Some observations on GIST research and a tumor bank

The simplicity of GIST makes it a model for research, treatments

By Jerry Call

The success of Gleevec for GIST has created an enormous interest in molecularly targeted therapy and GIST. Prior to Gleevec, GIST was notorious for its resistance to all treatments except surgery. Gleevec changed GIST from a cancer that rarely responded to chemotherapy to a cancer where more than 80 percent of patients on Gleevec show clinical benefit (shrinkage or stability). For many patients the durability of response has been long lasting, reaching the 2 ½-year mark and further adding to the enthusiasm for molecular target approaches.

GIST is a genetically a relatively simple cancer. It has fewer karyotype abnormalities (losses or gains of chromosomal regions) than most types of cancer. The mutation of a single gene, c-kit, seems to be an early and critical event in up to 90 percent of GISTs. Recently, another mutation in a second gene, PDGFRA, was identified as an apparent early and critical mutation in about 5 percent of GISTs. Blockage of aberrant signaling caused by the mutation of a single gene, c-kit, by Gleevec is responsible for its tremendous success in GIST patients. Gleevec also inhibits PDGFRA signaling, but how effective it is in GIST patients with PDGFRA mutations has not yet been reported.

The relative simplicity of GIST and the overwhelming importance of KIT signaling make GIST a model for studying molecular therapy in solid tumors. The following charts (see Page 5) illustrate the complications facing the studies of genetically more complex cancers. In the event the molecularly targeted therapy fails, which cancer affords researchers a better chance of discerning the underlying mechanisms of resistance? (Note: the chart is for purposes of illustration and does not depict all of the genetic alterations that occur in GIST).

Other signals besides KIT are involved in GIST. KIT signaling causes activation of downstream signals such as PI3K, AKT, STAT1, STAT3 and MAPK. In general, all of these downstream signals are common in other cancers. The PI3K/AKT pathway is particularly interesting, and several new drugs are being developed to target some of these pathways. It is probable that some day treatment for GIST might consist of a “cocktail” of several different drugs with multiple molecular targets.

The first of these multi-drug clinical trials combines Gleevec with an “mTOR” (mammalian target of Rapamycin — a downstream target of AKT) inhibitor called RAD001. This trial recently began signing on patients in Europe and will soon start in the United States. Imagine trying to assess the relative contribution of inhibiting mTOR in the “complex cancer” shown in the chart (above). How many of these pathways signal through mTOR? It seems logical that assessing the effectiveness of inhibition of downstream pathways would be easier in a “simple” cancer like GIST.

If molecularly targeted therapy is going to play a major role in more common (and more complex) cancers, then techniques will have to be developed to match the properly targeted drug to patients carrying the corresponding molecular defect. Staining for the presence of KIT (CD117) in GIST cells is the simplest example of matching a molecular marker with a molecularly targeted drug (Gleevec).

Even this simple example is more complex than it appears. First, the pathologist must determine that the tumor is GIST (staining for KIT helps do this), because the mere presence of KIT on other cancer cells does not mean KIT is activated or plays any role in the cancer. Just as we learn to walk before we learn to run, it will probably be easier to develop some of the technology for matching treatments to molecular defects in a simple well-studied cancer rather than in a complex one fraught with variability.

All cancers need tumor tissue samples for research. Because GIST appears to be a model cancer, research on GIST may help not only GIST patients but could advance the methods and technologies for other cancers as well.

For the type of sophisticated research needed, both paraffinembedded samples and frozen-tissue samples will be needed. Frozen tissue samples require that tissue be “flash-frozen” immediately upon surgical removal.

Ideally, the Life Raft Group would like to see that all GIST researchers receive all of the tissue they need for their research. However, the Life Raft Group is not interested in research alone.

We would also like to see some of the results learned from members’ tissue samples shared with the patient in order to improve that patient’s medical management. We recognize that such disclosure of data to the patient is a radically different concept than most traditional tumor banks, but our survival may well depend upon this.

In summary, research of GIST has a high level of interest, and tissue donation will facilitate the pace for discovery about mechanisms of resistance to Gleevec in GIST.

Due in large part to the relative simplicity of GIST, it is a model cancer in several ways:
1. Studying resistance to treatment should be easier. Some of the lessons learned may be applicable to other cancers.
2. Downstream signaling may be easier to identify and evaluate its importance to the progression of the cancer.
3. The techniques/methods developed in GIST research/treatment can also serve as models for other cancers. Techniques and methods will have to be developed to:
     a. Identify multiple genetic defects.
     b. Determine if the identified defects are relevant (how much are they contributing to the cancer?).
     c. Match genetic defects with targeted therapies. Identify whether the target is affected (for instance,
     is KIT inhibited?).