Accurate estimation of MTD minimises safety and efficacy issues leading to late phase clinical trial failures and is of particular importance for oncology, which has one of the highest Phase III attrition rates among all major therapeutic classes.
“Inaccurate dose selection in Phase I can place a ceiling on drug efficacy in all subsequent phases, potentially dooming a good drug, or expose patients to needlessly high and harmful doses. Operational tools that improve accuracy of Phase I trials is a major objective for drug developers,” says Professor Andy Grieve, Senior Vice President of Clinical Trial Methodology at Aptiv Solutions.
The pharmaceutical industry is well aware that the traditional 3+3 approach, which is still used in more than 95% of Phase I trials, is slow and often underestimates the MTD. Alternative model-based approaches, such as the continual reassessment method (CRM) and the modified toxicity probability interval (mTPI) approach, have broader appeal. Regulatory agencies are supportive of the development of innovative dose escalation procedures. For example, Sue-Jane Wang of the Office of Biostatistics at the FDA co-authored a recent publication in the Journal of Clinical Oncology on the mTPI methodology.
Despite clear advantages, adoption of these model-based approaches has been slow due to the intensive programming required for confident implementation and avoidance of MTD over-estimation. Aptiv Solutions developed ADDPLAN DF 3.0 to provide a validated platform for simulating and comparing dose escalation designs without the need for intensive programming. The software, which benefited from beta testing feedback from ADDPLAN DF Consortium members at Eli Lilly, Janssen, and Novartis, simulates and allows comparison of the 3+3, mTPI, classical CRM, and a modified CRM approach known as Bayesian logistic regression with overdose control. The latter methodology was originally developed by Novartis.
“Several dose escalation methods exist but without a tool to compare these methods, the effectiveness of Phase I studies is significantly reduced. Simulations comparing the impact of dose-response assumptions, dose selection rules, and stopping rules increase the transparency of these methods and help statisticians select the appropriate dose escalation design,” says Grieve. “ADDPLAN DF 3.0 is a robust tool to help resolve dose-related drug failures and supports the industry’s push to adopt innovative design and execution strategies to improve drug development.”
Executing adaptive designs in Phase I and beyond can require sophisticated logistics and robust firewalls to maintain trial integrity. ADDPLAN DF software provides direct integration with AptivAdvantage®, the leading real-time execution technology platform for seamless implementation of dose adaptations and operational modifications in adaptive design trials.
In addition to its utility for Phase I studies, ADDPLAN DF is the first validated software to utilise the EMA-qualified MCP-Mod methodology for more accurate analysis of Phase II dose-finding designs. The ADDPLAN DF Consortium members are currently collaborating to develop the first toolset for applying the MCP-Mod methodology to adaptive trial designs, expected in late 2014.
Earlier this year, the FDA, EMA, and PMDA licensed the ADDPLAN family of innovative software for the design, simulation, and analysis of adaptive clinical trials.
The ADDPLAN DF development team, led by Grieve, has published a white paper on Phase I dose escalation designs, titled “ADDPLAN® DF for Designing More Accurate Dose Escalation Studies in Exploratory Drug Development,” that is available below.