Regulatory glossary

505(b)(2): The Pathway Between an ANDA and a Full NDA

A 505(b)(2) application is a New Drug Application that relies, in part, on data the applicant did not generate and does not own: FDA's previous findings of safety and effectiveness for an already-approved drug, or published literature.

It exists to fill the gap between the two other routes. A 505(b)(1) carries its own complete data. An ANDA argues sameness with a reference listed drug and generates almost no new clinical data. A 505(b)(2) does neither: it permits a difference from the listed drug, and supports that difference with its own data on top of FDA's prior findings.

The three routes

The row that decides your route is the third one. If your product is the same as the listed drug on every attribute an ANDA requires, the ANDA is cheaper and you should use it. The moment you have a genuine difference, the ANDA route closes and the 505(b)(2) is what exists.

What a 505(b)(2) is typically used for

• A new dosage form: an oral solution where the approved product is a tablet.
• A new route of administration.
• A new strength outside what the listed drug covers.
• A new indication for an approved active ingredient.
• A fixed-dose combination of approved actives.
• A different salt, ester or complex of an approved active.
• An OTC switch of a prescription product.
• A change in formulation that alters the product enough to need supporting data.

In each case the argument is the same. The active ingredient's safety and effectiveness is already established by FDA's findings on the listed drug, and your own data covers only what your change makes different: usually bioavailability or bioequivalence, sometimes clinical data where the change reaches efficacy.

What it requires that a 505(b)(1) does not

Patent certification. Because you are relying on a listed drug, you certify against the patents listed for it in the Orange Book, the same Paragraph I through IV structure an ANDA uses. A Paragraph IV certification triggers notice to the patent holder and the possibility of a 30-month stay, which is a litigation timeline rather than a regulatory one and often dominates the launch date.

A bridge to the listed drug. You have to establish a scientific bridge, usually a comparative bioavailability study, that justifies relying on the prior findings. Without an adequate bridge there is nothing connecting your product to the data you are leaning on, and the application fails on its own terms.

A defensible choice of listed drug. The same care applies as for an ANDA's reference: confirm the product is still listed, still marketed, and not withdrawn for safety or effectiveness reasons.

Where 505(b)(2) applications go wrong

• Choosing it when an ANDA would do. More expensive, slower, and it invites patent certification issues an ANDA route may have handled more simply.
• An inadequate bridge. The most common substantive deficiency. The bridging study has to support the specific reliance you are claiming.
• Underestimating the certification. A Paragraph IV certification is a litigation decision as much as a regulatory one, and it should be made with counsel before the application is drafted, not after.
• Assuming exclusivity. Three-year exclusivity attaches to the change supported by new clinical investigations, not to the product as a whole, and only where those investigations were essential to approval.
• Relying on literature you cannot support. Published literature can support a 505(b)(2), but it has to be adequate and applicable to your product, and FDA assesses that rather than accepting the citation.