The ANDA Process: A Step-by-Step Guide to Generic Drug Approval

An Abbreviated New Drug Application (ANDA) is the U.S. Food and Drug Administration (FDA) pathway, under section 505(j) of the Federal Food, Drug, and Cosmetic Act, for approving a generic drug without repeating the brand product’s clinical trials. Instead of proving safety and efficacy from scratch, the applicant shows the generic is therapeutically equivalent to the reference listed drug (RLD): pharmaceutically equivalent (same active ingredient, strength, dosage form, and route of administration) and bioequivalent (the same rate and extent of absorption).

Development runs through nine connected stages, from product selection and reverse engineering the RLD through API sourcing, formulation, analytical work, bioequivalence, stability, manufacturing, and dossier submission. Under the third Generic Drug User Fee Amendments (GDUFA III), the FDA aims to act on a standard ANDA within ten months, though most applications need more than one review cycle. Formulation, bioequivalence, facility, and analytical deficiencies are the usual causes of delay [1, 2, 16].

Table of Contents

• What Is an ANDA?
• ANDA, NDA, and 505(b)(2): How They Differ
• Step 1: Product Selection and FDA Guidance
• Step 2: Characterizing the Reference Listed Drug
• Step 3: API Sourcing and the Drug Master File
• Step 4: Formulation Development and Excipient Selection
• Step 5: Analytical Methods and Validation
• Step 6: Demonstrating Bioequivalence
• Step 7: Stability Studies
• Step 8: Manufacturing and the Pre-Approval Inspection
• Step 9: Compiling and Submitting the ANDA
• The ANDA Process at a Glance
• Complex Generics: When the Standard Path Changes
• Why ANDAs Get Delayed or Rejected
• FAQ
• Key Takeaways
• Sources

What Is an ANDA?

An ANDA is the application a company submits to market a generic version of an already-approved drug. The pathway, created by the 1984 Drug Price Competition and Patent Term Restoration Act (the Hatch-Waxman Act), lets applicants rely on the FDA’s prior finding that the RLD is safe and effective, so an ANDA contains no new preclinical or clinical safety and efficacy trials [1]. It does contain full chemistry, manufacturing, and controls (CMC) data plus evidence of bioequivalence.

Approval rests on therapeutic equivalence, which combines two requirements:

• Pharmaceutical equivalence: the same active ingredient, strength, dosage form, and route of administration.
• Bioequivalence: no significant difference in the rate and extent of absorption [2].

The FDA records its conclusions in the Orange Book, assigning an “A” code to products it considers substitutable for the brand and “B” to those it does not. An ANDA that meets both halves of the standard earns an A rating, which underpins generic substitution at the pharmacy [2].

See and download our infographic on The ANDA Process:

ANDA, NDA, and 505(b)(2): How They Differ

The ANDA is one of three FDA approval routes, distinguished by how much new clinical evidence each requires. A New Drug Application (NDA) under 505(b)(1) carries full clinical trials, an ANDA under 505(j) substitutes bioequivalence, and a 505(b)(2) application falls in between, relying partly on data the applicant does not own [1].

Table 1. Comparison of FDA drug application pathways.

Attribute	NDA, 505(b)(1)	505(b)(2)	ANDA, 505(j)
Clinical safety and efficacy trials	Full	Partial, may reference others’ data	None required
Basis of approval	Applicant’s own data	Own plus referenced data	FDA findings for the RLD
Must match the RLD	No	No, modifications allowed	Yes: same active ingredient, strength, form, route
Key evidence generated	Clinical	Bridging studies	Bioequivalence

See our video on the ANDA process:

 

Step 1: Product Selection and FDA Guidance

Product selection is the first stage and the point at which many programs are won or lost. The applicant identifies a target RLD in the Orange Book, confirms the designated reference standard used for bioequivalence testing, and reviews patents and exclusivity.

For each Orange Book-listed patent, the ANDA files one of four certifications:

• Paragraph I: none listed.
• Paragraph II: patent expired.
• Paragraph III: the generic will await patent expiry.
• Paragraph IV: the patent is invalid, unenforceable, or not infringed [3].

