Science Spotlight

The rare make the most

from the Chapuis lab (Clinical Research Division)
The figure represents the frequency clonal evolution of dominant clone from autologous T cell product in vivo.
Individual TCRs were identified in the infused T cell product and their evolution was tracked in vivo in two patients: P2225-1 or P2225-7 that showed complete remission (CR) following adoptive T cell therapy. The frequencies of dominant clones (red on the left panel, purple on the right panel) match the frequencies of clones not detected in the pre-infused cell population (blue line with circle symbols), indicating that the dominant clones that were detected at very low frequency ex vivo present some survival and proliferation advantage in vivo following infusion and become dominant relative to other clones (colored lines). Each color represents an individual clone from the infused T cell product. Figure adapted from the publication and courtesy of Dr. Aude Chapuis.

Adoptive T cell immunotherapy is a promising way to treat tumor patients. In this approach, autologous (self) T cells reactive to the tumor antigen can be isolated from the patient’s blood and expanded ex vivo to achieve greater number of potent tumor-specific reactive immune cells called cytotoxic T cells (CTLs). The expanded cells are then infused back into the patient. In studies performed in the early 2000, patients were initially infused with a monoclonal cell population (single cell clone). However, such products took months to expand and were not persistent in vivo. A more recent approach consists in the expansion and infusion of a heterogeneous cell population or polyclonal T cells, reducing the ex vivo culture time while increasing the cell number in the final product (1, 2, 3). This approach has resulted in better persistence of the infused T cells infused.

Of all the thousands of T cells infused, characterizing the T cells that ultimately persist in the patients, identifying which clonotypes are associated with complete remissions and discern what could have been the origin of these ‘super’ cells could ultimately inform what subset of cells should be targeted for future trials and increase the success rate of adoptive immunotherapy. In a recent study from Dr. Aude Chapuis (Clinical Research Division) in collaboration with colleagues from the Fred Hutch, University of Washington and MD Anderson Cancer Center, high-throughput deep sequencing was used to track the T cells after in vivo infusion in patients. The results of this study were published in Science Immunology.

Individual T cell clones from a polyclonal population can be distinguished from each other by the sequence of their T cell receptor (TCR). The TCR is a protein presented at the cell surface and responsible for binding to the tumor antigen. Dr. Chapuis and her colleagues used high-throughput T cell receptor Vβ sequencing (HTTCS) to identify and track clones and their frequency in patient’s T cell isolate. This technique allows detection of very rare clones with a high sensitivity of 1 of 100 000 (0.001%).

Samples from 10 melanoma tumor patients who were infused with polyclonal T cell product were sequenced to determine TCR profiles. A broad range of diversity was observed with 56 to 2036 individual clones being identified in the different samples. The most prevalent clone represented between 4 and 77% of the population depending on the patient.

Two patients presented complete remission following infusion of the adoptive T cells. In both cases a single clone dominated the population frequency in vivo (0.056% of the total T cells in the circulating blood at 280 days post-infusion for one patient and 0.093% at 175 days for the second patient) and represented most of the tumor-antigen specific T cells. Further analyses of the dominant cells’ half-lives revealed that cells present in patients with complete remissions lived longer, followed by cells in patients with partial remissions and stable disease. In patients with progressive disease, the dominant clones had very shorter half-lives.

For all patients, the clonotypes identified in vivo post-infusion that composed the bulk of persistent clones corresponded to very rare clones in the pre-infusion T cell samples (detected at a frequency <0.001% or below detection levels). The clonotypes that ultimately persisted and which presented a proliferation and survival advantage in vivo after transfer, came from very rare, barely detectable populations in the patient before any therapy was performed.  Although it was impossible to definitely determine from this study, evidence suggests that these clonotypes might have predominantly been from ‘young’ T cell populations such as the naïve T cell pool (ie. cell type that has never encountered a tumor antigen as opposed to effector T cells). As such, the impact of ex vivo culture conditions could have different outcomes on cells from naïve, active or memory phenotypes and impact the quality of the T cells infused.

Additional studies will be necessary to develop culture conditions favoring greater expansion of these rare T cells with shortened ex vivo expansion period and improved anti-tumor efficiency. In regards to the future of the project, Dr. Chapuis said “from this study we figured out the cells that can control the tumor are rare. However, here we relied on the autologous cells that were present in the patient. Moving forward we are looking into gene-therapy approaches in which the most efficient cell type (or substrate cell) is selected and isolated from the patient, and then a highly-avid TCR that has more likelihood to control tumor is inserted. The two patients in this study that underwent complete remissions are a great model to understand what the ideal T cell infusion product should be made of. We are now investigating whether selecting T cells with a “younger” phenotype that can be identified by the CD62L ligand would allow us to reach this goal. We are also shortening the length of ex vivo cell culture time down to two weeks to try to keep them ‘young’ before they are transferred into the patient”.


This work was funded by the Cancer Research Institute, the Stand Up To Cancer program (American Association for Cancer Research), Cancer Immunology Dream Team Translational Research (Cancer Research Institute), the National Institute of Health, Damyon Runyon Cancer Research foundation and the Burroughs Wellcome Fund.


Chapuis AG,Desmarais C,Emerson R,Schmitt TM,Shibuya K,Lai I,Wagener F,Chou J,Roberts IM,Coffey DG,Warren E,Robbins H,Greenberg PD,Yee C. 2017. Tracking the Fate and Origin of Clinically Relevant Adoptively Transferred CD8+ T Cells In Vivo. Science Immunology, 2(8).

see also:

1- Pollack SM, Jones RL, Farrar EA, Lai IP, Lee SM, Cao J, Pillarisetty VG, Hoch BL, Gullett A, Bleakley M, Conrad EU 3rd, Eary JF, Shibuya KC, Warren EH, Carstens JN, Heimfeld S, Riddell S, Yee C.  2014. Tetramer guided, cell sorter assisted production of clinical grade autologous NY-ESO-1 specific CD8(+) T cells. Journal for Immunotherapy of Cancer. 2(1):36.

2- Chapuis AG, Roberts IM, Thompson JA, Margolin KA, Bhatia S, Lee SM, Sloan HL, Lai IP, Farrar EA, Wagener F, Shibuya KC, Cao J, Wolchok JD, Greenberg PD, Yee C. 2016. T-Cell Therapy Using Interleukin-21-Primed Cytotoxic T-Cell Lymphocytes Combined With Cytotoxic T-Cell Lymphocyte Antigen-4 Blockade Results in Long-Term Cell Persistence and Durable Tumor Regression. Journal of Clinical Oncology.

3- Chapuis AG, Lee SM, Thompson JA, Roberts IM, Margolin KA, Bhatia S, Sloan HL, Lai I, Wagener F, Shibuya K, Cao J, Wolchok JD, Greenberg PD, Yee C. Combined IL-21-primed polyclonal CTL plus CTLA4 blockade controls refractory metastatic melanoma in a patient. The journal of Experimental Medicine. Jun 27;213(7):1133-9.