There is no single road to FDA approval. A 505(b)(2) application lets you rely on published literature or FDA's prior findings for an approved drug. Accelerated approval trades a surrogate endpoint for a confirmatory obligation. The Animal Rule, priority review, and real-world evidence each solve a narrower problem. The right path depends on your asset, your data, and how much market exclusivity you need.
What is a 505(b)(2) application, and who should use it?
Section 505 of the Federal Food, Drug, and Cosmetic Act describes three kinds of drug applications. A 505(b)(1) NDA contains full reports of safety and effectiveness studies run by or for the applicant. A 505(j) generic shows sameness to a listed drug. A 505(b)(2) NDA contains full reports, but some required information comes from studies the applicant did not conduct and has no right of reference to [1].
In practice, that outside information is published literature or FDA's prior finding that a listed drug is safe and effective. FDA's guidance on choosing between a generic application and a 505(b)(2) is direct: an applicant can reference the agency's findings for a listed drug, then bridge the differences with its own data [2].
The pathway fits products that change something about an approved drug. A retrospective review of 505(b)(2) NDAs approved from 2012 through 2016 identified 226 applications, with new dosage forms, new combinations, and new formulations the most common types [3]. It does not lower the evidence bar. If exposure, population, or indication differ from the listed drug, expect bridging pharmacokinetics and often a new efficacy trial.
What does 505(b)(2) cost you in exclusivity?
This is the tradeoff founders most often underestimate. Under 21 CFR 314.108, a drug containing a new chemical entity receives 5 years during which FDA will not accept a competing 505(b)(2) or generic application, or 4 years if the competing application challenges a patent with a Paragraph IV certification. An NDA that instead relies on new clinical investigations essential to approval, which describes most 505(b)(2) products, receives 3 years of protection limited to the conditions those studies supported [4].
Three years is short, and it covers only the change you studied. A competitor can reach the same molecule by a different route. If your business case depends on market protection, ask whether patents, orphan exclusivity, or formulation know-how extend the runway.
How does accelerated approval work, and what changed in 2023?
Accelerated approval lets FDA approve a drug for a serious condition based on an effect on a surrogate endpoint that is reasonably likely to predict clinical benefit, or on a clinical endpoint measured earlier than irreversible morbidity or mortality [6]. The regulations sit at 21 CFR 314 Subpart H for drugs and 601 Subpart E for biologics, and both require further study to verify clinical benefit [7], [8].
The price of early approval is a confirmatory obligation. If the confirmatory trial fails to verify the predicted benefit, or the sponsor fails to pursue it with due diligence, FDA may withdraw the approval through expedited procedures that include notice, a meeting, and public comment [6].
The Consolidated Appropriations Act, 2023, often called FDORA, changed the leverage. Congress amended section 506(c) to let FDA require that confirmatory studies be underway before approval or within a specified period after it [6]. FDA's January 2025 draft guidance describes how the agency will judge whether a trial is underway and signals that FDA intends to use this authority in most cases [9].
For planning, the confirmatory trial is no longer something to design after the surrogate readout. Enrollment should be open before you file. On a small program that can double the near-term capital need, which is why I treat accelerated approval as a financing question as much as a regulatory one.
How often do accelerated approvals get withdrawn?
The record is better than the headlines suggest, but the risk is real. A 2025 cohort study in JAMA Network Open examined 166 oncology indications granted accelerated approval from 1992 through 2022. As of August 2024, 102 (61%) had converted to regular approval, 30 (18%) had been withdrawn, and 34 (20%) were ongoing. Median time was 3.1 years to conversion and 3.8 years to withdrawal [10].
FDA publishes its own ledger. The agency's list of withdrawn cancer accelerated approvals held 35 indications as of September 2026, alongside companion lists of ongoing and verified approvals [11]. Investors read these pages, and an assessment that ignores them will not survive diligence.
When do the Animal Rule, priority review, and real-world evidence apply?
The Animal Rule, published May 31, 2002, applies only when human efficacy studies are not ethical and field trials are not feasible. FDA requires a well-understood mechanism, efficacy in more than one species or one well-characterized model, an endpoint tied to survival or major morbidity, and enough pharmacokinetic data to select a human dose [12].
Priority review is a review-speed tool, not an approval pathway. FDA's goal is to act on a priority application within 6 months instead of the standard 10 months, and it decides on the designation within 60 days of receiving the application [13]. The scientific standard does not change.