A Paragraph IV certification can trigger litigation but can also earn the first qualifying applicant 180 days of marketing exclusivity [3]. Early in development, applicants follow the FDA’s product-specific guidance (PSG) for the drug, which states the bioequivalence studies the agency expects, and may use controlled correspondence or a pre-ANDA meeting to resolve questions before submitting [18].

Step 2: Characterizing the Reference Listed Drug

RLD characterization is the analytical reverse engineering of the brand product so the generic can be built to match it. The applicant determines:

• the active pharmaceutical ingredient (API) and its solid-state form;
• the qualitative and, where deducible, quantitative excipient composition;
• the dissolution behavior; and
• the impurity and degradation profile [4].

The label lists inactive ingredients only qualitatively, so quantities, grades, and particle sizes must be inferred through analysis. Strong characterization here prevents the most expensive failure mode in generic development: a product that looks similar on paper but fails bioequivalence in the clinic.

 

Step 3: API Sourcing and the Drug Master File

The API is the same active ingredient as the RLD, but the generic manufacturer rarely makes it in-house. It is sourced from suppliers who each file a Type II Drug Master File (DMF) with the FDA describing the substance’s manufacturing, controls, and impurity profile; the ANDA references the DMF, and the two are reviewed together [5].

Quality work centers on:

• impurity control to International Council for Harmonisation (ICH) Q3A thresholds;
• control of elemental impurities under ICH Q3D; and
• a nitrosamine risk assessment, now a standard expectation that also considers excipient-derived nitrite [20].

Because a single API deficiency can stall an entire ANDA, applicants often qualify a second source.

 

Step 4: Formulation Development and Excipient Selection

Formulation development designs the generic to perform like the RLD, and excipient selection is its most consequential decision. The objective is to match the reference product’s dissolution, physical stability, and in vivo performance using inactive ingredients suited to the dosage form and supported by precedent [6].

How closely the excipients must match depends on the route:

• Oral solids: the FDA allows different excipients if the product is bioequivalent and ingredient levels are established as safe.
• Parenteral products: 21 CFR 314.94 generally requires the same inactive ingredients at the same concentration as the RLD, with narrow exceptions for preservative, buffer, and antioxidant.
• Ophthalmic and otic products: a similar same-ingredient, same-concentration expectation applies [7].

This is the basis of Q1/Q2 sameness:

• Q1 (qualitative): the same inactive ingredients as the RLD.
• Q2 (quantitative): each within plus or minus 5 percent of the RLD concentration [8].

Even where sameness is optional, every excipient should stay within the maximum potency for its route in the FDA Inactive Ingredient Database (IID), since exceeding precedent requires added safety justification [6]. Compatibility studies between the API and each excipient close out the stage.

 

Step 5: Analytical Methods and Validation

Analytical methods measure the product’s identity, strength, purity, and performance, and they must be validated before they generate regulatory data. The core methods are:

• an assay for drug content;
• impurity and related-substance methods;
• a dissolution method; and
• stability-indicating methods.

For immediate-release oral products, dissolution profiles are compared using the similarity factor f2, where a value of 50 or higher indicates similarity [9]. Validation follows ICH Q2, evaluating accuracy, precision, specificity, linearity, range, robustness, and detection and quantitation limits [10]. The FDA accepts data only from validated methods, so a weak method can undermine an otherwise sound package.

 

Step 6: Demonstrating Bioequivalence

Bioequivalence shows the generic delivers the active ingredient to the bloodstream at the same rate and extent as the RLD. The standard design for a systemically absorbed oral product is a single-dose, randomized, two-period, two-sequence crossover study in healthy volunteers, often under fasting and fed conditions [11].

It measures two parameters:

• AUC (area under the concentration-time curve): total exposure.
• Cmax (maximum concentration): rate of absorption.

The product passes when the 90 percent confidence interval of the generic-to-reference ratio for both AUC and Cmax falls entirely within 80.00 to 125.00 percent on log-transformed data [11]. Highly variable drugs use reference-scaled limits.

Not every product needs a clinical study:

• Under the Biopharmaceutics Classification System (BCS), some immediate-release oral drugs qualify for a biowaiver based on comparative dissolution [12].
• A lower strength can often be approved without its own study when it is proportionally similar in composition and matches the tested strength on dissolution.
• Locally acting products, such as many topicals and inhalation drugs, may instead require comparative clinical endpoint or in vitro studies.

