Life Raft Research Team meets with European researchers

At 7:00 a.m. on Friday morning, prior to the start of the formal CTOS agenda, the Life Raft Group met with Dr. Jaap Verweij, Erasmus University Medical Center, Rotterdam, the Netherlands and Dr. Martine Van Glabbeke, EORTC Data Center, Brusssels, Belgium. Verweij and Van Blabbeke were two of the investigators in the large phase III European Gleevec trial for GIST patients. The primary purpose of the meeting was to discuss some of the methodology differences between the European phase III study and the Life Raft Group (LRG) study on relapse. Another great benefit of the meeting was just a chance to meet and strengthen relationships with some of the top European researchers.

The LRG study had found a significant difference in relapse rates in GIST patients using Gleevec when comparing actual higher doses (600 mg/day or more) to actual lower doses (less than 600 mg/day). The European study also found a significant difference (in “progression-free survival”) when looking at high vs. low starting doses. A companion phase III trial formed the LRG that their study did not find a difference when comparing starting dose to actual dose. Why these different findings occurred was one of the primary items that we wanted to discuss at this meeting. There were several differences between the European and LRG methodologies that could explain this difference: 1. The European study was a randomized controlled clinical trial and the LRG study was a nonrandomized retrospective study. Non-randomized trials/studies may be biased as a result. 2. The patient population that each looked at when comparing actual dose vs. starting dose was different. The Europeans looked at all patients from 6 months of treatment forward. The LRG looked at patients from 12 months of treatment done in the United States and Canada did not find a significant difference in its early interim analysis. Pictured from left to right are: Dick Singleton, Martine Van Glabbeke, Jerry Call, Norman Scherzer and Jaap Verweij

When looking at the dose patients started on (also known as “intent-totreat” dose), the results for the European study and the LRG study were similar.

The LRG study also found a larger difference when comparing the actual dose patients took vs. the starting dose prescribed to the patient. The LRG study had found that patients receiving a lower dose of Gleevec had an 8% higher relapse rate based on starting dose, but a 23% higher relapse rate based on the actual dose.

Dr. Verweij had previously informed the LRG that their study did not find a difference when comparing starting dose to actual dose. Why these different findings occurred was one of the primary items that we wanted to discuss at this meeting. There were several differences between the European and LRG methodologies that could explain this difference:
1. The European study was a randomized controlled clinical trial and the LRG study was a nonrandomized retrospective study. Non-randomized trials/studies may be biased as a result.
2. The patient population that each looked at when comparing actual dose vs. starting dose was different. The Europeans looked at all patients from 6 months of treatment forward. The LRG looked at patients from 12 months of treatment forward. The Europeans chose the 6 month starting point because of their assumption that almost all dosage changes would occur within the first 6 months.
3. The European study looked at all patients that completed 6 months of the trial, and the LRG only looked at patients that had initial shrinkage on Gleevec. Patients who were judged to be “stable” during initial treatment stages were not included in the LRG study.
4. The LRG attempted to consider dose changes done by the patients themselves. If a patient reported that they were prescribed a higher dose, but they actually took a lower dose, then that patient was considered in the low dose group when looking at actual dosage. A new unpublished study on Gleevec compliance suggests that patient compliance in taking Gleevec may be considerable worse than that reported by patients to the LRG. It seems likely, even though the LRG attempted to consider this issue, that it was probably underreported in our study.