background treatment. Rather than excluding patients in whom inadequate adherence was suspected, it sought to capture adherence data systematically and explore reasons for this non-adherence. It was also the only trial, to the best of our knowledge, that attempted to record corticosteroid AEs systematically, include an objective measure of the impact of corticosteroid on behaviour and include a health economic analysis.

Challenges of the trial

The long duration of the trial did present a challenge to try and minimise recruitment fatigue among local research teams. However, from an initial slowing of recruitment, a rate of 4.6 recruits per month was maintained for the last 4 years of the trial, although this might also have reflected a ready prevalent population at the start of the trial, moving to incident patients and slower recruitment once that initial eligible pool had been recruited.

The finding of a lower than anticipated event rate combined with a higher than anticipated number of children not experiencing an URTI within the 12 months of the trial prompted a review of the study that recommended two changes: (1) changing the primary outcome from the incidence of URR following first URTI to incidence of URR following any URTI within the trial duration and (2) increasing the sample size to allow for a 30% attrition rate. In the light of the trial results, the decision taken by the TMG to continue the trial with this change in primary outcome has been vindicated.

Towards the end of the trial, there was a change in chief investigator due to a change in situation for the previous incumbent. The new chief investigator had been a member of the TSC and changeover occurred smoothly.

Possible weaknesses

The trial may be criticised for the large number of excluded children who did not experience an URTI during their 12 months’ follow-up. With hindsight, an alternative strategy might have been to offer re-recruitment for a further 12 months as a more cost-effective way to reach the target population, but those children may not have fulfilled the inclusion criteria of a minimum of two relapses in the 12 months prior to their re-recruitment, therefore bringing into question the application of the trial’s findings.

In contrast to previous trial design, we did not review and examine these children at the time of their URTI and it is possible that some excluded children had experienced URTIs that were not captured within the trial definition. If we had made a type II error by failing to find a significant result that is present, we would have expected to see an increased number of relapses. As discussed above, the small increase in patients experiencing any relapses in the placebo arm was non-significant (p = 0.33).

The economic analysis also had some limitations. First, the cross-walked/mapping technique used to derive utility values for children aged < 5 years in the absence of a validated HRQoL instrument may have introduced additional uncertainty to the economic model. Second, the model ‘allowed’ patients to change their background therapy every 3 months and not more frequently and, although this assumption was based on clinical expert opinion, it may have biased the results. Any uncertainty related to these assumptions was explored in a sensitivity analyses by excluding the under-5s cohort in a subgroup analysis and by relaxing the change in background therapy to every 1 month, respectively, although neither had an impact on the direction of the ICER.

Conclusion

• This is, to the best of our knowledge, the largest by far and most methodologically robust trial addressing this specific clinical question. We have shown that giving 6 days of daily low-dose prednisolone at the time of an URTI does not reduce the risk of relapse of nephrotic syndrome in children on a range of background treatments for relapsing nephrotic syndrome and in a temperate country.

• Although the reduction in relapse associated with the intervention was not statistically significant, health economic analysis has shown a benefit to the intervention (with the high cost and reduction of HRQoL from having a relapse alongside background therapy costs and the low cost of prednisolone).

• Results from previous trials that supported use of daily low-dose corticosteroid at the time of an URTI were small and underpowered. The direction of the much larger, mixed-ethnicity PREDNOS2 trial result was the same, but with non-significant results. This may mean that previous trials produced false-positive results or that there is an ethnicity effect. The size of the PREDNOS2 trial population will outweigh the combined results of all previous studies when future meta-analyses are undertaken.

Recommendations for future research

The following research recommendations, in priority order, follow on from the PREDNOS2 trial conclusions:

• The PREDNOS2 trial data has highlighted the importance of relapse in terms of cost and quality-of-life impact. More research is needed to understand how this might be explained by variations in treatment and across different patient populations.

• More research is needed to establish if the findings of the PREDNOS2 trial, compared with previous trials, are purely a result of a more robust trial design or whether or not disease, georacial or viral epidemiological factors also play a role.

• Large registry studies, such as the RaDaR rare disease registry,54 have the potential to study differences in SSNS disease course and the reasons for them in different racial groups within a diverse population, such as the UK.

