A Declaration of Action

The FDA's Critical Path Initiative is an important step in meeting the goal of providing safe and novel products to the public in a more efficient and cost-effective manner. All stakeholders would benefit from a reversal of the current trends of escalating cost and elongated timelines in the drug development process. In March 2004, the FDA report, "Challenge and Opportunity on the Critical Path to New Medical Products," stated that "the investment required to launch a new drug had risen 55 percent between 1995–2000 and 2000–2002." This same report suggested that the Phase II and Phase III steps in the drug approval process were the main contributors to the cost increase. Rather than focusing attention on the development of tools to identify products that do not have a high probability for success at an early stage of their life-cycle, significant cost and time savings may be realized by investigating and implementing efficiencies in Phases II and III.

David Rucker

Dr. Lukas Makris, CEO of BioCor, suggests there is also a need for better strategic planning and up front study design to minimize the financial risk of clinical development and to ensure sponsors meet their main objectives later in the process.

Jeff Parno

We often read about Phase III trials failing to meet their objectives although Phase II data demonstrated significant promise and likelihood for success. Is this outcome merely statistical chance or are we cutting too many corners in this middle stage of drug development? It is not uncommon for Phase II studies to be designed as a "boutique" collection of data rather than a simulation of the broader experience. In many cases the preset Phase II targets are much easier to meet than those in Phase III, leaving a considerable degree of risk for the third stage of development and creating a false sense of timeliness and economies.

In a Phase II study we are trying to optimize the design by maximizing the "signal to noise" ratio, whether the signal reflects efficacy or safety attributes of the product. But in the process we may eliminate part of the "signal" as well as the "noise." We rely on a select set of investigators; add a refined patient population based on exhaustive inclusion and exclusion criteria; implement an optimized assessment of efficacy through endpoints with precise measurements; apply strict dosing guidelines and an intensive follow-up for toxicity; and create a possible patient selection bias considering the limited prior exposure of the product. These are all undoubtedly key elements for an optimal Phase II study. But in Phase III these criteria are relaxed while our assumptions are not adjusted accordingly. Instead we compare our Phase II results to known Phase III results for a competitive control agent, and our advantage is artificially inflated.

In addition, failure to adjust the assumptions of the Phase III drug effect often leads to underpowered Phase III programs, with all of their associated risks. Beyond significant p-values, a successful Phase III program must prove to the medical community that the original assumptions for effectiveness and safety have been met.

BioCor is a clinical research organization (CRO) that specializes in clinical data services to support Phase I through Phase IV programs from clinical plan design to defense of global regulatory submissions. Our position within the industry as a CRO allows us to gain a wide range of experience working with different organizations while maintaining confidentiality. The knowledge gained through our work includes the identification of both good practices and inefficiencies. Implementing good practices and avoiding inefficiencies are crucial for our organization to be competitive.

For example, BioCor has implemented a standard data format called the Study Data Tabulation Model (SDTM) developed by the Clinical Data Interchange Standards Consortium (CDISC). This model, adopted by FDA in July 2004, supports the concept of a standard method for electronic acquisition, exchange, submission, and archiving of clinical trial data. The primary goal of the standard is to reduce the time FDA reviewers spend reorganizing and reformatting data and to increase time spent evaluating data. The use of CDISC models allow efficiencies to be gained by establishing standard processes for data collection, data analysis, and documentation that can expedite the time from data collection to NDA submission.

There are many possible solutions, but sound preliminary strategic planning and clinical design as well as establishing standards such as the CDISC models may significantly reduce costs. Standardization of data collection templates, reporting methods, and analysis of safety data could also reduce both time and cost.

The first step is to recognize that a problem actually exists. We believe the FDA's Critical Path Initiative represents the challenge facing our industry. As an industry, we must acknowledge the need to make improvements towards providing new products to the public in a safer, more cost-effective manner. The FDA's Critical Path Initiative needs to exist as the industry's declaration of action.

David Rucker , a Manager, Clinical Programming

Jeff Parno , Associate Director, Statistics, BioCor