Ann: Invion alliance with Hudson Institute of Medical Research, page-14

Phase III
For a standard 505(b)(1) NDA, Phase III generally consists of two large, well-controlled studies (often referred to as pivotal studies) that include hundreds or even thousands of patients and are designed to produce statistically and clinically meaningful evidence of the product’s safety and efficacy. Once again, because a 505(b)(2) submission can rely in part on existing data, Phase III studies are often not necessary.

• If a Phase III study is required for a 505(b)(2), such as when approval is sought for a prodrug of a previously approved active ingredient, only one study is often necessary versus the two generally required for 505(b)(1).
• Fewer patients may be needed for 505(b)(2) product clinical trials due to the existing large exposure information available in the public literature or in the FDA’s databases.