Functional characterization of HLA-A2–restricted CTLs specific for hTERT. (A) hTERT CTLs were analyzed after 10 cycles of stimulation with either hTERT_865-873 or hTERT_30-38 peptides. Recognition of HLA-A2 tumor cells by hTERT-specific T cells. Polyclonal CTLs specific for hTERT_865-873 and hTERT_30-38 epitopes were tested both in IFN-γ ELISA (left) and in ⁵¹Cr release assay (right). 293-A2 cells were pulsed with the hTERT peptides as positive control or with HER-2/neu_369-377 peptide as a negative control. Data are mean ± SD of 3 independent experiments. (B) Therapeutic effectiveness of ACT by use of hTERT_865-873-specific CTLs. Rag-2^−/−γ_c^−/− mice were implanted with either SW480 or DG-75 tumor cells and when tumor area was ∼10 mm², either hTERT_865-873-specific or hTERT_30-38-specificCTLs were injected. Data are the summary of 3 cumulative experiments (n = 15/group). Mantel–Haenszel statistics: hTERT_865-873 vs hTERT_30-38, P < .001. (C) Characterization of cancer stem cells isolated from colon adenocarcinoma surgical specimens. Telomerase activity was measured by TRAP assay (left). Dilutions of TRAP products were loaded on 10% polyacrylamide gel. HI indicates heat-inactivated samples. Expression of HLA-A2 was analyzed by FACS after staining with anti-human HLA-A2 mAb or isotype control (right). (D) Cancer stem cells are in vivo target of hTERT_865-873-specific CTLs. Survival curves of Rag-2⁻/γ_c^−/− mice implanted with CTSC12 (HLA-A2⁻) or CTSC13 (HLA-A2⁺) cells. Mice were then injected with either hTERT_865-873-specific or hTERT_30-38-specific CTLs. Data are from 3 cumulative experiments (n = 5/group). Mantel–Haenszel statistics: CTSC 12: hTERT_865-873 vs hTERT_30-38, P = .43; CTSC 13: hTERT_865-873 vs hTERT_30-38, P < .001.