 

Step 7: Stability Studies

Stability studies establish the product’s shelf life under defined storage conditions. Following ICH Q1A(R2), applicants generate two data sets:

• long-term data at 25 degrees Celsius and 60 percent relative humidity; and
• accelerated data at 40 degrees Celsius and 75 percent relative humidity [13].

FDA guidance recommends data from at least three primary batches, with the test product made at an exhibit-batch scale of at least one-tenth of the proposed commercial batch or 100,000 dosage units, whichever is greater, for solid oral forms [14]. The data support the proposed expiry date and storage statement on the label.

 

Step 8: Manufacturing and the Pre-Approval Inspection

Before approval, the generic must be shown to be reproducible at commercial scale under current good manufacturing practice (CGMP). The applicant makes exhibit batches, then validates the commercial process to confirm it consistently meets specifications [19]. The proposed manufacturing site, the API supplier, and the testing facilities must all be in a state of CGMP readiness, because facility status is one of the pillars the FDA assesses on the GDUFA review clock.

For many applications the FDA conducts a pre-approval inspection of the finished-dose and API sites; an inspection that finds significant deficiencies is a common reason an otherwise complete ANDA is not approved on the first cycle. Confirming site readiness early protects the review timeline.

 

Step 9: Compiling and Submitting the ANDA

The final stage assembles every result into the ANDA dossier. Submissions use the electronic Common Technical Document (eCTD), required for ANDAs since 2017, in five modules:

• Module 1: regional administrative information.
• Module 2: summaries.
• Module 3: quality and CMC data.
• Module 4: nonclinical data (generally not applicable).
• Module 5: clinical and bioequivalence data [15].

The dossier also includes proposed labeling, which must match the RLD’s aside from permitted differences; an applicant may use a “section viii” carve-out to omit an indication still protected by patent or exclusivity, producing a so-called skinny label [7].

The FDA first decides whether the application is complete enough to review; major omissions cause a Refuse to Receive decision. Under GDUFA III (fiscal years 2023 to 2027), the agency aims to act on a standard ANDA within ten months, or eight for a priority application with a timely facility correspondence [16]. A final action can be one of:

• Approval.
• Tentative approval, when a patent or exclusivity blocks final approval.
• Complete Response Letter, listing deficiencies to resolve in a new cycle.

 

 

The ANDA Process at a Glance

Table 2. The nine stages of the ANDA process.

Step	Stage	Objective
1	Product selection	Pick a feasible RLD, clear patents, and follow the FDA’s product-specific guidance
2	RLD characterization	Reverse engineer the reference product’s composition and performance
3	API sourcing	Secure a qualified API with a Type II DMF and controlled impurities
4	Formulation and excipients	Match performance within IID and Q1/Q2 limits
5	Analytical methods and validation	Build and validate assay, impurity, and dissolution methods
6	Bioequivalence	Show AUC and Cmax fall within 80.00 to 125.00 percent
7	Stability	Establish shelf life under ICH Q1A(R2) conditions
8	Manufacturing and inspection	Validate the commercial process and pass the pre-approval inspection
9	ANDA submission	File the eCTD dossier and obtain FDA approval

 

Complex Generics: When the Standard Path Changes

Some products do not fit the standard crossover-study model, and these complex generics are a growing share of unapproved targets. Complexity can lie in:

• the active ingredient (peptides, mixtures, naturally derived substances);
• the formulation (long-acting injectables, liposomes, suspensions, gels);
• the route or delivery system (inhalation, ophthalmic, transdermal, nasal sprays); or
• a drug-device combination.

For these, a pharmacokinetic study alone often cannot establish equivalence, so the FDA’s product-specific guidance may call for additional evidence: comparative in vitro tests (for example, in vitro release testing for topical semisolids), comparative clinical endpoint or pharmacodynamic studies, and device and human-factors data [7, 11, 18]. Excipient selection carries even more weight here, because for many complex products the inactive ingredients must match the reference closely on a Q1/Q2 basis. Early engagement with the FDA through controlled correspondence is the norm for these programs.

