How Generics Challenge Patents: A Hatch-Waxman Act Guide

Pharmaceutical patents grant brand‐name drug companies decades of marketing exclusivity, enabling them to recoup R&D costs. However, high drug prices and public health concerns have driven policymakers to create pathways for timely generic entry (^[12]). Before 1984, generic manufacturers had no safe way to prepare or market generics while a brand’s patent life remained in force, leading to “at risk” launches and costly injunctions. The Drug Price Competition and Patent Term Restoration Act of 1984 (the Hatch-Waxman Amendments) instituted a balance: generics could prepare for market entry early and even challenge patents, but had to follow formal procedures under the Food, Drug, and Cosmetic Act (FDCA). Specifically, the Hatch-Waxman Act added parts of 21 U.S.C. §355 to allow abbreviated NDAs (ANDAs) based on bioequivalence, linking FDA approval to patents listed by brand sponsors (^[3]) (^[2]).

Under this regime, a brand must submit to FDA (and thus the Orange Book) any active ingredient or composition patents or method-of-use patents included in its new drug application (^[3]). This list of patents becomes the starting point for generic challenge. The FDA’s role is “ministerial” – it publishes patent data provided by the NDA-holder but does not judge patent validity (^[13]). Controversy has arisen over allegedly improper listings: in late 2023 the U.S. FTC sent warning letters to dozens of drugmakers to delete “improper” Orange Book patents (^[14]) (^[15]), claiming these bog down generic competition. The FTC’s 2023 “Orange Book Challenge” targeted over 400 patents on dozens of drugs, illustrating federal interest in ensuring that patent linkage does not unfairly delay generics (^[14]) (^[15]).

In return, generics gain a shortened approval process and can lawfully use the brand’s data. Specifically, 21 U.S.C. §355(j) allows a generic (a subsequent applicant) to rely on the FDA’s finding of safety and efficacy for the reference listed drug (RLD), conducting only bioequivalence studies (^[16]) (^[17]). The generic still must meet good manufacturing practices and labeling requirements, but can avoid duplicating clinical trials. Critically, the generic must include one of four patent certifications with its ANDA (^[2]) (^[18]). As described below, filing a Paragraph IV certification – that the patent is invalid or not infringed – is the prescribed statutory way for a generic to legally challenge the brand’s patent in court while seeking FDA approval (^[1]) (^[6]).

Hatch-Waxman ANDA Pathway and Patent Certifications

The cornerstone of the generic entry process is the Abbreviated New Drug Application (ANDA) under 21 U.S.C. §355(j). Under this provision, an ANDA applicant must demonstrate that its drug has the same active ingredient, dosage form, route, and strength as the listed drug, and is bioequivalent (^[16]) (^[17]). In its ANDA, the generic must make one of four possible paragraph certifications concerning each patent listed in the Orange Book for the RLD (^[2]):

• Paragraph I: “No patent information on this drug product or method of use is listed in the Orange Book.” (Generic claims no patents apply.)
• Paragraph II: “Each patent on the Orange Book has expired.” (Generic waits for expiration.)
• Paragraph III: Generic will not market its product until all listed patents expire; i.e., it does not challenge the patents.
• Paragraph IV: Generic asserts that the listed patent(s) is invalid or will not be infringed by its product (i.e., a challenge).

An ANDA with a Paragraph I or II certification is typically approved immediately (at the first possible date) because no contested patent issue remains (^[2]). A Paragraph III ANDA must wait until patent expiry. Crucially, a Paragraph IV certification is the mechanism for a generic to legally challenge a patent. The Hatch-Waxman statute explicitly provides that a Paragraph IV certification constitutes an act of patent infringement triggering a lawsuit, but only after following regulatory notice rules (^[18]) (^[5]). In practice, when a generic files a Paragraph IV ANDA, the law requires the applicant to notify (by certified mail) the NDA holder and patent owner within 20 days of FDA receiving the ANDA (^[4]). Receipt of this notice starts a 45-day window during which the patent owner can file suit for infringement (^[5]). If the patent owner sues within that period, the FDA is barred from approving the ANDA for up to 30 months, unless the lawsuit is resolved earlier (^[5]) (^[6]). This “30-month stay” is a statutory patent linkage provision linking regulatory approval to pending litigation. If no suit is filed within 45 days, the FDA may approve the ANDA immediately or soon thereafter (subject only to patent terms expiring).

