Numerous metaphors have been used to describe a company’s interactions with regulatory agencies, but the ones of greatest importance are those that help one plan one’s strategies. The author’s favorite in this regard is that of the game of chess. Although two opponents usually try to checkmate each other to win the game, this is not how this metaphor is being used. Companies should not have that as their goal or approach. People who understand chess know that many games end in a draw, and in fact, many skilled chess players try to reach a draw, particularly when they are at a small or moderate disadvantage with the pieces they have available. Companies should be seeking to reach a draw, which in the context of drug development, represents a win-win situation.

It is a widely held myth that great chess players can think many moves ahead to choose the perfect move for them to make at any point of a game. Interviews with grandmasters have shown, however, that they look at patterns of the pieces on the board to decide which move to make in order to enhance or optimize their position. Pattern recognition is also critically important for pharmaceutical companies in their interactions with regulatory agencies. A highly experienced drug development “grandmaster” who has seen and learned from many situations in his or her career, can anticipate from the pattern they observe which moves are most likely to lead to advancing the drug toward the market. In most
situations, a variety of options are present, particularly when an issue arises in clinical, manufacturing or any other function. Choosing the option with the highest likelihood of success is a skill that takes years to master and does not solely depend on logic. As many people have observed, negotiations is as much an art as a skill.

Chess can be used to describe several aspects of a company’s negotiating posture with regulators. The company is trying to reach a draw that is perceived as a win-win situation, where both sides are pleased with the outcome. It is rarely ever possible to reach this situation without some give and take. In looking at any aspect of the standards used to design a study or judge the number of patients to study for a New Drug Application (NDA), there is usually a wide range of answers or standards that are reasonable and a wider range that is possible. This implies that one may propose to regulators to conduct more or fewer studies, enroll more or fewer patients and to use one or another parameter as an endpoint. It is assumed that a company will negotiate within a scientifically and medically acceptable range of options, while still maintaining a high ethical standard.

Therefore, in making a proposal to the FDA regarding drug development or on one aspect of a study, it is possible to propose to obtain a somewhat lesser amount of data than one expects the FDA to agree to. If the FDA does not agree with one’s proposal, then one can hopefully negotiate to a position that both sides are comfortable with. It is a positive experience when one negotiates to a fallback position that represents what both the company and the FDA believe is fair and acceptable. If the company had initially suggested that fallback position to the agency, experience shows that the initial company proposal is often rejected.

A few specific examples can readily illustrate this process. For example a company may believe that it is ethically, medically and scientifically acceptable to submit an NDA with 400 patients treated with a specific investigational product. They realize that a very conservative regulatory agency may ask for 1,000 patients and the company may think it too extreme to propose an NDA with 400 patients. In this situation, the company may initially propose that the FDA accept for review an NDA with 475 patients, when the company is prepared to negotiate up to 750 patients. If their rationale is sound and their evidence convincing, it is reasonable to expect the regulatory agency to negotiate down from 1,000 to a mutually acceptable number. Another example is that the company could propose an ethically acceptable, but less strict clinical trial design in a situation where they are prepared to design the trial using a higher standard. Finally, one may propose using one set of legitimate statistical analyses to use with the dataset, when one is prepared to use another. If the Agency refuses to accept one’s initial position, it is then possible to propose a backup position to seek a “draw” which is considered as a win-win situation. There are a large number of issues in drug development that are subject to legitimate debate and disagreement, such as a situation where the regulators ask for more data than International Conference of Harmonisation guidelines, but there are other areas where little wiggle room exists. The area where least give and take and often no negotiation occurs is in the chemistry, manufacturing, and controls or quality area of technical development and manufacturing activities.

Of course, it is desirable for a company to generate a series of back-up positions so that one can go progressively from one to another, seeking agreement at each stage. If one is able to achieve agreement before reaching one’s final position, then the company will benefit. Sometimes, the agency is unwilling to even accept the final back-up position of the company. Clearly, one must engage in any negotiations with the best arguments to buttress each of one’s negotiated positions.

The chess metaphor is in actuality far more complex than mentioned above, as one is playing chess on multiple boards at one time and not on a single board. The multiple chessboards represent such disciplines or functions as clinical, toxicology, manufacturing, pharmacokinetics, regulatory submission content and format, scale-up and technical development issues (e.g., chemistry, stability tests, analytical test methodologies, formulation issues, combination drug issues). Even the type of regulatory submission [e.g., a 505(b)(1) versus a 505(b)(2)] is often subject to negotiations.

Almost every company the author has observed makes many assumptions about what regulatory agencies are thinking and how they will react toward a specific situation, question, or product. This conjecture is usually based primarily on the company’s own thinking, which is regarded (by them, and possibly by others as well) as logical and scientifically correct. A company often forgets the wide scope of possible perspectives that an organization or individual can take on almost any issue. The fact that regulatory agencies are highly conservative and are oriented to protecting public health as their primary responsibility means that they are not likely to accept at face value each of the perspectives favored by the company, which are influenced by many additional factors besides public health.

In addition, the regulatory agency has more information about class effects than any single company. They have reviewed data on numerous or many similar or related drugs that are terminated or in development and in today’s global regulatory environment, they are likely to have exchanged information with other regulators. This means that in numerous situations, a single company is not fully aware of the information that influences their views.

The problems that other drugs have had in development, particularly in safety, will influence their attitude and suggestions (or requirements) about which trials or parameters you will have to study that may not have been obvious or apparent to you. In negotiating with regulators, these points must be kept in mind.

The primary goal and motivation of regulators is not geared to approving your product, but to ensure that it will help the public while causing minimal harm (i.e., have a positive benefit-to-risk balance). Even in situations where the product appears to be safe, every regulatory agency knows that their jobs and even their careers are vulnerable if they make a mistake in judgment. This leads them to adopt a highly cautious approach to approving new therapies and sometimes to adhere to what is referred to as “paralysis by analysis.” This is frequently observed when the agency requests, if not requires, additional data, particularly to increase the safety database, because there is never a level at which one could ever say that a drug is safe. Although there are some guidelines on the size and nature of a safety database needed before a new drug is approved (e.g., International Conference of Harmonisation E1A), the quality and quantity of safety data desirable from a regulator’s viewpoint can be highly subjective and often differs widely among regulatory agencies. It is always possible for regulators to request additional safety data, something that is difficult to argue against. However, when the patient population available for study is limited, which it often is, this creates a difficult and sometimes an almost impossible situation to resolve. This is particularly true for indications with a very limited patient population, as occurs for rare diseases as well as for those diseases where there is adequate, if not excellent, therapy already available.

Many companies have submitted a protocol or other document to a regulatory agency and then received suggestions for changes. The company typically will make some, but not all of the suggested changes and resubmit the protocol. After a sufficient period of time passes without any response from the agency, the company may assume that the agency is in agreement with the latest changes and initiates the trial. A year or two (or even more) later, the data from that trial are presented to the agency. The response is often that “we never agreed to points a, b, and c and even wrote to you specifically stating our disagreement with the approach you took.” The company is caught off-guard, as their only defense (apart from the medical or scientific one) is to say that they assumed the agency was in agreement because they did not call or write to say otherwise. The author has personally seen this issue come up at least six times in the past few years, and several of those were in relatively large companies that made unwarranted assumptions.

In conclusion, moving ahead without confirmation about the assumptions you are making is very dangerous, particularly if the point of contention is an important one and could easily affect the status of the drug’s approval (e.g., a question on which efficacy parameter should be used as the primary endpoint in a pivotal trial).