ROADMAP Clinical Trial

Over 120,000 hip or knee replacements are performed in Australia every year, and there are ~1 million Australians living with one or more joint replacements. Approximately 3,900 prosthetic joint infections (PJI) are diagnosed in Australia every year. The burden is similar in other resource-rich countries. For example, in Canada, over 130,000 hip and knee replacements are performed yearly and 2% of primary arthroplasties require a revision within 2 years. Infection has been identified as the most common cause for a revision surgery, which translates to $42.1 million CAD in inpatient costs to the Canadian health care system.

PJIs are difficult to treat, because of the presence of a foreign body, which encourages biofilm formation, and the need for combined surgical and prolonged antibiotic management. Outcomes are generally poor, with substantial variation in reported treatment success (ranging from 10 to 90%, median 50% after 2 years) and PJIs have striking health care costs. Existing treatment guidelines are informed by low quality evidence, with fewer than 1,700 participants ever enrolled into published randomised controlled trials comparing management approaches. Defining the optimal surgical and antibiotic approaches to PJI is a critical unmet need.

The Investigators are using an adaptive platform trial to allow the investigators to simultaneously address these questions in the management of PJI. The trial will include 3 silos (Early, Late Acute and Chronic PJI) and evaluate treatments and outcome of treatments over a 12 month period. Time points for data collection and analysis will be Day 100 and Day 365 post platform entry.

The Investigators plan to test interventions within 3 initial domains of:

• Surgical Domain comparing outcomes of surgical strategies - Debridement and Implant Retention (DAIR) vs Revision arthroplasty.
• Antibiotic Choice Domain comparing outcomes of antibiotic treatment - Backbone antibiotic regimen plus the addition of oral rifampicin (AKA rifampin) vs Backbone antibiotic regimen without rifampicin.
• Antibiotic Duration Domains comparing the duration of antibiotics treatment in relation to one stage and also two stage surgical Revision arthroplasty strategies

• Part A - PJI treatment with a single-stage revision - 6 weeks (+/-7 days) of antibiotics following the single stage revision procedure ('short course regimen') vs 12 weeks (+/-7 days) of antibiotics following the single stage revision procedure ('long course regimen')
• Part B - PJI treatment with a two-stage revision and negative cultures following reimplantation (second stage) - Extended antibiotic prophylaxis: 12 weeks vs No extended prophylaxis.

The ROADMAP trial infrastructure will also have the potential to add further domains to the platform.

RANDOMISATION Participants will be randomly allocated to one arm within each domain for which they are eligible, which their site is participating in, and for which they have consented, using a web-based module available 24h a day 7 days a week. Randomisation within the patient's known silo, for all available domains will occur immediately following provision of consent (which is considered the time of platform entry), however, the reveal of each treatment allocation(s) will be delayed subject to confirmation of domain eligibility.

BLINDING This will be an open-label study unless otherwise specified in a domain-specific appendix. For the overall data and results, only specified members of the statistical analytical team and Data and Safety Monitoring Committee (DSMC) will have access to unblinded results, with other trial investigators and staff remaining blinded to the aggregate results until completion of final analysis for a domain or cell.

ANALYSIS The ROADMAP trial will repeatedly fit Bayesian hierarchical logistic models to the accumulating data (interim analyses), over the life of the trial, to estimate model parameters and evaluate pre-specified decision criteria either within a cell or a whole domain.

CORE PRIMARY OUTCOME MEASURE Treatment success assessed at 12 months post-platform entry, defined as all of: i) Alive; ii) Clinical cure (no clinical or microbiological evidence of infection); iii) No ongoing use of antibiotics for the index joint; and iv) "Destination prosthesis" (the prosthesis present after the initial management strategy is complete) still in place

CORE SECONDARY OUTCOME MEASURES

A core set of secondary endpoints will be evaluated across all domains:

