Is it necessary to eliminate the posterior dose margin in prostate brachytherapy to achieve an acceptably low risk of late rectal morbidity?
PURPOSE: The use of a posterior dose margin in (125)I prostate brachytherapy is controversial. The posterior margin is often eliminated to lower the risk of late rectal morbidity (Radiation Therapy Oncology Group protocols 9805 and P-0019), but this may compromise the posterior prostate dose coverage. The purpose of this work is to determine whether it is necessary to eliminate the posterior margin to achieve an acceptably low risk of Grade 2 (bleeding/ulceration) late rectal morbidity. METHODS AND MATERIALS: The present work is an extension of a previous study in which we reported the probability of Grade 2 late rectal morbidity (bleeding/ulceration) as a function of the maximum rectal dose. Here, we define the relationship between the maximum rectal dose and the width of the posterior dose margin. From this relationship, and the probability of late morbidity determined earlier, we assessed the probability of late rectal morbidity as a function of the width of the posterior dose margin. The present work was based on the preplans of 15 permanent (125)I prostate seed implants having volumes ranging from 19 to 78 cm(3). All used peripheral loading with (125)I sources in the range 0.35-0.45 mCi. The maximum rectal dose was taken to be the isodose line coincident with the posterior edge of the prostate. The maximum rectal dose and the corresponding margin width were determined from the isodose distribution. The sensitivity of the relationship between the maximum rectal dose and the margin width to the prostate volume, the seed density, and the distance of the most posterior seeds from the posterior edge of the prostate was investigated. RESULTS: The maximum rectal dose is directly proportional to the margin width. This relationship is relatively insensitive to the prostate volume and the seed density, but is sensitive to the location of the posterior seeds relative to the posterior edge of the gland. Moving the seeds from 5 mm to 3 mm from the edge typically increased the maximum rectal dose by 17%. With the posterior seeds 3 mm from the edge, the maximum rectal doses that corresponded to 1 mm, 2 mm, 3 mm, 4 mm, and 5 mm margins were 187 +/- 6 Gy, 222 +/- 8 Gy, 257 +/- 11 Gy, 292 +/- 14 Gy, and 327 +/- 17 Gy, respectively. The corresponding probabilities that a patient will experience late rectal morbidity are < or =1%, < or =2%, < or =3%, < or =5%, and < or =7%, respectively. CONCLUSIONS: Our results indicate that a 2-3-mm posterior dose margin can be used in prostate brachytherapy with a relatively low (2-3%) risk of Grade 2 (bleeding/ulceration) late rectal morbidity, provided the sources in the posterior row are implanted at least 3 mm from the edge of the prostate. A practical guideline is to keep the maximum rectal dose below 150% of the target dose.