Assessing a 12-month course of oral alendronate for adults with avascular necrosis of the hip: MANTIS RCT with internal pilot

Article history

The research reported in this issue of the journal was funded by the HTA programme as project number 15/39/06. The contractual start date was in June 2017. The draft report began editorial review in March 2021 and was accepted for publication in March 2022. 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 Glyn-Jones et al. This work was produced by Glyn-Jones 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 Glyn-Jones et al.

Trial summary

The Managing Avascular Necrosis Treatments: an Interventional Study (MANTIS) trial was, to our knowledge, the first Phase IV UK trial designed to determine the clinical effectiveness and cost-effectiveness of a 12-month course of oral alendronate (70 mg weekly) compared with a matched placebo in reducing the progression of avascular necrosis (AVN) of the hip.

The trial was due to open in a minimum of 10 NHS hospitals, with 280 participants planned to be randomised to one of two treatments:

1. 70 mg of alendronate given as an oral tablet weekly for 12 months

2. placebo given as an oral tablet weekly for 12 months.

Background and rationale

Avascular necrosis of the femoral head is an uncommon but devastating condition that affects around 0.001% of the population in developed countries1–3 and occurs when the bone of the hip joint dies, resulting in significant disability in many cases. In the USA, an estimated 10,000–20,000 cases of AVN are diagnosed each year. 3 AVN typically affects younger people, with 45% of patients developing the condition before the age of 30 years. 1 In the early stages of AVN, people often feel pain, followed by a rapid collapse of the joint, which leads to severe immobility and suffering in over 60% of patients. This means that patients often cannot walk comfortably, cannot work and have sleep disturbances. 3,4

The natural history of AVN is well documented; however, its exact pathophysiology is poorly understood. There have been several postulated mechanisms based on whether AVN is classified as post-traumatic or atraumatic. 5 There are also several established risk factors for AVN; glucocorticoid (steroid) therapy and alcoholism are two of the most common associations. Less commonly associated risk factors include sickle cell disease; radiotherapy; cocaine, methadone and heroin use; and Caisson’s disease. 3

Trial design

The MANTIS trial was designed to be a 66-month project with an initial pilot phase leading into a definitive, multisite, two-arm, parallel-group, placebo-controlled, double-blind, Phase IV randomised controlled superiority trial. Its objective was to evaluate and compare the clinical effectiveness of oral alendronate with a plaebo-matched control for participants with AVN. We planned to assess primary outcomes after 12 (short term) and 36 months (long term), and secondary outcomes at 6, 12, 24 and 36 months.

The trial was preceded by a feasibility study, which identified challenges in identifying patients with AVN of the hip, owing to a low prevalence rate and significant regional variations in at-risk groups and service delivery. To address this concern, the trial was designed with an internal pilot phase, with the intention of identifying hospitals/specialties that would be able to recruit adequate numbers of patients. According to the prespecified go/no-go criteria, the trial would be stopped if key recruitment targets could not be met.

The trial was designed as a placebo-controlled blinded trial with participants, clinicians, pharmacists and the research teams all blinded to the treatment allocation. The trial statistician who set up and monitored the randomisation schedules and prepared the unblinded data for review at independent data monitoring committee meetings was unblinded. Active and placebo oral medication were produced and encapsulated and packaged identically. The trial had two groups (Figure 1):

1. intervention group – active alendronate (70 mg taken as one tablet weekly for 12 months)

2. control group – matched placebo (taken as one tablet weekly for 12 months).

FIGURE 1.

Trial schema. PROM, patient-reported outcome measure; SmPC, summary of product characteristics. a, Treatment must be collected within 14 days of randomisation.

Research governance

The trial was sponsored by the University of Oxford, with regulatory compliance oversight by Oxford Clinical Trials Research Unit (OCTRU) through the Surgical Interventional Trials Unit (SITU).

An ethics application under the Integrated Research Application System (IRAS) ID 230545 was submitted to the South Central – Oxford A Research Ethics Committee on 17 April 2018 and the MANTIS trial received a favourable Research Ethics Committee (REC) opinion on 29 May 2018 (reference number 18/SC/0247) and Health Technology Assessment (HTA) approval on 26 June 2018.

A clinical trial authorisation application under the European Union Drug Regulating Authorities Clinical Trials (EudraCT) number: 2017-002798-21 was submitted to the Medicines and Healthcare products Regulatory Agency (MHRA) on 23 April 2018, updated with further information on 8 June 2018 and the trial gained MHRA approval on 19 June 2018.

To ensure transparency, the trial was registered with the International Standard Randomised Controlled Trial Number (ISRCTN) Registry on 11 June 2018 and was assigned ISRCTN14015902. 23

The trial applied for National Institute for Health and Care Research (NIHR) Clinical Research Network (CRN) Portfolio status and was eligible for this support. The lead supporting network was the Thames Valley and South Midlands CRN.

