The ability of mutational analysis to predict the response of GIST tumors to Gleevec has been known since Dr. Charles Blanke’s GIST presentation at the plenary session of the 2001 American Society of Clinical Oncology (ASCO) meeting. A new paper by Maria Debiec- Rychter and colleagues from the European Organization for Research and Treatment of Cancer (EORTC) may prove to be a tipping point that causes mutational analysis to become routine for GIST patients, especially if their data is confirmed by the slightly smaller U.S./Canadian phase III study.
As many oncologists and GIST patients know by now, the most important therapeutic target in GIST is the KIT protein (or for a small minority the PDGFRA protein). Since the 2001 Blanke presentation it has also been known that where the mutation occurs in the c-kit gene (which contains the instructions for making the KIT protein) is also important. Mutations can occur in different parts of the gene called exons.
Dr. Michael Heinrich and Dr. Christopher Corless of Oregon Health & Science University (OHSU) have been two of the pioneers in the field of mutational analysis for GIST. Analysis that identifies the mutated gene as well as the exon where the mutation occurs is known as genotyping and has been available since 2003, but recent data from the GIST reGISTry has shown that only 3.5 percent of GIST patients have had genotyping done. A limited role in the treatment decision process, possible denial of insurance coverage, and a lack of availability in some parts of the world have been the most cited reasons for not having genotype testing done.
Genotype has been shown to be the most important predictor of how GIST patients’ tumors will respond to Gleevec. Patients with KIT exon 11 tumors have the best response rate, those with exon 9 tumors have an intermediate response rate and those with wild-type GIST (no mutations in KIT or PDGFRA) have a poor response rate. Patients with a PDGFRA mutation in exon 12 also respond well while those with exon 18 mutations have a poor response rate.
Last year at the 2005 ASCO meeting, Dr. Heinrich presented the first information that patients with KIT mutations in exon 9 have responses to the drug much more often at higher doses of Gleevec. About 15 percent of GIST patients have this type of mutation and these patients have an 8 times greater chance of having a response (shrinkage of their tumor(s) by approximately 50 percent). Even though this data had a very high P value, it was not considered “statistically significant” because it was an “interm analysis.”
In a recent paper published in the European Journal of Cancer, Maria Debiec- Rychter and the EORTC team expanded the work of Heinrich and the U.S./Canadian group to show that the higher doses of Gleevec also significantly prolonged progression-free survival for the exon 9 GIST patients. They reported that “. . . treatment with the high-dose regiment resulted in a significantly superior progression-free survival (P=0.0013), with a reduction of the relative risk of 61 percent.
Included in the paper were graphs called “Kaplan-Meier” curves that showed in a more visual way the difference between doses. When we look at the graph for exon 9 patients (see page 3, Figure 1A) we see that the median progression time (time point at which half of the patients have progressed) is only about 4 months for the low dose group. In stark contrast, the median time to progression for the higher dose group is about 20 months. This represents tumor control for five times as long as the low-dose regiment.
Even though the exon 9 high-dose group’s median progression-free survival is five times as long, we do not know the full story because 57 percent of the patients on the lower dose were able to “cross-over” to a higher-dose of Gleevec. The length of benefit for these crossover patients was not given in the paper.
The EORTC paper concluded that “… These results suggest that imatinib (Gleevec) should be dosed at 400 mg twice a day in patients with tumours bearing KIT exon 9 mutations.” The only way to tell which patients have an exon 9 mutation (and need the higher dose) is to do mutational testing (genotyping) on all GIST patients (or at least all new GIST patients).
Additionally, the EORTC study showed that for wild-type GISTs “… higher dose did not significantly change the progression-free survival, with a trend even toward the benefit for the lower dose” (see page 3, Figure 1B). The authors concluded that interpretation of response in this group was difficult because of the diversity of wild-type GISTs. Hopefully when the data is merged between the two phase III trials the picture will become a little clearer.
The response to crossover from a lower dose of Gleevec to the higher dose was highly dependent on genotype as well. Responses (as measured by the growth modulation index) after crossover were:
•Exon 11, 7%
•Exon 9, 57%
Because KIT exon 11 tumors are the most common, an exploratory subset analysis of this group was possible. It turns out that not every exon 11 tumor has the same chance of responding to Gleevec (see Figure 2). Debiec-Rychter et al. discovered that patients with mutations in “the more distal end of exon 11” did not seem to respond as well as the total group of exon 11 patients. The hazard ratio for progression was significantly higher in tumors with mutations in codons 562-567 and particularly high in tumors with mutations at codons 577- 579. The authors hypothesized that “It is possible that mutations inducing conformational changes, such as large deletions or insertions (the latter being reported mainly in the distal 3’ end of exon 11) may reduce the affinity of KIT for imatinib and moderate the efficacy of the drug.” Mutation/deletion of codon 579 was the most significant factor in a univariate analysis and had a hazard ratio of 3.37 (P >0.0013).
Genotyping also plays a prognostic role for Sutent response as well, however, our understanding of response to Sutent is limited to Gleevec-resistant GIST.
Gleevec-resistant patients with exon 9 mutations have the highest response rate to Sutent, those with wild-type GIST have an intermediate response rate and those with exon 11 mutations have a lower response rate.
Sutent has also been tested against the secondary mutations that seem to be the largest cause of secondary resistance. As reported by Dr. Michael Heinrich at 2006 ASCO, Sutent is active against secondary mutations in exons 13 and 14, but not in the exon 17 and exon 18 mutations that were tested.
The Heinrich 2005 ASCO presentation and the Debiec-Rychter/EORTC paper are the most compelling arguments to date for moving genotyping from a research procedure to a routine clinical procedure. The difference in response between low-dose and high-dose Gleevec for exon 9 patients is simply too large to ignore.