Picking a Proprietary Name for a New Drug is No Simple Task – FDA Issues a Draft Guidance on Best Practices in Developing Proprietary Names

By James C. Shehan & David B. Clissold –  

Good brand names have been important to drug marketers as far back as Biblical times – the Book of Ecclesiastes tells us that “a good name is better than precious ointment.”  In a new draft guidance applicable to Rx and OTC products, FDA has laid out a detailed system to assist sponsors in selecting precious proprietary names that will pass FDA review.  The system includes four chronological steps: (1) a prescreening process to weed out things like obvious similarities in pronunciation or spelling; (2) a secondary review to detect misleading and error-prone attributes such as inclusion of the product dosing interval (e.g., “BID”); (3) a misbranding review to determine whether the proposed name suggests safety, efficacy or other unique attributes that are not supported by scientific evidence (e.g., including “best” in a name); and (4) a final look-alike sound alike (“LASA”) safety review that utilizes searches of public databases and extensive name simulation studies designed to mimic multiple real world use scenarios of the proposed product. 

The primary motivation for the new guidance is patient safety, with reduction of inappropriate promotional claims a secondary interest.  FDA states that the draft guidance partially fulfills its PDUFA IV commitment to take steps to reduce medication errors and cites a 1999 Institute of Medicine report that attributed 7,000 annual US deaths to such errors.  Kohn LT, Corrigan JM, Donaldson MS, eds. To Err Is Human: Building a Safer Health System. Institute of Medicine, National Academies Press: Washington, DC, 2000.  The new draft guidance partially reflects input from public meetings held in June and December of 2003 and July of 2008 and an FDA pilot program (see our previous posts here and here) and is intended to complement the existing guidance entitled “Contents of a Complete Submission for the Evaluation of Proprietary Names.”

The prescreening process is designed to identify and weed out readily identifiable aspects of proposed names that are very likely to raise concern for FDA.  Among these obvious non-starters are names obviously similar in spelling and pronunciation to existing names, names that include medical and/or coined abbreviations, inactive ingredients or the United States Adopted Name (“USAN”) stem, and names that are the same as that of a discontinued product.

The next step in the review process recommended by FDA is intended to weed out misleading and error-prone names.  Name that may be weeded out in this step are those that include product attributes such as dosage form (tabs), route of administration (oral), manufacturing characteristics (lyophilized) and dosing interval (BID). Certain name modifiers (e.g., Roman numerals) are also identified as problematic, as are names that are identical or similar to non-US products with different active ingredients (thus applicants apparently should conduct global name searches), names that include symbols and names that include the sponsor’s name.   Note that while discussing the prescreening process FDA uses hortatory language such as “should not,” in this second step, FDA uses softer words and phrases such as “discourages” and “recommends … against.”

The misbranding review called for in the third step suggested by the draft guidance is relatively simple.  Sponsors are advised to not suggest that a drug is safer or more effective than what is supported by available scientific evidence and to avoid names that falsely suggest “some unique effectiveness or composition.”

FDA devotes the largest part of the draft guidance to the last step in the suggested process, the LASA safety review.  Sponsors are recommended to conduct name simulation studies and a lengthy methodology for conducting them is laid out.  FDA calls for parallel group observational studies that simulate “real-world use conditions” that “reflect the full range and variety of tasks involved in the prescribing, transcribing, dispensing, and administration of drugs,” include “common and easily simulated characteristics of real use” and “approximate the diversity of real-world prescribing conditions.”  Although sponsors are advised to test several different handwriting samples, it is not clear just how poor the handwriting is supposed to be, and a poorly written prescription with the potential for “confusion” has been the basis for FDA rejecting proposed proprietary names in the past.  Given a bad enough handwriting sample, one can make “Ecclesiastes” look like “Elephants.”  Sponsors are advised to test a minimum of 20 scenarios.  In a table provided in the draft guidance as an example, FDA advises that the test of a proposed fictional oral product include 70 healthcare participants in 23 different scenarios, with each participant asked non-leading scripted follow-up questions with verbatim responses recorded.  Sponsors are also advised to conduct orthographic and phonetic searches using the FDA’s Phonetic and Orthographic Computer Analysis system and drug reference databases such as Drugs@FDA and RxNorm.

A separate section describes some additional considerations for selecting OTC product names, such as the inclusion of consumers in the simulation tests and the importance of being able to distinguish products in the same family of brands. 

Of course, many companies have been following these steps for several years, or have engaged outside branding experts to perform the same analyses, prior to submitting a proposed proprietary name for FDA review.  Although FDA currently reviews those reports, it does not seem that much weight is given to them since FDA conducts its own name review in any event, and continues to reject a substantial number of proposed proprietary names.  The draft guidance does not propose to treat sponsor’s submissions any differently whether “best practices” are followed or not.  It would be interesting to know whether following such practices leads more often to an accepted name, which is data that only FDA is likely to have.

The draft guidance notes at the outset that it is applicable to biologics.  No reference is made, however, to the most hotly-debated current topic in drug naming – whether or not biosimilar products must bear distinguishable non-proprietary names (see our posts here, here, and here).  But FDA may issue a draft guidance on that topic in the near future.