“This operationally seamless approach to the Phase 2b/Phase 3 clinical trial design for AV-101, with continued enrollment and collection of multiple endpoints, underscores Aerovate’s commitment to making new treatment options available to patients with PAH as soon as possible without compromising safety and scientific rigor,”
- Jeff Maca, et al, 2006 Adaptive Seamless Phase II/III Designs— Background, Operational Aspects, and Examples, Drug Information Journal
- Chen, YJ, et al, 2015, A Seamless Phase IIB/III Adaptive OutcomeTrial: Design Rationale and Implementation Challenges, Clin Trials
- Jenkins, M, et al, 2011, An adaptive seamless phase II/III design for oncology trials with subpopulation selection using correlated survival endpoints, Pharm
stat.
In FDA's final guidance "Adaptive Designs for Clinical Trials of Drugs and Biologics", the terms 'seamless design' or 'adaptive seamless design' was no longer used, instead, they were described under the 'Adaptations to Treatment Arm Selection' section. The term 'seamless' occurred only once throughout the guidance in the following sentence:
In general, seamless designs that incorporate both dose selection and confirmation of efficacy of a selected dose (based on data from the entire trial) can be considered if the principles outlined in section III (Principles for Adaptive Designs) are followed.
Operationally Seamless (or simply Seamless Design)
A "Seamless Design" combines two separate trials (individual Phase 2 and Phase 3 trials) into one trial. "Operationally seamless design" specifically refers to the strategic and efficient organization of various aspects of the clinical trial process to ensure smooth and effective operations. Clinical trials are complex endeavors involving multiple stages, from protocol development and participant recruitment to data collection, analysis, and reporting. Operationally seamless design in this context aims to optimize these processes for enhanced efficiency and effectiveness.
Key features of operationally seamless design in clinical trials may include:
Separate Statistical Analyses: Statistical analyses for Phase 2 and Phase 3 trials are separated, not combined. The data from the Phase 2 study will not be included in the statistical analysis of the Phase 3 data - therefore, the issues with multiplicity adjustment, handling of the immature data during the interim analysis, ... may be avoided.Integrated Workflows: Streamlining the various stages of the clinical trial, from patient recruitment to data collection and analysis, to minimize delays and improve overall trial efficiency.
Technology Integration: Leveraging technology solutions for data management, patient tracking, and communication to enhance the overall efficiency of the trial. This may involve using electronic data capture (EDC) systems, remote monitoring tools, and other technologies.
Collaboration and Communication: Fostering effective communication and collaboration among different stakeholders, including researchers, sponsors, regulatory bodies, and clinical sites, to ensure a cohesive and coordinated approach.
Patient-Centric Approaches: Implementing strategies that prioritize the experience of trial participants, making it easier for them to participate and comply with the trial requirements. This might involve the use of telemedicine, remote monitoring, or other patient-centric technologies.
Regulatory Compliance: Ensuring that the trial design and operations adhere to regulatory requirements, which helps in avoiding delays and ensuring the validity and reliability of the trial results.
Risk Management: Proactively identifying and managing potential risks throughout the trial to mitigate issues that could impact the overall progress and success of the study.
In summary, operationally seamless design in clinical trials is about creating a well-integrated and efficient process from the planning stages through to the conclusion of the trial. This approach aims to improve the quality of clinical trial data, reduce operational costs, and accelerate the development of new treatments.
The IMPAHCT study mentioned above is a operationally seamless design and the study design is described as the following:
PART 1 (Phase 2b): Part 1, which is the Phase 2b portion of the trial, will assess the safety, tolerability, and efficacy of three twice-daily doses (10, 35, or 70 mg) of AV-101 against placebo and establish an optimal dose for Phase 3. The primary endpoint for this part is change in pulmonary vascular resistance (PVR) after 24 weeks compared to placebo. (note: it is said that approximately 200 patients (50 per arm) will be enrolled in this part of the study)PART 2 (Intermediate, Phase 3): PART 2 begins immediately following enrollment of the last participant in the Phase 2b part of the trial and signifies the start of enrollment in the Phase 3 trial. Part two uses the same dosing as in the Phase 2b part of the trial with participants randomized across three AV-101 doses and placebo. Enrollment in part two will continue until the optimal AV-101 dose is selected based on results from the Phase 2b analysis.PART 3 (Phase 3): This part of the trial will start once an optimal dose of AV-101 has been selected based on the Phase 2b results. All patients enrolling during this part of Phase 3 will be randomized to either the optimal dose of AV-101 or placebo. The primary endpoint for Phase 3 is change in six-minute walk distance (6MWD) at 24 weeks for the optimal dose of AV-101 compared to placebo.