Global NHS permissions were obtained through the IRAS and local NHS permissions were obtained from each participating NHS trust. A clinical trial site agreement based on the model non-commercial research agreement was signed by each participating NHS trust and the sponsor.

Following NIHR guidelines, a fully independent Data Safety and Monitoring Committee (DSMC) and a 66% independent Trial Steering Committee (TSC) were convened. Both committees met throughout the lifetime of the trial.

Management of the trial

The trial manager (Gemma Greenall) was responsible for day-to-day management of the trial, with support from the data manager (Alistair Gray) and trial statistician (NP). The Trial Management Group (TMG) was responsible for overseeing day-to-day management of the trial and comprised the chief investigator (CI) (SGJ), the lead investigator (Kassim Javaid), the trial statisticians (IR, RK and NP), the trial manager (Gemma Greenall) and health economist (MD).

Design and development of the protocol

Clinicians (including surgeons, rheumatologists and nurses), as well as patient representatives, were invited to discuss the trial protocol. Their feedback was utilised by the applicants when designing and developing the protocol.

Amendments to the trial protocol

Following receipt of a favourable opinion of the trial protocol from the REC on 29 May 2018, no changes were made to the trial protocol – the trial started and ended on version 1.0, dated 10 April 2018.

Amendments to the trial

Following the main REC approval, nine substantial amendments were submitted and received favourable opinion. As the trial was a clinical trial of an investigational medicinal product (IMP), any new sites had to be added as a substantial amendment. The nine substantial amendments were as follows:

1. Change to the investigational medicinal product dossier (IMPD) and updated summary of medicinal product characteristics (SmPC) to be used by the trial (24 August 2018).

2. Addition of one new trial site (Royal Orthopaedic Hospitals NHS Foundation Trust) (12 October 2018).

3. Addition of four new trial sites (Milton Keynes University Hospital NHS Foundation Trust, Taunton and Somerset NHS Foundation Trust, Royal National Orthopaedic Hospital NHS Foundation Trust and St George’s University Hospitals NHS Foundation Trust) (23 January 2019).

4. Addition of two new trial sites (Ashford and St Peter’s Hospitals NHS Foundation Trust and Royal Cornwall Hospitals NHS Trust) (19 March 2019).

5. Addition of two new trial sites (University Hospitals of North Midlands NHS Trust and Royal Free London NHS Foundation Trust) (6 June 2019).

6. Addition of one new trial site (Ayrshire & Arran Health Board) (23 July 2019).

7. Addition of six new trial sites (North Bristol NHS Trust, Maidstone and Tunbridge Wells NHS Trust, South Tyneside and Sunderland NHS Foundation Trust, Sandwell and West Birmingham Hospitals NHS Trust, Cambridge University Hospitals NHS Foundation Trust and Swansea Bay University Health Board) (13 November 2019).

8. Notification of temporary halt of the trial owing to recruitment being substantially below targets set in the original and revised recruitment timelines. Following discussions between the funder and TMG on how to proceed with the trial, it was decided that no further participants would be recruited (5 December 2019).

9. Confirmation that the trial would be closed after the temporary halt. Addition of patient-facing documentation to inform participants of what is happening to the trial, and addition of some patient-facing questionnaires to ask about their experience of the trial (7 February 2020).

Support costs

NHS support and treatment costs were calculated in writing the protocol from health-care resource group (HRG) codes and the patient-level informatics system (PLIS) at the lead site. The PLIS allowed individual treatments, hospital bed occupancy, physiotherapy, outpatients costs and staff costs to be accurately calculated on a patient-level basis.

Routine NHS costs relating to the management of AVN typically consist of the cost of plain radiographs and a MRI scan (to make the diagnosis and stage the disease) and up to eight sessions of outpatient physiotherapy and three outpatient appointments with a clinician within the first year. Each session costs £50 plus £5 for equipment.

We estimated that NHS treatment costs for analgesia would be approximately £5–10 per week for 1 year in patients undergoing observation and investigations during the early stages of AVN (i.e. Ficat and Arlet stages 1 and 2).

NHS treatment costs for bisphosphonate therapy are not currently considered to be routine practice, so all drug costs were costed as a research cost and funded by the grant.

Patient and public involvement

Patient and public involvement was present at all stages of the development and execution of the trial, including during the feasibility and pilot stages. A Research Design Service-appointed PPI representative was involved in the initial trial design and during the development of the patient information sheets.