• A "desirability of outcome ranking" (DOOR - an ordinal outcome which ranks participants based on a combination of clinical cure, patient-reported joint function and quality of life).
• Patient-reported joint function (Oxford hip or knee score) at 12 months after platform entry.
• Patient-reported quality of life (EQ5D5L) at 12 months after platform entry
• Cost effectiveness
• All-cause mortality at 12 months after platform entry
• "Clinical cure" at 12 months after platform entry
• Proportion alive and no longer taking any antibiotics for the index joint at 12 months after platform entry.
• Proportion in whom the destination prosthesis is still in place at 12 months after platform entry.
• Microbiological relapse (Positive joint fluid or tissue culture for one or more of the index isolates between Day 100 and 12 months after platform entry)
• Microbiological reinfection (Positive joint fluid or tissue culture with a different organism to all of the index isolates between Day 100 and 12 months after platform entry).
• Time alive and free from any revision procedure on the index joint captured by a national joint replacement registry within 24 months of platform entry

ROADMAP Registry ROADMAP registry is a planned adjunct to the ROADMAP main platform study. The ROADMAP registry aims to describe the entire group of people who get large joint prosthetic joint infections, many of whom have never been a part of other previous research studies. The ROADMAP registry is interested in the different factors that can be examine to identify patterns in the medical history of people that get prosthetic joint infections, what sorts of testing and treatments they receive, and how these might affect their outcome. The registry information when analysed will be used to improve knowledge about prosthetic joint infections and also patient care.

The registry will include participants of the main platform study and also seek independent consent from patients with prosthetic joint infection that choose not to participate in the main platform study. Data for the registry will be collected from participants medical records with participant provided permissions. In addition, all ROADMAP registry participants will be asked to provide specific consent for data linkage that may occur with existing datasets, such as hospital records, national joint replacement registries, and registries of birth, deaths and marriages.

Apart from collecting available information from medical records, there are no additional tests or treatments involved.

Inclusion criteria

• "Confirmed" or "Likely" Prosthetic joint infection of a large joint according to European Bone and Joint Infection Society (EBJIS) criteria (2021)
• Physically present at participating hospital at time of eligibility assessment
• "Current" prosthetic joint infection, meaning symptoms and/or signs of the PJI are present at the time of eligibility assessment.

Exclusion criteria

Potentially eligible participants meeting any of the following criteria at the time of eligibility assessment for platform entry will be excluded from the randomised platform (but may still participate in the registry):

• The index prosthetic joint is a shoulder, elbow, wrist or ankle.
• Known previous participation in the randomised ROADMAP randomised platform for the index joint.
• Known previous participation in the randomised ROADMAP platform for a joint other than the index joint within the 12 months prior to eligibility assessment
• Treating clinician believes that death is imminent and inevitable.
• Treatment is not with curative intent.
• Patient is not classifiable into one of the three defined silos.
• Patient is unlikely to be accessible for follow up over the 12 months following platform entry.
• Treating team deems enrolment in the study is not in the best interests of the patient.

To be included in any of the following Domains the participant must be eligible for the PLATFORM ( as listed above) and none of the platform-level exclusion criteria AND all of the domain-level inclusion and none of the domain-level exclusion criteria.

Domain-specific eligibility criteria Each domain may have additional criteria for eligibility. Participants who fulfil the above criteria will be assessed for enrolment into all domains active at a participating site. At least 2 interventions (which may include standard of care) within a domain must be available to an eligible participant in order for that participant to enter the domain. The minimum number of interventions within a domain is two. The availability of interventions within domains will be region- and site-specific, although the default position is that all interventions within a domain will be available at all sites.

Surgical Strategy Domain: Late Acute Silo inclusion criteria

Patient is in the Late Acute silo, meaning all of the following 3 criteria are met:

• Onset of symptoms is >30 days after implantation of the index joint
• Treating team feel that either DAIR or revision is appropriate for this patient.
• The infected prosthesis is a primary arthroplasty. Surgical Strategy Domain: Late Acute Silo exclusion criteria

<!-- -->

• Any previous episode of native joint septic arthritis or PJI in the index joint
• A definitive DAIR has already occurred for this episode of infection.
• A revision has already occurred for this episode of infection
• Loosening or instability of any component of the prosthesis
• Predicted inadequate soft tissue coverage for wound closure
• One or more of the causative organisms is a fungus or a "difficult to treat" bacterium.
• Treating team deems enrolment in this domain is not in the best interests of the patient.

