Experimental and Clinical Therapeutics

Monday, October 17, 2005 3:00 PM-5:00 PM Exhibit Hall

(PP124) Clonogenic survival does not uniformly increase with split time between halves of 2-Gy fractions: implications for IMRT.

Khan, Atif*,1, Mehta, Par1, Foutch, Jennifer1, Kirk, Michael1, Chu, Jim 1, Abrams, Ross1, Blazek, Ed1, 1 Department of Radiation Oncology, Chicago, IL, USA

ABSTRACT- Purpose/Objective: Intensity-modulated radiation therapy (IMRT) can deliver highly conformal doses to complicated anatomical structures. Inherent to dose delivery by IMRT, however, is increased delivery time for each treatment fraction. There is concern that if sublethal damage repair (SLDR) were sufficiently rapid, it might be necessary to increment dose relative to more rapid delivery methods to achieve adequate tumor cell killing. We sought to measure SLDR on a time scale relevant to IMRT (0-60 min) in two glioblastoma (GBM) human tumor cell lines. Materials/Methods: U87MG and T98G GBM cells were plated into 96-well plates as single-cell suspensions. Beginning 4 h after plating, cells received 2 Gy per day in 1 Gy half-fractions separated by intervals of 0, 5, 10, 15, 20, 30, or 60 minutes. The radiation field was homogeneous within 6%, and plates were thermally insulated during irradiations. Irradiations continued for 5 consecutive days to a cumulative dose of 10 Gy, with 4 replicates per condition and 4 unirradiated plates to measure plating efficiency (PE). After incubation for at least two weeks, colonies were fixed and stained. Clonogenic survival was determined by dilution analysis: SF(D)=-ln(WCF)/(N*PE); WCF = no. of wells without colonies/96 and N = no. of cells plated per well. The mean values of SF2x5 from each experiment were normalized to the laboratory standard value for each cell line. Statistical tests were performed using PSI-Plot. Results: The figure [see poster] shows the SF2x5 values for T98G and U87MG. Points are means for 3 experiments with 4 plates per experiment (n=12); error bars are standard errors of the mean. Significantly different survivals for T98G were: 0 vs. 60 (p=0.025) and 30 vs. 60 (p = 0.048), while 15 vs. 60 was nearly significant (p =0.077). Significantly different survivals for U87MG were: 0 vs. 60 (p=0.020), 5 vs. 20 (p=0.038), 5 vs. 60 (p=0.042), 10 vs. 20 (p=0.009), 15 vs. 20 (p=0.007), and 20 vs. 60 (p =0.002), while 20 vs. 30 was nearly significant (p= 0.076). The U87MG, but not the T98G, p-values are valid for multiple comparisons by Fisher's Protected LSD test. Conclusions: A 60-min interval between half-fractions spared both cell lines, whereas a 30-min or shorter split spared neither. This implies that IMRT fractions delivered within 30 min should not be less effective than the same dose delivered at 0.8 Gy/min without interruption. U87MG survival was smaller, however, for the 20-min split than for several shorter or longer splits. Although thermal or other artifacts cannot yet be excluded, we may have detected a complicated variation in cell vulnerability to subsequent irradiation in the first few minutes after a 1-Gy dose. The extreme range of survivals observed in this study, if extrapolated to 30 fractions, could cause 1.9-fold and 7.8-fold differences in residual tumor clonogen number for T98G and U87MG, respectively, corresponding to tumor cell killing by 2 (2-Gy) fractions for T98G or 3 fractions for U87MG.

Key words:

Internet Services provided by
Allen Press, Inc. | 810 E. 10th St. | Lawrence, Kansas 66044 USA
e-mail assystant-helpdesk@allenpress.com | Web www.allenpress.com
2005 RRS