Why ANDAs Get Delayed or Rejected

Most ANDAs do not gain approval in the first review cycle, and the number of deficiencies the FDA identifies correlates with the number of cycles a dossier takes [17]. The recurring problem areas are:

• bioequivalence (failed studies or dissolution mismatches);
• formulation (excipient levels above IID precedent);
• CMC and manufacturing gaps;
• DMF deficiencies on the API side;
• inadequate stability data;
• pre-approval inspection findings; and
• labeling errors.

Because each cycle adds months, first-cycle quality is the largest lever on time to market. Median approval times have improved under GDUFA but still run near 20 months in recent FDA reporting [17].

FAQ

How long does the ANDA process take?

The FDA aims to act on a standard ANDA within ten months of submission under GDUFA III, but that covers only agency review. Development, API sourcing, and bioequivalence and stability studies usually take several years beforehand, and because most applications need more than one cycle, median approval times have recently run near 20 months.

What is the difference between an ANDA and an NDA?

An NDA seeks approval for a new drug with complete clinical evidence of safety and efficacy. An ANDA seeks approval for a generic copy of an approved drug, contains no new clinical trials, and instead proves therapeutic equivalence to the reference listed drug through pharmaceutical equivalence and bioequivalence.

What are Q1 and Q2 sameness?

Q1 (qualitative) sameness means the generic uses the same inactive ingredients as the reference listed drug; Q2 (quantitative) sameness means each is within plus or minus 5 percent of the reference concentration. Sameness is generally required for parenteral, ophthalmic, and otic products under 21 CFR 314.94.

How is bioequivalence demonstrated?

For most oral drugs, a crossover pharmacokinetic study in healthy volunteers must show the 90 percent confidence interval for AUC and Cmax falls within 80.00 to 125.00 percent of the reference. Some products qualify for a dissolution-based biowaiver, while locally acting products may need clinical endpoint or in vitro studies.

Why are ANDAs often delayed?

Most ANDAs need more than one review cycle. Common deficiencies include failed bioequivalence studies, dissolution mismatches, excipient levels above IID precedent, CMC and stability gaps, DMF issues, and findings from the pre-approval inspection. Each adds months, so first-cycle quality is the main lever on time to market [17].

What is the Orange Book?

The Orange Book, formally Approved Drug Products with Therapeutic Equivalence Evaluations, is the FDA’s public list of approved drugs and their therapeutic equivalence ratings. Generic developers use it to identify the reference listed drug, find the reference standard for bioequivalence testing, and review the patents and exclusivity periods that gate approval.

What are complex generics?

Complex generics are generic drugs that are difficult to copy or evaluate because of a complex active ingredient, formulation, route, or delivery device, such as inhalers, long-acting injectables, topicals, and drug-device combinations. They often cannot rely on a standard pharmacokinetic study, so the FDA’s product-specific guidance may require in vitro, clinical endpoint, or device-comparison evidence instead.

Key Takeaways

An ANDA wins approval by proving therapeutic equivalence to the reference listed drug, pharmaceutical equivalence plus bioequivalence, rather than repeating clinical trials.

Bioequivalence passes when the 90 percent confidence interval for AUC and Cmax sits within 80.00 to 125.00 percent of the reference.

Excipient selection is the highest-leverage formulation decision; parenteral, ophthalmic, and otic generics generally require Q1/Q2 sameness under 21 CFR 314.94, within IID limits.

Early alignment with the FDA’s product-specific guidance, and a nitrosamine risk assessment, are now standard parts of planning an ANDA.

Facility readiness and the pre-approval inspection are frequent first-cycle blockers, so the FDA’s ten-month GDUFA goal usually spans more than one review cycle.

Complex generics (complex actives, formulations, routes, or devices) often need alternative bioequivalence evidence and closer excipient matching, so early FDA engagement is the norm.

