Predicting rectal and bladder overdose during the course of prostate radiotherapy using dose-volume data from initial...

To evaluate whether information from the initial fractions can determine which patients are likely to consistently exceed their planning dose-volume constraints during the course of radiotherapy for prostate cancer.

Ten patients with high-risk prostate cancer were treated with helical tomotherapy to a dose of 60 Gy in 20 fractions. The prostate, rectum, and bladder were recontoured on their daily megavoltage computed tomography scans and the dose was recalculated. The bladder and rectal volumes (in mL) receiving ≥100% and ≥70% of the prescribed dose in each fraction and in the original plans were recorded. A fraction for which the difference between planned and delivered was more than 2 mL was considered a volume failure. Similarly if the difference in the planned and delivered maximum dose (D(max)) was ≥1% for the rectum and bladder, the fraction was considered a dose failure. Each patient's first 3 to 5 fractions were analyzed to determine if they correctly identified those patients who would consistently fail (i.e., ≥20% of fractions) during the course of their radiotherapy.

Six parameters were studied; the rectal volume (RV) and bladder volumes (BV) (in mL) received ≥100% and ≥70% of the prescribed dose and maximum dose to 2 mL of the rectum and bladder. This was given by RV₁₀₀, RV₇₀, BV₁₀₀, BV₇₀, RD(max), and BD(max), respectively. When more than 1 of the first 3 fractions exceed the planning constraint as defined, it accurately predicts consistent failures through the course of the treatment. This method is able to correctly identify the consistent failures about 80% (RV₇₀, BV₁₀₀, and RV₁₀₀), 90% (BV₇₀), and 100% (RD(max) and BD(max)) of the times.

This study demonstrates the feasibility of a method accurately identifying patients who are likely to consistently exceed the planning constraints during the course of their treatment, using information from the first 3 to 5 fractions.