At the ASCO Gastrointestinal Cancer Symposium held in late January, 2006, Dr. Charles Blanke presented a poster detailing 4-year follow-up results of the patients in the first randomized Phase II clinical trial of Gleevec for treatment of GIST, known as the B2222 trial. These patients represent the longest ongoing data set about Gleevec used to treat GIST. 147 patients were enrolled during summer and fall 2000. The follow-up data track them through May 2005, with a median follow up time of 52 months.
The abstract of the poster is reprinted below. An Adobe pdf file of the entire poster can be viewed (CLICK HERE) to see all the figures and tables, plus additional text. You can use the zoom function of the Adobe reader software to enlarge the image (try 200% to view two columns of the poster at a time, 400% for one column at a time). If you do not already have Adobe Reader software on your computer, it is a free download.
About 33% of patients continue with no progressive disease at the 240-week mark. The median overall survival on Gleevec is 58 months, compared to 15 months without Gleevec therapy. Starting dosage of 400 mg/day versus 600 mg/day made no substantive difference, though this trial was not designed to demonstrate statistical significance regarding dosage differences. The onset of a partial response typically occurred by 12 weeks (median value), but for 25% of patients achieving a partial response required longer than 23 weeks.
The mutation status of the patients was correlated with their drug responses, as shown in Figure 6 of the poster. Responses were measured according to RECIST: complete response (no measurable disease remaining), partial response (greater than 50% shrinkage by area) and stable disease (i.e. benefit less than 50% shrinkage by area or progression limited to less than 20%). Over 90% of patients with a mutation in exon 11 of the KIT gene showed benefit from Gleevec (either objective response or stable disease), while over 70% of exon-9 KIT-mutant patients showed benefit. There were very small numbers of patients with other mutations or with no detectable mutations (wild-type). Of 5 PDGFRA-mutant patients, 2 showed benefit (partial response). Of 9 wild-type patients, 3 showed benefit (stable disease).
Patients who achieved objective responses and those who experienced stable disease had similar durations of response, as shown in Figure 4. However, the exon-11-mutant patients had a longer-lasting response than the exon-9-mutant patients (see Figure 7). The conclusions section of the poster states “Late responses are often seen in patients with initial stable disease.” There were 2 patients who achieved a complete response, as described in the abstract.
ABSTRACT
Outcome of advanced gastrointestinal stromal tumor (GIST) patients treated with imatinib mesylate: Four-year follow-up of a phase II randomized trial
Authors: C. D. Blanke, H. Joensuu, G. D. Demetri, M. C. Heinrich, B. Eisenberg, J. Fletcher, C. L. Corless, E. Wehrle, K. B. Sandau, M. von Mehren
Imatinib mesylate induces an objective response in the majority of pts with advanced, surgically incurable GIST, but secondary resistance and late relapses do occur. A phase II randomized clinical trial was previously presented (PASCO 20:1A, 2001; Combined GI Symposium 2:43, 2004), and data from an updated long-term analysis of this study are now available, with median f/u of 4.4 years. 147 pts were randomized to treatment with continuous imatinib, at 400 or 600 mg po daily. Two pts (1%) achieved a complete response, and 98 (67%) achieved a partial response (PR), for an overall objective response rate of 68% (95% CI 59.8-75.5). An additional 16% had prolonged stable disease. No significant response differences were seen between the two doses. Of the two pts who achieved complete responses, one died 3.2 years later without evidence of recurrence, having remained on imatinib until shortly before his death. The latter pt did not achieve a CR until he had been on drug for 3.5 years, and he remains in remission at 3.8 years. KIT and/or PDGFRA mutational status were highly significant in predicting response. Pts with an exon 11 KIT mutation had a PR rate of 87%, while those with an exon 9 mutation had a PR rate of 48%. Those with no detectable mutation of KIT or PDGFRA had a response rate of 0%. Overall, median time to response was 13 weeks, the median duration of response was 118 weeks, median time to treatment failure was 84 weeks, and median survival was 4.8 years (median not attained for pts with exon 11 mutations). Overall, 63% have died of progressive disease. While secondary resistance and late progression can be seen in GIST pts treated with imatinib, many continue to benefit from drug therapy for a prolonged time, and responses can evolve over an extended treatment period. In particular, pts with exon 11 KIT mutations (the most common exon affected) have very high response rates and superior survival.