We outline the four certification pathways in Table 1 below. This illustrates how each route affects the approval timeline:

Table 1: ANDA patent certification types (FDA ANDA process). Paragraph IV is the pathway by which a generic formally challenges patents while still seeking FDA clearance. (^[1]) (^[2]).

Orange Book Patent Listing and Regulatory Exclusivities

A critical precondition for a Paragraph IV challenge is that the patent in question must be listed in the FDA’s Orange Book for the reference drug (^[3]). By statute, only patents that “claim the drug” (i.e. cover the active ingredient or dosage form) or “method-of-use” patents are eligible for listing (^[3]). Process patents (e.g., manufacturing steps), packaging, metabolites, and intermediates are explicitly excluded by regulation (^[19]). The Orange Book thus serves as the official link between FDA approval and patent affairs. Brand sponsors have a strong incentive to list any and all applicable patents to delay approval, while generics in turn must inspect these listings carefully.

The FDA’s role in patent listings is purely ministerial. FDA does not independently verify or examine the patents submitted by NDA holders (^[13]). It relies on patent owners to submit complete and accurate listings and merely publishes them. This has drawn scrutiny: for example, a Congressional report notes that the FDA “lacks expertise” in patent law and tables for stakeholders the different viewpoints, including calls for clearer listing rules (^[13]). The FTC has challenged allegedly improper Orange Book listings by sending warning letters and filing disputes with FDA. Under FDA's dispute process, the brand manufacturer has 30 days to withdraw or amend the listing, or certify that it complies with applicable requirements; some companies have voluntarily delisted patents in response. (^[20]) In late 2023 the FTC announced it had challenged over 100 Orange Book patents as improper and sent warning letters for dozens of drugs, and expanded a second round of challenges in April 2024 covering more than 300 additional patents on inhalers, EpiPens, and other devices (^[14]) (^[15]). In Teva Branded Pharmaceutical Products v. Amneal (Fed. Cir. Dec. 20, 2024), the Federal Circuit affirmed delisting of five Teva ProAir HFA inhaler-device patents, holding that patents that do not claim the active ingredient or an approved method of use are not properly listable under §505(j) — the first appellate ruling to set substantive limits on Orange Book listings. These actions underscore how patent listing is central to generic entry: a patent must be listed for generics to certify it – and generics cannot use the Paragraph IV process if the patent‐holder declines to list the patent. (Brand patent holders can still sue for infringement of unlisted patents under general patent law, but those patents do not trigger an ANDA stay or certification requirement.)

In addition to patents, FDA grants separate periods of regulatory exclusivity that temporarily bar generic approval. For example, a New Chemical Entity (NCE) receives 5 years of data exclusivity, during which no generic ANDA relying on that data may be approved. Orphan-drug exclusivity lasts 7 years, and pediatric exclusivity generally adds 6 months to existing patents and exclusivities. These exclusivities run concurrently with (and do not prolong) the patent term, but they can delay generic approvals independently of patent litigation. The combined effect of patents and exclusivities means that “competition-free” periods for new drugs typically extend 12–14 years after FDA approval (^[21]); only after those monopolies lapse can generics enter absent a successful patent challenge.

Safe Harbor and Experimental Use Exemption

The U.S. patent law itself grants a broad exemption for drug development. Specifically, 35 U.S.C. §271(e)(1) provides a “safe harbor” such that it is not patent infringement to make, use, or sell a patented invention “solely for uses reasonably related to the development and submission of information” to the FDA (^[8]). This statutory exemption codifies what courts had declined to do (the narrow “experimental use” exception) and was added precisely to benefit generic drug research and development. The Supreme Court in Merck KGaA v. Integra (2005) interpreted this language very broadly, holding that even preclinical research involving patented compounds is protected if it will support future drug approval. Thus, generics may synthesize a patented active ingredient and run bioequivalence tests without infringing, so long as it’s for FDA filings. This safe harbor is key: it ensures generics can prepare for marketing (and even challenge patents) without being enjoined for simply conducting required studies (^[8]). It has been further extended by courts to cover post-approval quality testing, compounding for IND studies, and even exploratory work by university researchers related to drugs (^[8]).

