by Shelia Hoar Zahm, Sc.D.
The National Council on Radiation Protection & Measurements estimates that the proportion of annual radiation exposure in the United States from medical sources has approximately tripled since the 1980s. In 2010, the President’s Cancer Panel called for the elimination of unnecessary testing as well as improvements in equipment and operator skill to minimize radiation exposure. Information from DCEG’s Radiation Epidemiology Branch (REB) helped provide the basis for the panel’s concerns, and, since the report, REB has led new studies that shed further light on cancer risks related to low-dose diagnostic and screening procedures and to high-dose radiation therapy.
Amy Berrington de González, D.Phil., senior investigator in REB, has focused her research attention on computed tomography (CT), which is the largest contributor to the recent increase in medical radiation. Although a lifesaving tool when used to help diagnose internal injuries or illness, CT scans often deliver a radiation dose that is up to 10 times higher than that of conventional x-rays. In addition, CT use has increased from approximately three million scans in 1980 to 70 million in 2007. Using information on the number of scans in the United States and radiation doses to various body organs, along with existing data on radiation-related cancer risks, Dr. Berrington de González estimated that approximately 29,000 future cancers could be related to CT scans performed in the United States in 2007 alone. This risk projection prompted NIH to require that makers of scanners used at the NIH Clinical Center incorporate software to measure and track patients’ radiation doses over time. Dr. Berrington de González was also a member of a National Academy of Sciences workshop committee that recently published recommendations on monitoring medical radiation exposures.
Pediatric CTs are of particular concern for several reasons: (1) children might receive a higher radiation dose than necessary if CT settings are not adjusted for children’s smaller body size; (2) children are more sensitive to radiation than adults are; and (3) children have a longer life expectancy than adults do, providing more time for radiation exposure–related cancer to develop. With Dr. Mark Pearce and colleagues from Newcastle University in the United Kingdom (UK), Dr. Berrington de González and Mark Little, D.Phil., senior investigator in REB, recently published the results of a U.K.–NCI study of 180,000 children who underwent CT scans before age 22. The study is the first of its kind to show directly that pediatric CT scans are related to subsequent cancer risk. The report in Lancet showed small but statistically significant increased risks of leukemia and brain tumors within the first decade after the children’s exposure to radiation.
Choonsik Lee, Ph.D., a tenure-track investigator in REB, developed radiation dose estimates for the study using innovative models (called “phantoms”) that approximated the size, shape, and composition of children’s anatomies (see Figure 1). Dr. Lee also developed new software, called NCICT, to assess organ doses from CT exams; the software is being used internationally in other studies of CTs.
The REB team plans to continue to follow the U.K.-NCI cohort and to collaborate with researchers conducting similar studies on pediatric CT use in Canada, Israel, and Australia, which will allow the team to access data for 1.2 million children. Later, the team also will be able to pool data with the ongoing European EPI-CT study to evaluate risk among approximately 2 million children.
The research on risks associated with CTs is affecting clinical practice. Radiology professional groups have joined forces to conduct the Image Wisely® and Image Gently® campaigns to address concerns about the increasing exposure of adults and children to radiation from medical imaging. Dr. Berrington de González noted, “Pediatricians are particularly interested and receptive to the issues of over-imaging in children. In the U.K., there are attempts to change the approach from immediately ordering a CT to having longer clinical observation times in the emergency room and conducting magnetic resonance imaging the following day if imaging is still deemed necessary.”
In addition to diagnostic radiation from CTs, REB investigators study the long-term effects of high-dose radiation used to treat cancer. Lindsay M. Morton, Ph.D., a tenure-track investigator, is leading a large investigation of second cancers of the gastrointestinal tract among patients who received radiotherapy for Hodgkin lymphoma or breast, testicular, or cervical cancer. The study involves collaborators from seven countries, including Dr. Lois B. Travis, former member of REB and currently at the University of Rochester Medical Center in New York, who initiated the study.
The first report from this complex project, published in the Annals of Oncology, described the risks of esophageal cancer among approximately 290,000 women treated for breast cancer. The researchers obtained detailed radiation dosimetry data for 252 women who developed esophageal cancer and 488 women who did not. Because the radiation dose varied more than 20-fold along the length of the esophagus, depending on the location of the radiation treatment fields used to target the breast cancer, Dr. Morton based her analysis on the dose at the location of the esophageal tumor. Esophageal cancer risk increased with increasing radiation dose to the esophageal tumor location, reaching 8.3-fold at doses at or above 35 Gray (see Figure 2). Radiation and other esophageal cancer risk factors, such as smoking, alcohol consumption, and family history of cancer, had multiplicative effects.
Dr. Morton and her colleagues are analyzing the other components of the study and expect to report soon on the risks for stomach and pancreatic cancers after radiation treatment for Hodgkin lymphoma, testicular cancer, and cervical cancer. Dr. Morton noted, “Future research should focus on identifying patients at highest risk for developing radiation-related second cancers—whether because of the specific treatments they received, other exposures, or inherited predisposition.” Understanding the basis of individual susceptibility has important clinical and public health implications for both prevention and treatment of malignancies.
Ruth A. Kleinerman, M.P.H., a staff scientist in REB, Margaret A. Tucker, M.D., Acting Director of DCEG and Director of the Human Genetics Program, and long-time collaborators Dr. David Abramson of Memorial Sloan-Kettering Cancer Center and Dr. Johanna Seddon of Tufts Medical Center have evaluated cancer susceptibility factors in a cohort of retinoblastoma (RB) survivors who have been followed for almost 30 years. RB survivors experience elevated risk of sarcomas, mainly in the area that was treated with radiotherapy. The investigators, along with Dr. Little and Dr. Chu-Ling Yu, a former REB research fellow, recently reported in the Journal of Clinical Oncology that RB survivors with bilateral disease and an inherited germline mutation in the RB1 tumor suppressor gene are at slightly higher risk of a second cancer, particularly melanoma, than RB survivors with a de novo germline mutation, perhaps because of shared genetic alterations. Dr. Morton also is working with Ms. Kleinerman, Dr. Tucker, and Jeannette Wong, M.P.H., a predoctoral fellow in REB, to identify the specific RB1 mutations associated with the development of sarcomas or melanomas among RB survivors compared with RB survivors who did not develop a second cancer.
Although REB investigators conduct research on radiation modalities that were used in the past or that, like CTs, increased in use during the previous decades, the investigators also are alert to the need to evaluate new technologies as they are introduced or as they become widespread in use. Martha S. Linet, M.D., M.P.H., Chief of REB, noted, “We have found that often the best first step in evaluating new imaging technologies is to study the clinicians or radiologic technologists who administer the tests or treatments. They have repeated exposure and wear radiation dosimetry badges, which helps us determine their exposures accurately.” Studying these occupational groups can inform risk projections for patients and help indicate whether a full-scale study of patients might be worthwhile. REB investigators are currently assessing cancer and other mortality among physicians performing fluoroscopically guided interventional procedures and among radiologic technologists assisting with fluoroscopic or nuclear medicine procedures. Branch scientists plan to investigate the feasibility of studying second cancer risks among patients receiving intensity-modulated or proton radiotherapy.