|2004N-0181 - Critical Path Initiative; Establishment of Docket|
|FDA Comment Number :||EC4|
|Submitter :||Dr. Richard Cramer||Date & Time:||07/21/2004 11:07:06|
|Organization :||Tripos, Inc.|
| 4. For each problem identified, if a solution would facilitate the development of drugs, biologics, and/or devices for a particular disease or categories of disease, please indicate which diseases would be affected? |
|All diseases for which pharmaceutical remedies might be appropriate.|
|3. For each problem identified, please indicate the type of drug, biologic, or device to which the hurdle applies.|
| 1. Hurdle Identification. Please describe the product development issue, the nature of the evaluation tool that is out-of-date or absent, how this problem hinders product development, and how a solution would improve the product development process. |
| How might the cumulative data on FDA-submitted compounds (toxicity, side effects, bioavailability) be used to improve the selection of future drug discovery candidates? One major consideration is that the actual owners of these data (the drug discovery enterprises) have legitimate reasons for keeping confidential the associations of particular structures with biological observations. Another is that the biological observations themselves are very difficult to interpret and compare, being individual data points inconsistently measured with uncertain extrapolability to general clinical use. Finally this information would most effectively be applied as early as possible, far away from the contentious regulatory process, by the individual medicinal chemist selecting a particular compound series for investigation.|
| 5. Nature of the Solution. For each problem identified, please describe the evaluation tool that would solve the problem and the work necessary to create and implement the tool/solution. For example, would a solution come from scientific research to |
| The fundamental principle guiding medicinal chemistry is that similar structures tend to produce similar biological effects. (Indeed, it would seem that otherwise medicinal chemistry must be pointless.) This principle seems as applicable to undesirable biological effects as to desirable ones, with 'chemical space' also being so vast that early in discovery it would be relatively painless to avoid structures that are 'too similar' to structures having known liabilities.
However structural similarity has many possible dimensions, with most being irrelevant to biological similarity. For example, no one would argue that similarity in molecular weight tends to produce similar specific biological effects. It happens that extensive work at Tripos has yielded a 'topomer' (shape) similarity descriptor of structure, which as far as we can determine is the strongest predictor of 'biological similarity' that has yet been found. (Evidence supporting this admittedly strong claim can be found in a manuscript submitted for publication, currently available via email from firstname.lastname@example.org.)
Thus a solution to the hurdle of making FDA's cumulative data on compound liabilities useful for discovery would be to encode all those compounds as topomers associated with specific biological liabilities. A suitable 'black box' program could then allow any discovery scientist to
| check structural candidates against this encoded FDA data, reporting for evaluation the biological liabilities represented by any topomer similarity neighbors.
With respect to the above-mentioned confidentiality consideration, it should be appreciated that the mapping of a topomer code back to a single definite structure is seldom if ever possible, and further that the underlying technologies are complex, proprietary, and otherwise internally reserved by Tripos for 'lead-hopping' laboratory-based discovery collaborations. However the biological uncertainty consideration is not addressed by this strictly chem-centric solution.
We would not favor any sort of regulatory guidance or standard based on this solution. Its benefit would instead be to improve the chances of clinical success for future candidates, by making best use of existing data to minimize investments in predictably risky structures.
| 6. For each solution identified, please indicate which could be accomplished quickly, in less than 24 months, and which require a long-term approach? |
| The two programs envisioned by this solution (the second one being for FDA representatives to use for encoding structures into topomers) would need only a few months for development. However the many questions that would arise from widespread use of this solution should be addressed in advance by research. For example:
For various topomer similarity radii, what would be the risks of false positives and negatives? How dependent are these risks on the type of biological response?
Many biological liabilities are caused by metabolism to other structures. Might this issue be addressed at all usefully by predicting metabolites and comparing their topomers as well?
Should similarity to liability-free compounds be reported as well? Should the reliability of the biological observation underlying a topomer- encoded structure be somehow considered?
However, useful answers to questions such as these could easily be obtained within 24 months.
|7. For each problem identified, what role should FDA play and what role should be played by others?|
| To implement this solution, the FDA would clearly have responsibility for processing the liability data, along with obtaining any required permissions. External funding would also be required, and here perhaps the FDA might participate in approaching other possible sources.