Growing T Cells That Remember More Than the Obvious Target A Culture Recipe for Cell Therapy

This application covers ways of making a batch of a patient’s own T cells that recognise chosen targets, such as viral or tumour proteins, for use in cell therapy. The claimed method starts from a mixed blood cell population, stimulates it with target antigens and cytokines, grows it, tests it for antigen-specific reactivity and harvests the result. It was filed by Geneius Biotechnology, Inc. and names Alfred E. Slanetz, Terry Y. Nakagawa and Marissa A. Herrman as inventors. It is a divisional of Australian application 2023254998, which is itself a divisional of 2017290119.

The Problem

Growing a patient’s T cells outside the body and putting them back, called adoptive cell transfer, has worked against viruses such as cytomegalovirus and Epstein-Barr virus since the 1990s, as the specification recounts. Against tumours it has been less consistent. The specification argues that over the course of a cancer the immune response narrows onto a few dominant antigens that do not drive tumour shrinkage, and that cells selected for strong recognition of those antigens are mostly aimed at the same dominant targets. The tumour can then lose the antigen or stop responding, and the benefit is temporary.

The specification cites a study in which tumour-infiltrating lymphocytes correlated with regression when they reacted to gp100 but not when they reacted to MART-1, and notes that immune suppression by regulatory T cells can blunt the response. Its summary of the position is short: there remains a need for better adoptive T cell therapies.

What This Invention Does

Claim 1 is a method for making a composition comprising T cells. The steps are to obtain an initial cell population containing T cells, stimulate the T cells by exposing the population to one or more target antigens and to cytokines, culture the population in media containing cytokines, test it for antigen-specific reactivity, and harvest the resulting composition. The claims name no particular disease in claim 1.

The dependent claims add detail. The cytokines can be IL-2, IL-7, IL-15 or IL-21, with IL-7 and IL-15 as the preferred pair. Reactivity can be tested by flow cytometry for activation markers or by cytokine release assays such as intracellular staining, ELISA or ELISPOT. The population can be split and each part stimulated with a different antigen, then recombined. The antigens can be overlapping peptide sets from sub-dominant antigens, neoantigens or viral proteins, and claims 17 and 18 name the Epstein-Barr virus proteins LMP1, LMP2 and EBNA1 and the cytomegalovirus protein pp65, along with NY-ESO-1 and Survivin.

Claim 22 is a second, separate method. It obtains the initial population, selects T cells on expression of activation markers, applies polyclonal stimulation and harvests. In the examples the activation markers CD137 and CD25 are used to sort rare antigen-responsive cells from the culture around day 7 to 11, using a cell sorter, and the sorted cells are returned to culture. Later claims cover treating patients with the products, including claim 39 for non-Hodgkin’s lymphoma, gastric cancer or nasopharyngeal carcinoma using T cells reactive to Epstein-Barr virus antigens, and claim 40 for glioblastoma using cells reactive to pp65, NY-ESO-1 and Survivin. Claim 41 is a product claim for a composition of more than half a million CD3 positive cells, over 70 per cent CD3 positive T cells, mostly CD8 positive and mostly effector memory cells.

The worked examples give the figures. In the process scale example, peripheral blood cells from a healthy donor were pulsed with peptide mixes for each of three antigens, pooled, grown with IL-7 and IL-15 at 10 ng/ml, restimulated at day 7, activated with a CD3, CD28 and CD2 reagent at day 14 and harvested at day 28. The yield was over 2 billion viable cells, more than 95 per cent CD3 positive, with a product that was about 63 per cent CD8 positive. Flow cytometry of one day 28 harvest showed 98.7 per cent CD3 positive cells, 62.5 per cent CD8 and 33.5 per cent CD4, with 12.5 per cent expressing CCR7 and 53.0 per cent expressing CXCR3. In a small scale test, 79.0 per cent of one donor’s culture stained for an LMP2 specific pentamer by day 11. The specification also reports that stimulating with all three peptide mixes together suppressed the response to EBNA1, which led to pulsing each mix separately and then combining the cells.

Key Features

• Antigen plus cytokine stimulation. T cells are stimulated with target antigens and cytokines, then grown in cytokine media, rather than being expanded without regard to what they recognise.
• Reactivity testing built in. The method includes a test for antigen-specific reactivity before harvest, with a release threshold of over 70 per cent CD3 positive cells in the examples.
• Peptide pools and sub-dominant targets. Overlapping peptides, including from sub-dominant antigens and neoantigens, are meant to avoid expanding only the dominant clones.
• Pulse then pool. Each peptide mix can be applied separately and the cells combined, which the specification introduced after seeing one peptide mix crowd out the response to another.
• Sorting on activation markers. The second method selects cells that carry activation markers such as CD137 and CD25 and follows with polyclonal stimulation to boost rare responders.
• Product and treatment claims. The later claims cover giving the cells to patients, and a composition defined by cell counts, CD3 fraction, CD8 dominance and effector memory phenotype.

Who Is Behind It

Geneius Biotechnology, Inc. is an immuno-oncology company based in Natick, Massachusetts. A BioSpace announcement of its 2023 AACR poster describes a personalised RNA-enabled T cell therapy platform and names Alfred E. Slanetz as chief executive and founder, and as the presenter. The company also describes its own clinical collaborations there. Those statements come from the company’s own press release and are not independently verified here, and nothing in this specification mentions that platform.

On priority, the specification states that it claims priority to US Provisional Applications 62/355,458, 62/355,506 and 62/355,553, all filed 28 June 2016, so the priority country is the United States. It also states that it is a divisional of Australian application 2023254998, itself a divisional of 2017290119. The title page names FPA Patent Attorneys of Melbourne as the attorney.

Why It Matters

Cell therapy for solid tumours has struggled partly because it is hard to make a product that targets more than one thing and does not exhaust itself. This application is about the manufacturing step: how to take a patient’s blood cells, steer them toward several antigens at once, check that they respond, and harvest a defined, mostly CD8 positive population. Making that reproducible and testable at the 2 billion cell scale is the practical question a manufacturer would care about.

The document reports laboratory results on healthy donor cells and a handful of cell cultures. It does not report patient outcomes, and the treatment claims are method claims that depend on the manufacturing claims. The title and abstract speak of compositions, while claim 1 is a method of making them, and a composition claim appears only at claim 41.

Related Concepts

• Adoptive cell transfer – the therapy approach of expanding a patient’s immune cells outside the body and returning them.
• Tumour-infiltrating lymphocytes – the tumour-derived T cells whose dominant-antigen bias the background criticises.
• Epstein-Barr virus – the virus whose LMP1, LMP2 and EBNA1 proteins are the main example targets.
• Interleukin-15 – one of the two cytokines, with IL-7, used to grow the cells.
• Glioblastoma – the brain tumour that claim 40 aims to treat with cells reactive to pp65, NY-ESO-1 and Survivin.

---

AU 2026202260 was published in the Australian Official Journal of Patents on 9 April 2026 and is open for public inspection. Patent applications represent inventions that are sought to be protected and do not necessarily reflect commercially available products.