Interaction of Patent Law and FDA Regulation

Beyond the ANDA mechanism, generics have other limited avenues to challenge patents. For instance, under the Biologics Price Competition and Innovation Act (BPCIA) (2009), an analogous process exists for biosimilars, involving a “patent dance” and exchange of lists. This acts like a parallel patent linkage for biologic drugs, though biosimilars are typically outside the scope of small-molecule generics. More broadly, a generics company might file a declaratory judgment action after developing the drug, but FDA’s safe harbor acts as a safe way to tie the challenge to the ANDA process, making paragraph IV the standard route.

Legal battles often ensue from the Hatch-Waxman scheme. Brands frequently sue generics under 35 U.S.C. §271(e)(2) when a Paragraph IV certification is filed, seeking preliminary injunctions or patent holds. Courts interpret 271(e)(2) to say that filing an ANDA with a Paragraph IV is itself an act of infringement (by statute), which triggers jurisdiction. A successful generic in these suits can invalidate the patent or show non-infringement and get approval. Famous cases include Merck v. Biovail, Teva v. AstraZeneca (development of generic Nexium), and Warfarin v. Apotex – all illustrating the tight interlock of FDA and patent statutes. The Supreme Court in FTC v. Actavis, Inc. (2013) also ruled that certain “pay-for-delay” settlements between brands and generics can violate antitrust laws, although such settlements are often structured to skirt Hatch-Waxman judicial review. Throughout, both regulatory and patent statutes jointly define when and how generic firms may lawfully contest a patent: essentially only by using the FDA‐approved pathways (ANDA filings with appropriate certifications or, in Canada, NOAs) and relying on the safe harbor for research.

Besides the U.S., numerous countries have integrated patent status into drug approval processes – a concept known as patent linkage. These regimes vary but share the goal of balancing early access with patent rights. Table 2 summarizes the key features of linkage in several jurisdictions:

Table 2: International examples of patent linkage schemes. In each case, generic sponsors must navigate the national patent-based delays or litigation requirements to obtain marketing authorization. For instance, Canada’s PM(NOC) Regulations provide that if the brand sues on a patent within 45 days of the generic’s notice, Health Canada cannot issue approval for up to 24 months (^[10]). By contrast, the EU has no linkage, so generics file after patent/exclusivity expiration and deal with patents through courts (as seen in recent SPC cases like apixaban (^[25])).

Canada: As noted, Canada’s linkage system is highly structured. The Patent Act’s “early working” exemption (§55.2) allows generic (subsequent-entry) companies to use the patented invention solely for regulatory purposes, but the PM(NOC) Regulations then impose a check on actual marketing. A “Notice of Allegation” (NOA) must be served on the Patent Register’s first person (brand) whenever the generic seeks approval before patent expiry (^[9]). The brand can file an infringement action within 45 days; if it does, Health Canada issues a 24-month “Intellectual Property (IP) Hold” blocking any Notice of Compliance (NOC) to the generic (^[10]). Only after litigation concludes (or if the brand fails to sue) can the generic obtain its NOC. This mechanism effectively mirrors the U.S. stay system, though the timelines (45 days to sue, 24 months of delay) and procedural details differ (^[9]) (^[10]).

European Union: In the EU, patents and regulatory approval are largely separate. The EU’s pharmaceutical directives (e.g. 2001/83/EC) focus on safety/efficacy, not patents. An MA holder may seek SPC extensions beyond 20 years to compensate for regulatory delay. After 8 years of data exclusivity, generic manufacturers may submit applications; market protection generally prevents placement on the market until 10 years after the reference product’s initial authorization. Patent and SPC enforcement remains separate from the authorization process and is pursued through litigation. Notably, after approval has been granted, a patent holder can still seek interim injunctions to block marketing – a tactic used by originators in Europe. A recent Spanish case illustrates this: Teva’s generic apixaban (anticoagulant) was on the market, but in 2024 a Barcelona court retroactively reinstated BMS’s SPC on the drug, resurrecting the monopoly and forcing the generic to be withdrawn (^[26]). Such cases show that even without formal linkage, generics cannot safely market if patent validity is in question, since national courts may impose injunctions. In practice, generics must clear patent hurdles separately (often via lengthy court PICs) and cannot rely on an FDA-like patent certification process in the EU.

