CT tumor measurement for therapeutic response assessment: Comparison of unidimensional, bidimensional, and volumetric techniques - Initial observations

PURPOSE: To compare unidimensional, bidimensional, and volumetric techniques for evaluation of treatment response in patients with liver metastases from breast cancer in a phase III clinical trial. MATERIALS AND METHODS: Helical computed tomographic (CT) studies in 38 patients with liver metastases from breast cancer who were enrolled in a phase III clinical trial were evaluated before treatment and at 6 months after treatment. Two subspecialty radiologists measured all lesions on transverse CT scans with use of electronic calipers according to both unidimensional and bidimensional criteria. Volumetric measurements were made by tracing individual lesions. Measurements of individual lesions were summed to obtain patient response, which was categorized as complete response, disappearance of lesions; partial response, greater than 30% decrease in tumor diameter (unidimensional), greater than 50% reduction in tumor area (bidimensional), or greater than 65% reduction in tumor volume (volumetric); disease progression, greater than 20% increase in tumor diameter, greater than 25% increase in tumor area, or greater than 73% increase in tumor volume: or stable disease (size response other than that of complete response, partial response, or disease progression). RESULTS: In 37 patients, there was concordant treatment response with use of unidimensional and bidimensional techniques. Volumetric measurement produced results different from those of the unidimensional and bidimensional techniques in 12 and 13 patients, respectively. In six patients with partial response per unidimensional and bidimensional criteria, the response based on the volumetric technique was stable disease. In two patients with stable disease per bidimensional and unidimensional criteria, the response was partial response by volumetric measurement. In four patients with disease progression per bidimensional and unidimensional criteria, the response was stable disease per volumetric criteria. CONCLUSION: Volumetric measurement of tumor burden gives different results for treatment response compared with that of the unidimensional or bidimensional technique in a considerable proportion of patients.