505(b)(2) Literature Searches – Too much or too little?
- Posted by: Ken Phelps
- Published on: September 15, 2008
A 505(b)(2) submission relies on information in the public domain to fulfill some of the information required in an NDA for approval. This information comes from more than the reference drug’s NDA review documents. In fact, for older drugs, the amount of information can be overwhelming. A recent Google search on an RLD yielded almost 40,000 articles. Adding “human” reduced this list to about 10,000. The FDA expects the Sponsor to make the effort to sort out the level of evidence. The FDA will not accept a list of publications with the expectation that they do this work for the Sponsor to support their submission.
First, Google probably isn’t the best search engine for this type of work. There are more targeted search engines available, depending on the information that is sought. Searches on PubMed, MedWatch, Approved Product Labeling and the FDA Center for Drug Evaluation and Research (CDER) Website may be fruitful.
What you are looking for is evidence that can be used to support an FDA filing. For information to support clinical safety or efficacy, we’re looking for results of published controlled trials. Each article needs to be reviewed, prioritized for evidence and then summarized on the basis of dose, patient population, indication, duration, primary and secondary endpoints and safety (including drug-drug interactions). Articles that are review articles or research articles without supporting evidence are not considered. For prioritization of evidence we use the following criteria: Placebo-controlled trials are the highest level of evidence. Dose-response trials where patients are randomized to different fixed-dose groups are next. Active -controlled trials that show superiority of the proposed drug over the RLD follow, then active-controlled trials that show no difference would be last. These clinical trials are also distinguished based on whether they are blinded or open-label. Blinded trials have a higher level of evidence than open-label.
All of this information is needed at the pre-IND stage because what you’re looking for at this meeting is FDA concurrence with your plan. The plan is based on what is known and what is needed. Without a good grasp of what is known (the public information) the FDA may or may not be able to give adequate advice.
Camargo develops this information as part of our Assessment Report, to enable the client to determine whether to move ahead with the pre-IND meeting. The good news is that, once the public information is properly cataloged and summarized, it doesn’t take long to insert it into the IND and NDA filings.