Association of Computed Tomography Morphologic Criteria With Pathologic Response and Survival in Patients Treated With Bevacizumab for Colorectal Liver Metastases

Abstract

The standard criteria used to evaluate tumor response, the Response Evaluation Criteria in Solid Tumors (RECIST), were developed to assess tumor shrinkage after cytotoxic chemotherapy and may be limited in assessing response to biologic agents, which have a cytostatic mechanism of action. To validate novel tumor response criteria based on morphologic changes observed on computed tomography (CT) in patients with colorectal liver metastases treated with bevacizumab-containing chemotherapy regimens. A total of 234 colorectal liver metastases were analyzed from 50 patients who underwent hepatic resection after preoperative chemotherapy that included bevacizumab at a comprehensive US cancer center from 2004 to 2007; date of last follow-up was March 2008. All patients underwent routine contrast-enhanced CT at the start and end of preoperative therapy. Three blinded, independent radiologists evaluated images for morphologic response, based on metastases changing from heterogeneous masses with ill-defined margins into homogeneous hypoattenuating lesions with sharp borders. These criteria were validated with a separate cohort of 82 patients with unresectable colorectal liver metastases treated with bevacizumab-containing chemotherapy. Response determined using morphologic criteria and RECIST was correlated with pathologic response in resected liver specimens and with patient survival. Interobserver agreement for scoring morphologic changes was good among 3 radiologists (kappa, 0.68-0.78; 95% confidence interval [CI], 0.51-0.93). In resected tumor specimens, the median (interquartile range [IQR]) percentages of residual tumor cells for optimal morphologic response was 20% (10%-30%); for incomplete response, 50% (30%-60%); and no response, 70% (60%-70%; P < .001). With RECIST, the median (IQR) percentages of residual tumor cells were for partial response 30% (10%-60%); for stable disease, 50% (20%-70%); and for progressive disease, 70% (65%-70%; P = .04). Among patients who underwent hepatic resection, median overall survival was not yet reached with optimal morphologic response and 25 months (95% CI, 20.2-29.8 months) with incomplete or no morphologic response (P = .03). In the validation cohort, patients with optimal morphologic response had median overall survival of 31 months (95% CI, 26.8-35.2 months) compared with 19 months (95% CI, 14.6-23.4 months) with incomplete or no morphologic response (P = .009). RECIST did not correlate with survival in either the surgical or validation cohort. Among patients with colorectal liver metastases treated with bevacizumab-containing chemotherapy, CT-based morphologic criteria had a statistically significant association with pathologic response and overall survival.