Inventive step of CAR T cell inventions at the EPO

Over the last decade, there has been a rapid evolution in the research and development of chimeric antigen receptor (CAR) T cell therapies. As of September 2026, seven CAR T cell therapies are approved by the MHRA in the UK: Yescarta®, Kymriah®, Tecartus®, Breyanzi®, Abecma®, Carvykti®, and Aucatzyl®. The approved CAR T cell products target either CD19 or BCMA for the treatment of different haematological malignancies.

As the clinical development of CAR T cell therapies advances, so does the patent landscape. Early foundational patents were often able to obtain relatively broad claim scope. However, as the volume of patent filings and scientific publications in the field continues to grow, applicants may find it more challenging to establish inventive step because the available prior art often leaves fewer distinguishing technical features upon which an inventive contribution can be based.

Problem-solution approach 

The European Patent Office (EPO) applies the problem-solution approach when assessing inventive step. This involves the following steps (Guidelines for Examination, G‑VII, 5):

1. Determining the closest prior art;
2. Establishing the objective technical problem to be solved; and
3. Considering whether the claimed invention, starting from the closest prior art and the objective technical problem, would have been obvious to the skilled person.

Establishing the objective technical problem

In establishing the objective technical problem, it is necessary to identify the distinguishing features of the claimed invention over the closest prior art and determine the technical effect associated with those distinguishing features. The technical problem may thus be formulated as the aim and task of modifying or adapting the closest prior art to achieve the technical effect that the invention offers over the closest prior art. In practice, it is often the presence or absence of a technical effect upon which the patentability of CAR T cell claims turns.

To demonstrate an alleged technical effect, the EPO generally requires evidence linking the distinguishing feature to an improved technical effect over the closest prior art. Comparative data against the most relevant prior art embodiment are therefore often particularly persuasive. This evidence may be found in the application as filed or filed as a supplementary submission provided that “the skilled person, having the common general knowledge in mind, and based on the application as originally filed, would derive [the technical] effect as being encompassed by the technical teaching and embodied by the same originally disclosed invention” (G 2/21). 

A useful example can be found in the prosecution of EP3227432 (Memorial Sloan Kettering Cancer Center and Eureka Therapeutics, Inc.). In a response filed on 12 October 2020, the claims were limited to defining the specific heavy chain variable region and light chain variable region sequences of the ET140-54 CAR. The applicant also submitted additional experimental evidence demonstrating that the claimed ET140-54 CAR did not bind soluble BCMA, in contrast to the closest prior art CAR construct. By establishing a technical effect associated with the claimed CAR, the applicant was able to overcome the inventive step objection. In addition, in both T 0203/22 (Trustees of Dartmouth College) and T 1555/23 (The Trustees of The University of Pennsylvania and The Children's Hospital of Philadelphia), the Technical Boards of Appeal found that the applicants had demonstrated a technical effect associated with the claimed CAR constructs. Specifically, in T 0203/22, the technical effect identified related to higher IFN-γ production as a marker of T-cell activation, and improved survival rate of mice after tumour challenge. While in T 1555/23, the technical effect identified related to reduced serum cytokines in a patient. Accordingly, the objective technical problems were formulated as the provision of an “improved” CAR. 

However, even where a technical effect is accepted for a specific CAR embodiment, a separate question is whether that effect has been demonstrated across the whole scope of the claim. The EPO will therefore generally not accept that a technical effect demonstrated for a single CAR construct automatically applies across a broad genus of CARs unless there is a credible technical reason why the effect should extend throughout the claim scope. This issue arose during examination of EP3896091 (Legend Biotech Ireland Limited). In a communication dated 18 October 2021, the Examining Division acknowledged that the application demonstrated a favourable efficacy and safety profile for the specific bivalent BCMA CAR construct, LCAR-B38M, which comprises two anti-BCMA VHH domains that target different BCMA epitopes. However, claim 1 was not limited to that particular CAR. Claim 1 instead defined a CAR comprising an extracellular antigen binding domain comprising a first and second anti-BCMA single VHH domain antibody (the first having three CDRs of a specific VHH domain and the second not being defined), a transmembrane domain, and an intracellular signaling domain. The Examining Division thus concluded that the effect could not be extrapolated to each and every CAR falling within the scope of claim 1, and the claim as a whole was considered to lack inventive step. A similar approach was taken during the prosecution of EP3227432 (communication dated 2 April 2020) and EP3126390 (communication dated 30 July 2017; Cellectis). 

