Platform trial

Last updated May 16, 2023

A platform trial is a type of prospective, disease-focused,^[1] adaptive, randomized clinical trial (RCT) that compares multiple, simultaneous and possibly differently-timed interventions against a single, constant control group.^[2] As a disease-focused trial design (compared to an intervention-focused), platform trials attempt to answer the question "which therapy will best treat this disease". Platform trials are unique in their utilization of both: a common control group and their opportunity to alter the therapies it investigates during its active enrollment phase. Platform trials commonly take advantage of Bayesian statistics, but may incorporate elements of frequentist statistics and/or machine learning.^[3]

Purpose

Platform trials can be a particularly useful design when researchers predict that multiple therapies that would become available at different times require investigation. For example, when the COVID-19 pandemic began, researchers predicted that there would eventually be multiple different therapies that could be investigated, but that these therapies would be discovered at different times in the pandemic timeline, therefore making a platform trial a useful design.^[2] Similar to COVID-19, platform trials have found use in oncology, alzheimer's disease and pneumonia research.^[1] Platform trials can be a superior design compared to simple 2-arm clinical trials when multiple therapies need investigation, because it only requires a single control group. This means that platform trials can be conducted with fewer enrolled patients than a set of potentially redundant control groups in a series of separate 2-arm trials. This in turn allows for results to be published sooner for time-sensitive diseases, and for fewer patients to be exposed to the risks of a clinical trial.^[4] Platform trials may be appropriate for phase II-IV trails.^[3]

Design elements

Master protocol

Platform trials, like any clinical trial, have many elements that must be established before starting enrollment. While platform trials have the ability to alter their therapies of interest there are still many elements of these trials that remain constant and regulated. Such common, stable elements of platform trials described in the master protocol include: qualified trial staff members, trial sites, recruitment criteria, enrollment procedures, pre-set criteria for adding/discontinuing new therapies, adverse event reporting, communication plans, and statistical analysis plans. The master protocol is submitted to the IRB and once approved, only arm-specific appendices need to be submitted for Institutional Review Board (IRB) approval in the event of changes to the trial arms. Establishing a stable master protocol with adaptive therapy arms allows for faster, more efficient trial execution.

Platform trials are often large, multi-site investigations and as a result, master protocols frequently try to identify common human and physical infrastructure to maximize resource availability and efficiency. Examples of this include identifying/creating a single IRB to review the trial for all sites, creating a single database for collecting data, and creating a single randomization mechanism for all enrolled patients.^[4]

Common control group

One of the defining aspects of a platform trial is the shared control group that all interventional arms are compared to. Whereas a conventional RCT would generally have half of all enrolled patients in the control group; platform trials have a higher total number of patients in various interventional groups. This allows for fewer patients to be enrolled which saves money and accelerates completion time.^[4] A common statistical tool for determining allocation ratios, Dunnett's test,^[5] suggests that n√t patients should be allocated to the control group; where "n" is the sample size for each of the arms and "√t" is the number of active arms. As the number of arms increase, the ratio of patients allocated to control also increases. This results in the control group having a higher proportion of allocated patients than any one arm^[1] though platform trials still allow for more total patients to be in intervention arms than multiple 2-arm RCTs.

While the control group is not necessarily designed to change in the way that the treatment arms are, because platform trials can run for long periods of time, control groups may have to evolve to stay current with standard of care. When this is the case, or if there is a change to patient demographics with time, later analysis of the trial must be careful to consider comparing investigational patients to only the appropriate subset of control patients.

Adaptive intervention groups

The second defining aspect of a platform is that the therapies under investigation can change during the active enrollment phase of a trial. By comparison, conventional RCTs must specify the therapies under investigation before active enrollment and then discontinuation of a therapy results in discontinuation of the entire trial. Platform designs allow for addition and/or discontinuation of therapy arms. Importantly, the addition or discontinuation of an arm must follow pre-set protocols such as reaching a certain demonstrated efficacy or being recommended by a set panel of experts. There are frequently caps to the number of arms that can be active at once which are pre-determined by the research team. The number of possible arms is influenced by considerations of cost, time available for the trial, operational feasibility, complications with organization large quantities of patient data and the number of total patients available for enrollment.^[1] While an arm most frequently represents a single therapy, advanced designs may have multiple therapies in a single arm. When this is the case, one arm may have different therapies in different therapy classes (i.e. one antibiotic and one immunomodulator).^[2] Another advanced strategy is for each arm to utilize the same treatments, but with each arm representing a different sequence of intervention administration.^[3] Advance trials may also be designed such that some arms are only activated depending on the results of other arms. For example, a higher-dose arm may only be activated if a lower-dose arm shows few side effects but also low efficacy.

