These results were presented today at ASTRO's 45th annual meeting in Salt Lake City, Utah, by Eric M. Horwitz, M.D., radiation oncologist at Fox Chase Cancer Center and lead author of the paper.
Using the largest previously pooled dataset of prostate cancer patients treated with radiation therapy alone, the study examined multiple definitions of biochemical control (success or failure of treatment) to determine more accurate guidelines.
The data set included 4,839 patients from nine institutions treated with radiation therapy alone. The data provided biochemical failure and clinical outcomes using the ASTRO definition.
"We reanalyzed this data to correlate alternate biochemical failure definitions with provided clinical endpoints to see which definitions were the most sensitive and specific," explained Horwitz.
The biochemical failure definitions were examined based on distant failure or on clinical failure as endpoints.
The researchers examined about 100 different definitions to develop a more accurate formula for determining treatment failure. Of the 100, only three provided both specificity and sensitivity for distant failure and clinical failure.
The specificity and sensitivity of the ASTRO definition to predict distant failure were 55 percent and 68 percent, respectively. "Using the three alternate definitions of distant failure, we obtained specificity and sensitivity results of 69 and 73 percent, 76 and 72 percent and 72 and 70 percent, respectively," Horwitz said.
The specificity and sensitivity of the ASTRO definition to predict clinical failure was 60 percent and 72 percent, respectively. Again, using three alternate definitions of clinical failure, the researchers demonstrated specificity and sensitivity results of 67 and 78 percent, 66 and 77 percent, and 64 and 74 percent.
"These alternate definitions potentially offer more accuracy in predicting who will remain free of cancer after treatment and who will truly develop metastatic disease based on a patient's post-treatment PSA levels," Horwitz said. "What's more, these results add to the current discussion in the radiation oncology community and to the body of evidence suggesting modifications to the standard ASTRO definition to predict treatment failure."
Other authors of the study include Howard D. Thames, Ph.D., department of biomathematics, University of Texas M. D. Anderson Cancer Center, Houston; Deborah A. Kuban M.D., division of radiation oncology, University of Texas M. D. Anderson Cancer Center; Larry B. Levy, M.S., department of biomathematics, University of Texas M. D. Anderson Cancer Center; Patrick A. Kupelian, M.D., department of radiation oncology, The Cleveland Clinic Foundation, Cleveland, Ohio; Alvaro A. Martinez, M.D., William Beaumont Hospital, Royal Oak, Mich.; Jeffrey M. Michalski, M.D., department of radiation oncology, Mallinckrodt Institute of Radiology, St. Louis, Mo.; Thomas M. Pisansky M.D., division of radiation oncology, Mayo Clinic, Rochester Minn.; Howard M. Sandler, M.D., department of radiation oncology, University of Michigan, Ann Arbor; William U. Shipley, M.D., department of radiation oncology, Massachusetts General Hospital, Boston; Michael J. Zelefsky, M.D., department of radiation oncology, Memorial Sloan-Kettering Cancer Center, New York City; Gerald E. Hanks, M.D., retired from the department of radiation oncology, Fox Chase Cancer Center, Philadelphia, Pa.; and Anthony L. Zietman, M.D., department of radiation oncology, Massachusetts General Hospital.
Fox Chase Cancer Center, one of the nation's first comprehensive cancer centers designated by the National Cancer Institute in 1974, conducts basic, clinical, population and translational research; programs of prevention, detection and treatment of cancer; and community outreach. For more information about Fox Chase activities, visit the Center's web site at www.fccc.edu or call 1-888-FOX CHASE.