Preparing and drafting a clinical investigation plan
A clinical investigation plan for a medical device is the document an investigation is approved and conducted on. Its content is set by ISO 14155, in an order the writer may choose. It rests on the clinical evaluation and risk documents and on inputs gathered before drafting, and it is written with contributors of several professions, starting from an agreed synopsis and checked for consistency.
What the plan is and who signs it
A clinical investigation plan is also called a protocol, a clinical study protocol, a clinical protocol or a clinical study plan. In medical devices, one document often goes by several names. Under any of these names, the document states the rationale, objectives, design and proposed analysis, methodology, monitoring, conduct and record keeping of the clinical investigation. A plan for a medical device must outline all objectives of the investigation and justify them on scientific and ethical principles, present and justify the investigational design and methodology, and state the principal features of the statistical analysis. It must also include practical aspects such as the organisation, conduct, monitoring and record keeping of the investigation.
The plan is the key document of the clinical investigation, and the basis of the application for approval sent to the ethics committee and any competent authority. The coordinating investigator and the sponsor must sign off its content before the application is submitted. The principal investigators of all participating sites must agree to conduct the investigation according to it, typically by signing the final plan, the version the ethics committee and competent authority approved. Any change to the plan after its approval must be described in an amendment, which must also be approved if it is considered substantial.
Compliance and required content
The first step in writing a plan is to make sure it complies with the applicable regulations, standards and guidelines. One guide for writers new to device documents holds that, since writing them is not very different from writing the corresponding pharmaceutical documents, the most important things are to follow the applicable guidelines, use the correct terminology, and be aware of certain differences from pharmaceutical documents, such as in safety reporting and its timelines.
ISO 14155 lists the minimum content requirements for a clinical investigation plan. A plan needs all that content, but it need not copy the standard's structure. There is no regulatory requirement to present the content in a particular order, as long as all the required information is provided. One experienced writer finds the structure of ISO 14155 a little odd in some parts and does not write protocols according to it.
ISO 14155 can be used as a content checklist, and the writer should know the guidelines inside out. A writer who does may even improve the company's protocol by finding missing items, and that is already a good start, provided the deficiencies are pointed out humbly.
Although all the information the applicable regulations and guidelines require should be present in a plan, some of it need not always be described in full, such as data management, statistics and monitoring. One option is to describe it briefly in the plan and refer to a separate document for the details. This may save time and reduce costs, because the separate documents can be updated without affecting the plan, which means fewer amendments and approval rounds. For less complex investigations, though, it can be easier to keep everything in the plan.
The documents it is built on
A writer starting a plan should first read the clinical evaluation plan, including the clinical development plan, the clinical evaluation report, the risk management report and, if available, the post-market clinical follow-up plan. If these have not been performed or updated recently, the writer should stop and take a step back, because they are the essential building blocks on which planning an investigation rests. A well-performed clinical evaluation identifies the need for a clinical investigation and the appropriate endpoints, acceptance criteria and investigational design, and so lays the basis for planning it.
The risks of the investigational device and of any related clinical procedure should also be estimated when the investigation is planned, in accordance with ISO 14971. The residual risk from an initial risk analysis, and the risks to the subject of the clinical procedure or any required follow-up procedure, must be balanced against the anticipated benefits.
Inputs to gather before drafting
The study design mostly comes directly from the company, but a writer experienced in the disease area can provide input and recommendations. Information on the study device can come from previous protocols with the device, from the clinical investigator's brochure or from the engineers. The plan should also summarise any previous experience with the device.
The statistical section usually comes from the statistician, but the writer often needs to help the statistician find the right numbers for the sample size calculation. The writer may also want to advise on analysis groups and subgroups, depending on their experience, though this is not necessarily expected. Data management needs and their input also need careful consideration, even though they matter most when the case report forms are created.
Published studies and protocols to learn from
The writer should be familiar with the key publications and presentations in the area. These feed the introduction, let the writer learn from mistakes others have made, and may give input for selecting the inclusion and exclusion criteria, the endpoints and the definitions.
Journals that publish the protocol together with the paper, or require it as supplemental material, such as the New England Journal of Medicine, show what a plan can look like, and a writer who is new can search them for similar studies in the indication area to get input. In PubMed, for instance, a search for transcatheter and the New England Journal of Medicine returned 24 hits, and the supplementary material of a hit holds its protocol. EMWA's webinar archive also shows what a plan might look like.
ClinicalTrials.gov and other clinical investigation plans give ideas for inclusion and exclusion criteria, endpoints and the like. A search on ClinicalTrials.gov for an intervention such as transcatheter can be narrowed further, for example to transcatheter and aortic for aortic valves or transcatheter and mitral for mitral valves. A search for transcatheter alone returned 322 hits, and a study record gives a full-text view, a tabular view, the outcomes, the endpoints and the inclusion and exclusion criteria. One experienced writer holds ClinicalTrials.gov to be a very useful source.