Unlike conventional RCTs, intervention arms do not necessarily need to start at the same time chronologically. This feature is particularly useful when investigating diseases that have new therapies being discovered regularly since these new therapies can be added to the trial without needing to start a new trial each time a therapy is discovered.

Response-adaptive randomization

Response-adaptive randomization is not a necessary component of platform trials but unique aspects of platforms allow for this feature to be incorporated. Response-adaptive randomization refers to the capability of redistributing the patient allocation ratio when one arm is showing superior/inferior outcomes compared to other arms after an interim analysis. Allocation ratios can therefor be adjusted to put more patients into more successful arms; however the ratio of patients randomized to the control group does not change.^[4] Allocation ratios are determined through a mix of empirical interim evidence and simulation modeling. Care must be taken, especially early in the trial when limited sample sizes are available, to avoid extreme swings in allocation ratios as such swings could cause early biasing of data.^[3]

Limitations

While platform trials offer many advantages for investigating a single disease, their adaptive nature and potential for numerous and complicated arms limit the ability to execute platform trials. Platforms require a large number of experts for trial design, Data Monitoring and Safety Boards and operations leading to high cost and communication complexity. That platform trials can run for long periods of time may mean that updates to the standard of care in the control group are necessary, complicating analysis. Further, care must be taken to ensure that the data from late-added arms are compared to appropriate sub-sections of the control group, further adding to statistical complexity. Too, publishing results of terminated arms may be complicated if the whole trial has not yet completed, as shared data in the trial may still need to remain blinded. Additionally, the complexity of platform designs, which may have multiple sponsors and funding sources as well as changing treatment arms, can make them difficult to register in standardized databases. Platform trials, again due to their complexity, require long planning times and can therefore be a poor choice of design for therapies that require immediate investigation. Finally, funding mechanisms can become complicated if a trial is investigating different therapies from different pharmaceutical companies; and their ill-defined trial lengths make them less desirable funding targets from federal funding agencies.^[3]

Related Research Articles

A randomized controlled trial is a form of scientific experiment used to control factors not under direct experimental control. Examples of RCTs are clinical trials that compare the effects of drugs, surgical techniques, medical devices, diagnostic procedures or other medical treatments.

Clinical trials are prospective biomedical or behavioral research studies on human participants designed to answer specific questions about biomedical or behavioral interventions, including new treatments and known interventions that warrant further study and comparison. Clinical trials generate data on dosage, safety and efficacy. They are conducted only after they have received health authority/ethics committee approval in the country where approval of the therapy is sought. These authorities are responsible for vetting the risk/benefit ratio of the trial—their approval does not mean the therapy is 'safe' or effective, only that the trial may be conducted.

In a blind or blinded experiment, information which may influence the participants of the experiment is withheld until after the experiment is complete. Good blinding can reduce or eliminate experimental biases that arise from a participants' expectations, observer's effect on the participants, observer bias, confirmation bias, and other sources. A blind can be imposed on any participant of an experiment, including subjects, researchers, technicians, data analysts, and evaluators. In some cases, while blinding would be useful, it is impossible or unethical. For example, it is not possible to blind a patient to their treatment in a physical therapy intervention. A good clinical protocol ensures that blinding is as effective as possible within ethical and practical constraints.

A cohort study is a particular form of longitudinal study that samples a cohort, performing a cross-section at intervals through time. It is a type of panel study where the individuals in the panel share a common characteristic.

Clinical study design is the formulation of trials and experiments, as well as observational studies in medical, clinical and other types of research involving human beings. The goal of a clinical study is to assess the safety, efficacy, and / or the mechanism of action of an investigational medicinal product (IMP) or procedure, or new drug or device that is in development, but potentially not yet approved by a health authority. It can also be to investigate a drug, device or procedure that has already been approved but is still in need of further investigation, typically with respect to long-term effects or cost-effectiveness.

A hierarchy of evidence is a heuristic used to rank the relative strength of results obtained from medical research. There is broad agreement on the relative strength of large-scale, epidemiological studies. More than 80 different hierarchies have been proposed for assessing medical evidence. The design of the study and the endpoints measured affect the strength of the evidence. In clinical research, the best evidence for treatment efficacy is mainly from meta-analyses of randomized controlled trials (RCTs). Systematic reviews of completed, high-quality randomized controlled trials – such as those published by the Cochrane Collaboration – rank the same as systematic review of completed high-quality observational studies in regard to the study of side effects. Evidence hierarchies are often applied in evidence-based practices and are integral to evidence-based medicine (EBM).

In clinical trials and other scientific studies, an interim analysis is an analysis of data that is conducted before data collection has been completed. Clinical trials are unusual in that enrollment of subjects is a continual process staggered in time. If a treatment can be proven to be clearly beneficial or harmful compared to the concurrent control, or to be obviously futile, based on a pre-defined analysis of an incomplete data set while the study is on-going, the investigators may stop the study early.

