By Jonathan Fletcher, M.D. & Sebastian Buaer, M.D.
Within the past decade, GISTs have emerged from being poorly defined, treatment-resistant tumors to a well recognized, well understood, and treatable tumor entity. Rapid advances in the understanding of GIST biology have made this tumor a paradigm for molecularly targeted therapy. Approximately 85 percent of GIST harbor activating mutations in KIT or in the closely-related receptor tyrosine kinase PDGFRA gene. These mutations are an early event in GIST development, and create constitutively activated oncoproteins that can be therapeutically inhibited by the smallmolecule tyrosine kinase inhibitors imatinib and sunitinib, among other drugs.
The challenge for the LRG Research Team in the coming years is to ensure that increasing numbers of individuals will be alleviated of suffering from GIST, and that GIST will therefore remain a paradigm of success in cancer therapeutics.
In particular, we must maximize the dramatic clinical successes of imatinib and like drugs by identifying additional therapeutic regimens that can synergize with these KIT/PDGFRA kinase inhibitors in lengthening clinical remissions, and in eventually curing persons afflicted with GIST. Such clinical successes, particularly for those with metastatic or otherwise inoperable GIST, will doubtless require treatment with combinations of drugs, because medical history tells us that cancer cells, in most people, will eventually develop resistance to a single drug, no matter how effective that drug is initially. In this sense, we are transitioning to a new era in GIST research and therapeutics, where the recent clinical focus on single drugs, such as imatinib and sunitinib, will increasingly be enhanced by studies in which cocktails of effective drugs are administered, either concurrently or sequentially (Figure 1).
The downside of treating with drug cocktails, rather than single drugs, is the likelihood of increased side-effects, since each drug can have additive toxicities against normal cells. However, the crucial benefits of drug combinations include not only the opportunity to maximize initial clinical response by killing more GIST cells, but also the possibility that fewer cells will develop resistance against multiple drugs, compared to a even a highly-effective single drug, particularly if the drugs have different mechanisms of action.
Most individuals with unresectable or metastatic GIST respond to imatinib, with the GIST tumors remaining stable for a time under treatment, after undergoing initial clinical response. Although imatinib thus succeeds in blocking GIST cell growth (the residual surviving GIST cells are referred to as “quiescent”), it is imperative to identify new therapies that can kill the surviving GIST cells. One of the priority aims of the LRG Research Team is to identify drugs that more effectively induce GIST cell death (apoptosis) and that can be safely administered together with a KIT kinase inhibitor drug such as imatinib. Such methods may be particularly useful in consolidating the initial clinical response, for those whose GIST has responded and stabilized on imatinib, sunitinib or some other KIT kinase inhibitor.
One such option, as highlighted in previous LRG newsletters, might involve the addition of PI3-K inhibitors to imatinib, in order to maximize inhibition of GIST cell survival pathways. However, like many evolving fields in cancer research, this is complex, with many possibilities to consider. Preliminary studies (unpublished) show that various PI3-K inhibitors differ widely in their effects on GIST cells, with some having primarily cytostatic (growth-inhibiting) properties, whereas others can induce a dramatic apoptotic response, resulting in GIST cell death. These observations emphasize a general rule, which is that different drugs against a particular GIST biological “target” do not necessarily have equivalent effects on the tumor cells. Such varying drug efficacy, even within a particular class of drugs and among drugs with similar pharmacological effectiveness, can result from differential success in inhibiting the various forms of the target proteins that regulate GIST growth and survival, and can also result from so-called “off target” effects, in which a particular drug might inhibit other proteins (above and beyond those the drug is known to inhibit) which can, in unanticipated manner, contribute to GIST growth inhibition and GIST cell death. Therefore, it is imperative to test as many drugs as possible against GIST laboratory models, even when the drugs are expected to have similar clinical properties, so as to identify those with the greatest clinical promise, which should then be prioritized for evaluation in new clinical trials.
Recently, the LRG Research Team demonstrated substantial inter- and intralesional heterogeneity in drug resistance mutations in patients treated with imatinib alone or imatinib and sunitinib: 83 percent of patients in this study had secondary drug -resistant KIT mutations, including 67 percent with two to five different secondary mutations in separate metastases, and 34 percent with two secondary KIT mutations in the same metastasis. This substantial heterogeneity of resistance mutations highlights the therapeutic challenges involved in substantially extending survival, especially after clinical progression on KIT/PDGFRA kinaseinhibitor monotherapies. This clinical reality suggests that although newer generations of broad-spectrum, increasingly potent, KIT/PDGFRA kinase inhibitors may benefit those progressing on imatinib therapy, such drugs – on their own – are unlikely to cure many patients with imatinib (or sunitinib) resistant disease. Therefore, it seems clear that multiagent treatment modalities are needed in the future, incorporating classes of drugs that inhibit the KIT/PDGFRA oncoproteins in novel ways. These are not “handcuffed” by the molecular heterogeneity of KIT/PDGFRA kinase-domain resistance mutations selected for during imatinib and sunitinib therapies. Combination therapies with various such inhibitors will prolong GIST remissions, in a manner analogous to treatment approaches used against HIV in patients with AIDS, i.e. by suppressing a broader spectrum of GIST clones from the outset of therapy. Similarly, therapeutic options less dependent on specific molecular mechanisms of KIT or PDGFRA activation are needed to overcome the substantial heterogeneity of secondary KIT/ PDGFRA kinase mutations ultimately responsible for treatment failure in many persons.