• The PREDNOS2 trial recruited more efficiently than the PREDNOS trial. 9 The PREDNOS trial research network, comprising local research teams in > 100 local hospitals and an experienced central clinical trials team, has now delivered two large multicentre interventional trials. This puts the network in a unique position to deliver further efficient trials towards evidence-based management of SSNS in children.

• Virology studies of the specific causes of URTIs experienced by children with SSNS would help us to understand whether or not specific organisms are more likely to trigger relapses and, in so doing, provide the potential for greater understanding of the pathophysiology that results in a relapse.

• The PREDNOS2 trial has shown that low-dose corticosteroid at the time of an URTI before the onset of any proteinuria does not reduce the risk of relapse and, moreover, the risk of an URTI resulting in a relapse is relatively low at 21%. A related research question is whether or not a relapse can be prevented by the earlier-than-usual commencement of a lower-than-usual dose of prednisolone. In the last few years, there has also been renewed interest in whether or not relapses can be treated with lower doses of prednisolone. 55–57 In addition, in the latest Cochrane update published in 2020,4 authors have called for a large trial to verify these findings, which would open the gateway to exploring more creative, personalised ways to manage relapses that limit the cumulative dose of corticosteroid. With its strong research network, the PREDNOS trial group would be well placed to continue to deliver the multicentre trials that are required to provide the evidence basis for this debilitating chronic childhood illness.

• The role of the adrenal axis in nephrotic syndrome and the potential association of adrenal suppression with an increased risk of relapse is a long-forgotten area of SSNS that has been highlighted anew in the latest Cochrane update of corticosteroid use in SSNS. 4

Trial Management Group

Professor Nicholas Webb (Royal Manchester Children’s Hospital, Manchester University NHS Foundation Trust, chief investigator up to May 2018); Dr Martin Christian (Nottingham Children’s Hospital, chief investigator from June 2018); Miss Natalie Ives (Senior Statistician, Birmingham Clinical Trial Unit); Professor Emma Frew (Senior Lecturer in Health Economics, University of Birmingham); and Mrs Elizabeth Brettell (Renal Trials Manager, BCTU).

Patient and public representatives

Mrs Wendy Cook (Director, NeST) and Mrs Sandra Cope.

Trial Steering Committee

Dr Megan Thomas (Consultant Community Paediatrician, Blackpool Teaching Hospitals NHS Trust, independent chairperson); Professor Nicholas JA Webb (Consultant Paediatric Nephrologist, Manchester Children’s Hospital, chief investigator, non-independent member); Dr Martin T Christian (Consultant Paediatric Nephrologist, Nottingham Children’s Hospital, chief investigator, non-independent member); Mrs Sandra Cope (consumer representative, independent member); Dr Andrew Duncan (Consultant Paediatrician, Borders General Hospital, Melrose, independent member); Dr Kate Hillman (Consultant Nephrologist, Manchester Royal Infirmary, independent member); Dr Zala Ibrahim (Consultant Paediatrician, Russells Hall Hospital, Dudley, independent member); Dr Ly-Mee Yu (Senior Medical Statistician, Nuffield Department of Primary Care Health Sciences, University of Oxford, independent member); Dr Nigel Coad (Consultant Paediatrician, University Hospitals of Coventry and Warwickshire NHS Trust, independent member); Natalie Ives (co-investigator, Senior Statistician, non-independent member); and Darren Green (Consultant Acute Physician and Nephrologist, Salford Royal NHS Foundation Trust, independent member).

Data Monitoring Committee

Professor Philip Kalra (chairperson, Consultant Nephrologist and Honorary Professor, Salford Royal Hospital NHS Trust); Professor Saul Faust (Reader in Paediatric Immunology and Infectious Diseases, Southampton General Hospital); and Dr Andrea Marshall (Senior Research Fellow in Medical Statistics, Warwick Clinical Trials Unit).

Participating centres and PREDNOS2 trial collaborative group members

Addenbrooke’s Hospital, Cambridge University Hospitals NHS Foundation Trust: Dr Birgit Ulbrich (principal investigator), Dr Peter Heinz and Sara Foster.