Real-world evidence has a narrower regulatory footprint than many assume. The 21st Century Cures Act of 2016 directed FDA to evaluate its use, and FDA's 2018 framework focused on supporting a new indication for an approved drug and satisfying postapproval study requirements [14]. FDA's 2023 final guidance states that the substantial evidence standard is unchanged and urges early engagement on study design [15].
How do I choose a path for my program?
| Pathway | Basis | Who qualifies | Evidence standard | Main risk |
|---|---|---|---|---|
| 505(b)(2) NDA | FD&C Act 505(b)(2); 21 CFR 314.108 | Modified versions of approved drugs | Reliance on literature or FDA findings plus bridging data | 3 years of exclusivity |
| Accelerated approval | FD&C Act 506(c); 21 CFR 314 Subpart H, 601 Subpart E | Serious conditions with a predictive surrogate | Controlled trial on a surrogate, then confirmation | Expedited withdrawal |
| Animal Rule | 21 CFR 314 Subpart I, 601 Subpart H | Countermeasures where human efficacy trials are unethical | Animal efficacy plus human safety and PK | Narrow applicability |
| Priority review | PDUFA review goals | Significant improvement for a serious condition | Unchanged | Faster clock only |
| Real-world evidence | 21st Century Cures Act | New indications; postapproval requirements | Substantial evidence unchanged | Data quality challenged |
On a radiopharmaceutical program I was involved with, we assessed whether published literature on the targeting ligand and FDA's prior findings for the isotope could support a 505(b)(2) filing. The answer was partly yes. The literature covered class safety, but our dose, schedule, and population were new enough that FDA would expect our own efficacy data, and 3 years of exclusivity alone would not justify the spend. However, we identified an alternative nonclinical development path that would expedite moving the radiopharma program quickly into a Phase 0 trial.
On a small-molecule prodrug program, the analysis ran the other direction. The parent drug's FDA finding of effectiveness was directly relevant, and the prodrug existed to change the exposure profile. A 505(b)(2) with bridging pharmacokinetics and one efficacy study was feasible, so the commercial question became whether patents, not exclusivity, protected the product.
My rule is simple. Match the pathway to the question your data can answer. 505(b)(2) answers how your product differs from a known drug. Accelerated approval answers how early you can show a believable signal and whether you can afford to confirm it. Then check the exclusivity math.
If you are weighing one of these pathways for a live asset, I am happy to compare notes on the evidence and exclusivity tradeoffs. Reach me at info@hamiltonatlas.com.
References
- U.S. Food and Drug Administration. Applications Covered by Section 505(b)(2). FDA, 1999.
- U.S. Food and Drug Administration. Determining Whether to Submit an ANDA or a 505(b)(2) Application. FDA, 2019.
- Freije I, Lamouche S, Tanguay M. Review of Drugs Approved via the 505(b)(2) Pathway: Uncovering Drug Development Trends and Regulatory Requirements. Therapeutic Innovation & Regulatory Science. 2020;54(1):128-138.
- U.S. Government Publishing Office. 21 CFR 314.108 - New drug product exclusivity. Code of Federal Regulations, 2024.
- U.S. Food and Drug Administration. Accelerated Approval Program. FDA, 2026.
- U.S. Government Publishing Office. 21 U.S.C. 356 - Expedited approval of drugs for serious or life-threatening diseases or conditions. United States Code, 2023.
- U.S. Government Publishing Office. 21 CFR 314.510 - Approval based on a surrogate endpoint or on an effect on a clinical endpoint other than survival or irreversible morbidity. Code of Federal Regulations, 2024.
- U.S. Government Publishing Office. 21 CFR 601.41 - Approval based on a surrogate endpoint or on an effect on a clinical endpoint other than survival or irreversible morbidity. Code of Federal Regulations, 2024.
- U.S. Food and Drug Administration. Accelerated Approval and Considerations for Determining Whether a Confirmatory Trial is Underway. FDA, 2025.
- Tibau A, Hwang TJ, Romano A, et al. Factors in Time to Full Approval or Withdrawal for Anticancer Medicines Granted Accelerated Approval by the FDA. JAMA Network Open. 2025;8(3):e252026.
- U.S. Food and Drug Administration. Withdrawn | Cancer Accelerated Approvals. FDA, 2026.
- U.S. Food and Drug Administration. Animal Rule Summary. FDA, 2024.
- U.S. Food and Drug Administration. Priority Review. FDA, 2018.
- U.S. Food and Drug Administration. Real-World Evidence. FDA, 2025.
- U.S. Food and Drug Administration. Considerations for the Use of Real-World Data and Real-World Evidence To Support Regulatory Decision-Making for Drug and Biological Products. FDA, 2023.