Furthermore, where no evidence can be provided of an alleged technical effect, the problem must be reformulated as the provision of an “alternative” product (i.e. a product that is structurally different from the prior art, but which displays no functional difference from the prior art). For example, during examination of EP3896091 (communication dated 18 October 2021), the Examining Division argued that the mere cloning of an alternative CAR comprising two anti-BCMA VHH domains is regarded as routine and therefore did not involve an inventive step. Similarly, during examination of EP4004055 (communication dated 20 February 2025; Janssen Biotech, Inc.), the Examining Division argued that the subject matter of the claims appeared to involve the routine combination of CARs and humanized anti-Kallikrein 2 antibodies known in the state of the art, resulting in a predictable outcome, and therefore did not involve an inventive step. 

CAR T cell inventions are particularly vulnerable to this reasoning. This is because, for a claimed CAR T cell construct, the antigen target may already be known, the antigen binding domains (e.g. scFvs) against the target may already exist, standard signalling domains (e.g. CD28, 4-1BB, CD3ζ) are well known, and routine molecular biology methods can be used to assemble constructs. As a result, in the absence of evidence of a technical effect, the EPO may regard the claimed CAR as an obvious combination of known elements. This approach is confirmed in the prosecution histories of both EP3149044 (The U.S.A. as represented by the Secretary, Department of Health and Human Services) and EP3227432. In communications dated 22 November 2017 and 2 April 2020 respectively, the Examining Divisions referred to decision T 735/00 (Iatron Laboratories, Inc.), which states that if there are no unexpected effects achieved with a further monoclonal antibody compared with a monoclonal antibody with essentially the same properties as desired, the case law denies inventive step. Although T 735/00 concerned antibodies rather than CAR constructs, the Examining Division considered that analogous principles could be applied to the claimed CAR inventions. Thus, the claims were found to lack inventive step. 

In view of the examples above, it appears that the EPO frequently regards the construction of an alternative CAR from known components as routine. However, it may be possible to argue for inventive step where the skilled person would not have had a reasonable expectation of obtaining the claimed CAR T cell construct.

Practical considerations

Recent prosecution practice before the EPO suggests that inventive step objections directed at CAR constructs increasingly turn on whether applicants can demonstrate a technical effect attributable to the distinguishing features over the closest prior art. Where such an effect is convincingly established, claims directed to specific CAR constructs or architectures may be patentable. Conversely, where no technical effect can be shown, the objective technical problem is often reduced to the provision of an alternative CAR construct, which the EPO may regard as an obvious implementation of known components.

To maximise the prospects of establishing inventive step before the EPO, applicants should thus consider:

• demonstrating a clearly defined technical effect such as improved efficacy, improved persistence, reduced exhaustion, improved safety, or enhanced selectivity;
• including comparative experimental data against the closest prior art CAR construct;
• providing evidence supporting the full breadth of the claims; 
• including sufficient disclosure of the technical effect in the application as filed to permit reliance on post-filed evidence where necessary;
• drafting applications with multiple fall-back positions directed to specific CAR constructs, binding domains, signalling architectures, or experimentally validated embodiments.

Taken together, recent EPO practice highlights the increasing importance of experimental evidence in support of CAR T cell patents. As the field continues to mature and the prior art landscape becomes more crowded, the ability to demonstrate a technical effect arising from the distinguishing features of a claimed CAR construct is likely to remain a critical factor in establishing inventive step.

Subscribe to receive more articles like this here.