Purpose: Response Evaluation Criteria in Solid Tumors (RECIST) are insensitive in evaluating imatinib-treated gastrointestinal stromal tumors (GISTs). Response by Choi criteria, a 10% decrease in size or a 15% decrease in density on contrast-enhanced CT, correlated well in a small training set of patients who showed response as measured by positron emission tomography, and was more predictive of time to tumor progression (TTP) than response by RECIST. This study was designed to validate these observations in an independent data set.
Patients and methods: Fifty-eight patients with imatinib-treated GISTs were evaluated by RECIST and Choi criteria. TTP was compared with TTP in the training set. Patients were analyzed initially with follow-up to 28 months, extended to 60 months for survival analysis.
Results: Patients who met Choi response criteria on CT at 2 months had significantly better TTP than those who did not (P = .0002), whereas response group by RECIST was not significantly correlated with TTP. Even when the 98 patients from both sets were analyzed together, the response group by RECIST did not correlate significantly with TTP, whereas response group by Choi criteria did correlate significantly with TTP. Disease-specific survival (DSS) was also significantly correlated with response group by Choi criteria (P = .04), but not with response group by RECIST.
Conclusion: Choi response criteria are reproducible, more sensitive, and more precise than RECIST in assessing the response of GISTs to imatinib mesylate. Response by Choi criteria, unlike response by RECIST, correlates significantly with TTP and DSS. Response by Choi criteria should be incorporated routinely into future studies of GIST therapy. We should desist using RECIST, at least in GIST.