In this month’s March 2007 newsletter, Estelle LeCointe, the director of Association Française des Patients du GIST: Ensemble contre le GIST, has shared the story of the work of Professor Laurence Zitvogel. Professor Zitvogel and her colleagues are engaged in a project combining Gleevec and immunotherapy. It has been very interesting to watch this work evolve over the last few years. This research has culminated in the discovery of a new type of immune cell (in mice) and is poised to enter clinical trials in 2007.

Unexpected responses in GIST

A 2004 paper in the Journal of Clinical Investigation, “Novel mode of action of c-kit tyrosine kinase inhibitors leading to NK cell-dependent anti-tumor effects” by Dr. Christophe Borg, Dr. Laurence Zitvogel and colleagues was the first in a series of papers describing the work of Professor Zitvogel’s team.

They reported that 6 cases (3 in a phase I/II French study and 3 in a phase II U.S. study) of GISTs that did not have KIT or PDGFRA mutations (wild-type GISTs) still exhibited objective tumor responses. Two of these patients that had liver, stomach or lung metastases had complete responses to Gleevec with 26 months of disease-free survival. Because patients with wild-type GISTs are not expected to respond as well to Gleevec, the finding prompted a search for an alternate mode of action of Gleevec.

Inhibiting KIT signaling in dendritic cells stimulates NK cells

Zitvogel and colleagues tested Gleevec against the B16F10 melanoma cell line, and found no inhibition in the test tube (in vitro), but found that Gleevec significantly hampered the formation of lung metastases in mice (melanoma). Gleevec also induced significant anti-tumor effects against the AK7 mesothelioma and the MCA102 fibrosarcoma models. They found that these effects could be potentiated by the addition of FL, a blood growth factor (the ligand for FLT- 3). A short administration of Gleevec (4 days) combined with FL (Gleevec + FL) promoted synergistic anti-tumor effects against AK7 mesothelioma with up to 45 percent tumor-free mice using FL+ Gleevec versus 14 percent using FL only and zero percent using Gleevec alone. They concluded that even in cancers where Gleevec had no direct anti-tumor effect, Gleevec somehow caused an indirect anti-tumor effect (later shown to be due to Gleevec’s effect on the immune system).


Through observations, Zitvogel et al. noticed that the Gleevec + FL combination was unable to protect mice against some tumors. Because these tumors did not express proteins needed to activate a type of immune cell, called natural killer (NK) cells, they hypothesized that NK cells might be the critical cells activated. In experiments where NK cells were neutralized with anti-NK1.1 monoclonal antibodies, the anti-tumor effects of the combination were significantly hampered.

The team noted that treatment with Gleevec + FL did not increase the number of NK cells and could not directly boost the secretion of interferon gamma (IFN-γ). Gleevec did not seem to directly affect the NK cells. They noticed, however, that IFN-γ was produced when the NK cells were cultured together with another type of immune cell, dendritic cells. Dendritic cells help target the immune system by presenting antigens to other immune cells, effectively helping to identify targets for the immune system. Apparently Gleevec was acting on the dendritic cells which then affected the NK cells. One untreated dendritic cell was unable to “activate” NK cells, but when treated with Gleevec, one dendritic cell was able to activate 10 NK cells.

The combination of Gleevec and FL was able to increase the number of dendritic cells compared to FL alone. In addition, the team showed that KIT (ckit) signaling in dendritic cells inhibited NK cell activation. Therefore, treatment with Gleevec helped stimulate NK cell activation by disrupting KIT signaling in the dendritic cells.

GIST patients with NK cell activation appear to do better
than those without NK cell activation

The team was intrigued by their findings. They wondered whether NK cells might be responsible for some of the therapeutic effects in GIST patients. They assessed NK cell function in 49 GIST patients. Using IFN-γ as a marker of NK cell activation, they found that NK cells were “activated” in 24 of 49 GIST patients treated with Gleevec. In contrast, only 21 percent of untreated GISTs or 11 percent of normal volunteers had signs of NK cell activation. They noted that NK cell activation was increased in 9 of 11 patients after treatment with Gleevec compared to before treatment with Gleevec. More importantly, they noted that patients with NK cell activation seemed to do much better than patients without NK cell activation (See Figure 1). This effect was statistically significant. None of 10 patients with progressive disease had enhanced NK cell function.


According to Zitvogel and colleagues, GIST tumors have molecular features of NK cell sensitivity: TAP-1 deficiency, loss of MHC class I molecules, high expression on NKG2D ligands and GIST recognition by NK cells comparable to that of K562 cells.