Japan: Uniquely, Japan has a quasi-linkage administered by the Ministry of Health (^[27]). A Guideline (“Notification”) from 2009 (updated 2025) bars generic approval altogether if any patent claims the active ingredient (^[11]). In effect, generics are only approved for non-patented uses or after patents expire. If a compound patent on the API exists, the MHLW simply refuses approval of the generic product (^[11]). This is an administrative filter rather than litigation: the Ministry decides patent overlap. (Critics note this lacks transparency, since MHLW staff are not patent experts (^[28]).) In practice, Japanese generics enter much later, effectively giving brand products an extended monopoly beyond patent life (especially for molecules with broad composition patents). There is no 30-month stay or formal notice; rather, generics must design around or wait out patents. This system is often cited in trade talks: Japan strongly resisted adopting a US-style linkage in trade agreements, defending its approach as consistent with WTO law.

Other Regimes: Many other countries have partial linkage. For example, under South Korea’s 2015 patent notification law, generics must notify patent holders and await 30 days; if sued, the approval is delayed (though stays are shorter than in North America). Australia’s Therapeutic Goods Act uses patent certificates for specified applicants relying on another person’s safety or efficacy data. An applicant proposing pre-expiry marketing must notify the patentee, and a patentee commencing infringement proceedings must certify good faith, reasonable prospects of success, and that the proceedings will not be unreasonably delayed; the Act does not establish a 45-day automatic approval stay. Trade agreements (like USMCA, formerly NAFTA) often require some form of drug-patent linkage among signing countries, prompting changes in Canada, Mexico, etc. In each case, the specifics (length of stay, courts involved, etc.) vary, but these frameworks govern how a generic applicant addresses patents while seeking marketing approval, commonly through a certified application or notice. They do not foreclose other patent-challenge procedures: a person other than the patent owner may petition for USPTO inter partes review on the statutory grounds and subject to the statutory requirements for that proceeding (^[29]).

Empirical studies shed light on how these frameworks play out. For instance, recent research shows that larger-market drugs are much more likely to attract patent challenges. A 2025 study of 210 new small-molecule drugs (FDA-approved 2007–2018) found that 55% of them faced at least one generic patent challenge within one year of eligibility (^[30]). Market value was by far the strongest predictor: high-revenue drugs almost invariably prompt Paragraph IV filings, whereas drugs with small sales often see no challenge. Indeed, that study notes that “nearly half of all new drugs are unlikely to see a patent challenge from a generic competitor” (^[31]). This empirically confirms that generic challengers target profitable franchises (often leading to a cascade of litigation on “blockbuster” drugs) and that many patents moot small-market drugs may never be tested. The implication is that weak or trivial patents may endure unexamined unless there is sufficient commercial incentive to contest them (^[30]) (^[31]).

FDA publishes a Paragraph IV Certifications List that identifies drug products for which one or more substantially complete ANDAs containing a Paragraph IV certification have been submitted; the list also provides information relevant to 180-day exclusivity for listed products (^[32]).

The cost impact of generic entry is dramatic. In the U.S., generic prescribing accounts for roughly 90% of all retail prescriptions (reflecting the ubiquity of off-patent medicines) (^[12]). Every successful challenge and subsequent launch can sharply reduce prices. A concrete example: Spain estimated that Teva’s generic apixaban (Eliquis) was saving the public health system over €380,000 per day in 2024 (^[25]). When that generic was halted by patent litigation, the court noted the public would lose those savings. This underscores the stakes – regulatory and legal battles over generics have massive budgetary implications for health systems and patients. (In fact, cost-savings studies often cite generic competition as essential to sustaining healthcare affordability, which is why regulatory frameworks aim to prevent frivolous patent enforcement from unjustly delaying entry.)

United States: Field Reports

• Clopidogrel and Plavix: In a landmark case (BMS/Sanofi v. Apotex), a Canadian generic launched clopidogrel (Plavix) immediately after a court invalidated a secondary patent. The generic captured market share until settlement. Later, U.S. generics Teva and others challenged Pay-for-Delay deals over Plavix patents, illustrating both ANDA battles and antitrust scrutiny.
• Insulin Patents: Small-molecule analog insulin patents have been repeatedly challenged. Time magazine (2023-11) and FTC reports document how competitors systematically infringe claims with generics. Currently, the FTC and DOJ are reviewing insulin brand patent listings as “misuse” (^[33]), indicating governmental pushback on expansive patent strategies.
• Dedicated Hatch-Waxman Battles: Many first-to-file generics spend years in court. The FDA’s Paragraph IV list shows hundreds of filings. For instance, Teva’s launch of generic simvastatin (Zocor) and generic atorvastatin (Lipitor) involved early Paragraph IV filings and long court profiles, demonstrating a playbook in which generics often litigate vigorously to invalidate key patents. First generic launches, including launches by applicants eligible for 180-day exclusivity, are closely watched industry events.