Airedale General Hospital, Airedale NHS Foundation Trust: Dr Pronab Bala (principal investigator).

Altnagelvin Area Hospital, Western Health and Social Care Trust: Dr Damien Armstrong (principal investigator), Julie Brown and team.

Arrowe Park Hospital, Wirral University Teaching Hospital NHS Foundation Trust: Dr Elizabeth Breen (principal investigator), Sharon Hughes and team.

Barnsley Hospital, Barnsley Hospital NHS Foundation Trust: Dr Rajeev Gupta (principal investigator) and Miranda Murray.

Doncaster Royal Infirmary & Bassetlaw Hospital, Doncaster and Bassetlaw Hospitals NHS Foundation Trust: Dr Mathew Kurian (principal investigator), Dr Sunitha Sampath, Dr Sarah Didier, Sarah Shortland, Sharon Allen, Lisa Warren and team.

Birmingham Children’s Hospital, Birmingham Women’s and Children’s NHS Foundation Trust: Dr David Milford (principal investigator), Dr Mordi Muorah, Alison Watson, Fatima Bibi and team.

Bradford Royal Infirmary, Bradford Teaching Hospitals NHS Foundation Trust: Dr Eduardo Moya (principal investigator).

Bristol Royal Hospital for Children, University Hospitals Bristol NHS Foundation Trust: Professor Moin Saleem (principal investigator), Dr Alison Kelly, Michelle Ross, Natalie Fineman and team.

Calderdale Royal Hospital & Huddersfield Royal Infirmary, Calderdale and Huddersfield NHS Foundation Trust: Dr Eilean Crosbie (principal investigator), Rachel Swingler, Susan Kilroy and team.

Chesterfield Royal Hospital, Chesterfield Royal Hospital NHS Foundation Trust: Dr Oyekunle Ayonrinde (principal investigator) and Amanda Smith.

Colchester General Hospital, Colchester Hospital University NHS Foundation Trust: Dr Andrea Turner (principal investigator), Dr Jonathan Campbell and Aine Turner.

Countess of Chester Hospital, Countess of Chester Hospital NHS Foundation Trust: Dr Stephen Brearey (principal investigator), Caroline Burchett and Sarah De-Beger.

Croydon University Hospital, Croydon Health Services NHS Trust: Dr Theo Fenton (principal investigator).

Cumberland Infirmary, North Cumbria University Hospitals NHS Trust: Dr Glyn Jones (principal investigator) and Nicci Kelsall.

Darent Valley Hospital, Dartford and Gravesham NHS Trust: Dr Selwyn D’Costa (principal investigator).

Darlington Memorial Hospital, University Hospital of North Durham, County Durham and Darlington NHS Foundation Trust: Dr Dinakaran Jayachandran (principal investigator), Dr Asha Nair, Catherine Tarn Nozedar and Dawn Egginton.

Derriford Hospital, Plymouth Hospitals NHS Trust: Dr Oliver Cuthell (principal investigator), Dr Catherine Derry, Dr Kathiresan Natesan and Sarah-Jane Sharman.

Dewsbury and District Hospital, Pinderfields Hospital, The Mid Yorkshire Hospitals NHS Trust: Dr Rajeeva Singh (principal investigator), Dr Kathryn Deakin, Gail Castle and team.

Diana, Princess of Wales Hospital (Grimsby), Northern Lincolnshire and Goole Hospitals NHS Foundation Trust: Dr Bukar Wobi (principal investigator), Dr Bemigho Etuwewe, Caroline Burnett and team.

East Surrey Hospital, Surrey and Sussex Healthcare NHS Trust: Dr Kamal Khoobarry (principal investigator).

Eastbourne District General Hospital, Conquest Hospital, East Sussex Healthcare NHS Trust: Dr Graham Whincup (principal investigator) and Anne Cowley.

Evelina London Children’s Hospital, Guy’s and St Thomas’ NHS Foundation Trust: Dr Ania Koziell (principal investigator), Dr Manish Sinha, Dr Christopher Reid, Mohammad Ahmad and team.

