Gilead Sciences v. Academy of Military Medical Sciences — UPC Revokes Chinese Military Institute’s Remdesivir COVID-19 Patent for Lack of Inventive Step

Background

In the early days of the COVID-19 pandemic, China’s Academy of Military Medical Sciences (AMMS) — the research arm of the People’s Liberation Army — was among the first institutions to test remdesivir against the novel SARS-CoV-2 virus. On January 21, 2020, just ten days after the virus’s genome was publicly sequenced, AMMS filed a Chinese patent application claiming the use of remdesivir to treat SARS-CoV-2 infections. That application eventually matured into European Patent EP 3,854,403, which the EPO granted on June 18, 2025 with unitary effect across 18 UPC contracting member states.

Gilead Sciences — the company that originally developed remdesivir as a broad-spectrum antiviral — filed a revocation action the same day the patent was granted. The dispute was valued at €20 million, reflecting the commercial significance of what became one of the first authorized COVID-19 treatments worldwide.

The patent’s two independent claims were straightforward: claim 1 covered remdesivir “for use in treating a disease or an infection caused by SARS-CoV-2,” and claim 6 covered a pharmaceutical composition containing remdesivir for the same purpose. Nine dependent claims specified disease types (respiratory disease, sepsis, COVID-19) and formulation details.

The Court’s Holding

The UPC Central Division unanimously revoked the patent in its entirety, finding that the claimed invention lacked inventive step under Article 56 EPC. The court also dismissed AMMS’s main request and all eight auxiliary requests to amend the patent claims.

The court’s analysis centered on two prior art publications — designated HRM 09 and HRM 10 — both published on January 10, 2020, just ten days before the priority date. HRM 09 was a commentary by Professor Sheahan that explicitly addressed the newly discovered Wuhan coronavirus and stated: “Previous publications show that remdesivir works against SARS and SARS-like viruses found in bats, as well as all other coronaviruses we have tested it against. Thus, it is likely to work against coronaviruses that are emerging now in Wuhan.” HRM 10 was a companion editorial titled “Preparing for Future Pandemics, Today with Broad-Spectrum Antivirals” that reached the same conclusion about remdesivir’s likely efficacy against the new virus.

The court defined the technical problem as “finding a drug with antiviral activity against SARS-CoV-2” and asked whether a skilled virologist at the priority date would obviously have turned to remdesivir. Given that the prior art already demonstrated remdesivir’s efficacy against SARS-CoV-1, MERS-CoV, bat coronaviruses, and multiple other RNA viruses — and that SARS-CoV-2’s RNA polymerase shared approximately 96% sequence homology with SARS-CoV-1 — the court found the answer was clearly yes.

Critically, the court distinguished between a “reasonable expectation of success” and a “mere hope of success.” It held that there is a reasonable expectation when scientific data indicate a tested solution can yield a positive result, despite the inherent uncertainty of experimentation. The skilled person, the court emphasized, “does not display fear of failure” and “would not reject a solution due to subjective concerns about possible failure.” Confirming remdesivir’s in vitro efficacy against the new coronavirus was “routine” experimentation, not an inventive act.

Key Takeaways

• Second medical use patents face a high bar in pandemic contexts. When a known broad-spectrum drug is already proven effective against closely related viruses, applying it to a new member of the same viral family is likely obvious — even if the specific virus is novel.
• Speed of filing does not create inventive step. AMMS filed within ten days of the genome’s publication, but the court found that being first to test was not the same as being inventive. The scientific community had already publicly identified remdesivir as the leading candidate.
• The UPC adopted a sophisticated inventive step framework. The decision applied both the EPO’s problem-solution approach and the UPC Court of Appeal’s newer “holistic approach” from Amgen v. Sanofi (UPC_CoA 528/24), noting both should reach the same result. The court articulated a clear “could-would” standard: the question is whether the skilled person would have arrived at the claimed solution, not merely whether they could have.
• The skilled person is objective, not fearful. AMMS argued that a real virologist would have been uncertain about remdesivir’s efficacy against an unknown virus. The court rejected this, holding that the skilled person under Article 56 EPC is a “notional figure” who cannot be identified with any real person and is not influenced by subjective attitudes like fear of failure.

Why It Matters

This is one of the first UPC decisions to revoke a patent related to COVID-19 treatments, and it carries implications well beyond pandemic preparedness. The ruling signals that second medical use patents for known drugs applied to closely related diseases will face rigorous scrutiny in Europe — especially when pre-existing research already points toward the drug as a candidate.

For pharmaceutical companies, the decision reinforces that racing to file a patent application during a health emergency does not guarantee patent protection if the underlying science was already pointing in that direction. The court effectively held that confirming a well-founded scientific hypothesis through routine testing does not constitute an inventive step.

The geopolitical dimension adds further significance: a Chinese military research institute’s patent was challenged and revoked by an American pharmaceutical company before a European court. The case underscores how the UPC is becoming an important venue for global patent disputes, with its decisions having unitary effect across 18 EU member states.

The parties agreed to mutual costs of €800,000, with AMMS bearing the expense as the losing party. The patent revocation eliminates any licensing obligations for remdesivir’s COVID-19 use across all UPC contracting states.