Inter partes review and drug patents

Between September 16, 2012, and July 31, 2025, challengers filed 574 petitions at the Patent and Trademark Office's Patent Trial and Appeal Board against patents listed in FDA's Orange Book, according to the agency’s own study of drug and biologic patent challenges⁵. Another 343 petitions went after patents on biologics. Nearly all of them were inter partes reviews, a trial-like proceeding Congress created in 2011 so the patent office could take a second look at patents it had already granted.

That route has narrowed fast. In fiscal 2026, which ended September 30, the board instituted trials on 41 percent of the petitions it decided, down from 68 percent in fiscal 2024, and only 500 petitions of every kind were filed, 35 of them against biotech and pharmaceutical patents, USPTO statistics show⁷.

For generic and biosimilar makers, inter partes review has been a cheaper and quicker companion to the court fights that the Hatch-Waxman Act sets up. This explainer covers how the proceeding works, how companies on the generic side have used it, what the patent office's Orange Book counts show and what changed at the agency in 2025 and 2026.

How an inter partes review works

The Leahy-Smith America Invents Act¹, signed September 16, 2011, rewrote chapter 31 of the patent code. Under 35 U.S.C. 311², anyone other than the patent owner can ask the board to cancel one or more claims, but only on the ground that the claimed invention was not new or was obvious (sections 102 and 103 of the code), and only on the basis of earlier patents or printed publications. Other invalidity arguments, such as an inadequate written description, still have to be made in court.

Reviews became available on September 16, 2012, and covers patents issued before or after that date, the USPTO says³. The board may start a trial only if the petition shows “a reasonable likelihood that the petitioner would prevail with respect to at least 1 of the claims challenged,” and it must issue a final written decision within a year of instituting, a deadline the director can stretch by six months for good cause.

Two rules matter most to a generic company. The challenger has to prove unpatentability only by a preponderance of the evidence, a lower bar than the clear and convincing evidence a court demands before it will invalidate an issued patent. And a petition is barred if it comes more than one year after the challenger was served with an infringement complaint, so a company sued after sending a Paragraph IV notice has a fixed window. Losing carries a price as well. After a final written decision, the statute bars the petitioner from arguing in court any invalidity ground it “raised or reasonably could have raised” in the review.

Those fee amounts come from the USPTO fee schedule⁴, which lists the same figures regardless of the size of the company filing. Each claim challenged beyond 20 adds $470 at filing and $940 after institution.

Why generic and biosimilar makers use it

Hatch-Waxman cases are tried in federal district court, where a listed patent carries a presumption of validity. Inter partes review offered a panel of technically trained administrative judges and a lower burden of proof. A 2025 commentary in Health Affairs Scholar¹² by law and medicine researchers put the cost of a life-sciences administrative challenge at about $725,000 and 12 months, against about $6.1 million and several years for patent litigation, citing earlier estimates.

Early results drew attention. A 2019 study in Applied Health Economics and Health Policy¹³, which covered petitions through April 2017, found that generic manufacturers had overturned all challenged claims in 43 percent of the patents they targeted and some claims in another 8 percent. The reviews of drug patents were consistently finished within the 12 months the statute allows, and successful challenges most often hit patents on new formulations or methods of use rather than patents on the active ingredient. Those secondary patents make up much of what critics call a patent thicket.

A board challenge usually runs alongside the court case rather than replacing it. FDA's patent certification page¹⁴ describes the 30-month approval delay as lasting unless the patent expires or is “judged to be invalid or not infringed” first, and the events that cut a 30-month stay short are court rulings. More than 80 percent of inter partes reviews have parallel district court litigation in which the petitioner also challenges validity, the USPTO said in an October 2025 proposed rule¹⁰, citing a 2016 academic study.

Biosimilar companies have used the route against patents on biologics, which often number in the dozens per product. The 2025 commentary cited earlier research finding that administrative reviews invalidated 14 of 20 patents that stood in the way of biosimilar entry, and that median revenue for the brand product fell 79 percent in the year after a patent was struck. The biosimilar law of 2010 has its own patent-exchange procedure, which runs in parallel.

What the USPTO's Orange Book counts show

According to the patent office's July 2025 study update⁵, Orange Book patents were the target of 3 percent of the 18,648 petitions filed under the 2011 law, and biologic patents 2 percent. Of the 574 Orange Book petitions, 548 (95 percent) were inter partes reviews. Their share of all petitions peaked at 7.5 percent in fiscal 2016 and fell below 1.4 percent in fiscal 2025.

Panels instituted trials on 62 percent of the Orange Book petitions it decided (307 of 499), close to the 64 percent rate across all technologies. Measured by claims, 17 percent of challenged Orange Book claims and 33 percent of claims that went to trial were found unpatentable. For biologic patents the figures were 27 percent and 59 percent.

GxRadar tallied the agency's patent-level outcome file⁶ for Orange Book patents, which assigns each challenged patent a single outcome. A patent appears in a final-decision row if it ever received one.

In other words, a challenged drug patent was about as likely to come out of a final decision fully intact as fully cancelled, and close to a third never reached trial at all.

What changed in 2025 and 2026

In 2025 the patent office leaned on “discretionary denial,” its power to refuse a review for reasons other than the merits. The Health Affairs Scholar commentary¹² counted 506 requests for discretionary denial decided through September 30, 2025, of which 304 (60 percent) were granted. Drug patents accounted for 42 of the requests, and 14 of those were granted. One new rationale, “settled expectations,” shielded patents that had been in force for more than about six years. On October 17, 2025, the agency's director announced he would personally decide whether to institute every review.

On that same day, the USPTO published a proposed rule¹⁰ that would bar a review unless the petitioner stipulates it will not raise section 102 or 103 invalidity arguments in any other forum, bar reviews of claims already upheld by a court, the International Trade Commission or the board, and bar reviews when a parallel trial is likely to reach a decision first. The comment period was extended¹¹ to December 2, 2025. GxRadar found no final rule in the Federal Register as of early November 2026.

Fiscal 2026 numbers show the effect. Of 782 petitions the director considered on discretionary grounds, 438 (56 percent) were denied and 344 were referred to a panel, according to the September 2026 report⁷. Panels granted 379 of 493 merits decisions. Across the 1,246 petitions that concluded during the year, 47 percent ended with institution denied, while 19 percent ended in a final decision cancelling every challenged claim, the board's year-end outcome roundup⁸ shows.

None of this touches the proceeding's constitutional footing. In 2018 the Supreme Court held¹⁵ that inter partes review does not violate Article III of the Constitution, because reconsidering a patent grant falls within the public-rights doctrine. What has changed is how often the agency agrees to hold one.

October 2026, the first month of fiscal 2027, gives the latest reading. Challengers filed 33 petitions that month, 27 of them inter partes reviews and one against a biotech or pharmaceutical patent, and the board instituted 18 of the 24 petitions it decided, the October 2026 statistics⁹ show. The next marker for drug patents is whether the October 2025 proposal is finalized, and in what form.