Influence of volumes of prostate, rectum, and bladder on treatment planning CT on interfraction prostate shifts during ultrasound image-guided IMRT.

PURPOSE: The purpose of this study was to analyze the relationship between prostate, bladder, and rectum volumes on treatment planning CT day and prostate shifts in the XYZ directions on treatment days. METHODS: Prostate, seminal vesicles, bladder, and rectum were contoured on CT images obtained in supine position. Intensity modulated radiation therapy plans was prepared. Contours were exported to BAT-ultrasound imaging system. Patients were positioned on the couch using skin marks. An ultrasound probe was used to obtain ultrasound images of prostate, bladder, and rectum, which were aligned with CT images. Couch shifts in the XYZ directions as recommended by BAT system were made and recorded. 4698 couch shifts for 42 patients were analyzed to study the correlations between interfraction prostate shifts vs bladder, rectum, and prostate volumes on planning CT. RESULTS: Mean and range of volumes (cc): Bladder: 179 (42-582), rectum: 108 (28-223), and prostate: 55 (21-154). Mean systematic prostate shifts were (cm, +/-SD) right and left lateral: -0.047 +/- 0.16 (-0.361-0.251), anterior and posterior: 0.14 0.3 (-0.466-0.669), and superior and inferior: 0.19 +/- 0.26 (-0.342-0.633). Bladder volume was not correlated with lateral, anterior/posterior, and superior/inferior prostate shifts (P > 0.2). Rectal volume was correlated with anterior/posterior (P < 0.001) but not with lateral and superior/inferior prostate shifts (P > 0.2). The smaller the rectal volume or cross sectional area, the larger was the prostate shift anteriorly and vice versa (P < 0.001). Prostate volume was correlated with superior/inferior (P < 0.05) but not with lateral and anterior/posterior prostate shifts (P > 0.2). The smaller the prostate volume, the larger was prostate shift superiorly and vice versa (P < 0.05). CONCLUSIONS: Prostate and rectal volumes, but not bladder volumes, on treatment planning CT influenced prostate position on treatment fractions. Daily image-guided adoptive radiotherapy would be required for patients with distended or empty rectum on planning CT to reduce rectal toxicity in the case of empty rectum and to minimize geometric miss of prostate.

The potential for dose dumping in normal tissues with IMRT for pelvic and H&N cancers.

The purpose of this study is to understand the potential for dose dumping in normal tissues (>85% of prescription dose) and to analyze effectiveness of techniques used in reducing dose dumping during IMRT. Two hundred sixty-five intensity modulated radiation therapy (IMRT) plans for 55 patients with prostate, head-and-neck (H&N), and cervix cancers with 6-MV photon beams and >5 fields were reviewed to analyze why dose dumping occurred, and the techniques used to reduce dose dumping. Various factors including gantry angles, depth of beams (100-SSD), duration of optimization, severity of dose-volume constraints (DVC) on normal structures, and spatial location of planning treatment volumes (PTV) were reviewed in relation to dose dumping. In addition, the effect of partial contouring of rectum in beam's path on dose dumping in rectum was studied. Dose dumping occurred at d(max) in 17 pelvic cases (85% to 129%). This was related to (1) depth of beams (100 SSD [source-to-skin distance]), (2) PTV located between normal structures with DVC, and (3) relative lack of rectum and bladder in beam's path. Dose dumping could be reduced to 85% by changing beam angles and/or DVC for normal structures in 5 cases and by creating "phantom structures" in 12 cases. Decreasing the iterations (duration of optimization) also reduced dose dumping and monitor units (MUs). Part of uncontoured rectum, if present in the field, received a higher dose than the contoured rectum with DVC, indicating that complete delineation of normal structures and DVC is necessary to prevent dose dumping. In H&N, when PTV extends inadvertently into air beyond the body even by a few millimeters, dose dumping occurred in beam's path (220% for 5-field and 170%, 7-field plans). Keeping PTV margins within body contour reduced this type of dose dumping. Beamlet optimization, duration of optimization, spatial location of PTV, and DVC on PTV and normal structures has the potential to cause dose dumping. However, these factors are an integral part of IMRT inverse planning. Therefore, understanding these aspects and use of appropriate technique/s would reduce or eliminate the dose dumping and minimize time to obtain optimum plan.

Single course IMRT plan to deliver 45 Gy to seminal vesicles and 81 Gy to prostate in 45 fractions.

We treat prostate and seminal vesicles (SV) to 45 Gy in 25 fractions (course 1) and boost prostate to 81 Gy in 20 more fractions (course 2) with Intensity Modulated Radiation Therapy (IMRT). This two-course IMRT with 45 fractions delivered a non-uniform dose to SV and required two plans and two QA procedures. We used Linear Quadratic (LQ) model to develop a single course IMRT plan to treat SV to a uniform dose, which has the same biological effective dose (BED) as that of 45 Gy in 25 fractions and prostate to 81 Gy, in 45 fractions. Single course IMRT plans were compared with two-course IMRT plans, retrospectively for 14 patients. With two-course IMRT, prescription to prostate and SV was 45 Gy in 25 fractions and to prostate only was 36 Gy in 20 fractions, at 1.8 Gy/fraction. With 45-fraction single course IMRT plan, prescription to prostate was 81 Gy and to SV was 52 or 56 Gy for a alpha/beta of 1 and 3, respectively. 52 Gy delivered in 45 fractions has the same BED of 72 Gy3 as that of delivering 45 Gy in 25 fractions, and is called Matched Effective Dose (MED). LQ model was used to calculate the BED and MED to SV for alpha/beta values of 1-10. Comparison between two-course and single course IMRT plans was in terms of MUs, dose-max, and dose volume constraints (DVC). DVC were: 95% PTV to be covered by at least 95% of prescription dose; and 70, 50, and 30% of bladder and rectum should not receive more than 40, 60, and 70% of 81 Gy. SV Volumes ranged from 2.9-30 cc. With two-course IMRT plans, mean dose to SV was non-uniform and varied between patients by 48% (54 to 80 Gy). With single-course IMRT plan, mean dose to SV was more uniform and varied between patients by only 9.6% (58.2 to 63.8 Gy), to deliver MED of 56 Gy for alpha/beta - 1. Single course IMRT plan MUs were slightly larger than those for two-course IMRT plans, but within the range seen for two-course plans (549-959 MUs, n=51). Dose max for single-course plans were similar to two-course plans. Doses to PTV, rectum and bladder with single course plans were as per DVC and comparable to two-course plans. Single course IMRT plan reduces IMRT planning and QA time to half.