Canada: PM(NOC) in Action

An illustrative Canadian example is the case of Baycol (cerivastatin): in 1998, Canadian generic companies sought approval of a cerivastatin generic by serving NOAs. Bayer (the first person) sued and obtained a 24-month NOC hold, delaying generics until Bayer’s patent expiry in 2002. Similarly, the 2006 Apotex v. Janssen case (citalopram/Cipramil) saw Apotex successfully invalidating patents after the full NOA litigation timeline, finally winning some market share. These show how the Notice of Allegation mechanism forces generics to litigate patents before marketing, achieving a definite resolution albeit at discrete intervals.

European Union: SPC Litigation

As noted, EU generics lack a centralized challenge process. Instead, patent battles often hinge on Supplementary Protection Certificates (SPCs). More recently, the Eliquis (apixaban) dispute illustrates this point: Teva launched an apixaban generic after initial court rulings invalidated AstraZeneca/BMS SPCs. However, in 2024 a Spanish appellate court reversed that, renewing BMS’s SPC and forcing Teva’s product off the market (^[26]). This highlights how European generics, though not regulated to prevent listing patents, remain subject to post-approval patent validity fights with massive economic stakes (the Spanish public health system lost an estimated €138 million/year in savings when the generic was removed (^[26])).

Other EU cases involve forum shopping in patent litigation. For example, generics might launch in one country and get injunctions elsewhere. The EU’s Unified Patent Court (once operational for medicines) may eventually centralize some disputes. Meanwhile, several EU antitrust enforcers have fined originators for misuse of the regulatory framework: for instance, Teva’s €462M fine (2024) for anticompetitive behavior around Copaxone (^[34]) shows that even indirect abuses (spreading false information to stall a rival) are being sanctioned, a phenomenon tied to the tension between patents and competition.

• FTC’s 2023-24 Patent Review: The FTC’s active challenges of Orange Book patents illustrate regulatory concern. By late 2023, FTC staff notified companies of dozens of patents it considered “improperly listed” (e.g., covering devices or methods not tied to the drug itself) (^[14]). For example, patents on EpiPen injector devices were contested. Some companies responded by voluntarily delisting challenged patents; others intend to defend them. These actions remain administrative (the FTC cannot force delisting, but can bring antitrust suits), yet they signal enforcement scrutiny of the patent linkage system.

• Legislative Initiatives: U.S. policymakers have proposed or enacted changes to facilitate generic challenges. The CREATES Act (codified in 2019) addresses anticompetitive delays in generic approval by, for example, forcing brand companies to provide samples if certain obstacles are used. The Inflation Reduction Act (2022) added a new dimension by allowing Medicare to negotiate prices on selected drugs, with the first round of negotiated prices for ten high-spend Part D drugs taking effect January 1, 2026 — a development that interacts with patent-challenge incentives because already-negotiated drugs offer smaller upside to first-filer generics. More recent bills aim to cut generic approval timelines and to reform citizen-petition abuses. The bipartisan Affordable Prescriptions for Patients Act (^[35]) was reported in the Senate on April 10, 2025; it addresses infringement of patents claiming biological products, so its patent-assertion provisions concern biologic and biosimilar disputes rather than Hatch-Waxman ANDA litigation. Globally, trade treaties under negotiation often include provisions about patent extensions and linkage. Such policy shifts will influence how quickly generics can contest patents.

• Innovation vs. Competition Debate: The tension between protecting patent rights and fostering competition is contentious. Industry advocates (as reflected in a recent CSIS analysis) warn that weakening patent protections could harm innovation and cede ground to foreign competitors (^[23]). They point out that U.S. entities filed about 38% of global biotech patents from 2015–2020 (^[23]); the argument is that robust enforcement, including patent linkage, helps sustain that lead. Conversely, consumer advocates argue that blocking “evergreening” patents will improve patient access, citing cases where trivial patents delayed generics. The evolving legal framework thus must strike a balance.