Sources

1. U.S. Food and Drug Administration, Abbreviated New Drug Application (ANDA), FDA, 2025. https://www.fda.gov/drugs/types-applications/abbreviated-new-drug-application-anda (accessed 2026-06-09).
2. U.S. Food and Drug Administration, Approved Drug Products with Therapeutic Equivalence Evaluations (Orange Book), FDA, 2026. https://www.fda.gov/drugs/drug-approvals-and-databases/approved-drug-products-therapeutic-equivalence-evaluations-orange-book (accessed 2026-06-09).
3. U.S. Food and Drug Administration, Patent Certifications and Suitability Petitions, FDA, 2025. https://www.fda.gov/drugs/abbreviated-new-drug-application-anda/patent-certifications-and-suitability-petitions (accessed 2026-06-09).
4. U.S. Food and Drug Administration, ANDA Submissions – Content and Format Guidance for Industry, FDA, 2019. https://www.fda.gov/media/125312/download (accessed 2026-06-09).
5. U.S. Food and Drug Administration, Drug Master Files (DMFs), FDA, 2025. https://www.fda.gov/drugs/forms-submission-requirements/drug-master-files-dmfs (accessed 2026-06-09).
6. U.S. Food and Drug Administration, Inactive Ingredient Database (IID), FDA, 2026. https://www.fda.gov/drugs/drug-approvals-and-databases/inactive-ingredients-database-download (accessed 2026-06-09).
7. Electronic Code of Federal Regulations, 21 CFR 314.94 – Content and Format of an ANDA, U.S. Government, 2026. https://www.ecfr.gov/current/title-21/chapter-I/subchapter-D/part-314/subpart-C/section-314.94 (accessed 2026-06-09).
8. U.S. Food and Drug Administration, Considerations When Preparing the Q1/Q2 Sameness Information for an ANDA, FDA, 2024. https://www.fda.gov/media/186015/download (accessed 2026-06-09).
9. Diane J. Burgess et al., Scientific and Regulatory Standards for Assessing Product Performance Using the Similarity Factor, f2, The AAPS Journal, 2015. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4365094/ (accessed 2026-06-09).
10. International Council for Harmonisation, Q2(R2) Validation of Analytical Procedures, ICH, 2023. https://database.ich.org/sites/default/files/ICH_Q2-R2_Guideline_2023_1130.pdf (accessed 2026-06-09).
11. U.S. Food and Drug Administration, Bioequivalence Studies With Pharmacokinetic Endpoints for Drugs Submitted Under an ANDA, FDA, 2021. https://www.fda.gov/media/87219/download (accessed 2026-06-09).
12. U.S. Food and Drug Administration, M9 Biopharmaceutics Classification System-Based Biowaivers, FDA, 2021. https://www.fda.gov/media/148472/download (accessed 2026-06-09).
13. U.S. Food and Drug Administration, Q1A(R2) Stability Testing of New Drug Substances and Products, FDA, 2003. https://www.fda.gov/media/71707/download (accessed 2026-06-09).
14. U.S. Food and Drug Administration, ANDAs: Stability Testing of Drug Substances and Products, FDA, 2014. https://www.fda.gov/media/87051/download (accessed 2026-06-09).
15. International Council for Harmonisation, M4 The Common Technical Document, ICH, 2016. https://www.ich.org/page/ctd (accessed 2026-06-09).
16. U.S. Food and Drug Administration, GDUFA Reauthorization Performance Goals and Program Enhancements Fiscal Years 2023-2027, FDA, 2021. https://www.fda.gov/media/153631/download (accessed 2026-06-09).
17. Bhargav Vyas et al., Focusing on First Cycle Approval in ANDA Submission: Understanding Common Deficiencies and Case Study Insights, Therapeutic Innovation and Regulatory Science, 2025. https://link.springer.com/article/10.1007/s43441-025-00755-5 (accessed 2026-06-09).
18. U.S. Food and Drug Administration, Product-Specific Guidances for Generic Drug Development, FDA, 2026. https://www.fda.gov/drugs/guidances-drugs/product-specific-guidances-generic-drug-development (accessed 2026-06-09).
19. U.S. Food and Drug Administration, Process Validation: General Principles and Practices, FDA, 2011. https://www.fda.gov/media/87425/download (accessed 2026-06-09).
20. U.S. Food and Drug Administration, Control of Nitrosamine Impurities in Human Drugs, FDA, 2021. https://www.fda.gov/media/141720/download (accessed 2026-06-09).

This article is for informational purposes for pharmaceutical industry professionals and does not constitute regulatory advice. Always refer to the current pharmacopoeial monograph, the supplier’s current technical data sheet, and applicable regulatory guidance for your dosage form, route of administration, and market. Pharma Excipients International AG is not a manufacturer of the excipients discussed.