Federal Circuit Reinforces Standard for Judicial Correction of Patent Error

The US Court of Appeals for the Federal Circuit reinforced when judicial correction of drafting errors is allowed, emphasizing the potential importance of intrinsic evidence, reasonable debate, and prosecution history. This LawFlash outlines how the Federal Circuit approached the correction and offers practical considerations for patent owners and accused infringers.

In its November 12, 2025 decision,[1] the Federal Circuit reinforced that judicial correction of an error in patent claims remains available where:

1. an error is evident from the face of the patent;
2. the correction is not subject to reasonable debate; and
3. the prosecution history does not suggest a different interpretation.

FACTUAL AND PROCEDURAL BACKGROUND OF THE DISPUTE

Canatex Completion Solutions Inc. (Canatex) sued Wellmatics, LLC and associated entities (Wellmatics or Defendants) in September 2022, alleging infringement of multiple claims (1, 4–13, and 15–19) of US Patent No. 10,794,122 (the ’122 Patent). The ’122 Patent generally relates to a releasable connection for a downhole tool string used in oil-and-gas well operations. [2] The tool comprises two parts: a distal (downhole) part and a proximal (surface-ward) part that can be locked together and then selectively disengaged if the lower portion becomes stuck.[3]

Defendants challenged the validity of the ’122 Patent, arguing that all independent claims containing the phrase “the connection profile of the second part” do not have proper antecedent basis, thus rendering the claims indefinite.[4] In response, Canatex agreed that the required antecedent basis was missing.[5] But Canatex argued that the phrase contained what a relevant artisan would recognize as “an evident error with simple correction, i.e., that a relevant artisan would clearly understand the intended meaning to be ‘the connection profile of the first part.’”[6]

The district court disagreed, holding that “the error is not evident from the face of the patent and the correction to the claim is not as simple as [Canatex] makes it seem,” concluding that the claims were invalid for indefiniteness.[7] The district court further reasoned that the “pervasiveness of the error” in both claims and specification reflected an “intentional drafting choice” combined with Canatex’s failure to seek correction from the US Patent and Trademark Office (USPTO) under 35 USC § 255, suggesting the error was neither minor nor evident on the face of the patent.[8] Thus, the district court rejected Canatex’s plea for judicial correction.[9]

FEDERAL CIRCUIT REVIEW: JUDICIAL CORRECTION DOCTRINE

On appeal, the Federal Circuit conducted a de novo review of the district court’s claim construction.[10] As to the substantive standard, the Federal Circuit explained that judicial correction is allowed only for “obvious minor typographical and clerical errors in patents” where (1) “the error is evident on the face of the patent,” (2) “the correction is not subject to reasonable debate based on the interpretation of the claim language and the specification,” and (3) “the prosecution history does not suggest a different interpretation of the claims.”[11]

The Error Was Evident on the Face of the Patent

The Federal Circuit concluded that the drafting error was evident on the face of the patent.[12] From the outset of its analysis, the Federal Circuit emphasized that the defect was not a matter of competing interpretations or claim-construction nuance but a plain and objectively recognizable mistake that a skilled artisan would immediately identify when reading the patent in its entirety.[13] The disputed phrase “the connection profile of the second part” lacked any antecedent basis, and the claim structure made clear that the only component capable of being “released” by the releasable engagement profile was the connection profile of the first part.[14] Nothing in the specification or figures depicted or described a second-part connection profile, confirming that each reference to one was an inadvertent error rather than a deliberate drafting choice.[15]

Rejection of Alternative Interpretations Resulting in a Finding of No Reasonable Debate

The Federal Circuit held that the only reasonable correction was to change “second” to “first” in the claim language.[16] The Federal Circuit rejected alternative corrections proposed by defendants (e.g., that “tubing connector 22” could be a second-part connection profile, or the drafter intended to add “a” rather than “the”) because none aligned with the claim structure, the figures, or the specification.[17] Because every functional description and embodiment identified the “first part”—and only the “first part”—as possessing the connection profile that could be engaged and released, the Court concluded that the proper correction was not subject to reasonable debate.[18]