Fairfield General Hospital, North Manchester General Hospital, The Royal Oldham Hospital, Rochdale Infirmary, The Pennine Acute Hospitals NHS Trust: Dr Beena Padmakumar (principal investigator), Dr Talaivirichan Magadevan, Grainne O’Connor and Louise Woodhead.

Royal Lancaster Infirmary & Furness General Hospital, University Hospitals of Morecambe Bay NHS Foundation Trust: Dr Mireille Formosa (principal investigator), Dr Nayan Peepah Nardeosingh and Kathryn Allison.

Glan Clwyd Hospital, Betsi Cadwaladr University Health Board: Dr Markus Hesseling (principal investigator), Annette Bolger, Lucie Hobson and team.

Gloucestershire Royal Hospital, Gloucestershire Hospitals NHS Foundation Trust: Dr Adamu Sambo (principal investigator), Dr Lyda Jadresic and Susan Beames.

Great Ormond Street Hospital, Great Ormond Street Hospital for Children NHS Foundation Trust: Professor Detlef Bockenhauer (principal investigator), Dr Daljit Hothi, Elizabeth Vella, Corinne Linton, Shaima Yussuf, Tendai Bazaya, Mahmoud Abou-Rayyah and team.

Great Western Hospital, Great Western Hospitals NHS Foundation Trust: Dr Nick West (principal investigator).

Homerton University Hospital, Homerton University Hospital NHS Foundation Trust: Dr Rajiv Sood (principal investigator), Hilarious De Jesus and team.

Hull Royal Infirmary, Hull and East Yorkshire Hospitals NHS Trust: Dr Vikas Gupta (principal investigator) and Dr Verghese Mathew.

James Paget University Hospital, James Paget University Hospitals NHS Foundation Trust: Dr Esi Bentsi-Enchill (principal investigator) and Allyson Davison.

John Radcliffe Hospital, Oxford University Hospitals NHS Trust: Dr Janet Craze (principal investigator).

Kent and Canterbury Hospital, Queen Elizbaeth the Queen Mother Hospital, William Harvey Hospital, East Kent Hospitals University NHS Foundation Trust: Dr Elhussein Rfidah (principal investigator), Janine Musselwhite, Angela Moon and team.

Kettering General Hospital, Kettering General Hospital NHS Foundation Trust: Dr Harsha Bilolikar (principal investigator), Sonia White and team.

King’s Mill Hospital, Sherwood Forest Hospitals NHS Foundation Trust: Dr Simon Rhodes (principal investigator) and Caroline Moulds.

Leeds General Infirmary, Leeds Teaching Hospitals NHS Trust: Dr Hitesh Prajapati (principal investigator), Dr Eric Finlay, Dr Pallavi Yadav, Dr Amanda Newnham, Dr Kay Tyerman, Majorie Allen, Lucy Wellings and team.

Leicester Royal Infirmary, University Hospitals of Leicester NHS Trust: Dr Angela Hall (principal investigator), Jackie Philps and team.

Leighton Hospital, Mid Cheshire Hospitals NHS Foundation Trust: Dr Subajini Kaviethasan (principal investigator), Sally Smith and team.

Pilgrim Hospital & Lincoln County Hospital, United Lincolnshire Hospitals NHS Trust: Dr David Broodbank (principal investigator), Dr Sourabh Mukhopadhyay, Dr Ruchika Gupta, Amanda Roper, Susie Butler and team.

Luton and Dunstable University Hospital, Luton and Dunstable Hospital NHS Foundation Trust: Dr Tomasz Rajkowski (principal investigator), Dr Michael Eisenhut, Karen Duncan, Karen Samm, Samantha Clough and team.

Macclesfield District General Hospital, East Cheshire NHS Trust: Dr David Wright (principal investigator), Dr Krishnakumar Thattakkat, Dr Ignatius Losa, Natalie Keenan and team.

Maidstone Hospital, Tunbridge Wells Hospital, Maidstone and Tunbridge Wells NHS Trust: Dr Krishnan Balasubramanian (principal investigator).

Manor Hospital, Walsall, Walsall Healthcare NHS Trust: Dr Muhammad Javed (principal investigator), Sharon Kempson, Marie Phipps and team.