The 2004 paper was extremely interesting. The experiments seemed to be well thought out. The only caveat with this paper is that there was no attempt to relate NK cell activation to response by exon status. An intriguing question is whether Gleevec sensitive tumors become more susceptible to the immune system while taking Gleevec as compared to non-responding tumors; in other words, does Gleevec downregulate signals/ proteins on the tumor cell that are involved in immune system evasion?

Too many Treg cells reduce NK cell activity

Regulatory T cells (Treg cells) discriminate between self and non-self. A tilt towards over-recognition of self results in autoimmune diseases, but with perhaps more activity against some cancer cells. A tilt in the other direction results in less autoimmune disease, but with perhaps less anti-tumor activity.

In 2005 at the American Society of Clinical Oncology (ASCO) conference, Zitvogel submitted Abstract 2516 that identifies the reason why half of the GIST patients did not have NK cell activation. Patients with high numbers of CD4+CD25high regulatory T cells (Treg) in the blood at entry were shown to have less NK cell function. In GIST patients with good NK cell function/activation, the percentage of Treg cells were not elevated compared to normal volunteers (1.1% vs. 1.2%). In the group with no NK cell activation, Treg cell levels were three times higher (3.2% vs. 0.8%).

Reducing Treg cells has therapeutic implications

In the 2005 ASCO abstract, Zitvogel suggests a possible way to reduce the number of Treg cells with a low-dose type of cytotoxic chemotherapy, cyclophosphamide. She noted that a combination of Gleevec + cyclophosphamide was synergistic in a mouse model of lung melanoma metastases. She notes that NK cell activation is a novel surrogate marker of efficacy of Gleevec which is critical for time to progression and could be enhanced by pretreatment of GIST patients with Treg inhibitors.

Cyclophosphamide is also given in a low-dose regiment to kill endothelial cells, thereby having an antiangiogenic effect. This type treatment, known as metronomic dosing, is under investigation as an antiangiogenic treatment. At ASCO 2006, Ghiringhelli (with Zitvogel as one of the co-authors) submitted Abstract 2561, “Metronomic cyclophosphamide regimen electively depletes CD4+ CD25+ regulatory T cells in patients with advanced solid tumors.” In 6 patients (not GIST), they found that the percentage of Treg cells was reduced by half (from 7.7% before versus 3.3% after) after treatment with cyclophosphamide. They concluded “metronomic cyclophosphamide has not only effect on tumor angiogenesis, but also strongly curtail immunosuppressive Treg, which could favor a better control of tumor progression.” A paper on this subject was published on September 6, 2006 in Cancer Immunology Immunotherapy.

By this time, Zitvogel and colleagues have found two possible ways to stimulate the immune system to work better with Gleevec: adding the growth factor FL and pre-treatment with cyclophosphamide (an approved chemotherapy).

The discovery of a new type of immune cell

gleevec_revolutionizing_treatment_clip_image002_0001By early 2006, Zitvogel and her team (with Dr. Julien Taieb as the lead author) had published a new paper in Nature Medicine, “A novel dendritic dell subset involved in tumor immunosurveillance.” At virtually the same time, another team (C.W. Chan et al.) from Johns Hopkins University published another article in Nature Medicine, “Interferon-producing killer dendritic cells provide a link between innate and adaptive immunity.” Both groups discovered in mice a new type of immune cell that appeared to be a cross between a dendritic cell and a NK cell. This cell plays the role of both assassin and messenger. When stimulated, the new cells produce large amounts of IFN-γ. They named the new cells “interferon-producing killer dendritic cells” (or IKDC).

IKDCs begin their lives behaving like a NK cell. After the cell encounters a pathogen, the cell switches roles from killer to dendritic-like messenger and, according to the researchers at Johns Hopkins, the swap occurs only once. Then, the cell dies and is replenished by the bone marrow.

“When an IKDC cell switches to its messenger function, the transformation is quite astonishing,” says Drew Pardoll, M.D., Ph.D., of Johns Hopkins Kimmel Cancer Center. The cell sprouts long, hairy tentacles called dendrites. It uses its “arms” to increase the amount of surface area it reaches to communicate and interact with other immune cells.

Professor Zitvogel and her French colleagues found that they could expand the number of IKDC in mice by fourfold during treatment with Gleevec + Interleukin- 2 (IL-2). IL-2 is a type of growth factor that stimulates the production of some types of immune cells (it is especially noted for stimulating the production of T cells). IL-2 is one of the more common treatments used in immunotherapy today.

Although the discovery of IKDC still has not been confirmed in humans, Zitvogel appears to have found a third potential way (IL-2) to stimulate the immune system to synergize with Gleevec. Now, armed with a better understanding of how Gleevec affects the immune system and how to augment the immune response, it is time for the ultimate test—clinical trials in humans.