The current legal framework has produced a predictable cycle: generics wait 4–5 years post-approval to file, challenge patents by ANDA/NOA, and often spend years in court. Looking forward, several implications and trends stand out:

• Accelerated Competition on Big-Market Drugs: Studies indicate generic challenges cluster on high-revenue drugs (^[30]). As biologics and specialty drugs dominate sales, expect increased lobbying to either open pathways for biosimilars (via changes to the BPCIA) or extend exclusivities. For high-demand small-molecule drugs, the Hatch-Waxman framework will remain central, but petitions to FTC/DOJ, legislative amendments, or rulemakings (for example redefining “bioequivalence” or reforming citizen petitions) could adjust the balance.

• Regulatory Reform and Enforcement: Both the Biden and second Trump administrations have signaled interest in reducing patent barriers, though through different levers. FDA’s Drug Competition Action Plan, FTC’s ongoing Orange Book policy statements (continued under Chair Andrew Ferguson in 2025), and Medicare drug-price negotiation under the IRA all suggest closer scrutiny of patents that delay generics. The Federal Circuit’s 2024 Teva v. Amneal decision now gives generics a clearer judicial mechanism — counterclaims under 21 U.S.C. §355(j)(5)(C)(ii) — to force delisting of improperly listed patents. Administrative tools like the Purple Book for biologics and international reference pricing might also pressure patents’ value. Future policy could include more explicit review of patent listings (perhaps codifying FTC’s ad-hoc challenges), faster dispute resolution, or incentivizing patent clearinghouse reviews.

• Global Harmonization and Tensions: Internationally, generic access remains uneven. Countries like India and Brazil might continue using TRIPS flexibilities (compulsory licenses, parallel imports, strict patentability standards) to open markets, which could invite trade disputes. Conversely, major markets (US, EU, Canada) will likely stick to patent-linkage plus selective reforms. The December 2019 USMCA revisions removed the concept of "hard" patent linkage from the trade-agreement template, while allowing Mexico to continue its linkage system under a new annex, illustrating these pressures (^[36]). The rise of R&D in China and other markets adds another dimension: patents in those jurisdictions often grant stronger exclusivity, pushing generics to forum-shop or delay launches outside Western markets.

• Economic and Clinical Impact: Every change in the framework will have cost and health effects. Accelerating generics can save governments billions; delaying them preserves brand revenues (supporting R&D funds) but at higher price for payers. Courts and policymakers will continue to weigh institutional priorities: ensuring challengers have a clear path (through headquarters filing, pre-litigation mechanisms) versus preventing opportunistic challenges to still-viable patents. Biopharmaceutical firms will adapt their patenting and litigation strategies accordingly, potentially leading to more patent clusters (“evergreening”) if litigation barriers are raised, or to leaner patent portfolios if accelerated challenges become easier.

The legal and regulatory scaffolding for generic patent challenges is intricate and multi-faceted. In the U.S., it is defined by the Hatch-Waxman ANDA process, Paragraph certifications, and court litigation under patent law – with concepts like the safe harbor and FDA’s Orange Book serving as unique tools. Abroad, each jurisdiction imposes its own timing and procedural requirements, from Canada’s NOA regime to Japan’s administrative patent-review process. In the U.S. Paragraph IV process, the generic applicant certifies and gives notice; the NDA holder or patent owner may then bring an infringement action within 45 days. A generic applicant's declaratory-judgment options are limited by statute. This framework can resolve patent disputes before market entry, or permit entry when the relevant requirements are satisfied. (^[32]) (^[37])

The balance achieved – protecting genuine innovation while expediting competition – has been dynamic. Empirical evidence confirms that the current system spurs generic entry in lucrative markets but also leaves some patents untested (^[30]) (^[31]). Ongoing reforms by regulators and legislators aim to fine-tune this balance. As the pharmaceutical landscape evolves (with more complex products, specialty drugs, and international supply chains), the patent‐linkage framework will likewise adapt. Future developments may involve more streamlined dispute resolution, changes in how patents are listed and challenged, and perhaps new incentives for generics. However accomplished, any changes must be scrutinized: as one policy analysis warns, weakening patent protections could undermine the incentives that support drug innovation (^[23]), even as we seek to reduce prices and expand access through timely generic competition.

In sum, a robust body of law and regulation – from 21 U.S.C. §355 and §271(e) in the U.S., to analogous statutes in Canada, to international trade obligations – governs when and how a generic drug can legally contest a patent. This framework has been tested in countless court cases and policy debates. Its contours will continue to shape the lifecycle of every off-patent drug, influencing who gets to make generic versions, when they can make them, and under what conditions, all with major economic and public-health stakes (^[1]) (^[10]).