Medway Maritime Hospital, Medway NHS Foundation Trust: Dr Janette Cansick (principal investigator) and Maines Msiska.

Milton Keynes Hospital, Milton Keynes University Hospital NHS Foundation Trust: Dr Lazarus Anguvaa (principal investigator), Dr Mya Aye, Sally Conway, Natalie Beer, Francesca Wright and team.

Musgrove Park Hospital, Taunton and Somerset NHS Foundation Trust: Dr Jennifer Langlands (principal investigator), Kirsty O’Brien, Nicola Thorne and team.

New Cross Hospital, The Royal Wolverhampton Hospitals NHS Trust: Dr Karen Davies (principal investigator), Sharon Kempson, Marie Phipps and team.

Newham University Hospital, The Royal London Hospital, Whipps Cross Hospital, Barts Health NHS Trust: Dr Ami Parikh (principal investigator), Dr Nimze Gadong, Dr Bahadur Anjum, Nicolene Plaatjies, Ivone Lancoma-Malcolm, Hilarious De Jesus and team.

Norfolk and Norwich University Hospital, Norfolk and Norwich University Hospitals NHS Foundation Trust: Dr Vipan Datta (principal investigator), Dr Chris Upton, Louisa Fear, Louise Coke and team.

North Devon District Hospital, Northern Devon Healthcare NHS Trust: Dr Dermot Dalton (principal investigator), Becky Holbrook and team.

Northampton General Hospital, Northampton General Hospital NHS Trust: Dr Imogen Norton (principal investigator).

Nottingham Children’s Hospital (Queens Medical Centre), Nottingham University Hospitals NHS Trust: Dr Martin Christian (principal investigator), Dr Andrew Lunn, Olivia Vincent, Helen Navarra, Neelam Khan and team.

Peterborough City Hospital, Peterborough and Stamford Hospitals NHS Foundation Trust: Dr Mona Aslam (principal investigator) and Paula Goodyear.

Poole Hospital, Poole Hospital NHS Foundation Trust: Dr Steve Wadams (principal investigator), Susan Power, Amy Roff and team.

Princess Royal Hospital, Royal Shrewsbury Hospital, The Shrewsbury and Telford Hospital NHS Trust: Dr Manish Gupta (principal investigator), Dr Naeem Ayub, Charlotte Owen and team.

Queen Alexandra Hospital (Portsmouth), Portsmouth Hospitals NHS Trusta: Dr Judith Scanlan (principal investigator), Sharon McCready and Andrew Gribbin.

Queen’s Hospital, Burton, Burton Hospitals NHS Foundation Trust: Dr Mansoor Ahmed (principal investigator), Dr Dominic Muogbo, Dr Heather Carswell, Stephanie Boswell, Claire Backhouse and team.

Queen’s Hospital, Romford, Barking, Havering and Redbridge University Hospitals NHS Trust: Dr Junaid Solebo (principal investigator) and Helen Smith.

Raigmore Hospital, NHS Highland: Dr Alan Webb (principal investigator), Ing-Marie Logie and Sandra Dekker.

Rotherham Hospital, The Rotherham NHS Foundation Trust: Dr Sanjay Suri (principal investigator), Janet Shackleton and team.

Royal Aberdeen Children’s Hospital, NHS Grampian: Dr Craig Oxley (principal investigator), Margaret Connon and team.

Royal Albert Edward Infirmary, Wrightington, Wigan and Leigh NHS Foundation Trust: Dr Vineeta Joshi (principal investigator), Nicola Pemberton and team.

Royal Alexandra Hospital, NHS Greater Glasgow and Clyde: Dr Heather Maxwell (principal investigator), Dr Amita Sharma, Elizabeth Waxman and team.

Royal Belfast Hospital for Sick Children, Belfast Health and Social Care Trust: Dr Karl McKeever (principal investigator), Muriel Millar and team.

Royal Berkshire Hospital, Royal Berkshire NHS Foundation Trus: Dr Ann Gordon (principal investigator), Dr Susan Edees, Susan Hallett and team.

Royal Blackburn Hospital, East Lancashire Hospitals NHS Trust: Dr Javed Iqbal (principal investigator), Dr Beate von Bremen, Heather Collier, Andrew Lancaster and team.

