939 F.3d 1375
Fed. Cir.2019Background
- OSI's U.S. Patent No. 6,900,221 claims methods of treating non-small cell lung cancer (NSCLC) using erlotinib (Tarceva); asserted claims 44–46 and 53 have an invention date of March 30, 2000.
- Prior art relied on by the Board: Schnur (discloses a class of EGFR‑inhibiting quinazoline compounds and lists erlotinib among many examples), Gibbs (a review article stating that ZD‑1839 and erlotinib have anti‑cancer activity, with a phrase referencing NSCLC), and OSI’s 1998 Form 10‑K (stating erlotinib targets cancers including NSCLC and had entered Phase II trials).
- The Board held the claims obvious over Schnur combined with Gibbs or OSI’s 10‑K, finding a person of ordinary skill would have had a reasonable expectation of success treating NSCLC with erlotinib.
- The record contained no clinical, preclinical (animal), or in vitro data showing erlotinib’s efficacy specifically for NSCLC, and oncology drug development was highly unpredictable (over 99.5% failure rate for Phase II NSCLC candidates cited).
- The Federal Circuit reviewed factual findings for substantial evidence and reversed: it concluded the Board misread Gibbs and that, on this record, there was insufficient evidence to support a reasonable expectation of success. The court also addressed and rejected (as foreclosed) OSI’s constitutional challenge to IPR application.
Issues
| Issue | OSI's Argument | Apotex/Board's Argument | Held |
|---|---|---|---|
| Whether the Board had substantial evidence to find a reasonable expectation of success (obviousness) | Board lacked substantial evidence; no erlotinib‑NSCLC data; art is unpredictable so expectation of success unsupported | Combination of Schnur with Gibbs or OSI 10‑K would give a skilled artisan reason and expectation to treat NSCLC with erlotinib | Reversed — no substantial evidence of reasonable expectation of success; obviousness not established |
| Whether Gibbs (and its cited refs) show erlotinib has anti‑NSCLC activity | Gibbs does not supply erlotinib‑NSCLC data; Gibbs’ citations (Woodburn, Moyer) do not support erlotinib efficacy in NSCLC; Gibbs author confirmed lack of such data | Gibbs’ language indicates erlotinib has anti‑cancer activity, particularly in NSCLC, supporting expectation of success | Reversed — Board misread Gibbs; cited references do not support erlotinib‑NSCLC efficacy and Gibbs’ author disavowed such data |
| Whether OSI’s 10‑K and Phase I/IND status supply an expectation that erlotinib would work for NSCLC | 10‑K contains no NSCLC‑specific efficacy data; Phase I/IND status does not prove NSCLC preclinical efficacy; high failure rate defeats reasonable expectation | 10‑K’s statement that erlotinib targeted NSCLC and entry into Phase II (implying preclinical data existed) supports a reasonable expectation of success | Reversed — 10‑K is insufficient to show NSCLC‑specific efficacy; cannot reasonably infer success given unpredictability and lack of data |
| Constitutionality of applying IPR to pre‑AIA patents | Retroactive application of IPR to pre‑AIA patents is unconstitutional | IPR is constitutional and applicable to pre‑AIA patents (as defended by government and Fed. Cir. precedent) | Court treated issue as foreclosed by precedent; OSI conceded panel‑stage challenge was foreclosed; no constitutional defect found |
Key Cases Cited
- Intelligent Bio‑Sys., Inc. v. Illumina Cambridge Ltd., 821 F.3d 1359 (Fed. Cir. 2016) (standard of review: Board legal conclusions de novo; factual findings for substantial evidence)
- Regents of Univ. of Cal. v. Broad Inst., Inc., 903 F.3d 1286 (Fed. Cir. 2018) (motivation to combine and reasonable expectation of success are factual inquiries)
- In re Kubin, 561 F.3d 1351 (Fed. Cir. 2009) (obviousness is a legal conclusion based on underlying facts)
- In re Gartside, 203 F.3d 1305 (Fed. Cir. 2000) (substantial‑evidence review requires examining the whole record)
- Dickinson v. Zurko, 527 U.S. 150 (U.S. 1999) (APA requires meaningful judicial review of agency factfinding)
- Anderson v. Liberty Lobby, Inc., 477 U.S. 242 (U.S. 1986) (assessment of what a reasonable factfinder could find guided by substantive evidentiary standards)
- Cuozzo Speed Techs., LLC v. Lee, 136 S. Ct. 2131 (U.S. 2016) (burden in IPR to prove unpatentability is preponderance of evidence)
- Oil States Energy Servs., LLC v. Greene's Energy Grp., LLC, 138 S. Ct. 1365 (U.S. 2018) (IPR constitutionality addressed; cited in discussion of OSI’s constitutional challenge)