The validity and value of cell therapy safety protocol inventions (T 1555/23)

24 views

Skip to first unread message

Rose Hughes

unread,

Mar 2, 2026, 2:58:16 AM (2 days ago) Mar 2

to ipkat_...@googlegroups.com

The validity and value of cell therapy safety protocol inventions (T 1555/23)

Dr Rose Hughes Friday, February 27, 2026 - cell therapy, epo, ip strategy, manufacturing IP, patents, PatKat, pharma ip, Rose Hughes

The patent in T 1555/23 is impressively broad. The claims covered a medical use invention comprising treating the side effects of CAR-T cell therapy using the now industry-standard class of cytokine inhibitor. The claim was not limited to any particular CAR-T cell therapy, target, type of cancer or specific inhibitor. The data supporting the claim, whilst clinical, was also relatively brief, being from only a few patients. Nonetheless, the Board of Appeal found the claim sufficiently disclosed and inventive. However, the inventive story and amendments required to maintain the claim may also have undermined the value of the patent by limiting the scope

The problem of cytokine release syndrome

CAR-T cell therapy is associated with the serious side effect cytokine release syndrome (CRS). CRS occurs when the infused CAR-T cells expand and trigger a massive, uncontrolled release of pro-inflammatory cytokines into the patient's system. Symptoms of CRS include fever, dyspnea, hypotension, and hypoxia. At a molecular level, Interleukin-6 (IL-6) was a known hallmark cytokine of CRS. Managing this toxicity is considered critical because the very cytokines that drive the anti-tumour response can also become life-threatening to the patient. The availability of IL-6 inhibitors is thus considered essential by regulatory authorities for the Risk Evaluation and Mitigation Strategy (REMS) for CAR-T cell therapies.

Cells

Case background: A protocol for IL-6 inhibition

The patent at issue, EP 2872171 (The Trustees of The University of Pennsylvania), related to CAR-T cells for use in treating cancer. Claim 1 of the Main Request (amended claim 1 found allowable at Opposition) is worth reading, for its broadness:

"A T cell genetically modified to express a chimeric antigen receptor (CAR), wherein the CAR comprises an antigen binding domain, a transmembrane domain, and an intracellular signaling domain comprising a 4-1BB signalling domain,

for use in a method of treating a disease, disorder or condition associated with an elevated expression of a tumor antigen,

the method comprising administering a first-line therapy and a second-line therapy to a patient in need thereof, wherein the first-line therapy comprises administering to the patient an effective amount of said genetically modified T cell,

wherein following the administration of the first-line therapy, cytokine levels in the patient are monitored to determine the appropriate type of second-line therapy to be administered to the patient,

wherein the appropriate second-line therapy is a cytokine inhibitory therapy which comprises administering to the patient an effective amount of a cytokine inhibitory compound in order to manage the toxicity resulting from administration of the genetically modified T cell to the patient,

wherein the cytokine is IL-6,

further wherein said monitoring indicates an increase in the level of said cytokine, and wherein the disease, disorder or condition is a cancer."

The claim therefore did not specify any particular CAR-T cell therapy, nor did it define the specific target of the cell therapy. By requiring only that the CAR comprise a 4-1BB costimulatory domain, the claim essentially covered the therapeutic method as applied to all CAR-T therapies using a 4-1BB costimulatory domain. CARs generally use either a 4-1BB or CD28 costimulatory domain, with many on-market therapies using 4-1BB, including the anti-CD19 therapies Kymriah (tisagenlecleucel), Breyanzi (lisocabtagene maraleucel), Carvykti (ciltacabtagene autoleucel) and the anti-BCMA therapy Abecma (idecabtagene vicleucel).

The claim was also not limited to any particular cancer. The dependent claims included a long list of potential cancers. Further the claim was not limited to any specific "cytokine inhibitory compound", and so covered any inhibitor of IL-6. The examples, by contrast, only included clinical data for a specific autologous anti-CD19 CAR-T cell therapy and a specific IL-6 inhibitor (tocilizumab, Actemra). The use of tocilizumab following anti-CD19 CAR-T cell therapy to reduce CRS whilst maintaining treatment efficacy was also only demonstrated in 3 patients. An additional patient who received tocilizumab had no response to the cancer therapy.

The Opponents (mostly Strawman opponents) argued that the invention lacked an inventive step and was not sufficiently disclosed in either the priority document or the application as filed. In particular, they questioned whether a single agent like tocilizumab could manage the complex toxicity arising from CAR-T treatments.

