Early morphological change (EMC) assessed with computed tomography (CT) may be a useful tool in evaluating early responses to anti-tumor treatments in patients with gastrointestinal stromal tumor (GIST).

A new retrospective study investigated the usefulness of EMC in predicting the clinical outcomes of imatinib treatment for GIST patients. The results of the study, published in Gastric Cancer, showed that patients with tumors that displayed a morphological response (MR) had a statistically longer progression-free survival (PFS) than patients with tumors that did not show a response (49 vs 23 months, respectively; P =.0039).

The authors mentioned that previous clinical trials showed that 5% of patients with unresectable or metastatic GIST had primary resistance to imatinib, with an additional 14% having early resistance. They also pointed out that half of patients treated with imatinib develop resistance within 2 years.

Continue Reading

“Considering the development of subsequent treatment after imatinib, including sunitinib and regorafenib, the importance of predicting the efficacy of imatinib is increasing in order to provide every patient with the best therapy,” the researchers wrote. “Our results have a possibility to contribute to proper decision-making in the course of treatment, including withdrawal of imatinib, surgical intervention, and switching to subsequent drugs in clinical practice.”

Read more about GIST therapies

The current study also compared EMC to 2 other treatment response estimation methods — the response evaluation criteria in solid tumors (RECIST) and Choi criteria. Kaplan-Meier curves showed that EMC status was a better predictor of PFS than RECIST and Choi criteria.

The authors mentioned that RECIST, which is based on tumor size, may underestimate the response to imatinib because it often takes several months for GISTs to shrink and that internal changes within tumors can occur without changes to the tumor size. The Choi criteria were found to overestimate treatment response in the current study and may not reflect heterogeneous attenuation within the tumor as it is based on the mean attenuation value of the region of interest.

A total of 66 patients were included in the retrospective study, with 45 (68.2%) being classified as having an active MR and the remaining 21 (31.8%) not showing MR. Patients were initially categorized into 3 groups based on the morphological appearance of the tumor on CT scans.

Group 1 included tumors with homogeneous attenuation with sharply defined edges while group 3 was heterogeneous with a poorly defined outline. Group 2 was between the other groups. Patients received baseline CT scans and then a follow-up scan within a few months of treatment (median of 52.0 days).

Patients whose grouping improved from group 3 or 2 up to group 1 were classified as having an “optimal” response with those moving from group 3 to group 2 were classified as having an “incomplete” response. The optimal and incomplete response groups were included in the active MR group while patients who did not change groups or whose grouping got worse were said to have no MR.


Ishida T, Takahashi T, Nishida T, et al. New response evaluation criteria using early morphological change in imatinib treatment for patients with gastrointestinal stromal tumor. Gastric Cancer. Published online August 20, 2021. doi:10.1007/s10120-021-01234-0