Patient’s immune system could help predict success of CAR-T cell therapy

Researchers, including scientists at Queen Mary University of London, have identified biological factors that could help predict whether patients will respond successfully to CAR-T cell therapy when lower numbers of therapeutic cells are available.

The research, published in Nature Communications, found that treatment success was linked not simply to the overall number of CAR-T cells given to a patient, but to the presence of a particular group of highly functional cells.

The researchers also found that important indicators of how well a CAR-T cell product might ultimately perform could already be detected in a patient’s blood before the manufacturing process began.

Professor Simon Haas, of the Precision Healthcare University Research Institute (PHURI) at Queen Mary University of London and the Berlin Institute of Health at Charité (BIH), is co-senior author of the study.

CAR-T cell therapy involves taking a patient’s own immune cells and genetically modifying them in the laboratory so they can recognise and attack cancer cells. The therapies have transformed treatment for some people with advanced blood cancers, but manufacturing sufficient numbers of cells can be difficult and patients do not always respond successfully.

Using high-resolution single-cell analysis, the researchers studied blood samples from patients with blood cancers alongside the CAR-T cell products manufactured from them.

They identified a particular subgroup of CAR-T cells whose abundance was associated with successful treatment at lower doses. These cells had a molecular profile associated with a strong ability to kill tumour cells.

The research also showed that the condition of a patient’s immune system before their cells entered the manufacturing process was important.

Professor Haas said: “Patients whose blood contained larger numbers of functional immune cells were more likely to yield effective CAR-T cell products.”

Conversely, where patients’ original blood samples contained a higher proportion of cells that suppress immune responses, fewer highly functional CAR-T cells were produced and patients were less likely to respond to treatment.

The findings could eventually help clinicians better predict the effectiveness of CAR-T cell therapies and make treatment more personalised. Further research will be needed to validate the potential biomarker and establish how it could be used in clinical practice.

Professor Haas said the findings could also have implications for healthcare resources. CAR-T cell therapies can cost hundreds of thousands of pounds per patient.

“If our findings are confirmed in further studies, CAR-T cell manufacturing could become much more targeted in the future,” he said. “This would benefit not only patients but also help conserve healthcare resources.”

The research was led by researchers from Heidelberg University Hospital, Heidelberg Faculty of Medicine at Heidelberg University and the Berlin Institute of Health at Charité (BIH), with collaborators including Queen Mary University of London, the German Cancer Research Center (DKFZ) and the National Center for Tumor Diseases (NCT).

The study, A distinct CAR-T cell phenotype mediates therapeutic response at limited doses, was published in Nature Communications in July 2026.