Clinical equipoise, also known as the principle of equipoise, provides the ethical basis for medical research that involves assigning patients to different treatment arms of a clinical trial. The term was first used by Benjamin Freedman in 1987, although references to its use go back to 1795 by Dr. Edward Jenner. In short, clinical equipoise means that there is genuine uncertainty in the expert medical community over whether a treatment will be beneficial. This applies also for off-label treatments performed before or during their required clinical trials.

In fields such as epidemiology, social sciences, psychology and statistics, an observational study draws inferences from a sample to a population where the independent variable is not under the control of the researcher because of ethical concerns or logistical constraints. One common observational study is about the possible effect of a treatment on subjects, where the assignment of subjects into a treated group versus a control group is outside the control of the investigator. This is in contrast with experiments, such as randomized controlled trials, where each subject is randomly assigned to a treated group or a control group. Observational studies, for lacking an assignment mechanism, naturally present difficulties for inferential analysis.

AGILE is a platform trial for early-phase evaluation of new treatments for SARS-CoV-2 infection. The trial platform is a collaboration led by the University of Liverpool, working with the Southampton Clinical Trials Unit, the Liverpool School of Tropical Medicine, the MRC Biostatistics Unit at the University of Cambridge and the National Institute for Health and Care Research (NIHR) Clinical Research Facilities. The AGILE platform is funded by the Medical Research Council, and the Wellcome Trust, with additional funding for specific candidate evaluations from the pharmaceutical industry and Unitaid. The Chief Investigator of the trial is Saye Hock Khoo.

Treatment of chronic fatigue syndrome/myalgic encephalomyelitis (CFS/ME) is variable and uncertain, and the condition is primarily managed rather than cured.

A glossary of terms used in clinical research.

<span class="mw-page-title-main">Placebo-controlled study</span>

Placebo-controlled studies are a way of testing a medical therapy in which, in addition to a group of subjects that receives the treatment to be evaluated, a separate control group receives a sham "placebo" treatment which is specifically designed to have no real effect. Placebos are most commonly used in blinded trials, where subjects do not know whether they are receiving real or placebo treatment. Often, there is also a further "natural history" group that does not receive any treatment at all.

The JUPITER trial was a clinical trial aimed at evaluating whether statins reduce heart attacks and strokes in people with normal cholesterol levels.

The phases of clinical research are the stages in which scientists conduct experiments with a health intervention to obtain sufficient evidence for a process considered effective as a medical treatment. For drug development, the clinical phases start with testing for drug safety in a few human subjects, then expand to many study participants to determine if the treatment is effective. Clinical research is conducted on drug candidates, vaccine candidates, new medical devices, and new diagnostic assays.

The United States Cochrane Center (USCC) was one of the 14 centers on the world that facilitated the work of the Cochrane Collaboration. The USCC was the reference center for all 50 US states and US territories, protectorates, and districts: the District of Columbia, American Samoa, Guam, the Northern Mariana Islands, Puerto Rico, and the US Virgin Islands. The USCC was also the reference Center for the following countries: Antigua and Barbuda, Bahamas, Barbados, Belize, Dominica, Grenada, Guam, Guyana, Jamaica, Japan, Saint Kitts and Nevis, Saint Lucia, Saint Vincent and the Grenadines, and Trinidad and Tobago. The USCC discontinued on February 7, 2018.

The Solidarity trial for treatments is a multinational Phase III-IV clinical trial organized by the World Health Organization (WHO) and partners to compare four untested treatments for hospitalized people with severe COVID-19 illness. The trial was announced 18 March 2020, and as of 6 August 2021, 12,000 patients in 30 countries had been recruited to participate in the trial.

The Randomised Evaluation of COVID-19 Therapy is a large-enrollment clinical trial of possible treatments for people in the United Kingdom admitted to hospital with severe COVID-19 infection. The trial was later expanded to Indonesia, Nepal and Vietnam. The trial has tested ten interventions on adults: eight repurposed drugs, one newly developed drug and convalescent plasma.

In an adaptive design of a clinical trial, the parameters and conduct of the trial for a candidate drug or vaccine may be changed based on an interim analysis. Adaptive design typically involves advanced statistics to interpret a clinical trial endpoint. This is in contrast to traditional single-arm clinical trials or randomized clinical trials (RCTs) that are static in their protocol and do not modify any parameters until the trial is completed. The adaptation process takes place at certain points in the trial, prescribed in the trial protocol. Importantly, this trial protocol is set before the trial begins with the adaptation schedule and processes specified. Adaptions may include modifications to: dosage, sample size, drug undergoing trial, patient selection criteria and/or "cocktail" mix. The PANDA provides not only a summary of different adaptive designs, but also comprehensive information on adaptive design planning, conduct, analysis and reporting.