The Young Adult Hip Group (Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, Oxford, UK. URL: www.ndorms.ox.ac.uk/clinicaltrials/hipyoungadult) was established in conjunction with the local CRN. This was an interest group for patients with conditions such as femoroacetabular impingement (FAI), hip dysplasia, osteoarthritis and AVN, based at the lead site. The trial was developed in conjunction with this group. A member of the group (Peter Lovell) is a co-applicant. In conjunction with advice from INVOLVE (URL: https://involve.org.uk/resources/knowledge-base?gclid=CjwKCAjwu5yYBhAjEiwAKXk_eCs-6b8p7fHnnbTC04L8chUi43tz2KQLZLOBcrIsznSZHHIoub91HRoCLpYQAvD_BwE) and the James Lind Alliance (Southampton, UK), the group advised on research design, developing patient information resources and monitoring screening in outpatient clinics. The co-applicant PPI member sat on the TSC. The Young Adult Hip Group were planning to help disseminate the trial findings through local (hospitals and PPI groups), regional (Thames Valley CRN) and national [National Musculoskeletal Clinical Research Network (Clinical Research Network Coordinating Centre, University of Leeds, Leeds, UK), Arthritis Research UK (Chesterfield, UK) and the James-Lind Alliance] entities at the end of the trial.

During the internal pilot we planned to conduct a focus group discussion with the help of the local NIHR Research Design Service to test the patient acceptability with the use of electronic patient-reported outcomes, the collection of drugs and compliance with drug therapy, and with follow-up and drug delivery method (blister or bottle). As the trial closed prematurely, the focus group discussion did not take place.

Site initiation

At a minimum the trial manager (Gemma Greenall) attended all site initiation visits (SIVs). The CI (SGJ) also attended some of the SIVs. At the SIV a presentation was given that introduced the MANTIS trial – this included the team, the condition being studied and the trial premise and processes. In addition, the management of the MANTIS trial at sites including screening, recruitment, the taking of consent, case report form (CRF) completion, drug management and safety was presented, alongside trial responsibilities and the current status of the trial.

The principal investigator (PI) was always present at these visits and a register was taken of those who attended the visit.

To encourage surgical trainees to become involved in the trial, a MANTIS trial collaborator points scheme was introduced that resulted in certification and the potential to be cited in publications. An additional incentive was the MANTIS trial’s inclusion in the NIHR Associate Principal Investigator Scheme in March 2019, a Royal College of Surgeons-endorsed scheme that aims to engage, recognise and promote trainees involved in NIHR portfolio research, and also to integrate clinical research into routine clinical training.

Investigator site file

The investigator site files (ISFs) consisted of essential documents relating to trial conduct at a specific site (i.e. the location where participant-related trial activities were actually conducted), and were designed to enable both the conduct of the clinical trial and the quality of the data at that site to be evaluated. Among other essential documents, the ISF contained source documents, such as participant screening logs, consent forms, drug accountability records, delegation logs and CVs. Confirmation of receipt of the ISF was requested from sites and then filed centrally.

The ISF was then maintained by the PI at that site; the trial manager (Gemma Greenall) provided instructions on how to maintain the ISF, including the need for direct access to monitors, auditors and inspectors.

Site management

Inclusion criteria

Each participant in the trial had to meet both of the following criteria to be included in the trial:

• diagnosed with early symptomatic atraumatic AVN of the hip (Ficat and Arlet stage 1 or 2 using MRI)

• aged ≥ 18 years.

Exclusion criteria

Otherwise eligible individuals who met any of the following criteria were excluded from the trial:

• diagnosis (Ficat and Arlet stage 1 or 2) had not been confirmed using MRI within the last 12 months

• renal function (creatinine clearance) of < 30 ml/minute/1.73 m² (tested within the last 3 months)

• adjusted serum calcium levels outside local reference range (tested within the last 3 months)

• established osteoarthritis (Kellgren–Lawrence score of ≥ 2)

• previous AVN, femoral head deformity, prior hip surgery or hip fracture in the index hip

• current pathology (e.g. osteoporosis) requiring treatment with bisphosphonates

• received previous anti-osteoporosis therapy (excluding calcium or vitamin D supplements) that lasted > 4 weeks for oral treatment or any length of parenteral treatment

• contraindications to MRI

• contraindications to alendronate therapy (including hypocalcaemia) as listed in the SmPC

• planning a pregnancy in the next 24 months or currently pregnant or breastfeeding

• of childbearing age and not using appropriate contraception

• due to receive a joint-preserving surgical procedure of the hip

• unable or unwilling to provide informed consent

• unable to commit to follow-up regime

• already enrolled in an interventional clinical trial.

Screening and recruitment

Following attendance at a site initiation meeting, screening and recruitment were commenced at participating sites once the clinical trial site agreement had been signed and all necessary approvals were in place.

Potentially eligible patients were identified and approached by authorised members of staff about taking part in the trial. Patients interested in participating were referred to a research facilitator at recruiting sites, who then assessed whether or not they met the trial inclusion criteria. Information about the trial, which included its purpose, potential risks and benefits, the type of data, and when and how data would be collected during the trial, and who was funding and sponsoring it, was provided to the patient. This information was provided on a site-specific localised patient information sheet (PIS) that included the name and contact details of the local PI. All those approached to participate were able to keep the PIS.

Routinely MRI and radiography images and reports were then screened by the local care team to confirm the diagnosis and the Ficat and Arlet stage. For the diagnosis to be relevant to the trial, radiographs and MRIs had to have been taken no more than 12 months from the point of recruitment.