Templates
One experienced writer holds that a writer who has never written a protocol can easily write one by following a few practical tips, and that though the first may not be the best protocol in the world, it will certainly work. The first of these tips is to ask the company whether it has a template. Most companies have one, so the writer basically just fills in the gaps.
A writer preparing plan and report templates from scratch can follow the order of the template in ISO 14155. A writer who already has a template under ICH E6, the guideline for pharmaceutical products, might use it as a basis, adjusting it where needed to comply with the regulations and guidelines for medical devices. Other sources of templates may depend on the country where the investigation is conducted: the Swiss Association of Research Ethics Committees, for example, has published a plan template on the SwissEthics website.
Who contributes, and what the writer owns
Preparing the investigation documents requires collaboration with numerous professionals with expertise in clinical practice, statistics, data management, monitoring and regulatory requirements. A writer developing a protocol works with the clinical research department and the statistician, and may also work with the clinical science department, data management and, if the preclinical studies have not been summarised adequately, the preclinical group. The writer may also work with the regulatory affairs department, which usually reviews a protocol for a CE-mark study or a pre-market study. One experienced writer recommends that the safety department also look at the protocol. How many departments are involved depends very much on the company. At one company, one experienced writer basically wrote protocols alone and had the others sign, whereas a writer with a very complex protocol at an unfamiliar company has to find out how many departments they will work with.
Everybody contributes their knowledge, the writer included. Because the writer writes the documents, it is crucial that they fully understand the objectives, endpoints and methodology of the investigation, and that takes communicating with people of other professions.
A writer who has written such documents before, for pharmaceuticals or for medical devices, will have gained experience in study design and can make a valuable contribution. Even a writer doing so for the first time more than likely has valuable experience and a perspective different from the rest of the team, perhaps from other medical writing or from designing laboratory experiments. So the writer should not be afraid to suggest alterations, to ask questions when something is unclear, or to offer input.
The writer's role may differ with the work situation and the support requested, but in every case the writer is responsible for conveying the core ideas of the investigation, providing the information the applicable regulations and guidelines require, and coordinating comments and creating consistency throughout the documents. To get started, the writer should read up on the applicable guidelines and believe in their own abilities.
The synopsis, drafted and agreed first
The synopsis summarises the investigation with all relevant information, such as the inclusion and exclusion criteria, the number of subjects, the duration of the investigation, follow-up, the objectives and the endpoints. It is likely the only part of the plan the investigators will read, which is not ideal but is reality, so anything paramount to the success of the study should be added to it. One experienced writer usually also includes a table of the assessments to perform at each visit, because investigators often look at that alone.
Including a schematic figure of the overall investigational design is one way to make the investigation easier to understand and present. It is also common, and advisable, to include in the plan a table of the frequency and timing of the clinical visits and of what is done at each, such as procedures and laboratory tests. This table is called the schedule of events or schedule of activities, and is equivalent to the similar table in a clinical study protocol for an investigational medicinal product.
The synopsis should be drafted first and reviewed thoroughly before the plan is drafted in its entirety. This can save a lot of time, because most questions and discussion concern the synopsis and all its content also appears in the main document, so the plan does not need updating in several places multiple times. One experienced writer also includes the full section on investigational design in this first draft, with the figure of the overall design and the schedule of events, because they often spark discussion and, together with the synopsis, set the basis for the plan.
The coordinating or principal investigator and any other medical expert should be involved as early as possible in drafting the synopsis, to obtain input on the investigational design and study procedures and to ensure a study setup as close to standard clinical practice as possible. ISO 14155 requires access to a medical expert with relevant knowledge for the investigation, who should be available to advise on its design and answer related medical questions. Specific questions should be discussed immediately with the clinician or another professional concerned, such as the investigation's statistician, and more general concerns with someone familiar with the project.
Reviews and the writing schedule
The synopsis and the investigational design, and later the full plan, should be reviewed by people of various professions and expertise, to catch potential problems with the plan as early as possible and to ensure that it is practically feasible. If possible, to cover all theoretical and practical aspects, the reviewers should include a statistician, a monitor and a data manager in addition to the clinical project manager, the manufacturer and the coordinating investigator. Depending on their own experience, the writer may also want to include someone more senior with regulatory knowledge.
When planning the writing to meet set deadlines, the writer should leave enough time for answers from the investigational team and experts, for reviews and revisions, and for other projects on the side. Whether the writer works at a consultancy, freelance or for a manufacturer, it is good to involve the manufacturer, colleagues and experts early in drafting the plan, though exactly how this looks varies with the work situation, experience and assignment.
Consistency and clarity
One experienced writer holds that consistency, clarity and unambiguousness are the most important things in writing a protocol. A writer who manages to write a consistent protocol is already in the upper league. Consistency matters because the protocol describes how people should conduct a study, so conflicting information in it is fatal, and such conflicts occur quite frequently. Clarity and unambiguousness matter too, because the people reading the protocol are busy and have little time, so a protocol should be as clear as possible and use visual aids such as graphics and tables.