To overcome the limitations of direct KIT-inhibitors we need to identify the Achilles’ heels of oncogenic KIT. KIT mutations allow GIST cells to evade natural self-inhibitory mechanisms, and in that sense are like a master-switch stuck in the “on” mode, resulting in socalled “constitutive activation” of KITdependent growth and survival signaling pathways. Nonetheless, there are undoubtedly ways to halt these constitutive activation signals. For example, even the “switched-on”, mutant KIT proteins are strongly dependent on various “helper” proteins, which are not mutant, but which bind to KIT and assist in its functions. Some of these helper proteins might represent ideal therapeutic targets in GIST.
Novel GIST therapeutic strategies, increasingly, will also focus on blocking production of the mutant KIT or PDGFRA proteins. Production of KIT/ PDGFRA proteins requires transcription and translation (“reading and writing”) from the KIT gene within the GIST cell DNA. Initiation of this production process is directly regulated by various transcription factors, and indirectly regulated by proteins that activate or inhibit such transcription factors.
Kinases are proteins, which – like KIT and PDGFRA – regulate the activities of cell growth and survival by transferring energy in the form of phosphorylation. We and others have shown that the kinase protein, PKCtheta, is uniquely and strongly expressed in GIST. Indeed, in GIST cells, PKCtheta binds to KIT, and can likely phosphorylate KIT, whereas KIT regulates phosphorylation of PKCtheta. Therefore, it seems that certain aspects of KIT and PKCtheta function are interdependent in GIST. Notably, expression of KIT itself depends on the presence of PKCtheta in the GIST cells. In laboratory experiments we reduced the amount of PKCtheta in GIST cells, which resulted in a decrease in KIT gene transcription, and also decreased the amount of mutant KIT proteins in the cells. PKCtheta seems to be a potentially useful, and highly specific therapeutic target, since few normal cells in the body contain large amounts of PKCtheta. Drugs against PKCtheta would therefore be expected to have very limited toxicity, and would not likely be constrained by the presence of imatinib-resistance KIT mutations. A near-term goal is to identify and validate potent clinically-useful inhibitors of PKCtheta, particularly those which might lead to decreased amounts of KIT oncoproteins.
Transcription and translation of the KIT mutant oncogene results in a string of amino acids (the starting point for a functional protein) that, left to it’s own devices, would assemble into a more or less random configuration, like an unkempt ball of wool, and which would be identified as useless and promptly diverted into the cell’s trash shredder: the “proteasome”. However, GIST cells contain chaperone proteins which ensure that nascent KIT oncoproteins are properly folded and localized within the cell, and are protected from premature destruction. We recently showed that inhibition of the chaperone HSP90 can result in KIT oncoprotein destruction, irrespective of the imatinib-resistance mutations present. And initial clinical trials of HSP90 inhibition (IPI-504) have served as a crucial proof-of-concept for HSP90 as a master-regulator of KIT in the GIST cell. Several HSP90 inhibitors are being developed by pharmaceutical companies and apparently the ability to inhibit KIT varies among them. Future studies will need to identify the most potent and clinically useful HSP90 inhibitors for GIST. An urgent priority, in future clinical trials, will be to determine the extent to which HSP90 can be truly shut down by HSP90 inhibitor drugs, and to determine what clinical efficacy (and what toxicities) result when that pharmacologic aim is achieved.
Similarly to HSP90 inhibitors, the histone deacetylase (HDAC) inhibitors – as shown by Drs. Debiec-Rychter and Bauer of the LRG Research Team (unpublished) – can also destroy the crucial KIT oncoproteins in GIST cells. While the exact mechanism for this remains to be elucidated, HDAC inhibitors not only inhibit HSP90 but may also inhibit transcription of KIT oncogenes. Other drugs that block KIT gene transcription, hence impairing KIT synthesis in the GIST, include flavopiridol, where the therapeutic efficacy, as for HSP90 and HDAC inhibitors, is not expected to be derailed by imatinib-resistance mutations in the KIT coding sequence.