Before patients were officially recruited into the trial, they needed to have blood samples taken within the previous 3 months to confirm their renal function and adjusted serum calcium and 25(OH)-vitamin D levels. As such blood tests are routinely taken every 3 months in this patient population, all patients should have had these results available for consideration of eligibility.

Following consent, if pregnancy was suspected, participants of childbearing age were asked to take a pregnancy test to confirm their eligibility.

Screening log

To enable full and transparent reporting for the trial, brief details of all patients screened for the trial were recorded at each site. Sites also recorded the date patients were considered for the trial and, for those that were ineligible, the reasons for ineligibility. Those who were considered eligible but declined to participate had their reasons for declining recorded.

Informed consent

Participants had to personally sign and date the latest approved version of the ICF before any trial-specific procedures were performed.

Written and verbal versions of the PIS and ICF detailing the exact nature of the trial, what it involved for the participant, the implications and constraints of the protocol; the known side effects and any risks involved in taking part were presented to participants. Both documents clearly stated that the participant was free to withdraw from the trial at any time and for any reason without prejudice to future care, without affecting their legal rights and with no obligation to provide a reason for withdrawal.

The participant was allowed as much time as needed to consider the information and had the opportunity to question the investigator, their general practitioner (GP) or other independent parties when deciding whether or not they wished to participate in the trial. Written informed consent was obtained by means of a participant-dated signature and dated signature of the person who presented and obtained the informed consent. It was expected that some of the patient population may be difficult to follow up and may not have a fixed address. Therefore, the consent form also asked the participant to consent to their clinical care team contacting their GP for information, which may have helped the clinical care team query missing data. The participant was also made aware of the fact that screening tests and previous tests and scans may need to be reviewed to confirm eligibility, and that images and reports of scans may need to be accessed from their routine appointments by the clinical care team to aid follow-up.

The person who obtained the consent had to be suitably qualified and experienced, and have been authorised to do so by the CI or relevant PI. A copy of the signed ICF was given to the participant, and the original signed ICF was retained at the trial site.

Randomisation and allocation procedure

Randomisation to the interventions was undertaken via the centralised secure web-based randomisation service Registration/Randomisation And Management of Product [RRAMP (OCTRU, Oxford, UK)], run through OCTRU. In the event that sites were unable to randomise patients using RRAMP, they had to contact the central research office, and a member of the trial team was then able to randomise the patient via RRAMP. An emergency backup randomisation system was in place if the central research office was also unable to use RRAMP.

Participants were randomised on a 1 : 1 basis to receive either alendronate or the placebo-matched control. Randomisation was performed using a minimisation algorithm including a random element (p = 0.8) to ensure balanced allocation of participants across the two arms stratified by:

• randomising site

• Ficat and Arlet stage (1 or 2)

• main AVN risk factors (steroid/alcohol/other) obtained from clinical notes

• bilateral vs. unilateral AVN.

However, the first 28 participants (10% of the expected sample size) were to be randomised using simple randomisation to seed the minimisation algorithm. Owing to the small numbers recruited, the trial did not move out of this seeding stage and so the stratification factors were not taken into account.

The intervention was expected to start within 4 weeks of randomisation.

Treatment groups

A total of 280 participants with AVN of the hip as per the trial inclusion and exclusion criteria were to be randomised to one of two arms to receive one of the following:

1. alendronate

2. placebo-matched control.

Dosing regimen

Both the active substance and the placebo control were manufactured and Qualified Person released for use in the trial by the holder of the Manufacturing and Import Authorisation (IMP) licence. Thereafter, bottles containing 13 weeks' supply of both the active substance and the control were sent to recruiting sites. All IMPs were overencapsulated and the bottles labelled with a MHRA-approved clinical trial label in accordance with Annexe 13 (Investigational Medicinal Products) of the European Union Guidelines on Good Manufacturing Practice. 24 Both the active substance and the placebo control were administered alongside standard care for patients with AVN.

Drug administration

The active substance and the placebo control were both oral tablets to be taken once per week for 52 weeks. Patients received their supply of tablets in bottle form and received a patient card and automated weekly text reminders to take their tablet. The dispensing pharmacy was blinded to the treatment allocation. The tablets were provided to the patient in bottles containing 13 weeks’ worth of treatment, which they needed to collect from the pharmacy within 14 days of randomisation. The patients received the next 13 weeks’ supply at 3-monthly intervals.

To ensure sufficient absorption, the SmPC suggested that alendronate be taken at least 30 minutes before the first food, beverage or medicinal product of the day with plain water only. Therefore, we advised participants that it was preferable to take the tablet first thing in the morning, and take it whole without crushing or chewing it, as doing so could potentially cause oropharyngeal ulcers. Once the alendronate had been taken, patients were advised not to lie down or have the first food of the day for at least 30 minutes.

