Abstract
Abstract only
93
Background: To evaluate the spatial pattern of permanent prostate brachytherapy (PBT) learning curve using post-implant multi-sector dosimetric analysis. Methods: Two hundred patients treated with I-125 PBT at a single institution were assessed. A 12 sector analysis of post-implant CT was created for each case by dividing the prostate base, mid gland and apex into 4 sectors each. Dosimetric analysis was evaluated for whole prostate and individual sectors assessing: minimum dose to 90% of prostate (D
90
) and dose to 0.1 cm
3
of the rectum (D
0.1cc
). Linear regression analysis and Pearson’s correlation co-efficient were used to investigate changes in D
90
with case number. Results: Global prostate analysis demonstrated a post-implant learning curve with significant improvement in D
90
(p<0.01) in the last 50 cases (157.0 Gy ± 9.7) when compared to the initial 50 (141.6 Gy ± 14.5). Linear regression confirmed a significant improvement in global D
90
with case number (r
2
=0.19, p=0.0001) at a rate of 0.11 Gy/case. Sector analysis demonstrated greatest improvement in base sectors with increasing case number with an improvement of 0.16 Gy/case (Table; p=0.049 when compared to global D
90
). Further, D
90
in posterior mid-gland reduced with case number and correlated with reduced rectal D
0.1cc
. Conclusions: Sector analysis demonstrated increased spatial information on the PBT learning curve which may enable on-going improvements in implant quality. [Table: see text]