The appropriate number of projects to keep in the pipeline depends on whether preclinical research projects, full-scale clinical development projects, or both are being discussed. The number of preclinical drug discovery projects may be further broken down into active projects involving exploratory nonfocused studies in broad therapeutic areas and research projects on highly targeted diseases. Research projects may or may not be required to have a chemical lead before they qualify as an official research project.

Once a commitment is made to take a compound to humans for clinical testing, it may be considered as a full-scale development project. Some companies divide this group of full-scale development projects into two categories, those that will be taken to market (if possible) and those that will only be evaluated in humans to address a specific question.

Although projects on individual drugs are described as the lowest common denominator between pharmaceutical companies, each project may also be defined based on one or all of a drug’s individual indications, dosage forms, routes of administration, and dosing regimens being studied. Therefore, a single drug may be represented by one or a dozen separate projects. An equal or greater amount of work may be expended on studying a minor route of administration for one drug as the entire effort spent on another drug project. Also, a single indication of a major project may require far more resources than all the indications of several other projects combined. Although the rest of this chapter discusses projects, it is necessary and important also to consider and evaluate within each project the resources allocated for each indication, dosage form, and route of administration currently under study or planned.

One approach to determining the number of drugs to develop simultaneously involves a theoretical approach. This is perhaps best accomplished through mathematical modeling. Mathematical modeling of the appropriate number of projects for a company to develop requires consideration of the company’s goal (e.g., a new drug generating $200 million or more is desired every two years). Estimates may also be required of any number of additional parameters that the company wishes to identify and control. These parameters include the (a) rate at which projects will be successful, (b) average number of projects to be initiated each year, (c) average length of time and costs (resources) required to achieve a successful project, (d) length of time required to terminate an unsuccessful project, and (e) total costs to be spent on unsuccessful projects. Instead of using a single estimated average for any of these parameters, it is possible to use two or more different averages or a range of values (assuming a normal distribution). These estimates should be derived from past company experience, present trends, and future goals. When the past, present, and future estimates for a parameter differ greatly, it would be desirable to derive the model using each set of values or to use the two extreme values.

When this type of mathematical simulation is run, wide fluctuations are observed in most of the outputs. Because of uncertainties of assumptions used in the model and variations in numbers used, the appropriate number of projects to develop simultaneously turns out to be a relatively wide range.

Any company’s portfolio of investigational drugs may be improved in quality by adding better projects in terms of medical and commercial value. An issue discussed in the following text is how to trim less attractive projects from the portfolio. An important issue for a research and development-based company is to determine how many clinical development projects and how many research projects with chemical leads are required to keep the drug development pipeline full. No one can answer this question with certainty at any one company because many factors determine whether compounds and drugs will be successful. Moreover, only some of these factors are able to be controlled. One means of attempting to keep the pipeline full is to determine mathematically the steady-state number of
projects necessary to achieve one’s goals, as previously described. It is then the responsibility of senior managers to ensure that this number of projects is present. A more practical means is to evaluate the contents of the company’s portfolio to determine whether the pipeline is full. This may be done by investigating the activities of all departments involved in drug discovery, development, and sales.

Many managers believe that each component part of research and development should be operating at or near maximal capacity in a highly efficient and productive manner. By keeping the overall research and development system (as well as individual departments) taxed beyond their capacity to handle the current work load, it will be more likely that the system is being used efficiently. The drawback to this approach is that it ensures that every department is diligently working but does not ensure that they are working on the most important projects, on the most important activities within those projects, or in an efficient manner.