Resistant GIST may fall to new inhibitor

Young researcher finds way to eradicate the driver of GIST tumors

During the 2005 Connective Tissue Oncology Society (CTOS) meeting held Nov. 19-21 in Florida, Dr. Sebastian Bauer was presented with a young investigator award for his research into methods for overcoming Gleevec-resistant GIST. Bauer currently works as a medical oncologist in the interdisciplinary Soft Tissue and Bone Sarcoma Group at the West German Cancer Center, chaired by Professor Seeber, in Essen, Germany. Bauer did his research while working in the lab of Dr. Jonathan Fletcher at Brigham and Women’s Hospital in Boston, Mass. Bauer presented the results of his group’s research at the CTOS meeting. The other authors of this abstract include Drs. Lynn Yu, George D. Demetri and Fletcher.

Life Raft board of directors member Jim Hughes, left, meets Young Investigator Award winner Dr. Sebastian Bauer of Germany

While Gleevec initially works for about 85 percent of GIST patients, Gleevec will eventually fail for many of these patients despite long lasting responses or disease stabilization. The problem of resistance is not unique to Gleevec; it is common in traditional chemotherapy given to patients with metastatic disease.

In GIST, however, mechanisms of resistance are increasingly well understood, unlike many other cancers treated with classical chemotherapy. The most common cause of “secondary resistance” is the acquisition of secondary mutations in KIT, mostly in the area of the protein where Gleevec binds. Depending where the surface of KIT changes, Gleevec’s ability to bind is somewhat to greatly diminished. As a result, some secondary mutations may be sensitive to higher doses of Gleevec, but many resist even the highest doses of Gleevec.

One strategy in targeting secondary mutations is to design drugs that bind to KIT just like Gleevec but that have a somewhat altered structure that fits better into the altered binding pocket of KIT. AMN107, for example, is a modified form of Gleevec that may inhibit some secondary mutations in KIT that are insensitive to Gleevec.

However, targeting secondary mutations with another KIT inhibitor has some limitations. Since many different types of secondary mutations have been found (and some probably exist that have not been found), the new KIT inhibitor must fit into the binding pocket of as many of these different mutations as possible.

The inhibitor must also retain sensitivity against the original mutation, as many parts of the tumor may still have the original mutation and no secondary mutations.

It is even possible that different parts of a tumor within the same patient may contain different secondary mutations or other mechanisms of resistance. Combining two different KIT inhibitors (as in the current trial combining AMN107 with Gleevec) may therefore have a broader activity against the different mutations — similar to a broad spectrum antibiotic used to treat strains of bacteria that are prone to resistance against single drugs.

However, whether the combinations will be more effective is not yet clear.

In the research reported at CTOS, Bauer and his associates took a different approach. Rather than targeting the changing KIT protein directly, they targeted a protein that is required to stabilize the KIT protein. This protein, called HSP90 (short for heat shock protein 90), is not known to be mutated in GIST and thus should not be a “moving target.”

WHAT IS HSP90?

HSP90 is a member of the “chaperone” family of proteins. It 1) helps proteins to fold into their correct three dimensional shapes; 2) stabilizes a variety of other proteins, among which many are involved in the development of cancer, and; 3) protects them from degradation. Very recent preclinical work by other researchers has shown that some of these mutant proteins can be effectively inhibited by interrupting the HSP90 function. These diseases include cancers such as leukemia (bcr-abl, the other Gleevec target) and lung cancer (EGF-receptor), as well as a rather benign disease of skin cells (mast cell disease) that is driven by a mutant form KIT which usually is highly resistant to Gleevec.

A NEW THERAPEUTIC STRATEGY IN GIST?

Based on these results, Bauer and his associates tested an inhibitor of HSP90, 17-Allylamino-17- demethoxygeldanamycin (17-AAG) against four GIST cell lines. One was sensitive to Gleevec (GIST882) and three were resistant. The resistant cell lines were derived from patient biopsies and represent different mechanisms of resistance just as they can be found in patients: While all cell lines have typical Gleevec-sensitive primary mutations, GIST430 and GIST48 contain two different secondary mutations of KIT; GIST62 lost KIT as a “driver” of tumor growth.