Data collection

Embedded pilot

An internal pilot was planned that would progress to the definitive trial if predefined evaluation criteria regarding recruitment, patient characteristics, equipoise and compliance were met. The pilot trial was to mirror the procedures and logistics to be undertaken in the main definitive trial. We planned for the data that were collected in the pilot trial to contribute to the final analysis.

The pilot aimed to randomise 50 patients or more over a 12-month period from at least 10 sites, with a target recruitment rate of one patient per site per month. If any issues were identified, it was proposed that they be discussed with the TSC; any changes required for the definitive trial were to be discussed with the funder and submitted as amendments for approval.

Outcome measures

Safety monitoring

Alendronate is a licensed medication, with a comprehensive safety profile. Only the adverse events (AEs) detailed in Table 2 were to be recorded on the trial AE form.

The listed AEs are known to be expected and related to the trial drug and are detailed in the SmPC. The recording of the listed AEs was to help us assess reasons for non-compliance or withdrawals in this patient population. All AEs were to be recorded on the AE form as soon as the site or central trial office team became aware of them.

The following information was to be recorded on the AE form: description, date of onset, end date and action taken. Follow-up information was to be provided as necessary.

The investigator was to use their clinical judgement to decide whether or not an AE was of sufficient severity to necessitate the participant’s removal from treatment. A participant was able to voluntarily withdraw from treatment because of what he or she perceived as an intolerable AE. In either of these scenarios, the participant was to undergo an end-of-trial assessment and receive appropriate care under medical supervision, until either the symptoms ceased or their condition became stable.

Trial oversight

A DSMC (comprising two medically qualified clinicians and a statistician, all independent) was appointed to safeguard the interests of the trial participants, assess the safety and efficacy of the interventions during the trial, and monitor the overall conduct of the trial, protecting its validity and credibility. The DSMC was independent of the trial investigators and sponsor and adopted a DAta MOnitoring Committees: Lessons, Ethics, Statistics (DAMOCLES)-based charter30 that defined its terms of reference and operation in relation to the oversight of the trial. We planned for this group to meet at least every 12 months over the duration of the trial.

The DSMC also monitored and reviewed accruing data from the internal pilot trial to determine if they would recommend to the TSC that the trial should continue and, if so, whether or not any amendments to the sample size or trial design were required. The DSMC fulfilled this role and, alongside the TSC, concluded that the trial should be closed prematurely.

Throughout the definitive trial, the DSMC would have continued to review accruing data and summaries of those data presented by the alendronate therapy group, and to assess the screening algorithm against the eligibility criteria. It would also have considered emerging evidence from other related trials or research and reviewed any related serious adverse events (SAEs) that were reported.

In addition to a DSMC, a TSC, whose primary function was to act as an oversight body for the trial on behalf of the sponsor and funding body, was appointed. The TSC was chaired by an independent member and considered and acted, as appropriate, upon the recommendations of the DSMC. The TSC also adopted a charter that defined its terms of reference and operation in relation to the oversight of the trial and agreed to meet at least every 12 months over the duration of the trial.

Data management

A Data Management and Sharing Plan was produced for the trial and included reference to confidentiality, access and security arrangements. All data received were processed in accordance with data protection rules. The trial was set up for direct access to be granted to authorised representatives from the sponsor, host institution and the regulatory authorities to permit trial-related monitoring, audits and inspections; however, this was not invoked during the pilot phase of the trial.

All trial data were collected on trial-specific documents, such as questionnaires and CRFs. All trial-specific documents, except for the signed consent form and follow-up contact details, referred to the participant by a unique trial participant number/code and not by name. Participant-identifiable data were stored separately from trial data and in accordance with OCTRU standard operating procedures. All trial data were stored securely in offices accessible only by swipe card by the central co-ordinating team staff in Oxford and authorised personnel.

All data were entered into the trial’s instance of OpenClinica (OpenClinica, LLC, Waltham, MA, USA), which was set up specifically for the trial.

Sample size

Statistical analyses

Owing to the early halt in trial recruitment, only 21 participants were randomised. Therefore, no formal statistical analyses were planned, as the sample size was too small to provide adequate power for hypothesis testing.

The primary statistical analysis was carried out on the basis of intention to treat (ITT), with all participants being analysed according to their allocated treatment group, irrespective of which treatment they actually received. The analysis was descriptive in nature, and the principal analysis compared results between the intervention and control arms. Numbers and percentages were provided by treatment arm for categorical variables, and numbers, medians and interquartile ranges (IQRs) were provided by treatment arm for continuous variables. For clinical outcomes, means and SDs, and medians and IQRs were provided for both treatment arms.

Issues identified for the MANTIS trial

Delayed set-up

The set-up of the trial was subject to significant delays, which in turn delayed the start of recruitment. These delays have been summarised in Figures 2 and 3.