Royal Bolton Hospital, Bolton NHS Foundation Trust: Dr Fiona Watson (principal investigator), Joanne Henry and team.

Royal Derby Hospital, Derby Hospitals NHS Foundation Trust: Dr Richard Bowker (principal investigator) and Coral Smith.

Royal Devon & Exeter Hospital (Wonford), Royal Devon & Exeter NHS Foundation Trust: Dr Hannah Cottis (principal investigator), Dr Rebecca Samuel, Caroline Harrill, Suzanne Wilkins and team.

Royal Hospital for Children (Glasgow), NHS Greater Glasgow and Clyde: Dr Heather Maxwell (principal investigator), Dr Ben Reynolds, Dr David Hughes, Elizabeth Waxman and team.

Royal Hospital for Sick Children (Edinburgh), NHS Lothian: Dr Ben Reynolds (principal investigator), Dr David Hughes, Tracey McGregor, Maxine Ramsay, Julie Baggott, Naomi Matos and team.

Royal Liverpool Children’s Hospital (Alder Hey), Alder Hey Children’s NHS Foundation Trust: Dr Caroline Jones (principal investigator), Dr Henry Morgan, Dr Richard Holt, Dr Louise Oni, Theresa Moorcroft, Joanne Shakeshaft and team.

Royal Manchester Children’s Hospital, Central Manchester University Hospitals NHS Foundation Trust: Dr Mohan Shenoy (principal investigator), Professor Nicholas Webb, Dr Amrit Kaur, Dr Dean Wallace, Dr Nicholas Plant, Dr Shaila Sukthankar, Angela Branson, Helen Blackburn, Jane Howell, Jess Nichols and team.

Royal Stoke University Hospital, The County Hospital (Stafford Hospital), University Hospitals of North Midlands NHS Trust: Dr Furqan Basharat (principal investigator), Dr Saeeda Raja, Marie Phipps, Helen Parker, Joanne Tomlinson, Eric Roe and team.

Royal United Hospital (Bath), Royal United Hospital Bath NHS Foundation Trust: Dr Lynn Diskin (principal investigator) and Alison Barratt.

Russells Hall Hospital, The Dudley Group NHS Foundation Trust: Dr Subramanian Mahadevan-Bava (principal investigator), Abigail Weston and Daljit Kaur.

Scarborough Hospital, York Teaching Hospital NHS Foundation Trust: Dr Udupa Venkatesh (principal investigator), Emma Temlett, Simon Dyer, Kerry Elliott, Rosie Furness and team.

Sheffield Children’s Hospital, Sheffield Children’s NHS Foundation Trust: Dr Andrew Lunn (principal investigator), Janet Shackleton, Sarah Shortland, Miranda Murray and team.

Southampton Children’s Hospital, University Hospital Southampton NHS Foundation Trust: Dr Rodney Gilbert (principal investigator), Dr Matthew Harmer, Dr Shuman Haq, Lisa Fairhead, Louise Haskell, Victoria Bingham and team.

Southend Hospital, Southend University Hospital NHS Foundation Trust: Dr Anupam Shrivastava (principal investigator), Onie Hove and Bernard Hadebe.

St Mary’s Hospital, Isle of Wight NHS Trust: Dr Christopher Magier (principal investigator), Bettina Harms and Sian Butterworth.

St Peter’s Hospital, Ashford and St Peter’s Hospitals NHS Foundation Trust: Dr Tariq Bhatti (principal investigator), Dr Aisling Parker, Lorna Walding and team.

St Richard’s Hospital, Worthing Hospital, Western Sussex Hospitals NHS Foundation Trust: Dr Nicholas Brennan (principal investigator).

Stepping Hill Hospital, Stockport NHS Foundation Trust: Dr Chris Cooper (principal investigator), Sara Bennett and team.

Tameside Hospital, Tameside Hospital NHS Foundation Trust: Dr Anjali Date (principal investigator), Dr Anjali Petkar, Wendy Hulse and team.