Atoltivimab/maftivimab/odesivimab, sold under the brand name Inmazeb, is a fixed-dose combination of three monoclonal antibodies for the treatment of Zaire ebolavirus. It contains atoltivimab, maftivimab, and odesivimab-ebgn and was developed by Regeneron Pharmaceuticals.

References

1 2 3 4 Park JJ, Harari O, Dron L, Lester RT, Thorlund K, Mills EJ (September 2020). "An overview of platform trials with a checklist for clinical readers". Journal of Clinical Epidemiology. 125: 1–8. doi: 10.1016/j.jclinepi.2020.04.025 . PMID 32416336. S2CID 218670123.
1 2 3 Angus DC, Berry S, Lewis RJ, Al-Beidh F, Arabi Y, van Bentum-Puijk W, et al. (July 2020). "The REMAP-CAP (Randomized Embedded Multifactorial Adaptive Platform for Community-acquired Pneumonia) Study. Rationale and Design". Annals of the American Thoracic Society. 17 (7): 879–891. doi:10.1513/AnnalsATS.202003-192SD. PMC 7328186 . PMID 32267771.
1 2 3 4 5 Angus, Derek C.; Alexander, Brian M.; Berry, Scott; Buxton, Meredith; Lewis, Roger; Paoloni, Melissa; Webb, Steven A. R.; Arnold, Steven; Barker, Anna; Berry, Donald A.; Bonten, Marc J. M. (October 2019). "Adaptive platform trials: definition, design, conduct and reporting considerations". Nature Reviews Drug Discovery. 18 (10): 797–807. doi:10.1038/s41573-019-0034-3. ISSN 1474-1784. PMID 31462747. S2CID 201652338.
1 2 3 4 Park JJ, Detry MA, Murthy S, Guyatt G, Mills EJ (January 2022). "How to Use and Interpret the Results of a Platform Trial: Users' Guide to the Medical Literature". JAMA. 327 (1): 67–74. doi:10.1001/jama.2021.22507. PMID 34982138. S2CID 245670604.
↑ Dunnett, Charles W. (December 1955). "A Multiple Comparison Procedure for Comparing Several Treatments with a Control". Journal of the American Statistical Association. 50 (272): 1096–1121. doi:10.1080/01621459.1955.10501294. ISSN 0162-1459.

This page is based on this Wikipedia article
Text is available under the CC BY-SA 4.0 license; additional terms may apply.
Images, videos and audio are available under their respective licenses.

[Park_2020-1] 1 2 3 4 Park JJ, Harari O, Dron L, Lester RT, Thorlund K, Mills EJ (September 2020). "An overview of platform trials with a checklist for clinical readers". Journal of Clinical Epidemiology. 125: 1–8. doi: 10.1016/j.jclinepi.2020.04.025 . PMID 32416336. S2CID 218670123.

[Angus_2020-2] 1 2 3 Angus DC, Berry S, Lewis RJ, Al-Beidh F, Arabi Y, van Bentum-Puijk W, et al. (July 2020). "The REMAP-CAP (Randomized Embedded Multifactorial Adaptive Platform for Community-acquired Pneumonia) Study. Rationale and Design". Annals of the American Thoracic Society. 17 (7): 879–891. doi:10.1513/AnnalsATS.202003-192SD. PMC 7328186 . PMID 32267771.

[:0-3] 1 2 3 4 5 Angus, Derek C.; Alexander, Brian M.; Berry, Scott; Buxton, Meredith; Lewis, Roger; Paoloni, Melissa; Webb, Steven A. R.; Arnold, Steven; Barker, Anna; Berry, Donald A.; Bonten, Marc J. M. (October 2019). "Adaptive platform trials: definition, design, conduct and reporting considerations". Nature Reviews Drug Discovery. 18 (10): 797–807. doi:10.1038/s41573-019-0034-3. ISSN 1474-1784. PMID 31462747. S2CID 201652338.

[Park_2022-4] 1 2 3 4 Park JJ, Detry MA, Murthy S, Guyatt G, Mills EJ (January 2022). "How to Use and Interpret the Results of a Platform Trial: Users' Guide to the Medical Literature". JAMA. 327 (1): 67–74. doi:10.1001/jama.2021.22507. PMID 34982138. S2CID 245670604.

[5] Dunnett, Charles W. (December 1955). "A Multiple Comparison Procedure for Comparing Several Treatments with a Control". Journal of the American Statistical Association. 50 (272): 1096–1121. doi:10.1080/01621459.1955.10501294. ISSN 0162-1459.

[1]

[2]

[3]

[4]

[5]