Abstract 3729

Comprising <10% of T lymphocytes in human blood, gamma delta T cells (GDTc) are cytotoxic immune surveillance cells that are an attractive option for adoptive immunotherapy. We have previously shown that healthy donor GDTc expanded in vitro are cytotoxic to EM-2 and K562 Ph+ leukemia cell lines but not to autologous peripheral blood mononuclear cells (PBMCs). Additionally, we have noted an oligoclonal expansion of GDTc in interferon alpha (IFNα) monotherapy treated chronic myeloid leukemia (CML) patients. The current study is aimed at elucidating the role that these cells play in the remission of IFNα-treated CML patients.

Immunophenotyping of thawed PBMC samples from CML patients undergoing (IFN-ON) or who had discontinued IFNα monotherapy (IFN-OFF) revealed similar central CD45RA-CD27+ (5.7% ± 2.7%, mean ± standard error, n=6) and CD45RA-CD27- effector memory (3.7% ± 2.0%, n=6) GDTc pools compared to those from healthy age-matched donors (4.4% ± 1.2% and 4.7% ± 1.4% respectively, n=3), suggesting that GDTc pools are normal in these patients. However, differences approaching significance were observed between IFN-ON (n=3) and healthy controls (n=3), with IFN-ON exhibiting less CD45RA+CD27+ naïve GDTc (24.6% ± 6.7% compared to 48.2% ± 7.7%, p=0.08) and more terminally differentiated CD45RA+CD27- RA+ effector memory GDTc (66.4% ± 11.8% compared to 42.7 ± 7.3%, p=0.16). The percentage of GDTc expressing the activating Natural Killer cell receptor NKG2D was more variable and significantly lower in freshly thawed IFN-OFF patient samples (53.1% ± 4.6%, p=0.03) as compared to age-matched healthy donor controls (68.4% ± 1.5%). A similar trend was also observed between IFN-ON patients and healthy controls (IFN-ON 54.5% ± 8.8%, p=0.08).

Although GDTc are typically difficult to recover and expand from cryopreserved PBMCs, we have developed a new protocol with which we can expand patient GDTc, potentially enabling the generation of therapeutic doses on demand. We achieved variable expansions of 41- to 9097-fold from patient samples over 21 days, with four of six patient cultures achieving over 100-fold GDTc expansion. Of these four cultures, two were from each patient group, suggesting that IFN-ON or –OFF status did not correlate with the ability of these cells to expand in vitro. Importantly, expanded GDTc were mostly terminally differentiated CD45RA+CD27- RA+ effector memory GDTc (52% ± 11.7%, n=6). Of the total GDTc population, 65.9% ± 4.6% expressed NKG2D (n=6), suggesting that our protocol can generate terminally differentiated cytotoxic GDTc. We are now investigating whether the cells expanded in vitro are the clonal GDTc population previously identified in our patients and whether these expanded cells are able to kill Ph+ leukemia cells. Our findings will be important not only for understanding the mechanism of IFNα-induced cure of the patients, but also for developing GDTc therapies for CML and other malignancies.


Mustjoki:Bristol-Myers Squibb: Honoraria, Research Funding; Novartis: Honoraria.

Author notes


Asterisk with author names denotes non-ASH members.