Worcestershire Royal Hospital, The Alexandra Hospital (Redditch), Worcestershire Acute Hospitals NHS Trust: Dr Munir Ahmed (principal investigator), Dr Tom Dawson, Connie Rowlands and Stephanie Chamberlain.

The County Hospital, Hereford, Wye Valley NHS Trust: Dr Simon Meyrick (principal investigator), Dr Iain Darwood and Emma Collins.

The Friarage Hospital, The James Cook University Hospital, South Tees NHS Foundation Trust: Dr Rajesh Lall (principal investigator), Dr Elizabeth Onifade, Joanna Green and team.

The Great North Children’s Hospital, The Newcastle Upon Tyne Hospitals NHS Foundation Trust: Dr Sally Johnson (principal investigator), Dr Heather Lambert, Dr Yincent Tse, Dr Michal Malina, Dr Vijaya Sathyanarayan, Jenny Booth, Kathryn Bell, Stephen Crulley and team.

The Ipswich Hospital, The Ipswich Hospital NHS Trust: Dr Jackie Buck (principal investigator), Deborah Beeby, Louise Hunt and team.

The York Hospital, York Teaching Hospital NHS Foundation Trust: Dr Sundeep Sandhu (principal investigator), Dr Gur Millman, Dr Murray Wheeler, Anna Clayton, David Thompson and team.

University Hospital Crosshouse, NHS Ayrshire and Arran: Dr Bridget Oates (principal investigator), Claire Bell, Joanna Wardrop and team.

University Hospital of North Tees, North Tees and Hartlepool NHS Foundation Trust: Dr Vijay Tandle (principal investigator), Carolyn Campbell, Dawn Egginton and team.

University Hospital of Wales, Cardiff and Vale University Health Board: Dr Shivaram Hegde (principal investigator), Dr Rajesh Krishnan, Zoe Morrison, Jennifer Muller, Louise Yendle and team.

Warrington Hospital, Warrington and Halton Hospitals NHS Foundation Trust: Dr Delyth Webb (principal investigator) and Natalie Rogers.

West Middlesex University Hospital, Chelsea and Westminster Hospital NHS Foundation Trust: Dr Nour Elhadi (principal investigator), Dr Dipali Shah, Amrinder Sayan and team.

West Suffolk Hospital, West Suffolk NHS Foundation Trust: Dr Karine Cesar (principal investigator), Dr Raman Lakshman and Helen Cockerill.

Wexham Park Hospital, Frimley Health NHS Foundation Trust: Dr Zilla Huma (principal investigator).

Wishaw General Hospital, NHS Lanarkshire: Dr Thin Thin Saing (principal investigator), Angela Brown, Karen Leitch and team.

Wythenshawe Hospital, University Hospital of South Manchester NHS Foundation Trust: Dr Gopi Vemuri (principal investigator), Claire Holliday, Jessica Carey, Louise Woodhead and team.

Ysbyty Gwynedd, Betsi Cadwaladr University Health Board: Dr Madalitso Kubwalo (principal investigator), Annette Bolger and team.

Contributions of authors

Martin T Christian (https://orcid.org/0000-0003-1642-1961) (subsequent chief investigator, Consultant Paediatric Nephrologist) took over as chief investigator in 2018, led and developed the trial design, led the TMG and provided causality assessment for all SAEs, was an original co-applicant of the grant application, was a principal investigator for their hospital and was the lead author of the report.

Nicholas JA Webb (https://orcid.org/0000-0001-8572-5446) (original chief investigator, Consultant Paediatric Nephrologist) led and developed the trial design, led the TMG and provided causality assessment for all SAEs until 2018, was an original co-applicant of the grant application and was a principal investigator for their hospital.

Rebecca L Woolley (https://orcid.org/0000-0001-5119-1431) (Statistician) provided statistical support throughout the study and developed and led the planned statistical analyses.

Nafsika Afentou (https://orcid.org/0000-0003-1337-7029) (Fellow in Health Economics) designed, conducted and wrote the health economic evaluation.

Samir Mehta (https://orcid.org/0000-0001-7096-513X) (Statistician) designed and undertook the final statistical analysis.

Emma Frew (https://orcid.org/0000-0002-5462-1158) (Professor in Health Economics) designed, conducted and wrote the health economic evaluation.