Overall, set-up resulted in a 14-month delay, with recruitment to the pilot beginning in month 17 rather than month 3. Consequently, the recruitment period was reduced from 36 months to 22 months and the recruitment predictions were remodelled based on estimates from sites. A major contributor to the delay in set-up was our sponsor office’s, which was delayed owing to high local demand and staff shortage; therefore, although it was thought that this review would require only 1 month for completion, it took over 4 months (from 29 November 2018–11 April 2018). In addition, the original submission to the MHRA was not accepted, owing to its operating under the incorrect IMP Product Licence and insufficient details being provided regarding the testing of the over-encapsulated drug product.

FIGURE 2.

Delays experienced with sponsor approval for the MANTIS trial.

We then encountered delays during the manufacture of the trial IMP; these are specified and detailed in Figure 3.

FIGURE 3.

Delays attributable to the IMP supplier. QP, qualified person.

Mitigation of issues identified

The trial team mitigated the issues identified in the following ways:

• Exploring the use of primary care – ‘GP champions’ in two regions across the UK ran searches for patients with AVN to explore whether these patients may be identified in primary care. Unfortunately, both GP champions confirmed that musculoskeletal conditions are poorly coded in primary care, and both regions identified only two potential patients over a 3-year period. It was agreed with the GPs that this would not be a cost-effective way to improve recruitment.

• Engaging the surgical trainee networks – trainees were engaged through the NIHR Associate PI Scheme and via trainee research collaboratives across the UK. This helped to raise awareness of the trial, encouraged trainees to get involved in research, and resulted in an increased number of patients being screened at sites where trainees were involved.

• Attending relevant conferences – this again helped us to raise awareness of the trial and engage with potentially interested clinicians. The trial team attended the British Hip Society annual congress in 2018 and 2019, and the British Orthopaedic Association annual congress in 2019, where we engaged with approximately 50 clinicians, some of whom went on to join the trial as members of a recruiting site.

• Circulating potential patient pathways – as more potential pathways were identified, this information was circulated to existing sites as well as being fed back to potential sites early on in feasibility discussions. Encouraging the engagement between specialties enabled sites to review their recruitment predictions more thoroughly and addressed some capacity issues, as some sites were able to engage more than just the orthopaedic teams.

Further mitigation of issues identified

The trial team researched some additional avenues to be explored at the review with the funder, which took place when recruitment was behind target. These included the following:

• Advertising the trial through primary care – this would have been in addition to existing posters used in clinic rooms and waiting rooms at recruiting sites. It was thought that this method may have increased recruitment, as patients with early-stage AVN may be unaware of their condition and will sometimes present to the GP only with generalised hip pain. If more patients had been aware of the trial, they may have asked their GP if they could self-refer to a recruiting site for more information.

• Re-engaging with sites that had been contacted previously – capacity issues may have been resolved since contact was last made and sites may have been more willing to take part in the MANTIS trial owing to the connections that were available through the trainee networks and a better understanding of the patient pathway.

• Reviewing the per patient payment fee – sites often queried whether or not more money was available to them (to allow them, for example, to cover archiving fees and pay for the time required by research nurses at visits). Although we believed the offered per-patient fee of £100 sufficiently covered the cost to sites (especially when taking into account that the IMP was provided to sites free of charge), the addition of a ‘screening’ fee may have encouraged more sites to take part in the trial. This practice is commonly seen in commercial trials, which often then provide funding for rare disease trials. The trial team could have explored ways that the ‘screening’ fee could be incorporated into the MANTIS trial budget: for example, the extra cost could have been offset by savings made on imaging costs.

However, owing to the early closure of the trial, the trial team did not have the chance to pursue these issues further.

Explanatory statement

Following the HTA meeting held on 21 November 2019, at which the difficulties in recruiting patients were discussed, it was decided that the MANTIS trial should be closed with immediate effect. Participants were unblinded by the trial team and those in the intervention arm were contacted, and the question of whether they wished to continue to receive alendronate was discussed.

A brief final report is presented in this chapter. No formal statistical tests were performed owing to the small number of participants recruited into the trial.

Sites open to recruitment

The trial aimed to open 20 sites. By December 2019, when the trial was shut prematurely, only eight sites had opened to recruitment.

Figure 4 shows the target number of sites compared with the actual number of sites open to recruitment month by month.

FIGURE 4.

Month-by-month target number vs. actual number of sites open to recruitment for the MANTIS trial.

Of the eight open sites, only six recruited at least one participant before the trial was closed to recruitment.

Study participants

The flow of participants through the trial from screening through randomisation to follow-up is presented in a CONSORT (Consolidated Standards of Reporting Trials) flow diagram (Figure 5). Most randomised participants had not yet reached the primary outcome time point for conversion to hip replacement when the trial was stopped.

FIGURE 5.

Participant flow diagram. a, Patients who received at least one course of treatment.

