TABLE 1

Revitalizing preclinical contributions to drug discovery—a data-independent assessment

This table was generated at the request of one of the reviewers. It could no doubt be enhanced and refined with additional input from the collective wisdom of those active in preclinical research.

Innovation
    Individual
        The published literature is not the only source of new ideas for first-in-class drugs—some of the most successful ideas came from within the industry (e.g., fluoxetine and cimetidine).
        Get the data to support a new concept that you believe in before you try to sell it—ideas are a dime a dozen, and many people are unable to reduce a concept to a peer-reviewable data set.
        Data are golden.
        When working on a best-in-class generation project, make sure you understand what “improved” looks like, how it stacks up to the initial drugs in the class (always benchmark to the competition), and whether the postulated improvements can be realistically tested and proven in the clinic.
    Management
        Allow a scientist who is really passionate about a concept the time to gather the necessary data to convince his/her peers and you. A minimal investment allows the separation of the true drug hunters from the dilettantes, the misguided, and the politically motivated.
        Mentor such people and recruit more of them.
        Tolerate dissent and avoid group-think. Always look for the “devil's advocate” viewpoint, and try not to shoot the messenger.
        The grass is infrequently greener. An expert consultant is not necessarily objective, and may not know the correct solution to, or even understand, your problem. Your experts should be in house.
        To paraphrase Orson Welles' famous ad for Paul Masson wine, “Serve no drug before its time.” If the compound is first in class, make sure it is the very best it can be to avoid whiny post mortems along the lines of “right target, wrong compound,” which again will convince no one.
Data generation/experimentation
    Individual
        Use the null-hypothesis approach.
        Apply appropriate statistical analysis to the data. An sample size of 1 convinces no one.
        Avoid qualitative science. Apply the law of mass action, the concentration/dose response effect (Maddox, 1992), in using compounds to define target pharmacology and pathway involvement. An antagonist used at a single concentration of 1 μM to invoke a response mediated at a target where it is active at 2 nM may result in an erroneous conclusion.
        Run appropriate controls. Do not rely on historical data or literature data obtained 15 to 20 years ago in a laboratory that no longer exists.
        Do not ignore data because it does not conform to what was anticipated and/or wanted. The unexpected often represents a paradigm shift (e.g., if you expect a decrease and get an increase). Mother Nature may also be telling you something about whether you should be running experiments.
The null-hypothesis culture
        Make decisions based on data.
        Apply the concept of the null hypothesis to the project portfolio—assume that none of them are on target to deliver and be pleasantly surprised when you find they are.
        Use a SWOT (strengths, weaknesses, opportunities, and threats) analysis approach to encourage transparency and dissenting viewpoints and to avoid complacency and the status quo.
        Develop and maintain a balanced risk, diverse target portfolio with a 3- to 5-yr timeframe (not 18–24 months).
        Review projects regularly but not in a timeframe that precludes the time and effort necessary to generate key data.
        Integrate absorption, distribution, metabolism, excretion and pharmaceutics in the lead optimization paradigm.
        Although Scientific Advisory Board (SAB) meetings are usually events where everyone has a win-win agenda, remember that 90% of what is done in drug discovery, the events that occur between project initiation and selection of a compound as an investigational new drug do not build on either the experience or interest of an SAB.
        Use retired pharmaceutical researchers as part of your SAB. Their prior experience can be very helpful, and at the very least, they will commiserate with you.
        Avoid:
            - dumbing down the scientific focus of a meeting to accommodate nonscientist participants.
            - sunk-cost scenarios.
            - not making decisions.
        Prohibit the use of wireless Internet devices at any and all meetings where you expect informed decision making.
Building a robust pipeline
        Make the translational medicine paradigm a reality in your organization, but first make sure everyone agrees to precisely what it is (Enna and Williams, 2009b; Wehling, 2009).
        Attempt to develop a seamless preclinical/clinical interface through phase IIa involving preclinical champions.
        Involve clinical in the drug-discovery process, but diminish the impact of risk-averse clinicians who prefer working on “derisked” phase IIb in-licensing candidates.