Quantitative imaging in oncology patients: Part 2, oncologists' opinions and expectations at major U.S. cancer centers.

Abstract

The purpose of this article is to examine oncologists' opinions and expectations concerning imaging and tumor measurements in patients with cancer. An electronic mail survey was sent to 2,400 medical, gynecologic, and radiation oncologists at 55 U.S. National Cancer Institute-funded cancer centers. The survey contained questions about departmental demographics, opinions regarding imaging for patients with cancer, PET/CT utilization, and utilization of the Response Evaluation Criteria in Solid Tumors (RECIST) system in therapy protocols that use imaging as a therapeutic end point. A total of 492 responses (21%) were received. Sixty percent (294) of respondents were medical oncologists, 9% (45) were gynecologic oncologists, 26% (127) were radiation oncologists, and 5% (25) answered "Other." Ninety-eight percent (431/438) of respondents provide clinical care, and 99% (420/425) have participated in clinical trials. Most respondents (94% [410/438]) expect some or all tumors to be measured at the time of standard initial clinical imaging. Over half (65% [275/426]) think that tumor measurements should be bidimensional. Only 25% (101/400) of respondents' institutions have department rules on the implementation of RECIST measurements. Sixty-eight percent of participants (269/397) think that RECIST is flawed but serviceable. Over half of respondents (56% [221/398]) were not familiar with RECIST 1.1 modifications. Most oncologists at National Cancer Institute-sponsored cancer centers expect tumor measurements to be made in the routine imaging of patients with cancer. Almost two thirds of respondents think that bidimensional measurements of index lesions are satisfactory in routine oncologic imaging. Little consensus exists in the implementation of RECIST measurements for clinical trials at these centers.