Bladder movement during radiation therapy for bladder cancer: implications for treatment planning.

PURPOSE To describe and quantify bladder movement during radical radiation therapy (RT). To attempt to identify factors that predict for excessive alterations in bladder position. To use the above information to assist in defining the "adequate" planning target volume margin. METHODS AND MATERIALS Thirty patients with bladder cancer suitable for radical courses of RT were followed prospectively. Patients had an initial planning computerized tomography (CT) scan of the pelvis and three subsequent scans performed weekly during the treatment period. The following measurements were made on each scan in the midbladder slice: maximum anteroposterior (AP) and lateral bladder dimensions, AP rectal diameter, and the distance (margin) between the bladder walls (anterior, posterior, right, and left lateral) and the 95% isodose line. Various patient and tumor data, including bladder and bowel symptoms, were recorded to attempt correlation with bladder movement. RESULTS Bladder size: the median bladder size (area) over all scans in all patients was 36.9 cm2 (range: 16.2 to 80.9 cm2). The change in bladder area across each sequence varied from 3.3 to 29.1 cm2 (7-55% change in area between scans). Patients with bladders of larger than the median size on the planning scan (despite emptying) were more likely to have alteration in size than those with small bladders, and this change was in the direction of contraction (p = 0.01). Bladder displacement: bladder wall movement of > 1.5 cm was defined as "significant." Eighteen of 30 patients (60%) demonstrated "significant" movement of at least one bladder wall relative to the original isodose plot. Movement resulting in margin reduction occurred in 10 patients (33%). Two patients required treatment replanning due to consistently altered bladder position. There was no pattern to displacement through RT, and all walls were at approximately equal risk of movement. Factors influencing bladder movement: posterior bladder wall movement appeared to relate to "marked" (>2 cm) rectal diameter change. There was a trend for patients with larger amounts of residual bladder tumor (greater than the median) to exhibit more bladder movement; 11 of 14 "moved" compared with 7 of 16 patients with less residual tumor. Other clinical factors including age, sex, body size, acute RT reaction, and tumor stage did not appear to relate to bladder movement. CONCLUSION Bladder movement during RT is clinically relevant and is random with respect to both time and direction. We recommend, at least with respect to tumor-bearing regions of the bladder, that no less than a 2.0 cm margin should be allowed.