Another therapeutic approach to potentially circumvent imatinib-resistance mutations might involve immunotherapies, using antibodies against the part of the KIT protein that extends outside the GIST cell. Such immunotherapies might be successful even in GISTs that have developed resistance to drugs like imatinib, that bind the KIT kinase domain within the cell.
As mentioned above, recent studies from our group and others show that the PI3-kinase pathway is strongly activated in most GISTs, and is indeed dependent on KIT/PDGFRA activation In a simplified model, KIT activates PI3-K, which then activates the AKT-kinase, which then activates the mTOR kinase as a linear pathway (like dominos falling) promoting GIST cell growth and survival. In reality, these so-called “signaling pathways” are not linear, but have intricate cross-connections, and the details of the cross-connections can differ between the GIST growing in a person versus the GIST laboratory model, such as an immortal GIST cell line, used for drug-testing. Therefore, the clinical efficacy of a given GIST signaling pathway inhibitor drug cannot be definitively predicted from its efficacy (or lack thereof) in laboratory tests. Nonetheless, such studies are crucial, because KIT or PDGFRA mutant oncoproteins are the major drivers of growth and survival in most GISTs, and an improved understanding of the mechanisms of KIT/PDGFRA oncogenic signaling will undoubtedly enable more effective therapeutic strategies.
Targeting of KIT/PDGFRA signaling pathways, particularly when coupled with KIT/PDGFRA kinase inhibitors, will likely be useful clinically in maximizing initial clinical responses, prolonging remissions, and treating GISTs that have progressed on imatinib or sunitinib therapy. These more complex combination-drug therapies will require highly-coordinated efforts between laboratory and clinical researchers, to more efficiently identify therapeutically relevant drugs, and to determine the optimal dose schedules and treatment sequences for these drugs.
- Corless CL, Fletcher JA, and Heinrich MC: Biology of gastrointestinal stromal tumors. J Clin Oncol 2004, 22: 3813-3825
- Blanke CD, Demetri GD, von MM, Heinrich MC, Eisenberg B, Fletcher JA, Corless CL, Fletcher CD, Roberts PJ, Heinz D, Wehre E, Nikolova Z, and Joensuu H: Long-term results from a randomized phase II trial of standardversus higher-dose imatinib mesylate for patients with unresectable or metastatic gastrointestinal stromal tumors expressing KIT. J Clin Oncol 2008, 26: 620-625
- Liegl B, Kepten I, Le C, Zhu M, Demetri GD, Heinrich MC, Fletcher CD, Corless CL, and Fletcher JA: Heterogeneity of kinase inhibitor resistance mechanisms in GIST. J Pathol 2008, 216: 64-74
- Duensing A, Joseph NE, Medeiros F, Smith F, Hornick JL, Heinrich MC, Corless CL, Demetri GD, Fletcher CD, and Fletcher JA: Protein Kinase C theta (PKCtheta) expression and constitutive activation in gastrointestinal stromal tumors (GISTs). Cancer Res 2004, 64: 5127-5131
- Zhu MJ, Ou WB, Fletcher CD, Cohen PS, Demetri GD, and Fletcher JA: KIT oncoprotein interactions in gastrointestinal stromal tumors: therapeutic relevance. Oncogene 2007,
- Ou WB, Zhu MJ, Demetri GD, Fletcher CD, and Fletcher JA: Protein kinase C-theta regulates KIT expression and proliferation in gastrointestinal stromal tumors. Oncogene 2008,
- Bauer S, Yu LK, Demetri GD, and Fletcher JA: Heat shock protein 90 inhibition in imatinib-resistant gastrointestinal stromal tumor. Cancer Res 2006, 66: 9153-9161
- Sambol EB, Ambrosini G, Geha RC, Kennealey PT, Decarolis P, O’Connor R, Wu YV, Motwani M, Chen JH, Schwartz GK, and Singer S: Flavopiridol targets c-KIT transcription and induces apoptosis in gastrointestinal stromal tumor cells. Cancer Res 2006, 66: 5858-5866
- Rossi F, Ehlers I, Agosti V, Socci ND, Viale A, Sommer G, Yozgat Y, Manova K, Antonescu CR, and Besmer P: Oncogenic Kit signaling and therapeutic intervention in a mouse model of gastrointestinal stromal tumor. Proc Natl Acad Sci U S A 2006, 103: 12843- 12848
- Bauer S, Duensing A, Demetri GD, and Fletcher JA: KIT oncogenic signaling mechanisms in imatinib-resistant gastrointestinal stromal tumor: PI3-kinase/AKT is a crucial survival pathway. Oncogene 2007, 26: 7560- 7568