The RECIST criteria may be limited in assessing response to biologic agents such as bevacizumab (BVZ). Computed tomography-based morphologic criteria (CTMC) might be an alternative to evaluate response in these cases. The aim of this trial was to evaluate the correlation of overall objective responses evaluated by RECIST with CTMC and the pathologic response (pR) after the resection of liver metastases from colorectal cancer.
Patients with ≤ 4 resectable CLM, with no prior chemotherapy for metastatic disease, received 3 cycles of capecitabine/oxaliplatin (XELOX)+BVZ. Response was evaluated using RECIST criteria and CTMC. For CTMC, metastasis was assigned to 1 of 3 groups (Table): CTMC were defined as optimal (metastasis changed from a group 3 or 2 to a 1), incomplete (group changed from 3 to 2), and none (the group had not changed or increased). Those patients without progression, received another cycle of XELOX prior to surgery. After surgery, BVZ + XELOX was given for up to 4 additional cycles. pR was scored as minor (≥50% of residual tumor cells), major (1-49% residual tumor cells), and complete (no residual tumor cells detected).Table:
|Computed Tomographic Morphologic Groups|
|Computed Tomographic Tumor Characteristics|
|Morphology Group||Overall Attenuation||Tumor-Liver Interface||Peripheral Rim of enhancement|
|3||Heterogeneous||Ill defined||May be present|
|2||Mixed||Variable||If initially present, partially resolved|
|1||Homogeneous and hypoattenuating||Sharp||If initially present, completely resolved|
83 patients were recruited. 68 were evaluated by RECIST and 67 of them by CTMC; 60 underwent surgery, 51 with metastases resection and histologically analysis. 33 of 68 pts reached partial response by RECIST (49%, 95%CI 37-60%) and 58 had optimal (26) or incomplete (32) response by CTMC (85%, 95%CI 75-92%). Complete (12) or major (30) pR was achieved in 42 of 51 pts (82%, 95%CI 70-90%). Although there was no correlation between the RECIST criteria, CTMC and pR, CTMC was more specific for predicting complete/major pR than RECIST (36 of 42, CTMC vs. 24 of 42, RECIST, p=.0038). In patients with liver resection overall survival at 48 months was 65.8%.
CTMC was more specific for predicting complete/major pR than RECIST criteria. After liver metastases resection, 82% of patients had complete or major pR. CTMC seem to be a better surrogate marker of objective pR than RECIST.
Clinical trial identification
Legal entity responsible for the study
All authors have declared no conflicts of interest.