In order to be operational seamless, the IMPAHCT employed a PART 2 (intermediate, Phase 3) portion of the study. During this stage, all patients for Phase 2b portion of the study have been enrolled, but are being followed up for reaching the endpoint (24 weeks) and for conducting the interim analyses. In PART 2 (intermediate, Phase 3) stage, patients are still randomized to one of three dose arms or placebo. The data collected from two active arms that are not selected for PART 3 will not be included in the final analyses and will be wasted. For example, if the 35 mg BID dose is selected as the optimal dose for PART 3, the final analyses will be comparing the 35 mg BID dose with the Placebo. The data from the 10 mg BID and 70 mg BID treatment arms will be wasted. Suppose it takes 28 weeks from the last patients randomized in Phase 2b to the sponsor establishing the optional dose for Phase 3, if additional 100 patients were enrolled into the PART 2 (Intermediate, Phase 3), 50 of these patients (not in the optimal dose or placebo) will be wasted.
With the operationally seamless design, the efficiency is sacrificed for the speed.
Inferentially Seamless Design (Adaptive Seamless Design)
An adaptive seamless design makes use of information (data) from patients enrolled before and after adaptation (pulls together data collected in both the Phase 2 and Phase 3 trials) in the final analysis.
The primary purpose of using the adaptive seamless design is to combine both the dose selection and confirmation phases into one trial, so information from the learning stage (Phase 2) can be combined with the confirmatory analyses of Phase 3.
"Inferentially seamless design" refers to an approach that emphasizes the seamless integration of data and statistical methodologies to derive meaningful inferences and insights throughout the course of the trial. The goal is to enhance decision-making by continuously analyzing data, drawing inferences, and adapting the trial design based on emerging findings.
Adaptive seamless design is one of the adaptive designs. The typical setting for adaptive seamless design (inferentially seamless design) is to start the study with multiple active dose groups. At the interim analysis at the end of Phase 2 or close to the end of Phase 2, the data monitoring committee will review the unblinded data accumulated so far to select an optimal dose for Phase 3 portion of the study. The phase 2 data from the selected dose group and the placebo group will be included in the final analyses and contribute to the inferential analyses. The appropriate statistical methods need to be applied. For example, Grifols is conducting a phase 2/3 study with adaptive seamless design "Study of the Efficacy and Safety of Immune Globulin Intravenous (Human) Flebogamma® 5% DIF in Patients With Post-polio Syndrome (FORCE)" where a method proposed by Posch et al "Testing and estimation in flexible group sequential designs with adaptive treatment selection" was used to combine the data from phase 2 and phase 3.
The inferentially seamless design can be illustrated as the following (cited from the paper by Maca et al "Adaptive Seamless Phase II/III Designs— Background, Operational Aspects, and Examples"
Putting them side by side, here is the table to compare the operationally seamless design and the inferentially seamless design:
Operationally Seamless Design
Inferentially Seamless Design
Integration of processes and systems for smooth operation.
Integration of data and insights for seamless decision-making.
Not an adaptive design - since the data from the phase 2 study is not included in the inferential analysis
One type of adaptive designs - Adaptations to Treatment Arm Selection
Emphasizes operational efficiency and workflow integration.
Focuses on integrating data and deriving meaningful insights.
Data collected from the phase 2 and phase 3 portions of the study is analyzed separately and separate clinical study reports may be written
Data collected from phase 2 the selected arms (the selected dose and the placebo arm) is combined into the phase 3 portion of the study and contributed to the final inferential analyses
Data collected from the phase 2 portion of the study is not included in the final inferential analyses
Overall sample size is larger since the data from Phase 2 and the data from the unselected arms in Phase 3 will not be used in the final inferential analyses.
Overall sample size is smaller since the data from some patients in the phase 2 portion of the study are used in the final inferential analyses
No pre-specified rules for dropping the inferior arms or selecting the optimal dose arm for confirmatory portion of the study
Pre-specified rules for dropping the inferior arms or selecting the optimal arm for the confirmatory portion of the study.
Dose selection is at the hand of the sponsor (the sponsor can unblind the phase 2 portion of the study data upon the completion)
Dose selection is implemented through the data monitoring committee (only they can review the unblinded data).
No need to deal with the multiplicity issue since the Phase 2 data is not used in the final analyses
Multiplicity issue needs to be considered and alpha needs to be adjusted since the portion of the phase 2 data will be included in the final analyses
Overrunning issues and handling of immature data (i.e. incomplete data at the time of data cut for interim analysis for patients who haven't reached the study endpoint) need to be considered
Additional reading:
- the blog article by Esha Senchaudhuri (2014) "Operationally Seamless & Inferentially Seamless Adaptive Designs"
- Bhatt and Mehta (2016) Adaptive Designs for Clinical Trials
- Aerovate Therapeutics Announces Simultaneous Completion of Enrollment in Phase 2b Portion and Enrollment of First Patient into Phase 3 in the IMPAHCT Trial Evaluating AV-101 for the Treatment of Pulmonary Arterial Hypertension - the press release said IMPAHCT is "a seamless adaptive Phase 2b/Phase 3 design", but it is not. It is an operationally seamless design, not adaptive since the Phase 2b data will not be used for the final analysis.
No comments:
Post a Comment