Elizabeth A Brettell (https://orcid.org/0000-0002-0669-3085) (Clinical Trials Team Leader) led the trial team at BCTU, overseeing the day-to-day management of the trial and data collection.

Adam R Khan (https://orcid.org/0000-0001-5366-233X) (Trial Manager) led the trial team at BCTU, overseeing the day-to-day management of the trial and data collection.

David V Milford (https://orcid.org/0000-0003-4737-2744) (Consultant/Senior Paediatric Nephrologist) was an original co-applicant of the grant application and was a principal investigator for their hospital.

Detlef Bockenhauer (https://orcid.org/0000-0001-5878-941X) (Consultant Paediatric Nephrologist) was principal investigator for his hospital.

Moin A Saleem (https://orcid.org/0000-0002-9808-4518) (Consultant Paediatric Nephrologist) was an original co-applicant of the grant application and was a principal investigator for their hospital.

Angela S Hall (https://orcid.org/0000-0003-2300-5421) (Associate Specialist Paediatrician with Interest in Nephrology) was an original co-applicant of the grant application and was a principal investigator for their hospital.

Ania Koziell (https://orcid.org/0000-0003-4882-0246) (Consultant Paediatric Nephrologist) was principal investigator for her hospital.

Heather Maxwell (https://orcid.org/0000-0002-5904-1718) (Consultant Paediatric Nephrologist) was an original co-applicant of the grant application and was a principal investigator for their hospital.

Shivaram Hegde (https://orcid.org/0000-0003-3573-9940) (Consultant Paediatric Nephrologist) was an original co-applicant of the grant application and was a principal investigator for their hospital.

Eric R Finlay (Consultant Paediatric Nephrologist) was an original co-applicant of the grant application and was a principal investigator for their hospital.

Rodney D Gilbert (https://orcid.org/0000-0001-7426-0188) (Consultant Paediatric Nephrologist) was an original co-applicant of the grant application and was a principal investigator for their hospital.

Caroline Jones (https://orcid.org/0000-0001-9442-9286) (Consultant Paediatric Nephrologist) was an original co-applicant of the grant application and was a principal investigator for their hospital.

Karl McKeever (https://orcid.org/0000-0002-5010-0398) (Consultant/Senior Paediatric Nephrologist) was an original co-applicant of the grant application and was a principal investigator for their hospital.

Wendy Cook (https://orcid.org/0000-0001-8360-2813) (Director for NeST) provided patient and public involvement at the point of trial concept and design, was a member of the TSC and read the report manuscript.

Natalie Ives (https://orcid.org/0000-0002-1664-7541) (Reader in Clinical Trials, Senior Statistician) provided the statistical support throughout the study, and developed and led the planned statistical analyses.

Publications

Webb NJA, Frew E, Brettell EA, Milford DV, Bockenhauer D, Saleem MA, et al. Short course daily prednisolone therapy during an upper respiratory tract infection in children with relapsing steroid-sensitive nephrotic syndrome (PREDNOS 2): protocol for a randomised controlled trial. Trials 2014;15:1–12.

Christian MT, Webb NJA, Mehta S, Woolley RL, Afentou N, Frew E, et al. Evaluation of daily low-dose prednisolone during upper respiratory tract infection to prevent relapse in children with relapsing steroid-sensitive nephrotic syndrome: the PREDNOS randomized clinical trial [published online ahead of print December 20 2021]. JAMA Pediatr 2021.

Data-sharing statement

All data requests should be submitted to the corresponding author for consideration. Access to anonymised data may be granted following review.

Patient data

This work uses data provided by patients and collected by the NHS as part of their care and support. Using patient data is vital to improve health and care for everyone. There is huge potential to make better use of information from people’s patient records, to understand more about disease, develop new treatments, monitor safety, and plan NHS services. Patient data should be kept safe and secure, to protect everyone’s privacy, and it’s important that there are safeguards to make sure that it is stored and used responsibly. Everyone should be able to find out about how patient data are used. #datasaveslives You can find out more about the background to this citation here: https://understandingpatientdata.org.uk/data-citation.

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 and Social Care. 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 and Social Care.