Dosimetric evaluation of interstitial high-dose-rate implants for localised prostate cancer

Abstract

Quantitative evaluation of dose distributions of high-dose-rate prostate implants in order to make a later comparison with clinical outcome. Treatment plans of 169 implants for 161 patients were evaluated using dose-volume histograms. The planning was based on transrectal US imaging and 10 Gy (100%) dose was prescribed to the surface of the prostate. The tolerance dose to urethra and rectum was 125% and 80%, respectively. The volume of the prostate was measured, and its fraction receiving 90%, 100%, 150% and 200% of the prescribed dose was calculated (V90, V100, V150, V200). The dose delivered to 90% of the prostate volume (D90) and the minimum dose in the prostate (Dmin) were determined. The dose nonuniformity ratio (DNR) and the dose homogeneity index (DHI) were calculated to quantify the dose homogeneity. The coverage index (CI) was determined, and the dose conformality to the target volume was assessed with the use of the conformal index (COIN). Maximal dose to rectum (Dr) and urethra (Du) reference points, dose to volume of 2 cm3 of the rectum (D2) and 0.1 cm3 and 1% of the urethra (D0.1, D1) were determined, too. Correlation analysis was performed between point and volume doses. In most patients in-vivo dose measurement was performed in the rectum with semiconductor detectors. The median number of needles was 16, the mean prostate volume was 25.5 cm3. The mean V90, V100, V150 and V200 were 98%, 94%, 41% and 14%, respectively. The mean D90 was 107%, and the Dmin was 82%. The mean dose to rectum and urethra reference points was 75% and 120%, respectively. The mean volume doses were D2=49% for the rectum, D0.1=128% and D1=143% for the urethra. The correlation coefficients were: R(Dr,D2)=0.69, R(Du,D0.1)=0.55, R(Du,D1)=0.23. The mean DNR was 0.39, while the DHI was 0.57. On average, 94% of the target volume received at least the prescribed dose (CI=0.94) and the mean COIN was 0.64. The mean maximal measured dose in the rectum was 2.67 Gy. Our US-based treatment plans based on the real positions of catheters provided acceptable dose distributions. In the majority of our cases the dose to urethra and rectum was kept below the defined tolerance level. The dose of rectal reference points correlated well with rectal dose-volume parameters but for urethra dose determination the use of the D1 volumetric parameter is recommended. Finding correlations between dose-volume parameters and clinical side effects requires further analysis.