It is now possible to modify the antigen specificity of T cells through gene transfer with chimeric antigen receptors, and to increase their response still further with the addition of a co-stimulatory domain.

This technique can be used to produce T cells that respond to tumour-specific antigens, but its usefulness is limited by the fact that these antigens are not well defined in most tumour types. However, the B-lymphocyte antigen CD19, which is expressed only in normal and malignant B cells and their precursors, is considered a useful target in B-cell leukaemias and lymphomas.

David Porter and his colleagues at the Perelman School of Medicine, University of Pennsylvania, Philadelphia, USA have designed a self-inactivating lentiviral vector to introduce such a chimeric antigen receptor, the CART19 receptor, combined with the CD3-zeta and the co-stimulatory CD137 signal transduction domain into T cells.

Following pre-clinical evaluation, they established a pilot clinical trial to assess the safety and efficacy of this novel therapy in B-cell cancers, using autologous T cells.

The initial trial involved three patients; the researchers have now published results from one of these as a case report.

The male patient concerned was first diagnosed with chronic lymphoid leukaemia (CLL) in 1996. His disease was stable without treatment for six years and remained well controlled with chemotherapy for another seven. By 2009, his disease had progressed to refractory, advanced p53 deficient CLL.

In July 2010, T cells previously collected from the patient were thawed, transfected with the chimera-expressing lentivirus and infused back into the patient with a low dose of approximately 1.5 x 105 cells / kilogram of body weight. There were no initial toxic effects from the infusion.

The main side effect from this treatment was grade 3 tumour lysis syndrome, which presented with fever, chills, nausea and diarrhea and was diagnosed 22 days after infusion.

The patient was hospitalised and made a full recovery. By 28 days after infusion, the patient's bone marrow showed no evidence of CLL, with normal karyotypes and no evidence of p53 deletion.

When the paper was written, the patient had remained in remission for ten months; the only other moderate or severe side effect was lymphopenia. Hypogammaglobulinemia was observed as expected, but this was mild and easily correctable.

The patient's serum and bone marrow were tested for immune responses throughout the trial, and a temporary increase in the inflammatory cytokines interferon-g, interleukin-6, CXCL9 and CXCL10 was observed that peaked at approximately the same time as the clinical symptoms of tumour lysis syndrome.

T cells expressing the CART19 receptor were observed from the first day after infusion, with the peak concentration indicating a more than 1000-fold expansion of these cells from the infused levels. The genetically modified T cells persisted in the patient's bone marrow for at least six months and continued to express the receptor throughout that time. Neither normal CD19-expressing B cells nor leukaemia cells could be observed in the bone marrow.

In discussing their work, Porter and his colleagues noted that T-cell responses do not depend on a patient's HLA molecules, so these treatments should be widely applicable and will not need patients to be HLA typed before treatment.

They also noted the persistence of both the patient's T-cell response and his remission, suggesting from their previous pre-clinical studies that that might result from the addition of the co-stimulatory CD137 domain.

However, they also note that this persistent response, itself, may lead to B cell depletion and suggest that patients may become susceptible to infection. They recommend from this small trial and case study that investigation of the use of anti-CD19 targeted T cells to treat B-cell neoplasms should be continued.

DOI: 10.1056/NEJMoa1103849