Effects of Imatinib (Gleevec) on GIST Cells

GSI asked Jonathan C. Trent, M.D. Ph.D. several questions about the effects of Gleevec treatment on the cells within gastrointestinal stromal tumors. Dr. Trent is a medical oncologist at the Sylvester Cancer University of Miami. GIST is one of his main interests in his practice and in his research.

Read a recent free-access summary paper on GIST authored by Dr. Trent and others, click this link:

QUESTION 1. Some patients’ GISTs shrink rapidly when they take Gleevec, while the tumors of other patients shrink very slowly or show stable disease. Is imatinib “cytostatic” or “cytotoxic” in the treatment of patients with GIST? In other words, does Gleevec merely arrest GIST or can it kill GIST?

a. In cancer the net tumor volume is a balance between the rate of tumor cell proliferation (cell division) and the rate of tumor cell death. In untreated cancer patients the tumor is composed of cells in which there are more cells dividing than dying, pushing the balance toward net tumor growth (reference 1). Perhaps an effective therapy should slow or prevent tumor cells from dividing (a cytostatic mechanism) while at the same time actively killing tumor cells (a cytotoxic mechanism).

b. There is evidence to support cytotoxic effects of imatinib mesylate in GIST. It is a well-described phenomenon now that tumor cells are replaced by myxoid degeneration after GIST patients have been treated with imatinib for as little as 4 weeks (reference 2). Myxoid degeneration refers to the proteinaceous material left behind after cell death. This decrease in cellularity suggests that the GIST cells have undergone cell death. Moreover, there have been laboratory studies that have shown that imatinib mesylate induces apoptosis in GIST cells (reference 3). In our studies we have found that when GIST patients are treated with imatinib for as few as 3 days, tumor cells in the surgically resected specimen are undergoing apoptosis (reference 4). Although the mechanism is not clear, treatment with imatinib can induce cell death. In other words, imatinib is cytotoxic in this situation.

c. There appears to be a cytostatic effect of imatinib mesylate as determined by a decrease in proliferation of GIST cells. In a GIST patient treated with imatinib mesylate therapy for 4 weeks most of the tumor cells were replaced by myxoid degeneration. Interestingly, the remaining tumor cells did not appear to be actively dividing since the proliferation marker, Ki-67, was not expressed by the tumor cells. Some researchers speculate that these GIST cells will resume cell division if imatinib mesylate therapy is discontinued in patients whose tumor has residual cells after imatinib mesylate therapy.

d. In summary, it appears that imatinib mesylate is both cytostatic and cytotoxic to GIST cells. Studies such as the RTOG-S0132 and the MDACC ID03-0023 may shed light on the mechanisms behind these effects. In each of these trials, patients undergo a tumor biopsy followed by preoperative therapy with imatinib mesylate and a surgical resection of their GIST. Patients then receive imatinib for 2 years postoperatively. Laboratory correlations from these studies will aid in understanding the early effects of imatinib mesylate in GIST, will help oncologists treat GIST more effectively, and possibly lead to breakthroughs in the therapy of other types of cancer.

QUESTION 2. How does imatinib induce apoptosis (controlled cell death) if it acts by inhibiting a growth signal?

a. Tumor cells depend on activation of aberrant growth signal pathways for both proliferation and survival (reference 1). These two phenomena are intricately tied together rather than existing as separate events (reference 5). In many experimental models the withdrawal of a growth factor pathway stimulatory signal leads to cell death by a mechanism such as apoptosis. The treatment of a “Kit-driven” GIST cell with imatinib effectively results in withdrawal of growth factor support. Thus, there is rationale behind the idea that inhibition of Kit signaling may inactivate molecules that are needed for GIST cell survival.

b. One survival pathway thought to be downstream of Kit involves phosphotidyl inositol-3-kinase activation of a molecule known as AKT (reference 3). Other molecules which are sometimes important in tumor cell survival are Bcl-2, mutant p53, telomerases, among others. Interaction between these molecules and the Kit signaling pathway is possible.

QUESTION 3. What potential apoptosis-inducing additions to drug treatment could be considered together with Gleevec for patients who are still responding to Gleevec? For example, a patient with an inoperable tumor may have achieved some shrinkage with Gleevec, but not enough to make the tumor resectable. As another example, patients with resected primary tumors can harbor GIST cells after Gleevec treatment, as evidenced by recurrence after Gleevec is withdrawn. Can you hypothesize about combined therapies to help finish the job of killing GIST cells?

a. Patients whose GIST is shrinking on imatinib, no matter how slowly, should continue imatinib therapy.

b. Patients with potentially resectable GIST (primary or with limited metastases) may require a year or longer of imatinib therapy to reach maximum tumor shrinkage prior to surgical resection.

c. If a patient has recurrent, metastatic GIST after discontinuation of imatinib, I would consider reinstitution of imatinib as my first line of therapy.

QUESTION 4. What strategies in drug treatment could potentially restore apoptosis and tumor control for patients who have grown resistant to Gleevec?

a. Patients who have growth of their tumor on imatinib have limited therapeutic options. One potential mechanism by which GIST cells are resistant to imatinib is the failure of imatinib to induce cell death.

b. Restoration of imatinib-induced tumor cell death in patient’s whose GIST has become resistant to imatinib mesylate is an area of ongoing clinical and laboratory research. One approach in the clinics is to use a new compound which inhibits Kit as well as other tyrosine kinases. Two such multi-targeted kinase inhibitors are currently open to patients in a clinical trial: AMG706 and SU11248. These agents appear to inhibit Kit signaling at a lower serum concentration than imatinib mesylate and inhibit other kinases such as the vascular endothelial growth factor receptor. More potent inhibition of Kit and inhibition of additional pathways may result in a restoration of tumor cell death.

c. An alternative to this approach is to add an additional therapy to imatinib to reverse the mechanism of resistance and reinstitute imatinib-induced GIST cell death. In the case that imatinib resistance is due to a failure to induce apoptosis, then addition of an agent that can restore apoptosis may provide benefit to patients. Examples of agents currently of interest are antisense Bcl-2 and inhibitors of the AKT pathway such as perifosine and RAD001.

QUESTION 5. What is your assessment of current and future GIST treatments?

The therapy of advanced GIST with imatinib mesylate has been markedly successful whether by cytostatic, cytotoxic, or both mechanisms. New agents to treat patients with metastatic, imatinib-resistant GIST should be selected carefully by their potential ability to induce GIST cell death while blocking proliferation.


1. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000;100(1):57-70

2. Joensuu H, Roberts P, Sarlomo-Rikala M, et al. Effect of the tyrosine kinase inhibitor STI571 in a patient with a metastatic gastrointestinal stromal tumor. New England Journal of Medicine 2001;344(14):1052-1056.

3. Tuveson DA, Willis NA, Jacks T, et al. STI571 inactivation of the gastrointestinal stromal tumor c-KIT oncoprotein: biological and clinical implications. Oncogene 2001;20(36):5054-5058.

4. Trent J, Choi H, Hunt K, et al.: Apoptotic and anti-vascular activity of imatinib in GIST patients (Abstract # 9001) The American Society of Clinical Oncology 2005; 24.

5. Trent JC, 2nd, McConkey DJ, Loughlin SM, Harbison MT, Fernandez A, Ananthaswamy HN. Ras signaling in tumor necrosis factor-induced apoptosis. Embo J 1996;15(17):4497-4505.