Recruitment breakdown

Screening and randomisation data, including conversion rates, are summarised by recruitment site in Table 3. Out of 108 patients screened for the MANTIS trial, 48 patients (44%) were deemed ineligible for reasons detailed in Table 4. The most common reason was advanced disease, defined as either a Ficat and Arlet score of > 2 or a Kellgren–Lawrence score of ≥ 2 (indicating osteoarthritis).

Out of 108 patients screened for the MANTIS trial, 17 declined to take part (17.6%) for reasons detailed in Table 5. Table 6 shows a breakdown of patient recruitment by site, including the numbers recruited at each site for the months in which the trial was open. Table 7 provides more information about the average number of recruits per site per month.

Patient demographics

Tables 8 and 9 show information on the baseline characteristic comparability by allocated treatment arm.

Treatment compliance

Of the 21 participants randomised, eight (38.1%) completed the full course of treatment as shown in Table 10; four were allocated to the placebo treatment arm and four were allocated to the alendronate treatment arm. Tables 11 and 12 show the information from the self-completed compliance CRF for each treatment arm, where compliance with taking the allocated treatment once per week was high (> 80%), as measured from those responding to the CRF.

Follow-up of patients

The number of follow-up questionnaires containing patient-reported treatment compliance and outcome measures, both expected and received, at 1 and 12 months post randomisation is summarised in Table 13. The compliance rate has been calculated as a percentage of those received out of the total expected. Please see Appendices 1 and 2 for more details on treatment allocation and individual participant details.

Owing to the early closure of the trial, no participants reached the 24 or 36 months post-randomisation time point.

Withdrawals and protocol deviations

Out of the 21 patients randomised, two withdrew from the trial. The details of both withdrawals are summarised in Table 14. In addition, Table 15 shows information about the two protocol deviations that occured during the trial and the actions taken to address these.

Clinical outcome results

Safety reporting

Details of AEs and SAEs are summarised in Table 22. Among 21 patients, two SAEs were recorded, both of which were considered to be unrelated to the treatment.

Limitations

The small recruitment figures, and the fact that the pilot phase closed prior to the primary outcome being recorded for most participants, makes it very difficult to interpret findings.

Although it is impossible to draw meaningful conclusions from the trial, several observations can be made.

First, we found that the later stages of AVN were those predominantly identified, meaning that many patients classed as Ficat and Arlet stage 3 or above were excluded from the trial. In addition, two-thirds of those included in the trial were classed as having Ficat and Arlet stage 2. This mirrors clinical observations of the condition, which is usually identified at a late stage, when joint-preserving treatments are of no value. These observations are supported by the Non-Arthroplasty Hip Register, which records very few joint-preserving surgical procedures (< 80) per year in early-stage patients. This contrasts starkly with the National Arthroplasty Register, which identifies several hundred joint replacements for late-stage AVN every year. The main barriers to targeting early disease is the fact that symptoms in the early stages of the condition are non-existent or mild. Consequently, even clinicians with a high awareness of the risks may miss early disease, and the few identified cases of early-stage AVN (Ficat and Arlet stage 0 or 1) are often diagnosed during routine screening of the contralateral hip.

Second, we observed that patient-reported outcomes worsen quickly over time in all patients. This further supports the clinical observation that only a small window of opportunity for intervention with joint-preserving treatments in early-stage AVN patients exists, beyond which joint replacement becomes inevitable. Ultimately, this may be the rate-limiting step to treatment.

Last, it seems that the intervention, alendronate, was well tolerated by most participants. The trial team deliberated between different formulations of bisphosphonate while designing the study. Concern was raised regarding the side effects and frequency of dosing in alendronate, but this does not seem to have adversely affected compliance, which was over 80%. However, given the longer rate of onset for oral medication, it remains to be tested whether or not a parenteral therapy would be more effective.

Future work

By the end of the pilot phase, this trial was recruiting small but steady numbers in most sites. Recruitment projections performed at this stage demonstrated that it would take over 10 years to reach the full sample size. In its current form the trial is achievable; however, the cost may be prohibitive. It is likely that many more sites would be required to recruit adequate numbers over a conventional NIHR funding cycle. Our pilot study therefore demonstrates that such a trial is unachievable without a better understanding of the referral pathways and geographic foci of the disease. Such a trial may be feasible in a health-care environment where AVN is more prevalent (e.g. South East Asia).

Early AVN of the hip should be treated as a rare disease in the UK. Although uncommon, the earlier stages of the condition are extremely hard to identify, further reducing its reported prevalence. This makes trials difficult and expensive to conduct, but more importantly, provides a very small window for treatment even if the condition is detected in time. An improved understanding of the condition’s natural history is required, along with an understanding of the geographic locations where the condition is located. During the pilot study, we found it very difficult to identify areas of high disease prevalence. Despite identifying target demographics, we were unable to identify sites with large patient populations who met these criteria. Clinical Specialty Groups and specialist societies engaged extensively in this process; however, they were often unable to provide any useful advice.

