Estimates of lung tumor size based on computed tomography (CT) scans can vary from scan to scan even when the actual size of the tumor remains the same. These results were published in the Journal of Clinical Oncology.
CT scans are used for imaging body tissues and organs. X-ray transmissions are converted to detailed images using a computer to synthesize X-ray data. The images are displayed as a cross-section of the body at any level from the head to the feet. Repeated CT scans may be used to assess response to cancer treatment and to monitor patients for cancer progression.
Although cancer imaging tools have become increasingly precise, it’s possible for two back-to-back scans to produce slightly different results. Several factors may contribute to this variability, including how the machine captures the images and how the radiologist performs the measurements.
Understanding the extent of this variability is important when interpreting the results of a repeat CT scan. For example, if the results of a repeat scan indicate that the tumor is 2 millimeters larger than it was previously, could this be due to variability in measurement or is it likely that the cancer actually grew?
To explore variability across CT scans, researchers conducted a study among 30 patients with non-small cell lung cancer that measured at least one centimeter in diameter. Patients underwent two separate CT scans within a period of about 15 minutes (a period during which the actual size of the cancer would not have changed). The images for each patient were viewed side-by-side by three experienced radiologists. The radiologists were not told how much time had elapsed between the two scans.
- For more than half the patients, the estimates of tumor size produced by the two scans varied by more than 1 mm. For one-third of patients, the estimates of tumor size varied by more than 2 mm.
- The differences between the two scans ranged from a 23% reduction in tumor size to a 31% increase in tumor size.
- For a majority of patients (84%), the second measurement differed from the first by 10% or less.
The results of this study suggest that small changes in lung tumor size on repeat CT scans should be interpreted with caution. The change may be due to variability in the CT scanning process rather a change in the tumor itself.
Reference: Oxnard GR, Zhao B, Sima CS et al. Variability of lung tumor measurements on repeat computed tomography scans taken within 15 minutes. Journal of Clinical Oncology. Early online publication July 5, 2011.