Antibiotic Choice Domain inclusion criteria

• The time of eligibility assessment is between 48h and 7 days following initial surgical treatment.
• One or more of the causative organisms is a Gram-positive of interest OR the infection is culture negative.

Antibiotic Choice Domain exclusion criteria

<!-- -->

• Previous hypersensitivity reaction to rifampicin
• The patient is receiving a concomitant medication with an expected clinically significant drug interaction with rifampicin, which cannot be ceased, substituted for an alternative agent, or mitigated by dose adjustment.
• One or more causative Gram-positive organisms of interest have in-vitro evidence of resistance to rifampicin
• Known pregnancy or breastfeeding
• Treating team deems enrolment in this domain is not in the best interests of the patient.

Antibiotic Duration Part A Domain inclusion criteria

• A single stage revision procedure has been performed

For the purposes of this domain, the definition of a single stage revision procedure is as follows:

A one-stage revision means:

• Arthrotomy

• Removal of all prosthetic components

• Comprehensive synovectomy

• Placement of "definitive" new components
• could be primary or revision components • With no plans for a second stage operation

Additional recommended elements (not part of definition, but considered 'ideal' elements of this procedure):

• Removal of all loose cement; well-fixed cement may be retained.
• Double set up (using a separate set of sterile instruments and drapes for the reimplantation)

Antibiotic Duration Part A Domain exclusion criteria

Patients will be excluded from this domain if, at the time of eligibility assessment, they have any of the following:

<!-- -->

• Any causative organism is a fungus or a 'difficult-to-treat' bacterium at the time of platform entry
• Treating team deems enrolment in this domain is not in the best interests of the patient
• The single stage revision was >7 days ago
• Patient no longer willing to participate in the domain

Antibiotic Duration Part B Domain Eligibility assessment and allocation reveal will be performed at day 7 (+/- 3 days) from the time of the reimplantation operation - i.e. between 4 and 10 days following the reimplantation operation. The purpose of this time frame is to confirmation of negative reimplantation cultures.

Antibiotic Duration Part B Domain inclusion criteria

• A two-stage revision procedure has been performed

For the purposes of this domain, the definition of two-stage revision and reimplantation procedures are defined as follows:

A two-stage revision means:

• Removal of all existing prosthetic components (including metal, ceramic, polyethylene) and any loose cement
• Placement of a temporary prosthesis or spacer with the intention to replace this at a later time point with a permanent prosthesis
• The intention to subsequently replace of the temporary spacer with a definitive prosthesis at a reimplantation operation in the future

Additional recommended elements (not part of definition, but encouraged):

• Use of antibiotic-loaded cement as the spacer
• Use of an articulating spacer which is capable of withstanding weight-bearing

A reimplantation procedure as the second stage of a two-stage revision means:

• Removal of the temporary prosthesis and all associated cement
• Implantation of a new prosthesis; considered the 'destination prosthesis'
• The infected prosthesis was a primary (not revision) arthroplasty
• At least 6 weeks of antibiotics have been received following the first stage revision, intended as treatment of the index infection
• The reimplantation operation was done 4-10 days ago

Antibiotic Duration Part B Domain exclusion criteria

• >6 months has passed since platform entry
• One or more of the causative organisms is a fungus or a 'difficult-to-treat' bacterium
• No perioperative tissue or fluid cultures were collected at the reimplantation operation
• Perioperative cultures collected at reimplantation grow a significant organism (s)
• Long-term antibiotic suppressive therapy is planned following the reimplantation operation
• Treating team deems enrolment in this domain is not in the best interests of the patient
• Patient no longer willing to participate in the domain