This demonstrates the need for a registry or rare disease database for osteonecrosis of the hip. Several databases of rare diseases exist and have been shown to be effective. Many are patient-reported, such as the Rudy Study,34 which provide a repository for patients with rare, long-term conditions. Such a database for osteonecrosis of the hip would help define the population and help clinicians better understand the natural history.

Overall conclusion

The MANTIS trial was terminated at the end of the pilot phase, as it did not meet its go/no-go criteria. The main issues were a low recruitment rate, owing to lower than expected disease prevalence, and difficulties in identifying the condition at an early enough stage.

We would not recommend that a short-term interventional study is conducted on this condition until its prevalence, geographic foci and natural history are better understood. This is likely to be a barrier in most health-care markets.

One means of developing this understanding would be the introduction of a database/registry for AVN of the hip.

Trial Steering Committee

Roger Francis (chairperson) (Newcastle University), Sally Hope (Oxford University Hospitals NHS Foundation Trust), Nick Harvey (University of Southampton), Mark Edwards (University of Southampton), Keith Tucker (Chair of Orthopaedic Data Evaluation Panel and Beyond Compliance), Caroline Fairhurst (University of York) and Michael Sheperia (patient representative).

Data Monitoring Committee

Bo Abrahamsen (chairperson) (Holbæk Hospital), John Belcher (Keele University) and Nicola Peel (Sheffield Teaching Hospitals NHS Foundation Trust).

Contributions of authors

Sion Glyn-Jones (https://orcid.org/0000-0002-9130-3167) (Professor of Orthopaedic Surgery and Honorary Consultant Orthopaedic Surgeon) contributed to the design of the trial, the analysis and interpretation of the data, and drafted and critically reviewed the manuscript.

Muhammad K Javaid (https://orcid.org/0000-0001-7985-0048) University Lecturer in Metabolic Bone Disease and Consultant Rheumatologist) contributed to the design of the trial, the analysis and interpretation of the data, and drafted and critically reviewed the manuscript.

David Beard (https://orcid.org/0000-0001-7884-6389) (Director, SITU) contributed to the design of the trial and protocol development, and critically reviewed the manuscript.

Julia Newton (https://orcid.org/0000-0001-7930-299X) (Consultant Rheumatologist) contributed to the design of the trial and protocol development, and critically reviewed the manuscript.

Robert Kerslake (https://orcid.org/0000-0002-3694-2345) (Consultant Musculoskeletal Radiologist) contributed to the design of the trial and protocol development, and critically reviewed the manuscript.

Callum McBryde (https://orcid.org/0000-0002-2500-5313) (Consultant Orthopaedic Surgeon) contributed to the design of the trial and protocol development, and critically reviewed the manuscript.

Tim Board (https://orcid.org/0000-0002-3295-2087) (Consultant Orthopaedic Surgeon) contributed to the design of the trial and protocol development, and critically reviewed the manuscript.

Susan J Dutton (https://orcid.org/0000-0003-4573-5257) (Lead Statistician, OCTRU) contributed to the design of the trial and protocol development, and critically reviewed the manuscript.

Melina Dritsaki (https://orcid.org/0000-0002-1673-3036) (OCTRU Senior Health Economist) contributed to the design of the trial and protocol development, and critically reviewed the manuscript.

Vikas Khanduja (https://orcid.org/0000-0001-9454-3978) (Consultant Orthopaedic Surgeon) contributed to the design of the trial and protocol development, and critically reviewed the manuscript.

Magbor Akanni (https://orcid.org/0000-0003-1864-1319) (Consultant Haemotologist) contributed to and critically reviewed the manuscript.

Shaun Sexton (https://orcid.org/0000-0002-1954-0950) (Consultant Orthopaedic Surgeon) contributed to and critically reviewed the manuscript.

John Skinner (https://orcid.org/0000-0002-1901-9057) (Consultant Orthopaedic Surgeon) contributed to and critically reviewed the manuscript.

Nicholas Peckham (https://orcid.org/0000-0003-1066-0726) (Medical Statistician) contributed to the analysis and interpretation of the data, and drafted and critically reviewed the manuscript.

Ruth Knight (https://orcid.org/0000-0001-6810-2845) (Senior Medical Statistician) contributed to the analysis and interpretation of the data, and drafted and critically reviewed the manuscript.

Ines Rombach (https://orcid.org/0000-0003-3464-3867) (Senior Medical Statistician) contributed to the protocol development and critically reviewed the manuscript.

Loretta Davies (https://orcid.org/0000-0002-4721-356X) (Senior Trial Manager, SITU) drafted and critically reviewed the manuscript.

Vicki Barber (https://orcid.org/0000-0001-9631-3666) (Operations Manager, OCTRU) contributed to the design of the trial, and drafted and critically reviewed the manuscript.

Data-sharing statement

All data requests should be submitted to the corresponding author for consideration. Access to available 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 and Care 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, 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, the HTA programme or the Department of Health and Social Care.