Sufficiency: The credibility requirement for CAR-T cell therapy inventions

The Opponents argued that the first priority document did not disclose a causal link between monitoring IL-6 and treatment with the IL-6 inhibitor. However, the Board of Appeal found that the priority document disclosed generating a profile of soluble factors and carrying out a therapy directed against the elevated factor. The Board of Appeal thus found the claimed therapeutic method to be credible. This finding was based on clinical observations of a patient who improved following tocilizumab administration, which the Board of Appeal took at face value.

The Board of Appeal also dismissed the Opponents' argument that the inherent complexity of CRS and the potential for isolated treatment failures undermined the sufficiency of the disclosure. A major point of debate was the timing of IL-6 inhibition, with the Opponents pointing to the data in the patient in which early administration of the IL-6 inhibitor appeared to interfere with CAR-T cell proliferation. However, the Board of Appeal was not convinced by this argument, noting that "the disclosure of an isolated failure of the CAR-T cell therapy which cannot clearly be linked to the administration of an IL-6 inhibitor does not raise serious doubts".

Interestingly, the Board of Appeal emphasised that for life-threatening diseases like cancer, "a therapeutic effect is not to be equated with a complete remission or cure but also includes prolonging life expectancy and alleviating symptoms", wherein "therapeutic efficacy may be acknowledged even with low percentages of responders". The detailed treatment history of the single patient who developed severe toxicities resistant to corticosteroids but improved "within hours" of receiving tocilizumab was considered credible evidence that toxicity could be managed while maintaining efficacy.

The broad claim was thus found to be sufficiently disclosed with respect to any CAR-T cell therapy (with a 4-1BB co-stimulating domain), any cancer, and any inhibitor of IL-6.

Inventive step (Article 56 EPC) of safety protocols

In assessing inventive step, the Board of Appeal defined the objective technical problem as the provision of an improved CAR-T cell therapy of cancer. The closest prior art described CAR-T cell therapy and the monitoring of IL-6 levels. However, the Board of Appeal found that the prior art did not disclose using the monitoring of IL-6 to determine the appropriate second-line therapy. Instead, treatment decisions in the prior art were typically based on clinical symptoms rather than specific biomarkers.

The Board of Appeal rejected the argument that combining the closest prior art with other documents, such as one suggesting the use of TNF-alpha inhibitors, would lead to the invention. Whilst it was accepted by the parties that IL-6 was known as a hallmark of CRS, the Board of Appeal found that there was no experimental evidence in the prior art that inhibiting it specifically would successfully manage toxicity while preserving the therapeutic effect of the CAR-T cells. The Board of Appeal concluded that the shift from symptom-based management to a precise, biomarker-driven protocol was thus found to be non-obvious.

Final Thoughts

This case is therefore a legally interesting example of the kind of broad therapeutic use inventions that can be protected at the EPO. Surprisingly, the equivalent granted US patents in the same family are just as broad (US 11273219 B2). However, aside from the legal interest, what is the value of a therapeutic claim effectively covering a safety protocol?

Even the broadest patent may be difficult to enforce or irrelevant to real-world therapeutic practice. Interestingly, a critical part of the Board of Appeal's reasoning in this case included interpreting the phrase "cytokine levels in the patient are monitored" as specifically requiring monitoring of IL-6 (r. 30), whilst also coming to a narrow interpretation of IL-6 inhibitor that excluded standard steroids. The claims were amended in opposition so as to be limited specifically to IL-6 out of a long list of potential cytokines. The patent in the same family that was not amended to include the monitoring step was revoked at opposition (EP3338794).

If we consider for a second what infringement of the granted claim might look like, the value of the patent is undermined if CAR-T product labels do not specify the step of monitoring IL-6 levels. None of the EMA or FDA labels for Kymriah, Breyanzi, Carvykti, or Abecma require monitoring IL-6 levels. Some CAR-T cell therapy labels do mention the treatment of CRS with the IL-6 inhibitor tocilizumab. However, CRS is generally monitored and identified by clinical presentation (e.g. fever, hypotension). Diagnostic tests for cytokines can take many hours, whilst measuring fever is instant, and the seriousness of CRS requires immediate IL-6 treatment.

The arguments and amendments required to establish patentability and maintain the patent in this case also potentially rendered the patent valueless. Perversely, given that the patent is not owned by a major marketer of cell therapy products, CAR-T cell therapy manufacturers may also be disincentivised by the patent from employing an IL-6-based diagnostic assessment of CRS, even if there was a clinical reason to pursue this.