Imaging in clinical trials calls for more than one-dimensional measurements

CHICAGO-Using imaging to monitor drug response is becoming an increasingly important field of research, and radiologists who are not involved in clinical trials should at least become aware of it, said presenters at an RSNA 2008 refresher course (RC730A).

The radiology community needs to recognize that clinical trials are a special case of treatment, said C. Carl Jaffe, MD, of the National Cancer Institute in Bethesda, Md. “Imaging to monitor response to treatment now serves as a major role in FDA drug approval and clinical trial therapy efficacy assessment, but few images are directly engaged in the discipline of formal scientific processes demanded by clinical trials,” said Dr. Jaffe.

Among the limitations of most clinical studies using imaging to predict tumor response, said Dr. Jaffe, is that they are mainly single-center, nonblinded studies with a limited number of patients and locally defi ned image acquisition protocols. They also “tend to be diffuse in terms of rigor,” he said, adding that “validation by multicenter trials is needed.”

Cancer therapies are in transition, with the cost of drugs on the rise, said Dr Jaffe. He pointed out that the traditional end point of FDA approval is survival, which comes with itsshare of drawbacks. “It requires a very large sample size and a long follow-up period, and crossover therapy may ‘wash out’ a survival effect.”

Dr. Jaffe and co-presenter, Lawrence H. Schwartz, MD, highlighted the changes required for imaging to be used to monitor treatment responses, including the need for uniformity of instrumentation and protocol-specifi c acquisition.

They also examined the principles of response assessment, including understanding the role of imaging in FDA cancer drug therapy approvals, knowing the appropriate use of different types of tumor response assessments-anatomic, dynamic, and molecular-and comprehending the necessary requirements to assure reliable and reproducible quantitative imaging data.

Dr. Schwartz, associate attending radiologist, vice-chair for technology development, and director of MR imaging at Memorial Sloan-Kettering Cancer Center in New York, discussed whether or not radiologists have been measuring treatment response accurately.

“We really have to be careful of the measurements we use in clinical trials and there really must be a set of rules governing the standards of response assessment,” he said. “The question is, can we do better than a unidimensional measurement of tumors?”

According to the NCI, plans call for improving RECIST by incorporating volumetric anatomic, dynamic contrast-enhanced, and functional (molecular) imaging (see Table below).

Reprinted from the RSNA Daily Bulletin newsletter, December 3, 2008.