Quantitative imaging in oncology patients: Part 1, radiology practice patterns at major U.S. cancer centers

The objective of our study was to examine radiologists' opinions and practice patterns concerning tumor measurements in cancer patients.

An electronic mail survey was sent to 565 abdominal imaging radiologists at 55 U.S. National Cancer Institute (NCI)-funded cancer centers. The survey contained questions about departmental demographics, procedures for interpretation of imaging in oncologic patients, and opinions concerning the role of radiologists in using the Response Evaluation Criteria in Solid Tumors (RECIST) system for tumor measurements.

Two hundred ninety-six responses (52%) were received. The distribution of the size of the respondents' abdominal imaging groups was as follows: 1-5 (16/295, 5%), 6-10 (112/295, 38%), 11-15 (77/295, 26%), and > 20 (73/295, 25%). Most respondents dictate some but not all tumor measurements in the first clinical scan (236/270, 87%). For follow-up imaging, 95% (255/268) of respondents dictate tumor measurements for selected index lesions. Most respondents believe inclusion of tumor measurements in the first scan is the responsibility of the radiologist (248/262, 95%). Ninety percent of respondents (235/261) believe inclusion of several index lesion measurements is satisfactory to document disease activity. Eighty-two percent (214/260) of respondents were familiar with RECIST. Forty-two percent (110/262) of respondents' departments have a centralized process for approval of industry-sponsored oncologic trials in which imaging is an important component of the protocol end point.

Most oncologic imaging at NCI-sponsored cancer centers includes tumor measurements on initial and follow-up imaging. Very few radiology departments have a centralized process for approval of clinical trial protocols that require imaging.