37PBP CANCER CLINICAL TRIALS PARTICIPATION RATES IN FLORIDA

Monday, October 20, 2008
Columbus A-C (Hyatt Regency Penns Landing)
Margaret M. Byrne, PhD1, Julie Kornfeld, MPH2, Stefan Gluck, MD, PHD2, Judith Hurley, MD2 and Michael Antoni2, (1)University of Miami, Miami, USA USA, (2)University of Miami, Miami, FL
Purpose: Participation in cancer clinical trials in Florida has previously been shown to be very low. The purpose of this research project was to determine rates of adult participation in cancer clinical trials in Florida, and to examine how these rates vary by age, sex, race and cancer site.

Methods: For the years 1998-2005 we collected data from two sources: 1) incidence of cancer in the state of Florida from the Florida Cancer Data System; 2) participation incidence in clinical trials run by the National Cancer Institute’s Clinical Trials Cooperative Groups (CTCGs). These two sources of data were used to calculate participation rates. Data include demographic and cancer site information. In addition, although CTCG participation data is the most comprehensive and often used, it does not collect information from all types of trials. Therefore, we also explored data from 2004 and 2005 on clinical trial participation incidence (no demographic information) at facilities accredited by the American College of Surgeons Commission on Cancer.

Results: Participation rates calculated using data from CTCG trials showed low participation rates (mean 0.75%, st.dev. 0.15%) over the 8 years of the study, with no obvious trends over time in rates. Regression analysis shows that there are significant differences among cancer types in participation rates, with breast and gynecologic cancer patients having the highest participation rates (p<0.001). Paired t-tests showed that African Americans and Whites did not have different rates of participation (p=0.09) but Hispanics had lower participation than non-Hispanics (p<0.05). Women had higher rates than men (p<0.001). An examination of ACOS data showed that these data captured a much higher rate of participation (5.05% in 2004, 8.41% in 2005).

Conclusions: Although CTCG data has been used to describe participation in cancer clinical trials, and is the only data that includes patient demographics, it has important limitations. Given the fact that many clinical trials conducted in Florida are funded from sources other than the National Cancer Institute, the higher rate of participation found in ACOS cancer data points to the need for further data collection and analysis. Nevertheless, if participation by subgroups is not biased in CTCG data, we can conclude that women and non-Hispanics are significantly more likely to participate in cancer clinical trials.