Bauer showed that removing the protector (HSP90) of KIT using 17-AAG completely inhibited KIT function. Unlike Gleevec that just switches off KIT, 17-AAG actually “trashed the whole switch,” causing degradation of KIT. Surprisingly, using a non-GIST cell line that expresses normal, non-mutated and non-activate KIT, 17- AAG did not cause degradation of KIT. This suggests that 17-AAG mainly affects the mutant, activated form of KIT.

Bauer also compared the effects of 17-AAG and Gleevec on the cascade of “electrical” signals activated by KIT that tell the cell to proliferate and survive. He found that 17-AAG inhibited both KIT and its signaling into the cell in both Gleevec sensitive and Gleevec resistant cells that still expressed KIT. In line with these findings, 17-AAG strongly inhibited growth of these cells and caused cell death at doses that have been shown in clinical trials to be achievable in patients. However, very little effect was seen in the cell line that lost KIT expression, suggesting that HSP90 may not work in GIST with this mechanism of resistance (possibly 10 to 20 percent of patients with Gleevec-resistant GIST). Certainly more work needs to be done to understand the biology of these cells.

Nonetheless, Bauer summarized his work as a proof of concept for a new strategy to treat KIT positive Gleevec-resistant GIST. Targeting KIT indirectly via HSP90 inhibition might help to overcome resistance by inhibiting activated KIT regardless of its mutations, unlike direct KIT inhibitors such as Gleevec and Sutent that are unable to inhibit all forms of mutant KIT.

SIDE EFFECTS?

As mentioned in the introduction, HSP90 has been shown to protect more than a hundred other “client” proteins besides KIT. Many of the “client” proteins are crucial for the function of normal cells. This naturally raises the question as to the side effects that could be expected by 17-AAG. Several phase I trials have already been conducted in patients with other types of solid tumors and side effects have been rather mild, compared to effects usually seen with classical cytotoxic chemotherapy. This can be explained by the fact that HSP90 apparently exists in two different states, a native form found in most normal cells and a form bound in a “super-chaperone complex” that is found mostly in cancer cells. The native form has a low affinity for 17AAG, while the super-chaperone complex has a high affinity for 17AAG. Thus, there could be a “therapeutic window” where HSP90 inhibitors are able to affect cancer cells more than normal cells.

CLINICAL TRIALS FOR GIST BEGIN IN JANUARY 2006

17-AAG is the HSP90 inhibitor that is furthest along in clinical development. It is entering or near phase II trials even though it is not an ideal drug as it only exists in an IV formulation and has very poor solubility. Therefore, it must be given with “carriers” like DMSO or Cremaphor, both of which have side effects of their own and limit the dosing of 17-AAG. However, HSP90 inhibitors are advancing fast and newer, synthetic, small molecule inhibitors of HSP90 are in development that will be available as oral drugs in the near future.

IPI-504, a modified 17-AAG, is an intravenous drug from Infinity Pharmaceuticals based in Cambridge, Mass., which is water soluble and does not require carriers such as DMSO.

Bauer and Fletcher presented a preclinical evaluation of IPI-504 in GIST cell lines at the joint National Cancer Institute-American Association of Cancer Researchers-European Organization for Research and Treatment of Cancer meeting on targeted therapies held Nov. 14-18 in Philadelphia, PA, showing that IPI-504 is as effective as 17-AAG at shutting down Gleevec-resistant GIST.

Based on this data, a clinical trial of IPI-504 for GIST has been developed by Dr. George Demetri and the phase I trial opened at Dana-Farber Cancer Institute in Boston in January 2006. This phase I trial is for patients with Gleevec-resistant GIST.

Bauer and the Dana-Farber team are cautiously optimistic that this new method of indirectly targeting KIT through inhibition of HSP90 may provide a broadly relevant therapy for Gleevec-resistant GIST.

By Jerry Call
Dr. Sebastian Bauer contributed to this article.