Correlation of target kinase genotype with clinical activity of imatinib mesylate (IM) in patients with metastatic GI stromal tumors (GISTs) expressing KIT (KIT+)

Reviewer: Walter Sall, MD
Last Modified: May 15, 2005

Presenter: M.C. Heinrich
Presenter's Affiliation: OHSU Cancer Institute
Type of Session: Plenary


  • Most GISTs express constitutively activated mutant isoforms of KIT.  These are therapeutic targets for IM (Gleevec).  The magnitude of clinical response of these tumors to IM shows marked heterogeneity.  The investigators hypothesized that different mutant subtypes of KIT may confer differing susceptibilty to IM perhaps allowing tumor genotyping to predict response and prognosis in individual patients.
  • The correlation between KIT mutant isoform subtype and clinical activity of IM were examined as part of a randomized phase III trial testing 400 vs 800mg daily IM dosing (CALGB 150105/SWOG S0033).

Materials and Methods

  • Pretreatment tumor samples from 324 patients with KIT+ GIST were examined for KIT mutations primarily by DNA array analysis.
  • Time to treatment failure (TTF) and Overall Survival (OS) were then analyzed by the absence or presence of KIT mutation and mutation subtype.


  • 280 of 324 patients with KIT+ GIST had KIT mutations (86.4%).
  • Exon 11 KIT mutation isoforms were associated with a higher Objective Response rate (OR=67%), than exon 9 mutation (OR=40%) or no kinase mutation (OR=39%).
  • IM dose had no effect on OR for any KIT mutation subtype.
  • Multivariate analysis showed exon 11 mutation genotype to be the single best predictor of OR.
  • TTF was 576 days vs 308 vs 251 for those with exon 11mutation, exon 9 mutation and no mutation, respectively.
  • There was a trend toward improved OS for patients with exon 11 mutation vs other subtypes; p=0.04 but was not significant when adjusted for multiple comparisons.

Author's Conclusions

  • Activating KIT mutations are found in the majority of KIT+ GISTs. 
  • The presence of exon 11 mutation correlates with the best clinical response to IM, as well as TTF and perhaps OS.
  • This study provides important evidence that gene expression profiling can be used to predict response to therapy and treatment outcome.

Clinical/Scientific Implications

In the new era of targeted cancer therapeutics, the rapidly increasing number of therapeutic options will make prediction of tumor response and prognosis important in determining choice of cancer therapy.  This trial provides some of the best evidence currently available that therapies can be customized using gene expression profiling. 

A similar trial in KIT+ GIST tested another KIT inhibitor, SU11248.  In this trial, exon 9 mutants were seen to have a better response to therapy and better ultimate outcome.  This further supports the continued use of gene profiling in chosing targeted therapies.  Resistance mechanisms to IM, SU11248 and other similar compounds are being investigated.  These mechanisms include KIT overexpression and further mutation. It may to useful to test these KIT inhibitor drugs in combination in order to combat these mechanisms of resistance.  This research is currently underway in the pre-clinical stage

Frequently Asked Questions