Prosecution History Considerations

The Federal Circuit further noted no aspect of the prosecution history suggested a different meaning or otherwise undermined the proposed correction. Nothing in the file history indicated that the patentee intentionally used “second part,” nor did the applicant rely on that language to distinguish prior art or advance any substantive argument. Unlike cases such as Chef America, Inc. v. Lamb-Weston, Inc.,[19] where the prosecution history confirmed that the patentee meant exactly what was written, the record here was silent, providing no basis to infer a deliberate drafting choice.[20]

The Lack of a Certificate of Correction

The Federal Circuit also noted that the lack of an attempt to seek a USPTO certificate of correction under § 255 did not preclude judicial correction, as the administrative standard and its prospective application differ materially from the court’s task of determining the claim’s original and proper meaning.[21] With no contrary evidence in the prosecution history, nothing stood in the way of adopting the single, reasonable correction dictated by the intrinsic evidence.[22]

HOLDING AND REMAND

For these reasons, the Federal Circuit judicially corrected the claim language, specifically changing “second” to “first” in the disputed phrase.[23] The case was remanded for further proceedings consistent with the corrected claims.[24]

PRACTICAL RECOMMENDATIONS AND STRATEGIC IMPLICATIONS

This decision is germane to indefiniteness challenges, judicial correction, and how parties may litigate antecedent-basis errors and specification inconsistencies. Practical recommendations and strategic implications for both patent owners and accused infringers include the following.

For patent owners:

• Drafting errors may be fixable if intrinsic evidence is strong. If an asserted claim has a scrivener’s error (e.g., wrong part name, duplicative word, incorrect reference numeral), judicial correction may be better than avoidance. Patent owners should emphasize consistency across intrinsic evidence, pointing to a single, reasonable correction.
• Build a strong intrinsic record and unified narrative regarding the correction. The court heavily relied on a reasonable interpretation of the claim (i.e., only one “correction profile” existed), figures clearly showing component relationships, and the specification’s description of the parts in a consistent, functional way.[25] As such, patent owners should focus on providing a clear, unified narrative that no other interpretation makes logical and technical sense. Further, when patent drafting, functional descriptions and figures must align in a way that makes the intended meaning unmistakable if an inadvertent error appears.
• Inaction on a certificate of correction is not fatal. Whereas the district court criticized the patentee’s failure to seek PTO correction,[26] the Federal Circuit explained that judicial correction is an independent option and sometimes the most appropriate.[27] Although seeking a certificate of correction may still be advisable, failure to do so is not fatal where intrinsic evidence clearly points to one correction.

For accused infringers:

• Indefiniteness attacks remain viable, at least where multiple plausible corrections exist. The Federal Circuit emphasized that correction is available only if the fix is not subject to reasonable debate.[28] As such, accused infringers should propose multiple, credible interpretations to demonstrate that there is reasonable debate regarding the proposed fix.
• The prosecution history can be a strong tool for advocating against judicial correction. The prosecution history did not help defendants in this case.[29] But, in other cases, an accused infringer may highlight arguments distinguishing prior art, examiner amendments, and statements indicating intentional drafting choices to show that the patentee intended the erroneous language. Accused infringers should look for any prosecution statement that could be cited as showing ambiguity or an alternative meaning.
• The “obviousness” of an error may be challenged by identifying repeated errors. The district court relied on the error appearing twice in the specification and in all independent claims.[30] The Federal Circuit rejected that reasoning here.[31] But, in other cases, repetition may be reframed as evidence of intent (not mistake) or evidence that the intended meaning is not obvious.[32] Stated differently, repeated use may be framed as evidence of a deliberate drafting choice or ambiguity in meaning.